- Email: [email protected]
HEPATITIS B IMMUNE GLOBULIN (HBIG) EFFICACY IN THE INTERRUPTION OF PERINATAL TRANSMISSION OF HEPATITIS B VIRUS CARRIER STATE
388
permit speedier assessment. On the other hand, bone damage to length of diving experience and at present we see a predominantly young group of men with relatively short
HEPATITIS B IMMUNE GLOBULIN (HBIG) EFFICACY IN THE INTERRUPTION OF PERINATAL TRANSMISSION OF HEPATITIS B VIRUS CARRIER STATE
is related
diving. When substantial numbers have remained long enough in the industry, will we see the development of careers
in
femoral head lesions? If so, without routine radiography, would evidence of this be the sudden onset of a stiff and painful hip joint, and what would be the medicolegal implications? There are no direct complications of shaft lesions, but they must be noted in case a diver with a shaft lesion is found to be abnormally susceptible to bone damage elsewhere. If bone scintigraphy should be effective in the early detection of bone necrosis, one solution might be a full radiological examination of the limb bones at the start of a diving career followed by a regular bone scan. When it is suspected that damage may have occurred, such as after a decompression accident, an additional bone scan should be carried out and further radiographs taken of those areas which appear suspicious or positive on the scan.
Initial
more
Placebo-Controlled Trial LU-YU HWANG CLADD E. STEVENS TSU-SHEN SUN WOLF SZMUNESS
R. PALMER BEASLEY CHIA-CHIN LIN KWEI-YU WANG FON-JOU HSIEH
University of Washington Medical Research Unit (UWMRU), Taipei, Taiwan; Institute of Public Health, College of Medicine, National Taiwan University; Lindsley F. Kimball Research Institute of the New York Blood Center, New York, New York; Maternity and Child Health Center of the Taiwan Provincial Junior College of Nursing, Taipei, Taiwan and U.S. Naval Medical Research Unit (NAMRU-2), Manila, Philippines A randomised double-blind placebo-controlled efficacy trial of hepatitis B immune globulin (HBIG) for prevention of the vertically transmitted HBsAg carrier state was conducted in Taiwan where the carrier rate in the general population is 15-20%. HBIG was given immediately after birth to infants of e antigen positive HBsAg carrier mothers and all infants were followed for 15 months or more. Among 35 placebo-treated infants the carrier rate was 91%. This compares with the carrier rate of only 23% among 40 infants who received 0· 5 ml HBIG at birth, three months, and six months and the 45% carrier rate among 42 infants receiving a single 1 0 ml dose of HBIG at birth only. Efficacy was 75% and 45% respectively for the two treatment schedules. Significantly, the most common response of HBIG-treated infants was passive-active immunisation.
Summary
Recommendations At present, commercial divers should have regular skeletal surveys of their long bones by conventional radiography as set out by the M.R.C. Decompression Sickness Panel (1981).21 We recommend that all commercial divers should have an initial skeletal survey which should be repeated annually, but with two exceptions: l.-Annual radiographs would not be required for divers using only compressed air who had not been exposed to pressure above 30 m seawater and whose total exposure time on any occasion had never exceeded 4 h. 2.-Radiographs at intervals of 3 yr would suffice for divers using only compressed air who had not been exposed to pressures over 50 m seawater and whose total exposure time on any occasion had never exceeded 4 h. Any diver who had had compression treatment for decompression sickness would be excluded from the above two
Report of a Randomised Double-Blind
Introduction
categories.
We thank all those doctors who issued certificates of medical fitness, in connection with the Diving Regulations, for supplying data, and Mrs D. Weightman, University of Newcastle upon Tyne, for statistical advice. We received grants and help from the Medical Research Council, the Health and Safety Executive, and the CIRIA Underwater Engineering Group.
Requests for reprints should be addressed to Prof. R. I. McCallum, Department of Occupational Health and Hygiene, University of Newcastle upon Tyne, 21 Claremont Place, Newcastle upon Tyne NE2 4AA.
TRANSMISSION of hepatitis B virus (HBV) from mother to infant in the perinatal period is common in East Asia and
Aseptic bone necrosis in clearance divers. Report no 1/74 Royal Navy Clinical Research Working Party. Alverstoke: Institute of Naval
10. Harrison JAB, Elliott DH.
Medicine, 1974. 11.
Trowbridge WP, Evans A, Walder bone necrosis (ABN) indicate
DN. Does the presence of shaft (B) lesions of aseptic increased susceptibility to juxta-articular (A)
an
lesions? Undersea Biomed Res 1979, 6: 19-20. REFERENCES 1. Harrison JAB.
Aseptic bone necrosis in naval clearance divers: radiographic findings.
Washington State (USA) decompression tables. In: Trapp WG, Bannister EW, Davison, AJ, Trapp PA, eds. Fifth international hyperbaric congress proceedings,
Proc
Roy Soc Med 1971; 64: 1276-78. Sphar RL, Harvey CA. Aseptic bone necrosis among US Navy divers: survey of 934 non-randomly selected personnel. Undersea Biomed Res
2. Hunter WL Jr, Biersner RJ,
3. Griffiths PD. An exposure to risk registry for compressed-air workers.
5.
6. 7.
8. 9.
Trans Soc Occup Med 1971; 21: 123-25. Davidson JK. Dysbaric osteonecrosis. In: Davidson JK, ed. Aseptic necrosis of bone. New York: American Elsevier Publishing Co., 1976: 147-212. Davidson JK, Harrison JAB, Jacobs P, Hilditch TE, Catto M, Hendry WT. The significance of bone islands, cystic areas and sclerotic areas in dysbaric
osteonecrosis. Clin Radiol 1977; 28: 381-93. Golding FC. The shoulder-the forgotten joint. Br J Radiol 1961; 35: 149-58. Diving operations at work regulations. London: HMSO, 1981. The merchant shipping (diving operations) regulations. Statutory instruments no 116. London: HMSO, 1975. Walder DN. Aseptic necrosis of bone. In: Strauss RH, ed. Diving medicine. New York: Grune and
Stratton, 1977;
Simon Fraser University, Burnaby, Canada. 1974: 904-11. Gregg PJ, Walder DN. Early diagnosis of dysbaric osteonecrosis. In: Pearson RR, ed The twelfth undersea medical society workshop (UMS report no 7-30-77). Earlv diagnosis of decompression sickness. Bethesda, USA. 1977: 268-76. 16. Weatherley CR, Dale G, McGurk J, Walder DN. Bone necrosis and urinary hydroxyproline excretion in rabbits, Clin Sci Mol Med 1977; 52: 523-26. 17. Gregg PJ, Eastham EJ, Bell JI, Walder DN. Serum ferritin and dysbaric osteonecrosis. 15.
1978; 5: 25-36.
4.
Suppl.
12. Williams B, Unsworth I. Skeletal changes in divers. Aust Radiol 1976; 20: 83-94 13. Ohta Y, Matsunaga H. Bone lesions in divers. J Bone Joint Surg Br 1974; 56: 3-16 14. Walder DN, McCallum RI. An objective appraisal of the Blackpool (UK) and
97-108.
Undersea Biomed Res 1977; 4: 75-79. 18. Stothard J. Caisson disease of bone—an experimental study of its cause, early diagnosis and management. MD thesis. University of Edinburgh, 1980. 19. Gregg PJ. The use of radioisotopes in aseptic bone necrosis. In: Evans A, Walder DN, eds. Aseptic bone necrosis. London: CIRIA Underwater Engineering Group, 1977 77-83. 20. Michael RH, Dorfman HD. Malignant fibrous histiocytoma associated with bone infarcts. Report of a case. Clin Orthop 1976; 118: 180-83. 21. MRC Decompression Sickness Panel. Radiological skeletal survey for aseptic necrosis of bone in divers and compressed-air workers. Radiography 1981; 47: 141-43
389
other parts of the world where hepatitis B surface antigen (HBsAg) carrier rates are high. 1-6 In Taiwan, approximately 40% of the offspring of carrier mothers are so infected and, when the mother is hepatitis-B e-antigen (HBeAg) positive, 85-95% are infected.7-9 Among infants infected in the perinatal period, approximately 95% become carriers: among carriers in Taiwan we estimate that 35-50% were infected perinatally by their mothers. The infants usually are healthy,I although serum transaminase levels sometimes rise slightly. However, increasing evidence suggests that HBsAg carriers are at considerably increased risk of chronic hepatitis, cirrhosis, and primary hepatocellular carcinoma. 10-14 From 1974 to 1978 we conducted a randomised placebocontrolled double-blind trial of the efficacy of hepatitis B immune globulin (HBIG) in the prevention of perinatal transmission among Chinese in Taiwan.9 A single dose of 02 ml of HBIG with a passive haemagglutination (PHA) titre of 1:100 000 was administered to the infant within 7 days of birth. The results showed a delay in appearance of HBsAg in infants who received HBIG but essentially the same frequency of HBsAg carriers by 12 months of age. However, infants treated within 48 hours of birth had considerably lower HBsAg persistence rates.. On the basis of this observation, we decided to conduct a second trial in which HBIG would be administered as soon as possible after birth. We also decided to use larger doses and to compare single with multiple dose schedules in the hope that HBsAg persistence would be lower if the onset of antigenaemia were delayed. We now report preliminary results of this second trial.
Subjects and Methods Study Population and Case Recruitment Women attending the prenatal clinic at the Maternity and Child Health Center of the Taiwan Provincial Junior College of Nursing in Taipei, Taiwan, were screened for HBsAg by reverse passive haemagglutination (RPHA). Since HBeAg testing was not regularly available at the time, we selected those with an RPHA titre >1: 1024 (i.e., those most likely to be HBeAg-positive and to transmit the virus) and interviewed them during their next regular prenatal visit at which written consent to join the study was obtained. Only infants delivered at the Maternity and Child Health Center weighing .2000 g with Apgar scores 8 at 1 minute were eligible to the study. 430 mothers agreed to join; 144 of their infants were subsequently not included either because they were delivered at other hospitals, or treatment preparations were not available, or because the infants had a low birth weight or low Apgar score. Because transmission to mfants is low from HBeAg-negative mothers (even when the RPHA titre is 26 treated infants whose mothers were later found to be HBeAg-negative were excluded from the analysis. Of 213 infants who received the initial injection, 202 completed the three-dose series. This report concerns the 117 infants who were 15 months old or more at the time of this analysis. enter
high),8
HBIG and Placebo Preparations and Treatment Schedule Infants were divided into three
treatment
groups:
Group A: given 10ml saline at birth, three, and six months; Group B: given 1.0 0 ml HBIG at birth, and saline at three and six months;
Group C: given 0’ 5 ml HBIG diluted in 0’ 5 ml of immune serum globulin (ISG) (for a total volume of 1 -0ml) at birth, three, and six months. The HBIG, provided by Abbott Laboratories, (commercially available, lots 7307 SF and 79070 SF2 with PHA titres of 1:204 000 and 1: 150 000 respectively) was repackaged in 1 ml vials and
code numbers at the New York Blood Center. The vials stored at 4°C. The initial preparations were given to study infants in the delivery room when it had been determined that the birth weight and Apgar score criteria were satisfied. The mean administration time was 36 minutes after birth: all were treated within 30 hours and 95% were treated within an hour.
assigned
were
Laboratory Methods Laboratory tests were performed on coded specimens. Initial screening of women for HBsAg was by RPHA (Organon ’Hepanosticon’) in the maternity hospital laboratory. Positive sera were rescreened at a 1:1024 dilution, titred to extinction by RPHA and tested for HBeAg by radioimmunoassay (RIA). Infants’ specimens were tested for HBsAg, anti-HBs, anti-HBc, and HBeAg by RIA (’AUSRIA-II’, ’AUSAB’, ’CORAB’, and HBeAg diagnostic kits respectively, Abbott Laboratories, North Chicago, 111.). Although anti-HBs testing of these specimens was the study, the results were not linked to the individual in order to avoid breaking the code inadvertently. Selected HBsAg-positive sera and anti-HBs positive sera with rising antibody levels were tested for anti-HBc to confirm active HBV infection. Specimens from infants of 12 months and older which were negative for both HBsAg and anti-HBs were also tested for anti-HBc. Serum aspartate or alanine aminotransferase (AST or ALT) levels were measured using a kinetic photometric method (ABA-50, Abbott Laboratories, South Pasadena, Calif.).
performed during
Cord Blood Collection Cord blood specimens were collected by venepuncture of the umbilical cord before delivery of the placenta: the surface of the cord was wiped clean with an alcohol swab and allowed to dry.
Follow-up Follow-up visits were scheduled at 3-monthly intervals until the infants were 18 months old with a final visit at 24 months. Strict adherence to the follow-up schedule was maintained; every family was asked to return within 3 days of the target date and those who did not were visited in their homes by project nurses. Among the 117 study infants, 58% had a perfect follow-up record, while 95% missed two specimens or less. The 3-month and 6-month injections were permitted during the 30 days following the scheduled date. If the treatment preparation had not been administered by that time, the baby was removed from the study and defined as lost or refused. In practice, adherence was excellent; the second injection was scheduled for 3 months and the actual time between the first two injections (mean ± standard deviation) was 99 ± 8 days; the third injection was scheduled for 6 months of age and the actual time was 190 ± 10 days. The code was first broken in August, 1980. Nurses responsible for continuing follow-up will remain blind until the final analysis.
Definition of HBV Events HBV
infection.-An infant was considered infected if any specimen (excluding the cord blood) was HBsAg-positive or if the infant was anti-HBs positive at 12 months or older. Anti-HBc results were taken to confirm infection, but by themselves were not considered part of the criteria for infection because virtually all infants of HBsAg carrier mothers are anti-HBc positive under 12 months (passive anti-HBc from the mother) and none of the infants in this study were positive only for anti-HBc after 12 months of age. Persistent HBs antigenaemia.-An infant was considered to have persistent HBsAg if serum specimens were HBsAg-positive for 6 months or more and all subsequent specimens were HBsAgpositive. Transient HBV infection.-An infant was considered to have had a transient infection if at any time during follow-up the serum was
390
and antigenaemia subsequently disappeared but anti-HBc and anti-HBs were detectable. Infants who had no HBsAg-positive sera but were anti-HBs positive and anti-HBc positive at 12 months or older were also considered to have had a transient infection with HBsAg, probably missed by the timing of routine specimen collection. Uncertain Cases.-3 infants (all in group C) had persistent antiHBs at 15 months or older, but were anti-HBc negative. One additional infant (group B) had a single HBsAg-positive sample, but no serological confirmation of infection on follow-up. None of these infants was counted as having had an HBV infection. Three additional infants (1 in group B and 2 in group C) were HBsAg positive for the first time when the last specimen was obtained; these were counted as infected but not as persistent and because they are as yet unclassifiable they are not shown in fig. 1.
HBsAg-positive
Methods
TABLE I-MATERNAL AND INFANT CHARACTERISTICS BY STUDY
GROUP
(ALL ENROLLEES)
of Analysis
The characteristics of the
treatment groups were
compared for
7WfP/’///’ff/f -’
*Log 2. Proteus mirabilis was isolated from the spinal fluid of one infant while no organism was isolated from the second. Both infants recovered with antibiotic treatment. There was no reason to attribute the illnesses to the HBIG treatment.
ZZ! Anti-HBc and Anti-HBs only
Transient HBsA6
1
Results
- Persistent HBsA6
Comparability of Study Groups The three study groups were compared for factors which might have affected the risk of HBV infection or the effectiveness of HBIG; as can be seen in table I, all differences were small and not statistically significant. Characteristic HB V Infection Patterns by Study
Infants age (months) of HBV infection by study group, age of infant when infection was first detected serologically, and outcome of infection.
Fig. 1-Temporal patterns
Group
Placebo Recipients.-All but 2 of the 35 placebo recipients were infected and all of these were HBsAg-positive by 3 months of age. In all but one infected infant, every subsequent specimen was HBsAg-positive; the single infected baby who did not become a carrier lost its HBsAg and developed anti-HBs by the sixth month (figs. 1, 2). HBIG Recipients.-Two patterns of infection occurred in HBIG recipients (fig. 1):
(1) Transient HBV infection’occurred in 30; in 11 of
these HBsAg
A=placebo; B=single dose HBIG; C=three doses HBIG. Arrow (y) indicates the times injections were given. Three infected individuals in table I are not included here because only their last specimen was HBsAg positive and therefore they cannot yet be classified.
statistical significance using the Student’s t test. HBIG efficacy was determined by comparing the proportion of infants with persistent HBsAg in each group using Fisher’s exact test. The times of onset of infection in infants with persistent antigenaemia were also compared using the life table method of analysis. As an approximation for this analysis, the time of onset was defined as the age of the infant at the time HBsAg was first detected serologically.
Inadvertent Events One of the study infants died at 6 days of age while still in the hospital. Labour and delivery had been uncomplicated, birth weight was 3700 g and the 5-minute Apgar score was 9. The infant had received 0 - 5 ml ofHBIG lot 79079 SF2. No antecedent event to death was recognised. A necropsy was not performed. In 2 infants neonatal meningitis developed during late June and early July, 1980, when similar cases were recognised in several nonstudy infants in the nursery. One study infant had received 0 - 5 ml of HBIG and the other 1.0 ml of HBIG (both lot 79079 SF2).
explain the
Infants age (months) Fig.2-Cumulative onset of chronic HBV carrier state by study group.
(Babies followed .15 months
- life table
analysis.)
391
appeared, usually in low titre, from 6 to 12 months of age (fig. 3a). HBsAg then disappeared and high titre anti-HBs developed and was present in all subsequent specimens. In the remaining 19 infants with transient infections, HBsAg was never detected but was presumed to have been present for a short period between blood specimens (fig. 3b). After an initial decline (loss of passive antibody), high titre anti-HBs developed in these infants (usually at 9 to 12 months) and remained present at high levels in all subsequent specimens. (2) Persistent antigenaemia developed in 30 HBIG-treated infants (fig. 3c). These cases differed from those in the placebo group in that HBsAg was usually first detected at a later age (fig. 2). There was no way to distinguish infants who would become chronic HBsAg
TABLE II- HBV INFECTION AND
HBsAg PERSISTENCE
BY STUDY GROUP
p VALUE BY FISHER’S EXACT TEST
carriers from those who would have transient infections except that their HBsAg levels when first detected were generally higher than in infants with transient antigenaemia.
Infants who failed to become infected during the first 15 months of life included only 2 in group A (5’ 7%), and a somewhat higher proportion of HBIG-treated infants (21% and 25% respectively for infants in groups B and C). When passive antibody was no longer detectable in HBIG-treated infants, their sera were negative for all HBV markers including anti-HBc and they were presumably still susceptible to HBV.
HBIG Efficacy The
primary goal
of HBIG
HBIGsuccesstransient antigenemia
treatment
HBIG no
in
perinatal
success-
antigenemia
1
Case 15 GroupC HBIG 0-5mlx3
Case 66
6raup C HBlG 0.5ml x 3
Age in months
Age in months
(a)
(b)
HBI6 failure
No infection
transmission of HBV is prevention of HBsAg persistence, rather than prevention of clinical hepatitis or HBV infection per se because clinical hepatitis is rare following perinatal transmission and HBV infection is desirable if it is subclinical, transient, and results in permanent immunity. Of the 96 infants who had HBV infections during the study, 62 developed persistent antigenaemia. As shown in table II, 91% of the placebo recipients developed persistent HBsAg. In contrast, only 50% of those who received a single dose of’ HBIG and 23% of those who received three doses developed persistent HBsAg. From these data the efficacy of the three 0’ 5 ml doses of HBIG is calculated to be 75% and the single 1 0 ml dose to be 45%. The differences between each HBIG group and the placebo group and between the two HBIG groups are highly significant statistically (P<0. 01). Of considerable interest is the fact that in the majority of infants treated with HBIG an active immune response develops indicating that HBIG prevents the carrier state, while allowing passive-active immunisation to occur. Among the HBIG-treated infants who did not develop persistent antigenaemia, 11 of 21 in group B and 19 of 31 in group C developed persistent anti-HBs in levels exceeding those achieved by passive immunisation. Among the placebo recipients, only 3 did not develop persistent HBsAg; one was transiently HBsAg-positive followed by persistent high titre anti-HBs and the others remained uninfected. Clinical hepatitis occurred in only one infant, one who had received the placebo and was jaundiced at 54 days after birth at which time HBV markers were not tested. HBsAg was present at the routine 3-month follow-up and remained positive through the last follow-up at 18 months. Aminotransferase levels were slightly elevated until 9 months and have been normal thereafter.
Case 55
Case 90 Group C 3 HBlG 0.5ml x
GroupB HBlG 1.0ml xl
Age in months
Age in months
(c)
(d)
Fig. 3-Temporal anti-HBs response patterns in HBIG recipients by outcome, age of infant, and anti-HBs titre (SD units).
of Onset of HBV Infection by Treatment Group Fig. 1 shows the onset of HBsAg positivity for infants who developed HBsAg persistence. This figure shows not only differences in the cumulative persistence rates for infants in Time
each of the three groups, but differences in the time of onset as well. All of the chronically infected babies in the placebo group were HBsAg-positive by the time of the first follow-up specimen at 3 months. Among the infants who received a single dose of HBIG at birth, new cases appeared at each follow-up bleeding over the first year with no additional ones thereafter. All of those who had received HBIG three times in 6 months and became HBsAg carriers were, with only three exceptions, HBsAg-positive for the first time at 12 months of
392 age with
thereafter. The infant’s age when first detected correlated with persistence. Among HBsAg 38 infants who were first positive at 3 months, 97% developed persistent antigenaemia, compared with 25 of 36 (69%) of those first positive thereafter. This difference was statistically no new cases
was
significant. Infants in the two HBIG treatment groups were evaluated for factors which might have predicted the infection outcome-maternal HBsAg titre, cord blood HBsAg positivity, age at first treatment, and number of older siblings. No predictive factors were found. The absence of correlation of cord blood (CB) positivity with outcome among HBIG treated babies is especially noteworthy. Among 66 CB-negative infants 20 (30-3%) became carriers, while among 16 who were CB-positive, 6 (37 - 5%) became carriers.
Discussion HBV infection early in life often leads to the HBsAg carrier
state. 1,3,15 Thus, interruption of early transmission from infant is extremely important. This study’s unequivocal results founded on a randomised double-blind placebo-controlled design, large number of high-risk infants, and long-term follow-up provided strong evidence of the value of HBIG in reducing the incidence of the HBsAg carrier state following perinatal transmission. A single dose at birth was 45% successful and the three-dose series was 75% successful in preventing persistent antigenaemia. Both results are highly significant statistically. Previous evidence on HBIG protection in perinatal transmission includes single case reports, small series without controls and one small nonrandomised unblind series with limited follow-up. 16-20 Our first study in Taiwan was the largest previous trial of HBIG efficacy in infants of carrier mothers.9 That study, like the present one, was randomised, double-blind, and placebocontrolled. In it we gave infants a single 0’ 2 ml dose of HBIG (average 0’ 06 ml/kg) up to 7 days after birth (at an average time of 46 hours). We found a delay in the onset of infection but only a slight reduction in the carrier rate among HBIG recipients compared with controls. The mean age (hours) at the time of HBIG administration was significantly lower in infants with transient HBV infection suggesting that the timing of HBIG administration was critical. In another report concerning the efficacy of HBIG in vertical transmission, Reesink and his colleagues used a nonrandomised, unblind design to study the effect of large doses of HBIG at birth followed by monthly injections for 6 months.2O HBsAg was not detected during follow-up in any of 6 HBIG-treated infants born to HBeAg-positive mothers compared with 4 of 6 infants not treated with HBIG. As in our study, most of the HBIG-treated infants had evidence of HBV infection, developing anti-HBs during follow-up. The primary goal of our second trial was to decrease as far as possible the time of initial HBIG administration. The data from our previous study demonstrated the critical importance of the timing of the initial HBIG administration. Fewer infants who received HBIG under 48 hours of age developed persistent antigenaemia than those treated at 48 hours or more. In the present trial in which HBIG was administered at an average time of 36 minutes there was a substantial and highly significant reduction in persistent HBsAg among all HBIG recipients compared with placebo controls. However, because almost all infants in this study
- mother
to
received HBIG within 1 hour of birth, it is impossible to determine whether administration somewhat later, but within 48 hours, will also be efficacious. Presumably, however, the earlier administration occurs the better. A single dose of HBIG at birth reduced HBsAg persistence although not as effectively as a multiple-dose series. In many areas of the world where follow-up treatment may be difficult to organise, a reasonable strategy might be simply to give a single dose immediately at birth. Even if follow-up treatment is not feasible, the HBIG given immediately will be of value. The present high cost of HBIG may be thought a barrier to its use despite its obvious and unique value in perinatal HBV infection. However, the current cost of the three small doses (0 - 5 ml) recommended for this regimen should not be prohibitive in developed countries. HBIG is expensive partly because demand has been low and the required high titre donors are rare. High titre donors should be easy to create in the near future using the new HBV vaccines,21-22 and the efficacy of HBIG demonstrated in this study may increase the demand. Thus, the cost of HBIG should eventually fall, making its worldwide use feasible. This study unequivocally showed that the primary effect of HBIG in the newborn was prevention of HBsAg persistence rather than prevention of HBV infection per se. HBIG . recipients had somewhat lower infection rates than placebo recipients. However, HBIG recipients had more transient infections, while placebo recipients were much more likely to acquire a chronic antigenaemia. It is especially noteworthy that HBIG is equally effective in both cord blood negative and cord blood positive infants. This observation, and the fact that cord blood titres are usually quite low, suggest that a positive cord blood is usually a reflection of maternal-fetal transfusion occurring during labour and delivery. Ifa positive cord blood reflected in-utero viral transmission, HBIG should be ineffective in preventing the HBsAg carrier state, since HBIG has no value in eliminating the HBsAg carrier state. Immediate administration of HBIG is essential in the prevention of perinatal HBV infection suggesting that HBIG works initially by blocking or decreasing viral access to the liver. The very high attack rates in untreated babies and the rapid onset suggests high infectivity of the mother during labour and/or delivery. It is reasonable to believe that the baby of an HBeAg-positive mother not infected at birth remains at high risk from continuing exposure. The fact that multiple doses of HBIG work better than a single dose suggests that postponing the time of infection is advantageous in reducing the probability of persistent
antigenaemia. Some
infants
developed persistent HBsAg despite
HBIG prophylaxis. It seems unlikely that larger doses of HBIG at birth would have been more effective. In fact, in our first trial in Taiwan,9 a much smaller, 0 - 2 ml, dose of HBIG when given within 48 hours of birth produced reduction in chronic antigenaemia identical to that seen in the present study where a single dose five times larger was used. Whether more frequent or more prolonged prophylaxis would be more effective is not known. Reesink et al. 20 gave monthly injections ofHBIG for 6 months or longer to a group of 6 high risk infants. Although he observed no chronic infections among these children, because of the small numbers of infants treated his results do not differ statistically from those reported here among infants given only three injections. Another approach to prophylaxis may include use of the new hepatitis B vaccine.21In a recent efficacy trial, this
prolonged
393
vaccine has proven highly effective in preventing HBV infection.22 This study and earlier ones among chimpanzees also suggest that the vaccine may be effective in post-exposure prophylaxis.z3 If newborn infants have a vigorous, rapid immune response to the vaccine, the vaccine alone may prove effective. An alternative approach would be the combined use of both HBIG and vaccine; HBIG to provide immediate protection and delay the onset of infection until the infant developed an active immune response to the vaccine. Recent studies in adults revealed that antibodies from HBIG do not interfere with an active immune response to the vaccine.24 Options utilising vaccine alone or in a passive-active regimen must be explored in the future to carefully assess efficacy.
We gratefully acknowledge the support of Miss Pao-Tien Chu, President of the Taiwan Provincial Junior College of Nursing and the staff of the college. We specially thank Miss H. C. Lin, Director of Nursing, and the nursing staff who assisted in enrolment of mothers and administration of the initial injections. Our deep appreciation is also due to Miss Susan Chen, UWMRU Nursing Supervisor, who established and supervised our perinatal transmission and HBIG studies until her retirement in 1979. Miss Lucy Chiang, Miss Linda Lo, and Miss Eileen Chen, UWMRU nurses, made this project their full-time work for several years. We also thank Dr Saul Krugman, who served as consultant for the study. This research was supported by funds provided by NIH National Institute of Allergy and Infectious Diseases Contract C2101800 (HBG-2908) and the Bureau of Biologics, Food and Drug Administration. Requests for reprints should be addressed to R. P. B. at UWMRU, 15-2 Kung Yuan Rd, Taipei, Taiwan, ROC. _
REFERENCES 1. Stevens CE, Beasley RP, Tsui J, Lee WC. Vertical transmission of hepatitis B antigen in Taiwan N Engl J Med 1975; 292: 771-74. 2 Okada K, Yamada T, Miyakawa Y, Mayumi M. Hepatitis B surface antigen in the serum of infants after delivery from asymptomatic carrier mothers. J Pediatr 1975; 87: 360-63. 3 Lee AKY, Ip HMH, Wong VCW. Mechanisms of maternal-fetal transmission of hepatitis B virus. J Infect Dis 1978; 138: 668-71. 4. Woo D, Cummins M, Davies PA, Harvey DR, Hurley R, Waterson AP. Vertical transmission of hepatitis B surface antigen in carrier mothers in two west London hospitals. Arch Dis Childh 1979; 54: 670-75. 5 Derso A, Boxall EH, Tarlow MJ, Flewett TH. Transmission ofHBsAg from mother to infant in four ethnic groups. Br Med J 1978; i: 949-52. 6. Shiraki K, Yoshihara N, Kawana T, et al. Hepatitis B surface antigen and chronic hepatitis in infants born to asymptomatic carrier mothers. Am J Dis Child 1977; 131: 644-47. 7. Okada K, Kamiyama I, Inomata M, et al. e antigen and anti-e in the serum of asymptomatic carrier mothers as indicators of positive and negative transmission of hepatitis B virus to their infants. N Engl J Med 1976; 294: 746-49. 8 Beasley RP, Trepo C, Stevens CE, Szmuness W. The e antigen and vertical transmission of hepatitis B surface antigen. Am J Epidemiol 1977; 105: 94-98. 9. Beasley RP, Stevens CE. Vertical transmission of HBV and interruption with globulin. In. Vyas GN, Cohen SN, Schmid R, eds. Viral hepatitis. Philadelphia: Franklin Institute Press, 1978: 333-45 10. Nishioka K, Hirhyama T, Sekine T, et al. Australia antigen and hepatocellular carcinoma. Gann Monogr Cancer Res 1973; 14: 167-75. 11 Prince AM, Szmuness W, Michon J, et al. A case/control study of the association between primary liver cancer and hepatitis B infection in Senegal. Int J Cancer 1975; 6: 376-83. 12 Ohbayashi A, Okochi K, Mayumi M. Familial clustering of asymptomatic carriers of Australia antigen and patients with chronic liver disease and primary liver cancer Gastroenterology 1972; 62: 618-25. 13 Szmuness W. Hepatocellular carcinoma and the hepatitis B virus: Evidence for a causal association. Prog Med Virol 1978; 24: 40-69. 14. Beasley RP, Lin CC, Hwang LY. The risk of hepatocellular carcinoma in hepatitis B virus infection: a prospective study m Taiwan. In: Szmuness W, et al, eds. Viral hepatitis. Philadelphia: Franklin Institute Press (in press). 15 Schweitzer IL, Dunn EAG, Peters RL, et al. Viral hepatitis in neonates and infants. Am J Med 1973; 55: 762-71. 16 Kohler PF, Dubois RS, Merrill DA, et al. Prevention of chronic neonatal hepatitis B virus infection with antibody to the hepatitis B surface antigen. N Engl J Med 1974; 291: 1378-80. 17. Varma RR. Hepatitis B surface antigen carrier state m neonates. JAMA 1976; 236: 2302-04. 18 Dosik H, Jhaveri R. Prevention of neonatal hepatitis B infection by high-dose hepatitis B immune globulin. N Engl J Med 1978; 298: 602-03.
LACTOBACILLI DO NOT CAUSE FREQUENCY AND DYSURIA SYNDROME W. BRUMFITT
J.
H. LUDLAM
M. T. HAMILTON-MILLER ANN GOODING
Department of Medical Microbiology, Royal Free Hospital, London NW3
Mid-stream specimens (MSU) of urine were collected from 142 healthy women (pregnant and non-pregnant) and cultured for lactobacilli and other fastidious bacteria. The latter either require CO2 or are obligate anaerobes. Lactobacilli were present in counts of 104/ml or more in 34·8% of the women, and in counts of 105/ml or more in 20·2%. Besides lactobacilli, which were the bacteria most frequently isolated, anaerobic grampositive cocci (peptococci and peptostreptococci) were often found. This flora is typical of that of the lower vagina, and none of these women had either symptoms of urinary infection or pyuria. Therefore, the bacteria isolated were commensals or contaminants. Cultures of MSUs taken from 26 women with symptoms of dysuria and/or frequency, but without significant numbers of conventional pathogens such as Escherichia coli, contained commensals and contaminants of the same variety and in similar numbers. Urine samples from 50% of these patients contained at least 104 lactobacilli/ml and 27% had 105 or more/ml. Lactobacilli were absent from the suprapubic urine specimens cultured from a further 44 women. There was no significant difference between the isolation rate of lactobacilli in urine cultures from healthy women and the rate in women with dysuria and
Summary
frequency.
Introduction
patients presenting with dysuria and/or a significant bacteriuria caused by conventional pathogens such as Escherichia coli (> 1 OS ONLY 50% of have
frequency
bacteria/ml).1,2
The illness with symptoms but without
significant bacteriuria is often called the urethral syndrome. However, this name has been rejected on the grounds that various workers have used the term in different contexts.33 Instead, the term "frequency and dysuria syndrome" has been recommended. This syndrome may be caused by organisms which are difficult to culture such as fastidious bacterial chlamydiae,S or even viruses.6 Corynebacterium
19. Matsumoto neonates
S, Tomigashi T, Fujmoto S, et al. Prevention of hepatitis B infection in by human anti-HBV immunoglobulin (HBIG). Perinatology 1979; 9:
615-22. 20. Reesink HW, Reerink-Brongers EE, Lafeber-Schut BJT, et al. Prevention of chronic HBsAg carrier state in infants of HBaAg-positive mothers by hepatitis B immunoglobulin. Lancet 1979; ii: 436-38. 21. Hilleman MR, Bertland AU, Buynak EB, et al. Clinical and laboratory studies of HBsAg vaccine. In: Vyas GN, Cohen SN, Schmid R, eds. Viral hepatitis. Philadelphia: Franklin Institute Press, 1978: 525-37. 22. Szmuness W, Stevens CE, Harley EJ, et al. Hepatitis B vaccine. Demonstration of efficacy in a controlled clinical trial in a high-risk population in the United States. N
Engl J Med 1980; 303: 833-41. 23. Purcell RH, Gerin JL. Hepatitis B vaccines: a status report. In: Vyas GN, Cohen SN, Schmid R, eds. Viral hepatitis. Philadelphia: Franklin Institute Press, 1978, 491-505. 24. Szmuness W, Stevens CE, Oleszko W, et al. Passive-active immunisation against hepatitis B: immunogenicity studies in adult Americans. Lancet 1981; i: 575-77.
permit speedier assessment. On the other hand, bone damage to length of diving experience and at present we see a predominantly young group of men with relatively short
HEPATITIS B IMMUNE GLOBULIN (HBIG) EFFICACY IN THE INTERRUPTION OF PERINATAL TRANSMISSION OF HEPATITIS B VIRUS CARRIER STATE
is related
diving. When substantial numbers have remained long enough in the industry, will we see the development of careers
in
femoral head lesions? If so, without routine radiography, would evidence of this be the sudden onset of a stiff and painful hip joint, and what would be the medicolegal implications? There are no direct complications of shaft lesions, but they must be noted in case a diver with a shaft lesion is found to be abnormally susceptible to bone damage elsewhere. If bone scintigraphy should be effective in the early detection of bone necrosis, one solution might be a full radiological examination of the limb bones at the start of a diving career followed by a regular bone scan. When it is suspected that damage may have occurred, such as after a decompression accident, an additional bone scan should be carried out and further radiographs taken of those areas which appear suspicious or positive on the scan.
Initial
more
Placebo-Controlled Trial LU-YU HWANG CLADD E. STEVENS TSU-SHEN SUN WOLF SZMUNESS
R. PALMER BEASLEY CHIA-CHIN LIN KWEI-YU WANG FON-JOU HSIEH
University of Washington Medical Research Unit (UWMRU), Taipei, Taiwan; Institute of Public Health, College of Medicine, National Taiwan University; Lindsley F. Kimball Research Institute of the New York Blood Center, New York, New York; Maternity and Child Health Center of the Taiwan Provincial Junior College of Nursing, Taipei, Taiwan and U.S. Naval Medical Research Unit (NAMRU-2), Manila, Philippines A randomised double-blind placebo-controlled efficacy trial of hepatitis B immune globulin (HBIG) for prevention of the vertically transmitted HBsAg carrier state was conducted in Taiwan where the carrier rate in the general population is 15-20%. HBIG was given immediately after birth to infants of e antigen positive HBsAg carrier mothers and all infants were followed for 15 months or more. Among 35 placebo-treated infants the carrier rate was 91%. This compares with the carrier rate of only 23% among 40 infants who received 0· 5 ml HBIG at birth, three months, and six months and the 45% carrier rate among 42 infants receiving a single 1 0 ml dose of HBIG at birth only. Efficacy was 75% and 45% respectively for the two treatment schedules. Significantly, the most common response of HBIG-treated infants was passive-active immunisation.
Summary
Recommendations At present, commercial divers should have regular skeletal surveys of their long bones by conventional radiography as set out by the M.R.C. Decompression Sickness Panel (1981).21 We recommend that all commercial divers should have an initial skeletal survey which should be repeated annually, but with two exceptions: l.-Annual radiographs would not be required for divers using only compressed air who had not been exposed to pressure above 30 m seawater and whose total exposure time on any occasion had never exceeded 4 h. 2.-Radiographs at intervals of 3 yr would suffice for divers using only compressed air who had not been exposed to pressures over 50 m seawater and whose total exposure time on any occasion had never exceeded 4 h. Any diver who had had compression treatment for decompression sickness would be excluded from the above two
Report of a Randomised Double-Blind
Introduction
categories.
We thank all those doctors who issued certificates of medical fitness, in connection with the Diving Regulations, for supplying data, and Mrs D. Weightman, University of Newcastle upon Tyne, for statistical advice. We received grants and help from the Medical Research Council, the Health and Safety Executive, and the CIRIA Underwater Engineering Group.
Requests for reprints should be addressed to Prof. R. I. McCallum, Department of Occupational Health and Hygiene, University of Newcastle upon Tyne, 21 Claremont Place, Newcastle upon Tyne NE2 4AA.
TRANSMISSION of hepatitis B virus (HBV) from mother to infant in the perinatal period is common in East Asia and
Aseptic bone necrosis in clearance divers. Report no 1/74 Royal Navy Clinical Research Working Party. Alverstoke: Institute of Naval
10. Harrison JAB, Elliott DH.
Medicine, 1974. 11.
Trowbridge WP, Evans A, Walder bone necrosis (ABN) indicate
DN. Does the presence of shaft (B) lesions of aseptic increased susceptibility to juxta-articular (A)
an
lesions? Undersea Biomed Res 1979, 6: 19-20. REFERENCES 1. Harrison JAB.
Aseptic bone necrosis in naval clearance divers: radiographic findings.
Washington State (USA) decompression tables. In: Trapp WG, Bannister EW, Davison, AJ, Trapp PA, eds. Fifth international hyperbaric congress proceedings,
Proc
Roy Soc Med 1971; 64: 1276-78. Sphar RL, Harvey CA. Aseptic bone necrosis among US Navy divers: survey of 934 non-randomly selected personnel. Undersea Biomed Res
2. Hunter WL Jr, Biersner RJ,
3. Griffiths PD. An exposure to risk registry for compressed-air workers.
5.
6. 7.
8. 9.
Trans Soc Occup Med 1971; 21: 123-25. Davidson JK. Dysbaric osteonecrosis. In: Davidson JK, ed. Aseptic necrosis of bone. New York: American Elsevier Publishing Co., 1976: 147-212. Davidson JK, Harrison JAB, Jacobs P, Hilditch TE, Catto M, Hendry WT. The significance of bone islands, cystic areas and sclerotic areas in dysbaric
osteonecrosis. Clin Radiol 1977; 28: 381-93. Golding FC. The shoulder-the forgotten joint. Br J Radiol 1961; 35: 149-58. Diving operations at work regulations. London: HMSO, 1981. The merchant shipping (diving operations) regulations. Statutory instruments no 116. London: HMSO, 1975. Walder DN. Aseptic necrosis of bone. In: Strauss RH, ed. Diving medicine. New York: Grune and
Stratton, 1977;
Simon Fraser University, Burnaby, Canada. 1974: 904-11. Gregg PJ, Walder DN. Early diagnosis of dysbaric osteonecrosis. In: Pearson RR, ed The twelfth undersea medical society workshop (UMS report no 7-30-77). Earlv diagnosis of decompression sickness. Bethesda, USA. 1977: 268-76. 16. Weatherley CR, Dale G, McGurk J, Walder DN. Bone necrosis and urinary hydroxyproline excretion in rabbits, Clin Sci Mol Med 1977; 52: 523-26. 17. Gregg PJ, Eastham EJ, Bell JI, Walder DN. Serum ferritin and dysbaric osteonecrosis. 15.
1978; 5: 25-36.
4.
Suppl.
12. Williams B, Unsworth I. Skeletal changes in divers. Aust Radiol 1976; 20: 83-94 13. Ohta Y, Matsunaga H. Bone lesions in divers. J Bone Joint Surg Br 1974; 56: 3-16 14. Walder DN, McCallum RI. An objective appraisal of the Blackpool (UK) and
97-108.
Undersea Biomed Res 1977; 4: 75-79. 18. Stothard J. Caisson disease of bone—an experimental study of its cause, early diagnosis and management. MD thesis. University of Edinburgh, 1980. 19. Gregg PJ. The use of radioisotopes in aseptic bone necrosis. In: Evans A, Walder DN, eds. Aseptic bone necrosis. London: CIRIA Underwater Engineering Group, 1977 77-83. 20. Michael RH, Dorfman HD. Malignant fibrous histiocytoma associated with bone infarcts. Report of a case. Clin Orthop 1976; 118: 180-83. 21. MRC Decompression Sickness Panel. Radiological skeletal survey for aseptic necrosis of bone in divers and compressed-air workers. Radiography 1981; 47: 141-43
389
other parts of the world where hepatitis B surface antigen (HBsAg) carrier rates are high. 1-6 In Taiwan, approximately 40% of the offspring of carrier mothers are so infected and, when the mother is hepatitis-B e-antigen (HBeAg) positive, 85-95% are infected.7-9 Among infants infected in the perinatal period, approximately 95% become carriers: among carriers in Taiwan we estimate that 35-50% were infected perinatally by their mothers. The infants usually are healthy,I although serum transaminase levels sometimes rise slightly. However, increasing evidence suggests that HBsAg carriers are at considerably increased risk of chronic hepatitis, cirrhosis, and primary hepatocellular carcinoma. 10-14 From 1974 to 1978 we conducted a randomised placebocontrolled double-blind trial of the efficacy of hepatitis B immune globulin (HBIG) in the prevention of perinatal transmission among Chinese in Taiwan.9 A single dose of 02 ml of HBIG with a passive haemagglutination (PHA) titre of 1:100 000 was administered to the infant within 7 days of birth. The results showed a delay in appearance of HBsAg in infants who received HBIG but essentially the same frequency of HBsAg carriers by 12 months of age. However, infants treated within 48 hours of birth had considerably lower HBsAg persistence rates.. On the basis of this observation, we decided to conduct a second trial in which HBIG would be administered as soon as possible after birth. We also decided to use larger doses and to compare single with multiple dose schedules in the hope that HBsAg persistence would be lower if the onset of antigenaemia were delayed. We now report preliminary results of this second trial.
Subjects and Methods Study Population and Case Recruitment Women attending the prenatal clinic at the Maternity and Child Health Center of the Taiwan Provincial Junior College of Nursing in Taipei, Taiwan, were screened for HBsAg by reverse passive haemagglutination (RPHA). Since HBeAg testing was not regularly available at the time, we selected those with an RPHA titre >1: 1024 (i.e., those most likely to be HBeAg-positive and to transmit the virus) and interviewed them during their next regular prenatal visit at which written consent to join the study was obtained. Only infants delivered at the Maternity and Child Health Center weighing .2000 g with Apgar scores 8 at 1 minute were eligible to the study. 430 mothers agreed to join; 144 of their infants were subsequently not included either because they were delivered at other hospitals, or treatment preparations were not available, or because the infants had a low birth weight or low Apgar score. Because transmission to mfants is low from HBeAg-negative mothers (even when the RPHA titre is 26 treated infants whose mothers were later found to be HBeAg-negative were excluded from the analysis. Of 213 infants who received the initial injection, 202 completed the three-dose series. This report concerns the 117 infants who were 15 months old or more at the time of this analysis. enter
high),8
HBIG and Placebo Preparations and Treatment Schedule Infants were divided into three
treatment
groups:
Group A: given 10ml saline at birth, three, and six months; Group B: given 1.0 0 ml HBIG at birth, and saline at three and six months;
Group C: given 0’ 5 ml HBIG diluted in 0’ 5 ml of immune serum globulin (ISG) (for a total volume of 1 -0ml) at birth, three, and six months. The HBIG, provided by Abbott Laboratories, (commercially available, lots 7307 SF and 79070 SF2 with PHA titres of 1:204 000 and 1: 150 000 respectively) was repackaged in 1 ml vials and
code numbers at the New York Blood Center. The vials stored at 4°C. The initial preparations were given to study infants in the delivery room when it had been determined that the birth weight and Apgar score criteria were satisfied. The mean administration time was 36 minutes after birth: all were treated within 30 hours and 95% were treated within an hour.
assigned
were
Laboratory Methods Laboratory tests were performed on coded specimens. Initial screening of women for HBsAg was by RPHA (Organon ’Hepanosticon’) in the maternity hospital laboratory. Positive sera were rescreened at a 1:1024 dilution, titred to extinction by RPHA and tested for HBeAg by radioimmunoassay (RIA). Infants’ specimens were tested for HBsAg, anti-HBs, anti-HBc, and HBeAg by RIA (’AUSRIA-II’, ’AUSAB’, ’CORAB’, and HBeAg diagnostic kits respectively, Abbott Laboratories, North Chicago, 111.). Although anti-HBs testing of these specimens was the study, the results were not linked to the individual in order to avoid breaking the code inadvertently. Selected HBsAg-positive sera and anti-HBs positive sera with rising antibody levels were tested for anti-HBc to confirm active HBV infection. Specimens from infants of 12 months and older which were negative for both HBsAg and anti-HBs were also tested for anti-HBc. Serum aspartate or alanine aminotransferase (AST or ALT) levels were measured using a kinetic photometric method (ABA-50, Abbott Laboratories, South Pasadena, Calif.).
performed during
Cord Blood Collection Cord blood specimens were collected by venepuncture of the umbilical cord before delivery of the placenta: the surface of the cord was wiped clean with an alcohol swab and allowed to dry.
Follow-up Follow-up visits were scheduled at 3-monthly intervals until the infants were 18 months old with a final visit at 24 months. Strict adherence to the follow-up schedule was maintained; every family was asked to return within 3 days of the target date and those who did not were visited in their homes by project nurses. Among the 117 study infants, 58% had a perfect follow-up record, while 95% missed two specimens or less. The 3-month and 6-month injections were permitted during the 30 days following the scheduled date. If the treatment preparation had not been administered by that time, the baby was removed from the study and defined as lost or refused. In practice, adherence was excellent; the second injection was scheduled for 3 months and the actual time between the first two injections (mean ± standard deviation) was 99 ± 8 days; the third injection was scheduled for 6 months of age and the actual time was 190 ± 10 days. The code was first broken in August, 1980. Nurses responsible for continuing follow-up will remain blind until the final analysis.
Definition of HBV Events HBV
infection.-An infant was considered infected if any specimen (excluding the cord blood) was HBsAg-positive or if the infant was anti-HBs positive at 12 months or older. Anti-HBc results were taken to confirm infection, but by themselves were not considered part of the criteria for infection because virtually all infants of HBsAg carrier mothers are anti-HBc positive under 12 months (passive anti-HBc from the mother) and none of the infants in this study were positive only for anti-HBc after 12 months of age. Persistent HBs antigenaemia.-An infant was considered to have persistent HBsAg if serum specimens were HBsAg-positive for 6 months or more and all subsequent specimens were HBsAgpositive. Transient HBV infection.-An infant was considered to have had a transient infection if at any time during follow-up the serum was
390
and antigenaemia subsequently disappeared but anti-HBc and anti-HBs were detectable. Infants who had no HBsAg-positive sera but were anti-HBs positive and anti-HBc positive at 12 months or older were also considered to have had a transient infection with HBsAg, probably missed by the timing of routine specimen collection. Uncertain Cases.-3 infants (all in group C) had persistent antiHBs at 15 months or older, but were anti-HBc negative. One additional infant (group B) had a single HBsAg-positive sample, but no serological confirmation of infection on follow-up. None of these infants was counted as having had an HBV infection. Three additional infants (1 in group B and 2 in group C) were HBsAg positive for the first time when the last specimen was obtained; these were counted as infected but not as persistent and because they are as yet unclassifiable they are not shown in fig. 1.
HBsAg-positive
Methods
TABLE I-MATERNAL AND INFANT CHARACTERISTICS BY STUDY
GROUP
(ALL ENROLLEES)
of Analysis
The characteristics of the
treatment groups were
compared for
7WfP/’///’ff/f -’
*Log 2. Proteus mirabilis was isolated from the spinal fluid of one infant while no organism was isolated from the second. Both infants recovered with antibiotic treatment. There was no reason to attribute the illnesses to the HBIG treatment.
ZZ! Anti-HBc and Anti-HBs only
Transient HBsA6
1
Results
- Persistent HBsA6
Comparability of Study Groups The three study groups were compared for factors which might have affected the risk of HBV infection or the effectiveness of HBIG; as can be seen in table I, all differences were small and not statistically significant. Characteristic HB V Infection Patterns by Study
Infants age (months) of HBV infection by study group, age of infant when infection was first detected serologically, and outcome of infection.
Fig. 1-Temporal patterns
Group
Placebo Recipients.-All but 2 of the 35 placebo recipients were infected and all of these were HBsAg-positive by 3 months of age. In all but one infected infant, every subsequent specimen was HBsAg-positive; the single infected baby who did not become a carrier lost its HBsAg and developed anti-HBs by the sixth month (figs. 1, 2). HBIG Recipients.-Two patterns of infection occurred in HBIG recipients (fig. 1):
(1) Transient HBV infection’occurred in 30; in 11 of
these HBsAg
A=placebo; B=single dose HBIG; C=three doses HBIG. Arrow (y) indicates the times injections were given. Three infected individuals in table I are not included here because only their last specimen was HBsAg positive and therefore they cannot yet be classified.
statistical significance using the Student’s t test. HBIG efficacy was determined by comparing the proportion of infants with persistent HBsAg in each group using Fisher’s exact test. The times of onset of infection in infants with persistent antigenaemia were also compared using the life table method of analysis. As an approximation for this analysis, the time of onset was defined as the age of the infant at the time HBsAg was first detected serologically.
Inadvertent Events One of the study infants died at 6 days of age while still in the hospital. Labour and delivery had been uncomplicated, birth weight was 3700 g and the 5-minute Apgar score was 9. The infant had received 0 - 5 ml ofHBIG lot 79079 SF2. No antecedent event to death was recognised. A necropsy was not performed. In 2 infants neonatal meningitis developed during late June and early July, 1980, when similar cases were recognised in several nonstudy infants in the nursery. One study infant had received 0 - 5 ml of HBIG and the other 1.0 ml of HBIG (both lot 79079 SF2).
explain the
Infants age (months) Fig.2-Cumulative onset of chronic HBV carrier state by study group.
(Babies followed .15 months
- life table
analysis.)
391
appeared, usually in low titre, from 6 to 12 months of age (fig. 3a). HBsAg then disappeared and high titre anti-HBs developed and was present in all subsequent specimens. In the remaining 19 infants with transient infections, HBsAg was never detected but was presumed to have been present for a short period between blood specimens (fig. 3b). After an initial decline (loss of passive antibody), high titre anti-HBs developed in these infants (usually at 9 to 12 months) and remained present at high levels in all subsequent specimens. (2) Persistent antigenaemia developed in 30 HBIG-treated infants (fig. 3c). These cases differed from those in the placebo group in that HBsAg was usually first detected at a later age (fig. 2). There was no way to distinguish infants who would become chronic HBsAg
TABLE II- HBV INFECTION AND
HBsAg PERSISTENCE
BY STUDY GROUP
p VALUE BY FISHER’S EXACT TEST
carriers from those who would have transient infections except that their HBsAg levels when first detected were generally higher than in infants with transient antigenaemia.
Infants who failed to become infected during the first 15 months of life included only 2 in group A (5’ 7%), and a somewhat higher proportion of HBIG-treated infants (21% and 25% respectively for infants in groups B and C). When passive antibody was no longer detectable in HBIG-treated infants, their sera were negative for all HBV markers including anti-HBc and they were presumably still susceptible to HBV.
HBIG Efficacy The
primary goal
of HBIG
HBIGsuccesstransient antigenemia
treatment
HBIG no
in
perinatal
success-
antigenemia
1
Case 15 GroupC HBIG 0-5mlx3
Case 66
6raup C HBlG 0.5ml x 3
Age in months
Age in months
(a)
(b)
HBI6 failure
No infection
transmission of HBV is prevention of HBsAg persistence, rather than prevention of clinical hepatitis or HBV infection per se because clinical hepatitis is rare following perinatal transmission and HBV infection is desirable if it is subclinical, transient, and results in permanent immunity. Of the 96 infants who had HBV infections during the study, 62 developed persistent antigenaemia. As shown in table II, 91% of the placebo recipients developed persistent HBsAg. In contrast, only 50% of those who received a single dose of’ HBIG and 23% of those who received three doses developed persistent HBsAg. From these data the efficacy of the three 0’ 5 ml doses of HBIG is calculated to be 75% and the single 1 0 ml dose to be 45%. The differences between each HBIG group and the placebo group and between the two HBIG groups are highly significant statistically (P<0. 01). Of considerable interest is the fact that in the majority of infants treated with HBIG an active immune response develops indicating that HBIG prevents the carrier state, while allowing passive-active immunisation to occur. Among the HBIG-treated infants who did not develop persistent antigenaemia, 11 of 21 in group B and 19 of 31 in group C developed persistent anti-HBs in levels exceeding those achieved by passive immunisation. Among the placebo recipients, only 3 did not develop persistent HBsAg; one was transiently HBsAg-positive followed by persistent high titre anti-HBs and the others remained uninfected. Clinical hepatitis occurred in only one infant, one who had received the placebo and was jaundiced at 54 days after birth at which time HBV markers were not tested. HBsAg was present at the routine 3-month follow-up and remained positive through the last follow-up at 18 months. Aminotransferase levels were slightly elevated until 9 months and have been normal thereafter.
Case 55
Case 90 Group C 3 HBlG 0.5ml x
GroupB HBlG 1.0ml xl
Age in months
Age in months
(c)
(d)
Fig. 3-Temporal anti-HBs response patterns in HBIG recipients by outcome, age of infant, and anti-HBs titre (SD units).
of Onset of HBV Infection by Treatment Group Fig. 1 shows the onset of HBsAg positivity for infants who developed HBsAg persistence. This figure shows not only differences in the cumulative persistence rates for infants in Time
each of the three groups, but differences in the time of onset as well. All of the chronically infected babies in the placebo group were HBsAg-positive by the time of the first follow-up specimen at 3 months. Among the infants who received a single dose of HBIG at birth, new cases appeared at each follow-up bleeding over the first year with no additional ones thereafter. All of those who had received HBIG three times in 6 months and became HBsAg carriers were, with only three exceptions, HBsAg-positive for the first time at 12 months of
392 age with
thereafter. The infant’s age when first detected correlated with persistence. Among HBsAg 38 infants who were first positive at 3 months, 97% developed persistent antigenaemia, compared with 25 of 36 (69%) of those first positive thereafter. This difference was statistically no new cases
was
significant. Infants in the two HBIG treatment groups were evaluated for factors which might have predicted the infection outcome-maternal HBsAg titre, cord blood HBsAg positivity, age at first treatment, and number of older siblings. No predictive factors were found. The absence of correlation of cord blood (CB) positivity with outcome among HBIG treated babies is especially noteworthy. Among 66 CB-negative infants 20 (30-3%) became carriers, while among 16 who were CB-positive, 6 (37 - 5%) became carriers.
Discussion HBV infection early in life often leads to the HBsAg carrier
state. 1,3,15 Thus, interruption of early transmission from infant is extremely important. This study’s unequivocal results founded on a randomised double-blind placebo-controlled design, large number of high-risk infants, and long-term follow-up provided strong evidence of the value of HBIG in reducing the incidence of the HBsAg carrier state following perinatal transmission. A single dose at birth was 45% successful and the three-dose series was 75% successful in preventing persistent antigenaemia. Both results are highly significant statistically. Previous evidence on HBIG protection in perinatal transmission includes single case reports, small series without controls and one small nonrandomised unblind series with limited follow-up. 16-20 Our first study in Taiwan was the largest previous trial of HBIG efficacy in infants of carrier mothers.9 That study, like the present one, was randomised, double-blind, and placebocontrolled. In it we gave infants a single 0’ 2 ml dose of HBIG (average 0’ 06 ml/kg) up to 7 days after birth (at an average time of 46 hours). We found a delay in the onset of infection but only a slight reduction in the carrier rate among HBIG recipients compared with controls. The mean age (hours) at the time of HBIG administration was significantly lower in infants with transient HBV infection suggesting that the timing of HBIG administration was critical. In another report concerning the efficacy of HBIG in vertical transmission, Reesink and his colleagues used a nonrandomised, unblind design to study the effect of large doses of HBIG at birth followed by monthly injections for 6 months.2O HBsAg was not detected during follow-up in any of 6 HBIG-treated infants born to HBeAg-positive mothers compared with 4 of 6 infants not treated with HBIG. As in our study, most of the HBIG-treated infants had evidence of HBV infection, developing anti-HBs during follow-up. The primary goal of our second trial was to decrease as far as possible the time of initial HBIG administration. The data from our previous study demonstrated the critical importance of the timing of the initial HBIG administration. Fewer infants who received HBIG under 48 hours of age developed persistent antigenaemia than those treated at 48 hours or more. In the present trial in which HBIG was administered at an average time of 36 minutes there was a substantial and highly significant reduction in persistent HBsAg among all HBIG recipients compared with placebo controls. However, because almost all infants in this study
- mother
to
received HBIG within 1 hour of birth, it is impossible to determine whether administration somewhat later, but within 48 hours, will also be efficacious. Presumably, however, the earlier administration occurs the better. A single dose of HBIG at birth reduced HBsAg persistence although not as effectively as a multiple-dose series. In many areas of the world where follow-up treatment may be difficult to organise, a reasonable strategy might be simply to give a single dose immediately at birth. Even if follow-up treatment is not feasible, the HBIG given immediately will be of value. The present high cost of HBIG may be thought a barrier to its use despite its obvious and unique value in perinatal HBV infection. However, the current cost of the three small doses (0 - 5 ml) recommended for this regimen should not be prohibitive in developed countries. HBIG is expensive partly because demand has been low and the required high titre donors are rare. High titre donors should be easy to create in the near future using the new HBV vaccines,21-22 and the efficacy of HBIG demonstrated in this study may increase the demand. Thus, the cost of HBIG should eventually fall, making its worldwide use feasible. This study unequivocally showed that the primary effect of HBIG in the newborn was prevention of HBsAg persistence rather than prevention of HBV infection per se. HBIG . recipients had somewhat lower infection rates than placebo recipients. However, HBIG recipients had more transient infections, while placebo recipients were much more likely to acquire a chronic antigenaemia. It is especially noteworthy that HBIG is equally effective in both cord blood negative and cord blood positive infants. This observation, and the fact that cord blood titres are usually quite low, suggest that a positive cord blood is usually a reflection of maternal-fetal transfusion occurring during labour and delivery. Ifa positive cord blood reflected in-utero viral transmission, HBIG should be ineffective in preventing the HBsAg carrier state, since HBIG has no value in eliminating the HBsAg carrier state. Immediate administration of HBIG is essential in the prevention of perinatal HBV infection suggesting that HBIG works initially by blocking or decreasing viral access to the liver. The very high attack rates in untreated babies and the rapid onset suggests high infectivity of the mother during labour and/or delivery. It is reasonable to believe that the baby of an HBeAg-positive mother not infected at birth remains at high risk from continuing exposure. The fact that multiple doses of HBIG work better than a single dose suggests that postponing the time of infection is advantageous in reducing the probability of persistent
antigenaemia. Some
infants
developed persistent HBsAg despite
HBIG prophylaxis. It seems unlikely that larger doses of HBIG at birth would have been more effective. In fact, in our first trial in Taiwan,9 a much smaller, 0 - 2 ml, dose of HBIG when given within 48 hours of birth produced reduction in chronic antigenaemia identical to that seen in the present study where a single dose five times larger was used. Whether more frequent or more prolonged prophylaxis would be more effective is not known. Reesink et al. 20 gave monthly injections ofHBIG for 6 months or longer to a group of 6 high risk infants. Although he observed no chronic infections among these children, because of the small numbers of infants treated his results do not differ statistically from those reported here among infants given only three injections. Another approach to prophylaxis may include use of the new hepatitis B vaccine.21In a recent efficacy trial, this
prolonged
393
vaccine has proven highly effective in preventing HBV infection.22 This study and earlier ones among chimpanzees also suggest that the vaccine may be effective in post-exposure prophylaxis.z3 If newborn infants have a vigorous, rapid immune response to the vaccine, the vaccine alone may prove effective. An alternative approach would be the combined use of both HBIG and vaccine; HBIG to provide immediate protection and delay the onset of infection until the infant developed an active immune response to the vaccine. Recent studies in adults revealed that antibodies from HBIG do not interfere with an active immune response to the vaccine.24 Options utilising vaccine alone or in a passive-active regimen must be explored in the future to carefully assess efficacy.
We gratefully acknowledge the support of Miss Pao-Tien Chu, President of the Taiwan Provincial Junior College of Nursing and the staff of the college. We specially thank Miss H. C. Lin, Director of Nursing, and the nursing staff who assisted in enrolment of mothers and administration of the initial injections. Our deep appreciation is also due to Miss Susan Chen, UWMRU Nursing Supervisor, who established and supervised our perinatal transmission and HBIG studies until her retirement in 1979. Miss Lucy Chiang, Miss Linda Lo, and Miss Eileen Chen, UWMRU nurses, made this project their full-time work for several years. We also thank Dr Saul Krugman, who served as consultant for the study. This research was supported by funds provided by NIH National Institute of Allergy and Infectious Diseases Contract C2101800 (HBG-2908) and the Bureau of Biologics, Food and Drug Administration. Requests for reprints should be addressed to R. P. B. at UWMRU, 15-2 Kung Yuan Rd, Taipei, Taiwan, ROC. _
REFERENCES 1. Stevens CE, Beasley RP, Tsui J, Lee WC. Vertical transmission of hepatitis B antigen in Taiwan N Engl J Med 1975; 292: 771-74. 2 Okada K, Yamada T, Miyakawa Y, Mayumi M. Hepatitis B surface antigen in the serum of infants after delivery from asymptomatic carrier mothers. J Pediatr 1975; 87: 360-63. 3 Lee AKY, Ip HMH, Wong VCW. Mechanisms of maternal-fetal transmission of hepatitis B virus. J Infect Dis 1978; 138: 668-71. 4. Woo D, Cummins M, Davies PA, Harvey DR, Hurley R, Waterson AP. Vertical transmission of hepatitis B surface antigen in carrier mothers in two west London hospitals. Arch Dis Childh 1979; 54: 670-75. 5 Derso A, Boxall EH, Tarlow MJ, Flewett TH. Transmission ofHBsAg from mother to infant in four ethnic groups. Br Med J 1978; i: 949-52. 6. Shiraki K, Yoshihara N, Kawana T, et al. Hepatitis B surface antigen and chronic hepatitis in infants born to asymptomatic carrier mothers. Am J Dis Child 1977; 131: 644-47. 7. Okada K, Kamiyama I, Inomata M, et al. e antigen and anti-e in the serum of asymptomatic carrier mothers as indicators of positive and negative transmission of hepatitis B virus to their infants. N Engl J Med 1976; 294: 746-49. 8 Beasley RP, Trepo C, Stevens CE, Szmuness W. The e antigen and vertical transmission of hepatitis B surface antigen. Am J Epidemiol 1977; 105: 94-98. 9. Beasley RP, Stevens CE. Vertical transmission of HBV and interruption with globulin. In. Vyas GN, Cohen SN, Schmid R, eds. Viral hepatitis. Philadelphia: Franklin Institute Press, 1978: 333-45 10. Nishioka K, Hirhyama T, Sekine T, et al. Australia antigen and hepatocellular carcinoma. Gann Monogr Cancer Res 1973; 14: 167-75. 11 Prince AM, Szmuness W, Michon J, et al. A case/control study of the association between primary liver cancer and hepatitis B infection in Senegal. Int J Cancer 1975; 6: 376-83. 12 Ohbayashi A, Okochi K, Mayumi M. Familial clustering of asymptomatic carriers of Australia antigen and patients with chronic liver disease and primary liver cancer Gastroenterology 1972; 62: 618-25. 13 Szmuness W. Hepatocellular carcinoma and the hepatitis B virus: Evidence for a causal association. Prog Med Virol 1978; 24: 40-69. 14. Beasley RP, Lin CC, Hwang LY. The risk of hepatocellular carcinoma in hepatitis B virus infection: a prospective study m Taiwan. In: Szmuness W, et al, eds. Viral hepatitis. Philadelphia: Franklin Institute Press (in press). 15 Schweitzer IL, Dunn EAG, Peters RL, et al. Viral hepatitis in neonates and infants. Am J Med 1973; 55: 762-71. 16 Kohler PF, Dubois RS, Merrill DA, et al. Prevention of chronic neonatal hepatitis B virus infection with antibody to the hepatitis B surface antigen. N Engl J Med 1974; 291: 1378-80. 17. Varma RR. Hepatitis B surface antigen carrier state m neonates. JAMA 1976; 236: 2302-04. 18 Dosik H, Jhaveri R. Prevention of neonatal hepatitis B infection by high-dose hepatitis B immune globulin. N Engl J Med 1978; 298: 602-03.
LACTOBACILLI DO NOT CAUSE FREQUENCY AND DYSURIA SYNDROME W. BRUMFITT
J.
H. LUDLAM
M. T. HAMILTON-MILLER ANN GOODING
Department of Medical Microbiology, Royal Free Hospital, London NW3
Mid-stream specimens (MSU) of urine were collected from 142 healthy women (pregnant and non-pregnant) and cultured for lactobacilli and other fastidious bacteria. The latter either require CO2 or are obligate anaerobes. Lactobacilli were present in counts of 104/ml or more in 34·8% of the women, and in counts of 105/ml or more in 20·2%. Besides lactobacilli, which were the bacteria most frequently isolated, anaerobic grampositive cocci (peptococci and peptostreptococci) were often found. This flora is typical of that of the lower vagina, and none of these women had either symptoms of urinary infection or pyuria. Therefore, the bacteria isolated were commensals or contaminants. Cultures of MSUs taken from 26 women with symptoms of dysuria and/or frequency, but without significant numbers of conventional pathogens such as Escherichia coli, contained commensals and contaminants of the same variety and in similar numbers. Urine samples from 50% of these patients contained at least 104 lactobacilli/ml and 27% had 105 or more/ml. Lactobacilli were absent from the suprapubic urine specimens cultured from a further 44 women. There was no significant difference between the isolation rate of lactobacilli in urine cultures from healthy women and the rate in women with dysuria and
Summary
frequency.
Introduction
patients presenting with dysuria and/or a significant bacteriuria caused by conventional pathogens such as Escherichia coli (> 1 OS ONLY 50% of have
frequency
bacteria/ml).1,2
The illness with symptoms but without
significant bacteriuria is often called the urethral syndrome. However, this name has been rejected on the grounds that various workers have used the term in different contexts.33 Instead, the term "frequency and dysuria syndrome" has been recommended. This syndrome may be caused by organisms which are difficult to culture such as fastidious bacterial chlamydiae,S or even viruses.6 Corynebacterium
19. Matsumoto neonates
S, Tomigashi T, Fujmoto S, et al. Prevention of hepatitis B infection in by human anti-HBV immunoglobulin (HBIG). Perinatology 1979; 9:
615-22. 20. Reesink HW, Reerink-Brongers EE, Lafeber-Schut BJT, et al. Prevention of chronic HBsAg carrier state in infants of HBaAg-positive mothers by hepatitis B immunoglobulin. Lancet 1979; ii: 436-38. 21. Hilleman MR, Bertland AU, Buynak EB, et al. Clinical and laboratory studies of HBsAg vaccine. In: Vyas GN, Cohen SN, Schmid R, eds. Viral hepatitis. Philadelphia: Franklin Institute Press, 1978: 525-37. 22. Szmuness W, Stevens CE, Harley EJ, et al. Hepatitis B vaccine. Demonstration of efficacy in a controlled clinical trial in a high-risk population in the United States. N
Engl J Med 1980; 303: 833-41. 23. Purcell RH, Gerin JL. Hepatitis B vaccines: a status report. In: Vyas GN, Cohen SN, Schmid R, eds. Viral hepatitis. Philadelphia: Franklin Institute Press, 1978, 491-505. 24. Szmuness W, Stevens CE, Oleszko W, et al. Passive-active immunisation against hepatitis B: immunogenicity studies in adult Americans. Lancet 1981; i: 575-77.