1

•-- • :

HISTORY OF THE GREAT WAR

BASED ON OFFICIAL DOCUMENTS.

MEDICAL SERVICES DISEASES OF THE WAR.

VOL. I

EDITED BY

Major-General Sir W. G. MACPHERSON, K.C.M.G., C.B.

Major-General Sir W. P. HERRINGHAM, K.C.M.G., C.B.

Colonel T. R. ELLIOTT, C.B.E., D.S.O.

AND Lieutenant-Colonel A. BALFOUR, C.B., C.M.G.

LONDON :

PRINTED AND PUBLISHED BY

HIS MAJESTY'S STATIONERY OFFICE.

To be purchased through any Bookseller or directly from

H.M. STATIONERY OFFICE at the following addresses :

IMPERIAL HOUSE, KINGSWAY, LONDON, W.C.2, and

28, ABINGDON STREET, LONDON, S.W.I ;

37, PETER STREET, MANCHESTER ; ^1, ST. ANDREW'S CRESCENT, CARDIFF ;

23, FORTH STREET, EDINBURGH • or from EASON & SON, LTD., 40-41, LOWER SACKVILLE STREET, DUBLIN.

Price i\ Is. Od. Net.

CONTENTS.

CHAPTER PAGE

List of Contributors . . . . . . . . . . v

Preface . . . . . . . . . . . . vii

I. General Aspects of Disease during the War .. .. 1

II. Enteric Group of Fevers .. .. .. .. ..11

III. Dysentery 64

IV. Cholera 116

V. Typhus Fever 133

VI. Cerebro-Spinal Fever 147

VII. Influenza 174

VIII. Purulent Bronchitis and Broncho-Pneumonia . . . . 212

IX. Malaria : ^Etiology, Incidence and Distribution . . 227 X. Malaria (contd.) : Pathology, Symptoms, Diagnosis and

Treatment 264

XI. Blackwater Fever 294

XII. Trypanosomiasis . . . . . . . . . . . . 305

XIII. Relapsing Fever (Spirochaetosis) .. 316

XIV. East African Relapsing or Tick Fever 329

XV. Phlebotomus Fever 345

XVI. Trench Fever 358

XVII. Jaundice 374

XVIII. Scurvy . . 409

XIX. Beri-ben 430

XX. Famine Dropsy . . . . . . . . . . . . 450

XXI. Pellagra 470

XXII. Nephritis 485

XXIII. Cardie-Vascular Disorders 504

Index 539

iii (2306) Wt. 38692/4589/902 1,500 4/22 Harrow G. 51. «2

LIST OF COLOURED PLATES.

TO FACE PLATE PAGE

I. Cerebro-spinal Fever :

Erythematous Rash (Fig. 1)

Petechial Rash (Fig. 2) 152

1 1 Cerebro- spinal Fever :

Macular Rash . . .. .. ..154

III Cerebro- spinal Fever :

Purpuric Rash 156

IV Influenza:

The "Heliotrope Cyanosis" of I nfluenzo- pneu- monic Septicaemia . . . . 180

V. Influenza :

Whole Lung in a Case of Influenzal Pneumonia . . 198

VI Jaundice:

zht Lung from a Case of Spiroctaetal Jaundice (Ictero-hxmorrhagicu :*92

CONTRIBUTORS TO SUBJECTS IN VOLUME I.

Balfour, Andrew, C.B., C.M.G., M.D., B.Sc., F.R.C.P.E., D.P.H., Lieut.-Col. R.A.M.C.(T), Member of the Advisory Committee Eastern Mediterranean 1915-16; President of the Advisory Committee, Mesopotamia, 1916-17 ; Member Medical Mission, Expeditionary Force, East Africa, 1917.

Bradford, Sir John Rose, K.C.M.G., C.B., C.B.E., D.Sc., M.D.. F.R.C.P., F.R.S., Major-General A.M.S. (T), Consulting Physician B.E.F., France.

Byam, W., O.B.E., L.R.C.P., Brevet Lieut.-Col. R.A.M.C.

Dawson of Penn, Rt. Hon. Lord, G.C.V.O., K.C.M.G., C.B., B.Sc., M.D., F.R.C.P., Major- General A.M.S. (T). Consulting Physician, B.E.F., France.

Foster, Michael G., O.B.E., M.A., M.D., F.R.C.P., Colonel A.M.S.(T), Consulting Physician to Troops in France and Flanders.

French, Herbert, C.B.E., M.A., M.D., F.R.C.P., Lieut.-Col. R.A.M.C. (T), Consulting Physician Queen Alexandra Military Hospital, Millbank, London.

Gordon, Mervyn H., C.M.G., C.B.E., M.A., B.Sc., M.D., Lieut.-Col. R.A.M.C.(T), Member of Army Pathological Advisory Committee ; Consulting Bacteriologist for Cerebro-Spinal Fever, and Officer in Charge of the Central Cerebro-Spinal Fever Laboratory.

Hay, John, M.D., F.R.C.P., Lieut.-Col. R.A.M.C. (T.F.) ; Specialist Cardiac Disorders, Western Command.

Herringham, Sir Wilmot P., K.C.M.G., C.B., M.D., F.R.C.P., Major-General A.M.S.(T), Consulting Physician B.E.F., France, 1914-1919.

Hume, W. E., C.M.G., M.D., F.R.C.P., Colonel A.M.S.(T), Consulting Physician, B.E.F., France.

Hunter, Wm., C.B., M.D., F.R.C.P., Colonel A.M.S.(T), Consulting Physician Eastern Com- mand ; President Advisory Committee Eastern Mediterranean ; Officer in Charge British Sanitary Mission, Serbia, 1915.

Lelean, P. S., C.B., C.M.G., F.R.C.S., Brevet-Col. R.A.M.C., Professor of Military Hygiene R.A.M.C., A.D.M.S. (Sanitation) Egypt.

Manson-Bahr, P. H.. D.S.O., M.A., M.D., D.T.M. & H., M.R.C.P., Brevet-Major R.A.M.C.(T), Officer in Charge Malaria Diagnosis Stations and Military Laboratories Egyptian Expedi- tionary Force.

Relapsing Fever.

East African Relapsing

or Tick Fever. Phlebotomns Fever.

Purulent Bronchitis and

Broncho- Pneumonia. Nephritis.

Trench Fever. Jaundice.

Cerebro-Spinal Fever.

Influenza.

Purulent Bronchitis and Broncho-Pneumonia.

Cerebro-Spinal Fever.

Cardio- Vascular Dis- orders.

General A spects of Disease

during the War. Influenza. Trench Fever.

Cardio- Vascular Dis- orders.

Typhus Fever. Jaundice.

Pellagra.

Dysentery.

Malaria (Pathology, Symptoms, Diagnosis and Treatment).

CONTRIBUTORS TO SUBJECTS IN VOLUME I.

Cholera.

Blackwater Fever. Trypanosomiasis.

Famine Dropsy.

Mitchell. T. J., D.S.O.. M.D.. Major R.A.M.C.. D.A.D.M.S.. 15th Indian Division, Mcs. K

Newham. H. B., C.M.G.. M.D., M.R.C.R, D.P.H.. ol. R.A.M.C.(T). Consultant in Tropical East African Forces.

Nixon, J. A., C.M.G.. M.D.. F.R.C.R, Colonel M.S.(T), Consulting Physician B.E.F., France,

and Rhine Array. Robinson, O. L., C.B., C.M.G.. K.H.P., M.R.C.P.,

D.P.H., Colon-

Torrens, J. A., M.D , F.R.C P.. Major R.A.M.C.(T).

Waterston, J.. M.A.. D.Sc., Captain R.A.M.C.

. 0

Wenyon. C. M.. C.M.G., C.B.E.. B.Sc., M.B.. B.S., Colonel A.M.S.(T). In charge Malaria Investi- gations. Macedonia.

x. Sir \\ H., K.C.I.E., C.B., C.M.G., B.Sc., U.C.P., Colon. I), Consulting

Physician to Forces in Mesopotamia.

Cholera.

Enteric Group of Fevers.

Malaria (List of Mosquitoes).

Malaria (^Etiology, Incidence and Distribution).

Scurvy. Bert-beri.

Note. — (T) means temporary commission.

(T.F.) means Territorial Force commission.

PREFACE.

chapters of the volumes on the Diseases of the War have been prepared by officers who held regular, territorial force or temporary commissions in the Royal Army Medical Corps, and who had special knowledge and personal experience of the diseases about which they write. The material contained in official documents, supplemented by the numerous references appended to each chapter, has been at their disposal.

In the present volume a considerable amount of repetition will be found, notably in the chapters on influenza and purulent bronchitis, consequent upon these diseases having been considered from two separate standpoints, namely, the experience of the epidemic amongst the troops in France and the experience of the epidemic in the United Kingdom.

The second volume will contain chapters on nervous disorders, venereal and skin diseases, in addition to chapters on the medical aspects of aviation, gas warfare, and mine gas poisoning. Although these latter subjects have a wider significance than that of actual disease and might of themselves have formed a separate volume, it has been found convenient to introduce them into the volumes on the Diseases of the War.

The measures for preventing disease, and the methods and results of laboratory research are fully detailed, in the volumes which will be published separately on Hygiene and Pathology during the War, and reference must be made to them for fuller information on these subjects. But it has been considered advisable to introduce a certain amount of detail with regard to preventive treatment and pathology into the present volumes.

It has been preferable, in an historical record such as this, to adopt the form in which the subjects are now presented rather than the form which is customary in text- books or articles in journals. The chapters are based chiefly on such work as was done during the progress of the war. There has been little opportunity for further analysis and study of the accumulated records of medical cases. Consequently the final nature of the invalidism produced by the various diseases has not been described with that measure of accuracy which can only come when the

vii

viii MEDICAL HISTORY OF THE WAR

documents now in the hands of the Ministry of Pensions are analysed in detail. Moreover, even in respect of dealing with actual clinical experience, the contributors to the present volume have been handicapped by the fact that papers published during the war were comparatively few. This restriction of papers and consequently of clinical and path- ological studies was due to the general military policy which of necessity governed the publication even of medical reports.

The chief work of editing the chapters has been carried out by Major-General Sir Wilmot Herringham, Colonel T. R. Elliott, and Lieut. -Col. Andrew Balfour, who have devoted an immense amount of valuable time and care to doing so. They desire to acknowledge the able assistance which they received from Major T. J. Mitchell R.A.M.C., and the staff employed in the office of the Medical History of the War.

Acknowledgments are also due to the British Medical Journal, Journal of the Royal Army Medical Corps, Lancet, Quarterly Journal of Medicine, the Medical Society of London, the Royal Society of Medicine, the Medical Research Council, the Cambridge University Press, and Messrs. Bailliere, Tindall and Cox for permission to use blocks of various charts, illustrations and coloured plates, which have already appeared in their publications.

W. G. M.

CORRIGENDA.

(1) P. 12, Table I. France 1916: The correct

number of cases is 2568, not 2668 as printed.

(2) P. 56, Line 14 : The correct date is January,

1916, not January, 1915, as printed.

(3) Plates facing pp. 194 and 200 should be

marked Figures 1 and 2, instead of Plates III and IV as printed.

DISEASES OF THE WAR.

CHAPTER I.

GENERAL ASPECTS OF DISEASE DURING THE WAR.

DURING war popular attention finds its chief interest in the number of the wounded, and concerns itself much less with the amount of sickness amongst the troops, although in every war of which we have records from the days of Sennacherib onwards the inefficiency from disease has out- numbered many times the losses from killed and wounded. Medical science has advanced so much that the figures of wars fifty and sixty years ago afford no useful bases of comparison. Those of the last two great wars are as follows : —

Annual Ratio per 1,000.

	Wounded.	Sickness.
Admissions.	Deaths (incl. killed).	Admissions.	Deaths.
South Africa, 1899-1902 (31 months).	34-2	14-4	843-0	24-58
Manchuria, Russo- Japanese War, Japanese Force, 1904-1905 (18 months).	391-6	137-3	589-6	41-2

The ratios in the table below are the total sickness rates and are calculated in the same way as those of the South African and Russo-Japanese Wars.

Year.	France.	Italy.	Macedonia.	Egypt & Palestine.	Mesopo- tamia.	East Africa.
1915	„
1916	—	—	982-7	618-7	1409-7
1917	—	—	837-9	745-2	1301-3	1403-5
1918	533-1	670-8	1011-7	1000-1	980-9	2310-6

(2396)

2 MEDICAL HISTORY OF THE 'WAR

But these figures do not indicate the proportion of sick and wounded, and for comparing them the actual admissions in certain years are as follows : —

	Wounded.	Sickness.
	Total Admissions.	Total Deaths (incl. killed).	Total Admissions.	Total Deaths.
France, 1918	574,803	46,084	980,980	8,988
Egypt and Palestine, 1917-1918	32,255	9,451	359,855	3,360
Macedonia, 1917- 1918	12,552	2,843	331,753	3,031
Italy, 1918	4,671	470	54,626	661
Mesopotamia, 1916- 1918 (White troops only]	16,793	6,752	242,159	2,752

These figures show that the admissions for disease in other theatres than France were 14*6 times as numerous as those for wounds (988,393 : 66,271), while even in France, though the perfection of instruments of warfare and the constant fighting greatly increased the number of wounded, the admissions for disease were still much the more numerous. It was indeed anticipated that the disproportion would be even greater. The admission rates for sickness per 1,000 of strength in 1909, a year of peace, are given as : —

In the United Kingdom . . . . . . . . 378 • 4

In Egypt 672-9

In West Africa 1026-1

Further, in the Royal Army Medical Corps Training Manual published in 1911, it is stated that in wartime the excess of sickness admissions over those for injuries received in action will probably be as twenty-five to one, and that though the fatality of injury is greater than that of disease, the deaths from disease are usually five times the more numerous.

To the mere statement of numbers given above three other factors should be added before the effect of wastage from disease during the war can be realized, namely, the average number of days that patients remained in hospital, which in France was found to be 45*, the cost of transport and maintenance of the

* But this average refers only to those cases treated and discharged in

Many of the severer cases were transferred to England, and these

probably took much longer to convalesce. Thus cases of dysentery treated

GENERAL ASPECTS OF DISEASE 3

patient, and the cost of the training and transport of the man sent up to take a patient's place in the ranks.

These considerations are sufficient to show the importance of disease as a cause of inefficiency in an army, and the vast expense which it entails upon the country.

The natural circumstances of each country differ so widely that the prevalence of diseases varied much in the different theatres of the war, as is shown in the following table.

Rates per 1,000 of strength.

	France.	Italy.	Mace- donia.	Egypt & Palestine.	Mesopo- tamia.	East Africa.
Enteric —
1915 ..	3-1				__
1916..	2-3		6-3	14-2	54-4
1917 ..	•7		2-5	•7	14-2	4-76
1918..	•2	1-48	•8	•9	6-3	6-80
Dysentery —
1915..	•03	—	—	—	—
1916..	4-09	—	63-89	31-19	50.94
1917..	3-76	—	28-89	23-13	60-34	486-56
1918..	•79	9-54	58-23	21-80	51-12	116-51
Malaria —
1915..
1916..	•05		331-47	8-10	68-61
1917..	•48	—	353-18	44-66	94-20	2880-9
1918..	1-77	2-90	369-29	134-40	95-79	1278-0
Nephritis —
1915 ..	7-16
1916..	8-46
1917 ..	9-51
1918..	4-17	—	—	—	—	—

The enteric rate of admissions was nowhere over 10 per 1,000 of strength except in Egypt during 1916, and in Mesopotamia during 1916 and 1917. Dysentery was very prevalent in East Africa, Egypt and Mesopotamia. Malaria was exceed- ingly prevalent in East Africa, Macedonia and Egypt. Its great prevalence in Macedonia in 1916, and in Egypt in 1918, coincides with the advance into the infested valley of the Struma and plain of Esdraelon. This distribution was on the whole expected from previous experience.

Nephritis is not mentioned in other statistics than those of the forces in France, where it formed an appreciable item, nor was trench fever made a notifiable disease elsewhere, though it was

in France averaged 42-3 days under treatment, while those which were transferred to England averaged 118-3 days. The 45 days mentioned in the text is much less than the average number of days for all cases of illness contracted in France.

4 MEDICAL HISTORY OF THE WAR

seen in Macedonia after divisions had gone there from France.

Although a war carried on in many areas and climates cannot be closely compared as a whole with previous wars confined to one country, the admission rate for sickness in France in 1918 (533- 1 per 1,000 of strength) compares favourably with those of the South African War (843-0) and the Russo-Japanese War (589 • 6) . The climatic conditions in Manchuria were favourable to health, and Japanese sanitary methods as regards cleanliness of person and sanitary discipline were extremely good. Most of the Japanese sickness was due to beri-beri, from deficiency of vitamine in their ration, the chief constituent of which was polished rice. On the other hand, however, if there were in France, in the wet climate and in the conditions of trench war- fare, factors unfavourable to health, there were also in the absence of extreme temperatures and of endemic disease, in the shortness of the lines of communication and the consequent abundant supply of food, and in the facilities for sanitation and early treatment of illness, points which might be expected to tell heavily on the other side.

In other theatres of the war the sick rate was very much heavier. On the whole, a more favourable theatre than France could hardly be expected, and while an improvement on the Japanese figures may be regarded as eminently satisfactory, it is necessary to inquire whether in France the results could not have been better, and why in other countries they were much worse than the Japanese ratios. This is all the more necessary since of the epidemic diseases which are known to have been the scourges of previous campaigns — dysentery, malaria, enteric, smallpox and typhus — the last two have been practically absent amongst the British troops, and enteric was very much less prevalent than in any previous war.* Nor was there any disease, except malaria in certain theatres of war, which caused the same amount of inefficiency as beri-beri did in the Japanese armies.

The extraordinary improvement in the figures for enteric fever as compared with those in the South African War is remarkable. During the 31 months of the South African War, in which ration strength was probably never more than 250,000, there were 59,750 admissions for enteric, with 8,227 deaths. During 53 months of war in France, during which the ration

• In France, during 1914-1918, there were only eleven cases of smallpox, none of typhus. In Italy, in 1918, there were two cases of smallpox and none of typhus. The only theatres of war where there was any degree of prevalence of smallpox or typhus were Egypt, Palestine and Mesopotamia. Details regarding the former will be found in the volumes on the hygiene of the war, and regarding the latter in the chapter on typhus in this volume.

GENERAL ASPECTS OF DISEASE 5

strength rose from 269,711 in 1914 to 2,528,400 in 1918, the total number of admissions for enteric fever, including typhoid and the para-typhoids, as noted in Chapter II, was 6,907, and the number of deaths 260. Since enteric was prevalent among the civilians in the area which the British occupied during 1915, and since the French troops had a large number of cases up to the time at which they altered their system of prophylactic inoculation, it is fairly certain that the British troops would have been attacked but for the three measures specially designed to prevent it: the prophylactic inoculation, the strict water control, and the vigilant search for " typhoid carriers." The great improvement in the French figures which followed on the alteration of their system of inoculation is evidence that this measure played an important part. The small incidence of enteric was not confined to the expeditionary force in France. Except in Mesopotamia and in Egypt during 1917 the rate nowhere rose to double figures, and in every area except East Africa it sank lower year by year.

The same cannot be said of dysentery. In France the admis- sions for this disease did not reach any large total, but in other parts of the world, notably in East Africa, they rose to very high figures. These facts may indeed be used as evidence of the effect of inoculation as a preventive of enteric, for the channel of infection is the same in both diseases, the same sanitary precautions were taken for both, and in both the affected men were separated as quickly as possible from the healthy. But on the one hand the diagnosis of dysentery is more uncertain, so that segregation is more difficult, and on the other there is no prophylactic yet discovered for it. A lesson may be learnt, however, for the future. In France the diagnosis of dysentery was at first based upon bacteriological evidence alone. It was soon found that in a large number of cases the bacilli were not recovered and accordingly the presence of blood and slime in the motions were regarded as sufficient evidence for a diagnosis. But cases showing these symptoms in the trenches might show simple diarrhoea by the time they reached the casualty clearing station, and in that event, in spite of orders to the contrary, the diagnosis was not infrequently altered, although, as the sub- sequent course of the case in base hospitals showed, the original diagnosis of dysentery was correct. If wastage by dysentery is to be reduced in future, it is of the utmost importance to segregate all infected men at the earliest possible time, and it should be clearly understood that medical officers should strive not to minimize the number of cases or to refuse all but the most rigid proof, but rather to watch for and at once discover

6 MEDICAL HISTORY OF THE WAR

and segregate all cases which may fairly be suspected. An army will lose far fewer men eventually by adopting this procedure.

The figures for malaria in Macedonia, Egypt, and Mesopo- tamia were not much more satisfactory, while in East Africa they were so excessive that an official enquiry into the causes was instituted. Although there may have been failure on the part of individual administrators, the questions of interest in the present connection are the deficiencies in existing knowledge, the limits which circumstances must sometimes set to the application of such knowledge as exists, and the means, if any, whereby, in the future, methods of prevention may be increased and treatment improved. The life history of the infection of malaria is of course known, and the building of the Panama Canal is evidence of what can be accomplished in the prevention of insect-borne disease when conditions admit of the necessary measures. But measures such as would be taken in Panama cannot be carried out in actual warfare. If troops are pushed forward into infected areas, destruction of breeding places may in some kinds of country be quite impossible, and almost equally impossible may be the protection of the soldier in the open while on sentry duty or in advanced posts, and even perhaps in bivouac, tent or billet. It is generally allowed that quinine is of little use as a prophylactic in war time, and it must be recog- nised that the occupation of a malarious area will inevitably cause a high malarial sick-rate.

In considering the possibility of a long campaign in an area such as that of Macedonia, it must be realised that at present the medical services cannot control the outbreaks of malaria, which are bound to occur, and that to occupy a malarial district for long will be as serious a drain on the strength of an army as to hold a shell-swept front, such as that of the Ypres sector in France. Moreover, the price of malarial casualties continues to be paid for many years after the campaign itself. In the autumn of 1920 malaria was still responsible for 13 per cent, of the total number of men drawing pensions for disabilities due to diseases contracted in the war, and was indeed the chief source of all the chronic forms of disability.

The history of scurvy in Mesopotamia is interesting from several points of view. In the first place its incidence brought to light the fact that the ordinary peace diet of the Indian soldier, which was provided by himself out of a money allowance and not as a Government ration, lacked many essentials of a scientific dietary, so that many of the men who arrived in Mesopotamia were noticed from the first to be anaemic, debili- tated and below the proper level of health, and were liable to

GENERAL ASPECTS OF DISEASE 7

feel at once the slightest further deprivation which difficulties of communication might entail. In the second place, it is clear that the earlier war ration was not sufficient to overcome this tendency ; it, like the peace diet, had no surplus value available. Thirdly, the outbreak of the disease revealed that the remedies on which reliance had been placed, namely, dried vegetables and lime juice, were practically useless, whereas the really efficient substances, whether of old standing such as orange and lemon juice, or lately* discovered such as germinating pulses, were not available. Lastly, it is worthy of note that these commodities were eventually obtained, and also that by means of Arab and Indian gardeners a large amount of green vegetables was produced in the country.

But, after all, these infections count for little in the total sick-rate. It took a long time to realize that when the serious maladies were held in check it was time to attend to the minor diseases that made up the great total of wastage. In France a list of 21 diseases including all the eruptive fevers, together with the diseases just mentioned and some others, only accounted for 27-51 out of a total rate of 533-1 for 1918 ; in Mesopotamia in 1918 the dysentery and malaria rates amounted to 146-91 only, out of a total of 980'9 ; in Macedonia dysentery, malaria, and pneumonia with influenza made an aggregate rate of 538*85 out of a total rate of 1,011-7.

There are no official statistics as yet available to show what diseases constitute the remainder. A series of figures, however, was obtained from the casualty clearing stations of one of the armies in France during 1917, and was analysed by Colonel Soltau. The admissions numbered 106,267. As the total sick admissions for all the armies in France for 1917 are not known it is not possible to say what proportion Colonel Soltau's figures bear to the whole ; but they are little more than 10 per cent, of the total for 1918. Nor is it possible to compare his figures with the rates given above, since the strength of the army to which his figures refer is not available. But Colonel Soltau compares various diseases and classes of disease with one another under eight groups, and produces the following results :

Group " A ", which includes scabies, skin diseases, boils, and cases classed as inflammation of connective tissue, accounts for 26,879 of the admissions. " The main fact that emerges from a study of group ' A ' is that some 25 per cent, of the sick wastage was due to simple skin lesions, that of them the vast majority were due to scabies or some form of pyodermia, and as such were very largely preventible by careful inspection and personal

* But see note in Chapter XVIII. on Scurvy, page 420.

8 j MEDICAL HISTORY OF THE WAR

cleanliness of the men, and that even where infection was established, prompt treatment was efficacious in greatly reducing the loss of time."

Group " B," which includes pyrexia of uncertain origin, trench fever, myalgia and rheumatism, accounts for 26,024 admissions. Colonel Soltau considers that fully 20,000 of these were really trench fever, and, adding to this figure 1,500 of the cases of disordered action of the heart which is a frequent sequel of the fever, he ascribes 2 1,500 of his cases to trench fever, or, in other words, to infestation by lice.

From the two groups combined he concludes that 44 per cent, of the total admissions were due to diseases caused by dirt or lice and therefore preventible by sanitary measures.

Uncleanliness and verminous infection have consequently been brought into special prominence during the war as causes of sick wastage from this group of disease. In the South African War, diseases of the connective tissue and diseases of the skin together accounted for an admission rate of 46-83 out of the admission rate of 843 • 0 for all classes of sickness ; and, although these diseases may not in other areas and in other conditions rise to so large a proportion, yet they must in future be regarded as so powerful a cause of inefficiency that great efforts to prevent them are not only justifiable but necessary.

During the war four conditions, one hitherto undescribed by military surgeons and the other three barely mentioned, attracted much attention in France. Trench foot can be recognized in Larrey's notes of the winter campaign in East Prussia in 1806, and nephritis occurred to a considerable extent among the troops in the American Civil War, but trench fever is a form of disease which has escaped notice until now, and though gas gangrene had been occasionally seen in civil practice there is, according to Sir Anthony Bowlby, hardly any descrip- tion of it in military surgery. Upon all these a great deal of original and experimental work was expended, and if in nephritis no great advance has been made towards its prevention or cure, much has been gained in the other three cases. Trench foot was at once studied with the greatest care. Many experi- ments were made in various forms of boots and leggings, and eventually by the use of long loose thigh boots, by the strict application of prophylactic treatment to preserve proper circu- lation, and by improvements in the trenches, its incidence was greatly reduced. It still, however, in 1917 accounted for 3,294 of Colonel Soltau's admissions. Trench fever is an excellent instance of the practical value of research to an army in the field. It was recognized in 1915 and proved to be infective by inoculation of volunteers early in 1916. Had that method

GENERAL ASPECTS OF DISEASE 9

been pursued at the time, the pathology of the disease and the means by which it was spread would soon have been discovered, but the use of volunteers for the needful experiments at the time was not permitted, and accordingly these discoveries were postponed till 1917-1918, when, with the help of 60 or 70 volunteers, the American pathologists settled the question in three months. The delay probably meant that about 200,000 cases might have been prevented had the experiments taken place earlier. Another striking instance may be drawn from the surgical triumphs which immediately followed upon the knowledge gained in the pathological laboratories regarding the anaerobic infections which produced gas gangrene.

What, then, are the lessons which may be learnt from such figures, imperfect though they are, and how can the experience of the war teach us to lessen sickness and consequent wastage in future campaigns ?

In the first place, while the standard of sanitary discipline was excellent in such matters as water supply and disposal of excreta, the immense effect of uncleanliness in the production of disease must be recognized in future far more than it has been hitherto. Men are often crowded in dugouts and cellars, can only change their clothes or bathe at rare intervals, and are continually feeding in conditions which must convey infection if there is any infection to convey. The result is a sick rate from dirt diseases which amounts to nearly 50 per cent, of the total sickness in an army. That is sufficient to warrant the greatest possible effort to provide more baths, more laundries, more vermin destroyers, and to see that the men have opportunities of using them. Although measures to exterminate lice were energetically pursued from an early stage in the war, and the means of disinfestation constantly increased, it was not until the trench fever committee reported that the infection was carried by lice that the sanitary branch obtained the full equipment and facilities of which it had long been desirous.

Secondly, the war has shown the immense services which original research can render to preserve the efficiency of an army. The examples of trench fever, of cerebro-spinal fever, of gas poisoning, and of gas gangrene showed what wonderful results could be obtained by the union of clinical and pathological research not only at home, but also in the actual area of military operations. The mobile bacteriological laboratories were designed chiefly as aids to diagnosis and special treatment, but they went far beyond these limits and played a large part in the fresh discoveries of medicine and surgery. It cannot be doubted that in the future a prophylactic against dysentery

10 MEDICAL HISTORY OF THE WAR

will be discovered, and it can only be discovered by scientific experiment ; it is even possible that by the same method we may improve our means of preventing malaria.

Thirdly, the facts prove that in planning campaigns, especially in regions little known, the general staff should take the wastage by sickness into account as much as the wastage by wounds, and that not only should the army medical authorities be consulted concerning the probable loss from sickness and the consequent need for reinforcements, but their opinion should also be required concerning the best methods of prevention, including such local questions as sites for camps, destruction of insect carriers of disease, and purification of water, and also the wider and more general subjects of the provision of proper dietary, clothing and equipment. The medical side of the planning of a campaign is just as necessary for efficiency as the military, and the neglect of it must inevitably lead to an enormous amount of preventible wastage.

Lastly comes the great lesson of the war with regard to disease that, while to an army medical officer the fullest know- ledge of all that tends to prevent disease is of the utmost importance, the treatment of patients admitted to hospital for injuries or disease, in other words, the clinical medicine and surgery of war time, is not of necessity rough in method or imperfect in attainment, but is susceptible of a high and exquisite perfection and affords scope for the finest scientific work.

BIBLIOGRAPHY.

Bowlby . . . . The Hunterian Oration. British Mili- Lancet, 1919.

tary Surgery in the time of Hunter Vol. i, p. 285. and in the Great War.

Macpherson . . Russo-Japanese War. Medical and 1908.

Sanitary Reports. Report No. 15.

Simpson . . . . The Medical History of the South Jl. of R.A.M.C., African War. 1910. Vol.

xiv, p. 23, et. seq.

Soltau . . . . A Note on Sick Wastage . . . . Jl. of R.A.M.C.,

1920, Vol. xxxv, p. 152.

CHAPTER II.

ENTERIC GROUP OF FEVERS.

THE enteric group of fevers includes typhoid fever, due to infection with Bacillus typhosus of Eberth, and the paratyphoid fevers, due to infection with either Bacillus para- typhosus A or Bacillus paratyphosus B. The paratyphoid section may have to be enlarged to include at least one other variety, Bacillus paratyphosus C, which has strong claims to be considered as a definite and specific infection.

In the early months of the war there was confusion in the nomenclature of these diseases, owing to the lack of precision with which the term " enteric fever " was used by different medical officers ; by some it was considered synonymous with typhoid fever, by others it was only considered to imply a group infection.*

It was not until March 1915 that official sanction was granted by General Headquarters in France for the use of the diagnosis " Enteric Group " on clinical grounds, with the obligation to change it later to typhoid, paratyphoid A or paratyphoid B when the precise nature of the infection had been determined in the laboratory.

A certain number of cases, in which, for various reasons, accurate bacteriological or serological diagnosis cannot be made, retain the diagnosis " Enteric Group." In describing this group of diseases, as they occurred during the war, the term enteric fever comprises the group infection considered as a whole. The terms typhoid and paratyphoid A or B indicate specific infections by their respective bacilli.

Enteric fever has long been recognized as likely to be more deadly to an army on active service than the bullets of the enemy, and the truth of this is shown in the statistics from previous campaigns.

In the South African War the British Army employed 557,653 men, with an average strength of 209,404, and there were 59,750 cases of enteric fever, with 8,227 deaths. This is equivalent to an admission rate of 285, with a death rate of 36 per thousand of average strength.

* The nomenclature which army medical officers were required to follow was the official nomenclature of diseases drawn up by a joint committee appointed by the Royal College of Physicians. According to it enteric fever is a synonym for typhoid fever, and includes the sub-groups of paratyphoid A and B.

11

12

MEDICAL HISTORY OF THE WAR

In the Spanish-American War the American Army employed 107,973 men and there were 20,738 cases of enteric fever, with 1,580 deaths. This is equivalent to an admission rate of 88*5 per thousand for the year 1898. Sternberg, in " Sanitary Lessons of the War/' gives the annual death rate per thousand from typhoid fever at 14' 8 in the American camps in Florida and Virginia during this war.

In the Franco-German War the Germans despatched 1,146,000 men across the frontier ; these showed 73,393 cases of enteric fever, with 6,965 deaths.

In the French operations in Tunis, from a total strength of 20,000 men there were 4,200 cases of enteric fever, with 1,039 deaths.

In the Russo-Turkish War, the Russian Army of the Caucasus comprised 246,000 men and showed 24,475 cases of enteric fever, with 8,900 deaths.

No attempt was made in previous campaigns to differentiate typhoid from paratyphoid fevers, hence a considerable number of cases from which the foregoing statistics were compiled were probably paratyphoid. In view of the much lower mortality from paratyphoid fever it follows that the true percentage mortality of typhoid cases in previous wars has been higher than has been stated above.

With regard to the incidence of enteric fevers in 1914-1918, the official figures for the British Armies offer a welcome contrast to the experiences of previous campaigns.

The incidence of the enteric fevers in the expeditionary forces in the various theatres of war is shown in the following table : —

TABLE I.

Theatre of War.	Year.	Number of Cases.	Incidence per 1,000 of Ration Strength.	Number of Deaths.	Death Rate per 1,000 of Ration Strength.	Total Case Mortality per cent.	Mean Ration Strength of Force.
France	1914	388		47		12-1
	(Aug.-Dec.)
	1915	2,351	4-0	130	•22	5-5	588,000
	1916	2,668	2-0	30	•02	1-12	1,274,200
	1917	1,166	•61	33	•012	2-8	1,884,100
	1918	334	•12	20	•007	5-9	2,528,400
E. Africa . .	1917	102	5-0	27	1-3	26-4	20,600
	1918	116	7-8	33	2-2	28-4	(appro*.) 14,700
							(approx.)
Salonika	1916	1,105	11-79	40	•42	3-62	93,684
	1917	529	2-61	19	•09	3-78	202,260
	1918	135	•84	6	•03	4-44	159,947
Italy	1918	141	1-5	15	•15	10-6	94,000
Egypt (exclud- ing officers and Indian	1916 1917 1918	2,950 505 401	17-35 2-82 1-87	66 22 51	•32 •12 •23	2-2 4-3 12-7	170,000 179,000 213,000
troops)

ENTERIC GROUP OF FEVERS TABLE I. — cont:

13

			Incidence		Death	Total	Mean
Theatre of War.	Year.	Number of Cases.	per 1,000 of Ration Strength.	Number of Deaths.	Rate per 1,000 of Ration Strengfh.	Case Mortality per cent.	Ration Strength of Force.
Mesopotamia	1916	1,266	_	110	_	8-6
	(22 weeks
	only)
	1917	1,211	14-4	91	1-08	7-5	84,000
							(approx.)
	1918	640	6-0	70	•55	10-9	106,000
							(approx.)
Gallipoli	1915	4,241		uncertain		9-0	Not
						(approx.)	available.

No attempt has been made in the above table to differentiate between typhoid and paratyphoid fevers, and the totals do not represent all the cases of enteric fever that occurred in the British forces throughout the war, since cases occurring in Indian troops and native labour corps are not included, but it is believed that the figures are as accurate as can be determined at present for the periods and theatres of war concerned.

The total number of cases and deaths in this table shows that in upwards of four years and in six theatres of war, with an average mean ration strength of nearly two million troops, there were only 20,149 cases of typhoid and paratyphoid fever with 1,191 deaths, giving a total case mortality of 5*4 per cent.

The relative incidence of the three infections in the various theatres of war, as far as it was possible to identify them with certainty, is shown in the following tables : —

TABLE II.

Incidence of the Enteric Group of Diseases.

Theatre		Number of Cases.	Incidence per 1,000 of Ration Strength.
of War.	Year.	Typh.	Para. A	Para. B	En- teric Group	Typh.	Para. A	Para. B	En- teric Group
France	1914	253	5	31	99
	1915	805	281	1,043	222	1-3	•47	1.7	•3
	1916	729	580	1,009	350	•57	•45	•7	•27
	1917	227	173	471	295	•12	•08	•24	•15
	1918	90	43	156	45	•03	•015	•06	•015
Salonika . .	1916	97	212	203	593	1-03	2-26	2-16	6-32
	1917	81	136	92	220	•40	•67	•45	.1-08
	1918	30	47	20	38	•18	•29	•12	•23
Italy	1918	33	23	61	22	•35	•25	•6	•26
Egypt . .	1916	99	187	143	2,521	•58	1-1	•84	14-83
	1917	13	70	74	348	•07	•38	•41	1-94
	191S	31	66	46	258	•14	•3	•21	1-2

14 MEDICAL HISTORY OF THE WAR

TABLE III.

Incidence of Enteric Group of Diseases in Mesopotamia and

Gallipoli.

Theatre of War.	Year.	Typh.	Para. A	Para. B	Enteric Group.	Proved Cases.
Mesopotamia	July-Dec. 1916	12-3%	74-4%	13-2o/0	1,018	446
(quoted from Ledingham)	Jan. -June 1917	8-9%	77-2%	13-8%	239	101
	July-Dec. 1917 Jan.-June 1918	21-3% 36-6%	72-5% 50-4%	6-0% 12-8%	544 170	197 101
	July-Dec. 1918	37-7%	47-2%	14-90/c	209	127
Gallipoli (Based on a report by Martin and Upjohn)	1915	7-0%	61-0%	32-0%

The information afforded by these tables is not of equal value in all the theatres of war. The chief difficulty lies in the widely different proportion of cases which remain classified " enteric group."

Considering the great difficulties under which all bacterio- logical work laboured throughout the eastern campaigns, it is inevitable that the proportion of " group " to proved cases should be particularly high in these areas, while the figures for France are the most reliable owing to the low proportion of these undiagnosed cases.

It is nevertheless reasonably certain that the above tables represent with fair accuracy the relative incidence of typhoid and the two paratyphoid fevers.

It will be seen that only in France was typhoid fever responsible for as many as one half the total cases and that in the other campaigns the proportion was usually less than one quarter ; but that whereas in France and Italy paratyphoid B was about three times as frequent as paratyphoid A, in Salonika paratyphoid A was more frequent than paratyphoid B in the proportion of 32 to 27 ; in Egypt the proportion of para- typhoid A to paratyphoid B was as 9 to 7, while in Mesopotamia paratyphoid A was five times as frequent as paratyphoid B and three times as frequent as typhoid until 1918, when

ENTERIC GROUP OF FEVERS 4 15

there was a notable increase in typhoid and diminution in paratyphoid A.

It would not be fair to assume that the relative proportions of the three infections shown above obtained in previous cam- paigns because there can be no doubt that, taken as a whole, prophylactic inoculation has conferred a greater mass immunity against typhoid fever than against either of the paratyphoids. This must be so if only because triple vaccine was not introduced in any theatre of war until January 1916 and cannot have become efficiently established until the end of that year.

At the outbreak of war typhoid fever was endemic in every theatre. Paratyphoid B was very rare in England, but it was fairly common on the continent, especially in Flanders, Alsace, parts of Middle Europe and Macedonia, while paratyphoid A was practically unknown except in India, Africa, Asia Minor, Turkey and possibly a few seaport towns like Marseilles, where there is a constant interchange between Europe and Africa.

The natural sequence of events as regards the British forces in France would be that typhoid should develop within the first few weeks or months, accompanied or closely followed by paratyphoid B, both acquired locally from water or carriers, but that the advent of paratyphoid A should be delayed until contact had been established by our troops from England with men who had served in India or the East, or alternatively with French troops who had served in Africa or been associated with French colonial forces. This is precisely what occurred ; cases of typhoid fever developed in the latter part of September 1914, and were attracting serious attention by the second week in November,but it was not until December 5th that an undoubted case of paratyphoid B was detected, and the first proved case of paratyphoid A was admitted to hospital on December 14th ; further, it is noteworthy that the early cases of paratyphoid A were all in troops who had either come to France from India or who had been in close contact with such troops.

The steady relative increase in the number of paratyphoid cases in France, especially paratyphoid B — for paratyphoid A was always numerically insignificant — as well as the steady decline in all forms of enteric fever in the last quarter of 1915 is shown in Table IV, compiled from the admission and discharge books of No. 14 Stationary Hospital, which dealt with more than half the total number of cases from the entire force during the period under review. This table also shews that with properly organized laboratory work the proportion of cases in which final diagnosis is impossible is relatively small ; the percentage of cases under the heading " enteric group "

16

MEDICAL HISTORY OF THE WAR

I

*••*

'i

c/) Tf

> !2

3 fe

^ ^ PQ s

I I

<N CO n O5 m W W <N

I IX CO <N CO <N t> 1C O

O5 < O I>(N O O CO CO ^

I I

M W CO CC C<1 <N ~

Mil

O ID t^ O5 00 00 C<l N

| |

1 t^ 1 00

CO Tf CO rt Tf CO CN <N Tj< T-, *-,

CO CO Tf C^ CO C^ —«

ENTERIC GROUP OF FEVERS 17

drops steadily as the efficiency of the laboratory workers and the co-ordination between clinician and bacteriologist increase.

At the same time there will always be a small residuum, up to 5 per cent., in which the clinical picture is that of enteric fever but the bacteriological and serological findings do not support the diagnosis. This difficulty was apparent in 1915 and was increased considerably in later years by the adoption of triple inoculation ; the question will be considered again when the diagnosis of the enteric group is under discussion.

Just as paratyphoid A was conveyed to the western front from India and Africa, so was paratyphoid B conveyed to Mesopotamia by the divisions which proceeded thither from Europe and Egypt in 1916. Prior to the arrival of these troops the Mesopotamia force was composed exclusively of troops from India where, as in Mesopotamia, paratyphoid B was practically unknown ; so that enteric fever was re- stricted in 1915 and the early part of 1916 to typhoid and paratyphoid A. Boney, Grossman and Boulenger state that paratyphoid B was not diagnosed till March 1916, which coincides with the arrival of a British division from Gallipoli and Egypt. These authors find from an analysis of 650 cases after this date that the proportions were : typhoid 21 per cent., paratyphoid A 65 per cent., paratyphoid B 14 per cent., so that paratyphoid B obtained a firm foothold when once it had been introduced ; indeed, for 1918 the incidence per thousand of paratyphoid B, including Indian troops, is nearly twice that for 1917.

Figures dealing with the incidence of the enteric group in the Gallipoli expedition are not very reliable, owing to the nature of the campaign and the extreme difficulty of evacuating the sick, as well as the long distances between the fighting zone and the hospital bases. Coutts gives clinical notes of 66 cases of paratyphoid B and 63 cases of paratyphoid A ; Martin and Upjohn found paratyphoid A to be nearly twice as frequent as paratyphoid B. It is noteworthy, in connection with this campaign, to find that a considerable number of cases evacuated as dysentery were ultimately proved to be suffering also from paratyphoid fever, especially paratyphoid B.

In Salonika, paratyphoid A was more frequent than either paratyphoid B. or typhoid ; and paratyphoid B was more numerous than typhoid till 1918, when it became the least common of the three.

In Italy, in 1918, the relative proportions resembled those obtaining in France, except that paratyphoid A was rather higher ; the actual incidence of enteric fever per 1 ,000 of ration strength was, however, more than ten times as high.

(2396) B

18

MEDICAL HISTORY OF THE WAR

In Egypt the very large proportion of group cases in 1916 makes comparison difficult, but it appears that paratyphoid A was numerically preponderant, both paratyphoid A and para- typhoid B being higher than typhoid. In 1917 typhoid was seven times less common than either paratyphoid A or para- typhoid B but paratyphoid B was fractionally higher than paratyphoid A ; in 1918 typhoid was still the least numerous, but paratyphoid A was definitely higher than paratyphoid B.

Mortality.

The total case mortality per cent, for the three varieties of enteric fever grouped together can be determined fairly accu- rately, and, as has been shown in Table I, it varies considerably with the different campaigns but may be summarised as follows :

TABLE V.

Summary of case mortality from the enteric fevers in different- theatres of war.

France

Salonika

Egypt

Mesopotamia

Italy

East Africa

3-8 per cent.

3-9

6-4

8-7 10-6 27-4

The high death rate in East Africa may be explained by the extreme rigour of that campaign and the necessity for operating at a great distance from a properly equipped base in a very unhealthy climate. It is also probable that many mild cases of group infection were overlooked, and it is certain that many of the deaths were due rather to the presence of a coincident infection, such as malaria or relapsing fever, than to the enteric infections.

There is one point of special interest in the mortality columns in Table I., namely, the fact that the low water mark of per- centage case mortality was reached in 1916 and that a notable increase occurred in both 1917 and 1918. This increase was more or less apparent in every theatre of war where reliable figures are available, as follows : —

TABLE VI.

Showing increase in case mortality after 1916.

	1916.	1917.	1918.
France	1-12%	2-8%	5-9%
Salonika Egypt Mesopotamia East Africa	3-62% 2-2% 10-9%	3-78% 4-3% 7-5% 26-4%	4-44% 12-7% 10-9% 28-4%

ENTERIC GROUP OF FEVERS 19

It is necessary to enquire briefly into the possible reasons for this increase in case mortality. It will be remarked that the increase dates from the adoption of triple vaccine, so that it might be thought that triple vaccine to some extent decreases the immunity conferred against typhoid fever. If this were so, one would expect to see a definite increase in the case mortality from typhoid in protected men, and also to find that the increase is chiefly in typhoid as opposed to paratyphoid cases.

The increase in typhoid mortality is as follows :—

TABLE VII.

Case mortality from proved cases of typhoid. (Western Front.)

	Protected by Inoculation.	Unprotected by Inoculation.
1915 1916 1917 1918	7-54% 1-58% 7-73% 13-84%	23-2% 8-3% 12-12% 24-0%

A similar increase is thus shown in the figures for those who are unprotected, and it is obvious that there must be some other factor at work to explain the drop to 8-3 per cent, in the unprotected in 1916 with the subsequent rise to 24 per cent, in 1918. The difficulty of getting satisfactory re-inoculation at the end of 1917 and throughout 1918 would tend to produce a higher death rate among the partly protected.

The points which seem to be of great importance in this connection are, first, the undoubted lowering of all powers of resistance to infection in the nation as a whole and in the troops in particular by four years of continuous warfare, and, secondly, the increased average age and lower physical categories of an army which became to all intents and purposes a nation under arms.

In attempting to arrive at the individual case mortality of the three enteric infections, there is the difficulty, already men- tioned, of including the cases of the enteric group in which no final diagnosis has been possible. To ignore these cases might in some cases give unduly high results by eliminating a consider- able number of cases with a low death rate. It is probable that a fairly correct result will be obtained by assuming that the enteric group cases are made up of typhoid, paratyphoid A, and paratyphoid B in like proportion to the proved cases for the same area during the same period of time, and that the infections causing death in group cases are relatively propor-

B2

20 MEDICAL HISTORY OF THE WAR

tionate to those causing death in proved cases. This method gives the following results for France, Italy and Egypt : —

TABLE VIII.

Approximate percentage case mortality from typhoid, paratyphoid A and paratyphoid B.

Theatre of War.	Year.	Typhoid.	No. of Cases.	Para.	No. of Cases.	Para. B.	No. of Cases.
France	1914	13-5	340		7	4-9	41
	1915	12-0	889	•6	314	2-9	1,148
	1916	1-6	839	1-5	668	•7	1,161
	1917	6-5	304	•8	233	1-9	529
	1918	15-5	104	2-6	50	1-4	180
Italy	1918	22-3	39	3-7	28	6-4	72
Egypt	1916	3-7	659	2-03	1,287	1-6	1,004
	1917	Sufficient data" not available as no deaths were
		recorded in proved cases of paratyphoid.
	1918	37-0	87	6-3	179	7-8	135

The returns from Salonika and Mesopotamia do not permit of analysis on these lines.

For purposes of comparison the death rate from proved cases of the three infections is shown in the following table : —

TABLE IX.

Percentage case mortality from proved cases of typhoid and paratyphoid.

Theatre of War.	Year.	Typhoid.	No. of Cases.	Para. A.	No. of Cases.	Para. B.	No. of Cases.
France	1915	13-0	805	•71	281	1-91	1,043
	1916	1-9	729	1-7	580	•7	1,009
	1917	8-3	227	•56	173	2-1	471
	1918	16-5	90	2-3	43	•6	156
Italy	1918	24-2	33	4-2	23	6-5	61
Egypt	1916	6-06	99	3-2	187	2-08	143
	1917	23-0	13		70	—	74
	1918	41-0	31	6-06	66	8-6	46
Mesopotamia	1916	-|
	1917	>1 1 -4	320	3-6	532	7-5	120
	1918	[

The value of the figures in the foregoing tables depends largely on the totals of the cases, for when there are few cases the value

ENTERIC GROUP OF FEVERS 21

is slight ; but it is evident that the case mortality from all the enteric infections varies within wide limits from time to time in the same theatre of war, and also varies directly with the efficacy of the general hygiene, transport, and medical arrange- ments.

The proverbial severity of these infections in hot climates is noticeable in the figures from Egypt and Mesopotamia. The theory held by many who had worked in India that paratyphoid A was practically negligible as a cause of death appears to be fallacious when applied to active service conditions in the East, for there was in 1918 a case mortality of over six per cent, from this disease in the Egyptian forces, and in France the death rate for the same year was more than two per cent. This high death rate in Egypt was in part explained by an outbreak of malignant tertian malaria which complicated the enteric infections.

The relative mortality from paratyphoid A and B appears to vary greatly with time and place, as shown in Table IX., but here again the totals are often too small to be reliable, and the only safe deduction seems to be that they are both very much less severe infections than typhoid under like conditions.

The total figures available at present for proved cases from France, Italy, Egypt and Mesopotamia give a mortality table approximately as follows : —

TABLE X.

Typhoid 9-8% mortality in 2,472 cases.

Paratyphoid A 2-6% „ „ 2,023 ,.

Paratyphoid B 1-55% „ ,,3,160 „

Total Paratyphoid 2-1% „ 5,183 „

This is striking in one particular, namely, that the figure for paratyphoid A is considerably higher than that for paratyphoid B, a fact that is opposed to the general impression as gathered from the analyses of smaller series of cases made before the introduction of triple vaccine. Thus, in 1915, Torrens and Whittington found the mortality to be four per cent, for para- typhoid B, and less than one percent, for paratyphoid A, while Boidin in January 1916 reported a series of cases in the French Army with a mortality of six per cent, for paratyphoid B and 1-4 per cent, for paratyphoid A. Rathery in a large series of cases of paratyphoid B found a mortality of over six per cent.

A possible explanation of this difference in the mortality of the two infections is that the vaccine used from 1916-1918 conferred more protection against paratyphoid B than against

22 MEDICAL HISTORY OF THE WAR

paratyphoid A, a suggestion that is to some extent supported by the low titre to paratyphoid A, so often shown after triple inoculation ; or again the severity and frequency of paratyphoid A in tropical and sub-tropical climates may more than counter- balance the greater relative severity of paratyphoid B on the Western Front. Hence it may well be that in a civilian uninoculated population in Western Europe, an epidemic of paratyphoid B would be found to be attended with a higher death rate than would one of paratyphoid A, with a figure for either disease of from three to five per cent.

JEtiology.

With regard to the various factors affecting the aetiology of these diseases there is no reason to suppose that any which may be said to predispose to typhoid fever predispose also in any greater or less degree to either of the paratyphoid infections. The predisposing causes can be considered under two headings ; first, those of environment, which influence the presence and distribution of the infective material, and secondly, those of immunity, which influence the individual's capacity to neutralize a given dose of infective material.

Although a tropical or sub-tropical climate does not favour the growth of the bacilli of enteric fever outside the body, it nevertheless favours their distribution by flies and in dust, while the defective sanitary arrangements amongst the inhabi- tants of the East and Near East make enteric fever widely endemic in these regions. Before the war typhoid and para- typhoid A were very prevalent throughout the East, while paratyphoid B was practically unknown, so that in a sense it might be said that a tropical climate predisposes to typhoid and paratyphoid A rather than to paratyphoid B, and conversely that a cold or temperate climate predisposes to paratyphoid B rather than to paratyphoid A. Whether paratyphoid B will speedily die out in the East and paratyphoid A in the West, now that their respective sites of election have been enlarged, remains to be seen. Although epidemics may start at any time of the year, the summer and autumn are always likely to show the greatest number of cases and also the most severe ones. The effect of the external temperature is undoubted. Enteric fever is more frequent and more severe along the Mediterranean littoral than in the more northern parts of Europe ; for the same reason the disease persists in a serious form in Egypt, India, Central America, and the Philippines.

The number of bacilli present in subsoil water increases with the utmost rapidity as soon as men are occupying the surface of the soil. Vincent gives the follo\ving analysis from

ENTERIC GROUP OF FEVERS 23

a camp in which typhoid was constantly occurring. Before the arrival of the troops the water was very pure and contained only 100 ordinary bacteria per c.c. Six days later there were 770 bacteria, forty days later 6,960, sixty days later 14,900, and three months afterwards 38,000 per c.c.

In highly cultivated districts there is a great likelihood of the subsoil water becoming infected as a result of the practice of manuring the earth with human excrement. It is true that the typhoid bacillus does not survive in drinking water more than three to five days, but under suitable conditions the water is constantly being re-infected with fresh relays of virulent bacilli from a saturated soil. The importance of drinking water as a cause of enteric fever has been proved in numerous epidemics, and in war time in the field all drinking water should therefore be sterilized efficiently before use.

It has been shown that flies can carry typhoid and other pathological bacilli in their stomachs, on their feet, and on their probosces. Although the curve of enteric fever does not follow closely that of the fly pest, and the extent to which flies may be responsible for the spread of enteric fever is not fully established, these insects and the fingers of the " carrier " may, however, be regarded as playing the leading parts in causing the dissemination of typhoid infected material in war time. The specific bacilli are always likely to be present owing to the existence of some recent case in the neighbourhood or to the presence of a " carrier " among the population. A man sick- ening for enteric fever may be infectious for three or four weeks before he realizes he is ill. A " carrier " may convey infection for months or years after he has recovered from the disease and the bacilli may live in faeces or urine under favourable conditions of moisture for 100 days, and for upwards of 40 days in the absence of moisture.

Hence the most important factors predisposing to the occur- rence of enteric fever in war are the manifest impossibility of securing an absolutely perfect disposal of all faecal and urinary matter and the difficulty of excluding all " carriers " from an army. It has been shown experimentally that a large per- centage of men soil their fingers both during micturition and def aecation, especially the former ; and the contamination of food or water is more than likely to result.

Fletcher investigated bacteriologically one thousand men who were convalescent from enteric fever ; he found that prophy- lactic inoculation diminished the frequency of " carrier " development amongst infected men but did not abolish it, and that 0-6 per cent, of all convalescent male enteric cases are " carriers." Small epidemics have, in peace time, frequently

24

MEDICAL HISTORY OF THE WAR

been traced to cooks, waiters and others, who were " carriers/' and the same source of infection has been proved repeatedly during the war to explain a sudden crop of cases in the same unit when neighbouring units have been relatively or absolutely free.

With regard to individual immunity, there are numerous personal factors upon which immunity from enteric fever seems to depend. A previous attack confers a very great though not absolute immunity from re-infection with the same bacillus. It is estimated by Vincent and Muratet that not more than two per cent, of persons who have had typhoid fever can contract it a second time. But there is no experimental evidence that typhoid fever confers any immunity from paratyphoid fever or vice versa.

Prophylactic inoculation with triple vaccine confers rela- tively great immunity against typhoid and both forms of paratyphoid fever, the degree of immunity increasing up to a point with the number of injections employed.

Real immunity is only relative, but it appears that the Japanese and Chinese are not so susceptible as Europeans. Enteric fever, for example, has been stated to be less frequent in the Japanese than in the Russian Army in the Russo- Japanese War, and there was a similar experience in the Chinese expedition of 1901. The Hindu races appear to suffer but slightly from enteric fever in spite of their primitive hygienic and sanitary arrangements. It is held by some that the immunity of the Eastern races is apparent rather than real, as it is thought that the bulk of the population gets infected in childhood. This apparent relative immunity from enteric fever amongst the Asiatic races is borne out by the figures from our forces operating in Egypt and Mesopotamia.

TABLE XI.

Showing relative incidence in British and Indian Races.

	Incidence per 1,000 of Ration Strength.	Case mortality per cent.
Egypt :—	British.	Indian.	British.	Indian .
1916	17-35	1*15	2-2
1917	2-8	•9	4-3	7-6
1918	1-87	•5	10-4	9-5
Mesopotamia : — . .
1917	2-5	•4	10-8	22-3
1918	2-5	•8	6-4	18-3

The Indian figures for Mesopotamia, however, include a large number of followers, of whom only 20 per cent, were protected by inoculation in 1917 and 50 per cent, in 1918. Of the Indian

ENTERIC GROUP OF FEVERS

25

troops proper about 80 per cent, were protected in 1918 and 50 per cent, in 1917, and of the British troops 75 per cent, in each year. If we exclude the Indian followers, in order to obtain a better standard for comparison with British troops, we find that the mortality for Indians in 1917 was 22 per cent, and in 1918 12-8 per cent., with an incidence of 0-3 and 0-5 per 1,000 respectively. It thus appears that in the Indian races there is a real insusceptibility to acquiring enteric fever, but that there is a tendency for the infection when acquired to be exceptionally severe.

On the other hand, the high death rate in Indian troops can to some extent be discounted by the probability that many mild cases were never reported as enteric fever, but were allowed to run their course as pyrexia of uncertain origin.

If this apparent racial insusceptibility is due principally to immunity acquired as the result of disease in childhood, it would be expected that the incidence of paratyphoid B in Indian troops would be more nearly that obtaining in the British troops, at any rate in 1918 when the paratyphoid B infection, which was at first confined to the British troops who brought it with them to the country, had become more widely dis- seminated. This view is supported to some extent by the official figures for 1917 and 1918, dealing only with men unprotected by triple vaccine.

TABLE XII.

	British.	Indian.
Incidence of Paratyphoid B~] per 1,000 of ration strength ^ in unprotected men J	1917	3-92	•02 (one case only)
1918	•64	•4

The conclusion that enteric fever has run a graver course when it has attacked the Indian troops than when it has attacked the British is upheld by Ledingham, who published the following figures from Mesopotamia for 1916-17-18.

TABLE XIII.

Case Mortality in British and Indian Troops.

British.

Indian.

Typhoid Paratyphoid A	11-4 percent. 3-6	27-2 per cent. 11-3
Paratyphoid B	7-5	16-6
Enteric Group	10-0	20-7
Enteric Fever as a whole	8'7	20-5

26 MEDICAL HISTORY OF THE WAR

Age is recognized as playing an important part in the suscep- tibility to the enteric infections. No age is immune, but 46 -5 per cent, of all cases occur between the ages of fifteen and twenty-five 37ears. The statistics of the city of Paris for thirty years show that men are most frequently attacked between the ages of twenty and twenty-four years, while the liability to infection remains high up to thirty years of age. Further, between the ages of twenty and twenty-five, the death rate is nearly twice as high in men as in women, 67 • 1 per cent, to 37 '6 per cent. An army is therefore composed largely of those members of the community who are most liable to become infected with enteric fever in a severe form.

There are three other personal factors of great importance as predisposing in wartime both to a high incidence of, and to a heavy death-rate from enteric fever ; they are physical fatigue, mental strain, and the necessity for a more or less prolonged journey after the infection has begun to show its symptoms. No one who has worked amongst enteric fever patients can have failed to notice that those cases are most severe which have been longest delayed in transit to the enteric fever hospital.

As regards the exciting causes in the aetiology of enteric fever, the disease as at present understood includes infection by one of three specific micro-organisms and thus comprises three distinct though very similar diseases, namely : —

Typhoid Fever due to infection by Bacillus typhosus.

Paratyphoid A Fever due to infection by Bacillus para- typhosus A.

Paratyphoid B Fever due to infection by Bacillus para- typhosus B.

The specificity of these three micro-organisms has been proved beyond doubt by biochemical and serological tests. Bacillus typhosus was identified by Eberth in 1880-81, but it was not until 1896 that Achard and Bensaude gave the first account of a bacillus other than Bacillus typhosus recovered from the urine of a case of apparent enteric fever. This organism is now recognized as being Bacillus paratyphosus B. In 1898 Gwyn recorded a similar experience ; in his case the bacillus was recovered from the blood stream. In 1900 Gushing described an organism not Bacillus typlwsus, which he recovered from the pus of a chondro-costal abscess following an attack of apparent enteric fever. In 1900 and 1901 Schottmiiller described organisms which biochemically were intermediate between Bacillus typhosus and Bacillus coli and which did not

ENTERIC GROUP OF FEVERS 27

agglutinate with typhoid serum. In 1902 Buxton split the paratyphoid organisms into two groups A and B, A being closely allied to Bacillus typhosus and B to paracolon. In 1904 Firth described fully paratyphoid A as it occurred in British troops in India, work which was later amplified by Harvey, Grattan, Wood and other officers of the Royal Army Medical Corps.

In 1904 Bainbridge in the Milroy lectures differentiated clearly between the paratyphoid bacilli A and B on the one hand, and the organisms of food poisoning, Bacillus suipestifer, isolated in 1885 by Salmon and Theobald Smith, and Bacillus enleritidis on the other. A third member of the food poisoning group Bacillus aerlrycke was first described in 1898 by Durham and de Nobele, working independently ; this organism, though closely allied to Bacillus paratyphosus B, is nevertheless specifically distinct, as is shown by Perry and Tidy in their report on an epidemic of this nature published in 1918. Most bacteriologists now hold the view that Bacillus suipestifer and Bacillus aertrycke are identical.

Although we can thus dissociate completely from enteric fever a considerable group of infections by allied bacilli, there is nevertheless a distinct possibility that the legitimate para- typhoid group is not absolutely restricted to the two members A and B. Apart from blood infection with members of the food-poisoning group of organisms which, clinically, do not as a rule very closely resemble paratyphoid fever, there is a rare class of case which clinically is enteric fever but in which the agglutination curve of the patient's serum offers no corro- boration of the diagnosis. Occasionally in such cases a bacillus will be recovered from the blood, urine or faeces, which bacteri- ologically is not Bacillus typhosus, or paratyphosus A or B on the one hand, or a member of the food poisoning group on the other. This bacillus, however, agglutinates with the patient's own blood serum and is therefore almost certainly responsible for the infection concerned. Such bacilli are commonly reported by the bacteriologist to be culturally indistinguishable from Bacillus paratyphosus B. It is reasonable to regard such cases as being a variety of paratyphoid fever as yet unclassified. This view is corroborated by the experience of Mackie and Bowen, and MacAdam in Mesopotamia ; these workers, inde- pendently, while investigating cases of clinical enteric, isolated from a series of cases a bacillus culturally indistinguishable from Bacillus paratyphosus B which proved by agglutination and absorption tests to be an additional member of this series. A specific high titre serum was successfully prepared for this bacillus by Mackie and Bowen for the purpose of diagnosing

28 MEDICAL HISTORY OF THE WAR

other cases of the same infection. Ledingham regards this bacillus as an Eastern variant of Bacillus paratyphosus B but, in view of its persistent inagglutinability to ordinary para- typhoid B serum, it seems that the name Bacillus paratyphosus C, as suggested by Hirschfeld, would be justifiable. Ledingham states that he has lately received a strain of this organism from East Africa.

Similar cases have been reported from Macedonia, where Willcox found that 10 per cent, of the cases of clinical enteric were due to a non-agglutinable Bacillus paratyphosus B.

Archibald describes eight cases in Sudanese soldiers clinically resembling enteric fever, but proved by blood cultures to be due to organisms unidentified but definitely not typhoid or para- typhoid.

On the whole, it would be well to keep an open mind for the present on the question of the eventual enlargement of the true paratyphoid group of diseases.

Morbid Anatomy.

With regard to the morbid anatomy of the disease, the post- mortem appearances in cases of typhoid fever are too familiar to need description here, and all the lesions ordinarily described have found a place in the records of the fatal cases of the war. A great diversity of possible lesions is naturally to be expected in a disease like typhoid fever, which is essentially a baciUaemia at the time of onset of symptoms and often for the first two or three weeks of its course, as^well as during part of any relapses that may occur. Further the bacilli do not leave the system when they cease to be present in the blood stream, for in fatal cases they are always to be recovered after death from the gall bladder, nearly always from the spleen and bone marrow, usually from the mesenteric glands and frequently from the kidneys, the fauces, and the lungs if pneumonia has been a feature of the case.

The persistence of the bacilli in the body tissues is shown by the percentage of cases — about 2 per cent, in uninoculated persons — who remain either faecal or urinary carriers for months or years, and also by the fact that sub-periosteal and other abscesses occurring late in convalescence can often be shown to contain the specific organism.

Though fatal cases of typhoid as a rule show very marked intestinal lesions, yet the extent or severity of the utceration in the intestines is not necessarily an indication of the severity of the disease from the point of view of general systemic intoxi- cation. The following case illustrates this point : —

ENTERIC GROUP OF FEVERS 29

Rfm. H., age 22. (Not protected by inoculation.) — Admitted on tenth day of disease with a positive diagnosis of typhoid fever by blood culture. Clinically a very severe typhoid fever of toxic type, the rapidity of respirations being due to toxaemia rather than any local pulmonary condition. There was a plentiful crop of spots and moderate enlargement of the spleen, also a tendency to diarrhoea till the sixteenth day. The rate and character of the pulse indicated an unfavourable issue. The patient remained semi-conscious and delirious from the time of admission until he died seventeen days later on the twenty-seventh day of illness. The agglutination reaction to Bacillus typhosus was negative on the tenth day, positive on the fifteenth day and weakly positive on the eighteenth day.

At the post-mortem examination there was no trace of any ulceration of the intestines, nor were the mesenteric glands soft or swollen with the chocolate discoloration usual in typhoid fever. The liver was pale, soft and rather larger than normal, the spleen weighed 8 oz. and was soft and diffluent. The lungs showed capillary bronchitis at the bases. The heart was dilated and the myocardium showed fatty change. There was a row of recent soft, fleshy vegetations along the three aortic cusps indicating commencing ulcerative endocarditis ; Bacillus typhosus was recovered from the bile after death, but not from smears of the cardiac vegetations.

It is relatively rare for ulceration to be practically restricted to the large intestine in typhoid fever, though far from unusual in paratyphoid B.

The following notes illustrate such a case : —

Gr. C., age 22. (Inoculated January 13th and January 23rd 1915.) — Taken ill January 23rd, 1915. Admitted to hospital on sixteenth day of illness. Clinically a severe toxic case presenting no special features until the thirty-first day when there was a smart haemorrhage ; there was a smaller haemorrhage the next morning and a large one the same evening from which the patient never rallied. The bowels had been opened freely throughout the illness but there was no profuse diarrhoea at any time, nor was there tenesmus.

Post-mortem there were only six healing ulcers in the lower part of ileum ; the whole of the large gut from caecum to sigmoid, and especially the latter, was crowded with large ragged unhealthy looking ulcers, the general appearance being somewhat reminiscent of dysentery. Bacillus typhosus was cultivated from the gall bladder and from the spleen; no bacterio- logical evidence of dysentery was obtained, in spite of a most thorough investigation.

Prior to the war but little was known as to the differences, if any, in the morbid anatomy of the paratyphoid fevers as contrasted with typhoid. It has now been established that there is no essential difference ; any lesion that may be met with in typhoid may be encountered in either of the paratyphoids.

Since the gross mortality of paratyphoid is probably less than one quarter that of typhoid, it is obvious that the average lesion will be less intense in the former, but since only the very severe infections prove fatal it is natural that the post-mortem findings should approximate closely to those of typhoid. As a matter of practical experience they are indistinguishable. Dawson and Whittington, in an analysis of fourteen fatal cases of

30 MEDICAL HISTORY OF THE WAR

paratyphoid B and two of paratyphoid A summarized the cause of death as follows : —

Perforation . . . . . . 2 cases.

Peritonitis from infected appendix 2 cases.

Haemorrhage 2 cases.

Haemorrhage and toxaemia . . 3 cases.

Toxaemia 4 cases.

Pneumonia . . . . . . 2 cases.

Splenic abscess 1 case.

The same writers also noted the tendency for paratyphoid B to affect the large intestine as well as, or to the exclusion of, the ileum ; thus in two of their cases the large intestine alone was involved, in seven both small and large gut were affected, in four the small intestine only was concerned. In three cases of this series, two paratyphoid B and one paratyphoid A, the appendix was acutely inflamed and had determined the incidence of peritonitis ; in two cases, one paratyphoid B, one paratyphoid A, there was definite enteric ulceration in the appendix.

There is also a distinct tendency for metastatic pus formation in infections from Bacillus paratyphosus B ; thus in the fifteen cases mentioned above there were two spleen abscesses, two lung abscesses, one of which had caused a secondary empyema, and one abscess in the liver.

Since there are only two cases of paratyphoid A in this series, it is obvious that it is impossible to deduce very much as to the morbid anatomy of this disease. In a number of fatal cases of paratyphoid A, observed in Mesopotamia in 1916 by Torrens, the lesions were in the main identical with those of typhoid fever. Some predilection for the large intestine was noticeable, especially to the exclusion of the lymphoid tissue, but metastatic abscesses were not conspicuous. In some of the cases in which death occurred, rather from a complicating heat stroke than from the primary infection, the intestinal lesions were very trilling, sometimes amounting to no more than hyperaemia of Peyer's patches in the lower part ot the ileum ; occasionally even this was wanting.

Carles discussing a series of 170 cases of paratyphoid in the French Army, with eight deaths, confirms the frequency of the involvement of the large intestine, as also the tendency for abscess formation ; he also observes that there may be no intestinal lesion present even in fatal cases. MacAdam records a fatal case of paratyphoid B complicated by thrombosis of the upper end of the left internal carotid artery extending upwards into the middle cerebral artery and the lenticulo-optic

ENTERIC GROUP OF FEVERS 31

and lenticulo-striate branches. There was also thrombosis of the cortical branches of the right middle cerebral artery. No venous thrombosis could be made out in the brain or elsewhere, but the spleen showed two large haemorrhagic infarcts in which purulent softening had commenced.

Scott and Johnson describe a small brain abscess in the right optic thalamus, found post mortem in a case which developed left hemiplegia during the course of paratyphoid B infection ; unfortunately no attempt was made to recover Bacillus paratyphosus B from the abscess contents, so the possi- bility of a coincident infection cannot be absolutely excluded.

The great severity of the toxaemia as well as of the specific lesions in certain fatal cases of paratyphoid fever is shewn in a case of paratyphoid B published by Hichens and Boome. Clinically the patient presented all the features of advanced typhus fever including a maculo-petechial rash on the trunk. Death took place on the 14th day of the disease. Post mortem there was haemorrhagic infarction in the lungs with early grey hepatisation at the right base. The entire intestine, large and small, showed acute inflammatory change but no ulceration. The mesentery was inflamed, the mesenteric glands swollen and haemorrhagic, both kidneys were riddled with abscesses and the bladder showed acute purulent cystitis. The swollen spleen showed haemorrhagic areas on section. This man had had antityphoid inoculation in 1915 and two doses of triple vaccine in June 1917, three months before the onset of his fatal illness.

Symptoms.

As regards the clinical features of typhoid fever as seen in unprotected men in war time, these do not show any material differences from the clinical features noted in the many classical descriptions of this infection. The average of such cases was severe, very much more so than the average case seen in civil hospitals in England during the ten years preceding the war. The mortality was far higher and the graver complications were more frequent than in the civilian cases. This severity of infection is explained by the age and environment of the fighting man, the fatigue and hardship he is undergoing at the time of infection and the inevitable delay before he reaches the infectious diseases hospital.

On the other hand the average case of typhoid fever in a fully protected man is very much less serious, indeed it was difficult, if not impossible, in 1915 to judge clinically in certain cases whether the infection was typhoid modified by inoculation, or paratyphoid fever. In like manner during the later years of the war the clinical picture of the average paratyphoid case

32 MEDICAL HISTORY OF THE WAR

was itself modified by the use of triple vaccine, so that in certain cases there was practically no clinical indication that an enteric infection was present.

For the Western Front the figures show that typhoid fever, even in protected men, was decidedly more severe than para- typhoid, the case mortality being :—

1914 — Protected typhoid . . 5 -8 per cent.

Paratyphoid .. .. 2-0 „ „

1915— Protected typhoid .. 7 '5 „ „

Paratyphoid .. .. 1*6 ,,

Again in 1915, according to Willcox, the Gallipoli cases showed a paratyphoid mortality of not more than 5 per cent. There can be no doubt that the rate in protected typhoid cases was higher than this.

In a disease like enteric fever, which naturally varies in severity and duration within very wide limits, it is most difficult to state in precise terms the exact effect of a measure like prophylactic inoculation ; the general lessening of severity has been established and, as would be expected, analysis of individual cases tends to show that the average duration of fever is distinctly lessened in protected persons. In the cases observed by Torrens the average duration of fever in typhoid cases was five days less and in paratyphoid cases three days less in protected than in unprotected men.

There is no necessity to describe here the clinical manifes- tations of typhoid fever, but the following notes describe the paratyphoid infections and their differences from typhoid fever. It may be stated, however, at once that to distinguish clinically between paratyphoid A and paratyphoid B is impossible.

It is difficult in war time to establish the actual date of infection in any given case. General experience in the recent war has shown that, whereas the incubation period of typhoid fever is usually from 12 to 16 days, it may be much shorter or much longer in the paratyphoid infections. The shortest tune observed by Torrens was, apparently, five days and the longest twenty-eight. Most observers are agreed that the average incubation period for paratyphoid fever is less than for typhoid. Vincent gives it as from nine to fifteen days. Sacquepee states it may be reduced to five or six days, Lenglet from three to eight, while Miller considers from twelve to twenty days to be most usual. The length of incubation does not appear to be affected by prophylactic inoculation.

The onset of paratyphoid fever may be either gradual or sudden ; the gradual type, 20 per cent, of the cases, is rarely so gradual as in typhoid the fever usually being at its height by

ENTERIC GROUP OF FEVERS 33

the fourth day. The common early symptoms are general malaise, increasing headache, pains in back and legs and chilliness. The sudden type of onset, 60 per cent, of the cases, is commonly ushered in with fainting, vomiting, or a rigor. There is yet a third type of onset affecting 20 per cent., in which a period of trifling malaise, not sufficient to interfere with the performance of duty, and probably practically afebrile, terminates on the third or fourth day by sudden collapse with high fever and obvious illness.

As a general rule the cases with a sudden onset run a shorter course than those which develop gradually.

Fortescue-Brickdale has summarized the symptoms and early signs in 385 cases of paratyphoid B as follow :—

Headache . . .90 per cent. Generalized Pains . 25 per cent.

Diarrhoea Shivering Abdominal Pain Backache

45 ,, Vomiting

37 „ Cough .

32 „ Epistaxis

26 „ Vertigo

17 13 10 9

Sore Throat

Labial herpes is stated to be common.

The diarrhoea is not often persistent or severe ; it occurs early in the disease and is usually replaced by constipation after two or three days. Hence in war time the patient but rarely comes under observation while the diarrhoea is present ; when he does do so the stools have a putrid odour and the appearance and consistency of the ordinary typhoid fever stool.

The shivering does not often amount to a true rigor, though repeated rigors may occur just as in typhoid. Recurrent rigors appear to be more frequent in paratyphoid A than in either typhoid or paratyphoid B. Care must, of course, be taken to exclude a coincident malarial infection. Abdominal symptoms, apart from diarrhoea, are very much less conspicuous than in typhoid ; in upwards of 70 per cent, of cases there is no abdominal pain after the first two or three days ; quite often there is none throughout the whole disease.

Sweating is frequent and sometimes causes considerable exhaustion. Epistaxis, though only noted in 10 per cent, of the cases, is probably more frequent, but is often very slight and occurs so early in the disease as to be forgotten by the time the • history is taken. „

The average degree of toxicity is much less than in typhoid fever, therefore the typhoid state is the exception rather than the rule. Pronounced nervous symptoms may occur, but are relatively infrequent, confusional psychoses have been described, as also hemiplegia with sensory disturbance.

Meningismus of such degree as to simulate meningitis is far less common than in typhoid fever. Often the general

(2396) C

34 MEDICAL HISTORY OF THE WAR

appearance of the patient shows nothing more striking than a slight flush, some dilatation of the pupils and a general air of heaviness, even though the temperature may be 104° F. The tongue tends to be dry and coated, with dorsal slabs of fur, and red tip and edges ; this appearance depends largely on the diet and on the hygiene of the mouth. In very severe cases the tongue is dry, glazed and cracked, just as in typhoid.

The abdomen is often normal ; sometimes there is a certain sensation as of elasticity or tumidity on palpation. Caecal gurgling and tenderness are rare, but tenderness under the left ribs is fairly common.

The spleen is enlarged in more than 60 per cent, of cases ; it is palpable in nearly half of all the cases at some time during the illness. Quite often the spleen may not be felt until the third week or even later ; as a rule, however, the enlargement is apparent about the sixth day. Opinions differ as to whether the average splenic increase is so great as in typhoid. In the experience of Torrens the spleen of paratyphoid is harder than the spleen of typhoid, and for this reason it is easier to feel. The enlarged spleen is nearly always more or less tender, and sometimes there is perisplenitis with an audible friction rub. Fortescue-Brickdale noted a palpable spleen in 43 per cent, of his cases and the average weight in fatal cases was 6J ozs.

Chevrel states that the liver is almost always increased in size. Miller says the liver edge is occasionally lower than normal, and pain on deep pressure over the gall bladder is fairly common. In Torrens' experience definite enlargement of the liver is rare, as also real tenderness over the gall bladder.

The urine contains albumin in half the cases, apart from any co-existent bacilluria ; this, however, does not persist long and is of no special significance.

The respiratory tract is not conspicuously affected by para- typhoid fever ; cough is present at the outset if there be initial sore throat or laryngitis ; bronchitis and nasal congestion, usually mild, are fairly common during the first ten days, especially in soldiers who have been subjected to any consider- able journey after going sick. A considerable proportion of very severe and fatal cases, as would be expected, show pneu- monia of lobar or more commonly of lobular distribution ; the sputum in these sometimes contains paratyphoid bacilli.

Endocarditis and pericarditis, though recorded, must be very rare ; dilatation of the heart can but rarely be demonstrated by percussion and then only in the latest stages of severe cases. Shortening of the first sound, with some loss of intensity, is not infrequent during the second and subsequent weeks. The pulse is slow for the height of the temperature, relatively more so than

ENTERIC GROUP OF FEVERS

35

in typhoid, and noticeably soft often to the point of dicrotism. The blood pressure is low, 80-95 mm., and remains subnormal well into convalescence.

The temperature presents no very characteristic features. The rise may be abrupt or gradual ; the maximum is rarely more than 104° F. There is not the same tendency to plateau formation as in typhoid fever, and there is commonly a daily variation of nearly two degrees which produces a remittent or intermittent type of pyrexia. The duration of fever is very variable, from a few days to many weeks ; the average is difficult to state, probably about 20 days for both paratyphoid A and B. The termination is usually by lysis, but quite fre- quently by a form of modified crisis extending over about forty-eight hours. There is often a very marked disinclination for the temperature finally to settle down, even when convales- cence appears to be well established. Recrudescences are common and true relapses occur in about 10 per cent, of all cases.

The sub-normal temperature during convalescence, which is so common in typhoid fever, is not so marked in paratyphoid infections, though it is present in a considerable proportion of cases.

The following charts illustrate paratyphoid fever. Charts I. -VI. are from paratyphoid A. Charts VI I. -XI I. are from paratyphoid B. All these cases were proved bacteriologically, most of them by blood culture ; in none had triple vaccine been administered.

^I06«-

100°

£/

Resa

Chart I.

36

MEDICAL HISTORY OF THE WAR

Dauo/Dis

7

tt

9

1U

1 1

12

13

14-

15 ME

\t>

1 /

18

19

'^0

2]

x."^!

y;5

24

25

26

'<i7

v.y

2Q

'•50

21

'd'/d

33

Time

M I.

M E

M E

ME

.-1 f

M E

g|

ML

ML

Ht

ML

ML

M C

M E

M L

M [-

M«;

MF

n F

MF

•if

M F

M F

MF

MF

-i P

1Ofi*

fb t Q* *

t 104;°-

k , OT*

1*103 i-102*

q

^5 1 0 1 °-

\

A

1

A

A

</

\ /\

Q^j nn°.

V\

7

/

I

/

/

V^

^

ft 00'

V

r

1

(«"

•^

Normal^

\/

^

Jj

,<A

-f

98- 9T°

— *•

^X

vA

/'

/

-x

y\

/

v"

/|

7

P///CJD M

1

Vv

IF

Fes/) ft

"B3

-73-

-75-

^

^1

^

oo

104

1 16

IO6

96

fl4

ftri

»4

W

W1

100

-£r JioJ

w-

ss

•7*

^3- ±2j

70

6?

^1

"TO" 1 6

Chart II.

Afi^/J.

6

r

8

9

10

M

12

15

If

13

ts

1 7

18

19

5?0

2!

22

X2T

?.4

25

r/me.

ML

gj

^E

M t

n £

Mt

ME

M C

n E

Mf

n t

r,L

Mt

M t

M E

n L

M L

ME

ME

n t

mfl°

S>105.

J; f « *°.

^ 1 01?*

\

f|

8

^ .

l^s

^1

\*

A

A

S .

i

*"

^

fl

v

tt

^ 100 ft 99°.

\

Mmna/

-U

98°

lv

*

V

\

V^

yi

v*<

V-

Pu/re W Resp &

104

96

104

ioe

100

94

92

84

96

2e

/a

ea

7t>

6«

5^

56

bO

64-

bO

100

104 Ji4

100

IOO

94

9t ,

rfe

88

100

102

7b

72

«

tiO

•^

•^

S8

JsfL

if

Chart 111.

106

104-

102 10 1

JVormal 98'

M\ ze | so

100 ^B| 6S ao 72 100 Z2JJ

Chart IV.

ENTERIC GROUP OF FEVERS

37

20 1\ 11 23 24 i 5 26 \Z7 28 29 30 31 32 33 34- 35 36 37 38 39

Chart V.

|g ia \$_ 1 5 .1 6 IT" 18 J9 20 21 2?. 23 2^ ZS Z6_ 27 29 29. 3O

3I_ 32 55

106°-

105- >(»•• ^103°- 102° 101'- MOO'-

*•»•- Normal 96'

97'-

Pufsfi ffeso.

Chart VI.

9 MO

1HE

n£!±§

Li EL

ML

^106- '5 105°-

1 104°- I _o

100

99 Norms/ 98

97

7

Chart VII.

38

MEDICAL HISTORY OF THE WAR

^

tfff/S.

m

m

m

Mt

-IL

H t

L

EME

l£0i

Mt

1§

HE

106?

z

V. i Q2°- 101'

joo'

99'

AW^/

»^ii<o»4'i«r-i^^i»^orTu«^OBtif ^sfc->j»T.i^'i^^ji'^i»A'jii •B-t'i»i?j»i-iiiijf-r'j»^ii^iriJ

Chart Vlfl.

^/06' |,03'

?

iJ02'

\/

ft

99°.

ENTERIC GROUP OF FEVERS

39

7><)</o/3'S 15 14 13 16 17

7/me

102" 101-

ioo

98° 97°-

IE £3

Oa *3b 9fc>

8 19 2O 2 1 22

26 ?./r 28 29 I

tHE

31 32 3o o^»

H|MEF|r2_E

V?

Chart XL

Chart XII.

The rash in paratyphoid fever is present in about 60 per cent, of cases ; it does not conform so strictly to type as does the typhoid roseola. Miller describes the following varieties : —

(1) Rose-pink papules as in typhoid, occurring in successive

crops, and most evident on the lower part of the chest and abdomen.

(2) Larger spots of irregular outline, red with a bluish tinge,

raised, and not completely fading on pressure. This variety is characteristic of paratyphoid fever, when present ; the spots may be very profuse and have been mistaken occasionally for measles, German measles, varicella, and even smallpox.

(3) A rare variety of rash, which may occur alone or in.

association with the other types, consisting of cyanotic sub-cuticular patches of irregular shapes and sizes and indicating a severe infection.

40 MEDICAL HISTORY OF THE WAR

The spots vary in number from two or three to several hundreds ; they may be noticed any time from the end of the first week well into convalescence. Their first appearance may be delayed till the temperature has been normal for several days. An analysis of several hundred cases showed that the twelfth day is the most usual date for spots to appear. A feature of most paratyphoid cases, shared with a fair number of typhoid cases that have been inoculated, is to feel and look quite well about the twelfth day of illness even though the fever continues for another fortnight.

Convalescence in paratyphoid fever, even in quite uncom- plicated cases, is apt to be disappointing ; all goes well till the patient gets up and about ; thereafter progress is tedious. There is a great tendency for complaints to be made of persistent lassitude, headache, lack of appetite and insomnia. There is often considerable variation between morning and evening temperature and the latter may be slightly above normal. This is not an indication for further rest in bed ; these cases do better if encouraged to be out of bed and taking a reasonable amount of exercise. A small but definite proportion of patients manifest true cardiac dilatation during convalescence ; still more show the characteristic features of disordered action of the heart, praecordial pain, dyspnoea on exertion, tachycardia and palpitation, without any demonstrable lesion in valves or myocardium. In this last type of case there is usually vaso- motor instability, as shown by cold and livid hands and feet and tendency to perspiration without cause.

Definite neurasthenia is a not uncommon sequel to para- typhoid fever, but it is hard to say how much of this depends on previous war experiences and how much, if any, is directly attributable to the specific infection.

On the whole, convalescence from paratyphoid fever differs rather strikingly from that of typhoid fever, but principally in the subjective feelings of the patient, who does not manifest that sense of well-being and eagerness to be up and doing that is so often a feature of typhoid convalescence.

From the above brief clinical description it may be gathered that paratyphoid fever, whether A or B, is a miniature edition of typhoid fever so far as the average case is concerned ; it cannot, however, be too strongly emphasized that a severe case of paratyphoid fever is just as severe as the most serious case of typhoid, and that every complication or accident which may attend the latter may equally well be encountered in the former.

Serious complications are not so frequent in paratyphoid as in typhoid ; minor complications are not so serious when they

ENTERIC GROUP OF FEVERS 41

do occur. In a disease showing so many diverse clinical signs as paratyphoid, it is difficult to say where legitimate manifes- tations cease and complications begin. The preponderance in certain groups of cases of certain manifestations or complications has led some writers to attempt to classify paratyphoid fever into various clinical types ; thus Miller recognizes typhoid, dysenteric, biliary, rheumatic, respiratory, influenzal, and septic aemic types.

The typhoid type is by far the most common variety, and the foregoing remarks principally apply to it.

The dysenteric type, which is only admissible when co- existent dysentery has been rigidly excluded, is relatively infrequent, but is more common in paratyphoid B than in paratyphoid A. It is remarkable that it is not more often met with in severe cases, in view of the relative frequency of con- siderable large gut ulceration in paratyphoid B. Paratyphoid fever can, however, begin with symptoms that clinically closely resemble those of true dysentery, so that a certain amount of haemorrhage in quite the early days does not necessarily negative the diagnosis. At the same time, the great majority of this type of case was reported from the Eastern theatres of war, so that the possibility of double infections, especially paratyphoid grafted on to a bacillary dysentery, is difficult to exclude.

With regard to the biliary type there is great divergence of opinion as to the frequency with which infection of the bile passages and gall bladder, to the extent of causing signs or symptoms referable to these organs, may occur. Rathery comments on the rarity of jaundice or biliary symptoms in his series of 1088 cases of paratyphoid B. Torrens and Whittington state that jaundice and biliary symptoms were conspicuous by their absence on the Western Front in 1915. Torrens could not trace any special connection between the camp jaundice, which was common in Mesopotamia, and enteric fever. On the other hand, Dawson and Hume record twenty-four cases of infective jaundice attributable to enteric fever, namely, in typhoid, six cases ; in paratyphoid A, four cases ; and in paratyphoid B, fourteen cases.

It is probable that the paratyphoid fevers of the Gallipoli campaign were accompanied by jaundice and biliary symptoms in larger proportion than the same fevers in other areas. Morley and Battinson Smith record a case of " epidemic jaundice " which showed acute gangrenous cholecystitis ; Bacillus para- typhosus B was recovered from the stools and bile of this patient. Sarrailhe and Clunet recovered an inagglutinable paratyphoid bacillus from the blood of a number of cases of

42 MEDICAL HISTORY OF THE WAR

camp jaundice in Gallipoli ; subsequent investigation showed these organisms to be, for the most part, paratyphoid A.

Acute cholecystitis is met with from time to time, usually after the third week.

In the rheumatic type, acute articular rheumatism has been noted in a few cases of paratyphoid fever. Arthralgia and myalgia, without objective evidence, occur in close on 10 per cent, of cases. Nobecourt and Peyre consider articular rheu- matism to be a common manifestation, especially of paratyphoid B. Synovitis simulating infective arthritis and giving rise to suspicion of gonorrhoea was noted by Miller in several cases.

In the respiratory type, rapidity of respiration may be due simply to toxaemia, but some bronchitis is commonly present. Lobar and broncho-pneumonia are seen in cases either at the onset or at any time during the course ; in only a small pro- portion of these cases can the paratyphoid organism be recovered from the sputum, or from the lungs after death. Pleurisy is far from uncommon in paratyphoid fever ; often a little dry pleurisy is noted for a few days, and clears up completely. Sometimes an effusion develops very rapidly ; this may be lymphocytic in nature, and suggests a tuberculous process. On the other hand, a polynuclear effusion which rapidly goes on to empyema is not unlikely, especially in paratyphoid B. It is rare to recover paratyphoid bacilli from the simple pleural effusions, but they have been found in some of the empyema cases according to Weeks and others.

It has been suggested that paratyphoid infections may light up a latent tuberculosis ; certainly lymphocytic effusions with transient signs at the apices suggesting tuberculosis have been described, but more evidence is wanted on this point. Jol train and Petitjean noted 19 cases of pleurisy in 310 cases of para- typhoid fever; 18 were due to Bacillus paratyphosus B and 1 to Bacillus paratyphosus A, 15 were sero-fibrinous, 2 were purulent and 2 were dry.

In the influenzal type, paratyphoid fever can simulate closely the respiratory, the gastro-intestinal, or the nervous forms of influenza. This is especially the case in protected men. Isolated examples of these varieties are very likely to escape detection in a busy general hospital, since it is impossible to keep all such mild cases under observation sufficiently long to exclude enteric infections by serological tests.

The septicaemic type is rare. In it death occurs early in the disease ; often there are no local lesions found post mortem, nothing but the general features of septicaemia. Job and Ballet record three such cases and Sawasaki has met with similar ones in Japan. Gangrene of the extremities may precede death.

ENTERIC GROUP OF FEVERS

43

Some of the complications of paratyphoid fever have been sufficiently discussed in the foregoing clinical description. There remains a large number of which only three require special notice here. Haemorrhage occurs in less than 5 per cent, of all cases ; there seems little doubt there is a greater tendency to haemorrhage in paratyphoid B than in para- typhoid A. Perforation is definitely less common than in typhoid fever, but appears to be somewhat more frequent in paratyphoid B than in paratyphoid A. Nearly all published figures show that more than one quarter of the deaths are due to haemorrhage or perforation. Webb Johnson gives the incidence and mutual relationship of perforation and haemorrhage of a series of cases in France.

TABLE XIV.

				Haemorrhage
	Number of	Haemorrhage.	Perforation.	and
	cases.			Perforation.
Typhoid	1,118	50	9	3
Paratyphoid A . .	344	1	2	0
Paratyphoid B . .	1,038	16	3	1

Apart from dilatation of the heart and the symptoms of simple disordered action, a small proportion of cases give evidence of more definite damage to the heart muscle. For example, heart block, auricular flutter and auricular fibrillation may all occur. The lesions giving rise to these phenomena may be transient or permanent, and it is important from the patient's point of view that the clinician should be alive to these possibilities in order that appropriate treatment may be instituted as early as possible. The fact that typhoid patients may die quite suddenly when apparently doing well, almost at any period of the disease, has long been recognized ; the same mode of death is observed, but less frequently, in paratyphoid fever. It seems possible that the actual cause of death in these cases may be the sudden development of ventricular fibrillation.

The incidence of the remaining complications of paratyphoid, compared with the same in typhoid, is shown in Table XV., taken from Webb Johnson's report. The figures are based on the analysis of 2,500 cases of enteric fever treated in hospital at Wimereux, and it must be borne in mind that, just as the case mortality was far higher in other theatres of war, higher also, without doubt, was the incidence of the in dividual complications. The table, however, is of interest since it deals with a large number of cases all treated under practically uniform conditions,

44

MEDICAL HISTORY OF THE WAR

rt 2-3

GO CD ^"t CM 05 01 O

CD 10 o I-* o i-< co

Tt'Tf'OCOICMCNCDOOaCGCOlCDOOOTl'aecOO

oo-<ti'-'C<;ioco;oocDt>.'-ioic^c^ coc^

(MOOOOCO

OOOO^OOOOOOr-i

II il II II II II II II II II II II il II II II

666 l^666666cM^6666^6 ! II I! II II II il II II II II II II II II il II II i|

.«<£

1 O CM

CO

<§

Paratyphoid B inoculated.

p-y

£8 g£

oJ O

cu a

Paratyphoid inoculated.

II

A o

i!

a, 3

-xO

0^

i-* us *+ •**

-^C<I -^f

o - 1 1 16 Mo 1

!l II II II

10 CN

-vP"vP

o"^o^ iO O)

O "O 1C O «O

6^6 1 loo! I 6 I 6

li i| II II il II il

i ! 1 ! ! 1 -1 1 6 i i 1 i I 1 i i 1 II M 1

S

CO o^o •- TgcB

CO CD CO CO CO O O

16 61 |6^"l6'^(NCM|c0^i6l6^H6'-^

II II il II II II II II II II II II II II II

O5

CD ^c

r^ t>> o

lO O5 1C O5 CO CM i-<

G>

CM CM •* lO CD

T-t r-< CM 00 CO

66 166 166

ES

II II

oco-*1 i-i ft

i— >-<<N r>co

IT> r-4f*G*

*%

t

^^ ^ eg rt "73

iil|iilliiii4i«

|QP«MtJ>

Cases Deaths Mortality

ENTERIC GROUP OF FEVERS 45

and it shows the effect of prophylactic inoculation alike on typhoid and paratyphoid fever, as regards not only general severity but also incidence of complications.

When a number of men are exposed at the same time to the risk of infection by three specific micro-organisms, no one of which has the power of conferring immunity against the others, it is certain that mixed infection with two or all of the infecting agents will occur in a proportion which can be expressed mathematically.

A number of such cases have been recognized, but it is inevitable that many should escape diagnosis, since further work would not be undertaken as a routine in any case so soon as the presence of one infection had been established. It is probable that certain of the cases of anomalous course or of unduly prolonged duration, as well as those showing unexpected ag- glutination curves in fully protected men can best be explained on the hypothesis of mixed infections. An interesting case of mixed infection has been reported by Dawson and Whittington as follows : —

The patient had a double infection by the Bacillus paratyphosus A and the Bacillus typhosus. He had thrombosis of the left femoral and left external iliac veins. Four relapses occurred. In the last relapse he had pulmonary infarction, and death was due to the subsequent severe lung affection on the 127th day from the onset.

The following points are noteworthy : The patient had had no pro- tective inoculations. Admitted on the twelfth day of the illness, he appeared to be typical of a rather severe enteric group infection, and his blood gave a pure culture of Bacillus paratyphosus A. The serum on this day and on the eighteenth day strongly agglutinated the stock paratyphoid A bacillus, and gave no reaction with Bacillus typhosus or Bacillus paratyphosus B. . By the twenty-second day the patient was obviously improving and during this time he had a swinging temperature (rather characteristic of paratyphoid A infection) from 99° to 102°. On the twenty-third day, however, the temperature range became steadier, remaining between 102° and 104° for five days. On the twenty-fourth day the serum agglutinated Bacillus typhosus as well as Bacillus para- typhosus A . It gave the same reaction on the twenty-ninth day, but the reaction with Bacillus paratyphosus A had much diminished. On the twenty-seventh day thrombosis of the left femoral vein was first noted. The duration of this primary attack of fever lasted forty-eight days.

The patient had four relapses with four, twenty, sixteen and ten days' pyrexia respectively. During the second relapse he was given two injections of paratyphoid A vaccine without obvious effect. In the middle of the third relapse a blood culture was negative. At the post- mortem a pure culture of Bacillus typhosus was grown from every viscus examined (gall-bladder, spleen, mesenteric gland and thrombosed vein), thus proving the presence of a second infection.

The date of the second bacillary invasion is not quite clear. The agglutination reactions suggest that it was before the twenty-fourth day, but not much before the eighteenth day ; also the temperature range altered on the twenty-third day. Thus it seems likely that when the patient came to hospital he had reached the twelfth day of a paratyphoid A attack and was in the midst of the incubation period of typhoid, that for a while the two infections reigned together, and later the para- typhoid A disappeared, leaving the typhoid to reign alone. The relapses were thus probably due to Bacillus typhosus.

46 MEDICAL HISTORY OF THE WAR

The increased severity of enteric fever in the East and Near East is in part explained by the greater frequency of its asso- ciation with malaria or dysentery, as well as the liability to hyperpyrexia or even true heat-stroke. Latent malaria may be lighted up, often in a virulent form, by an enteric infection, while the extra strain of even a mild paratyphoid infection may determine a fatal issue in a case of dysentery of only moderate severity. A certain number of paratyphoid A cases developed heat-stroke in Mesopotamia in the hot weather of 1916 ; the majority of these proved fatal, sometimes during the first week of illness. At this time ice was not available.

A number of cases of combined infection with typhoid and diphtheria was noted by the French authorities. The mortality in these was very high. The severity of this double infection was confirmed by experience of a small number of similar cases in the British forces.

Prognosis.

The prognosis in enteric fever has been shown to vary with the specific infection, with the amount of time that elapses between " going sick " and reception into a hospital for per- manent treatment, with the climate, with the rigours of active service to which the individual has recently been exposed, and also with the presence or absence of protective inoculation against the particular infection which has been acquired. These factors have already been discussed ; but there are certain clinical features which may lead the clinician to regard any given case as likely to do well or badly and also indicate the average duration of " invalidism." It is important to estimate the proportion of cases likely to be unfit for further military service and the probable incidence of symptoms sufficiently serious to justify a more or less prolonged pension.

With regard to the clinical features bearing on prognosis, in all enteric infections the most reliable guide as to the patient's actual state of well-being is the pulse. The quality of the pulse is significant ; a pulse so soft as to be " dicrotic " betokens a relatively intense infection, but apart from the quality the all- important factor is the actual pulse rate. So long as the pulse rate is no more than 100 per minute the patient's condition is not likely to be very urgent; a pulse rate of 110 is serious and when the rate reaches 120 the prognosis becomes extremely grave. In adult male patients a pulse rate of more than 120 per minute continued for longer than 36 to 48 hours means death in all but a few very exceptional cases. The intensity of toxaemia, as shown clinically by the dry skin, flushed face and mental lethargy, has an obvious bearing on prognosis, but the

ENTERIC GROUP OF FEVERS 47

importance varies with the nature of the infection and period of disease to which the toxaemia persists. In paratyphoid fever manifest toxaemia persisting after the twelfth day indicates a severe infection ; for typhoid fever the same degree of toxaemia might be expected till the twentieth day.

Spots are most frequently seen in the more severe infections ; but Torrens considers that, granted a severe infection, a plentiful crop of spots is of favourable import and that such a case is likely to do better than a similar case in which spots are scanty or absent.

The degree of splenic enlargement does not seem to be of special import, except in so far as a big spleen usually indicates an infection of at least moderate severity.

A high temperature, apart from hyperpyrexia which is always serious and particularly likely to occur in the tropics, is not a sign of danger unless it is associated with a rapid pulse, when the prognosis should be based on the pulse rate rather than on the degree of pyrexia.

The chief risk of a relapse is that it prolongs the period in which perforation and haemorrhage may occur.

The complications of enteric fever, with the exception of pneumonia, haemorrhage and perforation, influence prognosis principally as regards the probable length of invalidism required before any work can be undertaken. Pneumonia, haemorrhage and perforation, however, are complications rather apart from all the rest ; they may all three, but especially haemorrhage and perforation, occur without any warning in the course of a case which has to all seeming been quite a mild infection. The advent of any one is of very grave prognostic import, but perforation is infinitely the most serious, since it is probable that not more than one in fifteen can be saved under active service conditions.

In a series of seventeen perforations observed, only one sur- vived, although practically all were operated upon within a very few hours of the complication occurring. In another series perforation was responsible for 14 out of 103 deaths in 2,500 cases according to Webb Johnson.

Haemorrhage is probably responsible for one-fifth of all deaths from enteric fever.

Invalidism.

The following table indicates the average duration of invalid- ism. It is based upon 2,000 cases treated in Addington Park Hospital and shows the number of days' treatment necessary for cases of enteric fever from the different theatres of war. It

48

MEDICAL HISTORY OF THE WAR

is noteworthy that the length of treatment appears to vary directly with the distance from England of the country where the infection was contracted.

TABLE XVI.

Duration of Treatment of Enteric Fever.

Force from which derived.	No. of cases.	Average number of days under Treatment.
France Gallipoli Egypt Salonika Malta East Africa Mesopotamia India Miscellaneous Cases	1122 143 206 192 117 10 11 3 196	102-92 140-59 151-36 152-69 156-44 208-30 234-00 227-00 126-52
Total number of Cases. 2000	Total number of days under treatment. 244,520	Average number of days under Treatment. 122-26

The average length of treatment is seen to be 122*26 days so that it is reasonable to suppose that an ordinary case is fit to resume duty six months after the date of infection. A further two to three months may have to be added to this period for those patients who were infected in the East. A small per- centage of all cases become carriers and therefore useless for further military service. In the unprotected this proportion is fully 2 per cent. In those protected by triple vaccine it is pro- bable that the proportion is much lower. The percentage of typhoid carriers is higher than that of paratyphoid, while that of paratyphoid B is higher than that of paratyphoid A.

The other principal reasons for discharge from military service after enteric fever are complications or sequelae affecting the cardio-vascular system, and neurasthenia.

Phlebitis and thrombosis occur in not more than 4 per cent, of all cases ; a small but definite proportion of these cases are left with permanent oedema of the limb and are unfit for further military service.

Disordered action of the heart is a more frequent reason for discharge, since symptoms may persist to the extent of pre- cluding any but a sedentary occupation for several years, in spite of careful treatment by graduated exercises. Such cases, however, should not be discharged for at least a year, since a large proportion will recover under suitable conditions.

ENTERIC GROUP OF FEVERS 49

Those few cases which manifest a more definite cardiac lesion, such as heart block, auricular flutter, or auricular fibrillation are probably unsuited for further military service.

The number of soldiers now receiving pensions for disabilities which are directly attributable to enteric fever is not great, either absolutely or relatively. The only cases of this sort seen by Torrens during 1919 may be grouped under the headings general debility, disordered action of the heart, other cardiac conditions, effects of thrombosis and affections of the gall bladder. In all these, with the exception of the first, a pensionable disability may persist for many years.

Statistics are not at present available to show the exact percentage of enteric cases who were discharged from the army or who are now drawing pensions.

Diagnosis.

The diagnosis of enteric fever depends upon its clinical manifestations and laboratory investigations. With regard to the former it is established that clinical signs can take one no further than a diagnosis of enteric fever ; the attempt to say that a given case is either typhoid or paratyphoid fever can only be a guess, since typhoid can be as mild as paratyphoid, and paratyphoid can be as severe as the worst case of typhoid. This statement applies alike to protected and unprotected persons, the only difference being that the experienced observer is more likely to guess correctly in the latter case than in the former.

Any case presenting several of the characteristic enteric features — headache, continued fever, slow pulse, diarrhoea, tumid belly, spots, enlarged spleen and mental lethargy — must at once be referred to the laboratory for more precise diagnosis ; but these are not the important cases, as they would justify a clinical diagnosis anywhere, and there is no risk of their failing to be isolated for an adequate period. The important cases are those which are so mild and atypical that, clinically, they do not suggest an enteric infection, for these may well dissemi- nate infective material should they be returned to duty while in a " carrier " condition. In the majority of cases there will be one or two isolated signs or symptoms that may put the wary observer on the track : such as, the quality of the pulse, a suggestion of undue lethargy, a history of looseness of the bowels or epistaxis at the onset of the illness, an increase in the area of splenic dullness, or a doubtful spot or two about the shoulders or abdomen.

It is well to remember that in the tropics malaria is more often confounded with typhoid than with any other disease.

(2396) D

50 MEDICAL HISTORY OF THE WAR

In view of the large number of cases which are not enteric, and in which the diagnosis of pyrexia of uncertain origin can never be replaced by one more scientific, as well as a host of trench fever and influenza cases, it is obvious that the ideal method of treating every case of unexplained fever of six days' duration as suspected enteric group is not practicable. Actually then the onus of diagnosis rests on the clinician, who must appreciate that any case of unexplained fever may be enteric, and who must be unceasingly alert to distinguish those lesser signs which may lead him to seek the aid of his bacteriological colleague only in those cases which will yield a reasonable proportion of positive results.

The atypical forms only of influenza or trench fever are likely to give rise to doubt and may be clinically indistinguishable from the modified varieties of enteric fever. An enumeration of leucocytes may serve to eliminate a certain number of " suspect " cases ; a definite leucocytosis excludes enteric fever, while a true leucopenia, (4,500 cells or less), is very suggestive of an enteric infection, especially when associated with a definite mono-nuclear increase. A mono-nuclear leucocytosis may persist throughout convalescence. Counts of from 5,000 to 7,000 white cells are, however, often found in influenza or trench fever.

The atropine test, introduced by Harris, is a useful aid to diagnosis, but its value lies chiefly in the fact that a series of negative results excludes enteric fever ; unfortunately it has been found that a positive result may be obtained in about 20 per cent, of cases of trench fever and possibly other febrile disorders as well, certainly also in cases of infection by B. aertrycke. The test depends on the variation in the pulse rate of the suspect after the hypodermic injection of 1/33 gr. of atropine sulphate. The injection is given one hour after a meal, the patient being recumbent, the pulse is counted every minute till it is of uniform rate, the atropine is injected and the pulse rate noted minute by minute for from 30 to 35 minutes. The maximum increase due to the atropine is thus ascertained. If the increase does not exceed 14 beats per minute, the reaction is positive. The test is applicable from the fifth to fourteenth day of fever ; it is not reliable when the initial pulse rate is over 100 beats per minute and should not be employed in patients over 50 years of age or those who are markedly art erio-scler otic.

It is thus possible by clinical means to earmark two classes of cases, first those that can be considered as certainly enteric fever, and secondly those that must be regarded as suspect cases till the diagnosis can be confirmed or refuted. Both classes

ENTERIC GROUP OF FEVERS 51

must be referred to the bacteriologist without delay for confirmation and for the identification of the specific infective agent. Topley, Platts and Imrie claim that about 5 per cent, of the cases invalided from the Western Front as pyrexia of uncertain origin were in reality suffering from enteric fever ; it is probable, however, that this figure is too high.

Cases of disease due to the food poisoning group of organisms, Bacillus enteritidis, Bacillus aertrycke and Bacillus suipestifer, can generally be diagnosed clinically from enteric fever by the sudden onset with severe diarrhoea and vomiting, the occurrence in epidemic form of several cases at exactly the same time, the short duration of fever and the absence of the classical features of enteric fever. Perry and Tidy, discussing an extensive epidemic due to Bacillus aertrycke, noted a latent period of 6 to 28 hours, sudden onset with diarrhoea and abdominal pain in many cases apyrexial, fever when present rarely lasting more than two days, tongue clean throughout and stools watery with but little faecal matter.

A considerable epidemic of disease due to Bacillus suipestifer was noted in Egypt in 1917. The cases resembled in the main those due to Bacillus aertrycke but vomiting was more pro- nounced a feature and the temperature remained high for 96 hours, thereafter falling by crisis.

With regard to laboratory diagnosis this is easy in unpro- tected persons. The specific bacillus can be recovered from the blood in most cases up to the fifth day and often for longer. If the blood fails, cultivation of the stools or urine will often give a positive result in the second and third weeks of the disease.

In war time it often happens that cases do not get within touch of a properly equipped laboratory till the second week or later ; it is then necessary to test for specific agglutihins in the patient's blood serum. In a positive case these will appear from about the tenth to the twelfth day, though the paratyphoid A agglutinins may be delayed till the third week.

In protected individuals the procedure is less simple because the percentage of cases in which the specific organism can be recovered from the blood, urine or faeces is much less, and the act of inoculation causes the specific agglutinins for the organisms, against which the person has been protected, to appear in the blood serum, quite apart from any infection having taken place.

It has been shown, however, that by the use of special technique and standard bacillary emulsions, as advocated by Dreyer, a positive diagnosis can be made in nearly every case by noting the variations in agglutination titre to the different

52 MEDICAL HISTORY OF THE WAR

organisms exhibited by the patient's blood serum throughout the course of the disease.

Infections due to Bacillus enteritidis may closely resemble some cases of paratyphoid B fever ; this organism possesses identical biochemical reactions with Bacillus paratyphosus B but can be distinguished readily by agglutination tests.

Infections due to Bacillus aertrycke or Bacillus suipestifer are more difficult, as their biochemical and agglutination reactions are the same as for paratyphoid B. Absorption tests, however, serve to differentiate Bacillus paratyphosus B from the others.

The diagnostic position of enteric fever may be summarized as follows : —

A diagnosis of enteric fever may be justified on purely clinical grounds even though unsupported by bacteriological or serological findings. The percentage of cases thus unsup- ported will be small, probably less than 5 per cent, in those protected by triple vaccine, and practically negligible amongst unprotected men, provided they are under observation sufficiently long for a series of agglutinations to be determined. An additional factor which applies also to unprotected men is the occurrence of infections by bacilli closely allied to, but not really belonging to, the typho-paratyphoid groups as at present defined.

The isolation of one of the specific bacilli from the blood is the simplest and most conclusive proof of infection. This should always be attempted as soon as enteric fever is suspected. Bacillus typhosus has been recovered by Torrens from the blood on the 26th day of illness quite apart from a relapse. Recovery of the bacilli from the stools or urine is the next most satisfactory proof of infection ; this procedure is most successful in the second, third, and fourth weeks of the disease.

If no bacilli can be recovered in those protected by triple vaccine, the accurate diagnosis must depend on the agglutina- tion curves of the patient's blood serum, as determined by three, four, or more successive readings at intervals of three, four, or five days. A variation of 150 to 200 per cent, or more in the agglutination titre to one of the bacillary emulsions between the twelfth and thirtieth days of illness implies an infection with that bacillus. A variation of as little as 100 per cent, is probably sufficient but may just fall within the limit of technical error. In unprotected men agglutination with any of the three bacilli in higher serum-dilution than 1-10 is proof of infection with that bacillus ; in the case of paratyphoid A a positive diagnosis is justified even if the maximum titre is no more than one in ten. To take these agglutination readings it is essential to use standard agglutinable bacillary emulsions, to

ENTERIC GROUP OF FEVERS 53

use the macroscopic method and to follow closely the technique laid down by Dreyer and Ainley Walker.

Treatment.

With regard to the general treatment of enteric fever the experiences of the war have done nothing to modify the old- established methods. The essential factors still remain, namely, good nursing, careful dieting, and enforced rest at the earliest possible moment in the nearest hospital set apart for the treat- ment of these cases. It cannot be too strongly emphasized that there is nothing so prejudicial to the interests of the patient as repeated transference from place to place or even from one ward to another in the same hospital.

It is generally agreed that only fluids and jellies should be permitted during the height of the disease, with the possible addition of milk-chocolate and rusks after the tenth day in mild non-toxic cases ; there is, however, a tendency to permit solid food to cases of paratyphoid fever relatively early in the disease. Nothing is probably gained by this course, and it is safer to adopt the old rule that no case should have solid food till the temperature has been normal for seven days ; the convalescence of cases treated in this manner is speedier and less interrupted by relapses than when solids are permitted at an earlier stage. It is of the greatest importance to encourage the patient to drink as much water as possible during the height of the disease.

No drug is of specific value in the treatment of enteric fever. An aperient should be administered if the case is constipated and seen in the first ten days ; after this date the bowels should be opened, if necessary, every other day by the administration of an enema of normal saline. Antipyretics and intestinal antiseptics are best avoided. Liquid paraffin may be given with advantage throughout the disease so long as there is no diarrhoea, as it tends to minimize the constipation which is often so obstinate during convalescence. Bromide is of service if insomnia is troublesome.

Stimulants are but rarely necessary or desirable until con- valescence ; the pulse must be the guide. It appears that very severe cases can be kept alive a few hours or days longer than would otherwise be the case by the free exhibition of brandy, but that rarely, if ever, is a fatal issue avoided by this means. This does not apply to cases who have developed pneumonia or who have been operated on for perforation ; or to the occasional administration of a tablespoonful of whisky in a little warm milk to induce sleep.

The foul condition of the mouth and the characteristic typhoid

54 MEDICAL HISTORY OF THE WAR

tongue can be greatly improved by careful attention, especially by encouraging the patient to use " chewing gum " which is an excellent prevention of the septic parotitis so frequently occurring in typhoid fever.

Immersion in baths can rarely be practicable in wartime, even if it be desirable. Tepid or cold sponging is, however, of the greatest value ; it should be done as a routine measure every four to six hours to all patients whose temperatures are 103° or over. Apart from the degree of pyrexia, sponging is the most valuable remedy for restlessness or insomnia.

With regard to the treatment of complications, meteorism is best treated by stopping milk and allowing only whey or albumen water for 48 hours or longer. A simple enema may be of service ; the turpentine enema should be used with caution and not during the third week of the disease, since there is no means of estimating the extent of ulceration in the large intestine.

Immediate operation offers the best chance in cases of perfor- ation. Peritonitis without perforation may be localized and unsuspected clinically ; such cases often recover. If generalized, operation should be undertaken as soon as the diagnosis is made.

In the event of haemorrhage occurring all fluids should be stopped for at least 48 hours and sufficient morphia injected to keep the patient absolutely at rest. The mouth must be care- fully attended to during this time. Feeding, when recommenced, must proceed with the utmost caution. It is amazing what a large amount of blood can be lost without death ensuing. One large haemorrhage is often less serious than a series of smaller ones. When the haemorrhage seems definitely to have ceased, subcutaneous infusion with saline solution up to 30 oz. may be permitted, if the condition of the patient remains unsatisfactory ; and this may be repeated if no further bleeding takes place. It is well to attempt to anticipate the occurrence of haemorrhage by increasing the coagulability of the blood about the time when the sloughs may be expected to separate. For this purpose 30 gr. of calcium lactate may be given thrice daily from the sixteenth to the twentieth day of typhoid fever and from the fourteenth to the eighteenth day of paratyphoid fever. In a considerable number of cases treated in this manner, and checked by controls not so treated, the results appeared dis- tinctly to justify the measure ; haemorrhage was less frequent and, when it did occur, of less severity.

In cases of thrombosis the administration of citrates is indicated. Marris claims excellent results from the intravenous injection of 10 oz. of 5 per cent. sod. citrate solution.

Pulmonary complications must be dealt with on their merits.

ENTERIC GROUP OF FEVERS 55

Cholecystitis and gallstones may require surgical intervention. The former will usually yield to aspirin and urotropine.

In addition to these general methods, certain special methods of treatment have to be considered.

Various writers have advocated from time to tune the therapeutic use of vaccines in enteric fever. A great variety of different forms of vaccine have been employed, varying from stock killed cultures, as used for prophylactic inoculation by Wiltshire and MacGillicuddy, to an autogenous living vaccine used by Bourke, Evans and Rowland. The dosage has varied within wide limits and the vaccine has been given subcuta- neously, orally, or intravenously. In most cases the evidence adduced in favour of vaccine treatment fails to carry conviction. The cases are few in number and there is no record of specially selected similar control cases treated at the same time without vaccines.

In January 1915, Torrens believed he was favourably influ- encing certain cases by injection of stock antityphoid vaccine ; many of these cases were later proved to be paratyphoid fever running their normal course. Subsequently, a considerable experience of vaccines both stock and autogenous led him to the belief that equally good results were obtained in both typhoid and paratyphoid fever without the use of such vaccines as he was able to procure. As regards the use of stock anti- typhoid vaccine for cases of Bacillus typhosus infection, Whittington has shown in a careful analysis of controlled cases that the results are no better with vaccine than without it, that there is " a distinct suspicion that the vaccine increases the incidence of haemorrhage," and that neither the duration of the fever nor the occurrence of complications is appreciably altered.

It thus appears that there is not sufficient evidence to justify a dogmatic opinion on the value of vaccine treatment in enteric rver, but the probability is that it is of little value as hitherto >ractised, while it is certain that its beneficial effect is by no leans striking.

Serum treatment, promising though it seems on theoretical rounds, does not appear to have been discussed in English icdical literature, although it has been used in France.

Prevention.

The measures employed for the prevention of enteric fever rere prophylactic inoculation and general measures of hygiene

id sanitation. In August 1914, in conformity with the usual >rocedure by which troops were not inoculated against enteric iver until they were proceeding on service abroad, only a small >roportion of the troops forming the expeditionary force was

56 MEDICAL HISTORY OF THE WAR

protected by inoculation at the time war was declared. But the work of inoculation was carried on energetically after the expeditionary force arrived in France, and eventually the proportion of inoculated men exceeded 90 per cent. The pro- gress of events during the five years of the war has proved conclusively that it is the best, most important and successful means at our disposal for combating typhoid fever. Inocu- lation and systematic re-inoculation at stated intervals should be rigidly enforced in every army. The success of anti-typhoid inoculation was assured by the autumn of 1915, and the question then arose as to the advisability of introducing a similar measure to deal with the paratyphoid fevers which threatened to become a distinct menace to the health of the army.

In January 1915 inoculation with triple vaccine was adopted as a routine for all the British expeditionary forces. One c.c. of vaccine contained 1,000 million of Bacillus typhosus and 750 million each of Bacillus paratyphosus A and B. Two injections were given at an interval of eight to ten days, the first dose being 0-5 c.c. and the second 1 -0 c.c.

It was also ordered that re-inoculation, one dose of 1 c.c., should be performed as a routine measure after an interval of from eighteen months to two years. It had been shown that the result of simple anti-typhoid inoculation was not only to reduce the incidence of typhoid fever but also to diminish the severity of the infection when acquired, as well as the liability to complications ; but it was possible that some of these beneficial effects might be impaired or abolished by the addition of paratyphoid bacilli to the vaccine. In the event, however, the experiment was amply justified ; the inci- dence of each infection steadily decreased year by year and was always conspicuously less in those who had been protected by inoculation. The death rate per 1,000 of ration strength was also consistently lower for each infection amongst the protected, as also the case mortality per cent., except for paratyphoid in 1918, when, however, there were too few cases to afford reliable evidence.

The increase in the case mortality from typhoid fever in 1917 and 1918 occurs in both protected and unprotected ; it therefore probably depends on other factors rather than on the adoption of triple vaccine. This has been referred to above, and even if it could be shown to depend entirely on triple vaccine the disadvantage would be many times counterbalanced by the very much lower incidence of the enteric infections in protected persons. The efficacy of prophylactic inoculation is shown in the following table, taken from the official returns for the Western Front from 1914 to 1918.

ENTERIC GROUP OF FEVERS TABLE XVII.

57

	1	II	CM «C — < 00 *-< CO CO-^tx ^^ CO 00 00 ^* CO C^ ^* CM CM CM O CM CD	8S§	1C CD O5 CM CM
	"o	Protected.	^H 1 1 f* 1 1 CO CD CM 1C ^H 1 O5 10 CD 1C CD CM CO	-^ O5 CD 05 CO-* »-< »- 1 CO	CD CO Cl
.	I	OTJ M 4)	CO 00 t>. COICM CMt^O) COt>>t>«	t!SS	8 | ,
^	£	53	t^. CO CO "-i 00 1-"	<N CM CO	S
1	Case Mortal	1	00 CO CM 00 | | » | | u-^cp 1C l> '-"-'	JC ^ !^	•^f oo 00 IN t^ CO CM
1	1	it	CO	CO CO CO i-< O »-«	cS 1 )
	b PH O	0*	1-1
1	-M 2 1	I 8	§CO CM 88	00 O CO SS8	£88 O O O
S)	P	£
i	|	||	^^CN	822	O5 r»< M -<0<N
i	i		' ' ' « • ^ ' co<^	i— i i— i ^<
3	1 'C	1	III CO 1 1 !>«>-" Ill 05 1 1 ICCMCO	o i>a-	Soc5
'"i	a	£
§
t	(	!	li Si i	fijjjj	jH« figg
	1	i	Tf* 1C CO »-« »-H »— t O5 O5 O5	O5	00 s
	Theatre of	M PS £	1

58 MEDICAL HISTORY OF THE WAR

It will be seen from this table that the influence of protective inoculation on the liability to infection from enteric fever is undoubted, and equally undoubted is the very much lower case mortality in typhoid fever. It is difficult to say from these statistics whether triple vaccine has any pronounced effect on the case mortality of paratyphoid fever. The 1917 figures suggest that it had, but the 1918 figures do not confirm this. It must be noted, however, that in this latter year the total number of cases of paratyphoid fever on the Western Front was too few to be of much value. Only two deaths occurred, one in paratyphoid A and one in paratyphoid B, and it is probably quite accidental that both these happened in protected men. The incidence of complications seems to be lowered in all three infections by the use of triple vaccine.

Since one result of infection by one of the organisms of enteric fever is the appearance of specific agglutinins in the patient's blood serum and identical agglutinins are produced by the injection of the appropriate vaccine, it is reasonable to suppose that the amount of agglutinin to Bacillus typhosus, Bacillus paratyphosus A and Bacillus paratyphosus B, respectively present in the blood serum after prophylactic inoculation with triple vaccine, affords some approximate idea of the relative immunity conferred against each of the three infections. It must be remembered, however, that the infections themselves do not produce identical amounts of agglutinin in every case and that the response to paratyphoid A is habitually very much less than that to either paratyphoid B or typhoid ; so that a quite low agglutination titre to paratyphoid A might conceivably indicate the same actual degree of immunity as a much higher agglutination titre to paratyphoid B or typhoid.

It might appear from the preceding remarks that prophylactic inoculation was the only necessary preventive measure against enteric fever. This is far from being the case, and the success of the campaign against enteric fever has been in no small measure due to the unremitting care and energy of the army sanitary authorities.

For the details connected with the diverse sanitary measures rendered necessary by the varying features of the different campaigns, reference must be made to the volumes on the Hygiene of the War; but the general principles may be summarised here.

The water supply must be beyond reproach ; in the case of the trenches this can best be secured by the daily provision in tins of an adequate supply which must be chlorinated before use, or by the individual use of bisulphate of sodium tabloids. Water which has not been either boiled or sterilized in this

ENTERIC GROUP OF FEVERS 59

manner must be used for no personal purpose whatsoever. Behind the line the precautions must be equally strict ; but it is, of course, easier there to arrange for the provision of large tanks of properly chlorinated water. All vessels used for the carrying of water for cooking and for washing up must be kept scrupulously clean and covered up. In European countries the town water supply usually requires careful testing and super- vision. At Rouen, Boulogne and elsewhere in France the supply was by no means safe, and it was found necessary to install a chlorinating plant at the source of supply.

In the Eastern theatres of war the troops may often be compelled to rely on a single water supply such as a river with its subsidiary canals ; such water is highly dangerous and should only be used after sedimentation or clarification with alum and chlorination or after prolonged boiling. Since it may often be lecessary for large bodies of mobile troops to be several hours

it of reach of their own water supply, the utmost care must be taken to ensure that their water bottles are clean, properly corked and as large as possible. Tablets of bisulphate of sodium

ly be issued when there is a likelihood of temporary shortage >f chlorinated water ; these destroy cholera vibrios and all

icilli of the coli group in twenty minutes. It is customary to

ink large quantities of soda water in the East, and the very

ictest supervision is necessary over all soda-water factories ensure that the returned bottles are properly washed in

tlorinated water before being refilled, and that only properly terilised water is used for aeration.

At the base and behind the line all excrement should be burnt in an incinerator ; the urine pails should be emptied twice daily into a suitable soakage pit. In the vicinity of the front line deep trenches must be dug when practicable, and should be covered with a board to exclude flies. The site of all ground used for this purpose must be carefully marked to prevent its being used again. All urine must be passed into special tins which are emptied regularly into properly constructed soakage pits. Cresol should be placed in every tin before it is used. In permanent camps urine must be disposed of in soakage pits or evaporated in incinerators.

Latrines and cookhouses must be rendered fly-proof as far as possible by the use of canvas screens, wire gauze, etc. Special attention must be directed to the breeding places of flies, and manure must be suitably treated and disposed of.

Vegetables and fruits must not be eaten uncooked except after efficient cleansing in pure water.

Every case suspected to be enteric fever should be notified, isolated at once, and sent without delay to a hospital for

60

MEDICAL HISTORY OF THE WAR

infectious diseases. The occurrence of a sporadic case should lead to strict investigation as to a possible carrier in the troop or company. Spot maps must be kept of all cases and their probable place of origin. Every patient should be kept isolated until he is definitely proved not to be a carrier.

In the event of an outbreak of enteric fever in the civilian population of a town or district necessarily occupied by troops, special hospitals must be provided and all cases should be compulsorily sent to them. Infected houses and areas must be recognized and placed strictly out of bounds. Immediate notification to the sanitary authorities of all suspects is essential. The efficiency of the measures outlined above depends very largely on cordial co-operation between the combatant and the medical or sanitary authorities. This co-operation will be very much closer if steps are taken to explain the reason for the various rules and regulations. This can readily be accomplished by means of an occasional short address by the medical or company officer.

BIBLIOGRAPHY. Achard & Bensaude Infections Paratypho'idiques . .

Archibald

Bainbridge Boidin

Boney, Grossman &

Boulenger Bourke, Evans &

Rowland

Buxton Carles .. Coutts Gushing

Enterica in the Soudan

Paratyphoid Fever and Meat

Poisoning. Sur la mortalite des fievres

typhoides

Report of Base Laboratory in

Mesopotamia. Autogenous living Vaccine in

the treatment of Enteric

Fever.

. . La Fievre Typhoide du Com- battant

. . Paratyphoid in the Army at the Dardanelles

. . A comparative study of some members of a pathogenic group of bacilli of the hog Cholera or B. enteritidis (Gartner) type. Dawson & Hume . . Jaundice of Infective Origin . .

Bull, et Mem. Soc.

Med. des H6p. de

Paris, 1896. 3e S.,

Vol. xiii, p. 820. Journ. Trop. Med.,

1918. Vol. xxi,

p. 229. Lancet, 1912. Vol. i,

pp. 705, 771, 849. Arch, de Med. et

Pharm. Mil., Paris,

1916. Vol. Ixvi,

p. 514. Jl. of R.A.M.C., 1918.

Vol. xxx, p. 409. B.M.J., 1915. Vol. i,

p. 584.

Journ. Med. Research, 1904-1905. Vol.viii, N.S., p. 431.

Journ. de M6d. de Bordeaux, 1916. Vol. xlvi, p. 65

Can. Med. Ass. Journ. Toronto, 1917. Vol. vii, p. 97.

Johns Hopkins Bul- letin, 1900. Vol. xi, p. 156.

Quar. Journ. Med., 1916-17. Vol. x, p. 90.

ENTERIC GROUP OF FEVERS

61

BIBLIOGRAPHY— cont.

Dawson & Whittington Paratyphoid Fever, of Fatal Cases.

Durham

A Study Quar. Journ. Med., 1915-16. Vol. ix, p. 98.

Dreyer & Ainley Walker The diagnosis of the Enteric Lancet, 1916. Vol. ii, Fevers in inoculated indi- p. 98. viduals by the Agglutinin Reaction.

Some theoretical considerations Jl. on nature of agglutinins, to- gether with further obser- vations on B. typhi abdomin- alis, B. enteritidis, B. colicom- munis, B. lactis aerogenis and some other bacilli of allied character. The Paratyphoid Problem in

India. Paratyphoid Infections

of Experimental Med., 1900-01. Vol. v, p. 353.

Firth

Fletcher

Fortescue-Brickdale Glynn & Lowe

Goodall

Grattan & Harvey Grattan & Wood Gwyn

Harvey

Hirschfeld

Hichens & Boome Job & Ballet

Ledingham MacAdam

Jl. of R.A.M.C., 1911.

Vol. xvii, p. 136. Jl. of R.A.M.C., 1904.

Vol. ii, p. 241. Jl. of R.A.M.C., 1918. Vol. xxx, p. 51.

Report on Bacteriological Ex- amination of Soldiers conva- lescent from Diseases of the Enteric Group.

Notes on the Symptomatology Lancet, 1917. Vol. i, of Paratyphoid Fever. p. 611.

Observations on the Serum Re- Jl. of R.A.M.C., 1916. action of 300 Unselected Vol. xxvii, p. 663. Cases of Enteric from the Eastern Mediterranean, with the Oxford Standard Agglu- tinable Cultures.

Enteric Fever in Flanders, 1914 Proc. Roy. Soc. Med., and 1915. 1918-19. Vol. xii

(Epid. Sect.), p. 18.

Inquiry into small Epidemic of Jl. of R.A.M.C., 1911. Paratyphoid Fever in Camp. Vol. xvi, p. 9.

Paratyphoid Fever in India . . Jl. of R.A.M.C., 1911.

Vol. xvii, p. 143.

On infection with a paracolon Johns Hopkins Bui- bacillus in a case with all the letin, 1898. Vol. clinical features of Typhoid Fever.

The Causation and Prevention of Enteric Fever in Military Service, with Special Refer- ence to the Importance of the Carrier.

A New Germ of Paratyphoid . .

ix, p. 54.

Jl. of R.A.M.C., 1915. Vol. xxiv, p. 491 ; Vol. xxv, pp. 94, 193.

Vol. i.

Lancet, 1919.

p. 296. A fatal case of Paratyphoid B.M.J., 1918. Vol. i,

B simulating Typhus Fever. p. 398. Contribution a 1'etude de Bull, et Mem. Soc. 1'anatomie pathologique des Med. des Hdp. de fievres paratyphoides. . . Paris, 1915. Vol.

xxxix, 3e S.,p.991.

Dysentery and Enteric Disease Jl. of R.A.M.C., 1920. in Mesopotamia. Vol. xxxiv., p. 306.

An account of an infection in Jl. of R.A.M.C., 1919. Mesopotamia due to a bacil- Vol. xxxiii, p. 140. lus of the Gaertner-para- typhoid Group.

62

MEDICAL HISTORY OF THE WAR

MacAdam Mackie & Bowen

Martin & Upjohn Marris

Miller

Morley & Battinson

Smith NobScourt & Peyre . .

Perry & Tidy

Perry ..

Rathery & Ambard . .

Rodet .

BIBLIOGRAPHY— cont.

Thrombosis of Cerebral Arteries Lancet, 1916. Vol. i,

in Paratyphoid B. p. 243.

Note on the characters of an Jl. of R.A.M.C., 1919.

anomalous member of the Vol. xxxiii, p. 154.

Paratyphoid Group met with

in Mesopotamia. The distribution of Typhoid Jl. of R.A.M.C., 1916,

and Paratyphoid Infection Vol. xxvii, p. 583.

amongst Enteric Fevers at

Mudros, Oct.-Dec., 1915. The use of Atropine as an aid B.M.J., 1916. Vol. ii,

to the Diagnosis of Typhoid p. 717.

and Paratyphoid A and B

Infections. Goulstonian Lectures on Para- Lancet, 1917. Vol. i,

typhoid Infections. pp. 747, 827, 901.

Acute Gangrenous Cholecystitis B.M.J., 1916. Vol. i,

p. 444. Complications observees au Bull, et Mem. Soc.

cours des fievres typhoi'des Med. des Hop. de

et paratyphoiides. Paris, 1916. Vol. xl

3eS. A Report on an Epidemic Med. Research Comm.

caused by Bacillus aertrycke. Spec. Report Series

No. 24. London, 1919. Illustrations of the Agglutina- Lancet, 1918. Vol. i,

tion Method of Diagnosis in p. 593.

Triple inoculated Individuals. Les Fievres Paratyphoides B Paris, 1916.

a 1'Hopital Mixte de Zuyd-

coote, de Dec. 1914 a Fev.

1916. Serotherapie antityphoi'dique : Bull. Acad. de Med.,

preparation du serum.

Rodet & Bonnamour. Serotherapy of typhoid fever .

Serotherapy in typhoid fever

Paris, 1916. Vol. Ixxvi, pp. 83-85.

Serotherapie antityphoidique : Bull. Acad. de Med., application. Paris, 1916. Vol.

Ixxvi, pp. 114-116. Bull. Acad. de Med., Paris, 1919. Vol. Ixxxi, p. 759. PresseM6dicale, Paris, 1920. Vol. xxviii, p. 81.

The Blood pressure in Typhoid Med. Press & Circular, Fever. 1916. Vol. i, p. 234.

A new chromogenous bacillus Proc. Amer. Ass. Ad- vanced Sc., 1885. Vol. xxxiv, p. 303.

Sarrailhe & Clunet .. La Jaunisse des Camps et 1'epi- Bull, et Mem. Soc. demie de Paratyphoide des Med. des H6p. Dardanelles. Paris, 1916. Vol. xl,

3e S., p 45. Brain Abscess in a Case of Para- Lancet, 1915. Vol. i,

typhoid B. p. 852.

Report on the probable pro- M.R. Committee, portion of Enteric Infections Spec. Rep. Series among undiagnosed Febrile No. 48. London, Cases invalided from the 1920. Western Front since Oct., 1916.

Rolleston

Salmon & Theobald Smith

Scott & Johnston . . Topley, Platts & Imrie

ENTERIC GROUP OF FEVERS

63

Torrens & Whittington Vincent & Muratet . .

Webb- Johnson

Weeks Whittington

Willcox

Wiltshire & McGilli- cuddy

BIBLIOGRAPHY— cont.

A Preliminary note on the Clinical Aspects and Diag- nosis of Paratyphoid Fever.

Typhoid Fevers and Para- typhoid Fevers.

Hunterian Lecture on the Sur- gical Complications of Ty- phoid & Paratyphoid Fevers.

Empyema due to infection by B. paratyphosus A.

Report on the use of Stock Vac- cine in infection by B. ty- phosus, with analysis of 230 cases.

Paratyphoid Fever, its clinical features and prophylaxis.

Experience in Treatment of Typhoid Fever by Stock Ty- phoid Vaccine.

Jl. of R.A.M.C., 1915. Vol. xxvi, p. 359.

Military Medical

Manuals. Trans.

by J. D. Rolleston.

London, 1917. Lancet, 1917. Vol.

ii, p. 813.

Lancet, 1916. VoL

ii, p. 433. Jl. of R.A.M.C., 1916.

Vol. xxvii, p. 422.

Lancet, 1916. Vol. i,

p. 454. Lancet, 1915. Vol. ii,

p. 685.

Note. — Vincent and Muratet also quote Chevrel, Joltrain and Petitjean, Lenglet and Sacquepee, and Miller quotes Sawasaki, to all of whom reference is made in the text of this chapter.

CHAPTER III.

DYSENTERY.

THE subject of dysentery is very extensive and comprises the knowledge of a considerable number of parasites, bacterial, protozoal and metazoal, which may cause inflammation and ulceration of the intestinal canal. The term " dysentery" is in many ways inappropriate and indicates solely the passage of blood and mucus in the stools accompanied by abdominal pain and tenesmus, symptoms which are common to several infections specifically distinct. The war presented an opportunity hitherto unrivalled for the study of bowel diseases, and this has been made full use of by bacteriologists and proto- zoologists. Notable advances in our knowledge of these sub- jects have been recorded by workers attached to the British forces.

Intestinal disorders, especially dysentery, furnished a con- siderable proportion of casualties on all fronts ; more especially was this the case in the tropical and subtropical theatres of war. In Gallipoli, Salonika, Egypt, Palestine, Mesopotamia, East Africa, and even in France and Flanders, dysentery at different times and seasons raged in epidemics of great magni- tude, and as a cause of invaliding and death it supplanted the enteric fever of British troops in more recent wars, though, taking the magnitude of the forces into account, there is no evidence to show that its incidence was higher than in the South African War.

So far as figures are available the incidence of dysentery in British Expeditionary Forces is shown in the following table : —

Table of Incidence of Dysentery (both Bacillary and Amoebic),

1914-1918.

	Aug.-Dec. 1914	1915	1916	1917	1918
		Ratio		Ratio		Ratio		Ratio		Ratio
	Total	per	Total	per	Total	per	Total	per	Total	per
	Cases	1000	Cases	1000	Cases	1000	Cases	1000	Cases	1000
France East A/rica . .	11	•05	20	•03	5,754	4-09	6,031 9,369	3-76 486-56	12,211 1,646	•79 116-51
Salonika					5,987	63-89	5,842	28-89	9,318	58-23
Italy									897	9-54
Egypt					5,599	31-19	4,341	23-13	4,906	21-80
Mesopotamia					1,839	50-94	4,960	60-34	5,445	51-12

64

BACILLARY DYSENTERY 65

Three types of dysentery, correlated to three different kinds of parasites, are now recognized. They are not mutually exclusive ; one type may be superimposed upon and complicate another. The principal types and their associated parasites are as follows : —

Bacterial . . . . Bacillary or epidemic dysentery.

Bacillus dysenteries (Shiga and

Flexner-Y). Protozoa! . . (a) Amoebic dysentery and hepatic

abscess (amcebiasis) .

Entamceba histolytica. (b) Balantidial dysentery.

Balantidium coli. Verminous . . (a) Bilharzial dysentery.

(Schistosoma mansoni, S. hcema-

tobium and S. japonicum) . (b) " Dysentery " associated with

(Esophagostomum apiostomum,

Ascaris lumbricoides, and Ankylo-

stoma duodenale.

Of these only the first two are of military importance, namely, the epidemic or bacillary, and the endemic or amoebic forms. They require, therefore, more lengthy consideration ; the re- maining types, together with the other conditions which they may simulate, are of importance chiefly in connection with differential diagnosis.

BACILLARY DYSENTERY.

The characteristics of bacillary dysentery are the acuteness of its onset, a well-marked initial pyrexia, severe abdominal pain and tenesmus, the presence of Bacillus dysenteries in the stools, and a tendency of the disease to occur in epidemic form. After recovery from the initial attack, there is little tendency to relapse. The . pathological process consists of an initial diphtheritic necrosis of the large intestine, together with a toxaemia of varying degree.

This type of dysentery was prevalent throughout the whole war. It first claimed serious attention when it broke out in epidemic form in Gallipoli in August 1915, where in three months it was responsible for a high proportion of the 120,000 casualties evacuated from the Peninsula on account of sickness.

From that date onwards it was much in evidence in all the Eastern theatres, being responsible for at least 90 per cent, of the acute clinical dysentery recorded.

In France and Belgium a milder form of bacillary dysentery

(2396) E

66 MEDICAL HISTORY OF THE WAR

commenced in July 1916 and reached its maximum in September of that year ; similar epidemics also occurred in the autumn of the succeeding two years of war. The maximum incidence recorded was 126-62 cases per 100,000 troops in September 1916.

In the other theatres of war it was also prevalent : the ad- mission rate to hospital per 100,000 of ration strength varied from 7,900 in Mesopotamia in 1916 to 1,300 in Egypt and 990 in Salonika in 1919 ; it exhibited also a distinct seasonal incidence, occurring in epidemic form as a disease of the late summer and autumn with a maximum prevalence in October, though minor outbreaks were noted during the spring months of March and April. Sporadic cases were apt to occur through- out the whole year ; but, on the other hand, during the hot summer months in Macedonia, Egypt, and Mesopotamia, the disease was almost entirely in abeyance. The case mortality rate is difficult to estimate ; probably it assumed its greatest virulence during the Gallipoli epidemic, though even there the death rate cannot have exceeded 5 per cent. ; statistics show that in Macedonia, Egypt and Mesopotamia from 1915 onwards it rarely exceeded 2-7 per cent.

The true importance of this disease, as a military factor, is not to be reckoned solely from the point of view of the death rate, but from the amount of invalidism it causes, for in indi- viduals recovering from a severe attack the mucous membrane of the intestine may be so damaged as to render them unfit for further service.

Mtiology.

With regard to its aetiology, circumstances which predispose to the development of bacillary dysentery are just those which are unavoidable under the conditions of modern warfare ; that is, close contact of one man with another, physical exertion, a monotonous diet of preserved food, and one must add to these another factor upon which sufficient importance does not seem to have been laid, namely, the mechanical irritation of the intestinal mucous membrane by dust or sand ingested in the food. In desert warfare, or in arid regions such as Gallipoli and Egypt, it is almost impossible at times to avoid swallowing a considerable amount of sand with the food. This in itself is sufficient to produce a lienteric diarrhoea and so prepare the way for the activities of the dysentery bacillus, which is mainly disseminated by means of flies and polluted water.

The first outbreak of dysentery on a large scale in France occurred during the first battle of the Somme, when the British occupied ground from which the enemy had been driven. It was known at the time that dysentery was prevalent in his

BACILLARY DYSENTERY 67

lines, and it was suspected that one source, at any rate, of the disease was the contamination of this ground. In 1918 dysentery prevailed at the time of the British advance under the same conditions. Indeed, at every phase of active movement and almost at any time in the fighting line, sanitary regulations could hardly be carried out with complete accuracy. Latrines could not be dug or kept so well as desirable, garbage and faeces could not be burnt, and the provision of water was often difficult. Under the latter difficulty the use of disinfecting water tablets was largely increased. Experience goes to show that all these dangers are increased when enemy lines are captured and occupied.

Another cause that temporarily predisposed towards in- creasing the spread of the disease was found in the crowded state of the infantry base depots in France. Camps designed for 1,200 men sometimes contained between 2,000 and 3,000. In the event of carriers being present — an accident which, in

)ite of all precautions, occurred and always will occur — such

mditions of overcrowding materially aided the spread of the

>ease.

It is known that epidemics occurred amongst British prisoners )f war in Germany, a fact which was brought to the notice of War Cabinet by the Admiralty, War Office, Air Ministry,

)lonial Office and Prisoners of War Department in a special >int memorandum dated 25th September, 1918.

The dysentery bacillus was discovered in Japan in 1897 by )higa, and in 1900 an organism, morphologically similar, but differing in its power of fermenting mannite, was isolated by Flexner in the Philippines. Since that date a great deal of attention has been paid to this subject, with the result that many variants of these two organisms have been described. The bacteriology of bacillary dysentery attracted a considerable amount of attention during the war. Interest centred chiefly around the mannite-f ermenting bacilli first described by Flexner and afterwards elaborated by Hiss and Russell, Strong and others. This work was important mainly from the point of view of laboratory diagnosis and the preparation of effective anti-sera, and it was undertaken by Murray, Gettings, Dudgeon, Andrewes and Inman.

The species of bacteria which are now recognized in bacillary dysentery are : Shiga's bacillus, the Flexner-Y group of bacilli, and certain atypical bacilli.

Shiga's bacillus, fermenting glucose only amongst the sugars and alcohols* employed as tests, and forming no indol, has

* The fermentable substances of real service in the classification of the dysentery group are four in number : glucose, mannite, lactose and dulcite.

68 MEDICAL HISTORY OF THE WAR

been abundantly proved to be the cause of dysentery. Being much the most toxic of dysentery bacilli, it is responsible for the most serious cases and for the greater number of fatalities. It was recorded commonly from all theatres of war. In the East it accounted for about half the number of cases and showed no special epidemic prevalence ; in France and Belgium, on the other hand, it apparently played a minor part, on the whole accounting for 15 per cent, of the dysenteries. It predominated at the commencement of the epidemic in August and September 1916, but was more or less replaced by Flexner bacilli later on in the late autumn. In about 2 or 3 per cent, of bacteriologi- cally diagnosed cases both Shiga and Flexner bacilli co-existed.

The Flexner-Y group of bacilli apparently belongs to a single species, fermenting glucose and mannite, but not lactose or dulcite. As in the case of the former organism, the evidence connecting this species with dysentery is complete. The researches of Gettings, Murray, Andrewes and Inman undertaken on a large scale have indicated that serological races of the species exist. It may be regarded as a group formed of at least four distinct antigenic components which have been provisionally lettered V, W, X and Z. Any of these four components may so preponderate in different strains as to impart a distinct serological facies. The corresponding agglutinins are not mutually absorbed, except in a slight degree. In addition to the four serological races thus denned, there remains the true Y bacillus of Hiss and Russell, which presents differences in its agglutinability and agglutinogenic capacities. These bacilli are not nearly so toxic as is Shiga' s bacillus and are responsible for the milder and more chronic forms of the disease, though occasionally they may become virulent and cause death. That is to say, two species, namely the Shiga and the Flexner-Y bacilli of various sero- logical races, are responsible for the vast majority of cases of bacillary dysentery.

Other organisms which have been described during the war* and have been called atypical bacilli are bacilli resembling Flexner's bacillus but fermenting dulcite and separable from the latter on serological grounds, and a bacillus resembling Shiga' s in its sugar reactions, but forming indol and not agglutinating with Shiga antiserum, first described by Schmitz, in Austria, as the cause of dysentery and apparently identical with the B. ambiguus of Andrewes and the one

* Dumas has lately shown that the atypical organisms may be distinguished further by their power of producing fluorescence in media impregnated with neutral red, as well as by reduction of 1 per cent, lead acetate, thereby produc- ing a black line in a stab culture of agar containing this substance.

BAGILLARY DYSENTERY 69

described by Remlinger and d'Herelle. Together with these may be grouped organisms which ferment lactose early and differ serologically from the Flexner-Y group. There is no evidence for inculpating either of these groups, and they may be classified as " atypical" or " inagglu tumble " strains.

These latter need not claim the serious attention of the expert bacteriologist, though they constitute a source of fallacy to the beginner.

Morbid Anatomy.

With regard to the morbid anatomy of bacillary dysentery, the gross pathological appearances of the organs vary con- siderably according to the acuteness of the process in different individuals, and indeed in different epidemics, though the under- lying process is essentially the same in all cases. It is probable that no such opportunity has ever before presented itself for studying the effects of the dysenteric toxins upon the intestinal mucosa as was afforded to pathologists in the Eastern theatres of war. In mild cases it is naturally difficult to define the exact appearances of the earliest lesions ; the inflammatory changes originate in the first instance in the solitary lymphoid follicles of the large intestine. From these, superficial " snail track " liberations spread across the bowel, especially upon the free transverse folds, and the surrounding mucous membrane is involved in a greater or lesser degree with hypersecretion of viscid mucus. The abdominal viscera do not exhibit any striking changes.

In very»acute cases, succumbing to an overwhelming infection, the chief change is seen in the intestinal mucosa, but there are present as well abundant signs of a widespread toxaemia in other organs of the body. At first the process consists of acute hyperaemia of the mucosa of the large intestine, which, should life be sufficiently prolonged, ends in colliquative necrosis of the mucosa with involvement of the last two feet of the ileum— though, rarely, the whole of the ileum and the greater part of the jejunum may be similarly affected.

It is not generally realized that the specific lesions are most developed in the lower part of the intestinal canal, especially the rectum and pelvic colon. On opening the abdomen a paralytic distension of the large intestine is often found ; the mucosa is bright red in colour, very friable, and may actually drip with blood. Few, if any, intestinal contents will be found and the lumen may be occupied by viscid blood-stained mucus, or it may be pure blood and serous fluid. A general lymphoid peritonitis has been observed with the escape of free serum into the peritoneal cavity and the deposition of lymph flocculi on

70 MEDICAL HISTORY OF THE WAR

the peritoneal surface, together with oedema of the mesentery, especially at its posterior attachment. Post-mortem intussus- ception may occur. The mesenteric glands are inflamed and diffuse. The right side of the heart is engorged, the liver en- larged and congested with consequent parenchymatous changes. The gall bladder usually contains scanty and viscid amber- coloured bile. The spleen is generally dark, engorged and slightly diffluent, weighing about ten ounces. The suprarenal glands are congested and may show central necrosis.

In cases which do not run such a rapid course the intestinal mucosa is of plum-red colour, stippled with submucous haemorrhages, and the whole gut wall infiltrated and cedematous. Should the patient survive a week or more, these inflammatory changes result in colliquative necrosis of the mucosa ; the mucous membrane is converted into an olive-green, or it may be blackish, substance, rigid to the touch, and often honey- combed in a peculiar manner ; this substance represents the dead and functionless mucous membrane and it is therefore incorrect to describe it as " diphtheritic " dysentery, a term used by German authors. Exceptionally, the whole bowel wall may be converted into such a gangrenous substance. The peculiar green tint which this necrotic mucosa assumes is thought to be due to staining of the defunct tissues by bile pigments.

The intestinal contents in these cases generally consist of a dark-grey fluid containing much altered blood without the addition of mucus, which cannot be secreted when once the destruction of the goblet cells has taken place. The colli- quative necrosis may have a patchy distribution and may be confined to limited areas, as for instance, the hepatic and splenic flexures, or the descending and pelvic colons.

Should the patient survive, as he seldom does, such an exten- sive destruction of the bowel wall, the now defunct membrane is exfoliated in much the same manner as a diphtheritic mem- brane, exposing a raw, bleeding, granulated surface underneath. In a bowel which has undergone such disintegration complete regeneration of the mucosa does not take place ; restoration of the mucous membrane proceeds from islands of mucous membrane which escape unscathed. The whole process of repair would appear to constitute a struggle between proli- feration of the specialized epithelium and fibrosis.

Chronic ulceration of the large bowel in bacillary dysentery takes place in varying degrees of severity. The smallest lesions consist of lenticular ulcerations of the mucous membrane, involving the mucosa alone ; the more advanced lesions consist of ulceration of limited tracts of the mucosa, rarely penetrating

BAGILLARY DYSENTERY 71

beneath the muscularis. That ulceration may, although very rarely, proceed to ante-mortem perforation appears to be beyond doubt. The ulcers are roughly ovoid in shape, and run transversely to the long axis of the gut.

The ulcers of bacillary dysentery may be distinguished from lesions in dysentery of amoebic origin by the fact that they commence on the free edge of the transverse folds and run transversely, not longitudinally, to the long axis. In shape they are irregular in outline, with ragged undermined edges, often intercommunicating with neighbouring ones in contradis- tinction to the oval and rather regular shaped, isolated amoebic ulcer. The intervening mucous membrane is hyper aemic, cedematous and plum-coloured and there is no compensatory thickening of the gut. In amoebic ulcers, on the other hand, the intervening mucous membrane is healthy and there is considerable hypertrophy of the gut wall.

Another pathological condition, which is the direct sequel of chronic bacillary dysentery, and has so far attracted little attention, is the presence of tapioca-like mucus-retention cysts, varying from microscopic proportions to the size of a cherry stone, which jut out on to the mucosa and are situated beneath the scars of old ulcers. These cysts become secondarily invaded by B. coli organisms and frequently a peculiar B. coli septicaemia results, leading to formation of pyaemic abscesses in the cortex of the kidneys and very often to a fatal termination. They appear to be formed as the result of an adenomatous downgrowth of Lieberkiihn's follicles into the submucosa. Apparently in the formation of scar tissue part of the fundi of the crypts is nipped off.

Polypoid outgrowths reaching f to 1 in. in length, scattered throughout the rectum, have been observed as the result of a chronic bacillary infection.

Typical dysentery bacilli can be isolated from the gut in all stages of the disease. When the mucous membrane is necrotic, it is necessary to remember that successful isolation depends upon procuring material from beneath the necrotic tissue, where alone the bacillus can be found ; for this purpose one should sear the tissue with a red-hot knife and then scrape it away. Failure to remember this results in the isolation of putrefactive organisms, such as B. pyocyanem, which are found in necrotic tissue and which have nothing whatever to do with the patho- genesis of acute dysentery.

The bacillus has been recovered from the mesenteric glands, but never from the bile or blood post-mortem, though Flexner-Y organisms have been isolated from the blood-stream during life by Ledingham, Boyd, and others, and it is recorded that Wilson

72 MEDICAL HISTORY OF THE WAR

in France recovered the bacillus on three occasions by haemoculture of 88 acute Shiga cases. He also obtained both organisms, Shiga three times, Flexner eight times, out of 1,113 urines cultured.

In the most acute stage the mucous membrane is infiltrated with lymphocytes and plasma cells, the capillaries are engorged and the submucosa is the seat of numerous capillary haemor- rhages. The goblet cells show signs of great secretory activity. The inflammatory changes are most intense in the lymphoid follicles.

In the necrotic stage the whole mucosa has undergone coagulation necrosis and is converted into a structureless layer, in which only polymorphonuclear leucocytes with disintegrated nuclei can with difficulty be distinguished. The submucosa is greatly thickened to twice or three times its normal dimensions owing to oedema and haemorrhage. In fact, the chief feature would appear to be the destruction, or endothelial spoiling, of the nutrient vessels.

In the majority of microscopic sections of such an intestine, numbers of large macrophage cells, derived apparently from the endothelium of blood capillaries and lymphatics, may be distinguished. These cells are often of a considerable size, 15 to 20 microns in diameter, and may contain ingested leucocytes and red blood corpuscles. When voided in the stools they constitute a characteristic feature of the cellular exudate and, owing to their large size, refractility and phagocytic propen- sities, are apt to be mistaken for Entamceba histolytica, a point which will be referred to later.

Amongst Eastern peoples who are subject to recurrent attacks of bacillary dysentery, acute lesions are occasionally seen in a bowel which has recovered from a previous attack, with consequent scarring and fibrosis. The amount of destruc- tion to which such a bowel may be subjected, compatible with life, has to be seen to be believed. Some of the large intestines of Turkish prisoners for instance resembled pieces of parchment with radiating fibrotic scars, the result of healed dysenteric ulcers.

Amoebic ulceration may be superimposed upon a healed bacillary dysentery, though it is more usual to find an acute bacillary process terminating the more chronic amoebic disease.

Symptoms.

The incubation period of bacillary dysentery is probably 2 to 7 days. The clinical symptoms are never so characteristic that the clinician can afford to neglect the advantages of a laboratory diagnosis, and there is no disease

BACILLARY DYSENTERY 73

in which the mutual co-operation of the clinician and path- ologist is so necessary. All degrees of severity may occur, from a mild diarrhoea of three days' duration with passage of blood and mucus in the stools, to fulminating cases with death supervening in the same period.

On clinical grounds, bacillary dysentery can be classified into five types (a) mild, (b) acute, (c) toxic or fulminating, (d) re- lapsing, and (e) chronic.

The fulminating type may be divided into two sub-groups — the choleraic and the gangrenous. The onset is acute, generally with vomiting ; collapse with its attendant phenomena sets in early. The temperature is subnormal, the tongue dry and glazed, the skin cold and clammy, and the patient may complain of cramps. There is an initial watery diarrhoea, which is soon replaced by dark-red mucus containing a high proportion of blood or, it may be, serum alone. It is hardly necessary, from their superficial resemblance to cholera, to emphasize the importance of these cases.

The gangrenous form also commences suddenly with a rigor, headache and vomiting and other evidences of a severe toxaemia. The face is flushed ; the pulse rapid and bounding. The ab- dominal pain and tenesmus are very severe, but as the toxaemia increases these wear off. This is a point in prognosis, and one should be suspicious of patients with pyrexia who become insensitive to abdominal pain ; it is by no means a favourable omen. The stools at first resemble " meat-washings," but towards the end are composed of dark-grey offensive fluid, containing much altered blood. The underlying pathological cause of the absence of mucus is to be found in the total destruction of the goblet cells.

Important points to remember about the chronic form, which is more frequently seen in debilitated natives, are its intractability and the nature of the stools, which may show no external signs of blood or mucus for many months at a time. The great improbability, amounting sometimes to an impossibility, of isolating a dysentery bacillus from the faeces, though the organism may be present in the intestinal wall and can be found at autopsy, renders the diagnosis of these cases during life a matter of very great difficulty indeed.

Several complications occur in connection with bacillary dysentery. Of these arthritis is the most frequent ; it generally affects one joint alone, but cases have been recorded in which both knees, wrist, fingers and even the temporo-mandibular joint have been involved. It is apparently quite common in some epidemics, and one small series of cases was observed in Egypt in which no less than 27 per cent, developed poly-

74 MEDICAL HISTORY OF THE WAR

arthritis. The joint effusion is ushered in by pyrexia, rarely during the acute stages of the disease, more generally after the tenth day of the disease when the stools have once again become faecal. The cases are usually Shiga infections, though Flexner cases have been recorded by Waller in Mesopotamia. The joint fluid is clear, never purulent, and is usually sterile on culture, though in one instance a culture of Shiga' s bacillus was obtained from the joint fluid by Elworthy. Waller has recorded that it usually contains specific agglutinins for this organism. In the majority of cases the fluid is completely absorbed and no permanent injury to the joint remains, albeit convalescence may be considerably protracted.

General cedema was- noted in Salonika in acute phases of some Shiga infections, in which there appeared to be a flooding of the tissues with dysentery toxins. In late stages also cedema was noted together with the development of ascites. No evidence of a coincident nephritis was obtained.

Conjunctivitis with pain, lachrymation and photophobia must now be regarded as due to the absorption of dysenteric toxins. It is liable to ensue from the 14th to the 34th day of the disease in convalescent cases and appears to have been specially common in Salonika. Iridocyclitis must also be regarded as a complication although a rare one. It bears no characteristic features and usually supervenes during convalescence. It is usually associated with arthritis.

Parotitis, either uni- or bilateral, may supervene, though it is by no means certain whether it can be regarded as a true complication.

Intussusception of the large intestine may occur, though it is more usually found in children.

Collapse may occur early in the illness from toxaemia, or later in the third or fourth week apparently from physical ex- haustion and the draining of fluid from the body by continuous evacuations. The clinician should always be on his guard to forestall, if possible, this serious condition.

Neuritis of one or both legs following bacillary dysentery has been noted in chronic cases. It is doubtful whether the complication is to be ascribed to dysentery toxins or to an in- dependent infection.

The sequelae of bacillary dysentery may be the result of mechanical alterations to the bowel wall, or the direct effect of the absorption of toxins. In the former instance stenosis of the large intestine may occur leading to an obstinate post- dysenteric constipation with painful peristalsis and dyspeptic symptoms.

Tachycardia subsequent to bacillary dysentery was

BACILLARY DYSENTERY 75

frequently observed in men in convalescent camps. It may be ascribed partly to the physical exhaustion this disease entails and partly to a toxic myocarditis. If neglected, or unrecognised, it may even lead to sudden cardiac failure.

Enright and Manson-Bahr have shown that invasion of the blood-stream by Bacillus coli is liable to take place through the chronic bacillary lesions, leading to formation of metastatic abscesses in the kidneys.

Prognosis.

The prognosis in bacillary dysentery depends very much

upon the virulence of the particular epidemic, the age and

• physical condition of the patient. The infection appears to be

specially virulent in those races, who for generations past have

not been exposed to infection.

The prognosis is not good in cases with a subnormal tem- perature, rapid pulse, and a tendency to collapse ; while vomit- ing and persistent hiccough may be regarded as constituting almost invariably fatal signs. In the majority of cases, as regards expectation of life, the prognosis may be considered good, but it is otherwise as regards the permanent injury to the intestinal canal which this disease involves.

A series of 70 cases specially observed in France by Captain H. Letheby Tidy may be quoted here as probably typical of the usual disposal of the patients. The cases fell into three groups :

(1) Evacuated to convalescent depot. 50=71 per cent.

(2) „ „ the United Kingdom 12=17 „ „

(3) Method of evacuation doubtful .. 8=12 „ „

In the cases of the first group the average duration from onset to evacuation to a convalescent depot was 30 days, and the average duration of diarrhoea 8 days, leaving 22 days in hos- pital after cessation of diarrhoea before the men were fit for con- valescent life. The factors which were found to be important in estimating such fitness were the condition of the bowels and the pulse.

A man was considered to have diarrhoea if he had more than two motions daily. Until diarrhoea in this sense had been absent for one week he was not fit for solid diet, and only if one week on solid diet produced no relapse of diarrhoea was he fit for the convalescent depot. Softness of the motions appeared to be of much less importance than their frequency. No cases were sent to the convalescent depot unless the stools had been negative for three consecutive examinations.

In some cases, usually in the fourth week, the pulse became rapid. When this occurred the patient needed a long convales- cence.

76 MEDICAL HISTORY OF THE WAR

The second group comprised all cases which had been classed as dangerous from their general symptoms, and all cases which on the 20th day from onset were still passing four stools daily. All such cases were found to need a long period of treatment and recovered very slowly.

From the military standpoint all cases of the disease, even if apparently mild, should be considered unfit for duty until a microscopic examination of the faeces shows an absence of any inflammatory cells or desquamated epithelium and until com- plete restoration of the digestive functions has been established.

Many clinically severe cases recover completely, while others continue to pass diarrhceic and dysenteric stools, it may be for several years after the initial attack ; in these the destruction of the bowel tissue is progressive and they ultimately end fatally. Cases initially acute with persistent diarrhoea should no longer be considered fit for active service. Those with chronic ulcera- tion of the bowel and continuous passage of mucopurulent stools are most intractable and distressing and should therefore be regarded as entitled to permanent pensions.

It is questionable how far the mucosa can regenerate after such a severe destruction, but undoubtedly many cases, especially in the young and vigorous, completely recover and should be judged upon their general condition. In contra- distinction to the amoebic form, bacillary dysentery is not prone to relapse and need not necessarily be pensionable. But it should be borne in mind that a previous bacillary ulcera- tion undoubtedly predisposes to the development of amoebic colitis. Cases of this nature are being frequently encountered among pensioners who suffered undoubtedly from bacillary dysentery in the first instance, but whose subsequent relapses were due to infection with the Entamceba histolytica.

Bacillary dysentery when complicated with other specific fevers is a dangerous combination ; in Gallipoli it co-existed frequently with paratyphoid fevers and it may be mentioned that subtertian malaria together with bacillary dysentery generally assumes a grave aspect and requires a most vigorous and thorough antimalarial treatment. A grave prognosis should also be given in cases complicated with lobar or broncho- pneumonia.

The average duration of invalidism from dysentery may be gathered from the following tables. In the first, compiled from index cards and admission and discharge books by the Medical Research Council, a series of 3,000 cases of dysentery has been taken from the records of patients treated in military hospitals in France and Gallipoli in 1915, and includes cases both of brief and of long duration. The second table shews a series of 2,000

BACILLARY DYSENTERY

77

cases from France, Salonika, and Egypt during 1917 and 1918, taken from the records of cases treated to a conclusion in the special convalescent depot for dysentery at Barton-on-Sea.

Cases of Dysentery in 1915.

Force from which derived.	No. of Cases.	Total No. of Days under Treatment.	Average No. of Days under Treat- ment.
France Gallipoli Total	681 2,319	28,823 175,365	42-3 75-6
3,000	204,188	68-1

Cases of Dysentery in 1917—1918.

Force from which derived.	No. of Cases.	Total No. of Days under Treatment.	Average No. of Days under Treat- ment.
France Salonika Egypt Total	1,586 330 84	187,666 82,672 12,018	118-3 250-5 143-1
2,000	282,356	141-2

The various forms of dysentery have not been differentiated.

Diagnosis.

Though acute dysentery, occurring in epidemic form in armies in the field, may be justifiably regarded as bacillary dysentery, yet it is always advisable to resort to laboratory diagnosis whenever possible. It was found, however, even in France, where the facilities for scientific work were probably greater than in any other of the theatres of war, that it was quite impossible for bacteriological examination to be applied to all cases admitted to casualty clearing stations. It was accordingly ordered that cases, in which the passage of blood and mucus was observed, should be diagnosed as " clinical dysentery," and that the bacteriologist's labours should be directed first to the cases in which these symptoms were not established. This examination presented many difficulties to the uninitiated. In order to economize in men and material it was advisable to employ as pathologists those who had been especially trained in this branch of work, for besides a knowledge

78 MEDICAL HISTORY OF THE WAR

of bacteriology, a considerable insight into cellular pathology and an intimate acquaintance with the varied protozoological fauna of the intestine are required.

The gross character of the stools passed during different stages of the disease varies considerably and certain rules may be laid down for the guidance of the military clinician, though it must be admitted that exceptions occur. The acute bacillary stool consists of pure blood and mucus, or more accurately " bloody mucus." It is in fact mucus tinged with bright red blood, of extreme viscosity, and tending to adhere to the bottom of the bed-pan or containing vessel. It is odourless or bears a faint smell of spermin. It represents, in fact, an acute inflammatory exudate, derived from the mucosa of the whole or major part of the large intestine. The amoebic stool, from which it is necessary to differentiate it, is composed of blood and faeces intimately mingled, is very offensive, not viscid, and represents the exudate and sloughs derived from ulcers throughout the canal, the dark altered blood being derived from small intermittent haemorrhages at the bases of these ulcers.

The diagnosis of bacillary dysentery can be made sufficiently accurate for all practical purposes by examination of the cellular exudate alone. In military practice what is most required is promptness ; it is necessary to diagnose early in order to save the patient's life. The clinician in a casualty clearing station cannot afford to wait twenty-four hours before applying the appropriate remedy. If a few hours' delay takes place, it may result in irreparable damage to the gut wall, and one cannot restore a once scarred and fibrosed intestine.

A provisional laboratory diagnosis may be made by direct examination of the cellular exudate under the microscope, by recognition of the predominant type of cell and by exclusion of the Entamceba histolytica. As seen under the one-sixth lens the characteristic cellular picture is one composed for the most part of undamaged polymorphonuclear leucocytes. They constitute over 90 per cent, of all the cells in the exudate. Willmore and Shearman have noted that the ringing of the nuclei of these cells is specially distinctive. The large macro- phage cells, which, as previously mentioned, are derived from the submucosa, constitute about 2 per cent, of the cells and are present in the mucus, especially in the early stages of the disease. They are large hyaline cells 20-30 microns in diameter ; sometimes they are round, oval or even bi-lobed in outline and in their protoplasm they contain vacuoles and fatty globules of various shapes and even ingested red cells or leucocytes. The pathologist should make himself familiar with these cells as

BACILLARY DYSENTERY 79

they are extremely liable to be mistaken for Entamceba his- tolytica and consequently lead to a mistaken diagnosis.

The following are the points which require attention, in order to avoid those mistakes in diagnosis which frequently occurred. In bacillary dysentery the macrophage cells are defunct and consequently non-motile ; they are by no means as refractile as is the Entamceba histolytica and their protoplasm has a bluish ground glass appearance. The characteristic endothelial nucleus can seldom be made out as it is usually undergoing chromatolysis. Columnar epithelial cells are frequently present with the macrophage cells and, in the later stages of the disease, intestinal protozoa such as Entamceba coli and intestinal flagel- lates ( Trichomonas and Chilomastix) may make their appearance. The Entamceba coli is specially liable to cause a fallacy in diagnosis, for, unless the pathologist is familiar with the morpho- logical characters of the non-pathogenic amoebae, a mistaken diagnosis, or even a suspicion of a double infection with the two main forms of dysentery, may arise.

For successful laboratory diagnosis it is essential that the stool should be fresh and passed early in the course of the disease.* It should, if possible, be collected in a bed-pan without admixture of urine and brought straight to the laboratory. On no account should the selection of a portion of the stool suitable for examination be left