The feasibility and acceptability of neuromuscular electrical stimulation to improve exercise performance in patients with advanced cancer: a pilot study

Patients with cancer frequently suffer problems common to other chronic diseases, namely long-lasting sequelae from their disease and its treatments [1] as well as progressive decline in overall functional status [2, 3]. In many chronic medical conditions e.g. chronic obstructive pulmonary disease (COPD), and congestive heart failure (CHF), exercise-based rehabilitation programs have a major impact on quality of life and functional status [4] and are widely deployed as part of standard clinical care. However, despite growing evidence that exercise training in cancer patients also contributes to improved quality of life and tolerance of anti-cancer treatment [5], exercise-rehabilitation has not been generally adopted for patients with cancer.

Exercise training programs for patients with chronic diseases typically involve hospital-based supervised training two or three times a week. However, attendance at hospital-based exercise training is difficult for debilitated patients with cancer; and though patients with advanced cancer are willing to exercise they strongly prefer to do so at home [6, 7]. However the beneficial impact of such home-based training is still unclear and is difficult to measure. For example, a recent study of home-based semi-supervised exercise was unsuccessful because of unexpected poor accrual and high dropout rates [8]. Outside of such studies, community or home-based supervised training is often not readily available for patients with medical conditions. Thus by default, a self-directed unsupervised home exercise program is often recommended for patients unable or unwilling to attend hospital training, despite the inherent difficulties in monitoring exercise adherence and intensity. Other approaches are therefore needed which can deliver a safe and effective exercise stimulus at home, in a manner which is suitable for use by patients with poor physical function.

Neuromuscular electrical stimulation (NMES) delivers a controlled contractile stimulus to underlying muscles via surface electrodes placed on the skin. It is widely used for rehabilitation of neurological or orthopedic conditions. NMES is often used to assist with rehabilitation following joint surgery [9], by helping to speed up the recovery of muscle strength and joint function. Moreover, NMES used in conjunction with voluntary contraction exercises, accelerates leg muscle strengthening after knee arthroplasty [10]. More recently NMES has been assessed for patients suffering chronic medical conditions. In debilitated patients with COPD or CHF, NMES of the large lower limb muscles was shown to increase muscle size [11], improve muscular strength and overall exercise capacity [12‐14], facilitate activities of daily living, and improve quality of life [15].

Despite the evidence for effectiveness of NMES in other patient populations, relatively few studies have been performed to test the potential usefulness of NMES in patients with cancer. One case study reported use of NMES in a single patient with extensive metastatic lung cancer, who achieved improvements in mobility, function and quality of life with NMES [16]. Another pilot study used NMES in a randomized trial involving 16 advanced stage non-small cell lung cancer patients with good performance status. Those receiving NMES found it highly acceptable, but there was only a non-significant trend towards improvement in leg muscle strength or other measures of endurance and habitual daily activity [17]. In a follow-up phase II study, the same authors recruited lung cancer patients undergoing first-line palliative chemotherapy and randomized them to NMES intervention (30 patients) or a control (19 patients) group [18]. In this later study there was no improvement in the physical function tests (quadriceps strength or physical activity), and a lower than expected level of adherence. As a result, the authors concluded that NMES does appear to warrant further study in advanced lung cancer patients at this stage in their treatment [18]. This does not rule out a role for NMES to reduce muscle loss and dysfunction [19] in other patients with cancer, but to date there is still little published data on the use of NMES in patients with cancer in general, especially those with poor performance status.

The McGill Cancer Nutrition-Rehabilitation Program clinic at the Jewish General Hospital (CNR-JGH clinic) in Montréal, is a multi-disciplinary team that specifically addresses the clinical needs of patients with cancer referred with weight loss, anorexia, and reduced physical function. The CNR-JGH clinic uses a comprehensive interdisciplinary approach to assess and control symptoms, along with individualized dietary interventions and exercise training wherever possible. In our experience, hospital-based exercise training is very difficult to deliver to patients with poor performance status and those who live far away from the hospital. To address this practical problem, a pilot study was performed to evaluate the acceptability and feasibility of a home-based NMES intervention in a heterogeneous group of patients attending the CNR-JGH clinic, the majority of whom had advanced stage cancer. The secondary objective of the study was to assess the impact of the NMES intervention on tests of physical function focused on walking endurance capacity, lower extremity strength and global functional performance.

111 new patients were seen in the CNR-JGH clinic between June 2010 and July 2011. Of these 52 (47%) were potentially eligible for the study (poor performance status and/or inability to attend hospital-based exercise training). After further screening, 37 patients were approached and of these 18 (49%) patients were recruited. The reasons that 19 patients were not consented for the study ranged from medical deterioration (3) or improvement (1), preference for standard exercise (4), feeling that the study was not of interest or too burdensome (6), and follow-up and communication difficulties with study coordinator (5). Of the 18 patients who gave signed consent, three withdrew from the study and did not proceed to baseline assessment (Figure 1). Reasons for early withdrawals included: rapid medical deterioration and death (n = 1), started exercising regularly at a local gym (n = 1), withdrawn due to language and communication difficulties which became evident only after consent was obtained (n = 1). The remaining 15 patients underwent baseline function testing and their demographic and disease characteristics are shown in Table 1. Two patients with good PS (PS = 1) were also included in the study as they were not exercising regularly and could not attend the usual hospital-based supervised training sessions. The majority of patients (73%) had metastatic (stage IV) disease. The most common tumour types were lung or gastrointestinal cancers, but five patients had other types of tumours including breast (n = 1), ovarian (n = 1), endometrial (n = 1) and haematological (n = 1) malignancy. One additional patient was being treated with cyclical intravenous chemotherapy for systemic amyloidosis without overt evidence of underlying malignancy. Overall 67% of the patients were undergoing active treatment with chemotherapy and 40% were taking, or had completed within the preceding 14 days, an extended (≥3 weeks’ duration) course of corticosteroids (Table 1). Results of functional testing at baseline are shown in Table 2. As a group, the patients in this study had impaired exercise capacity with mean 6 MWT of 257 m (50.2% predicted) and mean STS time of 8.0 s, which is 2.6 standard deviations above the mean expected in (i.e. slower than) healthy individuals of similar age (Table 2).

This is the first report of the use of NMES in patients with advanced cancer and poor performance status or other barriers to usual exercise training. Our results suggest that NMES is feasible and well-tolerated in this group of patients with advanced cancer who are unable to attend hospital-based exercise training. Only one patient (7%) withdrew from the study because they were unable to tolerate the NMES intervention, whereas four patients (27%) withdrew or could not complete the study, because of medical complications or death. Furthermore, qualitative feedback confirmed that home-based NMES is highly acceptable. There was a wide range in adherence with NMES over the 6-week period of intervention, but despite this, 70% of patients used the NMES for at least 40% of the available days. This equates to use at least three times a week, which is the usual frequency for standard exercise training programs. Repeated physical performance testing before and after NMES revealed wide intra-individual variability. However, there was no improvement in objective measurements of exercise function after NMES intervention (Table 2). Furthermore, no dose-effect of NMES was demonstrated, as there were no differences in physical functional outcomes, between adherent (>40% maximum usage) and poorly adherent (<40% maximum NMES usage) patients.

We recognize a number of limitations to interpreting our results, including the small number of analyzable patients (10) in this pilot study. The primary study objective was to assess feasibility and acceptability of NMES in a variety of cancer patients and for this reason the patients recruited for the study had a range of pathological types and stages of cancer. However, as a result the study participants also had a widely varying initial functional status and exercise performance, making it more difficult to reliably detect functional changes attributable to NMES. We did not include an untreated control group to assess the likely spontaneous changes in outcome measures over the same time period. In debilitated advanced cancer patients, progressive physical deterioration may be anticipated. Thus a reasonable minimal expectation of successful NMES, and other exercise interventions, would be to inhibit this deterioration, rather than achieve an improvement from baseline. Without an untreated comparison group we are unable to say whether there was any indication of this beneficial effect of NMES in the current study. Due to limited personnel resources, the same physiotherapist who instructed the patient in the use of NMES also performed the outcomes assessments. This raises the possibility of patient response bias, particularly for reported global performance status and responses to questionnaire items.

Maddocks et al. [18] reported only 15/30 lung cancer patients receiving palliative chemotherapy completed approximately 10 week NMES intervention. In contrast to the present study, they concluded that NMES was not an acceptable treatment based on a pre-defined 80% group adherence threshold [18]. However it is notable that in line with our results only 3(11%) of their patients withdrew because of NMES discomfort suggesting that the intervention, per se, is tolerable for the majority. Furthermore, without any correction for periods when patients were hospitalized etc., 50% of patients in the study by Maddocks et al. achieved the required three times a week treatment frequency, and another 3 patients fell below this threshold because of only one missed session. We would concur with the authors of that study who acknowledged that their conclusion, that NMES is “unacceptable”, may have been based on unrealistically stringent criteria [18].

Published studies of use of NMES describe a range of different electrical stimulation protocols [25] and there is no clear consensus as to the optimal stimulation frequency, electrode placement sites and duty-cycle etc. The original intention in this study was to use dual stimulation of quadriceps and gluteus muscles. However, it became clear that there were both practical and patient acceptability difficulties in repeatedly and consistently placing the gluteus muscle electrodes, especially in very sedentary patients or those requiring help from caregivers to do so. As a result a simplified protocol was used with electrical stimulation of quadriceps only, combined with active voluntary co-contraction of the stimulated limb muscles, to enhance the contractile stimulus. The use of active co-contraction is novel and proved both practical and acceptable as demonstrated by the feedback received (Table 3). Other informal comments received also suggested the patients appreciated the ability to participate in their exercise training even in this limited way.

Our study was not powered to detect differences in the physical outcome measures and many other confounding factors, including acute changes in medical status, can influence the results of the physical performance tests. The physical outcomes measures adopted (PS, STS, 6 MWT) have been widely used in studies of cancer patients and are relevant to normal daily functional activities [22]. However, direct measurements of muscle strength or cross-sectional area, have been used by other authors [11, 14, 17], and these approaches may be more sensitive to detect modest short-term improvements in muscle induced by NMES. Analysis of pooled data from several different studies of NMES in chronic medical conditions suggests NMES use leads to a mean improvement in 6 MWT of 40 m [26]. However we calculated that 127 patients would be needed to detect a difference of 40 m in 6 MWT (effect size 0.25 SD) with 80% power in any future study, without modifying the entry criteria. This is directly related to the wide variance in initial 6 MWT results. Thus stratifying analysis or selecting patients with a pre-defined baseline performance (e.g. by 6 MWT distance) would make it easier to detect subsequent changes after NMES intervention.

High drop-rate rates and the continuing uncertainty about which patients are most likely to respond to NMES physiologically, present significant challenges to successful completion of studies of NMES in cancer patients. It is noteworthy that many studies in COPD and CHF conclude that use of NMES may be best targeted to the most debilitated patients with very poor function [14]. However cancer patients who have both advanced disease and impaired functional status (e.g. 6 MWT <350 m at baseline [27]) have very poor prognosis, and even short-term studies in these patients are likely to suffer substantial dropouts [3, 8] due to medical deterioration and death. In this pilot study we were not able to confidently identify any subgroups of patients who have greater potential to complete the NMES intervention. Thus patients who did not complete the intervention had a trend towards lower mean baseline 6 MWT distance (166 m Withdrawn vs 303 m Completed, P = 0.11) (Table 2). However, neither this nor any other demographic or physical function results at baseline were significantly different between those that completed and those who were withdrawn from the study (Tables 1, 2).

One group of patients who may be more likely to benefit from NMES are those that have taken a recent prolonged course of corticosteroids, given that corticosteroids are both widely used in cancer therapy and can cause myopathy [28]. Consistent with this notion, the data from this study did confirm that those patients recently treated with steroids had poorer function at baseline (6 MWT mean(SD): recent steroids 166(139)m vs no steroids 394(88)m, P = 0.04). However there was only a non-significant trend towards a differential improvement in 6 MWT in patients treated with extended courses of steroids. Nevertheless, further studies of NMES may be warranted to establish if early use of NMES can be of particular benefit in cancer patients treated with prolonged courses of steroids. Similarly, it is conceivable that other subgroups of cancer patients should be considered for more focused studies, including the more elderly patients with comorbidities [29] and those being treated with antitumour agents, such as Sorafenib, which promote muscle loss [30].

The results of the current pilot study suggest that NMES is both feasible and acceptable in a mixed group of patients with cancer, most of whom had poor performance status. This study does not demonstrate that NMES leads to improved physical functioning in cancer patients with poor performance status despite the success of NMES in other chronic diseases and clinical settings. One reason for the difficulty in detecting functional benefits of NMES in cancer patients may be their marked heterogeneity in terms of medical status and physical functioning. As a result, though it is likely that NMES has a role in achieving or maintaining optimal functional ability in selected patients with cancer, this still remains to be confirmed by future prospective studies. Such studies should be designed to both identify the best candidates for NMES interventions and to confirm the optimal method of delivery of this type of therapeutic exercise.

TS and TW are physiotherapists who worked in the CNR-JGH team during the study. BV worked as study coordinator with the CNR-JGH team during the study. RTJ is an MD and Director of the CNR-JGH team.