LIVER DAMAGE AFTER PARACETAMOL OVERDOSE COMPARISON OF LIVER-FUNCTION TESTS, FASTING SERUM BILE ACIDS, AND LIVER HISTOLOGY

LIVER DAMAGE AFTER PARACETAMOL OVERDOSE COMPARISON OF LIVER-FUNCTION TESTS, FASTING SERUM BILE ACIDS, AND LIVER HISTOLOGY

579 LIVER DAMAGE AFTER PARACETAMOL OVERDOSE COMPARISON OF LIVER-FUNCTION TESTS, FASTING SERUM BILE ACIDS, AND LIVER HISTOLOGY OLIVER JAMES M. LESNA ...

371KB Sizes 0 Downloads 31 Views

579

LIVER DAMAGE AFTER PARACETAMOL OVERDOSE COMPARISON OF LIVER-FUNCTION TESTS, FASTING SERUM BILE ACIDS, AND LIVER HISTOLOGY

OLIVER JAMES M. LESNA S. H. ROBERTS L. PULMAN P. A. SMITH ADRIAN P. DOUGLAS A. J. WATSON

Departments of Gastroenterology, Clinical Biochemistry, and

Pathology, Royal Victoria Infirmary, Newcastle upon Tyne

Patients and Methods Patients 155 patients have been admitted after paracetamol overdose. 54 (aged fifteen to fifty) fulfilled the criteria necessary for admission to this study. The criteria were: (1) daily liver-function tests and fasting serum bile-acid estimations available for at least four days after admission; (2) liver biopsy (informed consent having been obtained from the patients or their relatives); (3) no history of liver disease within two years preceding admission; (4) no history of ingestion of other potentially hepatotoxic drugs or of excess quantities of alcohol in the twenty-four hours before admission; and (5) HBAg negative.

Methods

54 patients have been studied after Sum ary paracetamol (acetaminophen) overdose. Liver-function tests and fasting serum bile-acids were measured daily; liver biopsy was done in all cases, and slides were examined "blind" to assess liver damage. The plasma-paracetamol was measured on one occasion. A histological abnormality was present in the livers of 53 of the 54 patients, and was minor in 23, moderate in 16, and severe in 14. In 6 patients with moderate and 18 with mild histological abnormality liver-function tests were normal. A serumaspartate-aminotransferase above 400 units/1 was always associated with severe histological liver damage. Fasting serum bile-acids were raised in 51 of the patients with abnormal liver histology; the serum-bileacid seemed to be a more sensitive indicator of mild liver-cell damage than was the transaminase level. There was, however, little correlation between increase in bile-acid concentration and the degree of histological abnormality. As a result of these investigations empirically determined levels of plasma-paracetamol have been drawn which give a guide to the likelihood of liver damage after paracetamol overdose.

Introduction SINCE 1966 paracetamol (acetaminophen) overdose has been recognised as potentially fatal.1-3 It can cause centrilobular hepatic necrosis and, in severe The number of hospital cases, hepatic failure. admissions in the United Kingdom with paracetamol overdose is increasing rapidly and has been estimated at 1000 per year.4 This may now be a serious under estimate. The fatality of severe overdose is high,5 and it is now one of the commonest causes of hepatic failure in Britain.6 While fulminant hepatic failure after paracetamol overdose is now recognised, the incidence and possible complications of more minor degrees of liver damage are less clearly known. To study these effects and, if possible, to offer any advances in treatment which became available, a regional paracetamol overdose service was set up in 1974. Physicians in the North East region were informed. Every patient thought to have taken a paracetamol overdose was admitted to the Royal Victoria Infirmary, Newcastle, where a plasma-paracetamol was estimated. We report here the results of serial liver-function tests and fasting bile-acid estimations together with the liver histology in patients admitted in this way.

Plasma-paracetamol was estimated by spectrophotometry7 the time of admission to hospital. Liver function was tested daily by measurement of serum bilirubin, aspartate aminotransferase (A.s.T.) (normal range 4-20 units!l), and alkaline phosphatase (normal range 20-90 units/1) together with prothrombin ratio. Serum-bile-acids (fasting) were estimated by spectrofluometry 8 (normal < 14 jAmol/1). Liver biopsy was done on the fourth day after admission to hospital or as soon as the prothrombin ratio and patient’s condition allowed thereafter. In 3 patients liver tissue was obtained just after death. Paraffin sections were prepared and stained with haematoxylin and eosin, periodic-acid/ Schiff with and without amylase pretreatment, Foot’s reticulin, Van Gieson, picro-Mallory, and Perls’ prussianblue. On the basis of histological findings three grades of abnormality were established with the following main

at

features: Grade 1.-Excess of lipofuscin in centrilobular hepatocytes where inappropriate for the age of the patient, focal hyperplasia of Kupffer cells, not more than two foci of hepatocytolytic necrosis. Grade II.-Multiple foci of necrosis, focal reticulin collapse, inflammatory reaction. Grade III.-Severe centrizonal necrosis, liver-cell loss and inflammation, reticulin collapse, features of active regeneration.

The slides were assessed by two pathologists independently and without knowledge of the patient’s clinical course and investigations. Liver sections from other patients pre-

viously reported as showing no abnormality and from patients with other liver conditions were included in the final " blind " assessment.

In the few instances where reached after consultation but before clinical information was obtained.

grades differed

a consensus was

Treatment

Each patient received an intravenous infusion of 5% dextrose 2-3 litres daily together with phytomenadione (’ Konakion ’) 10 mg, folic acid 15 mg, and ’Parentrovite’ (vitamins B and C) twin ampoules. Potassium was added to the infusion as indicated by electrolyte estimations. If liver function was normal in forty-eight hours after admission clear fluids were given by mouth and the infusion was discontinued after a further twenty-four hours. Normal diet was then resumed. Where moderate or severe abnormalities were observed in liver-function tests, the infusion was continued and lactulose 10 ml six-hourly was given. If signs or symptoms of hepatic encephalopathy developed the full conventional treatment for acute hepatic failure was carried out. 2 patients who reached grade-iv hepatic coma were treated with charcoal column hsemoperfusion; both died.

Cysteamine 11 patients whose initial plasma-paracetamol level was considered to be high were given cysteamine9 as part of a

580 controlled trial of this treatment for paracetamol overdose. These patients have not been separated from the others, as the trial is still in progress. Their management was in all other respects the same as the patients who did not receive

cysteamine. Results Liver Histology (fig. 1) 1 patient had normal liver histology; 23 had minor changes (grade I); 16 patients had moderate (grade II) abnormality; and 14 patients" showed" severe (grade III) changes. Among the control slides examined, all previously reported " normal " appearances were again reported as being normal.

Liver Function

(fig. 1) 24 patients (44%) had no abnormally high A.S.T. during the four days after hospital admission. 2 of these patients had a serum-bilirubin above 12 µmol/l on one occasion (22 and 32 µmol/l). (For bilirubin In no patient with 1 mg/100 ml =17-1 µmol/l.) a normal A.S.T. was there an abnormal prolongation of prothrombin-time. In 6 patients with grade n abnormality on biopsy there was no increase of A.S.T. or

Fig. 2-Liver-biopsy grade against

maximum fasting

serum-

bile-acid level.

Upper limit of normal

14

µmol/l.

bilirubin.

patients whose A.s.T. exceeded 400 units/I had (severe) liver-biopsy changes. grade All

III

Bile Acids (fig. 2) In patients with abnormal liver biopsy, only 2 had a normal fasting serum-bile-acid, in both cases at the upper limit of the normal range (14 µmol/l or 0-68 mg/ 100 ml with reference to glycocholic acid). There was

Fig. 3--Plasma-paracetamol

on

admission to hospital.

mean levels of serum-bile-acid with patients grade i and grade n liver While there was considerable overlap of histology. results between the levels of all three biopsy grades, 6 of the 14 patients with severe liver damage had peak bile-acid concentrations over 100 µmol/l, including all 3 patients who died. No other patient had a level above 100 µmol/l. All grade-III patients had peak fasting serum-bile-acids above 40 µmol/l.

no

difference in the

between

Paracetamol Levels (fig. 3) Plasma-paracetamol concentrations

.

Fig. 1-Liver-biopsy grade against maximum A.S.T. Upper limit of normal 20 units/I.

were

widely

scattered. Because only one blood-sample was taken on admission or at two hours after ingestion, no extrapolation of results is possible. 9 out of 11 patients whose plasma-paracetamol level lay above the upper line in fig. 3, whether treated with cysteamine or not, had severe liver-biopsy changes. Only 2 of 17 patients whose plasma-paracetamol level lay below the lower line showed more than mild biopsy changes, and these 2 patients both had grade-n changes with slightly abnormal liver function and moderate elevation of fasting serum-bile-acid. In 13 of the 22 patients

581

plasma-paracetamol level lay between the two were grade 11. 1 patient had more severe (grade ill) damage. whose

to

-

Discussion

Liver It

severe liver damage. It is difficult know what influence cysteamine treatment or the routine management with intravenous dextrose had A controlled trial is at present in on these patients. progress to answer these questions.

subject sustained

1

lines, the histological changes

Damage

that almost all patients who take paraoverdose sustain some hepatic damage, as cetamol seems

liver biopsy. Where liver-function tests remain normal this damage is usually mild (gradeI in

judged by

classification), but 6 patients showed moderate changes despite normal liver function. It thus seems our

that transaminase levels are not a very sensitive indicator of mild or moderate degrees of liver damage. In this respect fasting serum bile-acid measurement was more sensitive, being abnormal in all patients with grade-n biopsies and in all but 2 of the patients with grade-i changes. Kaplowitz et al.10have shown that postprandial bile-acid levels may be more sensitive than fasting values, but our treatment regimen did not permit postprandial levels to be measured since patients had no food for the three days after admission. Korman et al. 11showed that serum-bile-acid level was a more sensitive indicator of activity in chronic active liver disease than conventional liver-function tests when liver histology was examined, and we have confirmed this finding in acute hepatic damage. There was, however, a large overlap in bile-acid levels in all three histological groups, and the highest concentration recorded in an individual patient was a poor predictor of the abnormality found at histology. Only above 100 mol/1 was there no overlap between patients with grade-ill histology and patients in the other two groups. 3 of 5 patients with peak bile-acid levels above 100 .mol/1 died. Where more severe liver damage had been sustained A.s.T. was more useful; there was no overlap between levels seen in patients with grade I and II changes on biopsy (all under 400 units/1) and levels in patients with grade-in changes (all over 400 units/I). Thus A.s.T. levels over 400 units/l gave a good prediction of severe liver

damage. Paracetamol Levels Because only one paracetamol plasma level was estimated, and in view of the unreliability of patients’ reports of the amount of paracetamol swallowed, extrapolation of blood-levels from different times to the same time after ingestion was impossible. The empirical lines in fig. 3 do, however, offer guidelines for management of paracetamol overdose. Patients whose plasma-paracetamol on admission lay under the lower line were very unlikely to sustain more than mild histological liver damage, and such patients can be discharged from hospital with little risk of liver damage. Individuals whose paracetamol level lay above the upper line were at great risk of developing severe hepatic damage whether they received cysteamine or not-indeed, 3 of these patients died. This

line is very similar to that of Prescott et al. who suggested that patients whose plasma-paracetamol level fell above the line should receive cysteamine. It is between the two lines that doubt arises; although most patients in this area sustained moderate liver damage some had only mild liver damage and

We thank the Scientific and Research Committee of the Newcastle University Hospitals for a research grant to 0. J. and L. P., and Prof. David Kerr for his help and advice with this project. The following performed the paracetamol estimations: Mr J. A. Fleetwood, Miss M. Goldfinch, Dr M. Watson, Dr J. Ahmed, and Miss J. Lawson. Requests for reprints should be addressed to 0. J. REFERENCES 1. Davidson, D. G. D., Eastham, W. N. Br. med. J. 1966, ii, 497. 2. Thomson, J. S., Prescott, L. F. ibid. p. 506. 3. Proudfoot, A. T., Wright, N. ibid. 1970, iii, 557. 4. ibid. 1975, i, 536. 5. Clark, R., Thompson, R. P. H., Borirakchanyavat, V., Widdop, B., Davidson, A. R., Goulding, R., Williams, R. Lancet, 1973, i, 66. 6. Murray-Lyon, I. M., Portmann, B., Gazzard, B. G., Williams, R. in Artificial Liver Support (edited by R. Williams and I. M. 7. 8. 9. 10. 11.

Murray-Lyon); p. 244. London, 1975. Routh, J. I., Shane, N. A., Arrendondo, E. G., Paul, W. D. Clin. Chem. 1968, 14, 882. Murphy, G. M., Billing, B. A., Baron, D. N.J. clin. Path. 1970, 23, 594. Prescott, L. F., Newton, R. W., Swainson, C. P., Wright, N., Forrest, A. R. W., Matthew, H. Lancet, 1974, i, 588. Kaplowitz, N., Kok, E., Javitt, N. B.J. Am. med. Ass. 1973, 255, 292. Korman, M. G., Hofmann, A. F., Summerskill, W. H. J. New Engl. J. Med. 1974, 290, 1399.

INCREASED URINARY EXCRETION OF ALBUMIN, LIGHT CHAINS, AND &bgr;2-MUCROGLOBULIN AFTER INTRAVENOUS ARGININE ADMINISTRATION IN NORMAL MAN E. VITTINGHUS C. E. MOGENSEN K. SØLLING

University Clinic of Internal Medicine, Kommunehospitalet, 8000 Aarhus C, Denmark

Second and First

Summary

arginine

in six

40-50 min

period

Infusion of 60 g of

healthy

men over a

caused a 16-fold increase in the excretion-rate of albumin from 8·6 to 142·4 µg/min. After injection of 3, 6, 9, and 12 g of arginine the mean albumin-excretion rate of another six normal subjects rose from 5·8 µg/min to 9·4, 13·2, 21·6, and 33·9 µg/min, respectively. Excretion of light chains of immunoglobulin increased from 5·5 µg/min to 18·0, 30·0, 40·5, and 52·0 µg/min, respectively, and the excretion rate of &bgr;2-microglobulin increased from 0·092 µg/min to 2·97, 9·23, 16·39, and 25·71 µg/min. Thus a clear dose response pattern was found. The results strongly suggest that the mechanism behind the arginine-induced protein excretion is an inhibition of tubular reabsorption. Introduction INCREASED protein excretion is consistently provoked in normal man by several factors such as heavy exercise, fever, and catecholamine infusion.1-4 We have found5 that a moderate exercise load that did not increase albumin excretion in normal man (600 kp m min-1 for 20 min) provoked an increase of albumin excretion in a group of diabetic patients