1104

would then form the initial lesions of disease, as tubercles.In this light the topography of tubercles is determined by theanatomy of the carry-off system for surplus alveolar fluid.

REFERENCES

Bertalanffy, F. D., Leblond, C. P. (1953) Anat. Rec. 115, 515.Bratianu, S., Guerriero, C. (1930) Arch. Anat., Strasbourg, 11, 423.Broderson, J. (1933) Z. mikr.-anat. Forsch. 32, 73.Carleton, H. M. (1934) Proc. roy. soc. B, 114, 513.Clara, M. (1936) Z. mikr.-anat. Forsch. 40, 147.Cowdry, E. V. (1952) Laboratory Technique. 3rd ed., Baltimore ;

p. 276.Evans, C. L. (1949) Principles of Human Physiology. 10th ed.,

London.Frey-Wyssling, A. (1953a) Submicroscopic Morphology. London ;

p. 2.- (1953b) Ibid, p. 55.

Guieysse-Pellissier, M. A. (1920) C. R. Acad. Sci., Paris, 170, 1411.Hale, C. W. (1946) Nature, Lond. 157, 802.Ham, A. W. (1953) Histology. 2nd ed., Philadelphia, London, and

Montreal; p. 586.Lancet (1947) i, 68.Leathes, J. B. (1925) Lancet, i, 957.Leblond, C. P. (1950) Amer. J. Anat. 86, 1.- Bertalanffy, F. C. (1951) Canad. Med. Ass. J. 65, 263.

Loosli, C. G., Adams, W. E., Thornton, T. M. jun. (1949) Anat. Rec.105, 697.

Low, F. N. (1952) Ibid, 113, 437.- (1953) Ibid, 117, 241.

Macklin, C. C. (1938a) Ibid, 70, suppl. 53.- (1938b) J. thoroc. Surg. 7, 536.- (1946) Trans. roy. Soc. Can. 40, 93.- (1949) Int. Congr. exp. Cytology, Stockholm, 1947, p. 383.- (1950a) Canad. J. Res. 28, 5.- (1950b) Proc. Inst. Med. Chicago, 18, 78.- (1951a) Lancet, i, 432.- (1951b) Anat. Rec. 109, 321.- (1953a) Ibid, 115, 343.- (1953b) Ibid, p. 403.- (1953c) Ibid, p. 431.

Policard, A. (1942) Bull. Histol. Tech. micr. 19, 15.Rinehart, J. F., Abul-Haj, S. K. (1951) Arch. Path. 52, 189.Short, R. H. D. (1950) Phil. Trans. 235, 35.Sikorsky, J. (1870) Zbl. med. Wiss. 8, 817.Sjöstrand, F., Sjöstrand. T. (1938) Z. mikr.-anat. Forsch. 44, 370.Terry, R. J. (1926) Anat. Rec. 32, 223.- (1945) Ibid, 91, 302.

von Hayek, H. (1942) Anat. Anz. 93, 149.- (1952) Ergebn. Anat. EntwGesch. 34, 144.- (1953) Die Menschliche Lunge. Berlin.

TUBERCULOUS MENINGITIS IN

CHILDREN TREATED WITH

STREPTOMYCIN AND P.A.S.

JOHN LORBERB.A., M.D. Camb., M.R.C.P.

SENIOR LECTURER IN CHILD HEALTH IN THE UNIVERSITY OF

SHEFFIELD

WITH the advent of streptomycin in 1947 it became

possible to treat patients with tuberculous meningitis.Since then there have been many modifications in thetreatment, but there have been four main stages in itsevolution.

In the first stage, in 1947, streptomycin alone wasavailable, and no-one knew the best way to use it.

In the second stage, from 1948 to 1950, streptomycintreatment was more standardised and consisted in

prolonged and uninterrupted intramuscular courses, withshorter and often interrupted intrathecal courses.

Alternative routes were introduced for the intrathecaladministration of streptomycin in spinal blocks andother obstructions in the pathways of the cerebrospinalfluid (C.s.F.). Major neurosurgical procedures were.

attempted occasionally. Adjuvants such as’ Sulphetrone’and streptokinase were tried but found to be of littlevalue. Towards the end of that period intrathecaltuberculin was introduced by Smith and Vollum (1950).The third stage in the treatment of tuberculous

meningitis is described here. It is characterised by theroutine use of oral p-aminosalicylic acid (P.A.s.) and theselective use of intrathecal tuberculin in conjunction withstreptomycin treatment, and ends with the introduction

of isoniazid in April, 1952. That date marks the beginningof the fourth stage, which is still in progress.

Several British centres have already reported theirresults relating to the first two stages. The combinedresults of five major English centres, including our own,were recently analysed statistically (Lorber 1954) ; 549

patients of all ages admitted to those centres between1947 and 1950 were observed for a minimum of two years,and many of them for five years or more. Observationfor two years was useful for the assessment of resultsbecause deaths were relatively frequent up to that time but

TABLE I—AGE-DISTBIBUTION ON ADMISSION

Age (yr.)

Less than 11-1 11/122-22 11/123-4 11/125-911/12

10 or more

Total

No. of cases

3} 14 ______

7 15 2 } 2438

--

very exceptional afterwards. These results form a base-line for comparison with newer methods of treatment.The two-year survival-rate of the whole series was 46.1 %.In 1947 the results were indifferent, 32% having survivedfor two years compared with almost 50% in 1948-50.The results were the same in each of the three years 1948,1949, and 1950.The most important prognostic factor was the stage

of the tuberculous meningitis on admission. In 1948-50,74% of the patients admitted in an early stage survived,compared with 54% of those in an intermediate stage, andonly 25% of those in an advanced stage. Most of thesurvivors of those admitted in either an early or an

intermediate stage were in good condition and withoutneurological sequelæ. The survivors of those admittedin an advanced stage often had permanent physical ormental damage. Children aged less than 3 years and thosewith miliary tuberculosis fared less well, but the mostimportant prognostic factor was the stage of thetuberculous meningitis on admission.The results in our first 82 cases (Illingworth and

Lorber 1951) showed an over-all one-year survival-rateof 44%. After the completion of that series our routinetreatment was altered in several respects. I report herethe results in our patients who were admitted in thetwenty months from August, 1950, to March, 1952. Allthe patients have been observed for two years or more,and all completed their course of treatment more thana year ago.

Material

38 consecutive cases of tuberculous meningitis inchildren were admitted without any selection. No patientwas refused treatment. Most of the patients were referredfrom other hospitals. The series includes 3 children whohad been unsuccessfully treated elsewhere for severalweeks. 1 other child is not included, because he receivedno treatment ; he was brought to hospital in status

epilepticus and died an hour after admission.

TABLE II-STAGE OF TUBERCULOUS MENINGITIS ON ADMISSION

Stage

Early ..Intermediate ..Miliary developing meningitis..Advanced ......

Total ......

Present series.No. of cases

9(23-7%)16 (42.1 %)2 (5.2 %)

11 (28.9%)

11 (28.9 %)38

First 82 cases(Illingworthand Lorber

1951). ).No. of cases

9 (11.0%)37 (451 %)10 (12.2%)26 (317%)

82

1105

Nearly two-fifths of the children were aged less than3 years, and only 2 were aged more than 10 years(table 1). Of the 38 children 26 were boys.The classincation suggested by the Medical Research

Council (1948) was again used for assessing the stage oftuberculous meningitis on admission. About a quarterof the patients were in the early stage, over two-fifthsin the intermediate stage, and nearly a third in theadvanced stage and unconscious. Only 2 patients devel-oped meningitis while under streptomycin treatment formiliary tuberculosis (table 11). The treatment of 1 of thesewas started in another hospital, and’ he developedmeningitis before he was transferred to us.

Miliary tuberculosis was shown by radiographs of thechest in 10 children (26%) (table ill) ; 5 of these hadclioroidal tubercles. In 5 other children the initial

radiograph of the chest suggested miliary tuberculosisbut was insufficient to enable a firm diagnosis to be made.Of these 5 children 2 died, and miliary tuberculosis wasconfirmed at necropsy. In another fatal case miliarytuberculosis was found at necropsy but had not beendetected during life. In view of our experience (Emeryand Lorber 1950) that miliary tuberculosis may oftenbe missed during life, the total number of children whohad miliary tuberculosis was probably more than the16 definite. or probable cases.Bacteriological confirmation was obtained from the

c.s.r. in life in 35 (92%) of the 38 patients. The organismwas of the human type in 22 of the 23 cases in whichtyping was successfully done.

In 3 children no bacteriological proof was available.All survived and all had a positive tuberculin test and acharacteristic c.s.F. 1 infant, aged 1 year, had miliarytuberculosis of the lungs. The 2nd, a boy aged 5 years,

TABLE III-TYPE OF CASE, AS DETECTED DURING LIFE

:.1:"llingjti aloiieTiliqrv m·i 11lpningitis ...

Poil;l,-’ milian- and meningitis ..

Total --- ....

No. of cases

23 (60-5%B10 (263°)5 (132%)

38 "

developed permanent optic atrophy and intracranialcalcifications. The 3rd, a girl aged 5 years, had miliarytuberculosis of the lungs and choroidal tubercles. Thediagnosis in these 3 patients is felt to be supportedsuincieiitly by indirect evidence.

Treatment

fS’Vp/t/Ci’M n(a) Irrtnanauscular streptomycin was given in two

divided doses (20 mg. per lb. body-weight daily) eitherfor a minimum of six months or for a minimum of twomonths after the last intrathecal injection of streptomy-cin, whichever period was the longer. If the patient’sclinical condition, the state of the C.S.F., or the radio-graphic appearance of the miliary lesions in the chestwas unsatisfactory, intramuscular treatment was con-tinued for as long as necessary. Of the 28 survivors 26received a single course. In 13 this lasted for the minimumperiod of six months ; in 8 it lasted for seven months ;and in 4 it lasted up to twelve months. None weretreated for more than a year. In 1 child treatment wasabandoned after three months because of extreme

hydrocephalus and decerebrate rigidity ; she neverthelesssurvived. The 2 other survivors completed one courseof treatment (six and seven months) but relapsed threeand a half and four and a half months later and weretreated for another six months in each case. All thedeaths took place within four and a half months ofadmission.

The total duration of treatment was substantiallyshorter than in our earlier series, in which 7 of 36survivors were treated for more than a year, and 2 formore than two years.

(b) Intrathecal streptomycin in daily doses of 25 mg. or50 mg. was given for forty-five injections in the first twomonths of treatment, a day’s rest being allowed each weekin the first four weeks and two days’ rest each week inthe next four weeks. Intrathecal treatment was con-tinued beyond this period without interruption if tuberclebacilli had been recovered from the c.s.F. during thepreceding six weeks, or if there were other signs of anactive meningeal process, judged by the clinical conditionor by the state of the c.s.F. If intrathecal treatment wasdiscontinued after forty-five injections it was resumed iftubercle bacilli were again found in the c.s.F. ; or if theC.S.F. cell-count did not fall below 100 per c.mm. soonafter intrathecal treatment was stopped ; or if the c.s.F.cell-count did not show a progressive fall after that periodor showed a sudden or sustained rise, especially inassociation with a falling c.s.F.-sugar and a rising c.s.F.-protein level. If any of these conditions was found,further similar courses of forty-five injections were givenas often as necessary.

In none of the fatal cases were more than fifty intra-thecal injections given. Of the surviving 28 patients 10had fewer than fifty injections, 9 had between fifty anda hundred, and 9 had between a hundred and one anda hundred and thirty-five ; none had more than a

hundred and thirty-five injections. The number of injec-tions given was much smaller than in our previous series.

Spinal blocks developed in 2 patients, of whom 1survived. The diagnosis was confirmed by the methodsreported elsewhere (Lorber 1950), and the treatment wasgiven by the cisternal route. This incidence of spinalblocks-2 (5%) out of 38-was much less than in ourprevious series: 20 (24%) out of 82. 1 other patientwas treated by the cisternal route because of tuberculosisof the lower thoracic vertebrae.The ventricular route was used for streptomycin

injections in 4 patients : in 3 because of either high c.s.F.pressure or gross papilloedema, and in the 4th because ofextensive tuberculous involvement of several vertebrse.

Para-aminosalicylic acidP.A.S. was given by mouth in divided doses three-hourly

(0-5 g. per kg. body-weight daily). No p..s. was givenbetween 9 P.M. and 6 A.M., but the 9 P.M. dose was doubled.P.A.s. treatment was continued until the child’s dischargefrom hospital two or three months after the discontinua-tion of streptomycin treatment. This treatment was welltolerated.

Intrathecal TitberculinPurified protein derivative (P.P.D.) was given selec-

tively to 9 patients in the advanced stage of tuberculousmeningitis on admission after a short period of observa-tion on streptomycin treatment, and to 3 patients in theintermediate stage on admission who showed definiteand progressive deterioration despite the routine treat-ment. Free access to the lateral ventricles was consideredessential during this treatment, and anterior burr-holeswere made in all the cases in which the fontanelle wasalready closed.

Intrathecal St’reptokinaseA controlled trial of intrathecal streptokinase was in

progress in 1950 (Lorber 1951a) and the first 8 patientsbelonged to that trial ; 3 of them received streptokinase,2 of them surviving.RelapseThe 2 patients who relapsed after the completion of

streptomycin treatment were treated again as if theywere new patients.

1106

General Treatment

In all the cases general supportive measures wereused, as in the earlier series. Emphasis was placed on theearliest possible mobilisation out of bed and on playtherapy. -

Method of Follow-upThe method of follow-up was the same as in our earlier

series. All the children attended a special clinic and wereexamined and supervised by one person throughout.Intelligence and hearing were tested regularly by thesame independent experts not connected with the depart-ment or with the treatment of the children. All thesurvivors remain under observation.

Results

Of the 38 children 28 (73-7%) survived, and 10

(26.3%) died. The minimal follow-up was two years. Ofthe survivors 9 have been observed from two to two and

TABLE IV-RESULTS OF TREATMENT OF 38 CASES OFTUBERCULOUS MENINGITIS

Condition on admission

Conscious ....

Unconscious

Total ..

No. of cases

27

11

38

Survivors

25 (92-6%)

3

28 (73-7%)

Period of observation 2-31/. years.

a half years, 11 from two and a half to three years, and8 from three years to three years and eight months.This survival-rate is almost exactly double the two-yearsurvival-rate in our previous series, if deaths are includedwhich took place after our report in 1951. All the sur-vivors in this series have been observed for more thanthirteen months after the completion of streptomycintreatment, and 26 of the 28 for more than eighteenmonths.

PROGNOSTIC POINTS

Tables iv and v show the importance of early diagnosis.Of the 27 patients who were conscious on admission 25(92-6%) recovered: Neither of the 2 deaths was due totuberculous meningitis. None of the 9 children admittedin the early stage of tuberculous meningitis died. The

only death among the 16 children admitted in theintermediate stage was in a boy aged 7 years, four and ahalf months after the beginning of his treatment. Bythat time his C.S.F. was practically normal, and no grossmeningeal lesions and no hydrocephalus were found atnecropsy. He had seven carious vertebrae with para-vertebral cold abscesses, a tuberculous knee-joint,ulcerative tuberculosis of his kidneys, and diffuse calci-fying tuberculous lesions in his spleen. All these lesionswere present on admission. His death was due toadvanced cachexia.

_

The other child who was conscious on admission anddied was a boy, aged 4 months, who had been admitted

TABLE V-RESULTS IN RELATION TO THE STAGE OF DISEASE

ON ADMISSION IN 38 PATIENTS

Recovered :No sequelae . -

Moderate sequelaeSevere sequelae ..

All survivorsDied ....

Total ..

Stage II Miliary Inter- Ad-

I

Early developing mediate vancedmeningitis

mediate vanced

7 1 13 02 0 2 10 0 0 2

-

9 1 15 3 ’

0 1 1 8-

9 I 2 j 16 11

Total

2152

-

2810

38

to another hospital with miliary tuberculosis and massiveconsolidation of the right upper lobe.

His c.s.F. was normal. Streptomycin treatment was begun,but meningitis developed within a fortnight, and he wastransferred to us. His response to treatment appeared satis-factory for nearly four months. His c.s.F. returned to normal,and the miliary shadows and consolidation could no longerbe detected in radiographs of the chest. His fever, however,returned, and a radiograph of the chest unexpectedly showeda gross pericardial effusion. The effusion was tapped, andpneumococci were grown from the thick pus obtained. In

spite of penicillin treatment the boy went downhill and diedsuddenly after two days. At necropsy a further unexpectedfinding was an extensive pneumococcal meningitis. Therewas evidence of healing disseminated tuberculosis on histo-logical examination. There had been no clinical indication ofacute meningitis ; and, when the C.S.F. was last examined threedays before his death, the cell-count was 9 per c.mm., protein40 mg. per 100 ml., and sugar 58 mg. per 100 ml.

The 25 survivors who were conscious on admission areall in excellent general condition ; none are mentallyaffected, but 2 are deaf, 1 is partially deaf, and 1 is

partially blind. 21 (84%) of the 25 are without anyneurological sequelae. 1 of them, however, is still in anorthopaedic hospital receiving treatment for tuberculousspondylitis. All the 3 deaf children received more than

ninety intrathecal injections. The need for prolongedintrathecal treatment was probably a factor in thecausation of their deafness (Lorber 1954). None of our

patients received dihydrostreptomycin.Patients who were unconscious when treatment was

started responded poorly. Of 11 children admitted withadvanced tuberculous meningitis only 3 survived. Noneof the 3 survivors are free from sequelae, 2 being hydro-cephalic idiots with gross neurological defects. The 3rdchild recovered without any subjective defects, althoughhe had a severe paraplegia in flexion during treatment,associated with ectopic ossifications in his psoas-majortendons, which are now resorbing. This case, togetherwith 4 others of ectopic ossifications, has already beendescribed (Lorber 1953).Compared with our first series, the quality of the sur-

vivors has shown great improvement. Altogether 75% ofthe survivors in the present series are free from anydefect (grade 1), compared with 61% in the previousseries. The sequelae in the remainder were less severe inthis series : 18% have some physical disability (grade 2),against 11% in the previous series, and only 7% havegrave mental and physical sequelae (grade 3), against22%. 10 of the first 19 survivors have developed someintracranial calcification, but it requires a little longerperiod of observation to assess its final incidence. Inour previous experience intracranial calcification devel-oped in about two-thirds of all the children who recoveredfrom tuberculous meningitis (Lorber 1952).

Of the deaths of the 8 children admitted in theadvanced stage 6 took place within a month, 1 on the

forty-eighth day, and the last on the sixty-third day afteradmission. 1 of these children had a double meningitis,meningococci and tubercle bacilli being obtained fromthe c.s.F.The child’s age appeared to have an influence on the

prognosis. Of the 14 children aged less than 3 on

admission 8 (57%) survived, including the 2 decerebratechildren. Of the 24 aged more than 3 years 20 (83%)survived, and none of them suffered mental damage.The influence of age, however, is only apparent, and isdirectly connected with the stage of the tuberculous

meningitis on admission. Of the 14 children aged lessthan 3 years 7 were in the advanced stage ; 5 of themdied, and the other 2 are decerebrate. Of the 7 who wereconscious on admission only 1 died-the infant, aged4 months, who developed pneumococcal meningitis andpericarditis. The 6 other children are in good health,1 being deaf. Only 4 of the 24 children aged more than

1107

3 years were in the advanced- stag@, and 3 of these died.Therefore the less favourable prognosis in young childrenwas directly due to delayed diagnosis, as in our nrstseries.

’

’ .

The presence of miliary tuberculosis did not adverselyaffect the prognosis. 8 of the 10 children recovered, andthe 2 who died were those in whom the cause of deathwas not directly attributable either to the miliary tuber-culosis or to the ’meningitis. The appearance- of miliarytuberculosis in the radiograph of the chest disappearedin all 10-in 7 within three months.

Of the 12 patients who were given P.P.D. 6 died, butwe gave P.P.D. selectively to the patients with the

poorest prognosis. Of the survivors 3 were in the iitter-mediate stage on admission but did not respond toroutine treatment. 1 of them became unconscious beforeP.P.D. was started. After a very stormy course this boy,aged 4 years, made an excellent recovery. He was thefirst child in our experience who recovered in spite ofbecoming unconscious during routine treatment. Intra-thecal tuberculin probably helped in these 3 cases.Without P.P.D. there might have been only 12 recoveriesinstead of 15 out of the 16. P.P.D. seemed to be of littlebenefit in most of our 9 advanced cases in which it wasused. Only 3 of the 9 children survived, and 2 of themare decerebrate. In 1 of the fatal cases death was

possibly accelerated by the P.P.D.

Discussion

These results represent a considerable advance over ourprevious figures. The over-all two-year survival-rate ismuch higher (73-7% against 46-1%) than in the largecombined English series relating to 1947-50 and repre-senting tive important centres in the country (Lorber1954).They are the highest long-term survival figures

published in this country. These results were obtainedwith shorter courses of intramuscular treatment andfewer intrathecal injections of streptomycin than in theprevious series. Fewer children had relapses, and manyfewer developed a spinal block. The incidence of hydro-cephaius was much less, although exact figures cannotbe given because encephalography was no longer a

routine procedure. Encephalograms were made only asa guide to treatment if progress was unsatisfactory, if

complications were suspected, or if a decision was requiredabout abandoning treatment (Lorber 1951b). Treatmentwas abandoned in 1 case only. The condition of thesurvivors was better, more are free from sequelae, andfewer became decerebrate. There have not been anylate deaths.The whole of the improvement in the results was

obtained in children who were conscious on admission.The two-year survival-rate in these children was 92.6%(25 of 27), compared with 43-6% (24 of 56) in our firstseries. The results in children who were unconscious onadmission were still very bad. The moral of this isobvious. All possible efforts must be made to avoid delayin diagnosis.

Increasing experience undoubtedly played a part inthe improvement in the results. The outstanding dif-ference, however, is the elimination of deaths afterthe second month of treatment. The prevention of thedevelopment of streptomycin-resistant organisms by theadjuvant action of P.A.S. was probably responsible forthis. No resistant organisms were found in this series.Deb re et al. (1952) presented evidence that streptomycin-resistant organisms played a large part in the latemortality in tuberculous meningitis.There is another major point concerning the absence’of

late deaths. We have never continued with intrathecaltreatment until the c.S.F. returned to normal. Indeed,the c.s.F. usually did not return fully to normal for fromnine to fifteen months with this method of treatment.

So long as the C.B.F. showed a steady tendency to improve,it did not seem to matter and did not lead to late deaths.This policy spares the patient much pain and reduces therisk of deafness. Possibly a further reduction of intra-thecal treatment might not be reflected in poorer resultsnow that isoniazid is available. The risks, however, indisregarding the good results with the method describedmight be serious- and need careful consideration.

It is also apparent that the routine use of intrathecaltuberculin is unjustifiable. Of our 27 conscious patients22 recovered without it, and we were able to use it ingood time in those 3 who were thought to be in needof it. ’ The last 2 deaths were not due to the tuberculous

meningitis. The value of P.P.D. in advanced cases hasnot yet been fully determined.

In conclusion, the chances of dying from tuberculousmeningitis -are very small if the treatment describedabove is followed and if the patient is conscious whentreatment begins.

Summary

Between- August, 1950, and March, 1952, 38 consecu-tive children admitted with tuberculous meningitis, 14being aged less than 3 years, were treated with intra-muscular and intrathecal streptomycin and oral P.A.s.12 selected patients with a poor prognosis were givenintrathecal tuberculin.The results of this treatment were compared with our

previous results and with the results of a large combinedEnglish series relating to 1947-50. Considerable improve-ment was noted. Of 27 conscious children on admission25 (92-6%) recovered, and the 2 deaths were not due totuberculous meningitis. Of 11 children admitted withadvanced tuberculous meningitis only 3 survived-2with severe sequel2e. The total survival-rate was 73.7%with a minimal period of observation of two years fromthe beginning of the treatment and thirteen months fromthe end of the treatment.The worse prognosis of infants in this series was directly

due to delayed diagnosis. Young children who wereconscious on admission did as well as older children.

Intrathecal tuberculin appeared to benefit 3 childrenin the intermediate stage on admission who did not

respond to routine treatment.There were no late deaths in this series. This is probably

due to the elimination of streptomycin-resistant organ-isms by the combined action of streptomycin and P.A.S.

It was not found necessary to continue with intra-thecal treatment until the c.s.F. returned to normal.

I have much pleasure in thanking Prof. R. S. Illingworthfor his criticism ; my colleagues Dr. Al. G. Philpott, Dr. D. G. H.Stone, and Dr. D. M. G. Beasley, who supervised the dailymanagement of many of the patients ; our house-physicians,Dr. N. Cole, Dr. A. George, Dr. M. Hunter, Dr. D. Judson,Dr. J. Owens, the late Dr. V. Radcliffe, Dr. K. A. Hallidie-Smith. Dr. E. Sutherland, Dr. A. Turyczyn, and Dr. M. Wylie,who did so much of the clinical work : Dr. J. L. Emery,Miss Sheila Stewart, and Miss E. Finch for the pathological,bacteriological, and biochemical work ; Dr. T. Lodge andSister Mallinder for the radiographs ; Dr. M. C. Taylor andMr. N. E. Whilde for the psychometric assessment of thechildren; Mr. G. C. Arnold for the audiometric examinations ;and the many consultants and general practitioners whoreferred cases to us.

REFERENCES

Debré, R., Brissaud, H. E., Noufflard, H. (1952) Sem. Hôp. Paris,28, 1676.

Emery, J. L., Lorber, J. (1950) Brit. med. J. ii, 702.Illingworth, R. S., Lorber, J. (1951) Lancet, ii, 511.Lorber, J. (1950) Arch. Dis. Childh. 25, 404.- (1951a) Lancet, i, 1334.- (1951b) Arch. Dis. Childh. 26, 28.- (1952) Ibid, 27, 542.- (1953) Ibid, 28, 98.- (1954) Amer. Rev. Tuberc. 69, 13.

Medical Research Council (1948) Lancet, i, 582.Smith, H. V., Vollum, R. L. (1950) Ibid, ii, 275