368 POST GRADUATE MEDICAL JOURNAL July 1948

be considerably delayed. Thus, a normal chloridelevel is not incompatible with the diagnosis oftuberculous meningitis, for the level may becomeappreciably reduced only at an advanced stage inthe disease process.

Table 2 summarizes the important changes inthe cerebro-spinal fluid. It will be seen thatvariations between the cellular and chemicalresponses are of great value in diagnosis, thoughfor ultimate identification additional investigationsusually prove necessary. The differences incellular and chemical responses also serve to em-phasize the classification of the causes of meningealirritation pr-vio,usly advanced on an aetiologicalbasis.

Differentiation of Purulent MeningitidesThe optimal treatment of purulent meningitis

depends on precise identification of the causal or-ganism. The species and type can be determinedonly by accurate bacteriological and, where

necessary, immunological methods. In the largemajority of cases an immediate result can beobtained by staining, either of a direct smear or ofone made after rapid centrifuging of the cerebro-spinal fluid. With skilled technique and ex-perience the culture of organisms from the fluid inpurulent meningitis is more frequently con-firmatory than diagnostic. The severity of the in-fection can rapidly be gauged by a simple quantita-tive test for glucose concentration.The identification of haemophilus influenzae by

the above methods is less frequently successful,since its morphological features are unpredictablyvariable. In difficult cases the capsular swellingphenomenon may be employed.

BIBLIOGRAPHY

COHEN, H. (1927), Brain, L., 6o0.FREMONT-SMITH, F., DAILEY, M. E., and THOMAS,

G. W. (I928-29), J. Clin. Invest., 6, 9.M. R. C. (1948), Streptomycin in Tuberculosis Trials Committee,

Lancet, April 17.

INFANTILE DIARRHOEA AND VOMITINGBy IAN M. ANDERSON, M.D. (Glasgow), M.R.C.P. (London)

Assistant Paediatrician, Westminster Hospital

Infantile diarrhoea is a disease of infancycharacterized by vomiting, frequent waterymotions, rapid dehydration, loss of weight and ahigh mortality rate.

Diarrhoea and vomiting has always been amajor problem to those physicians who have hadto deal with diseases of children. Epidemics havebeen reported since early times. In England aslong ago as the mid I7th century epidemics ofsuch magnitude occurred as to lead to the deathof 2,000 babies during the hot summer weather ofI669-71 (Garrison).' Carter2 in I893 drewattention to the infective nature of the disease andnoted its relationship to impure milk; hesuggested the use of subcutaneous saline. Wilson3(1927) discussed the bacteriology and pointed outthat the disease could be caused by manyorganisms.

Until the second quarter of the present centuryinfantile diarrhoea and vomiting was synonymouswith cholera infantum and the main incidence ofthe disease was in the summer months. Eventuallyhowever, investigations showed that contaminatedmilk and poor hygienic conditions were the most

important factors contributing to the seasonalincidence. With improvement in the milk supply,including the use of pasteurized and dried milksand improvement in sanitation associated with thedisappearance of the horse, the extermination offlies and the establishment of Welfare Centreswhere infant hygiene could be taught to themothers, there was no longer the same seasonalincidence nor mortality. It is probable that acertain amount of infantile diarrhoea and vomitinghas always occurred throughout the year and thatwhen the incidence of summer diarrhoea fell belowthe level of the all-season diarrhoea between 910oand 1930 this latter attracted attention. Marriott4considered that otitis media and mastoiditis com-monly caused infantile diarrhoea and vomitingand following on his papers several others ap-deared claiming success in treatment by mas-toidectomy. Findlay5 (I932) discussed the re-lationship of otitis media to infantile diarrhoea andvomiting and mentioned the social and seasonalincidence. Craig6 (I935) drew attention to thedisease as it occurred in the newborn and apartfrom summer diarrhoea and Dick,7 Barenberg8 and

copyright. on A

pril 18, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.24.273.368 on 1 July 1948. D

ownloaded from

ANDERSON: Infantile Diarrhoea and Vomiting

Frant9 also drew attention to the disease in thenewborn with its highly infectious nature. Frantconsidered the disease in the newborn as a specificillness and gave as his criteria fof diagnosisapparent limitation to babies in the first month oflife, apparent absence of parenteral disease andnon-identification with any known organisms.

In most countries cases have been encounteredin maternity hospitals, at home and in children'shospitals at varying times in the first year of life.It is generally felt that the disease may occur uptill the end of the first year, but that it is pre-ponderar)tly an illness of early infancy and that60 to 80 per cent. of cases occur before nine monthsof age.ClassificationAmongst numerous classifications, that of

Campbell and Cunningham seems adequate.They divide the causative factors into:-

I. Specific. Infective.2. Symptomatic of general diseases or due to

local infection.3. Dietetic.4. Alimentary-infectious complex. (a) Associated

with parenteral infection; (b) without evidenceof parenteral infection.These various groups will be discussed in detail.

i. Specific. Infective.(a) Known pathogenic bacteria. Many or-

ganisms come into this category including theDysentery Group of Sonne, Flexner and Shiga,the Salmonella Group, B. Pyocyaneus, etc. Inthe Dysentery Group the last outbreak reportedin infants was that by Watt°1 (I945) who describedan epidemic due to Shiga. McKinlay1 (I937) re-ported an outbreak due to Salmonella in which anurse was the possible carrier. Watt (1945)described an epidemic due to S. Typhimurium,and Seligmann 12(1947) reported a similar epidemic.Ensign and Hunterl3 (I946) reported an epidemicdue to P. aeroginosa (B. Pyocyaneus). Felsen14(1942) and Brown15 (1946) reported epidemics dueto Staphylococcus Aureus.

(b) Bacteria whose aetiological relationship issuspect but not known. (i) Coliform Group.-Crowley, Downie, Fulton and Wilson16 (1941)from their investigations in association with threeepidemics of diarrhoea in the newborn found thatthe proportion of coliform organisms in the stoolswas increased in those infants suffering from thedisease and that they were found in much thesame proportions in the faeces of healthy contactinfants in the same outbreaks, while they werepresent in much smaller numbers in the faeces ofhealthy, non-contact infants in normal maternityhomes. They came to the conclusion that it wasimpossible to say whether they were the primary

cause of the disease or whether their increase wassecondary to a disturbance of the bowel caused bysome other agent. Payling Wright17 (1948)suggested that through the cessation of breastfeeding and the resulting disappearance of aciduricflora from the intestines, the infant may becomeunduly vulnerable to infection with any of themore pathogenic strains of coliform organismsthat may happen to be prevalent in the locality atthe time.

(ii) Paracolon Group.-Neter and Clark18 sum-marize the knowledge of the pathogenicity of theparacolon group and consider that:-

I. Organisms may be merely saprophytes.2. May occur as harmless organisms in

diarrhoeal disease caused by enteric pathogens orother incitants.

3. They may be a primary cause of diarrhoealdisease in Susreptible individuals.They state4 that there was fairly conclusive

evidence that these organisms may cause gastro-enteritis (lab. infection) but they felt that furtherstudies were necessary before it was possible toestablish or disprove the etiological significance ofparacolon organisms in enteral disease. McClure19and Baker20 describe epidemics where P.Fscherichia was found in a high proportion ofcases.

(iii) B. Proteus.-Both P. Morgagni and P.Vulgaris have been incriminated as being thecausative agents in infective diarrhoea. James 21(1947) described an epidemic due to P. Vulgaris.This work awaits confirmation in view of thetendency for these organisms to follow in the wakeof a previous specific infection.2. Symptomatic of general diseases

Acute illnesses such as acute pyelitis, meningitisand pneumonia may present with diarrhoealsymptoms and in the same way it may be the mainevidence of chronic infections of a similar nature.

3. DieteticThe symptoms are usually more chronic in

nature and the motions are usually pale ol frothy.4. Alimentary infectious complex

(a) Associated with parenteral infection.-Find-lay (1932) disagreed with the American and Frenchworkers who stressed the relationship betweenotitis media and gastro-enteritis and stated that inhis opinion the ear was not commonly a causativefactor. He quoted Finkelstein as being emphaticthat otitis media was not the cause of gastro-enteritis. Since then various authors in thiscountry have expressed their views as to thecausal relationship. Cooper22 (1937) found 40 percent of cases with parenteral infection, Smellie23

July I948 369

copyright. on A

pril 18, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.24.273.368 on 1 July 1948. D

ownloaded from

POST GRADUATE MEDICAL JOURNAL

(I939) 30 per cent., Campbell24 (1941) I3 per cent.,Alexander and Eiser25 (I944) 88.6 per cent.,Gairdner26 (I945) 34 per cent. It is of interestthat, as the ears have been incriminated morethan any other focus, Patterson and Smith27(I944) reported a study of 120 necropsies ininfants and children and found pus in the'mastoidcavities in 71 per cent. and that although thehighest incidence was in patients with gastro-enteritis it was found almost as often in a widevariety of conditions. Culture of the pus revealedvarious organisms including Pneumococci, Strep.haemolyticus, Strep. viridans, Staph. pyogenes,B. Coli, etc.The subject of the relationship between infantile

diarrhoea and vomiting and otitis media andmastoiditis is a vexed one and it is further com-plicated by the fact that there is considerabledisagreement amongst paediatricians as to what ismeant by an infected ear. Few paediatricians havenot had the experience of asking the E.N.T.specialist to operate on what they thought was aninfected ear and finding no infection present.Furthermore, as commented upon by Wyllie,28the middle ear at post mortem usually containssome inspissated matter. From my own experienceI feel that while otitis media and mastoiditis mayundoubtedly cause infantile diarrhoea and vomit-ing, the number of cases in which this occurs isnot over 20 per cent., and probably lower.

(b) Without evidence of parenteral infection.-Barenberg, Levy and Grand (1936) suggestedthat the disease might possibly be caused by a virusand since then various papers have been writtensuggesting the possibility, and more recently inAmerica attempting to prove this aetiology.Lyon and Folsom29 (I94I) described three

epidemics occurring at the same time as an in-fluenzal epidemic was present in the community,and they claimed that convalescent serum frominfluenza patients had a preventative effect.Light and Hodes 30 (I943) in each of four separateepidemics isolated a filtrable agent which, whengiven nasally to young calves, produced a bloodydiarrhoea in two to five days which lasted up tothree weeks. With each of the four strainssuccessive calf passages were made and after re-covery immunity was conferred. Furthermore theserum from six infants after convalescence fromthe disease conferred complete or partial protectionagainst nasal injection. The material had eitherto be boiled for ten minutes, or to be exposed to atemperature of 80° C. for an hour, to be in-activated. Cummings31 (I947) claims to have dis-covered a similar virus but he could not maintainits potency beyond the fifth calf transfer. Sevitt32(1948) criticizes the work of Light and Hodes be-

cause the agent withstood boiling and becauseDeeney in Dublin could not obtain similar resultswith similar experiments.

Buddingli and Dodd33 in an epidemic ofdiarrhoea and stomatitis obtained positive re-actions in rabbits' cornea but Cummings, in anattempt to repeat their work, found it possible toproduce similar lesions to those described byBuddingh and Dodd by traumatizing cornea,traumatizing and applying alundum, and byapplying extracts of stools and mouth washings ofnormal infants and various bacteria, and he con-cluded that the scarification test was not specific.PathologyOne of the striking features of the disease,

except in those cases where there is an accom-panying acute infection such as otitis media,pyelitis, etc., is the wide discrepancy between theseverity of the illness and the findings at autopsy.There may be some congestion of the surfacevessels of the gut with marked distension of thelarge and small intestines. The mesentericlymph nodes are sometimes moderately enlarged.In the more chronic cases the liver may show fattyinfiltration, fibrosis and necrosis and the kidneys,spleen and heart may show parenchymatous de-generation. In cases of long standing, evidence ofsecondary infection may be found, such as otitismedia, bronchopneumonia, mastoiditis, cranialsinus thrombosis, etc.

EpidemiologyThe epidemiology of the disease is still in-

determinate but the American workers Buddinghand Dodd have shown that a causative agent couldbe isolated from the mouths of doctors and nursesand that it could be transmitted from infant toinfant by these carriers.That the disease is highly infectious is obvious

to anyone who has dealt with an epidemic of thedisease. It may spread rapidly from infant toinfant when it occurs in a hospital ward. PaylingWright, in examining the records of deaths fromgastro-enteritis in the Borough of Willesden,found that deaths occurred with unexpected fre-quency in nearby houses or in short streets. Hefound evidence of epidemic association and con-cluded that there was some common local factor,most likely some specific infective agent, responsi-ble for the focal outbreaks in the borough.Social IncidenceApart from epidemics in the newborn the disease

is preponderantly one which occurs in the lowerstrata of society where overcrowding and poorhygiene are liable to occur.

July 1948370

copyright. on A

pril 18, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.24.273.368 on 1 July 1948. D

ownloaded from

ANDERSON: Infantile Diarrhoea and Vomiting

Seasonal IncidenceCooper (I937) found 52.3 per cent. of cases

from June-October. Smellie (I939) found aslightly higher incidence from June-October.Campbell (i94i) found 53 per cent. from June-October. Payling Wright (I945) found a slightpreponderance in the winter., There is thus, asmentioned previously, a change in the incidenceof the disease from a previous summer pre-ponderance to an overall distribution throughoutthe year.Symptomatology

Incubation period.-This is usually from one toeight days though occasionally much longerperiods have been recorded.

Stage of invasion.-There is loss of appetite,disinclination to take feeds, drowsiness, suddenloss of weight, occasional vomiting, increase in thenumber of stools.

Toxic stage.-The infant changes rapidly frombeing reasonably healthy to acutely ill and suffer-ing from varying degrees of dehydration andcollapse. The stools become projectile, frequent,and vary in colour from yellow to greenish ororange. If unchecked the disease progresses untilthere is gross dehydration, acidosis, and theHippocratic facies. The tissues lose their elas-ticity, the fontanelle and eyes become sunken, andcoma gradually develops. Recovery is evidencedby an improvement in the tissue turgor, in themental state, recovery of appetite and a decreasein the vomiting and in the number of stools.Diagnosis

In the presence of an epidemic the diagnosis isrelatively simple such as in the epidemics describedby Ormiston. In the more sporadic cases, and inslightly older infants, the acute onset and thetypical projectile stools with acidosis and rapidloss of weight suggest the condition.

PrognosisThere is no doubt that, with improvements in

nursing care, prevention of cross infection, im-proved methods of treatment, and the use ofmodern drugs to tackle secondary infection whenit occurs, the mortality from the disease isgradually falling. If, however, one compares themortality rate of various British authors there isconsiderable discrepancy. For example Alexanderand Eiser (I944) in reviewing a series of I40 casesreported a mortality of 5.7 per cent., but as theyfound that 88.6 per cent. of their cases -hadevidence of parenteral infection it is doubtfulwhether the majority of their cases came underthe category of true infectious diarrhoea. Gairdner(I945) in a series of 216 cases had a mortality rate

of 5 per cent. but the series was adversely affectedby the fact that there were 71 cases with a 67 percent. mortality rate in the first month of life whenit is generally recognized that the mortality ishighest. Apart from Gairdner's series the mor-tality rate has been improving. Cooper (1937) 47per cent., Smellie (I939) 48 per cent., Campbell(I94I) 27.7 per cent., Alexander and Eiser (I944)5-7 per cent.

ProphylaxisBreastfeeding.-It is well known that breast-fed

infants are much less liable to contract the diseaseand that the illness is less severe if contracted.This fact has been stressed by Findlay, Cooper,Smellie and Campbell.Preventive MeasuresThere have been a number of epidemics which

occurred amongst the newborn and which couldhave been prevented from spreading if moreattention had been paid to the possible means ofsuch a spread. Pollock34 (I935) in discussing theovercrowding of nurseries due to lack of spacesaid ' all too frequently nurseries are assigned towhat might be designated as left over space andtheir size is determined by the number ofbasinettes that can be crowded in without anyregard to the spacing or the aisles between them.'Clifford36 writes 'outbreaks of epidemics ofdiarrhoea of the newborn in which no aetiologycould be established can be subdivided into twogroups. The first consists of outbreaks in nur-series with excellent techniques where careful in-vestigation of the affected patients and of theirenvironment by both bacteriological and epidemio-logical studies fails to reveal any aetiological agent.In the second, and by far the largest group, in-vestigations failed to reveal any specific organism,but such atrocious nursing and formula-roomtechnique was uncovered that infection with any-thing was possible.' Without going so far as eitherof those writers in their assertions a good deal canbe done both in the newborn and in the later agegroups by more care in the isolation of the patientsand proper aseptic precautions on the part of theattendants to prevent the spread of infection onceit has occurred. Once a case of diarrhoea hasoccurred it is essential that the child should beisolated and that if any further cases occur in theward it should be closed and all fresh admissionsshould be admitted to a separate ward.Treatment

(a) Treat the cause. Where a causal organismcan be isolated such as B. Sonne, etc., specifictreatment can be undertaken.

(b) Relief of diarrhoea. This is best obtained

July I948 371

copyright. on A

pril 18, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.24.273.368 on 1 July 1948. D

ownloaded from

POST GRADUATE MEDICAL JOURNAL

by stopping all fluids by mouth, with perhaps theexception of a few minims of boiled or distilledwater to moisten the mouth. This restrictionshould be maintained until the motions haveeither ceased or have so much improved incharacter as to warrant an increase.

(c) Correction of physiological disturbance.Gamble36 has shown that one can have a markedlydiminished intake from the gastro-intestinal tractwhich can be covered by a correspondinglydiminished output of water and electrolytes bythe kidney. When the limit of renal conservationis passed, deficits develop quickly. He has alsoshown that fluids and electrolytes are not sufficientand that calories must be supplied and bymeans of a chart what he estimates to be thesurvival rate in relation to fluid and caloric intake.In a 7 kilo. infant he estimates that the survivalperiod after a severe diarrhoea during which theinfant can be kept alive whilst awaiting recovery offood tolerance is 20 to 30 days, but he also pointsout that there are many diarrhoeas which lastlonger than this and that some method of supple-menting calories must be found. This will bediscussed later.

Immediate TreatmentDisturbances of chemical equilibrium occur on

account of the vomiting, diarrhoea, and subse-quent oliguria and dehydration. The vomitinggives rise to loss of hydrochloric acid and basechloride, and consequent alkalosis. The diarrhoeacauses loss of base bicarbonate and as a resultacidosis. Oliguria gives rise to an increase in theN.P.N. of the blood. When these physiologicalchanges were first observed and it was felt thatoral feeding was contra-indicated, saline was givenparenterally as sodium chloride. As there was amarked tendency towards retention of basechloride leading to a resultant reduction of basebicarbonate and the development of acidosis, thiswas found to be contra-indicated. Because of thisand because of the frequent loss of base bi-carbonate from the body, sodium bicarbonate wassubstituted. It was found however, that therewere several objections to the use of sodiumbicarbonate. If given intravenously in amountslarge enough to be effective it produced too rapidchanges in the reaction of the body fluids with theresult that there might be an immediate shift froman uncompensated acidosis to an uncompensatedalkalosis. In addition there were difficulties insterilizing the sodium bicarbonate before it wasgiven parenterally. Hartmann and Senn37 thenintroduced sodium lactate and lactate Ringerssolutions which by virtue of their buffer contentare extremely effective in the isotonic form in re-lieving severe acidosis and in the form of the

physiological buffer salt solution in relieving eithermild acidosis or alkalosis. By the more gradualtransformation they do not lead to the severe re-actions obtained with bicarbonate. In spite ofthese advances, however, the ultimate results oftreatment did not improve greatly and Darrow38(1946) stated 'the comprehensive plan for thetreatment of diarrhoea published by Powers39in I926 apparently produced as satisfactory resultsas any.' Darrow then claimed that diarrhoea ledto a loss of intracellular potassium and that ifduring the period of fasting one could retainintracellular potassium by adding potassiumchloride to the mixture containing sodiumchloride, sodium lactate, and glucose one couldsupply the substance which by its lack was con-tributing to the continued high mortality rate inthe disease. He did, however, point out that carehad to be taken to avoid potassium intoxicationbut that if attention was paid to establishment ofkidney function before administration there waslittle danger of this occurring. He quoted twoseries of cases in which three out of 50 childrenwith the new treatment and 17 out of 53 with theold died. He did not, however, use alternatecases, although he stated that the cases werestrictly comparable.Maintenance of Metabolic RequirementsAs mentioned earlier Gamble has pointed out

that whilst providing water and electrolytes par-enterally can sustain body fluids for a period thereis also a considerable energy deficit which com-mences with the onset of the diarrhoea. Energyrequirements can be provided by glucose, butmore than a io per cent. solution is irritant to theveins. Various people have tried to overcome thisdifficulty and provide the necessary energy re-quirements by using casein hydrolysates. Thesein their earlier forms were found to be irritant tothe veins in concentrated form and Hartmannsuggested that they be used subcutaneously with aPH of 6 and in a 3i per cent. solution. Therestill appears to be a difference of opinion as to theadvisability of their use intravenously, some claim-ing that any reactions produced are due to errorsin technique. They have been found to be ofvalue in the more chronic case, especially wherethere has been found to be evidence of liverdamage. Care must be taken to ensure that theperipheral circulation is satisfactory before theiruse.

It does appear however, that even when amino-acids are given, the greater part of them is oxidizedfor energy requirements and more recently it hasbeen suggested that what is still required in thelong term case is an emulsion of fat suitable forparenteral injection. Shafiroff and Frank40 have

YJuly I 948372

copyright. on A

pril 18, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.24.273.368 on 1 July 1948. D

ownloaded from

July 1948 ANDERSON: Infantile Diarrhoea and Vomiting 373

prepared an emulsion of fat suitable for intra-venous therapy and have given it for a period ofover eight days without ill effects. Magee41 quotesIvy as preparing a fat emulsion capable of keepingdogs in nitrogen balance and good health whengiven intravenously for over 42 days.Plasma and Whole Blood Transfusions

In cases where there is evidence of severe initialshock with poor response to glucose saline solutionor where the .plasma proteins are markedly de-pleted, or there is anaemia, these may be foundto be of value and also perhaps at a stage in thediarrhoea where there is little resistance and thereis the possibility of secondary infection.

VitaminsThese may also play a part in the chronic case

and in those cases where evidence of liver damagehas been found vitamin K may be of particularvalue.

FeedingThere is considerable agreement that starvation

should be continued until there is evidence thatthe diarrhoea has been controlled. This may takeseveral days. Initial feeding should commencewith the use of expressed breast milk, separatedlacidac or protein milk, and the caloric intakeshould be small at first, taking a week to reachfull caloric requirements. In the intermediateperiod before full requirements are reached fluidintake must be kept up parenterally. Once thebreast milk or separated milk is being digestedwithout upset, a gradual change-over should bemade to a half cream milk and then, according tothe age of the infant, a full cream one.

DiscussionThe history of infantile diarrhoea and vomiting

has been reviewed. The aetiology of the diseasehas been discussed and evidence produced to showthat, in some instances at least, the disease maybe caused by a virus. Stress is laid on the im-portance of ensuring that more stringent care istaken over the administration of nurseries for thenewborn, particularly as the incidence of thedisease in the first month of life is increasing alongwith an increasing tendency for babies to be bornin hospital rather than at home, for various reasonsincluding better obstetric care, shortage of homefacilities and economic necessity.

It is pointed out that the morbidity and mor-tality from the disease is diminishing as a result ofthe recognition of the infectious nature of thedisease, better isolation facilities, better nursingtechnique and better medical treatment includingthe treatment and prevention of complications by

the use of chemotherapy and antibiotics. Treat-ment is discussed in its various stages. Initialtreatment is aimed at resting the alimentary tract,treatment of collapse and restoration of fluid loss,replacement of electrolytes and restoration ofacid-base metabolism to normal. On how all thisis carried out there is not, as yet, complete agree-ment. For instance some recommend restorationof kidney function by means of intravenous plasma,whilst others consider glucose saline equallyeffective. Then some consider that if kidneyfunction is satisfactory and enough fluid is suppliedin the form of glucose saline or plasma any acidosisor alkalosis present may be corrected without theuse of bicarbonate or Hartmann's solution. Onthe whole, however, Hartmann's solution seems tobe used now by most people. Darrow's views onthe necessity for increased potassium intake havebeen mentioned. Although his views have notyet met with general acceptance, in the writer'sopinion, from limited personal use of his solution,it does appear to have special merit. Maintenancetherapy may be carried out by the use of o percent. glucose intravenously with, in the morechronic cases the addition of casein hydrolysates,remembering that the latter have the disadvantageof causing, in some cases at least, severe local re-actions. The need in some cases for blood plasmato replace protein loss and whole blood to combatanaemia is mentioned as is the use of vitamins,particularly vitamin K in the more chronic cases.The suggestion made by recent writers that solublefat is required in emulsion form, to supplysufficient food for body growth in addition toenergy requirements, is mentioned and also thefact that such an emulsion has apparently beengiven intravenously with some success.

Diet is discussed and it is pointed out that themain essentials are delay in commencement offeeding until the diarrhoea has ceased, initial useof either breast milk or a milk with a low fat con-tent such as separated lacidac; and a slow increasein feeding until full caloric requirements arereached, taking care that the full fluid requirementsare maintained.

REFERENCESI. GARRISON, F. H. (I929),' History of Medicine,' 4th Edition

Philadelphia.2. CARTER, A. H. (I893), Prov. Med. J., 12, 236, 292, 349.3. WILSON, G. S. (1927), Y. Hygiene, 26, 170.4. MARRIOTT, W. McK. (1923), Phys. Rev., iii, 275.MARRIOTT, W. McK. (1925), Trans. Amer. Ped. Soc., 37, 38.

5. FINDLAY, L. (1932), Arch. Dis. Child., 7, 307.6. CRAIG, W. S. (1936), Lancet, 2, 85.7. DICK (1928), Am. J. Dis. Child., 35, 955.8 BARENBERG, L. H., LEVY, W., GRAND, M. (1936),

J.A.M.A., o06, 1256.9. FRANT, S., ABRAMSON, H. (I937), J. Paed., II, 772.FRANT, S., ABRAMSON, H. (1938), Am. J. Dis. Child., 5,

I288.FRANT, S., ABRAMSON, H. (i939),' New York State Journal

of Medicine,' 39, 784.io. WATT, J., CARLTON, E. (1945), 'Public Health Reports,'

6o 734WAT', J. (x945), Am. J. Public Health, 35, 1205.

copyright. on A

pril 18, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.24.273.368 on 1 July 1948. D

ownloaded from

374 POST GRADUATE MEDICAL JOURNAL July 1948

ii. McKINLAY, B. (1937), Am. J. Dis. Child., 54, 1252.12. SELIGMANN, E., BARASH, L., COHLAN, S. Q. (I947),

J. Paed., 30, i82.I3. ENSIGN, P. R., HUNTER, C. A. (1946), J. Paed. 29, 620.I4. FELSEN, J., WOLANSKY, W. (1942), Arch. Paed., 59, 495.I5. DRAPER, F., BROWN, G. W. (1946), M. J. Australia, I, 469.I6. ORMISTON, G. (1941), Lancet, 2, 588.17. PAYLING WRIGHT, G., PAYLING WRIGHT, H. (1946),

Hyg. J., 480.PAYLING WRIGHT, G., PAYLING WRIGHT, H. (1946),

Hyg. J., I5.PAYLING WRIGHT, G. (1948), Proc. Royal Soc. Med., 41,

I82.18. NETER, E. R., CLARK, P. (1944), J. Digest. Dis., 356.I9. McCLURE, W. B. (1943), J. Paed., 22, 60.20. BAKER, C. J. (I939), J. Paed., 14, 183.21. JAMES, U. (1947), Arch. Dis. Child., 22, i8o.22. COOPER, E. D. (I937), Arch. Dis. Child., 7, 307.23. SMELLIE, J. M. (1939), Lancet, I, 969.24. CAMPBELL, R. M., CUNNINGHAM, A. A. (I94I), Arch.

Dis. Child., I6, 211.

25. ALEXANDER, M. B., EISER, Y. (1944), B.M.J., 2, 425.26. GAIRINER, P. (1945), Arch. Dis. Child., 20, 22.27. PATTEPRSON, W. H., SMITH, G. S. (1944), B.M.J., 2, 659.28. WYLLIE, W. G. (1948), Proc. Royal Soc. Med., 41, i82.29. LYON, G. M., FOLSOM, T. G. (1941), Am. J. Dis. Child.,

61, 427.30. LIGH'T, J. S., HODES, H. L. (1943), Am. J. Public Health

33, I451.31. CUMMINGS, G. D. (1947), J. Paed., 30, 706.32. SEVITT, S. (1948), Proc. Roy. Soc. Med., 41, i82.33. BUDDINGH, G. J., DODD, K. (1944), J. Paed., 25, 105.

BUDDINGH, G. J. (1946), S. Med. J., 39, 382.34. POLLOCK, H. (1935), Address delivered New England

Hospital Association, Boston, March 9th.35. CLIFFORD, S. H. (1947), New England Med. J., 26, 969.36. GAMBLE, J. L. (1947), J. Paed., 30, 488.37. HARTMANN, A. F., SENN, M. J. E., J. Clin. Invest., II, 337.38. DARROW, D. 0. (1946), J. Paed., 28, 515,'541, 550.39. POWERS, G. F. (1926), A. J. Dis. Child., 32, 232.40. SHAFIROFF, B. G. P., FRANK, C. (1947), Science, o06, 474.41. MAGEE, H. E. (948), B.M.J., I, 4.

LAY CLINICIANSNVo. 12

Often when reading a book or play one comesacross an excellent description of some disease. Suchaccounts are always of interest to the doctor and theyoften show surprisingly accurate powers of observa-tion and clinical acumen. We hope to print some ofthese ' case records ' from time to time and for thosewho would like to try their hand at naming the author,this information will be given at the end ofthe extract.

' Certainly the occupants of the landaulet werebizarre enough to attract attention even in theworldly Paris. But the figure perched on thedriver's seat held the added fillip of surprise.Seated there, his legs braced against the highfootboard, his knees covered with a driver's rug,his powerful arms and hands managing the twoneat chestnuts with a true horseman's deftness oftouch, there seemed nothing remarkable about thecoachman, Cupidon. It was only when he threwaside the rug, clambered down over the wheel asagilely as a monkey and stood at the kerb that yousaw with staring unbelief the man's realdimensions. The large head, the powerful armsand chest belonged to a dwarf, a little man notmore than three feet high. His bandy legs werelike tiny stumps to which the wee feet wereattached. This gave him a curiously rolling gaitlike that of a diminutive drunken sailor on shoreleave. The eyes in the young-old face weretender, almost wistful; the mouth sardonic, theexpression pugnacious or mischievous by turn.This was Cupidon, whom they fondly calledCupide; bodyguard, coachman, major-domo.When he spoke, which was rarely, it was in a sur-prisingly sweet clear tenor like a choirboy whosevoice had just changed. He might have been anyage-fifteen, twenty, thirty, forty. There were

those who said that, though white, he wasKakaracou's son. Certainly she bullied him andpampered him by turns. Sometimes you saw herwithered hand resting tenderly on the tiny man'shead; sometimes she cuffed him smartly asthough he were a naughty child. She managed tosave the choicest titbits for his plate after theothers had finished, she filled his glass with goodred wine of the country as they sat at the servants'table, he in his specially built high chair on whichhe clambered so nimbly up and down.

' " Drink your red wine, Little One. It will makeyou strong."

'Instead of thumping his chest or flexing his armhe would rap his head briskly. " I don't need redwine to make me strong. I'm strong enough.Here's my weapon." But he would toss down theglassful, nevertheless, giving the effect as he didso of a wickedly precocious little boy in his cups.Everyone in the household knew that his boastwas no idle one. That head, hard and thick as acannonball, was almost as effective when directedagainst an enemy. Thigh-high to a normallybuilt man, he would run off a few steps, thencharge like a missile, his head thrust forward anddown, goat fashion. On a frontal attack, mentwice his size had been known to go down withone grunt, like felled oxen.'

This colourful description of an achondroplasicdwarf appears in a recent novel, ' Saratoga Trunk'by Edna Ferber. Dwarfs of all descriptions areencountered in the world's literature but frequentlythey are the product of the author's imagination.Edna Ferber, however, gives us a pen picture of thecondition which was first described by vonSoemmerring in I791.

copyright. on A

pril 18, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.24.273.368 on 1 July 1948. D

ownloaded from