DAPSONE AND SEVERE HYPOALBUMINÆMIA

662

DAPSONE AND SEVERE HYPOALBUMINÆMIA A

normal,

Report of Two Cases

Departments of Medicine, St. Bartholomew’s Hospital and St. Mary’s Hospital, London, and Clinical Research Centre, Harrow

hypoalbuminæmia developed in patients on long-term dapsone treatment for dermatitis herpetiformis. The patients had been treated with dapsone for 3 and 11 years before the syndrome developed and both recovered completely when dapsone was withdrawn. Albumin-turnover studSevere

two

ies revealed a great increase in intravascular albumin catabolism and a modest decrease in synthesis.

Introduction FOR

leprosy.

30 years dapsone has been used to treat In the U.K. dapsone is principally used to treat

over

diseases of the skin, particularly dermatitis herpetiformis. Adverse effects were commonly reported in the early years when doses of 1000-2000 mg/day were administered. Red and white cell toxicity are important side-effects. Other less common reactions include skin manifestations, hepatitis, neurological and psychiatric disorders, gastrointestinal upsets, and a nephrotic syndrome.’ Toxic effects are uncommon now that much smaller doses are used. We report a hitherto unrecognised syndrome of almost fatal but reversible hypoalbuminaemia in two patients on long-term dapsone therapy for dermatitis her-

petiformis.

Case-reports Case1 A 51-year-old man in whom dermatitis herpetiformis had developed in 1971 was treated with dapsone, 100 mg daily. In 1974 dyspnoea, polyuria, facial and peripheral oedema, and a large pleural effusion developed. Serum albumin was 29 g/1

and the

erythrocyte-sedimentation

rate was

117

mm

in the

11. Ainbender E, Hirschhorn K. Routine alpha-fetoprotein studies in amniotic fluid. Lancet 1976; i: 597-98. 12. Kjessler B, Johansson SGO, Sherman M, Gustavson K-H, Hultquist G. Alpha-fetoprotein in antenatal diagnosis of congenital nephrosis. Lancet

1975; i: 432-33. 13.

14.

Seppälä M, Aula P, Rapola J, Karjalainen O, Huttunen N-P, Ruoslahti E. Congenital nephrotic syndrome: prenatal diagnosis and genetic counselling by estimation of amniotic-fluid and maternal serum alpha-fetoprotein. Lancet 1976; ii: 123-25. Brock DJH. Prenatal diagnosis of neural tube defects through amniotic fluid

In: Crandall BF, Brazier MAB, eds. Prevention of neural tube defects. New York: Academic Press, 1978: 155-62. 15. Rodeck CH, Campbell S. Sampling pure fetal blood by fetoscopy in second trimester of pregnancy. Br Med J 1978; ii: 728-30. 16. U.S. National Institute of Child Health and Human Development Study Group. Midtrimester amniocentesis for prenatal diagnosis: Safety and acmeasurement.

curacy. JAMA 1976; 236: 1471-76. Simpson NE, Dallaire L, Miller JR, Siminovich L, Hamerton JL, Miller J, McKeen C. Prenatal diagnosis of genetic disease in Canada: report of a collaborative study. Can Med Assoc J 1976; 115: 739-46. 18. Report to the Medical Research Council by their Working Party of Amniocentesis. An assessment of the hazards of amniocentesis. Br J Obstet Gynœcol 1978; 85: suppl no 2: 1-41. 19. Office of Population Censuses and Surveys. Mortality statistics: review of the Registrar General on deaths in England and Wales, 1975. London: H.M. Stationery Office.

17.

as were

hsematological and biochemical tests were bronchoscopy and renal, liver, and pleural

He remained on dapsone and his serum albumin fluctuated between 21 and 25 g/1. In May, 1975, the pleural effusion worsened and he was treated with prednisolone. He was reinvestigated because the serum albumin continued to fall. Gamma-globulins, liver function tests, electrolytes, and blood-counts remained normal but renal function was slightly impaired, with blood urea 9 nmol/1 (54 mg/dl) and creatinine clearance 70 ml/min. Enteric protein loss was suspected because the patient had dermatitis-herpetiformis enteropathy, with partial villous atrophy and a slight inflammatorycell infiltrate. A chromium-51 chloride study confirmed a modest protein loss of 2.5% per day (normal <1%). A barium meal and follow-through showed coarse mucosal folds. Gastroscopy with biopsy, sigmoidoscopy with biopsy, barium enema, Schilling test, fxcal-fat excretion, and xylose tolerance test were normal. Pleural and ascitic fluid had a protein content of 8-10 g/1 and were microscopically normal. A biopsy specimen of unaffected skin showed IgA granular fluorescence in the papillary dermis, consistent with dermatitis herpetiformis. Prednisolone and dapsone were continued and a gluten-free diet was started. His condition deteriorated as pleural effusions, ascites, hepatosplenomegaly, and pronounced oedema developed. Serum albumin continued to fall, to 8 g/1, and sodium fell to 129 mmol/1. A protein-turnover study (see accompanying table) demonstrated a high albumin catabolic rate, not accounted for by his mild protein-losing enteropathy, and a low albumin synthetic rate. A lymphoma was suspected and so a laparotomy was performed in November, 1975. Ascites and enlarged cedematous viscera were seen; biopsy specimens of liver, jejunum, peritoneum, and mesenteric lymph-nodes provided no new information. Postoperatively massive oedema, ascites, and heart-failure developed and he seemed unlikely to survive. It was suggested that dapsone might be responsible, and after withdrawal of this agent his clinical condition gradually improved and fluid retention responded to diuretics. In the next 6 weeks oedema disappeared and serum albumin rose to 40 g/l. Dermatitis herpetiformis was controlled with sulphapyridine. Turnover studies were repeated one year later. Albumin metabolism was almost normal (see accompanying table). 3 years later the patient remains completely well on sulphapyridine, with no biochemical abnormalities.

biopsy specimens.

PAUL SWAIN JEREMY G. C. KINGHAM EDWIN T. SWARBRICK JEFFREY G. WALKER ANTHONY M. DAWSON

Summary

first hour. Other

.

Case 2 In 1966 dermatitis herpetiformis developed in a 26-year-old He was treated with dapsone, 150 mg daily. In 1975 a jejunal biopsy specimen showed partial villous atrophy consis-

man.

20. Butler NR, Alberman E. Perinatal Problems: the second report of the 1958 British Perinatal Mortality Survey. Edinburgh: E. & S. Livingstone, 1969. 21. British Births 1970. London: William Heinemann Medical books, 1975. 22. Jones MD, Burd LI, Bowes WA, Battaglia FC, Lubchenco LO. Failure of association of premature rupture of membranes with respiratory-distress syndrome. N Engl J Med 1975; 292: 1253-57. 23. Chalmers I, Dauncey ME, Verrier-Jones ER, Dodge JA, Gray OP. Respiratory distress syndrome in infants of Cardiff residents during 1965-75. Br Med J 1978; ii: 1119-21. 24. Warburton D, Fraser FC. Spontaneous abortion risks in man: data from reproductive histories collected in a medical genetics unit. Human Genet 1964; 16: 1-25. 25. Alberman E, Elliott M, Creasey M, Dhadial R. Spontaneous abortion and reproductive history. Br J Obstet Gynœcol 1975; 82: 366-73. 26. Wald N, Barker S, Cuckle H, Brock DJH, Stirrat GM. Maternal serum alpha-fetoprotein and spontaneous abortion. Br J Obstet Gynæcol 1977,

84: 357-62. NJ, Cuckle H, Stirrat GM, Turnbull AC. Maternal serum alpha-fetoprotein and birth weight in twin pregnancies. Br J Obstet Gynœcol 1978, 85: 582-84. 28. Wald N, Cuckle J, Stirrat G. Maternal serum alpha-fetoprotein levels in triplet and quadruplet pregnancy. Br J Obstet Gynœcol 1978; 85: 124-26. 29. Wald NJ, Cuckle HS, Boreham J, et al. Antenatal screening in Oxford for fetal neural tube defects. Br J Obstet Gynœcol 1979; 86: 91-100. 27. Wald

663 IODINE-125-ALBUMIN

IV = intravenous. *The absolute synthetic

bolic rate because these this case.

rate

TURNOVER STUDIES IN CASE

1

is devised

rates are

by measuring the absolute cataequal in the steady state-i.e., as in

dermatitis-herpetiformis enteropathy. He continued In February, 1977, he suddenly became unwell, with gross facial and leg cedema and ascites. Serum albumin was 16 g/1, urea 7-1 mmol/1 (43 mg/dl) and sodium 130 mmol/1. Liver-function tests, gamma-globulin, full bloodcount, urine analysis, and tests for malabsorption were normal. Enteric protein loss was thought to be the cause of his hypoalbuminæmia and a chromium-51 chloride protein-losing enteropathy test demonstrated a modest increase in faecal loss of 5% per day (normal <1%). He was treated with a glutenfree diet, diuretics, albumin infusions, and dapsone. His condition deteriorated and he was readmitted to hospital in August, 1977, with cachexia, hepatosplenomegaly, and ascites. Serum with

tent on

dapsone.

albumin had fallen to 14 g/1 and sodium to 114 mmol/1 and liver-function tests had become abnormal, with alkaline phosphatase 630 iu (liver isoenzyme) and serum glutamic oxaloacetic transaminase 86 iu. The ascitic fluid contained 14 g/1 protein. The following investigations were normal: chest X-ray; barium meal, follow-through, and enema; abdominal ultrasound and computerised axial tomography; inferior vena cavagram and hepatic venogram; right atrial and wedge hepatic vein pressures; bone-marrow biopsy, liver biopsy, peritoneal biopsy; gastroduodenoenteroscopy; and laparoscopy. Multiple jejunal biopsy specimens showed no worsening of his mild enteropathy. The chromium-51 chloride protein-losing enteropathy test again showed a modest loss (4% per day). There was no evidence of malabsorption. A gluten-free diet, dapsone, and diuretics were continued and prednisolone was started. There was a slight clinical improvement. In November, 1978, he was readmitted confused and grossly wasted, with massive ascites and circulatory failure. His bloodpressure fell to 70/50 mm Hg, serum sodium fell to 114 mmol/l, and urea rose to 15 mmol/1 (91 mg/dl). Attempts to treat his fluid retention with diuretics exacerbated both his blood-pressure and his biochemical abnormalities. He was believed to be dying. All treatment except a low-dose of prednisolone was stopped in December, 1978. Within a week he was less confused, his circulatory state was improving, and fluid retention responded to diuretic therapy. After 2 weeks dermatitis herpetiformis recurred and sulphapyridine was started. 5 weeks after withdrawal of dapsone the ascites had cleared, he

1—Case 2, 3 months before dapsone withdrawal 6 months after dapsone withdrawal (right).

Fig.

case

until the very end because both

men

(left) and

had been

on

therapy for a long time before the illness started. The similarity of the illnesses and rapid recovery after withdrawal make it extremely likely that dapsone was responsible although the relevance of the underlying condition of dermatitis herpetiformis is unclear. 50-80% of dapsone is albumin-bound. 1.2 This binding may have caused the extraordinary increase in albumin catabolism. Perhaps dapsone acts as a hapten and in-

and serum albumin, urea, and electrolytes normal. He was sent home on sulphapyridine. 7 months later he is fit (fig. 1) and continuing his hobby of karate. The course of his disease and its relation to dapsone treatment and other therapy are shown in fig. 2. was

gaining weight,

were

Discussion

Both these patients with dermatitis herpetiformis had received dapsone 100-150 mg daily for several years (3 years and 11 years) without any ill effect when severe

hypoalbumaemia suddenly developed. were

similar. Dapsone

was not

Their illnesses

incriminated in either

Feb

ul

1977 Aug

1978

Sep Oct

∥Feb∥

Nov

Dec

1979 Jan

2-Serum sodium and serum albumin during course of illin case 2 and their relation to dapsone and its withdrawal. Girth is given as a measure of ascites.

Fig.

ness

664

duces immunological destruction of albumin by the reticuloendothelial system. The observed decrease in albumin synthesis is not explained, but may reflect generalised impairment of synthetic processes in a dying

patient. In view of the extensive use of dapsone this metabolic defect has probably been observed before but has remained undiagnosed. In drawing attention to this syndrome we hope that more cases may be recognised and deaths may be avoided. The mechanisms underlying this hypoalbuminæmia must be investigated in further cases.

Knight,

Dr A. H. 2 to A.M.D.

Stoke Mandeville

Hospital, kindly referred

case

Requests

for

reprints should be addressed to J.G.C.K., GastroenterBartholomews’ Hospital, West Smithfield, Lon-

ology Department, St. don EC1A 7BE.

REFERENCES 1. Graham WR. Adverse effects of dapsone. Int J Dermatol 1975; 14: 494-500. 2. Riley RW, Levy L. Characteristics of the binding of dapsone and monoacetyl dapsone by serum albumin. Proc Soc Exp Biol Med 1973; 142: 1168-70.

DOES THE NEPHROTIC SYNDROME INCREASE THE RISK OF CARDIOVASCULAR DISEASE? R. J. JARRETT V. J. WASS C. CHILVERS* J. S. CAMERON On behalf of the South-East Thames Renal Physicians† Renal Unit and Department of Community Medicine, Hospital, London SE1 9RT

Summary

Cardiovascular were

Guy’s

mortality and morbidity a mean follow-up

assessed, after

period of 5 years, in an unselected series of 159 adults presenting with the nephrotic syndrome between 1972 and 1975. 60% of the deaths were attributed to terminal renal failure, and the incidence of deaths from ischæmic heart-disease (IHD) was not significantly above normal. The proportion of patients experiencing angina and intermittent claudication and the prevalence of ischæmic electrocardiographic changes did not differ significantly from those of a London control population. At fol-

low-up, hypertension was significantly more common (p<0·001) in male nephrotic patients than in controls. Earlier reports of a greatly increased incidence of IHD in unselected patients with the nephrotic syndrome were not confirmed. Routine treatment of hyperlipidæmia in the nephrotic syndrome is not, therefore, recommended. Introduction

(predominantly a Fredrickson abnormality) is found in many patients with the nephrotic syndrome,1,2 but whether atherosclerosis is more common in such patients than in healthy people HYPERLIPIDÆMIA

type-n

*Present address: Institute of Cancer Research, Sutton,

Surrey.

†C. S. OGG and D. G. WILLIAMS, Guy’s Hospital London SE1 9RT. M. GOGGIN and D. PROSSER, Kent and Canterbury Hospital, Canterbury, Kent, CT1 3NG. A. J. P. TRAFFORD and P. SHARPSTONE, The Royal Sussex County Hospital, Brighton, Sussex BN2 5BE. V. PARSONS and M. J. WESTON, Dulwich Hospital, East Dulwich Grove, London SE22 8PT.

has never been clearly established. Although a greatly increased incidence of ischaemic heart-disease (IHD) has been claimed in adults with the nephrotic syndrome, this conclusion was based on two very small series of 153 and 174patients respectively and has been disputed by workers with a larger clinical experience.5-7 Necropsy evidence of increased atherosclerosis is similarly based on a group of only 20 nephrotic patients, 13 of whom had systemic lupus erythematosus (SLE) and 4 diabetes mellitus:8 both these diseases, in their own right, predispose to premature cardiovascular disease. Nevertheless, treatment of hyperlipidæmia in the nephrotic syndrome has been recommended,3.4but, unfortunately, increased clofibrate toxicity was subsequently reported in these patients.9 Because the evidence is unsatisfactory, we have followed up a larger and unselected series of adults with the nephrotic syndrome who formed part of the SouthEast Glomerulonephritis Survey established in 1972. The-incidence and prevalence of cardiovascular disease and associated risk factors in this population has been determined. Data on glucose tolerance, total lipids, and cholesterol fractions in these patients will be presented elsewhere. Patients From January, 1972, to December, 1975, details were kept of all patients with glomerulonephritis referred for renal biopsy in the South-East Thames Region (population 3.6 million). Patients presenting with oedema associated with a plasma albumin <3.0 g/1 (AutoAnalyzer dye method) and 24 h urinary protein >3.0 g/day (biuret method) were defined as having the nephrotic syndrome. 159 patients were traced, excluding children (aged < 15 years on referral) and, because of increased cardiovascular mortality associated with the disease itself, patients with diabetes mellitus. There were 101 men and 58 women with a mean age (±SD) of 44.1±18.5 years at the onset of the nephrotic syndrome. Renal histology findings are given in table i. At follow-up in 1978, 41 (26%) had died, 8 (5%) were on haemodialysis or transplant programmes, and 2 (1%) were abroad and could not be traced. Of the 108 (68%) surviving patients, 59 had had continuous and 37 intermittent episodes of proteinuria during the follow-up period, whereas 12 had had only one episode of the nephrotic syndrome, and this had lasted less than 6 months. 61 patients had been treated with steroids at some stage of their illness, but 1 had received treatment for hyperlipidaemia.

Methods Assessment of Cardiovascular Mortality Causes of death were established from hospital records, death certificates, and, where available, necropsy data. All patients who had started dialysis, been transplanted, or had a plasma creatinine >1000 µmol/l at the time of death were defined as "dying" from terminal renal failure (TRF).

Assessment

of Cardiovascular Morbidity

104 (69 men) of the 108 surviving patients (96%) completed the WHO cardiovascular questionnaire in its self-administered version, 10 and from this the prevalences of angina and intermittent claudication and a past history of myocardial infarction (MI) were assessed. In addition, 73 (48 men) of the patients (67%) who were still attending hospital had, after fasting for 12 h, a resting 12-lead electrocardiogram (ECG) which was coded by two independent non-medical observers according to the Minnesota Code.’° "Probable ischaemia" was diagnosed on Q-wave changes (codes 1-1, 1-2) and "possible ischaemia" on sT-segment and T-wave abnormalities (codes 1-3, 4.4-4,