Leg Cramps! Are You Part of the Problem?

Abstract & commentary

By Allan J. Wilke, MD, Professor, Department of Introduction to Clinical Medicine, Ross University School of Medicine, Commonwealth of Dominica. Dr. Wilke reports no financial relationships relevant to this field of study.

Synopsis: Nocturnal leg cramps are frequently preceded by the prescription of diuretics, statins, and inhaled long-acting ß2-agonists.

Source: Garrison SR, et al. Nocturnal leg cramps and prescription use that precedes them: A sequence symmetry analysis. Arch Intern Med 2012;172:120-126.

These investigators from the university of british columbia in vancouver used population-level data to study the anecdotal associations of the use of certain medications to the development of nocturnal leg cramps (NLC). The medications in question were diuretics,1 statins,2 and inhaled long-acting ß2-agonists3 (LABAs). They used a statistical tool called sequence symmetry analysis (SSA) to compare whether new prescriptions for quinine (which, despite recommendations to the contrary,4 are commonly prescribed for NLC) increased in the year following a first-time prescription for a diuretic, a statin, or an LABA. They broke down diuretics into potassium-sparing, thiazide-like, and loop, and LABAs into LABAs alone and LABA-corticosteroid combinations. They also compared the rates of use of inhaled anticholinergics and beta-blockers as negative controls. These two medications have similar indications as LABAs and diuretics, respectively.

They linked a province-wide pharmacy database to a British Columbia Ministry of Health physician services and hospitalization database and registration and vital statistics data. The population under review numbered 4.2 million. It was 62.5% female with a median age of 69 years. They excluded patients on dialysis and those with a diagnosis of malaria or amyotrophic lateral sclerosis, who may develop cramps secondary to their illnesses and be prescribed quinine.

Adjusted sequence ratios (95% confidence intervals) for the three drug classes were 1.47 (1.33-1.63) for diuretics, 1.16 (1.04-1.29) for statins, and 2.42 (2.02-2.89) for LABAs. For diuretic subclasses, adjusted sequence ratios were 2.12 (1.61-2.78) for potassium sparing, 1.48 (1.29-1.68) for thiazide-like, and 1.20 (1.00-1.44) for loop. For LABA subclasses, adjusted sequence ratios were 2.17 (1.56-3.02) for LABAs alone and 2.55 (2.06-3.12) for LABAs-corticosteroids. All of these, except for loop diuretics, were statistically significant. The negative controls (inhaled anticholinergics and beta-blockers) did not demonstrate an association with quinine prescription.


This is the first time I've run across SSA, but it seems like a very good way to identify possible medication side effects. Of course, it requires a comprehensive drug database, covering a very large population. Large health organizations (think Group Health) and retail pharmacy chains are already doing it. Wouldn't it be wonderful if we had access to state, regional, or national databases so that we could do this kind of study? The authors offer this conceptualization of SSA: "the rate of events in exposed individuals compared with what would be expected for a similar unexposed population...This is essentially a relative risk...."

The conclusions of this study are surprising and beg to be confirmed with randomized controlled trials (RCTs). While we tend to think of statin use when a patient presents with myalgias, we do not generally associate statins with NLC. The greater association of potassium-sparing diuretics compared to thiazide-like (2.12 vs 1.48) also goes against conventional wisdom, as we are inclined to ascribe muscle cramps to a potassium deficit. We don't generally think of LABAs having an effect beyond the local one of bronchial smooth muscle relaxation, but this study indicates a systemic effect. The authors propose that LABAs could cause leg cramps via stimulation of ß2-adrenergic receptors on peripheral nerves. Apparently, the addition of corticosteroids to LABAs does not attenuate the risk.

The strengths of this study are its huge population and its unique design. British Columbia (at least Vancouver) is ethnically diverse, so it is likely that these results are generalizable to a U.S. population.

Our dictum to do no harm is difficult to implement if we don't know that we are doing it. SSA is a tool to help us tease out when we may inadvertently cause harm. Specific to this study, if your patient complains of NLC, look back at his prescription drug use to see if he is using one of the offending medications. Then discuss with the patient whether continuing the medication is in his best interest or if there is a medication not associated with NLC that could be substituted. And, certainly, do not prescribe quinine to treat the cramps! I'm looking forward to more SSA studies and the RCTs that confirm or refute their findings.


1. Mosenkis A, Townsend RR. Muscle cramps and diuretic therapy. J Clin Hypertens (Greenwich) 2005;7:134-135.

2. Sinzinger H, et al. Two different types of exercise- induced muscle pain without myopathy and CK-elevation during HMG-Co-enzyme-A-reductase inhibitor treatment. Atherosclerosis 1999;143:459-460.

3. Bedi RS. Generalised muscle cramps with inhalation of salmeterol. Indian J Chest Dis Allied Sci 1995;37:51-52.

4. http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm218202.htm. Accessed January 6, 2012.