Wrapping up Knee Pain: Cabbage Leaf Wraps for Osteoarthritis of the Knee

Cabbage leaf wraps (CLWs) … the newest health craze trending on social media? A culinary masterpiece? Or an effective complementary treatment for osteoarthritis (OA) relief? A recently published study attempted to answer the latter question. Management of OA requires a multifaceted approach, including lifestyle changes, pharmacologic treatments, nonpharmacologic therapies, and occasionally surgery, depending on the severity of the patient’s pain and disability. Approximately 30 million American adults have osteoarthritis (OA) and the lifetime risk of developing OA of the knee is around 50%.¹ Symptoms of OA, including pain and functional impairment, can significantly impact patient quality of life (QoL).

Nonpharmacologic therapies for knee OA include increased physical activity and weight loss —often combined with medication treatment.^2-4 Topical medication options for knee OA include non-steroidal anti-inflammatory drugs (NSAIDs) and capsaicin. While topical NSAIDs have a lower risk of gastrointestinal, renal, and cardiovascular toxicity compared to oral formulations, many patients are not candidates for NSAIDs due to other comorbidities.^5,6 Capsaicin was not recommended in the 2012 American College of Rheumatology guidelines for knee OA and has not been studied in direct comparison to NSAIDs.³ Given that up to two-third of patients with OA have other medical comorbidities which preclude NSAID use, complementary and alternative medication therapies are often used.⁷

CLWs have been used for the treatment of OA for years. To prepare the wraps, the cabbage stem is removed and the leaves are placed on a hard surface such as a cutting board, and a bottle or rolling pin is used to bruise the leaves to release the plant juices. One to two leaves are then layered around the affected joint and fixed in place with a bandage. Although anecdotal reports regarding the effectiveness of CLW have been published⁸, until now there have been no published controlled studies evaluating the use of CLWs for knee OA.

This German study was a randomized controlled, open-labeled, 3-armed parallel group trial designed to compare the effects of CLWs to usual care (UC) and diclofenac topical pain gel (TPG) for symptom improvement and QoL in patients with knee OA.⁸ Usual care was defined as patients’ usual activities and therapies, including pain medications, used for symptom management. The primary outcome measure was pain intensity after four weeks of treatment as measured by the Visual Analogue Scale (VAS). Secondary outcome measures included: pain intensity, functional disability, QoL, and self-efficacy after four and 12 weeks; physical function and pressure pain sensitivity after four weeks; and pain intensity and medication use reported in participants’ daily log, as well as medication adherence. 

A total of 207 individuals were screened for the study and 81 were enrolled and randomized in a sequential order to CLWs, UC, or TPG (27 participants were enrolled in each group).  Study inclusion and exclusion criteria are listed in Table 1.

Table 1: Inclusion/Exclusion Criteria

^{*Defined by the Kellgren-Lawrence classification}

After randomization, patients received instructions and a four weeks’ supply of their assigned treatment. Patients in the CLW group were provided with detailed instructions on how to prepare the wraps. The investigators asked the patients to wear the wraps for a minimum of two hours per day, but suggested that the wraps be left on overnight. Patients in the TPG group were provided with diclofenac 10 mg/g gel and were advised to rub one to four grams of the gel to the knee up to four times per day. Patients in the UC group were advised to continue their usual care interventions (except for what was listed in the exclusion criteria), but not to initiate any new therapeutic regimens for symptom management.  Individuals in the CLW and TPG groups were allowed to continue their usual care treatments.

Baseline characteristics between the three groups were similar.  Slightly more than 50% of the participants were female and the mean age was 65 years. All patients were white and the majority had less than a high school level of education and were retired. Half of the patients reported prior use of medications for symptom management. The only significant difference noted in baseline characteristics between the three groups was BMI (TPG > CLW > UC).

Patients were assessed at baseline, four weeks, and 12 weeks after study enrollment, using an intention-to-treat analysis using the last observation carried forward. Outcomes were measured using the study tools listed in Table 2. Participants kept daily logs that were used to assess adherence, pain severity, and use of analgesics for symptom management. During the study, compliance was highest at the beginning of the four week intervention period and dropped to 80% in the last week. Over the four week period, 79.2% of the CLW patients and 92% of TPG patients were adherent for at least 80% of the days.

Table 2: Method of Measurement for Secondary Outcomes

After the four week intervention period, a significant difference between the CLW and UC groups was observed for the primary outcome of pain intensity as assessed by VAS. However, no between group difference was observed when comparing CLW and TPG (Table 3, 4). When pain intensity was assessed at 12 weeks, no difference could be found between CLW and UC or CLW and TPG. After adjustments were made for adherence, the effect on the primary outcome remained significant compared with the UC group and nonsignificant compared to the TPG group.

Table 3: Results of Analyses Between CLWs and UC

^{*: P < 0.05; NS=Not Significant}

Table 4: Results of Analyses Between CLWs and TPG

^{*: P < 0.05; NS=Not Significant}

When assessing daily logs for concurrent medication use, patients in the CLW group tended to use fewer analgesics and had the lowest daily doses of analgesics compared to the UC and TPG groups. However, a higher percentage of the CLW group used concurrent therapies. The therapies used most often by the CLW group include guideline-based recommendations such as acupuncture and physical therapy (See Table 5).³  In assessing CLWs for safety, one patient in the CLW group complained of dry cough, which resolved after discontinuation of an angiotensin II converting enzyme inhibitor. Another patient in the CLW complained of itching and burning during CLW application and stopped the treatment. The same patient later reported a zoster infection and resigned from the study.

Table 5: Most frequently used non-study interventions

In summary, this study found that CLW was more effective than UC with respect to pain, functional disability, and quality of life. However, CLWs were not found to be more effective than TPG. The strengths of this study include its randomized, three-arm design, including a comparison to UC and TPG. Limitations of the study include its lack of blinding to physicians and patients, which would admittedly be difficult due to the nature of the interventions. Furthermore, the wide variability in the use of treatment modalities within each group limits the ability to discern the impact of the study interventions on reducing knee pain. Moreover, patients in the CLW group had a higher use of acupuncture and physical therapy, which are often used to reduce pain and thus may have influenced the results. This study was designed to examine the benefits of CLW on patients with knee OA who suffered from moderate pain; however the average baseline VAS of the participants was roughly 40 mm, which is considered mild pain. Therefore, the results can’t be generalized to patients with more advanced OA. As with all studies regarding pain, the outcome measures were based on subjective reports including the quality of life assessments. All patients in the study were white from Germany, and thus the results may not be generalizable to other racial or ethnic groups. Finally, substantial differences in the amount of TPG applied and the variable duration of CLW application further confounds the interpretation of these results..

So, should we start looking for CLWs in our social media feeds or in the next version of knee OA guidelines? Given the limitations of the current study, we do not believe the evidence is strong enough to recommend CLWs for all patients, especially as monotherapy. However, CLWs do represent a lower-cost option for patients unable to afford topical diclofenac gel. We believe CLWs could be considered as one component of knee OA management, but further research is needed to define their place in therapy. What do you think? Should pharmacists suggest CLW for knee OA or continue to recommend lifestyle modification and pain medications  as the mainstays of treatment? Have patients in your practice reported using CLWs? What other complementary or alternative therapies do you recommend for your patients with OA?