Association Between Long-term Quinine Exposure and All-Cause Mortality

Quinine has been used since the 1930s to treat idiopathic muscular cramps. However, in 2006, because of efficacy and safety issues, the US Food and Drug Administration cautioned about this off-label use of quinine, citing “665 reports of adverse events with serious outcomes…including 93 deaths.”¹ Despite warnings, quinine is still prescribed to individuals with idiopathic muscular cramps. Furthermore, many drinks such as bitter lemon or tonic waters contain quinine, and hence, many may be exposed to quinine daily. This study explored the association between long-term quinine exposure and all-cause mortality.

This study used data recorded in The Health Improvement Network (THIN), a UK primary care database containing anonymized data on more than 12 million individuals representative of the UK population.² The protocol for THIN to obtain and provide anonymous patient data was approved by a national ethics committee in 2002; need for informed consent was waived. Adults who received incident quinine salt (sulfate, bisulfate, dihydrochloride) prescriptions for idiopathic muscular cramps or restless leg syndrome for at least 1 year from January 1990 to December 2014 (last follow-up December 2015) at a mean dosage of 100 mg/d or more were considered to be exposed. This definition was chosen because the period of risk is unknown and most patients stopped quinine within weeks of death. The start date (ie, start of at-risk period) was defined as the first day of the first prescription of quinine. Individuals with muscular cramps or restless leg syndrome, never exposed to quinine or its derivatives, and with at least 1-year follow-up after a randomly selected start date were eligible to be included in the unexposed sample. Three unexposed individuals were selected for every exposed individual. The samples were stratified by sex and age. Group characteristics were compared using the χ² test or the Wilcoxon rank-sum test. The primary outcome (all-cause mortality) was compared between the exposed and the unexposed populations using Cox proportional hazards models adjusted for sociodemographic data, underlying conditions, and concomitant prescriptions. Post hoc subgroup analyses were conducted by age and amount of exposure (averaged over the exposure period). The proportional hazards assumption for Cox models was checked graphically using the Schoenfeld residuals. Analyses were done using Stata (StataCorp), version 14.0. A 2-sided P value less than .05 denoted statistical significance.

The study population included 175 195 individuals (median age, 70 years [interquartile range, 61-78]; women, 57.8%; median follow-up, 5.7 years [interquartile range, 3.1-8.9]) (Table 1). Exposed persons received a median 203 mg/d (interquartile range, 163-252) of quinine. There were 11 598 deaths (4.2 per 100 person-years) among the 44 699 exposed individuals vs 26 753 (3.2 per 100 person-years) among the 130 496 unexposed individuals (adjusted hazard ratio [HR], 1.24 [95% CI, 1.21-1.27]). The increase in the risk of death was more pronounced in those younger than 50 years (adjusted HR, 3.06 [95% CI, 2.51-3.73]), whatever the indication for prescription (Table 2). A dose-effect was found for exposure of 200 to 299 mg/d (adjusted HR, 1.25 [95% CI, 1.20-1.30]), 300 to 399 mg/d (adjusted HR, 1.83 [95% CI, 1.72-1.94]), and 400 mg/d or more (adjusted HR, 2.24 [95% CI, 1.95-2.58]) compared with less than 200 mg/d (P value for trend, <.001).

In this study, long-term quinine exposure was associated with increased mortality. Individuals in this study received more than 100 mg/d of quinine, equivalent to a daily consumption of more than 1 L of bitter lemon or tonic waters.³

Quinidine, an isomer of quinine, increases by 3-fold the risk of death (mainly sudden cardiac arrests) compared with placebo or no treatment.⁴ Short-term quinine exposure may lead to life-threatening adverse events, such as thrombocytopenia, hypoglycemia, or cardiac arrhythmia.⁵ In a study, the incidence rate ratio of death in individuals with heart failure initiating quinine for leg cramps was 1.19 (95% CI, 1.14-1.24) compared with unexposed patients.⁶ However, mortality of long-term users has not been studied.

Limitations include that unmeasured confounders rather than exposure to quinine may predispose patients to higher mortality. No data were available regarding quinine beverage consumption. A competing risk may explain the lower increased risk in the elderly. Because death certificates were not accessible, the mortality causes are unknown.

The benefits of quinine in reducing cramps should be balanced against the risks.

Accepted for Publication: February 17, 2017.

Corresponding Author: Laurence Fardet, MD, PhD, Hôpital Henri Mondor, 51 Avenue du Maréchal de Lattre de Tassigny, 94010 Créteil, France (laurence.fardet@aphp.fr).

Author Contributions: Dr Fardet had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Fardet.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Fardet.

Administrative, technical, or material support: Nazareth.

Supervision: Nazareth, Petersen.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.