Practice Point: Vitamin K2 might work for nocturnal leg cramps. Or it might not. Don’t change your practice yet.
EBM Pearl: Researchers — please don’t cramp study validity with unclear published methods or irreproducible results. When the validity of study data is in question, going ahead with the intervention because it seems "benign" might be exposing people to the potential harm(s) of a treatment without the possibility of benefit.
Have you ever woken up in the middle of the night with a sudden, painful, leg-thrashing cramp? People who experience frequent nocturnal leg cramps would do just about anything to prevent them, and massage and stretching usually don’t work. Unfortunately, there are few treatments that are both effective and safe.
A highly publicized randomized trial published in JAMA Internal Medicine suggested new hope for avoiding nocturnal leg cramps. The study, performed in China, evaluated vitamin K2 (menaquinone) in adult volunteers ≥ 65 years old with at least two episodes of unexplained nocturnal leg cramps (average was two to three episodes) in the last two weeks. [Side note: Vitamin K2 is not found in standard vitamin K tablets or foods such as spinach that are conventionally considered to be high in vitamin K (which are usually high in K1) but can be found in a variety of probiotic-containing and fermented foods.]
Back to the study details: Cramps associated with specific metabolic diseases and neuropathies were excluded. The participants, who volunteered for the study, were on average 72 years old and weighed about 52 kg (approximately 115 pounds), and the majority had hypertension and/or diabetes. Vitamin K2 was prescribed as 180 mcg orally once nightly for eight weeks. Patients recorded the number, severity, and duration of cramps by diary, with investigators checking in regularly by phone. The authors reported that after eight weeks, vitamin K2 was found to reduce the number of cramps per week (mean 0.96 vs. 3.63 in the placebo group), severity of cramps, and duration of cramps. No adverse effects were reported.
Don’t get too excited, though. Trustworthy results require trustworthy (valid) methodology. When study methods are not clearly reported or are inconsistently reported, we lose confidence in the results.
In this case, there were some noteworthy discrepancies noted in how randomization and data collection methods were reported in a public peer review from April 2025. These concerns were presented by Professor Alison Avenell and organized according to the REAPPRAISED checklist. We agree there should be more details about how data was collected. The authors performed an intention-to-treat analysis, and although the published protocol states that the reasons why patients were excluded or dropped out from protocol would be reported, they failed to clearly do so. If patients drop out of a study because bad things happen to them (even things that don’t initially seem related to the research), readers should be made aware of that. In addition, concerns were raised related to the impossibility of the outcome data due to the precision of confidence intervals (CIs) for between-group differences (-2.67 cramps/week, 95% CI -2.86 to -2.49). When we calculate CIs from the data provided in the supplement, we get CIs that are wider than this, similar to what is presented in the peer review.
All these questions raise concerns. However, while writing this EBM Focus, a notice of retraction and replacement was published in which the authors clarified some discrepancies and committed to resubmit their publication with errors and corrections highlighted. While some might question whether these concerns should have been raised via the original peer review process, we’re going to go with better late than never.
This all leads to the bigger issue of the pressure to publish, which inevitably contributes to the growing problem of untrustworthy data. In 2023 alone, more than 10,000 research papers were retracted from medical literature. And how many more are in circulation yet to be discovered?
Online publishing and artificial intelligence (AI) make the critical appraisal of medical literature simultaneously more important and more difficult. It is essential that clinicians 1) know this problem exists and 2) either perform critical analysis themselves—a Herculean task—or find trusted tools (such as DynaMedex) to do it for them. The stakes are too high to believe everything you read.
We suspect that the original paper will have far more views than any retraction published by the authors. So, for now, you can trust that the evidence is unclear about the use of Vitamin K2 for nocturnal leg cramps.
For more information, see the topic Nocturnal Leg Cramps in DynaMedex.
