491

worth, happiness, and self-esteem, only to find that the cur-rency debases.

Lordswood House,Harborne, Birmingham B17 9DB ANTON R. DEWSBURY

LITHIUM-INDUCED URÆMIA?

SIR,-Dr Hestbech and Dr Aurell (Jan. 27, p. 212) reporta case of uraemia in a patient on lithium and conclude that"lithium treatment, even when well controlled, may causesevere renal damage". Poorly controlled lithium treatmentmay lead to toxicity with disturbance of renal function, andeven well-controlled treatment may be associated with inhibi-tion of tubular water reabsorption, resulting in polyuria andthirst. However, severe disturbance of glomerular filtrationwith uraemia had not been described before, so how well con-trolled was the lithium treatment in Hestbech and Avrell’s pa-tient ?

Their patient twice had raised serum-lithium values (1-4mmol/1 in December, 1975, and 2.1 mmol/1 in September,1977, on the second occasion accompanied by tiredness). InNovember, 1977, the patient was given antihypertensive treat-ment (unspecified) which could have led to loss of sodium, low-ered lithium excretion, and a rise in the serum-lithium concen-tration. This patient thus had one or more mild lithium

intoxications, and since he also had severe lithium-induced’polyuria, it is not surprising that his kidney showed morpho-logical changes which resembled those observed previouslyby Hestbech and colleagues in lithium-treated patients selectedfor having polyuria or having had lithium poisoning. The his-tory of nephrolithiasis and the development of hypertensionwith renal arteriosclerosis may or may not have played a rolein this case.Hestbech and Aurell state that the "the rapid progression of

renal insufficiency stopped when lithium treatment was with-drawn." Their data, however, show the opposite. Apart froma serum-creatinine of 152 umol/l (1-69 mg/dl) shortly beforelithium treatment was stopped there is no indication of kidneymalfunction during lithium therapy. It was not until 1-3months after discontinuation of lithium that severe renaldeterioration developed-namely, a rise in serum-creatinine to500 umol/1 (5-66 mg/dl) in February, 1978, and a fall in glo-merular filtration-rate from 38 ml/min in January, 1978, to 13ml/min in March. This pattern differs from that seen in Hest-bech’s earlier study, where seven patients had reduced creati-nine clearance due to lithium poisoning; after elimination oflithium from the body the creatinine clearance remained

moderately low in three patients and rose to normal in four.No patient showed progressive deterioration of renal functionwith development of uraemia.

In Hestbech and Aurell’s case lithium probably was respon-sible for the morphological kidney changes, but whether thefall in G.F.R. was an after-effect of treatment, whether theurxmia was caused by kidney disease that was independent oflithium (the patient had proteinuria, renal arteriosclerosis,hypertension, aneemia, and a raised E.S.R.), or whether kidneydisease led to uraemia because the patient’s kidneys hadbecome vulnerable as a result of lithium intoxications due to

inadequately controlled treatment remain open questions.Many patients are on long-term lithium for affective dis-

order, and in some serious organic illness may develop at sometime. Such cases should be assessed carefully to determinewhether the treatment caused the illness, and disturbed kidneyfunction calls for special attention. Large-scale investigationsof the effects of lithium on renal function which are under waymay help the doctor to weigh the benefits and hazards of lith-lum treatment.

Psychopharmacology Research Unit,Aarhus University Institute of PsychiatryPsychiatric Hospital,DK-8240 Risskov, Denmark PER VESTERGAARD

D.D.A.V.P. TEST IN ASSESSMENT OF RENALFUNCTION DURING LITHIUM THERAPY

SIR,-Lithium treatment impaires tubular function’ and isassociated with morphological kidney lesions.2 Dehydrationhas been used to assess the urinary concentrating capacity.This procedure is very uncomfortable for the polyuric patient,it may be hazardous, and it is impossible in some poorly co-operating psychotic patients. We have therefore been testing aD.D.A.V.P. (1-desamino-8-arginine vasopressin) test instead. Uri-nary osmolality was measured 3 h after 40 fLg D.D.A.V.P.

(’Minirin’, Ferring AB, Sweden) administered intranasally.The result was compared with that of a 14 h period of total

UTNYUKAIIUN ]t:1 UKINAK1’ UbMULALIIYimosm/Kgm2u

Comparison of D.D.A.V.P. and dehydration tests.

restriction of fluid intake. Eighteen inpatients on lithium werestudied. The D.D.A.V.P. test could readily be performed in allof them, but five patients refused, or were too psychotic to par-ticipate in, the dehydration test.The results in the remaining thirteen patients showed a

satisfactory correlation between the two tests over a wide

range of urinary concentrating capacities (see figure). No side-effects during the D.D.A.V.P. test were reported. Seven patients.agreed to prolong the dehydration period to 18 h, but therewas little further increase in urinary osmolality (9 + 35 mos-mol/kg water ; mean ± s.E.M.).We consider the D.D.A.V.P. test to be a safe, convenient, and

reliable measure to estimate renal concentrating capacity inpatients treated with lithium. Unlike the dehydration test, itcan readily be done in psychotic patients. We now include itas a routine method in our regular assessment of tubular func-tion in these patients.

Department of Medicine,University Hospital,S-901 85 Umeå, Sweden

KJELL ASPLUND

ANDERS WAHLIN

Umedalen Hospital, Umeå WALTER RAPP

COMPLICATIONS OF DISOPYRAMIDE

SIR,-Dr Nies and co-workers (Feb. 10, p. 330) report aninteresting case involving complications due to disopyramidetherapy. Whilst disopyramide was prescribed at a dose withinthe recommended range, the toxicity is hardly surprising inview of the high plasma concentrations of the drug: On thebasis of the 13 - 8 h half-life of disopyramide elimination in thispatient even the "trough" plasma-disopyramide concentrationduring therapy would have been above 6.0 0 mg/l.Our experience of measuring plasma-disopyramide concen-

trations by specific gas-liquid chromatography3 suggests that

1. Bucht, G., Wahlin, A. Lancet, 1978, i, 779.2. Hestbech, J., and others. Kidney Int. 1977, 12, 205.3. Hayler, A M., Flanagan, R. J. J. Chromatogr 1978, 153, 461.

492

the concentrations achieved in Nies’ patient were very high.The Denver group do not state their normal disopyramide con-centrations but here, in a group of 19 inpatients (13 males, 6females, mean age 54±13-3 [s.D.] years) receiving 600 mg dis-opyramide daily, the mean plasma drug concentration in ablood-sample collected close to the time of oral dosage was3-09±1-28 mg/1 (s.D.). Only one patient was reported to haveside-effects-namely, dry mouth, blurred vision, drowsiness,nausea, vomiting, and anorexia. The disopyramide concentra-tion was 3.9 mg/l but the plasma digoxin concentration at thistime was 3.0 ug/l; none of these symptoms had been present2 weeks earlier when the disopyramide and digoxin concentra-tions had been 4.3 mg/l and 1 . pLg/1, respectively. Our resultsare very similar to the mean plasma concentration found by C.Oshrain and others (G. D. Searle, unpublished) of 2.9 mg/1achieved in a previous study, in which patients received 150mg disopyramide four times a day for 8 weeks.

Other factors could explain the high plasma concentrationsin Nies’ case, such as impaired renal clearance or the effectsof other drug therapy on absorption, distribution, and metab-olism of disopyramide. It is not, therefore, advisable to rely tooheavily on the prescribed dose as a measure of therapeutic effi-cacy. Whilst it is right to urge due caution in the prescriptionof potent antiarrhythmic agents we would suggest thatmeasurement of the plasma drug concentration at an earlierstage might have helped to elucidate this complex pharmaco-logical problem.Cardiac Drug SectionPoisons Unit.

Guy’s Hospital,London SE1

DAVID W. HOLTANGELA M. HAYLER

THEOPHYLLINE PHARMACOKINETICS DURINGRESPIRATORY VIRAL INFECTION

SIR,-There has been much interest of late’ in factorswhich affect theophylline pharmacokinetics, including age,body-weight, smoking, diet, and concurrent illness, particu-larly liver dysfunction. Fleetham et al. found a doubled serumhalf-life of theophylline during an acute viral illness in a singlesubject, following an earlier report by Chang et al. of reducedtheophylline clearance during acute respiratory viral infectionin children with chronic asthma. The implication is that pa-tients previously well maintained on theophylline derivativesmight require a modification of their drug dosage if toxic side-effects are to be avoided during an acute respiratory viral ill-ness.

We have looked at side-effects in twelve patients receivingmaintenance therapy with a sustained-release theophyllinederivative (’Phyllocontin’). The patients had established un-complicated chronic bronchitis. Four had serological evidenceof acute viral infection with symptoms of respiratory-tract in-fection during the 6-week assessment period. The other eightpatients were screened over the same period and had no clini-cal evidence of respiratory infection or change in viral titres.Comparison of factors affecting theophylline pharmacokineticsin these two groups provided no evidence of an increased pro-pensity to reduced theophylline clearance in the four patientswith viral infection. Moreover, they were heavier smokers thanthose in the non-infected group and this would tend to lowerrather than raise serum-theophylline levels.4 Of the four pa-tients with viral infection three had side-effects (comparedwith four out of the other eight) and mean serum-theophyllinelevels were increased (11.2 ag/ml compared with 7.5 ug/ml).Estimation of theophylline levels in the post-infection periodwas possible in only one of the four patients because the othersexperiencing toxicity preferred to discontinue the theophylline

1. Ogilvie, R I. Clin. Pharmacokinet. 1978, 3, 2672. Fleetham, J. A., and others Lancet, 1978, ii, 892.3. Chang, K., and others ibid. 1978, i, 1132.4. Powell, J. R., and others Am. Rev. resp. Dis 1977, 116, 17.

preparation. This one patient had a serum-theophylline of 9.6(J.g/ml compared with 13.4 4 µg/ml during the viral illness.These findings support Chang’s comments relating to reducedtheophylline clearance during acute viral respiratory infection,and we support the view of Chang and colleagues that clin-icians should be aware of the increased liability to toxicity dur-ing these periods.Centre for Respiratory InvestigationRoyal Infirmary,Glasgow G4 0SF

C. J. CLARKG. BOYD

LYSOSOMAL ENZYMES AND PANCREATITISDURING RIFAMPICIN THERAPY

SIR,-Rifampicin has potent hepatic microsomal enzyme-inducing properties in man’ and we have recently extended ourstudies to its effect, in combination with isoniazid, on the lyso-somal enzymes &bgr;-glucuronidase and -N-acetyl-glucosamini-dase.2 Both enzymes were significantly increased in plasmawhereas neither the microsomal enzyme i-glutamyltranspepti-dase nor the cytosol enzyme alanine transaminase (monitoredto exclude hepatocellular damage) were affected. Rifampicinand isoniazid thus contrast with anticonvulsants which raiseboth &bgr;-glucuronidase and i-glutamyltranspeptidase.3 This

finding emphasises the variability of enzyme induction by dif-ferent active drugs.

MEAN+S.D. LYSOSOMAL ENZYME ACTIVITIES (nmol/1/min)BEFORE AND DURING RIFAMPICIN/STREPTOMYCIN THERAPY*

*Rifampicin 450 mg + streptomycin 0.75 or 1.00 g intramuscularlydaily for 3-43 days.tMethods of Woollen and Walker .4,5

We have now established that this lysosomal enzyme effectis due to rifampicin alone. Individual patients were followedup before and during early treatment with rifampicin com-bined for the purpose of this study with streptomycin (seetable). Both lysosomal enzyme activities rose significantly butsubsequent substitution of streptomycin by isoniazid did notfurther change enzyme activity.One patient had developed acute pancreatitis on the seven-

teenth day of treatment with rifampicin and streptomycin (ser-um-amylase 4300 i.u./I). Although subsequent investigationshowed multiple gallstones, pancreatitis can occasionally com-plicate rifampicin therapy6 and the question arises as to

whether this is connected with the observed rise in lysosomalenzymes. Two possible mechanisms, other than antibody-mediated reactions, include concentration of rifampicin or itsmetabolites in lysosomes with resulting toxicity, or a deficiencyof tissue inhibitors of these enzymes in susceptible patients.Drug-lysosomal interactions are well-recognised in experimen-tal pharmacology’ but more work needs to be done on these in-teractions in mon

W. PERRYMetabolic Unit,Royal National Orthopædic HospitLondon W1N 6AD

M. V. JENKINST. C. B. STAMP

1. Remmer, H., Schoene, B., Fleischmann, R. A. Drug. Metab. Disp 1973, 1,224.

2. Perry, W., Jenkins, M. V., Erooga, M. A., Stamp, T. C. B. Biochem Med1978, 20, 153.

3. Stamp, T. C. B., Flanagan, R. J., Richens, A., Round, J. M., Thomas. M,Jackson, M., Dupré, P., Twigg, C. A. in Endocrinology of Calcium Met-abolism (edited by D H. Copp and R. V Talmage) Amsterdam, 1978.

4. Woollen, J. W., Walker, P G. Clin. chim Acta, 1965, 12, 659.5. Woollen, J. W., Walker, P. G. ibid 1965, 12, 6476. British Medical Journal, 1977, ii, 1043.7. De Duve, C., De Barsy, T., Poole, B., Trouel, A., Tulkens, P , Van Hoof.

F. Biochem. Pharmac. 1974, 23, 2495.