• haemodialysis;
  • pharmacokinetics;
  • quinine


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Aims Quinine is often used to prevent muscle cramps in patients with chronic renal failure. A standard dose of 300 mg at bedtime is usually recommended, but little is known about the pharmacokinetics of quinine in the presence of renal failure.

Methods We studied the pharmacokinetics of quinine in eight normal subjects and eight patients with chronic renal failure on haemodialysis after a single oral dose of quinine sulphate (300 mg).

Results The concentration of α1-acid glycoprotein (AAG), the major binding protein for quinine, was increased in haemodialysis patients compared with control subjects (1.52 g l−1vs 0.63 g l−1[mean difference 1.033; 95% CI 0.735, 1.330]) whereas albumin levels were decreased (30 g l−1vs 40 g l−1[mean difference 9.5; 95% CI 3.048, 15.952]). Accordingly, the free fraction of quinine was decreased (0.024 vs 0.063 [mean difference 0.0380; 95% CI 0.0221, 0.0539]) and the apparent volume of distribution tended to decrease (0.95 l kg−1vs 1.43 l kg−1[mean difference 0.480; 95% CI 0.193, 1.154]). The quinine binding ratio correlated with the plasma concentration of AAG but not that of albumin. The clearance of free (unbound) quinine was increased in haemodialysis patients compared with controls (67.9 ml min−1 kg−1vs 41.1 ml min−1 kg−1[mean difference −26.8; 95% CI, –56.994, 3.469]), and the area under the curve (AUC) of the two main metabolites, 3-hydroxyquinine and 10,11-dihydroxydihydroquinine were increased.

Conclusions In patients with chronic renal failure, there is an increase in plasma protein binding and in the clearance of free drug, resulting in lower plasma concentration of free quinine.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Quinine is often used for the prevention and treatment of muscle cramps. A meta-analysis of randomized clinical trials evaluating its efficacy for the relief of nocturnal leg cramps supports its use for this indication [1]. Quinine and its diastereomer quinidine are thought to act by decreasing the excitability of the myoneuronal plate and by increasing the refractory period of striated muscle [2, 3]. The effect of quinine or quinidine appears to be dose-related: Lee et al.[4] have shown a significant relationship between serum quinidine concentration and attenuation of cramps, and trials using a larger dose of quinine (500 mg) have shown better efficacy than those using a lower dose (200 mg) [5–7].

In patients with chronic renal failure, muscle cramps occur most frequently during haemodialysis sessions and may be relieved by hypertonic saline or dextrose solutions administered intravenously [8]. Cramps may also be prevented by prazosin [9], although clinically significant hypotensive episodes are a major drawback to its regular use. Quinine remains the most commonly used agent to prevent cramps in patients with chronic renal failure. Controlled trials have shown that quinine reduces the frequency and severity of muscle cramps in haemodialysis patients [10, 11]. A standard dose of 300 mg quinine at bedtime is most often used, although not uniformly effective. No data are available on the pharmacokinetics of quinine in patients with severe chronic renal failure. Quinine is a weak base that is highly bound to α1-acid glycoprotein (AAG) in plasma [12–14]. AAG plasma concentration is known to be increased in chronic renal failure [15], and this could result in lower plasma concentration of free (unbound) quinine and reduced ­efficacy. The purpose of the present study was to define the pharmacokinetics of a single dose of quinine in chronic haemodialysis patients as compared with healthy volunteers.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References


Eight patients with chronic renal failure treated by regular haemodialysis and eight healthy volunteers participated in the protocol after providing informed consent. The protocol was approved by the hospital local Ethics Committee. Patients were excluded if they had severe anaemia (Hb < 70 g l−1), severe heart failure (NYHA class III or IV), liver disease (as assessed by clinical evaluation and liver function tests), if they were HIV-positive or if they were receiving quinine or quinidine. None of the patients had a history of gastrointestinal dysmotility. Two patients had diabetes.

On the morning of the first study day, after an 8 h fast, a small indwelling catheter was inserted in a forearm vein, opposite to the haemodialysis vascular access. This catheter was used for blood sampling. Liver enzymes, total bilirubin, creatinine, complete blood count and serum albumin were measured using standard techniques. AAG plasma concentration was measured by nephelometry. Patients and healthy volunteers received quinine sulphate 300 mg (Novopharm, Montréal, Canada) orally and were allowed to eat breakfast 1 h later. Heparinized blood samples for determination of serum concentrations of quinine were obtained before and at 0.5, 1, 2, 3, 4, 6, 8, 12, 24, 30 and 48 h after quinine administration. Samples    were    centrifuged    and    plasma    was    stored    at −70 °C until analysed. At 48 h, immediately before the haemodialysis treatment, each haemodialysis patient received another dose of quinine sulphate 300 mg orally; 1 h later, two pairs of blood samples were drawn from the arterial and venous haemodialysis lines to calculate the extraction of quinine by the haemodialysis filter. The ultrafiltration rate was reduced to minimum (0.1 l h−1) for 5 min before the blood samples were taken, in order to avoid overestimation of the venous concentration of quinine [16].

Determination of quinine and its metabolites in plasma

Quinine was assayed in plasma by high pressure liquid chromatography     (h.p.l.c.)     with     triamterene    µg ml−1 in water as an internal standard. Internal standard (50 µl) was added to 1.0 ml plasma samples in a glass stoppered tube with a Teflon-lined screwcap. Samples were alkalinized with 1.0 ml of NaOH 0.1 n, vortexed for 10 s, and 6 ml of methylene chloride : 2-propanol (8 : 2 v/v) were added. The mixture was shaken for 20 min on an alternative agitator, then centrifuged at 3000 rev min−1 for 10 min. The aqueous layer was discarded, and the organic layer   was   transferred   to   a   clean   tube   and   evaporated to dryness under a stream of nitrogen at 37 °C. The residue   was   dissolved   in   300 µl   of   mobile   phase   and an   aliquot   was   chromatographed.   A   Perkin-Elmer series 3B chromatograph was used, equipped with a Varian 9090 Autosampler. The column was a 3.9 × 300 mm : Bondapack C-18 reverse-phase column (Waters) eluted with a methanol :  acetonitrile : 1% acetic acid solution [2 : 1 : 7]: at a flow rate of 2.0 ml min−1. The effluent was monitored with a Hewlett Packard HP 1046 A fluorescence detector using a 340 nm excitation wavelength and a 425 nm emission wavelength. The detector was coupled to a Hewlett-Packard HP 3396 A integrator.

Retention   times   were   4.6 min    for   the   internal   standard    and    5.8 min    for    quinine.    Two    metabolites    of quinine were also visible on the chromatograms, with retention   times   of   1.9    and   2.4  min.   They   were   identified      using      gas       chromatography-mass       spectrometry      as      10,11-dihydroxydihydroquinine    (10,11-Q)    and    3-hydroxyquinine (3-OH-Q), respectively [17].

Calibration curves were constructed using plasma obtained from the Canadian Red Cross and spiked with increasing concentrations of quinine and metabolites. Purified metabolites were obtained from Dr James M. Cook of the University of Wisconsin. Calibration curves were linear for concentrations ranging from 0 to 10.0 µg ml−1 with correlation coefficients always above 0.97. The coefficient of variation of the assay for quinine was    3.6%    at    0.1    µg ml−1   and    1.7%     at    1.0 µg ml−1 of quinine.

Plasma protein binding

The plasma binding of quinine was measured by equilibrium dialysis at 37 °C with a Teflon cell system (Spectrum) and a Spectapor 2 membrane (molecular weight 12,000–14 000). Quinine was added to duplicate blank plasma samples (1.0 ml) obtained prior to drug administration, at a final concentration of 2.0 µg ml−1, and samples were dialysed against an equal volume of 0.067 m phosphate buffer pH 7.4 for 3 h. Quinine concentration in the plasma and dialysate samples was measured by h.p.l.c. The free fraction (f) of quinine in plasma was calculated as: dialysate concentration/plasma concentration. Preliminary studies indicated that equilibrium was achieved by 1 h and that binding was constant for concentrations ranging from 0.5 to 10.0 µg ml−1 in both healthy volunteers and haemodialysis patients; the variation of the protein binding was less than 5% for this range of concentrations (i.e. protein binding was not concentration dependent). The coefficient of variation for the determination of the free fraction of quinine was 6.35%.

To evaluate the influence of the plasma concentration of AAG and albumin on the binding of quinine, the relationship between the binding ratio (bound/free) and the concentration of AAG and albumin was examined according to the equation [18]:

  • Bound/Free = nKP

where P is the plasma protein concentration of albumin or AAG, K is the affinity constant and n is the number of binding sites.

Pharmacokinetic analysis

The area under the plasma drug concentration-time curve (AUC) was calculated using the trapezoidal rule with extrapolation to infinity. Cmax was defined as highest quinine plasma concentration measured and tmax was defined as the time when Cmax was measured. Oral clearance was calculated as (Dose/AUC) and oral free clearance as (oral clearance/free fraction). The apparent volume of distribution (Vd) was calculated as (CL/Kel).

Statistical analysis

Data are presented as mean ± s.d. Comparisons between patients and controls were done with a nonparametric unpaired test (Mann–Whitney) and the level of significance was set at P = 0.05. The confidence interval on the difference between means was also calculated (see abstract).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

The characteristics of the two groups studied are reported in Table 1. Haemodialysis patients were older than controls, had lower serum albumin and higher serum AAG concentrations. Liver function tests were normal in both groups (data not shown).

Table 1. Characteristics of controls and haemodialysis patients
  Controls (= 8) Haemodialysis (= 8)P value
  1. Data are presented as mean ± s.d.

Gender (male : female)5 : 35 : 3 
Age (years)38 ± 958 ± 130.003
Weight (kg)71 ± 1368 ± 140.798
Serum creatinine (µmol l−1)84 ± 16625 ± 2060.001
Serum albumin (g l−1)40 ± 330 ± 70.022
Serum AAG (g l−1)0.63 ± 0.151.52 ± 0.600.010

The mean plasma concentration vs time curves for quinine and its metabolites in control subjects and haemodialysis patients are shown in Figure 1 and pharmacokinetic parameters are presented in Table 2. Cmax was significantly  higher  in  haemodialysis  patients  whereas tmax was similar in the two groups (1.6 ± 0.7 vs 1.9 ± 0.4 h). The oral total clearance of quinine was decreased by 43% in haemodialysis patients. The clearance by the dialysis apparatus was poor with an extraction of 6.4 ± 5.5%. The volume of distribution of quinine appeared to decrease in haemodialysis patients as compared with controls, but this difference was not significant (P = 0.08).


Figure 1. Mean total plasma concentration vs time curves for quinine (●), 3-hydroxy-quinine (3-OH-Q) (bsl00151), and 10–11-dihydroxy-dihydro-quinine (10–11-Q) (□) in control subjects and haemodialysis patients.

Download figure to PowerPoint

Table 2. Pharmacokinetic parameters of quinine disposition, protein-binding of quinine and metabolism of quinine in control subjects and haemodialysis patients
  Controls Haemodialysis P value
  1. Data are presented as mean ± s.d.

Quinine disposition
C max (µg ml−1)2.60 ± 0.994.56 ± 1.580.004
CLo (ml min−1 kg−1)2.51 ± 1.591.41 ± 0.630.033
V d area (l kg−1)1.43 ± 0.780.95 ± 0.290.080
K el (h−1)0.106 ± 0.0310.090 ± 0.0230.221
t 1/2 (h)7.02 ± 1.728.26 ± 2.100.382
Protein-binding of quinine
Free fraction0.063 ± 0.0140.024 ± 0.0150.001
(Bound/Free) AAG (l g−1) 26.0 ± 6.031.0 ± 13.40.191
(Bound/Free) albumin (l g−1)0.4 ± 0.11.6 ± 0.40.001
Free CLo (ml min−1 kg−1)41.1 ± 24.067.9 ± 32.00.033
Metabolism of quinine
Free fraction0.063 ± .0140.024 ± .0150.001
AUC free quinine 2.11 ± 1.04 1.40 ± .550.080
AUC3-OH-Q 12.9 ± 4.345.1 ± 51.30.083
AUC10−11-Q2.42 ± 0.894.46 ± 1.070.003
Ratio 3-OH-Q/free quinine6.80 ± 2.1246.30 ± 74.380.049
Ratio 10–11-Q/free quinine 1.36 ± .664.12 ± 3.040.002

The free fraction of quinine was markedly decreased in haemodialysis patients (Table 2). The ratio (bound/free)/AAG was similar in the two groups and the correlation between bound/free ratio and AAG concentration was good (r = 0.789, P = 0.001). On the other hand, the ratio (bound/free)/albumin was higher in haemodialysis patients, but the correlation between the ratio and albumin concentration was poor (r = 0.234). Plasma concentrations of free quinine were lower and the oral free clearance was increased in haemodialysis patients compared with control subjects (Figure 2 and Table 2). The relationship between the plasma concentration of quinine and its metabolites is summarized in Table 2. The AUC for free quinine was significantly lower, and the ratio of AUC10,11-Q/AUCfree quinine and AUC3OH-Q/AUCfree quinine were significantly increased in haemodialysis patients compared with control subjects.


Figure 2. Plasma concentration (mean ± s.d.) vs time curves for free quinine in control subjects (●) and haemodialysis patients (bsl00151).

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Quinine pharmacokinetics has not been studied previously in patients with severe renal failure, although this drug is frequently used to prevent and treat muscle cramps associated with dialysis. In the present study, a group of haemodialysis patients was compared with healthy volunteers.

The absorption phase seemed altered in haemodialysis patients. Quinine is a weak base (pKa 8.4) and it is best absorbed when the gastric pH is high. In a more alkaline solution, the ratio of unionized to ionized quinine will be higher, promoting absorbtion. In patients with severe renal failure, the gastric pH may increase due to the presence of ammonia, derived from urea present in swallowed saliva and cleaved by gastric urease [19]. This could explain an increased absorption of quinine and a lower absorption of acidic drugs in haemodialysis patients. Other causes of altered absorption in severe renal failure include drug interactions, oedematous state and gastroparesis [20], none of which was present in our patients. An alternative explanation for the increased Cmax is the smaller apparent volume of distribution in haemodialysis patients.

In haemodialysis patients, there was an important decrease in the free fraction of quinine which was related to an increased serum concentration of AAG. This acute-phase reactant protein is known to be increased in chronic renal failure [15]. Quinine is mostly bound to AAG and much less to albumin [12–14]. Uraemia may alter drug-binding to albumin not only by a decrease in serum albumin concentration, but also by competitive displacement of the drug from its binding sites by accumulated substances, or by a change in the structural configuration of the binding sites [21]. Drug binding to AAG however, appears to depend almost exclusively on its serum concentration in uraemic patients [22]. In the present study, we observed a good correlation between quinine binding and the plasma concentration of AAG but not that of albumin, in both normal and haemodialysis subjects. Thus, quinine binding to AAG depends mostly on the binding protein concentration whereas quinine binding to albumin may be influenced by factors other than serum albumin concentration.

Quinine is eliminated almost exclusively by hepatic metabolism. Since quinine has a low hepatic extraction ratio, the binding is rate limiting for its hepatic uptake. Thus free (unbound) drug is available for hepatic uptake and metabolism, and free clearance is an appropriate measure of its rate of elimination. In haemodialysis patients, the clearance of free quinine was increased compared with controls. The elimination of other drugs such as propranolol and phenytoin has also been shown to be increased in uraemic patients [23]. The mechanism responsible for this increased hepatic elimination remains unclear. Quinine is metabolized by oxidation of the quinoline or the quinuclidine moieties into four primary metabolites and at least 11 secondary metabolites [17]. Cytochrome P4503A catalyses the formation of 3-OH-quinine [24], but the enzymes responsible for the formation of the other metabolites have not been identified. In our h.p.l.c. assay, two of the primary metabolites of quinine could be identified and quantified. There was an increase in the plasma concentration of both 3-OH-quinine and 10,11-dihydroxydihydroquinine in haemodialysis patient which could be related to increased formation of the metabolites and/or decreased renal elimination. None of the patients included in the study was receiving drugs known to induce cytochrome P450, but enzyme induction has been postulated in uraemic patients due to external agents such as plasticizers in haemodialysis tubing [25].

In summary, two major changes were seen in haemodialysis patients. The first relates to the binding of quinine, which is increased because of elevated serum concentrations of AAG. The second is an increase in the clearance of free drug, resulting in lower plasma concentrations of free quinine. The efficacy of quinine to prevent muscular cramps appears to be dose-related, and our data show that patients with severe chronic renal failure may have a lower free concentration of quinine; further studies are needed to confirm that patients who do not experience an improvement in the severity or frequency of muscular cramps may benefit from a larger dose.

We thank Mrs Ginette Raymond for technical assistance and Manon Bourcier for preparing the manuscript. Dr L. Roy was supported by the Fonds de la Recherche en Santé du Québec.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  • 1
    Man-Son-Hing M, Wells G. Metaanalysis of efficacy of quinine for treatment of nocturnal leg cramps in elderly people. Br Med J 1995; 310: 1317.
  • 2
    Quinine for night cramps. Med Letter Drugs Therapeutics 1986; 28: 110.
  • 3
    Pray S. Nocturnal leg cramps: the dangers of quinine therapy. US Pharm 1992; 17: 1720.62.
  • 4
    Lee FY, Lee SD, Tsai YT, et al. A randomized controlled trial of quinidine in the treatment of cirrhotic patients with muscle cramps. J Hepatol 1991; 12: 240.
  • 5
    Connolly PS, Shirley EA, Wasson JH, Nierenberg DW. Treatment of nocturnal leg cramps. A cross-over trial of quinine vs vitamin E. Arch Int Med 1992; 152: 18771880.
  • 6
    Sidorov J. Quinine sulfate for leg cramps: does it work? J Am Geriat Soc 1993; 41: 498500.
  • 7
    Warbutron A, Royston JP, O'Neill CJA, et al. A quinine a day keeps the leg cramps away? Br J Clin Pharmacol 1987; 23: 459465.
  • 8
    Blagg CR. Acute complications associated with hemodialysis. In Replacement of Renal Function by Dialysis ed. Maher JF. Dornrecht: Kluwer Academic Publishers, 1989.
  • 9
    Sidhom OA, Odeh YK, Krumlovsky FA, et al. Low-dose prazosin in patients with muscle cramps during hemodialysis. Clin Pharmacol Ther 1994; 56: 445451.
  • 10
    Kaji DM, Ackad A, Nottage WG, Stein RM. Prevention of muscle cramps in hemodialysis patients by quinine sulphate. Lancet 1976; ii: 6667.
  • 11
    Roca AO, Jarjourei D, Blend D, et al. Dialysis leg cramps. Efficacy of quinine versus vitamin E. Am Soc Artificial Internal Organs Journal 1992; 38: M481M485.
  • 12
    Wanwimolruk S, Denton JR. Plasma protein binding of quinine; binding to human serum albumin, α1-acid glycoprotein and plasma from patients with malaria. J Pharm Pharmacol 1992; 44: 806811.
  • 13
    Mihaly GW, Ching MS, Klejn MB, Paull J, Smallwood RA. Differences in the binding of quinine and quinidine to plasma proteins. Br J Clin Pharmacol 1987; 24: 769774.
  • 14
    Silamut K, Molunto P, Ho M, White NJ. α1 -acid glycoprotein (orosomucoid) and plasma protein binding of quinine in falciparum malaria. Br J Clin Pharmacol 1991; 32: 311315.
  • 15
    Pacifici GM, Viani A, Taddeucci-Brunelli G, Rizzo G, Carrai M, Schulz HU. Effects of development, aging, and renal and hepatic insufficiency as well as hemodialysis on the plasma concentrations of albumin and α1-acid glycoprotein: implications for binding of drugs. Ther Drug Monit 1986; 8: 259263.
  • 16
    Gibson TP. Problems in designing hemodialysis drug studies. Pharmacotherapy 1985; 5: 2329.
  • 17
    Bannon P, Yu P, Cook J, Roy LF, Villeneuve JP. The biotransformation of orally administered quinine in man. J Chromatogr B 1998; 715: 387393.
  • 18
    Kremer JMH, Wilting J, Janssen LHM. Drug binding to human alpha-1-acid glycoprotein in health and disease. Pharmacol Rev 1988; 40: 147.
  • 19
    Aronoff GR, Abel SR. Principles of administering drugs to patients with renal failure. Contemporary Issues Nephrol 1987; 17: 119.
  • 20
    Reed WE, Sabatini S. The use of drugs in renal failure. Seminars Nephrol 1986; 6: 259295.
  • 21
    Gulyassy PF, Depner TA. Impaired binding of drugs and endogenous ligands in renal diseases. Am J Kidney Dis 1983; 2: 578601.
  • 22
    Grossman SH, Davis D, Kitchell BB, Shand DG, Routledge PA. Diazepam and lidocaine plasma protein binding in renal disease. Clin Pharmacol Ther 1982; 31: 350357.
  • 23
    Reidenberg MM. The biotransformation of drugs in renal failure. Am J Med 1977; 62: 482485.
  • 24
    Guengerich FP. Human cytochrome P450 enzymes. In Cytochrome P450, ed. Ortiz de Mortellano PR. New York: Plenum Press, 1995: 473535.
  • 25
    Leber HW, Gleumes L, Schufferle G. Enzyme induction in the uremic liver. Kidney Int 1978; 13(Suppl 8): S43S48.