Authors: Arno Kuusipalo, Juhani Laitila, Eemil Lehtonen, Jenni Joensuu, Pertti Mustajoki, Pekka Mustonen, Heini Huhtala, Minna Kaila, Tuomas H. Koskela
Categories: Research, Leg cramp, Compression stockings, Magnesium, Secondary prevention
Source: Trials
Authors: Arno Kuusipalo, Juhani Laitila, Eemil Lehtonen, Jenni Joensuu, Pertti Mustajoki, Pekka Mustonen, Heini Huhtala, Minna Kaila, Tuomas H. Koskela
Leg cramps are common among older adults and often lead to sleep disturbances and reduced quality of life. However, there is no consensus on how to treat this condition. This study aimed to evaluate the effectiveness of compression stockings in preventing leg cramps in individuals aged 50 to 85.
This study was a three-arm, parallel-group, partially blinded, randomized placebo-controlled trial conducted in Finland. Participants were recruited nationwide through online advertisements and primary care centers. Eligible individuals had experienced at least two leg cramps per week during the previous 4 weeks. Participants were randomized to receive either knee-high medical compression stockings, magnesium hydrochloride, or placebo pills, to be used daily for 4 weeks. The primary outcome was the change in leg cramp frequency at week 8. Secondary outcomes included the number of leg cramp-related nocturnal awakenings and the perceived pain intensity on an ordinal scale.
A total of 121 participants were randomized, and 109 (90.1%) completed the trial. The primary outcome analysis included 114 participants. The mean age was 65.8 years (SD 7.8), and 87 (71.9%) were women. At baseline, the median number of weekly leg cramps was 4 (IQR 3–7). At week 8, the median weekly leg cramp frequency was 2 (IQR 1–2.5) in the compression stockings group, 3 (IQR 2–6) in the magnesium group, and 3 (IQR 2–5) in the placebo group. The baseline-adjusted mean difference in leg cramp frequency between the compression stockings and placebo group was −1.43 (95% CI −2.36 to −0.50; P = .001). No significant difference was observed between the magnesium and placebo groups, with an adjusted mean difference of −0.20 (95% CI −1.49 to 1.09; P = 0.929). Four participants discontinued compression stockings due to adverse reactions. No serious adverse events were reported.
Among older adults, daily use of compression stockings was effective in reducing the frequency and pain intensity of leg cramps, as well as the number of nocturnal awakenings caused by them.
ClinicalTrials.gov NCT04694417. Registered on 4 January 2021.
The online version contains supplementary material available at 10.1186/s13063-025-09370-z.
Muscle cramps are painful, involuntary muscle contractions that begin suddenly. They typically last from a few seconds to a few minutes before resolving spontaneously [1]. Actively stretching the affected muscle can accelerate recovery [2]. Leg cramps can occur at any time but most commonly arise at night, in which case they are referred to as nocturnal leg cramps (NLC) [1]. A substantial proportion of the population experiences leg cramps, and their prevalence increases with age, with up to 60% of adults reporting having had them at some point in their lives. NLC is strongly associated with secondary insomnia, which can lead to a reduced overall quality of life [1, 3]. The etiology of leg cramps is multifactorial, and the underlying pathophysiological mechanisms remain unclear, with most cases considered idiopathic in clinical practice [1].
Currently, there is no consensus on how to treat NLC. Various treatments—such as magnesium supplementation, quinine, massage, exercise, and stretching—have been studied, but evidence for massage and exercise is limited [1, 4–10]. However, regular calf stretching might be effective as a preventive measure [10]. Quinine has been shown to be somewhat effective, but it carries serious side effects and is difficult to use safely because of adverse drug interactions [3, 5, 11]. For idiopathic leg cramps, magnesium supplementation is the most studied treatment, but there is little to no evidence of its effectiveness [1, 3, 6–9, 12]. For muscle cramps caused by exercise or underlying disease, various medications have been tried, with limited evidence of effectiveness [3, 4, 10].
In a widely used Finnish online health portal article on muscle cramps, interactive feedback indicated that 25 individuals reported a noticeable reduction in their cramping symptoms when regularly wearing knee-high compression stockings [13]. However, to our knowledge, no previous research has examined the efficacy of compression stockings in preventing leg cramps. Young et al. mentioned “compression hosiery” as a potential preventive measure for idiopathic leg cramps but, similarly, found no research on the topic in their review article [14]. Venous insufficiency has been associated with leg cramps, and its treatment has been linked to alleviation of cramp symptoms [15–22]. It is therefore possible that interventions that enhance lower-extremity venous return could benefit individuals suffering from leg cramps. By passively improving lower-limb circulation, the use of compression stockings could potentially reduce symptoms.
The primary aim of this study was to evaluate the efficacy of commonly used Class I knee-high compression stockings in preventing leg cramps, compared with placebo pills or magnesium supplementation, in a randomized controlled trial. The secondary objectives were to assess the effects of compression stockings on the frequency of nocturnal awakenings caused by leg cramps and on the perceived pain intensity of the cramps.
This trial was a randomized, partially blinded, placebo-controlled study. A pilot study testing the research protocol by Joensuu et al. was published in 2021 [13]. The trial was registered at ClinicalTrials.gov (NCT04694417) on January 4, 2021. Participants were randomized into three intervention compression stockings, magnesium supplementation, and placebo pills. They were followed weekly for eight consecutive 4 weeks prior to the intervention (run-in) and 4 weeks during the intervention. The primary outcome was the change in average weekly leg cramp frequency between the run-in and intervention periods. Secondary outcomes included changes in the average weekly frequency of nocturnal awakenings caused by leg cramps and the perceived pain intensity of the leg cramps.
Participants were recruited nationwide between June 2021 and December 2022 through a Google advertisement, the Finnish Terveyskirjasto.fi health library, and health clinics in the Pirkanmaa region, which serves approximately 9.3% of the Finnish population. The recruitment advertisement included a link to a document providing detailed information about the study and a link to sign up. Individuals who signed up were assigned a unique identification code and given a separate link to provide baseline information, including age, sex, comorbidities, medications, supplements, lifestyle factors (alcohol use, smoking, exercise), and recent history of leg cramping and nocturnal awakenings caused by leg cramps.
The inclusion criteria were age between 50 and 85 years and experiencing lower limb muscle cramps at least twice a week for the preceding four consecutive weeks. The exclusion criteria were as diagnosed or suspected peripheral arterial disease, peripheral artery bypass surgery, severe peripheral neuropathy, severe sensory disorder, allergy to materials in compression stockings, severe kidney failure (estimated glomerular filtration rate [eGFR] < 30 ml/min), ongoing magnesium supplementation, pulmonary edema, massive leg edema due to heart failure, soft tissue problems affecting the legs (including skin grafts, thinned skin, leg ulcers, gangrene, skin inflammation, or inflammation of subcutaneous connective tissue), foot or leg deformities, or abnormal foot shape or size that would prevent proper fitting or use of compression stockings.
After reviewing the baseline information, participants who met the inclusion criteria and had no exclusionary conditions were enrolled in the study. The presence of symptoms required for inclusion and the absence of exclusionary conditions were not verified through external sources, such as electronic medical records or information from participants’ healthcare providers. Participants were not clinically examined by the research team. Enrollment decisions were based on participant self-reporting and, when necessary, clarification via e-mail or phone conversations.
Eligible participants were sent an email outlining the study procedures, their individual timetable, and starting date, along with a request to track their leg cramp symptoms—preferably in a notebook to minimize recall bias—and report them weekly. At the beginning of the follow-up period, they were also provided with a letter by post for written informed consent.
Participants who reported regular use of magnesium supplements were included if they agreed to discontinue independent magnesium supplementation before the start of the run-in period. All other participants were likewise advised not to take independent magnesium supplements for the duration of the study.
Participants were allocated 1:1 to receive either magnesium pills, placebo pills, or two pairs of compression stockings. Randomization was performed by a separate research assistant who had no contact with participants during the study. A stratified sampling method, based on participants’ ages, was used. Randomization occurred after a participant was accepted into the study, during the third week of the run-in period. A computer-generated sequence served as the template for randomization, employing a stratified permuted block design. Age strata of 50–60, 60–75, and 75–85 years were used to ensure balanced age distribution across groups, with a block size of six to promote equal numbers of participants per group.
All participants and the research assistants involved in participant follow-up and data collection were blinded to the magnesium and placebo pill interventions. The pill packages were identical, differing only by a unique identification number. Blinding was not possible for the compression stockings group (CS group) in this study.
The CS group received CE-marked (Conformité Européenne) Class I (25–40 mmHg) knee-high medical compression stockings. Stocking size was determined based on self-measurements taken by the participants, who were instructed to measure the maximum circumference of their calves and ankles. Using this information, the research assistants selected and sent the appropriate size according to the manufacturer’s guide. If participants reported fit issues with the initial stockings, alternative sizes were provided until an appropriate fit was achieved. Participants were instructed to put on the stockings immediately after waking and wear them until going to bed. The magnesium group received a package of 28 pills, each containing 620 mg of magnesium hydrochloride, equivalent to 250 mg of elemental magnesium per pill. The placebo group received an identical package of 28 pills with no active ingredients. Both groups were instructed to take one pill daily for 4 weeks.
Participants were sent a link to an online form each week via email to report their symptoms and intervention usage. Each participant was followed for a minimum of four run-in weeks and 4 weeks during the intervention, resulting in a minimum participation time of 8 weeks. In the follow-up questionnaire, participants reported the total number and location of muscle cramps experienced during the week, choosing from five ankle, calf, thigh, elsewhere, or no cramps. Additionally, participants reported the most commonly experienced pain intensity of the cramps on an ordinal numerical rating scale from 0 to 10 (NRS), the total number of times they were woken by a muscle cramp, and the number of days per week they used their assigned intervention. If a participant did not complete the weekly online form on time, a research assistant followed up via email or phone to remind them to respond. A participant was considered lost to follow-up if they stopped responding to the online forms, and no contact could be established after at least two reminder attempts.
If a range was provided instead of a single number in a response, the mean of the range was used in the analysis. If a participant did not experience leg cramps during a given week, the pain intensity for that week was recorded as 0. If a participant reported a number of nightly awakenings caused by leg cramps that exceeded the total number of leg cramps, the number of awakenings was interpreted as representing the total number of leg cramps for that week. This occurred only twice, in the case of two participants during the run-in period.
An interim analysis was not specified in the protocol. However, in October 2022, an independent, blinded statistician reviewed the data due to slower-than-expected recruitment and determined that the study population had sufficient power to detect significant differences between the intervention groups, even with a smaller sample than the originally planned target size. Based on this assessment, a decision was made to discontinue recruitment by December 31, 2022.
The study was designed to have sufficient power to detect a small effect size (Cohen’s d = 0.25) in an analysis of variance, with 80% power at a two-sided significance level of 0.05, corresponding to a required sample size of 159 participants. Accounting for an anticipated 30% loss to follow-up observed in the pilot study [13], the target sample size was increased to 225 participants, resulting in 75 participants per group.
All data analyses were conducted according to the intention-to-treat principle. Missing data were not imputed but were handled directly using a multivariate model with a mixed model for repeated measures (MMRM). For the secondary outcome of reported pain, a complete case analysis was performed.
The primary outcome was analyzed using a multivariate model based on a restricted maximum likelihood (REML)-based mixed model for repeated measures (MMRM). The model included group, week, and group-by-week interactions as fixed effects and the mean value of the response variable during the follow-up period and its interactions with time as covariates. Treatment groups and participants were included as random effects. An unstructured covariance structure was used to model within-patient errors. The Kenward–Roger approximation was applied to estimate denominator degrees of freedom. Significance tests were based on estimated marginal means using a two-sided α = 0.05 (corresponding to 95% confidence intervals). The primary endpoint was the difference in estimated least squares means (LSM) between treatments at week 8. Tukey’s method was used to adjust P-values for multiple comparisons.
For the secondary outcome of the number of nocturnal awakenings, the same approach as for the primary outcome was applied. The ordinal pain outcome was analyzed using a multivariable Bayesian mixed-effects proportional odds semiparametric ordinal logistic model, employing the same covariates and random effects as in the primary MMRM model, except that group was not treated as a random effect. Restricted cubic splines were used to accommodate potential nonlinear effects of baseline variables. The Bayesian model assumed compound symmetry for correlations between repeated measures. Treatment comparisons were expressed as the difference in odds of a higher pain score between treatment groups at week 8, with odds ratios less than 1 indicating better pain outcomes. The prior for the intercept followed a Dirichlet distribution, while priors for all other coefficients were assumed to follow a normal distribution with a mean of 1 and a standard deviation of 100.
A sensitivity analysis using multiple imputation was conducted to assess the potential impact of missing data. Statistical significance was defined as a two-tailed P = 0.05. All analyses were performed using R software (R version 4.1.0) with the MRMM and rmsb packages.
The study flow diagram is presented in Fig. 1. A total of 244 individuals enrolled in the study, of whom 123 (50.4%) met the eligibility criteria. Two participants discontinued before randomization, and of the 121 participants who were randomized, 109 (90.1%) completed the entire study, resulting in a dropout rate of 9.9% among randomized participants. The highest dropout rate was observed in the CS group (21.4%), while dropout rates were lower in the magnesium and placebo groups (5.4% and 2.4%, respectively). A total of 114 participants (94.2% of those randomized) provided at least one measurement during the intervention period and were included in the primary analysis.Fig. 1Flow diagram of the study. ^a^Diagnosed or suspected peripheral arterial disease, bypass surgery for a peripheral artery, severe peripheral neuropathy, severe sensory disorder, allergy to materials in compression stockings, severe kidney failure (eGFR < 30 ml/min), ongoing magnesium carbonate usage, pulmonary edema, massive leg edema caused by heart failure, soft tissue problems affecting legs (including skin graft, thinned skin, leg ulcer, gangrene, skin inflammation, or inflammation of subcutaneous connective tissue), foot or leg malformation, or abnormal shape or size of foot that would cause problems finding or using appropriate compression stockings
All groups were similar with respect to most baseline characteristics, as shown in Table 1. The mean age of randomized participants was 65.8 years (SD = 7.8), and 71.9% (87/121) were female. The magnesium group had a higher proportion of male participants and a greater number of participants with diabetes and hypertension compared to the other groups. Comorbidities for each group are provided in Additional File 1, and medications used are listed in Additional File 2. During the study, nearly all reported cramps (98.2%) were localized to the calf, foot, or thigh, with each episode causing at least some degree of local pain. Additionally, 50.5% of all reported cramps resulted in sleep disruption. Table 1Baseline participant characteristicsNo. (%)Characteristic^a^Stockings, N = 42Magnesium, N = 37Placebo, N = 42Age, mean (SD), y66.5 (8.4)66.5 (7.5)64.6 (7.4)Female sex31 (73.8)23 (62.2)33 (78.6)Body mass index, mean (SD)26.7 (5.6)28.2 (5.3)27.6 (5.7)Multiple diseases (2+)24 (57.1)27 (73.0)24 (57.1)Hypertension14 (33.3)20 (54.1)15 (35.7)Heart disease^b^7 (16.7)8 (21.6)5 (11.9)Diabetes2 (4.8)9 (24.3)3 (7.1)Use of magnesium supplements before enrollment14 (33.3)13 (35.1)12 (28.6)Excessive alcohol use^c^1 (2.4)1 (2.7)2 (4.8)Weekly exercise sessions (≥ 30 min)^d^ 0–211 (26.2)17 (45.9)14 (34.1) 3–416 (38.1)11 (29.7)15 (36.6) 4+ 15 (35.7)9 (24.3)12 (29.3)^a^Characteristics were self-reported^b^Any of the coronary artery disease, heart failure, heart arrhythmia^c^Defined per Finnish for men, consuming 24 standard drinks (12 g of pure ethanol) or more per week, and for women, consuming 16 standard drinks or more per week^d^Based on 120 participants
At baseline, the median weekly frequency of leg cramps across all groups was 4 (IQR 3–7) (see Additional File 3). Between the run-in and intervention periods, the change in the median number of weekly leg cramps was −2 in the CS group, −1 in the magnesium group, and 0 in the placebo group, as shown in Table 2. This is reflected in the baseline-adjusted mean change in weekly leg cramp incidence, illustrated in Fig. 2. At week 8, the baseline-adjusted mean number of leg cramps in the CS group was significantly lower than in the placebo group, with an adjusted mean difference of −1.43 (95% CI, −2.36 to −0.50; P = 0.001; see Fig. 2). In the magnesium group, the baseline-adjusted mean number of weekly leg cramps was largely similar to the placebo group throughout the intervention period, with an adjusted mean difference of −0.20 (95% CI, −1.49 to 1.09; P = 0.929) at week 8. All outcome measurements from weeks 5 through 8, along with between-group comparisons, are provided in Additional File 3. Table 2Summary of study outcomesMedian (IQR)Stockings, N = 42Magnesium, N = 36Placebo, N = 42Group comparisons at study week 8Run-inIntervention^a^Run-inInterventionRun-inInterventionStockings-placeboMagnesium-placeboStockings-magnesiumPrimary outcome Weekly number of leg cramps4 (3–6)2 (1–3)4 (3–8)3 (2–5)4 (3–8)4 (2–6)Mean difference^b^−1.43 (−2.36 to −0.50)−0.2 (−1.49 to 1.09)−1.23 (−2.33 to −0.13)Secondary outcomes Weekly number of nocturnal awakenings1 (1–3)0 (0–1)2 (1–4)1 (0–3)2 (1–5)2 (0–3)Mean difference^b^−1.12 (−2.05 to −0.19)−0.06 (−1.07 to 0.95)−1.06 (−1.73 to −0.39) Pain Numerical Rating Scale score4 (3–6)3 (2–4)5 (4–6)4 (3–6)4 (4–7)4 (3–6)Odds ratio^c^0.27 (0.09 to 0.85)0.56 (0.19 to 1.78)0.48 (0.14 to 1.57)Run-in, study weeks 1–4; intervention, study weeks 5–8^a^Based on 36 participants as 6 were lost to follow-up^b^Least squares mean difference scores from the mixed models for repeated measures (MMRM) model. Adjusted for baseline (run-in score). Group comparisons for intervention weeks 5–8 are presented in Additional File 3^c^Odds ratios from the Bayesian proportional odds model. OR < 1.0 indicated that the intervention had shifted the ordinal outcome toward lower pain scores. Intervals are the 0.95 highest posterior density intervals. Adjusted for baseline (run-in score). Group comparisons for intervention weeks 5–8 are presented in Additional File 3Fig. 2Mean change in weekly leg cramp incidence by intervention group. Results are from a mixed model for repeated measures. Error bars represent standard errors. The run-in value (star, 5.77) indicates the mean number of leg cramps during the 4-week screening period (run-in), which the model is adjusted for. Week 8 is the primary endpoint in the intention-to-treat P = 0.001 for the stockings vs. placebo comparison and P = 0.929 for the magnesium vs. placebo comparison
Between the run-in and intervention periods, the change in the median number of weekly nocturnal awakenings caused by leg cramps was −1 in the CS group. In the magnesium and placebo groups, the reductions were −1 and 0, respectively. At week 8, the baseline-adjusted mean number of weekly nocturnal awakenings was significantly lower in the CS group compared to the placebo group, with an adjusted mean difference of −1.12 (95% CI, −2.05 to −0.19; P = 0.015). In the magnesium group, the baseline-adjusted mean number of nocturnal awakenings throughout the intervention period did not differ significantly from placebo, with an adjusted mean difference at week 8 of −0.06 (95% CI, −1.07 to 0.95; P = 0.989; see Additional Files 3 and 4).
There was a shift toward lower leg cramp-associated pain levels in the CS and magnesium groups compared with placebo, which was more pronounced in the CS group (see Additional Files 3 and 5). At week 8, the CS group had a baseline-adjusted posterior probability of 0.990 for achieving a better outcome on the ordinal NRS scale of pain intensity compared with the placebo group, with a baseline-adjusted posterior odds ratio (OR) of 0.27 (95% credible interval [CrI], 0.09–0.85). At week 8, the magnesium group had a baseline-adjusted posterior probability of 0.844 for a better outcome than placebo, with a baseline-adjusted posterior OR of 0.56 (95% CrI, 0.19–1.78). The CS group had a 0.874 baseline-adjusted posterior probability of a better pain outcome than the magnesium group, with a baseline-adjusted posterior OR of 0.48 (95% CrI, 0.14–1.57).
To explore the impact of missing data, a post hoc sensitivity analysis of the primary outcome was conducted. This analysis compared the primary MMRM model with a similar model in which all missing data were fully imputed, as well as a model including only participants with complete responses (see Additional File 6). The intervention effects were similar across all models.
Four participants in the CS group (9.5%) experienced adverse effects severe enough to discontinue treatment (Fig. 1). One developed eczema, two experienced pain, and one reported itching and discomfort caused by the stockings. In both intervention groups, all other dropouts were due to health issues deemed unrelated to the treatment or loss of contact for unknown reasons.
Adherence to the treatment was lower in the CS group compared to the magnesium group (see Additional File 7). Participants in the CS group used the intervention on a daily basis for only 42.3% of the weeks, whereas complete adherence was reported for 83.9% of the weeks in the magnesium group. The CS group was also the only group to report complete nonadherence during 2.3% of the intervention weeks. Similar to the magnesium group, daily usage in the placebo group was reported for 85.3% of the weeks. Nevertheless, in the CS group, the intervention was used at least 5 days per week for 72.3% of the weeks.
This study represents the first randomized clinical trial to assess the efficacy of compression stockings as a treatment for idiopathic leg cramps. Our results demonstrated that compression stockings effectively reduced the frequency of leg cramps, as well as associated pain and nocturnal awakenings, in individuals aged 50–85. A similar clinically meaningful effect was not observed for magnesium supplementation in this study.
A recent Cochrane review on nondrug therapies for lower leg cramps reported that regular stretching was the only intervention studied, with potential efficacy in reducing the nightly frequency of cramps from 2 to 1 after 6 weeks [10]. An evidence-based review from 2009 concluded that no clinically significant trials had yet evaluated compression stockings for the treatment of leg cramps [14]. We postulate that the therapeutic effect of compression stockings observed in this trial may be mediated by increased venous return and reduced fluid retention in the lower extremities.
Magnesium hydrochloride supplementation did not provide significant benefits in reducing the frequency of leg cramps or nocturnal awakenings, consistent with previous studies reporting mixed results regarding the efficacy of various forms of magnesium across different patient populations and clinical settings [7, 23, 24]. However, in this study, magnesium supplementation did modestly alleviate the pain associated with leg cramps.
Perhaps the most important finding was that in the CS group, the median number of weekly nocturnal awakenings caused by leg cramps decreased to nearly zero. The CS group already had fewer nightly awakenings during the run-in period, but the baseline-adjusted mean number at week 8 was 1.03, compared with 2.09 and 2.15 in the magnesium and placebo groups, respectively. The significant reduction in nightly awakenings observed in the CS group may represent the most quality-of-life–enhancing effect of using compression stockings for the secondary prevention of NLC. It is possible, however, that some participants reported all nocturnal awakenings in the follow-up questionnaire, even though they were asked specifically about those caused by leg cramps. Therefore, the observed reduction in the CS group may not entirely reflect fewer leg cramps per se but could at least partly be explained by increased venous return during the daytime, which reduces fluid accumulation in the lower extremities. This fluid would otherwise be mobilized during the night in the supine position, potentially contributing to nocturia, a common condition in older adults.
The reported adverse effects in the CS group were generally minor and consistent with previous studies documenting side effects associated with compression stockings [25]. We acknowledge that wearing compression stockings may be challenging for some individuals. Although not reported in our study, donning or removing compression stockings can be physically difficult and frustrating, particularly for those with reduced muscle strength, limited mobility, back pain, or impaired coordination.
This study was the first with a randomized prospective design and a relatively large sample size to evaluate the effect of compression stockings on leg cramps. The magnesium and placebo groups were double-blinded. Nationwide representation was achieved through Internet-based recruitment, encompassing participants across the full targeted age range of 50–85 years.
In clinical practice, muscle cramps are typically diagnosed based on medical history, characterized by sudden, involuntary, and painful tightening or hardening of muscles that is usually relieved by stretching the affected body part [26]. Other causes of discomfort must first be ruled out, such as restless legs syndrome, in which individuals experience an uncomfortable, non-painful sensation in the lower limbs that induces an irresistible urge to move the legs, particularly at night before sleep [27]. In this study, no objective methods were used to confirm whether the symptoms reported by participants reflected actual muscle cramping, consistent with everyday clinical practice, nor to thoroughly exclude other pathologies that could be mistaken for leg cramps. Self-reporting and recall errors may therefore have introduced bias. Additionally, there may have been discrepancies in how participants interpreted muscle cramping; some may have reported other symptoms as leg cramps, such as restless legs syndrome episodes, periodic limb movement disorder episodes, or leg pain without muscle contractions caused by other pathologies.
The requirement to wear compression stockings strictly from morning to noon is less convenient than taking a pill, which may explain the lower adherence and higher dropout rate. Similar adherence challenges have been reported previously among individuals using compression stockings [25]. The effect observed in the CS group might have been even greater with optimal adherence. Nevertheless, we believe that the observed effect in this study reflects pragmatic outcomes that are achievable in routine clinical practice. Therefore, the robust results of this study may be clinically meaningful and relevant for patient outcomes in community settings, where leg cramps are a common nuisance [3].
Further studies are needed involving different participant demographics, varying daily usage patterns, leg cramps of different etiologies, and stockings of higher compression class to better quantify the effects of this accessible, affordable, and non-pharmacological treatment.
Our study provides evidence that unlike magnesium supplementation, the use of compression stockings appears to be an effective and plausible option for treating idiopathic leg cramps in the 50–85 age group without contraindicated conditions. To date, treatment options for this very common condition have been limited.
Additional file 1. Comorbidities by intervention group. Number and percentage of participants with each comorbidity, presented by intervention group. NR, no responses. NA, not applicable.Additional file 2. Medications by intervention group. A table presenting the number and percentage of participants with ongoing use of specific drugs or drug classes, shown by intervention group. NR, no responses. NA, not applicable.Additional file 3. Group comparisons. A table presenting the median and interquartile range (IQR) for leg cramp frequency, frequency of nocturnal awakenings caused by leg cramps, and the most commonly experienced pain intensity (on a 0–10 numerical rating scale) for each intervention group during weeks 1–8. Baseline-adjusted means with 95% confidence intervals (CI) are shown for leg cramp frequency and frequency of nocturnal awakenings in weeks 5–8. Baseline-adjusted posterior means with 95% CI are presented for the most commonly experienced pain intensity in weeks 5–8. The table also displays baseline-adjusted mean differences with 95% CI and p-values for group comparisons between Compression Stockings and Placebo, Magnesium and Placebo, and Compression Stockings and Magnesium in weeks 5–8. For pain intensity, baseline-adjusted odds ratios (OR) with 95% credible intervals (CrI) and posterior probability values are reported for comparisons between intervention groups in weeks 5–8.Additional file 4. Nocturnal awakenings group comparison. A graph showing the baseline-adjusted mean number of nocturnal awakenings caused by leg cramps, with 95% confidence intervals (CI), for each intervention group during weeks 5–8.Additional file 5. Pain intensity group comparisons. A graph illustrating the percentage of participants in each intervention group classified as having low, moderate, or high pain associated with leg cramps during weeks 1–8. Pain status for each participant in a given week was defined by the most commonly experienced pain intensity of leg cramp episodes on a 0–10 numerical rating scale, with 0–3 representing low pain, 4–6 moderate pain, and 7–10 high pain.Additional file 6. Sensitivity analysis. A table comparing the baseline-adjusted mean differences (with 95% CI) for the primary outcome measure, leg cramp frequency, between intervention groups during weeks 5–8. Estimates are shown from the primary MMRM model, a similar model with fully imputed data, and a model including only participants with complete data.Additional file 7. Adherence to treatment. A table showing the number and percentage of person-weeks classified by different levels of treatment adherence in each intervention group during the intervention period (weeks 5–8). Adherence for each individual person-week was categorized as every day (full adherence), 5–6 times per week, fewer than 5 times per week, unknown frequency, and 0 times per week (non-adherence).Additional file Changes from protocol.