OBSTRUCTIVE JAUNDICEAND HAEMOLYTIC DISEASE OF THE NEWBORN

BY

PETER M. DUNNFrom the Department of Paediatrics, United Birmingham Hospitals

(RECEIVED FOR PUBLICATON AUGUST 7, 1962)

Although in most cases of haemolytic disease ofthe newborn practically all the bilirubin in the serumis of the indirect reacting type, a quarter of a centuryago Still (1927) and Hawksley and Lightwood (1934)described cases in which the serum direct reactingbilirubin was raised and signs of obstructive jaundiceappeared. This condition came to be known as the'inspissated bile syndrome' (Hsia, Patterson, Allen,Diamond and Gellis, 1952). In 1959 Oppe andValaes reported a systematic study of this condition.They confirmed that the findings of hepatospleno-megaly (Lathe, 1955), gross anaemia and hyper-bilirubinaemia, including an elevated conjugatedfraction (Cornblath, Kramer and Kelly, 1955;Stempfel, Broman, Escard6 and Zetterstrom, 1956),were present at birth. They also noted that thejaundice in these babies took on a greenish hue andthat a tendency to bleed sometimes developed.

Recent papers (Billing and Lathe, 1958; Zuelzerand Brown, 1%1b) have shown that the 'indirect'fraction is made up of free bilirubin and bilirubinmonoglucuronide, both of which are largely formedat extrahepatic sites. The 'direct' fraction representsbilirubin diglucuronide which is probably formedonly in the liver.

TABLE I

No. Seen No.Aetiol Ygy of Neonatal in Neonatal TransferredObstructive Jaundice Present Deaths to Children's

Series Hospital

Congenital atresia of the bile-ducts 2

Bile-duct comoression 0 0 0Mucoviscidosis I 0 IHigh intestinal obstruction 0 0 0Haemolytic disease of the new-born. 14 2 l

Idiopathic hepatitis I 0 I'Infective' hepatitis (viral.cytonegalic inclusion dis-ease; bacterial; syphilitic;cholangitis, toxoplasmosis) 0 0 0

Toxic' hepatitis (galacto-saemia; drugs) .. 0 0 0

Miscellaneous (portal veinthrombosis, congestive car-diac failure) .. .. 0 0 0

54

Infants were designated as showing obstructivejaundice if the direct-reacting serum bilirubin roseabove 3 mg./100 ml., which is comparable with workby Tisdale, Klatskin and Kinsella (1959) and Zuelzerand Brown (1961b).

Neonatal Obstrctive Jaundice

During 1960 and 1961, 18 cases of obstructivejaundice were diagnosed among 6,854 live birthsat the Maternity and Queen Elizabeth Hospitals,Birmingham.

Inquiries among paediatricians in the City failedto reveal further cases of obstructive jaundice in theneonatal period, following discharge from thesematernity units.

In Table 1 these cases are presented under anaetiological classification based on a review of theliterature (Hsia et al., 1952; Harms, Andersen andDay, 1954; Nakai and Landing, 1961; Cameron andHou, 1962).

Thus, haemolytic disease of the newborn wasassociated with 14 of the 18 cases of obstructivejaundice, two of the four that died, and three of thefive that were transferred to a children's hospital.No cases of haemolytic disease without obstructivejaundice required transfer. It should be noted,however, that the incidence of haemolytic diseaseat these maternity units was six times as great asthe national average.

Obstructive Jaundice due to HaemolyticDisease of the Newborn

The haemolytic process responsible for theobstructive jaundice in all these infants was due tothe Rhesus factor. The results of a survey of thiscondition at the Maternity Hospital betweenApril 1, 1960, and September 30, 1961, are shownin Table 2. Obstructive jaundice occurred in10 (80) babies with Rh haemolytic disease.

copyright. on M

arch 25, 2020 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.38.197.54 on 1 February 1963. D

ownloaded from

OBSTRUCTIVE JAUNDICE AND HAEMOL YTIC DISEASE

Haemoglobin was estimated as oxyhaemoglobinby photoelectric absorptiometer (Unicam) (Hb100% = 14-8 g./100 nml.).On the cord blood specimens, total bilirubin was

estimated by King and Coxon's method (1950), andit was only on subsequent specimens of sera thatthe presence of an obstructive elermnt was demon-strated by the use of Lathe and Ruthven's (1958)modification of the diazo method. The totalbilirubin of these later specimnens was also checkedusing the spectrophotometric method of White,Haidar and Reinhold (1958).An indirect reacting serum bilirubin of 20 mg.,

100 ml. or above was considered an indication forexchange transfusion.

Case Material

The cases of obstructive jaundice due to haemolyticdisease of the newborn occurring in the 18-month survey(Cases 1-10) will now be reported along with the otherfour cases (Cases 11-14) seen during the full two-yearperiod. Their main clinical and biochemical featuresare summarized in Tables 3 and 4, and in Fig. 1.

It will be seen (Table 3) that there were no previoussibs with haemolytic disease of the newborn in eightcases, but that of the remaining six, three had had a pasthistory of hydropic stillbirth. A moderately high orhigh titre of Rh:D antibody was present in all motherssave one that had a high titre of anti Rh:c. The ABOgrouping of mother and baby is worth noting in that itbears out Levine's observation (1943) that ABO incom-patibility tends to protect the mother from Rhesussensitization. The colour of the liquor, the sex incidenceand the maturity ofthese babies will be referred to later.At birth all but two babies were very ill. Five pre-

sented resuscitation problems. All were noted to be

12

10

00-a

z

orBE

IC01

0

a

OBST JAUNNO EXCE.EXCHAWGEDDIED

to

_~~~~~~ so * It

'0

if.M Jr. a

: v: - 3,j ** . *- X . .

I I I I a I I

150

ee

a N. 00r

if if0

0a a

0

e

100 75 soCORD HAEMOGLOB*N %

0

FIG. 1.-Haemoglobin and total serum bilirubin salues in theumbiLical cord blood of 122 babies with Rh haemolytic disease(Table 2) showing the relation between obstructive jaundice and severeanaemia and hyperbilirubinaemia at birth and subsequent mortality.The four cases of obstructive jaundice (Cases 11-14) outside the

series are included, but are shown as 9.

TABLE 2

Numbers

Babies with Rh haemolvtic disea5e .. . 133Stillbirth*. . 9 (6-8)Neonatal deaths . . . 7 (5-2)

Total perinatal mortality.. . . .. 16 (12)Cases not requiring exchange transfusion .. .. 63 (505)Cases requiring exchange transfusion .. 61 (49-5)

I exchange transfusion: 35 babies (287*.)2 exchange transfusions: 24 babies (19-6.) .3exchangetransfusions: 2babies(l-7')/.

Cases showing obstructive jaundice .. .. 10 (8-0)

Percentages are given in parenthesis.

TABLE 3SUMMARY OF MAIN FEATURES OF CASES WITH OBSTRUCTIVE JAUNDICE

Details of Cases1 2 3 4

Birth rank 2 3 2Previous sibs with haemolytic disease

of newborn .. 0 0 0Past hydropic stillbirthsTyrpe maternal antibody (Rh) .. D D DMother/baby ABO groups .. 0,'O AB'A A/OColour of liquor amnii.. ? ? ?Type of delivery N N FGestationa! age (weeks) .. 38 39 38Sex .M M MPlacental weight (b. oz.) .. .3 1.3 1.9Birth ,eight (lb. oz) .7.15 6.8 6.5Condition at birth:

Gcneral . .P F PPallor .. ..JaundiceOedemaAbdominal distensionHepatospenomegaly (grade 1-4) 4 1 2

4

20DAAGC38M

7.7

Cases With Obstructive Jaundice

5 6 7 8 9 10

7 2 3 5 3 3

0

0

D DOO A/OG ?N N39 38F F1.8 ?7.6 7.7

DA/OyC35M

2.147.7

0

DA/O

N35M1.124.15

cB/OGC38M

1.55.10

0

DA/AyN39M1.127.10

11

2

0

DBIB

C40F

1.67.10

P F P P P F P P

4 4 4 4 4 3 3 4

G = Green; Y = Yellow; P = Poor; F = Fair; N = Normal delivery: F = Forceps: C = Caesarian section; B = Breech.

55

12 13 14

2 4 3

0

DAIAGB

M

5.3

0

D0/0GN37M

5.6

DABIAyC34M

4.11

P P P

2 4 4

a

6

copyright. on M

arch 25, 2020 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.38.197.54 on 1 February 1963. D

ownloaded from

ARCHIVES OF DISEASE IN CHILDHOODTABLE 4

SUMMARY OF MAIN FEATURESS OF CASES WITH OBSTRUCTIVE JAUNDICE

First Differential Maximum SerumCord B:ood Serum Bilirubin No. of Transfusions Bilirubin Corticosteroids Dura-

Case (mg./100 mL) (mg /100 ml.) tionof

Serum Drug and Dose Age No. Jaun-Hb Biirubin Age Total Direct Exchange Simple Age Total Direct per lb. Body When of dice(-') (mg (hrs) (days) Weight/Day Started Days (wks)

I 00 ml.) (days) !Given

1 25 6-1 3 9-0 2-1 2 1 2 23-3 7-5 None - - 12 62 13-2 5 21-0 7-6 2 0 2 254 14-3 Prednisolone 4 5 1

1 mg.3 35 6-7 6 10-0 4-1 1 3 4 13-2 9-4 Prednisolone 23 55 15

1 mg4 71 7-2 6 15-4 1-8 3 0 4 26-0 15-7 Cortisone 4 7 3

1-5 mg.5 42 8-9 3 12-9 8-5 1 1 5 37-4 31-2 Prednisolone 2 17 3

1 2 nig.6 43 8-6 3 9-6 1-5 2 0 8 8-4 7-6 Prednisolone 2 1 1 2

1 mg.7 35 9-9 1 6-2 3-9 1 Died 3 hr 5-5 4-9 None Died - -8 30 11-6 3 11-8 7-9 1 Died 8 hr 12-6 8-2 None Did -9 35 11-1 7 16-1 5-0 1 1 5 29-8 20-3 Cortisone 2 25 2

4 mg.10 49 9-0 2 8-2 4-5 1 1 -?2 ? ? None - - 1-211 54 11-6 46 19-5 5-5 2 1 2 20-5 10-7 Cortisone 2 6 1

I-5 mg.12 70 9-1 12 14-3 2-7 2 0 4 18-2 14-3 Cortisone 2 8 16

1mg.13 25 4-3 32 18-9 5-7 1 2 3 22-8 16-8 Prednisolone 2 11 2

1 mg.14 48 10-5 60 25-2 7-5 3 1 3 28-4 10-8 Prednisolone 3 27 3

1 mg.

pale and 13 jaundiced. At least six were oedematous,and hepatosplenomegaly, present in every case, waslargely responsible for the abdominal distension notedin ten.The 'cord blood' findings are shown in Table 4 and

are also plotted in Fig. I along with those of all the otherbabies with Rh incompatibility mentioned in the survey(Table 2). Both mortality and obstructive jaundice areshown to be closely associated with the degree of anaemiaat birth. No baby whose cord blood haemoglobin wasover 75% either died or developed obstructive jaundice.Where the Hb was below 75%, one-quarter died andnearly half became obstructed; and when Hb was below50%, about one-third died and at least three-quartersbecame obstructed. It will also be seen that all babieswith cord blood total bilirubins over 9 mg./100 ml.demonstrated obstructive jaundice. Of the seven deathsin the survey (Table 2), all of which took place within24 hours of birth, two showed obstructive jaundicebefore they died, one did not (Fig. 3) and the otherfour were insufficiently investigated.When first estimated, usually within a few hours of

birth (Table 4), the direct reacting bilirubin was alwaysabove 1-5 mg./100 ml. and often much higher. Itspeak was reached in all but one of these cases before thefifth day. Clinical evidence of obstructive jaundice wasseen by the passage of dark urine containing bile, paleor white stools for one to three days (five cases) andjaundice of a green hue. The latter made its appearancein 10 of the 12 survivors during the first three days oflife. It was observed when there was as little as 5 mg./100 ml. direct bilirubin out of a total bilirubin of20 mg./100 ml.

Complications were very common. Signs of heartfailure occurred in at least six cases, the respiratorydistress syndrome in eight, meconium-inhalation 'pneu-monitis' in two, a paralytic ileus causing temporaryintestinal obstruction in two, and cyanotic attacks innine. A tendency to bleed was seen in seven cases(skin, five; haematemesis, two; lungs, one; injectionsites, one; brain, three) and was sometimes present atbirth. Finally, signs of cerebral disturbance were notedin six babies and were thought to be due to kernikterusin at least two (Cases 4 and 14). One of these babiesnow exhibits selective deafness at I year. No othercases of kernikterus were seen in the neonatal periodduring these two years.

In the management of these cases, vitamin K (0 5-1 0mg. intramuscularly) was ordered routinely, anddigitalization, normally reserved by us for the mostseverely affected, was thought necessary in all theseinfants. Exchange transfusion was required in everycase and two-and-a-half times as frequently in this groupas in all babies with Rh haemolytic disease. Latetransfusion for anaemia was required over five times asoften as for the whole group. Corticosteroids weregiven to 10 of the 12 survivors.

Jaundice never lasted more than three and a half weeks(Table 4), except in Case 3 (15 weeks). It is worthrecording that in this case corticoid was not given tillthe age of 3 weeks. Liver function tests on this babywere grossly abnormal at I week (S.G.O.T. 132 units;S.G.P.T. 61 units; aLkaline phosphatase, 18 K.-A. units),and also at 1 month (S.G.O.T. 132 units; S.G.P.T.63 units; alkaline phosphatase 23 K.-A. units). Inthe only other infant (Case 9) where liver function tests

56

copyright. on M

arch 25, 2020 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.38.197.54 on 1 February 1963. D

ownloaded from

OBSTRUCTIVE JAUNDICE AND HAEMOL YTIC DISEASE

Fv,o. 2-Case 7: hydrops foetalis at autopsy. Note the gross hepato-

splenomegaly and the ecchymoses mainly on the face.

were performed (at 3 months), they were again abnormal(S.G.P.T. 87 units; alkaline phosphatase 38 K.-A. units),and hepatosplenomegaly was still present in this babyat the age of 6 months. No long-term follow-up is yetpossible.Two case histories follow.

Case 3. This baby was the first of twins; both wereaffected by Rh haemolytic disease, but only one showedobstructive jaundice. The mother, aged 31, had hadone normal infant. She was 0 Rh negative andhad a high titre of Rh(D) anti-bodies. At 38 weeks'gestation twins were born by the aid of forceps. Theirfeatures are compared together in Table 5.

TABLE 5

Twin I Twin 2

Sex .. .. .. Male MaleBirth weight .6 lb. 5 oz 5 lb. 7 ozPlacental weight .. 1 lb. 9 oz. I lb. I oz.Blood groups .. 0 Rh vositive 0 Rh positiveGenotyre .. .. cDe/cde cDe/,cde'Cord' haemoglobin 355 88%'Cord' total bilirubin 6-7mg-/100ml. 6-9mg./100ml.Condition at birth .. Poor FairHepatos-lnomegaly Moderate SlightObstructive jaundice Present Not pre%entTransusions .. One exchange trans- Two exchange trans-

fusion and three fusionssim-le transfusionsT

Complications .. Heart failure, res-ir- Noneatory distress syn-drom. haematem-esis, bleding frominjection sites. in-testinal iems

Case 7. The second case history is included becausethe baby died and post-mortem material is shown inFigs. 2 and 4. The mother, aged 36, had had two babiespreviously. The first was normal and the second a

hydropic stillbirth. She was A Rh negative, witha high titre of anti-Rh(D). In the present pregnancy,she was delivered at 35 weeks' gestation by caesariansection of a boy weighing 7 lb. 7 oz. and a placentaweighing 2 lb. 14 oz. Both were hydropic. At birththe baby demonstrated asphyxia pallida, congestivecardiac failure, gross hepatosplenomegaly and manyecchymoses especially on the face. In spite of digitaliza-tion and a small exchange transfusion undertaken atl hours, he died aged 4 hours.

Autopsy. The findings were typical of hydrops foetalisdue to erythroblastosis (Fig. 2). Effusions were presentin all body cavities.

Histology. The lungs showed hyaline membranedisease. In the liver (Fig. 4) there was considerableproliferation of the bile-ducts in the portal tracts. The

lobules consisted largely of syncytial masses of pigmentedmultinucleated liver cells. The snall cytoplasmicgranules were mainly of haemosiderin and the largerbodies were due to bile pigments. Bile thrombi were

present in some of the canaliculi and there w%as alsoexcessive extramedullary erythropoiesis.

DiscussionThe incidence of obstructive jaundice in haemo-

lytic disease of the newborn is probably between8 and O% (Table 6).

Haemolytic disease of the newborn is the mostcommon cause of obstructive jaundice in thenewborn, though this may not be so apparent fromthose babies admitted to a children's hospital.While all our cases were due to Rh incompatibility,obstructive jaundice may follow exceptionallysevere cases of incompatibility due to ABO (Harriset al., 1954), Kell (Ivemark, Hogman, Rudert andAndersen, 1959) or other groups.The occurrence of 11 males to three females in

this series may be of no significance. However,the preponderance of males among fatal cases andcases of kemikterus due to haemolytic disease iswell known (Walker and Mollison, 1957) and malesex hormone therapy may cause cholestasis in adults(Schaffner, 1960). Also Allen, Diamond andWatrous (1949) showed that the use of male, rather

TABLE 6

Haemolytic Disease of Newborn

Reported Series Total No. with % withNo. Obstructive Obstructive

Jaundice Jaundice

Stempfd et al. (1956) .. 83 7 8-4Oppe and Valaes (1959) .. 117 12 10-2Present Series . .j 124 10 8-0

5

57

copyright. on M

arch 25, 2020 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.38.197.54 on 1 February 1963. D

ownloaded from

58 ARCHIVES OF DISEASE IN CHILDHOOD

_;; b ji S ~~~ _S,.. ,.401t~~~~~

~ ;

--_. ^rs-_ +k iZ

FiG. 3.-Post -mortensection of the liver of a baby of 35 3wss'gegtion with Rh haemolytic ddsa (cord blood Hb 75 but wnhout obstruc-ti undice o ieawho died age8 hours. Shosdnorftallobulararchi ithaniattnoteramedusLryerthropoicticiman (H.mand E.x364.)

g s f b iwa

^* ", 0 ^Ws Ka >~~~~~~%.|e5t #. w- s C*~~~~A-~2 {-2....jjL i;:nA*} O.' ''-

-_

Fi.4Cs :pstmre eto of the live of a bayo 5wes to.wt eemR aroy+ iccr oHb3% adostutv jandce -wh didae or. Shw diogn4aino eai rhtcue mymliuiae in

ce otinggaul fbi pqpen an ha;oidrn xc iv rt ooissadblth mi.(.ndE x36

copyright. on M

arch 25, 2020 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.38.197.54 on 1 February 1963. D

ownloaded from

OBSTRUCTIVE JAUNDICE AND HAEMOL YTIC DISEASEthan female, donor blood for exchange transfusionwas accompanied by a much greater mortality.At that time, only one exchange transfusion wasthought necessary and the increased mortality waspresumably due to kernikterus.Oppe and Valaes (1959) suggested that babies of

less than 36 weeks' gestation could not developobstructive jaundice because of inability to con-jugate bilirubin. Yet, in this series, three wereunder 36 weeks' gestation (Table 3), and two of theselater died. The high mortality among immaturebabies with obstructive jaundice probably accountsfor it being rarely diagnosed in the premature infant.The obstructive jaundice of haemolytic disease

may often be detected at birth, a cord blood 'direct'bilirubin over I mg./100 ml. being highly sus-picious and over 2 mg./103 ml. pathognomonic ofthis condition. Thereafter, the biochemical diag-nosis depends on a rise of the 'direct' bilirubinabove 3 mg./100 ml. Clinically, all severelyaffected cases of haemolvtic disease of the newbornshould be suspect. Jaundice of a green hue is oftenof diagnostic value. As bilirubin diglucuronide isyellow, this green coloration may be caused by itsconversion to another pigment, possibly biliverdin.Green liquor amnii was reported in some of these

cases and was difficult to distinguish from meconiumstaining. This on occasion led to a faulty diagnosisof foetal distress. In this condition, where chole-stasis is often present before birth, urobilin andrelated pigments would be expected in the foetalurine and hence lend colour to the liquor.

Kove, Goldstein, Perry and Wroblewski (1959)have used the serum transaminase levels in thedifferential diagnosis of neonatal jaundice. Whilethese were within normal limits in mild or moder-ately severe haemolytic disease, the levels wereraised, indicating hepatocellular damage, in verysevere cases and when there was obstructivejaundice. The serum transaminase findings in ourtwo cases confirm this.The importance of recognizing this condition lies

in its complications and their prevention, its highmortality, the need to institute specific treatment,the possibility of avoiding unnecessary exchangetransfusions and finally the need to distinguish itfrom other causes of cholestasis. Should thejaundice persist, this distinction may become verydifficult later in infancy. Complications of theexchange transfusion such as serum hepatitis orportal vein thrombosis (Shaldon and Sherlock, 1962;Tizard, 1962) may further confuse the issue. A lesssevere form of obstructive jaundice that appearslater may follow relatively mild cases of haemolyticdisease of the newborn (Jouvenceaux, Brizard,

Michaud and Revol, 1959; Zuelzer and Brown,1961a). No cases of this type were seen in thepresent series and this may be due to early andadequate transfusion therapy.Haemorrhagic disease appears to be a frequent

complication of this condition and may be presentat birth. It is probably an expression of liverdamage. Besides hypoprothrombinaemia, theremay be a deficiency of convertin and other coagula-tion factors (van Creveld, 1959). Vitamin K, orpreferably K1 (Jouvenceaux et al., 1959), shouldbe given by intramuscular injection at birth andrepeated if necessary. Fresh blood or transfusionwith triple strength plasma may be required. It ispossible that citrated blood is safer than heparinizedfor transfusion in these cases.When contemplating exchange transfusion for

hyperbilirubinaemia in these infants, the direct-reacting fraction of the serum bilirubin was dis-counted as a factor in the causation of kernikterus.In the two cases of kernikterus reported above,occurring after three and two exchange transfusionswithin 48 hours of birth, the 'indirect' serumbilirubins, when symptoms first appeared, wereinitially reported to be below 20 mg./100 ml.However, re-estimation showed them to be wellabove this level. While there is a natural reluctanceto transfuse such seriously ill babies unnecessarily.it does appear that they are at especial risk fromkernikterus. This may be due to the associatedpathology or to coexistent complications causinghypoxia (Zuelzer and Brown, 1961b). Finally, thevery high levels of serum bilirubin that frequentlyoccur may lead to errors in estimation.The early use of corticoid in this condition was

thought to be of benefit, though this opinion wasonly based on clinical impression. Many reportsare available on the use of these drugs in bothobstructive jaundice and hepatitis (Summerskill,Clowdus, Bollman and Fleisher, 1961; Williams andBilling, 1961), in cirrhosis (Effersoe and Kjerulf,1962), and in hyperbilirubinaemia of prematurity(Crosse and Corney, 1961; E. E. Hill, 1961, personalcommunication). The general consensus of opinionsuggests that corticosteroids both reduce jaundiceand improve liver function and are of especialuse when there is liver cell damage. The continuedempirical use of this drug when cholestasis is presentseems justified.The microscopical appearance of the liver in this

condition has been fully recorded (Hawksley andLightwood, 1934; Craig, 1950; Cameron and Hou,1962). The findings in the two babies that died(Case 7 is reported above) appeared to be typical.In addition to necrosis, giant-cell regeneration and

59

copyright. on M

arch 25, 2020 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.38.197.54 on 1 February 1963. D

ownloaded from

60 ARCHIVES OF DISEASE IN CHILDHOOD

other evidence of hepatic damage, fibrosis may be The severity of the haemolytic disease in thesepresent at birth (Henderson, 1942; Cornblath et al., babies, associated with a high mortality and com-1955) or appear in the succeeding weeks (Hawksley plications such as kernikterus and haemorrhagicand Lightwood, 1934). The microscopical appear- diatheses, is stressed. Evidence is produced toance of the liver of Case 7 (Fig. 4) is contrasted with indicate that the obstructive jaundice in these casesthat of another baby of 35 weeks' gestation, who had is due primarily to liver cell damage.Rh haemolytic disease but no obstructive jaundice Treatment with corticosteroids appears to be(Fig. 3). The latter shows a normal lobular beneficial.architecture in spite of the presence of excessiveextramedullary erythropoietic tissue. No bile pig- I am greatly indebted to Dr. B. S. B. Wood for hisment or thrombi are present. This baby died from help and direction; also for both his and Dr. B. D.the respiratory distress syndronrie. Bower's permission to report their cases, and to thethe resirtoy isresynroobstetricians who supervised their antenatal care; toIn haemolytic disease of the newborn the liver, Dr. R. Gaddie and Dr. W. Weiner and their departmentswhich is the seat of excessive erythropoiesis, receives for their assistance with the biochemical investigations.blood rich in maternal antibodies straight from and to Dr. H. G. Kohler and Dr. A. H. Cameron forthe placenta. Whether the resulting intense haemo- their advice on the post-mortem material; finally, toMiss V. Macdonald for her secretarial assistance.lytic reaction damages the liver or whether the The author was in receipt of a grant from the Uniteddamage is caused by anoxia following anaemia or Birmingham Hospitals endowment fund.congestive cardiac failure, is not known. Certainlygross functional and often anatomical liver cell REFEREN-CESdamage is present in these cases. The structural Allen. F. H.. Jr.. Diamond. L. K. and Watrous. J. B. Jr. (1949).abnormality common to all forms of cholestasis, Ervthroblastosis fetalis. V. The va!ue of blood from femaledonors for exchange transfusion. New Engil. J. Med.. 241. 799.seen only under the electron microscope, is an altera- Billing. B. H. and Lathe. G. H. (1958). Bilirubin metabolism in

judc.Amer. J. Mfed., 2A. I111.tion of the bile canaliculi (Schaffner, 1960). Necro- Cameron. G. R. and Hou. P. C. (1962). Biliarv Cirrhosis. Oliversis of the liver cells involves rupture and destruction and Boyd, Edinburgh and London.Cornblath. M.. Kramrer. 1. and Kelly. A. B. (1955). Rh isoimmuniza-of the associated canaliculi which are merely modi- tion associated with regurgtation jaundice beginning in utero:

ficationsof the walls of the liver cells (Cameron a report of two patients. Amer. J. Dis. Child.. 90. 628.fications Cra~~~~~~~~~~~~ig.1J. M. (1950). Sequences in the development of cirrhosis ofand Hou, 1962). It is suggested that once cellular the liver in cases of erythroblastosis fetalis. Arch. Path.49, 665.van Creveld. S. (1959). Coagulation disorders in the newbornbreakdown has occurred, protein may enter the period. J. Pediat.. 54. 633.canaliculi and produce a precipitation of the bile Crosse. V. M. and Corney, G. (1961). The use of adrenocortico-trophic hormone in neonatal jaundice in premature babies.components (MacClure, 1931; Eppinger, 1937). Proc roy. Soc. Med.. 54. 737.

Effersoe. P. and Kierulf. K. (1962). Cirrhosis of 'iver; effect ofThe resulting bile thrombi may secondarily block prednisolone on frequency of biochemical rlapse. Lancet.the bile passages. It is worth recalling that as long ago Eppinger. .(1937). Die Lerkrankheiten. Soringer. Wienas 1933 Parsons, Hawksley and Gittins appreciated Harris, R. C.. Andersen. D. H. and Day. R. L. (1954). Obstructive

jaundice in infants with normal biliary tree. Pediatrics. 13. 293.that hepatic damage and obstructive jaundice might Hawksley. J. C. and Lightwood. R. (1934). A contribution to theoccur in infants with severe erythroblastosis foetalis. study of erythroblastosis: icterus gravis neonatorum. Quart.J. .Sfed.. 27. 155.While in most cases the prognosis appears to be Henderson. J. L. (1942). A fourth type of erythroblastosis foetalis

showing he:,atic cirrhosis in the macerated foetus. A reportgood (Hsia et al., 1952), as long as obstructive of three cases. Arch. Dis. Childh_. 17. 49.jaundice persists it remains in doubt (Lightwood Hsia. D. Y. Y. Patterson. P.. Allen. F. H.. Diamond. L. K. andGellis. S. S. (1952). Prolonged obstructive jaundice in infancy.and Bodian, 1946). Some of the most severe cases 1. General survey of 156 cases. Pediatrics, 10. 243.Ivemark. B. 1., Hogmfian. C.. Rudert. P.O. and Andersen. B. (1959).which, because of improved techniques, are now Kell-iso-immunization as the cause of fatal erythroblastosissurviving may well progress to cirrhosis, hepatic fetalis. Acta path. microbiol. scandd., 45, 193.surviving may ~~~~~~~~~~~Jouvenceaux, A.. Brizard. C. P.. Michaud, D. and Revol. L. (1959).failure and portal hypertension. Treatment of haemolytic disease of the newborn due to Rh-

immunization. Brit. med. J.. 2. 336.King. E. J. and Coxon. R. V. (1950). Determination of bilirubin

with preci:itation of the plasma proteins. J. clin. Path., 3. 248.zflISX~ Kove. S.. Goldstein. S.. Perry. R. E. and Wroblewski. F. (1959).

Eighteen cases of obstructive jaundice diagrosed Diagnosis of neonatal jaundice by serum transaminase. Arch.Eighteencass of obstruc ivejudc igoe Suirg.. 78. 157.in the neonatal period amongst 6,854 babies born Lathe. G. H. (1955). Exchange transfusion as a means of removingb,lirubin in haemolytic disease of the newborn. Brit. med. J.,at two maternity units during 1960-61 are reviewed. 1. 192.

haemo- an~d Ruthven. C. R. J. (1958). Factors affecting the rate ofOf these cases, 14 were associated with Rh haemo counuling of bBirubin and conjugated bilirubin in the van denlytic disease. The incidence of obstructive jaundice Bergh r,eacton. J. cli'i. Path.. 11. 155.

,Levine. P. (1943). Scrological factors as possible causes in sponin the latter appears to be 8-10%. Their common taneous abortions. J. Hered., 34. 71.clinical and biochemical features are analysed and Lightid wRthadt BoianvM. (na46 uBiliary obstDictionlassodiscussed. The diagnosis is also discussed and the 21, 209.MacClure, E. (1931). OJber icterus neonatorum gravis. Z. Kinder-importance of early detection by differential serum heilk., 51, 86.bilirubin studies is emphasized. Nakai. H. and Landing, B. H. (1961). Factors in the genesis ofbilirubin studiesis emphasized. ~~~bile stasis in infancy. Paediatrics, 27. 300.

copyright. on M

arch 25, 2020 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.38.197.54 on 1 February 1963. D

ownloaded from

OBSTRUCTIVE JAUNDICE AND HAEMOL YTIC DISEASE 61Oppe. T. E. and Valaes. T. (1959). Obstructive jaundice and haemo- Tisdale. W. A.. Klatskin. G. and Kinsella. E. D. (1959). The sirni-

lytic disease of the newsborn. Lancet. 1. 536. ficance of the direct-reacting fraction of serum bilirubin inParsons, L. G.. Hawksley. J. C. and Gittins. R. (1933). Studies in hersolytic jaundice. Amer. J. Med.. 26. 214.

the anaemias of infancv and early childhood. IV. The haemo- Tizard. J. P. M. (1962). Portal hvpertension follosing exchangelvtic (erythronoclastic) anaemias of the neonatal period; with transfusion throuigh the umbilical vein. Proc. roY. Soc. Med..social reference to erythroblastosis of the newborn. Arch. Di'. 55. 772.Childh.. 8. 159. Walker. W. and Mollison. P. L. (1957). Haemolytic disease of the

Schaffner, F. (1960). latrogenic jaundice. J. Amer. med. Ass.. nesborn. Deaths in England and Wales during 1953 and 1955.174, 1690. Lan:et. 1. 1309.

Shaldon. S. and Sherlock. S. (1962). Obstruction to the extra- White. D.. Haidar. G. A. and Reinhold. J. G. (1958). Spectro-henatic portal system in childhood Lacet. 1. 63. photometric measurement of bilirubin concentrations in the

Stempfel. R.. Broman, B.. Escardo. F. E. and Zetterstro.P.P 1956). serum of the newborn by the use of a microcapillary method.Obstructive jaundice comotlicating hemo'ytic disease of the Clin. Chem.. 4. 211.newborn. Pediatrics. 17, 471. Williams. R. and Billing. B. H. (1961). Action of steroid therapy in

Still. G. F. (1927). Common Disorders and Diseases of Childhool. jaundice. Lancet. 2. 392.5th ed., p. 375. Oxford University Press. London. Zuelzer, W. W. and Brown. A. K. (l96la). Neonatal jaundice.

Summerskill. W. H. J.. Clowdus. B. F-. Bollman. J. L. and Fleisher. Amer. J. Dis. Child.. 101. 87.G. A. (1961). Clinical and exoerimental studies on the effect (l961b). Studies in hynerbilirubinemia. IIJ. Separateof corticotrophin and steroid drues on bilirubinen-.ia. .4mer. metabolic defects in premature infants reflected in the partitionJ. med. Sci.. 241. 555. of serum bihirubin. ibid.. 102. 815.

copyright. on M

arch 25, 2020 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.38.197.54 on 1 February 1963. D

ownloaded from