POSTOPERATIVE CARDIAC DEATH

455 to normal or near normal levels within 4-8 weeks, and the clinical and histological findings improved 3-12 months after therapy. The pre-treatment duration of CAH (median 2 years in responders vs 10 years in non-responders) influences the outcome of therapy significantly (p 0-05). Patients without HBV sequences integrated into the hepatocyte genome tended to have a better response to therapy. In the control group of 10 untreated patients only one spontaneous remission occurred, in a 41-year-old woman. In contrast to Alexander et al’s study no attempts were made to stratify or match any variables to identify pretreatment factors affecting =

therapy outcome. Since the responder rates of the IFN-tx and our trial (6/23 and 5/10, respectively, for loss of HBeAg) greatly exceeded the spontaneous seroconversion rates (0/23 and 1/10) but do not differ significantly from each other, we propose consideration of the IFN-6 and IFN-y combination, given the advantage of a shorter treatment duration, for a larger trial. Department of Virus Research, Max-Planck-Institut für Biochemie,

8033 Martinsried, Munich, West Germany

W. H. CASELMANN P. H. HOFSCHNEIDER R. KOSHY

Department of Internal Medicine, Krankenhaus der Barmherzigen Bruder,

data that HDV is a cytopathic virus it can be assumed that the first enzyme peak is because of HDV-induced liver necrosis. Whether the second bout of hepatitis is caused by the immune response to a restored synthesis of HBV gene products after the acute phase of HDV hepatitis3 remains speculative. Finay, DeCock et al ask how long total anti-HD antibodies persist after self-limited HBV/HDV coinfection. Our data on 68 parenteral drug abusers in whom this antibody response was prospectively evaluated show persistence for a mean (SD) period of 97 (62) months after the acute episode (67 [5’5] months after the clearance ofHBsAg). Institute of Internal Medicine, Infectious Diseases, and Immunopathology, Ospedale L Sacco, 20157 Milan, Italy

FRANCESCO CAREDDA SPINELLO ANTINORI CARLA PASTECCHIA MAURO MORONI

1. Caredda F, Antinon S, Re T, Pastecchia C, Moroni M. Course and prognosis of acute HDV hepatitis. Prog Clin Biol Res 1987; 234: 267-76. 2. Amoroso P, Giorgio A, Fico P, et al. Delta infection in the Naples area: Epidemiologic and clinical significance. Hepatol 1986; 2: 11-18. J 3. Bonino F, Smedile A. Delta agent (type D) hepatitis. Semin Liver Dis 1986; 6: 28-33.

J. EISENBURG

Munich 1. Caselmann

EFFECT OF ACTIVATED CHARCOAL ON HYPERCHOLESTEROLAEMIA

WH, Eisenburg J, Hofschneider PH, Koshy R. Antiviral therapy of

chronic-active hepatitis B with interferon-beta and interferon-gamma: a controlled trial. J Med Virol 1987; 21: A128 (abstr).

HBV/HDV COINFECTION SIR,-We do not agree with Dr DeCock and colleagues (June 20, p 1438) that the long-term prognosis of acute hepatitis in patients simultaneously infected by the hepatitis B virus (HBV) and the hepatitis delta virus (HDV) is always good. The possible chronic evolution of HDV infection and disease in cases of HBV/HDV coinfection has been reported by us1 and independently by Amoroso et aI.2 Chronic hepatitis developed in 6 of 220 (2-7%) drug addicts admitted to our unit and prospectively studied during 1979-86 after apparently acute HBV/HDV coinfection. These 6 patients were males, aged 18 to 25. None had had clinical hepatitis before. On admission they were HBsAg and HBeAg positive, and had high titres (over 1:4000) of IgM anti-HBc. A transient HD antigenaemia was detected in 3 patients. In all the cases IgM and IgG anti-HD became detectable in rising titres during the following weeks. The 6 patients remained HBsAg positive during follow-up (mean 26 [SD 10] months, range 17-42), and IgM and IgG anti-HD remained positive at titres ranging from 10-2 to 10-5. In all the patients alanine aminotransferase (ALT) levels have remained high. A progressive liver disease (4 chronic active hepatitis and 2 active cirrhosis) was established just 5 to 18 months after the acute episode. It also seems unlikely that the severity of hepatitis prevents its chronicity, since 2 of our 6 patients had a severe illness (prothrombin activity less than half normal). In addition, as far as the pathogenesis of the double bout of hepatitis in HBV/HDV coinfected patients is concerned, we do not agree with DeCock and colleagues that the temporarily related appearance of anti-HD antibody to the second peak of ALT suggests that relapse is caused by HDV-induced hepatocellular necrosis. In fact we have not observed a correlation between the appearance of anti-HD’ antibodies and the second bout of ALT (table). Furthermore, a transient HD antigenaemia generally can be detected only during the first ALT elevation, coincident with the presence ofHDAg in the liver. According to the currently available SEROLOGICAL HDV MARKERS IN HBV/HDV

COINFECTED PATIENTS

WITH A BIPHASIC COURSE OF THE ACUTE DISEASE

SIR,-A decrease in plasma total cholesterol and low density lipoprotein cholesterol after treatment with activated charcoal was described in patients with hypercholesterolaemia by Kuusisto et al.l Their results contrast with data that we obtained in a double-blind trial in mainly primary hyperlipidaemic patients. In the first part of the trial 12 patients were randomised to receive either 15 g activated (n 6) or 15 g non-activated charcoal in granulated form as placebo (n = 6) over 12 weeks. In the second part the same patients were randomised again to receive either 30 g activated (n = 6) or 30 g non-activated (n = 6) charcoal. Mean serum cholesterol levels in patients treated with activated charcoal did not decrease significantly (table). In both groups during treatment with activated charcoal, 1 patient had a significant decrease in serum cholesterol level. No decrease in serum triglycerides was observed in 3 =

hypertriglyceridaemic patients. EFFECT OF

12 WEEKS’

TREATMENT WITH ACTIVATED CHARCOAL -

ON SERUM CHOLESTEROL LEVELS

(MEAN, SD)

The contrasting findings of Kuusisto et al may be due to the non-placebo design of their study. It may also be that there are responders and non-responders in treating hypercholesterolaemia

with activated charcoal. Diakonessenhuis, PO Box 80250, 3508 TG Utrecht, The Netherlands; and University Hospital,

J. B. L. HOEKSTRA

Utrecht

D. W. ERKELENS

1. Kuusisto P, Vapaatalo H, Manninen V, Huttunen JK, Neuvonen PJ. Effect of activated charcoal on hypercholesterolaemia. Lancet 1986; ii: 366-67.

POSTOPERATIVE CARDIAC DEATH

SiR,—Your July 11 editorial on postoperative cardiac death ignores a contributory factor that is probably important, namely the inadvertent abrupt withdrawal of cardiac drugs in many patients admitted for surgery. Duthie et all reported that 29% of patients taking cardiovascular drugs as outpatients failed to have their

456 normal medication prescribed in hospital before surgery. Evidently, information on prescription drug use fails to accompany many

patients as they enter hospital. The risk of postoperative cardiac death is greatly concentrated in patients with pre-existing cardiac disease. It is reasonable to assume that patients who have been prescribed cardiovascular drugs are likely to be at higher than normal risk of perioperative cardiac complications. The abrupt withdrawal from such patients of their usual drugs can only add to the risk because, as Duthie et al correctly point out, the abrupt withdrawal of certain cardiovascular drugs (eg, (3-blockers and calcium channel blockers) is hazardous. Part of the investigation of postoperative cardiac death should be to review the patient’s prescription drug history before and during their hospital stay. Part of the prevention of such death should be to close the gap between outpatient and inpatient prescribing. While the doctors involved ought to do better, improved liaison between hospital and community pharmacists may be the simplest way to close the gap. Capaciteitsgroep Epidemiologie/Gezonheidszorgonderzoek, Rijksuniversiteit Limburg, 6200 Maastricht, The Netherlands

JOHN URQUHART

Spece AA, Nimmo WS. Concurrent drug therapy in patients undergoing surgery. Anaesthesia 1987; 42: 305-06.

1. Duthie DJR, Montgomery JN,

ALUMINIUM-RELATED BONE DISEASE AND

HAEMODIALYSIS

(-))

TO = before first haemodialysis, Tl = after desferrioxamine, T2 = before second haemodialysis, and T3 after second haemodialysis. =

Medizinische Klinik III, Staedtisches Krankenhaus, Friedrich Engelstrasse 25, D-675 Kaiserslautern, West Germany

U.T. SEYFERT K. DIDION F.W. ALBERT

DJ, Dawborn JK, Ham KN, Xipell JM. Treatment of dialysis osteomalacia with desferrioxamine. Lancet 1982, ii: 343-45. 2. Malluche HH, Smith AJ, Abreo K, Faugere MC. The use of desferrioxamine in the management of aluminium accumulation in bone in patients with renal failure. N EnglJ Med. 1983; 311: 140-44 3. Iwamoto N, Ono T, Yamazaki S, Fukuda T, Kondo M. Clinical features of aluminium-associated bone disease m long term haemodialysis patients. Nephron 1. Brown

1986; 42: 204-09. BU, Becker GJ, Kincaid-Smith PS. Clinical and biochemical features of aluminium-related bone disease. Kidney Int 1986; 29: suppl 18: 80-86. Milliner DS, Hercz HG, Miller JH, Shinaberger JH, Nissenson AR, Coburn JW. Clearance of aluminium by haemodialysis: effect of desferrioxamine. Kidney Int 1986; 29: suppl 18: 100-03.

4. Ihle

SiR,—Aluminium intoxication is a severe complication in patients undergoing long-term haemodialysis. The clinical features can include dialysis dementia, dialysis osteomalacia, anaemia, myopathy, and skin lesions. There are no laboratory findings which reliably correlate with the clinical symptoms. Bone biopsy is the definitive test for establishing the diagnosis of aluminium-induced osteomalacia. The administration of desferrioxamine (2 g during haemodialysis), a chelating agent that can mobilise aluminium from bone and other tissues, can serve as a diagnostic as well as a therapeutic tooP for aluminium intoxication. In 9 patients with renal insufficiency and bone symptoms (on haemodialysis for 7-21 yr) plasma aluminium (normal range up to 50 g/1) and bone aluminium levels (histochemical staining technique) were measured.2,3 We found no significant difference in

Serum-aluminium levels in haemodialysis patients with and without (- - -) aluminium-related bone disease.

the basal levels of serum aluminium betwreen 3 patients with aluminium-related bone disease and 6 with non-aluminium-related bone disease (figure). After administration of desferrioxamine we observed a sharp rise in serum aluminium (150 pg/1) in patients with aluminium-related bone disease after 48 h. In our opinion baseline aluminium levels are not generally helpful in the diagnosis of aluminium intoxication. Symptoms improve on treatment with desferrioxamine.4,5 The sources of aluminium accumulation were aluminium-containing drugs, aluminium-rich dialysate, and contamination by the tube system. To prevent the disease advancing and to reach an early diagnosis we used the chelating agent desferrioxamine in a dose of 2 g during haemodialysis. The desferrioxamine test is simple, non-invasive, and generally well tolerated. Moreover, the symptoms of patients with aluminium related bone disease improved after treatment with desferrioxamine.

5.

SECOND MALIGNANCIES AND HODGKIN’S DISEASE

SIR,-Dr Pedersen-Bjergaard and colleagues (July 11, p 83) an increased incidence of acute leukaemia in patients previously treated with alkylating chemotherapeutic agents for Hodgkin’s disease. They felt, however, that there was no correlation between these drugs and the subsequent development of solid tumours. We report two patients with previously treated Hodgkin’s disease in whom small-cell bronchial carcinoma developed in the report

last year, at a young age. Case 1.-A 33-year-old male smoker (10-15 cigarettes per day for 15 years) was diagnosed as having nodular sclerosing Hodgkin’s disease (IIB) in 1975. He was treated with mechlorethamine, vincristine, procarbazine, and prednisone (MOPP) for six pulses until January, 1976, and then received standard mantle 4 MeV radiotherapy. In June, 1986, he presented with a 3-month history of cough and exertional dyspnoea. Bronchial biopsy revealed smallcell bronchial carcinoma. His disease was poorly responsive to treatment and he died within 2 months of diagnosis. Case 2.-A 48-year-old male smoker (15-20 cigarettes per day for 30 years) was diagnosed as having lymphocyte-depleted Hodgkin’s disease (IIIB) in 1984. He was treated with eight alternating pulses of doxorubicin (’Adriamycin’), bleomycin, vinblastine, and dacarbazine (ABVD) and chlorambucil, vinblastine, procarbazine, and prednisolone (ChLVPP) between March and October, 1984, and then received radiotherapy to the area of bulk disease in his left groin. 2 years later he had histologically proven small-cell bronchial carcinoma limited to the thorax, and died within 2 months of ,

diagnosis. Although small-cell lung carcinoma can occur in young smokers we were struck by the young age of presentation in these two patients. We speculate about the possible contributory role of previous chemotherapy or radiotherapy in the early presentation of these patients who were at risk of lung cancer because of their smoking. Continued surveillance of patients cured of Hodgkin’s disease is required to elucidate fully the cumulative risk of second malignancies. Medical Oncology Unit, Western General Hospital, Edinburgh EH4 2XU

ALISON J. CHEYNE SIMON G. ALLAN