158

CHOLERABy PROFESSOR BRIAN MAEGRAITH, M.A., M.B., M.R.C.P., D.PHIL.

School of Tropical Medicine, Liverpool

DefinitionCholera is an acute, self-limiting, often fatal

infectious disease of short duration caused by aspecific organism, Vibrio cholerae, which multipliesin the gut contents but does not invade the bloodstream or tissues. It is characterized by copiouswatery diarrhoea, vomiting, muscle cramps, severedehydration, vascular collapse and various com-plications, especially suppression of urine andacute uraemia.

DistributionCholera has a strictly endemic distribution in

certain tropical regions of high humidity andtemperature, namely Bengal and Madras in Indiaand the Yangtse valley in China. The mostimportant endemic focus is that in Bengal, whichis confined mainly to localities along the RiverHooghly. Other suspected but not confirmedendemic foci are situated in Burma and thePhilippines.

Local epidemic or pandemic extensions of thedisease have occurred from time to time alongtrade routes and other lines of communication.

AetiologyThe causative organism Vibrio cholerae belongs

to a group of bacteria which are morphologicallysimilar to, but antigenically and biochemicallydistinct from, other vibrios. The cholera vibriois small, comma-shaped and motile. It is gramnegative and grows easily at 37°C. in ordinarybacteriological media. Vibrios are divided intogroups depending upon their antigenic pattern.Those known to cause cholera have a common Hantigen and specific 0 antigens. There are threeprincipal strains, i.e. the Inaba, Ogawa and Hiko-jima strains. One of these usually predominatesin outbreaks, but sometimes the prevalent antigenicstrain changes.A fourth strain, called the El Tor vibrio,

appears, under some circumstances, to be able toproduce the disease. It is distinguished from theother cholera vibrios by its haemolytic action invitro.

In nature the organism is pathogenic only toman. Other animals are not infected. A con-

dition somewhat resembling cholera can, how-ever, be induced by artificial gut infection ofyoung guinea-pigs and rabbits.

In the human case the vibrio grows and multi-plies almost entirely in the lumen of the gut. Itdoes not penetrate beyond the submucosa of theintestinal wall and is never found in the bloodstream or in the urine. In the clinical attack it ispresent in enormous numbers in the faeces andvomitus.

In most cases the vibrio may be found in thefaeces and vomitus for about five days only.Occasionally it may persist for as long as a fewweeks. There are no true "carriers" of thedisease in whom the infection persists for longperiods.

It appears, therefore, that the maintenance ofendemicity in a given area must depend on thedirect spread of infection from case to case, themode of transmission being mainly through faeces.Individuals may become infected without showingany clinical evidence of the disease. It is thus notnecessary to have overt cases for transmission tooccur.The organism can survive on moist clothing for

up to three days. It dies rapidly in pure water,but survives for days (maximum of six weeks) inslightly dirty water containing salts and organicmatter. It is easily killed by moderate heat(55°C. for i hr.) and acid.

in a community the infection spreads mostcommonly through infected water (including iceand cola drinks) contaminated with faeces. Otherpotable fluids including milk, cold cooked foods,vegetables sprinkled with water and uncookedfruits may also be concerned with the spread ofthe infection. Spread may occur from case tocase through direct contact with faeces or vomitus.The only important living agent of transmissionis the domestic fly.

In endemic areas most cases occur in the hotmoist season, possibly because of the mechanicalflushing of local filthy water supplies. Thus theincidence is highest in Bengal in the early rains(May, June and July) and lowest in the dryweather.

Epidemics arise from the introduction of the

copyright. on A

ugust 30, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.28.317.158 on 1 M

arch 1952. Dow

nloaded from

March 1952 MAEGRAITH: Cholera I59

vibrio by infected individuals, who may or maynot show clinical signs of infection. Once thedisease is introduced it spreads by the same meansas in the endemic area. Where sanitary conditionsare good there is little chance of spread. Wherethey are so bad that water supplies can be con-tinually infected, cholera becomes rapidly estab-lished. Since the individual case is usuallyinfective for dnly a few days the extension of thedisease from endemic areas is largely limited bythe rate of travel. Modem transport is a potential.threat to insanitary areas outside the endemic foci,and rigid international control is necessary to keepthe disease confined.

In non-endemic areas prevention of the importof the disease is a matter of sanitary control andquarantine of ships and aircraft from infectedareas and the isolation of suspected cases. Wherecholera has appeared local water supplies must beexamined and protected from pollution. Chlorina-tion will rapidly destroy the vibrio. Individualdrinking and washing supplies should be boiledor chlorinated before use, and the strictest atten-tion must be paid to the preparation and consump-tion of food. Anti-fly precautions should beenforced.

Individual or mass protection by the use ofvaccines of dead vibrios, preferably local strains,should be carried out if possible. Individualinjection is given in two doses (first dose 0.5 ml.and second dose i ml.) a week apart. Whenvaccination is used for mass protection, forexample during an epidemic, a single dose of i ml.is employed.Modern vaccines do not usually produce febrile

reactions. They endow some immunity, which iseffective for four to six months. Repeated vac-cination is necessary in individuals residing forlonger periods in infected areas.

Resistance to InfectionIn a group of individuals exposed to the same

source of infection by no means all may becomeinfected. Of those infected some will show thefully developed syndrome, others may have noclinical ill-effects. It is believed that certainindividuals may thus present some natural resis-tance to infection. The acidity of the gastric juicemay act as a barrier to infection of the gut, thevibrios being highly susceptible to an acid environ-ment; reduction in gastric acidity may prediposeto infection.Newcomers to the endemic area are believed to

be more likely to become infected, but this is notcertain. There is little evidence of herd immunityin the sense that it exists in malaria; the localpopulation of an endemic area may be highlysusceptible during outbreaks.

The cholera outbreak in a community is self-limited. It tends to die out after reaching its peak.This may be in some measure due to the largenumber of subclinical cases which develop, or tothe acquisition of temporary individual immunity.It has been demonstrated, however, that one attackdoes not protect against subsequent infection,except partially for a very limited period.

In epidemics a smaller proportion of the localpopulations may be infected than might be anti-cipated. Napier, for instance, has pointed out thestriking difference between the infection rate ofcholera (about one in three) and smallpox, inan exposed community in which there is freeintercommunication.

Antibodies, including agglutinins, and somemeasure of protection against infection are pro-duced in animals by the inoculation of killed vibriocultures intramuscularly or subcutaneously. Suchinoculations may afford considerable protection inman for a few months at a time, certainly notmore than six months. It is said that inoculationwith the local strain is most efficacious.

Race, sex and age appear to play little part inthe incidence of the disease. Malnutrition andpoor health probably predispose to infection, butotherwise healthy subjects are often readilyinfected.

PathologyPathogenesis

In cholera the organisms remain in the gut anddo not invade the blood stream. In the gut thevibrios multiply rapidly and cause the loss ofenormous quantities of water and salt from thetissues, throughl the intestinal epithelium into thelumen and so outside the body. This is the basicphysiological lesion. Certain local effects on thegut wall may be produced. There is often con-siderable shredding of the epithelium and in thelater stages some bleeding into the lumen. Thereis never any deep ulceration of the intestinal wall.The mode of action of the infection is not

understood. No soluble toxin has been identified.It has been shown, however, that poisonous sub-stances or endotoxins are formed in vitro duringthe growth and lysis of the vibrio, and it may bethat these are involved. The endotoxin has beenshown in animals to affect the permeability of thegut wall to electrolytes. Its effect can beneutralized to some extent by antibodies derivedfrom injection of killed cultures of cholera vibrios.It is probable that the whole syndrome arises as aresult of the action of a vibrio or its products suchas this ' endotoxin ' on the physiological per-meability of the intestinal wall.The loss of fluid and electrolytes is rapid and

severe. A state of mixed salt and water dehydra-

copyright. on A

ugust 30, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.28.317.158 on 1 M

arch 1952. Dow

nloaded from

i6o POSTGRADUATE MEDICAL JOURNAL March 1952

tion is quickly achieved. Relatively greater lossesof sodium than chloride occur owing to the pre-ponderance of fluid loss from the gut. Somedegree of acidosis and haemoconcentration results.The serious dehydration itself affects the circu-

lating plasma volume, which is often further re-duced by the appearance of vascular collapse.Haemoconcentration is therefore pronounced insevere cases. The viscosity of the bloodincreases considerably and the efficiency of thecirculation is correspondingly diminished. Theappearance of vascular collapse accentuates thecirculatory difficulties. The combination ofdehydration and shock bring about functional andstructural tissue' damage, particularly in thekidney.Morbid AnatomyThe information regarding the morbid anatomy

of cholera is surprisingly incomplete. The tissuechanges are basically non-specific and the lesionsin the internal organs appear to be those whicharise from the prevailing' dehydration and/orvascular collapse.

Rigor mortis develops rapidly. The musclesare deep red and dehydrated; violent post-mortemcontractions may occur. The tissues, are dry. Theblood is viscid. There may be scattered petechialhaemorrhages in the mucous membrane of theintestines and in the pericardium.

Changes in the organs are commoner in caseswhich survived to the late stages before death.The anuric and uraemic case may present kidneylesions similar to those often met in other examplesof the renal anoxia syndrome. There may be ir-regular ischaemia of the cortex involving theglomeruli in a patchy manner, some medullarycongestion and epithelial degeneration anddesquamation evident particularly in the corticaltubules which, together with the collectingtubules may contain casts of albuminous materialand epithelial debris. On the other hand, theremay be little evidence of structural change in thekidneys of the anuric case, especially if death hasoccurred soon afler the renal failure. The livermay be congested and show some degenerativelesions, mainly centrilobular. The gall bladderand bile ducts :Lre filled with dark viscid in-spissated bile.Pulmonary oedema may be present in shocked

cases. Otherw'se the lungs are shrunken andanaemic.

Clinical PathologyBlood CellsThe dehydration and loss of plasma volume

cause rapid haemoconcentration. In severe casesthe blood is viscid and the erythrocyte count may

become as high as 8 or 9 million cells per c.mm.The white cell count is correspondingly increased.

The BloodThe viscosity and specific gravity is increased

roughly in proportion to the loss of plasma volume.The specific gravity of normal blood measured bythe method described in the section on treatmentis about 1054. In severe cases at the height of'fluid loss' it may be 0o60 or even more.The chemical constituents of the blood depend

primarily upon the state of dehydration and thesuccessful function or otherwise of the kidneys.In the severe case the total chloride concentrationis low; the sodium is lower in proportion; thepotassium is usually unchanged.The blood urea N concentration is raised above

the normal range in most cases. In the anuriccase it rises steadily to reach very high figures. Inrecovery after anuria it falls rapidly.UrineIn the dehydrated case the output is low.

There may be no urinary secretion. The specificgravity may be high and the urine deeply pig-mented. Nevertheless, the urea and electrolytecontent is low and there may be no chloride orsodium. Albumin and casts are present in theacute attack. After recovery the electrolytes re-turn rapidly, but owing to the slow recovery of therenal epithelium it may be some time before theconcentration of the urine is re-established.

Clinical PictureCholera may be mild or severe. Cases have

been described in which the infection has been sosevere that death has resulted before the classicalwatery diarrhoea has become established. On theother hand, infection of the intestinal contentsmay be present without causing any clinical signs.In epidemics it is often the case that early and latecases are mild, and those at the height of theepidemic most severe.The majority of clinically overt cases are severe.

Nevertheless, the condition is of short duration,seldom lasting more than five days.Incubation PeriodThe incubation period varies from a few hours

to five days. Commonly it is about three days.There are no prodromal symptoms.

The Classical AttackIt is usual to describe the development of a

severe case in three stages, which often mergeindistinguishably into one another. These stagesare those of (i) evacuation, (2) collapse, and (3)reaction.

copyright. on A

ugust 30, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.28.317.158 on 1 M

arch 1952. Dow

nloaded from

March 1952 MAEGRAITH: Cholera 161

The stage of evacuation commences withdiarrhoea which may at first be mild, but whichsoon empties the bowel of faeces and changes tothe urgent watery diarrhoea of the classical con-dition. The patient now passes frequent waterystools which amount to little more than largequantities of clear or slightly opalescent, non-offensive liquid containing practically no faeces.Against a dark background little flecks of mucosacan often be seen floating about. This appearancehas led to the descriptive name of ' rice waterstools.' The stools swarm with vibrios. In thelate stages there may be a little blood.

Bowel motions are frequent, effortless and un-controlled. Napier refers to them as the painlesspassage of pints of pale fluid. The stool comesout in spurts, sometimes with considerable force.The general appearance is that of the turning onand off a faucet. There is no pain and no colic,and the patient is often scarcely aware that he isevacuating his bowel. Contamination of bedclothing is therefore frequent unless watched for.Vomiting begins as a rule shortly after the

diarrhoea has started. There is no nausea. Thepatient has little or no control over the vomitus,which gushes out with considerable force andvolume. The watery vomit is essentially similarin appearance and content to the stool, and likethe latter contains enormous numbers of vibriosand is highly infective. It constitutes a realdanger to the unwary physician.

Evacuation continues for a variable time andfrequently persists into the second stage of thedisease, that of collapse. This algid stage may bereached in a few hours in acute cases and in oneor more days in less severe attacks. Evacuationseldom contirxues, however, for more than threedays. The fr quent bowel evacuation and vomit-ing lead rapi ly to tremendous loss of fluid andelectrolytes. Within a matter of hours thesevere'y ill patient becomes dehydrated. Thewhole body so .ms to shrink; sub-cutaneous fluid islost and the pale, clammy skin becomes inelasticand stretchedI over the underlying tissues. Theeyes are sunken, the cheeks hollow, the skin tightover the malar prominences. The mouth andtongue are dry; there is extreme thirst; the voiceis husky. The patient becomes anxious, fore-boding and restless but remains mentally clear.As the dehydration continues, the circulation be-comes inefficient. The blood pressures fall, thepulse quickens and may be impalpable at the wrist.The picture now becomes essentially one of de-hydration and vascular collapse and closely re-sembles similar conditions, such as severe heatexhaustion, in which medical shock has developed.Muscle cramps are common once dehydration

has become evident. They are severe and painful,

frequent, of short duration and arise particularlyin the legs. The abdominal muscles are affectedonly in the later stages. The rectal temperature isseldom raised above normal. Skin and oral tem-peratures are often subnormal.At the start of thesyndrome the urine volume is

reduced and there is usually a thin cloud ofalbumin; as the dehydration proceeds the urinaryoutput diminishes. Even in relatively mild casesthere is oliguria. In severe cases there may becomplete urinary suppression, which may be onlytemporary or may pass on to irreversible acuteuraemia. The urine has a low electrolyte con-centration. It contains albumin and granulartubular casts and may have a high specific gravityand pigment cc ntent.By the time dehydration is manifest the plasma

volume is considerably reduced. When vascularcollapse supervenes there is a further reduction.The result is an extreme loss of plasma volumewhich may amount to over half the initial volumeand a corresponding concentration of the cellularelements of the blood. The red cell count andhaemoglobin concentration rise steadily to amaximum which may be greatly in excess ofnormal. The viscosity of the blood is increased.Death is common at this point from the com-

bined effects of dehydration and shock.The fate of the patient is determined to a con-

siderable extent by the degree and duration ofevacuation and collapse, i.e. by the degree andduration of dehydration and vascular failure. Ifthe algid state has lasted only a few hours, thethird stage, that of circulatory recovery is com-monly followed by rapid return to health. Re-action is ushered in by a rise of blood pressure,slowing and improvement of volume of the pulse,a return of normal coloration to the skin and arise of body temperature sometimes to abovenormal and occasionally to hyperpyrexial levels.Evacuation and vomiting, if not already stoppedduring the stage of collapse, now cease.

If the algid state has been prolonged or verysevere, there may be no circulatory recovery or atemporary reaction stage may be followed byserious and sometimes irreversible changes invital organ function and the threat to life againbecomes urgent.

This is particularly so in cases in which anuriahas developed in collapse. If the anuria has beenof very short duration reaction may be im-mediately followed by recovery of urinary flow andrenal function. On the other hand, however, insome cases, especially those in which the anuriahas been prolonged, renal failure may persist intoacute uraemia. Such cases usually die within afew days; even in their late stages, however, re-covery may occur, as it does in other examples of

copyright. on A

ugust 30, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.28.317.158 on 1 M

arch 1952. Dow

nloaded from

I62 POSTGRADUATE MEDICAL JOURNAL March I952

the renal anoxia syndrome. Occasionally theanuria of the collapse may be followed by a shortperiod of oliguria in which small volumes ofurine containing albumin and casts may be passed;anuria then reappears and the patient dies in a fewdays in uraemia.

In some cases after brief improvement, theperipheral circulation may fail again and thepatient dies of shock. It must be noted that inmany fatal cases, the algid stage persists to deathand there is no recovery stage at all.The whole progress of the cholera case is only

a matter of a few days at the most. Recovery when'it occurs is usually rapid, especially with correcttreatment, when it is often remarkable, the de-hydrated corpse-like patient visibly swelling intosomething resembling normality almost while heis being watched.

Course and PrognosisThe so-called stages may completely merge in

severe cases. In milder cases there may be aninterval of recovery after the algid stage followedin turn by the complications of the reaction stage,especially renal or circulatory failure or both.The very severe case may perish literally in a fewhours. This is especially so in children.

Prognosis depends considerably on the lengthof time elapsing from onset to commencement oftreatment. If dehydration is very severe and shockdevelops, the outlook is bad in the individualsubject. The longer the dehydrated patient is leftuntreated, the worse the prognosis. Prognosis isbad if anuria has already developed and persistedfor some hours before treatment.The recovery rate is high with efficient treat-

ment, except in advanced and shocked patients.Complications of cholera other than renal

failure and shock, are rare. Pneumonia is some-times described, especially in outbreaks in coldclimates, and gangrene of the extremities has beenreported in a few neglected cases. There are nosequelae in the recovered case. Once the tissuewater-salt balance and plasma volume have beenadjusted, return to normal is very fast, usually amatter of a few days.The death rate in local outbreaks or in epidemics

is often very high. For instance, it was about 50per cent. at the height of the recent outbreak inEgypt. Even under epidemic conditions, how-ever, the death rate of patients treated in hospital,who must be regarded as a highly selected popula-tion, is usually not much greater than io per cent.

In any outbreak there are many mild cases whichrecover spontaneously or respond remarkablyquickly to treatment, and there are probably manymore cases without symptoms, even of milddiarrhoea.

DiagnosisIn an outbreak the clinical diagnosis of the

individual case is easy. Doubtful cases must betreated as cholera. Any condition leading toacute dehydration with watery diarrhoea andvomiting may be mistaken for cholera. Choleraicand algid pernicious malaria, acute food orchemical poisoning and heat exhaustion are allexamples of related clinical pictures.The diagnosis of an isolated case may be

difficult. The presence of vibrios-may be con-firmed in wet or stained stool preparations.Further identification of the organism is essential.The vibrio can often be isolated by inoculation ofspecimens of faeces into alkaline peptone waterand incubation at 37° C. for six to eight hours.The ordinary faecal organisms are partly in-hibited by the alkalinity hnd the vibrio becomesconcentrated at the surface of the medium.Further bacteriological identification is thensimplified.TreatmentTreatment of established cholera is essentially a

matter of non-specific measures for restoring thebiochemical balance of the body and the plasmavolume. Specific measures designed to destroythe vibrios or neutralize hypothetical poisonousproducts are of secondary importance.Non-Specific MeasuresThe fluid and salts lost by evacuation must be

replaced as rapidly as possible.The substances missing are essentially water,

sodium and chloride. The loss of sodium ex-ceeds that of chloride. It is possibly for thisreason that the use of hypertonic saline is ap-parently more effective than isotonic saline in theinitial stages of treatment. In shocked cases theplasma volume must be restored immediately andfor this purpose parenteral plasma injection isprobably better than saline.

Treatment of the severe case is a matter ofparenteral replacement of fluid and salts. Thefollowing details are modified from those followedin the Campbell Hospital in Calcutta:-

Solutions:

(a) Hypertonic salineSodium chloride 140 gr. (i6 gm.).Pyrogen-free distilled water i pt. (I 1.).

(b) Alkaline salineSodium chloride 80 gr. (9.o gm.).Sodium bicarbonate I80 gr. (20.5 gm.).Pyrogen-free distilled water i pt. (i 1.).

copyright. on A

ugust 30, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.28.317.158 on 1 M

arch 1952. Dow

nloaded from

March 1952 MAEGRAITH: Cholera 163

MethodIntravenous injection of saline is given to all

dehydrated or collapsed patients. The first pintof hypertonic saline (a) should be administeredimmediately the patient is admitted. Subsequentdosage may be calculated by clinical assessment ofthe degree of dehydration or shock or, if possible,by estimating the specific gravity of the blood,which is raised as the result of prevailing haemo-concentration.

If the systolic pressure is 80 mm. Hg or less,or if the specific gravity of the blood is 1058 toIo6o, i pt. of hypertonic saline are needed;specific gravity Io60 to Io62, 2pt.; over io62, 3 pt.The specific gravity is calculated as follows:

Blood is taken from the finger into a pipette. Adrop of blood is extruded from the pipette belowthe surface of the contents of a series of mixturesof glycerin and water of known specific gravity,ranging from 1054 to Io64 and starting from thehighest. The specific gravity of the blood is takenas being equal to that of the fluid in which thedrop of blood remains suspended where it wasdischarged.

Shock is counteracted by injecting plasma afterthe first injection of hypertonic saline.

It may be necessary in the collapsed patient tocut down on the vein. 'Otherwise the infusion isgiven through a wide bore needle into the cubitalor antecubital vein.

Rate of AdministrationThe first pint of hypertonic saline is given very

quickly, i.e. in about five minutes. The secondpint is given more slowly, in about 20 minutes.In a big man it may be necessary to give a furtherpint in about 30 minutes, but usually after thesecond pint the rate of administration is slowed toabout a pint in four hours.

In the first 24 hours the proportion of hyper-tonic to isotonic saline given should be about 2:1.Some authors advise continuing the hypertonicsaline on the following day in the ratio of i :2 toisotonic, but this is seldom necessary and in anycase the injection of saline must be carried out withcaution after the first phase of dehydration hasbeen adjusted.

Acidosis associated with the reduction in fixedbase may be counteracted by injection of thealkaline saline or of bicarbonate solution (I20 gr.to the pint), but the injection of saline is often initself sufficient to adjust the electrolyte balance.Check should be kept on the urinary chloride

concentration during treatment. Once the chloridecontent has been re-established, hypotonic salinemay be substituted for isotonic. The hypotonicsaline is made up by mixing isotonic saline andisotonic glucose in the proportion of I :i or i :2.

Infusion of fluid usually brings about rapid re-covery within a few hours. It is seldom necessaryto continueit for more than 24 horss, but in casesin which evacuation or shock persist or reappearit may be. cessary to continue longer or repeat.One of:the difficulties encountered in parenteral

treatment is the development of rigors arisingfrom pyrogens in the solutions used., .i-t may benecessary to stop Ithe infusion temporarily duringrigors. Cases which originally present with slightfever are more prone to pyrogenic reactions.These may often be the only signof fever duringthe attack. Infusion of saline must not be ex-cessive. An input/output account of fluid shouldbe kept in all patients, and after dehydration hasbeen adjusted the balance should be kept. It maybe fatal to overload the patient with fluid, especiallyif he is anunc.

Specific TreatmentDestruction of the vibrio can be effected to

some extent by the use of specific bacteriophage orsulphonamides. Other methods have been recom-mended from time to time with indifferent success.There is considerable doubt about the efficacy ofphage. At present the best approach to specifictreatment seems to be the use of the relatively in-soluble sulphonamides, such as sulphaguanidine,given in large doses corresponding to those ad-ministered in bacillary dysentery. Drug therapymust always be combined with the non-specifictherapy detailed above.

Dosageo. Io gm. per kilo. body weight given immediately.

Follo w by 0.05 gm. per kilo. body weight everyfour hours. The difficulty may be to administerthe drug against the tide of vomiting.Symptomatic Treatment

Infusion of saline will usually bring about im-mense relief in all symptoms including musclecramps, which may be very violent. Severe painmay be relieved by self-administered chloro-form on a handkerchief. Most authors adviseagainst the use of morphia. Atropine, I/75 gr.,is sometimes recommended in the early stages.The patient should be persuaded to take fluid bymouth if possible. Such fluid should containglucose. Careful nursing is essential. In theconvalescent stage the patient must be kept-quietto avoid the recurrence of vascular collapse.Acknowledgments

I am indebted to Prof. Chaudhuri, Director ofthe Calcutta School of Tropical Medicine and toProf. L. Everard Napier for much valuableinformation.

copyright. on A

ugust 30, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.28.317.158 on 1 M

arch 1952. Dow

nloaded from