979 artificial feeds (either to supplement breast-feeds or alone) month, 87% were using a low-solute preparation.




Northwick Park Hospital and Clinical Research Centre, Harrow, Middlesex HA1 3UJ.



read the paper by Dr Beasley and which (Oct. 18, p. 740) presented evidence against a for vertical transmission of as mechanism breast-feeding hepatitis B. I agree that, taking carrier mothers as a group, there seems to be no significant difference in the development of hepatitis-B surface (HBs) antigenaemia or antibody to HBsAg between the breast-fed and bottle-fed babies. However, it is difficult to assess the data without considering other variables, the most important of which I believe to be the babies’ antigen status at birth. Although this was mentioned in the paper, the results were not analysed with respect to this factor. If, at birth, it can be demonstrated that HBsAg is present in the infant’s circulation, then it is difficult to estimate the importance of further possible infection through the breast-milk. Although Dr Beasley’s group were unable to demonstrate HBsAg in the unconcentrated breast-milk of 32 carriers, the statistical treatment of the results necessarily assumed that all breast-milk had been positive. The amount of HBsAg found in breast-milk is small compared with serum, but in some cases it can be demonstrated by radioimmunoassay (R.I.A.) in unconcentrated samples and by immune electron microscopy (I.E.M.) and R.I.A. on concentrated samples.I The only satisfactory way of demonstrating breast-milk as an important route of vertical transmission is to examine closely the group of children whose cord-blood does not contain any HBsAg by R.I.A. and in whose mothers’ breast-milk HBsAg can be demonstrated by R.I.A. or I.E.M. after concentration by ultracentrifugation.2 The effect of breast-feeding on the prevalence of HBs antigenxmia in this group of infants would give the best indication of the possible importance of breast-feeding as a mechanism of vertical transmission. In my own studies3 in 58 infants, I found that at birth 33 (57%) had HBsAg in their cord-blood. Of 27 samples of breast-milk from carrier mothers, concentrated by ultracentrifugation, 15 (56%) samples were positive by R.I.A. HBsAg particles have also been seen in 4 out of 10 samples examined by I.E.M. thus demonstrating the specific presence of HBsAg, but only in very small quantities. 7 out of 12 mothers whose infants had HBsAg in their cordblood also had HBsAg in their breast-milk. In 3 cases both the breast-milk and the cord-blood were negative, and in 3 cases the breast-milk was positive and the cord-blood negative. The latter group of infants will provide the best evidence of the importance of breast-feeding as a method of vertical transmission.





Regional Virus Laboratory, East Birmingham Hospital, Bordesley Green East, Birmingham B9 5ST.



SIR,-Dr Galbraith and his colleagues (Sept. 20, p. 528)

reported fulminant hepatic failure

in three HBsAg-positive chronic carriers with leuksemia and choriocarcinoma after withdrawal from cytotoxic chemotherapy. They postulated that return of "immunocompetence" after stopping chemotherapy resulted in massive liver damage. We have seen two cases which support this hypothesis. These two patients were 1

Boxall, E. H., Flewett, T. H., Dane, D. S., Cameron, C. H., MacCallum, F. O., Lee, T. W. Lancet, 1974, ii, 1007. 2. Linneman, C. C., Goldberg, S. ibid. p. 155. 3 Boxall, E. H., Davies, H. ibid. p. 1513.

receiving 25 mg/kg of cyclophosphamide at 4 to 6 week intervals for multiple myeloma, and both had HBsAg in their serum while on chemotherapy (from multiple transfusions or hospital contacts’2). They had no symptoms and only a slightly raised serum-glutamic-oxaloacetic-transaminase (S.G.O.T.). Cyclophosphamide was discontinued after 5 to 7 months because of a good clinical response. Subsequently one patient developed a hepatitis-like syndrome with a rapid rise in s.G.o.T. and serum-bilirubin. Liver biopsy demonstrated subacute hepatic necrosis, and the patient died several weeks later from liver failure despite prednisone therapy. The other patient developed a clinical syndrome suggestive of chronic active hepatitis confirmed by liver biopsy. She responded well to prednisone, 40 mg/day, but died several months later from complications of multiple myeloma. Thus, fulminant hepatitis, subacute hepatic necrosis, and chronic active hepatitis may occasionally occur after the withdrawal of chemotherapy in HBsAg-positive patients with myeloproliferative and lymphoproliferative disorders. Gastrointestinal Unit, Massachusetts General Hospital, Boston, Massachusetts 02114, U.S.A.





interested in the


of Professor

Munro and his colleagues (March 29, p. 722) connecting changes in tryptophan metabolism with hepatic coma. They

patients with impairment of liver function a high plasma-insulin concentration leads to excessive removal of the branched-chain aminoacids by muscle, thereby lowering their plasma concentrations. In consequence, the competitive suggest that in

action of the branched-chain aminoacids on the entry of tryptophan into the brain is reduced, more tryptophan enters the brain, and 5-hydroxytryptamine (5-H.T.) is synthesised in excess. It was suggested that this extra 5-H.T. facilitates hepatic coma. We have been studying tryptophan metabolism in patients with hepatic cirrhosis and some of our data are relevant to this hypothesis. We found3 that patients with hepatic cirrhosis have normal total serum-tryptophan. The mean free (i.e., non-albuminbound) serum-tryptophan was raised only 50% above the mean control value. However, cerebrospinal-fluid (c.s.F.) tryptophan was increased almost fivefold. This suggests a large rise in the brain-tryptophan content of these patients. The c.s.F. tryptophan concentration varied widely from 49% to 830% above the control value. It is unlikely that the very large increase sometimes seen was due solely to a decrease in serum branchedchain aminoacids, especially since patients with hepatic cirrhosis have high plasma concentrations of phenylalanine and tyrosine both of which, like the branched-chain aminoacids, compete with tryptophan for transport into brain. If elevated brain tryptophan contributes to hepatic coma, it might be expected that brain tryptophan would be higher in patients in coma. However, we found no difference in the c.s.F. patients with tryptophan concentrations when comparing hepatic cirrhosis in and not in coma.3 We also looked at c.s.F. 5-hydroxyindoleacetic acid (5-H.I.A.A.).4 We found, like Knell et al.,’ that baseline concentrations of c.s.F. 5-H.I.A.A. were raised in patients with hepatic failure. However, after probenecid administration to patients in or recently recovered from coma, the ratio of increase Wands, J. R., Chura, C. M., Roll, F. J., Maddrey, W. C Gastroenterology, 1975, 68, 105. 2. Wands, J. R., Walker, J. A., Davis, T. T., Waterbury, L. W., Owens, A. H., Carpenter, C. C. J. New Engl. J. Med. 1974, 291, 1371. 3. Young, S. N., Lal, S., Sourkes, T. L., Feldmuller, F., Aronoff, A., Martin, J. B. J. Neurol. Neurosurg. Psychiat. (in the press). 4. Lal, S., Aronoff, A., Garelis, E., Sourkes, T. L., Young, S. N., de la Vega, C. E. Clin. Neurol. Neurosurg. 1974, 77, 142. 5. Knell, A. J., Davidson, A. R., Williams, R., Kantamaneni, B. D., Curzon, G. Br. med. J. 1974, i, 549. 1.

Letter: Subacute and chronic active hepatitis after withdrawal of chemotherapy.

979 artificial feeds (either to supplement breast-feeds or alone) month, 87% were using a low-solute preparation. at one E. C. COLES S. COTTER H. B...
158KB Sizes 0 Downloads 0 Views