TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE (1999) 93,69-72 69

Pharmacokinetics of quinine and 3-hydroxyquinine in severe falciparum malaria with acute renal failure

Paul Newton1p2, Duangsuda Keeratithaku13, Paktiya Teja-Isavadharm3., Sasithon Pukrittayakamee’ , Dennis Kyle’ and Nicholas White”* ‘Faculty of Tropical Medicine, Mahidal Untversity, Bangkok, Thailand; ‘Centre for Tropical Diseases, Nufield Department of Clinical Medicine, University of Oxford, UK; 3Department of Immunology and Parasitology, US Army Medical Component, Armed Forces Research Institute for Medical Science (AFRIMS>, Bangkok, Thailand

Abstract The plasma concentrations of quinine and its main metabolite, 3-hydroxyquinine (30HQn), were measured in 5 adult Thai patients with severe Plasmodium falciparum malaria and acute renal failure. Two patients required peritoneal dialysis but all survived. During acute renal failure plasma concentrations of 30HQn rose to reach up to 45% of the levels of the parent compound. The estimated median (range) quinine clearance was 0.83 mL/kg/ min (0.58- l-16), and 30HQn clearance/fraction of quinine converted was 3.40 ml/kg/mm (2.58-4.47). The estimated 30HQn terminal elimination half-life was 2 1 h (16.5- 32.5). These data suggest that 30HQn contributes about 12% of the antimalarial activity of the parent compound in patients with falciparum malaria and acute renal failure. It is also likely that 30HQn contributes to adverse effects, although this metabolite is not quantitated routinely by current high- performance liquid chromatography quinine assays.

Keywords: malaria, Plasmodiumfuleipamm, renal failure, chemotherapy, quinine, 3-hydroxyquinine, pharmacokinetics, Thailand

Introduction Acute renal failure is a common and serious manifes-

tation of severe Plasmodium falciparwn malaria in adults. Renal failure results from acute tubular necrosis. Qui- nine is still the most widely used treatment for severe malaria in the world and in recent years dose regimens have been modified on the basis of pharmacokinetic information (WI-IO, 1990; WHITE, 1992; KRISHNA & WHITE, 1996). A loading dose of 20 mg of the dihy- drochloride salt per kg is given initially and this is followed by 10 mg salt/kg every 8- 12 h (WHITE et al., 1983). It is generally recommended that the dose of quinine should be reduced in severe malaria by one-third to one-half after 48 h if there has been no significant improvement in the patient’s condition (WHITE et al., 1982; WHO, 1990). The principal route of quinine clearance is by hepatic biotransformation, first to 3- hydroxyquinine (30HQn), and then to more polar and less bioactive metabolites which are eliminated in the urine (PUKRITTAYAKAME E et al., 1997). About 20% of administered quinine is excreted unchanged in the urine, but renal clearance would be expected to be proportio- nately more important for the hydroxylated metabolites. 30HQn has antimalarial activity (NONTPRASERT et al., 1996) and is likely to have other pharmacodynamic properties that are similar to those of quinine. 30HQn has not been measured previously in patients with severe malaria.

Methods Patients with severe falciparum malaria were admitted

to Paholpolpayuhasensa Hospital, Kanchanaburi, Thai- land. On admission a clinical assessment was made, a malaria blood smear was taken, and urea and electro- lytes, creatinine, glucose, bilirubin and serum glutamic- oxaloacetic transaminase (SGOT) were measured. An intravenous loading dose‘ of quinine dihydrochloride (20 mg salt/kg: Government Pharmaceutical Organisa- tion of Thailand) was infused over 4 h and this was followed by 10 mg salt/kg infused over 2 h at 8-h intervals. The dose interval was increased to 12 h after 48 h. The parasite counts and haematocrit were recorded 6-hourly until parasite clearance time (defined as time to first negative malaria thick film). In 5 patients with

Address for correspondence: Professor Nicholas J. White, Faculty of Tropical Medicine, 420/6 Rajvithi Rd, Bangkok 10400, Thailand; phone + 66 2 246 0832, fax + 66 2 246 7795, e-mail fnnjw@diamond.mahidoI.ac.th

elevated serum creatinine, venous blood samples were taken before and after the first quinine infusion of each day after obtaining fully informed consent from the natient or attendant relatives. All blood samples were centrifuged immediately and the plasma was-stored at -70°C until analvsis. These studies were hart of a series of investigations< on severe malaria apbroved by the Ethical Review Sub-committee of the Thai Ministry of Public Health.

Quinine assay Quinine and 30HQn were measured by high-

performance liquid chromatography (HPLC) with fluorescence detection (EDSTEIN et al., 1990; PLXIUT- TAYAKAMEE et al., 1997). 30HQn for HPLC calibration was synthesized as described pre%ously (NONTPRASERT et al.. 1996). The HPLC internal standard was cincho- nine hydrochloride. The inter-assay coefficients of varia- tion for samples of known concentration were 2.6% (15.4 mg/L) and 5.2% (0.24 mg/L) for quinine and 4.6% (9.2 mg/L) and 8.9% (0.075 mg/L) for 3OHQn.

Pharmacokinetic estimates Plasma concentration-time profiles were plotted and

estimates obtained of transient steady-state total plasma quinine and plasma 30HQn levels. Quinine clearance (Clo) and 30HQn clearance (C13ou~~) were estimated from the steady-state equations:

ClQ = dOSeQ/PQ,,

C130HQn/f= (ClQ x PQss)/P30HQnss

Where Clq = quinine clearance, dOSeQ = dose of qui- nme base mfused, Pass = quinine plasma concentration at steady State, P3OH

is== 30HQn plasma concentra-

tion at steady state, 130HQn = 30HQn clearance and f = proportion of quinine biotransformed to 30HQn.

The steady-state values were the estimated average concentrations during the plateau phase before recovery or dialysis. In these patients with acute renal failure, the calculation of 3OHQn clearance assumes that no tm- changed quinine is eliminated in urine or bile during the oliguric phase. 30HQn first-order elimination rate con- stant in the period of acute renal failure was estimated from the accumulation profile assuming constant input throughout this period. Quinine and 30HQn ‘area under the curve’ (AUC) were estimated by the trapezoi- dal method with extrapolation to infinity. The total apparent volume of distribution was calculated as clear- ance/first-order elimination rate constant.

70 PAUL NEWTON ET/X..

Results Five adult Thai patients (3 male, 2 female) were

studied (Table 1). None had known previous renal disease. All had severe malaria (WHO, 1990) with acute oliguric renal failure and impaired level of consciousness, and 2 had convulsions. Two patients required perito- neal dialysis but all patients survived with near-normal renal function (serum creatinine cl86 ltrnol/L) at discharge.

fraction ofthe drug that is converted to other metabolites is not known.

In this small series of patients with acute oliguric renal failure in severe malaria, the plasma concentration of 30HQn rose to as high as 45% of that of the parent compound (median 24%, range 19-45%). This accu- mulation of 30HQn will not usually be detected as the metabolite is not routinely quantitated in current HPLC assays. We have shown previously that free 30HQn has

Table 1. Admission clinical and laboratory findings

Variable

Age (years) Weight (kg) Haematocrit (%) Parasite count (/liL) Urea (mmol/L) Creatinine (prnol/L) Total bilirubin @rnol/L) SGOTC (III/L) Parasite clearance time (h)

Median=

25 (18-60) 48 (47.0-53.5) 29 (12.0-33.5)

67 324b (14 300- 1026 340) 19.8 (13.7-44.6) 415 (167-769)

73 (41-219) 86 (68-l 50) 77 (38- 110)

a Range in parentheses. b Geometric mean. ’ Serum glutamic-oxaloacetic transaminase.

The variations in plasma quinine and 3OHQn con- centrations with time for individual patients are shown in Figures 1A and 1B. The estimated clearances of quinine and 30HQn and 30HQn half-life are shown in Table 2. Median (range) estimated value for quinine clearance was 0.83 (0.58-1.16) mL/kg/min, for 30HQn clear- ance/fraction biotransformed was 3.40 (2.58-4.47) mL/kg/min and for 30HQn half-life was 21 h (16.5- 325), respectively. The median (range) ratio of quinine to 30HQn clearance was 0.24 (0.18-0.45). The median (range) total apparent volume of distribution (Vd/f) for 30HQn was 7.1 L/kg (3.7-9.6). The median AUCO-.u was 2678 mgL-’ h for quinine and 392 mgLP1 h for 30HQn, with a median quinine/30HQn AUCoPm ratio of 5.54 (Table 2). There were sufficient data to calculate 30HQn accumulation rate constants for patients 1 and 2 which were 0.0 11 and 0.021 /h, respectively.

about 10% of the antimalarial activity of the parent compound in vitro (NONTPRASERT et al., 1996). How- ever, 30HQn is only about 50% protein bound, in comparison to over 90% binding of quinine in severe malaria (PUKRITTAYAKAMEE et aZ., 1997). The data presented here suggest median free plasma concentra- tions of quinine and 30HQn of about 1.4 mg/L and 1.7 mg/L, respectively. Therefore, 30HQn probably contributed an average of 12% and up to 23% of the antimalarial activity of the parent compound and, as these are additive (NONTPRASERT et aZ., 1996), total antimalarial activity was under-estimated by approxi- mately 11% .

Whether 3OHQn has the same cardiovascular, meta- bolic and neurological adverse-effect profile as the parent compound is not known. The diastereomer of 30HQn (3-hydroxyquinidine) has similar pharmacodynamic

Table 2. Quinine and 3-hydroxyquinine pharmacokinetic parameters

Steady-state Clearancea Elimination AUCe-, mgLP1 h concentration (mg/L) (mL/kg/min) half-life (/h) Vd/f

Patient Q/30HQn 30HQn no. Quinine 30HQn Quinine 30HQn 30HQn Quinine 30HQn ratio Wkg)

1 20 3.4 0.58 2.97 24.0 2776 479 5.79 7.09 2 10 4.5 1.16 2.58 16.5 1066 392 2.72 3.68 3 14 3.4 0.83 3.42 32.5 2837 512 5.54 9.62 4 14 2.6 0.83 4.47 19.0 2678 381 7.02 7.35 5 13 3.4 0.89 3.40 21.0 1306 349 3.74 6.19

Median 14 3.4 0.83 3.40 21.0 2678 392 5.54 7.09

aFor 30HQn this refers to Cl/f.

Discussion These data indicate that SOHQn, the principal meta-

bolite of quinine (NONTPRASERT et al., 1996), accumu- lates in acute renal failure. In patients with normal renal function about 20% of a quinine dose is excreted un- changed in the urine. The remainder is metabolized by liver enzymes, predominantly by CYP3A4 (ZHAO et al., 1996), to hydroxylated, more polar, metabolites allowing renal clearance. In humans 30HQn is the principal metabolite (PUKRITTAYAKAhJE E et al., 1997). Both metabolic and renal clearance of quinine are reduced in malaria in proportion to disease severity (WHITE et al., 1982). Obviously, in renal failure nearly all quinine clearance is by hepatic biotransformation, although the

properties to those of its parent compound, quinidine, and it is therefore likely that 30HQn does contribute to potential adverse effects with quinine. Although there is considerable inter-individual variance, free 3-hydroxy- quinidine has about 25% of the potency of the parent quinidine in terms of QT, prolongation (VOZEH et aZ., 1987). These effects are additive. If the same obtains for quinine and 30HQn then about 25% of the cardiac effects of quinine in renal failure result from its principal metabolite. There is therefore a potential for under- estimating potential quinine toxicity, particularly in pa- tients with protracted acute renal failure, if plasma quinine is measured by HPLC and 30HQn is not quantitated. Whether other hydroxylated metabolites,

3-HYDROXYQUININE IN ACUTE RENAL FAILURE 71

25 A

100 125 150 175 200 225

Time (h)

B

0 25 50 75 100 125

Time (h)

150 175 200 225

Fig. 1. Plasma quinine (A) and 3-hydroxyquinine (B) concentrations in the 5 patients with acute renal failure and severe malaria (+, patient no. 1; n , patient no. 2; X, patient no. 3; A, patient no. 4; 0, patient no. 5). Median values (0) are also shown.

such as 2-hydroxyquinine, also contribute significantly to antimalarial activity or toxicity in acute renal failure is not known. This would be resolved by a comparison of bio-activity with HPLC measurements of quinine and 3OHQn to determine whether there is additional un- explained antimalarial activity. Interestingly the earlier less specific extraction-fluorescence method for quinine measurement, which was used when current dose regi- mens were being devised (WHITE et aZ., 1982), does not distinguish between quinine and its hydroxylated meta- bolites.

Thus, although 3OHQn contributes relatively little to antimalarial activity or toxicity in the acute phase of the disease because of impaired liver function, in those patients who develop acute renal failure and whose liver function recovers faster than renal function there is steady accumulation of the more polar metabolite. Where possible 30HQn should be measured in addition to quinine in patients with acute renal failure particularly if toxicity is suspected.

Acknowledgements We are very grateful to the Director of Paholpolpayuhasensa

Hospital and his staff for their assistance with this study, and in particular the nurses of the Intensive Care Unit. Quinine assays were supported by the US Army Medical Research and Material Command. This study was part of the Wellcome-Mahidol University-Oxford Tropical Medicine Research Programme funded by the Wellcome Trust of Great Britain.

References Edstein, M. D., Prasitthipayong, A., Sabchareon, A., Chong-

suphajaisiddhi, T. & Webster, H. K. (1990). Simultaneous measurement of quinine and quinidine in human plasma, whole blood, and ecythrocytes by high-performance liquid chromatography with fluorescence detection. Therapeuh~ DrugMonitoring, 12,493-500.

Krishna, S. & White, N. J. (1996) Pharmacokinerics of quinine, chloroquine and amodiaquine. Clinical implications. ClinicuE Pharmacokinetics, 35 263-299.

Nontpraseq, A., Puknttayakamee, S., Kyle, D. E., Vanijanonta, S. & Wlute, N. J. (1996). Antimalarial activity and inter- actions between quinine, dihydroquinine and S-hydroxyqui-

72 PAUL NEWTON ETA/L.

nine against Plasmodium falcipantm in vitro. Transactions of the Royal Society of TropicalMedicine and Hygiene, 90, 553-555.

Pukrittayakamee, S., Looareesuwan, S., Keeratithaknl, D., Davis, T. M., Teja-Isavadharm, B., Nagachinta, B., Weber, A., Smith, A., Kyle, D. & White, N. J. (1997). A study of the factors affecting the metabolic clearance of quinine in malar- ia. EuropeanJournal of Clinical Pharmacology, 52, 487-493.

Vozeh, S., Bindschedler, M., Ha, H. R., Kaufrnarm, G., Guentert, T. W. & Follath, F. (1987). I’harmacodynamics of 3-hydroxyquinidine alone and in combination with quini- dine in healthy persons. Amen’can Journal of Cardiology, 59, 681-684.

White, N. J. (1992). Antimalarial pharmacokinetics and treat- ment regimens. British Journalof ClinicalPhannacology, 34,1- 10.

White, N. J., Looareesuwan, S., Warrell, D. A., Warrell, M. J., Bunnag, D. & Harinasuta, T. (1982). Quinine pharmacoki- netics and toxicity in cerebral and uncomplicated falciparnm malaria. American Journal of Medicine, 73, 564-572.

White, N. J., Looareesuwan, S., Warrell, D. A., Warrell, M. J., Chanthavanich, I’., Bnnnag, D. & Harinasuta, T. (1983). Quinine loading dose in cerebral malaria. AmericanJournal of Tropical Medicine and Hygiene, 32, 1-5.

WHO (1990). Severe and complicated malaria, 2nd edition. Transactions of the Royal Society of Tropical Medicine and Hygiene, 84, snpplement 2. - -

Zhao. X. 1.. Yokovama. H.. Chiba. K.. Wanwimolrnk, S. & Ish*&aki; -T. (1996). Identification of human cytochrome I’450 isoforms involved in the 3-hydroxylation of quinine by human live microsomes and nine recombinant human cyto- chromes P450. Journal of Pharmacology and Experimental Therapeutics, 279, 1327-1334.

Received 20 August 1998; acceptedforpublication 7 October 1998

Treatment of Tuberculosis: Guidelines for Na- tional Programmes (2nd edition). D. Maher, I’. Chaulet, S. Spinaci & A. Harries (writing committee). Geneva: World Health Organization, 1997. 78pp. Price L7.50. WHO/TB/97.220.

Between 1984 and 199 1, New York-hardly a devel- oping country+xpet-ienced a significant outbreak of tuberculosis, due in part to reduced funding for tuber- culosis control, in part to HIV, and in part to an increas- ing division between the rich and the poor: if such an outbreak can occur in the world’s richest country, it is hardly surprising that tuberculosis is out of control in the developing world, where the same contributing factors are even more pronounced. The World Health Organi- zation has adopted a new strategy for the treatment of tuberculosis based on DOTS (directly observed treat- ment, short course). Poor compliance is associated with a poor clinical outcome and increases the chances of the emergence of both secondary, and subsequently pri- mary, drug resistance. Already in some developing countries resistance to at least 1 front-line antitubercu- losis drug is found in over 10% of new isolates. If self administration of medication is not possible, then DOTS is the only alternative, and this booklet, in only 77 clear and well set out pages, describes in detail the whys and hows of DOTS. There are different strategies for differ- ent cultural settings-rural nomads clearly require a different form of observation from the urban poor- and descriptions of how to develop new concepts of direct observation. What to do if the patient interrupts the treatment, or in special cases (pregnancy, liver disease, renal disease, etc.), is briefly and succinctly described. The impact of HIV on tuberculosis, detailed in the WHO manual TBIHIV: a Clinical Manual 1996, is outlined. The essentials of drug supply, including quality control and financial management, are considered: it is always easy for the clinician to blame someone else when drugs fail to appear, yet drug procurement is a complex web of interlocking issues, and clinicians need to under- stand this.

This manual is a useful second edition which updates the first edition published in 1993, with great emphasis on DOTS.

Hugo R&e AIDS Medical Unit I? 0. Box 239 Albert Street Brisbane QLD 4002, Australia

Tuberculosis Control in Refugee Situations: an Inter-Agency Field Manual. Geneva: World Health Organization, 1997. 72pp. Price not known. WHO/TB/97.221.

It is estimated that there are over 26 million refugees, or other persons of concern to the United Nations High Commission for Refugees in the world today, and that number is likely to increase; also increasing is the difficulty in relocating many refugees, such that long- term residence in camps and other temporary refuges will be the outcome for many displaced persons. Tubercu- losis is not a priority in an acute situation, but control services should be developed once the basic daily needs of the population have been met, and once the acute mortality from diarrhoeal disease and acute-respiratory infections has been reduced. The key steps in planning a tuberculosis control programme in such conditions start with the assessment that tuberculosis is an important health problem: once that is established, the agency or agencies responsible for the refugee population have to assess their own capacity to develop and implement a control programme. There have to be sufficient funds for the training of health care workers, for administrators to set up systems to ensure regular and on-going drug supply, for the establishment-on or off site+f sputum microscopy services, and last but not least for the man- agement of the paperwork which is such an integral component of any control programme. Once the funds are ear-marked, then the agency can define its objectives, strategies, and policies, and prepare its work plans, budgets and personnel. If all these are adequately defined, and the money (and will) are in place, then the programme follows the usual control methodologies: clinical suspicion, sputum microscopy (with the tradi- tional algorithms depending on the result), and treat- ment with the emphasis on DOTS (directly observed treatment, short course). Among the detailed instruc- tions are interspersed useful vignettes that highlight some of the special problems that may arise. For example, in Goma (Zaire), in 1995, a control programme covering a population of some 750 000 people recruited 357 new, smear-positive tuberculosis patients: of these, 7 1% com- pleted their treatment, 13% died, 8% defaulted, and 6% transferred out (the manual includes a useful page on managing these); the high mortality rate was attributed to the high prevalence of HIV infection.

Refugees are a grossly disadvantaged population: in dry clear language, this booklet attempts to deal with one of the multitudinous problems from which minority groups always suffer.

Hugo RBe AIDS Medical Unit I? 0. Box 239 Albert Street