Is Electroacupuncture Contraindicated for Preventing and Treating Chemotherapy-Induced Peripheral Neuropathy?

Abstract

Background:

Chemotherapy-induced peripheral neuropathy (CIPN) is a common side-effect of several drugs used to combat cancer. Thus, researchers have sought better treatments for and prevention of CIPN, such as electroacupuncture (EA). Some trials show EA worsens or prolongs CIPN pain and recommend against further studies on this. This narrative review explores EA for preventing or treating CIPN, comparing positive and negative outcomes.

Methods:

PubMed, ScienceDirect, and Google Scholar were searched for electroacupuncture, CIPN, and peripheral neuropathy. A snowballing method was used to find systematic reviews and studies in systematic reviews.

Results:

Seven English-language trials were found on using EA for preventing or treating CIPN. In 3 prevention studies, 1 had significant benefits, 1 had modest benefits, and 1 had worse pain in an EA group at follow-up, compared to sham controls. In 4 treatment studies, 2 had significant benefits, 1 had no difference from 3 controls, and 1 had sham control was superior to verum EA.

Conclusions:

Most of the studies were limited by small sample sizes, and some studies used EA protocols and treatment doses (frequency and total number of sessions) that were potentially suboptimal. The quantity and quality of the studies are insufficient to draw firm conclusions on effectiveness and safety. More studies must test optimal EA protocols and treatment dosages. It is inappropriate to say that EA is not recommended for CIPN prevention or treatment, because there is no robust evidence to justify this. Generally, research has found benefits and no harms.

INTRODUCTION

Chemotherapy-induced peripheral neuropathy (CIPN) is a common side- effect of several cytotoxic drugs used to treat cancer. CIPN is often referred to as a “dose-limiting” side-effect, as the symptoms caused by this disorder can be severe enough to necessitate a dose reduction or even suspension of chemotherapy, thus, negatively impacting treatment success.¹ While CIPN symptoms may resolve in some patients after cessation of chemotherapy, in others, those symptoms may linger for years or even be irreversible. Those symptoms include pain but also numbness and tingling with loss of nerve functioning leading to increased falls, gait and writing difficulty, and a host of complications limiting quality of life.²

While some medications provide a measure of relief from CIPN-related pain, those medications also carry risks of side-effects, and there are no approved treatments that lead to nerve regeneration and reduction of the full range of CIPN symptoms.³ The combination of CIPN's dose-limiting potential and lack of effective treatment to either prevent CIPN or its long-lasting impact on quality of life (QoL) has led to much research looking for better treatment and prevention options.

Over the past 25 years, numerous studies have been conducted to measure the effectiveness and safety of acupuncture or electroacupuncture (EA) to reduce the symptoms in postchemotherapy CIPN and, more recently, CIPN prevention. This narrative review examines conflicting evidence on EA for CIPN prevention and treatment with some studies reporting beneficial outcomes, while others reporting either no changes or even exacerbation of pain.

METHODS

An initial search was conducted on PubMed, ScienceDirect, and Google Scholar, using the search terms electroacupuncture, CIPN, and peripheral neuropathy. Then a snowballing method was adopted to search for the systematic reviews in a review of systematic reviews and for individual studies that were included in systematic reviews. A total of 7 relevant trials in the English language were identified, 3 of which related to prevention of CIPN with EA and 4 on the treatment of CIPN with EA.^4–10

RESULTS

CIPN Prevention Studies

In addition to CIPN being potentially dose-limiting, it is also “dose-dependent” (i.e., the severity of neuronal damage and CIPN symptoms generally increase with higher chemotherapy dosage). Hence, any slowing of the onset of CIPN symptoms or reduction in severity would be evidence of some form of a neuroprotective effect of EA.

Lu et al. conducted a substudy nested within a larger trial involving 21 women with ovarian cancer receiving chemotherapy.⁴ The substudy measured QoL scores in women receiving either manual acupuncture and EA or sham EA at nonpoints connected to an inactive electrostimulator. Ten treatments were given, 2–3 times weekly over 4 weeks, during active chemotherapy beginning 1 week before the second cycle of chemotherapy. QoL subscores for social function, pain, and insomnia (measured with the EORTC-QLQ-C30 [European Organization for Research and Treatment of Cancer-Quality-of-Life Questionnaire-Core 30 Item]) improved in the group receiving acupuncture/EA only; however, after adjustment for baseline differences, only social function improvement was significant. There were no significant differences between the groups in ovarian cancer–specific QoL scores (QLQ-OV28[The Quality of Life Questionnaire-Ovarian Cancer Module-28 Item]).

Greenlee et al. conducted a pilot randomized trial of EA versus sham EA during active chemotherapy to investigate if EA exerted a neuroprotective effect.⁵ Patients with breast cancer undergoing taxane chemotherapy were given either EA, 2 Hz for 30 minutes weekly, for 12 weeks or sham EA using Park sham device needles on nonpoints attached to an inactivated electrostimulator. There was no difference between the groups in worst pain at week 12. At week 16 (a fourth-week follow-up), the sham group had improved compared to week 12, but the EA group had worse pain than at week 12. This led the researchers to express concern that the EA may have made the participants' pain worse, and the researchers even went so far as to recommend against further study of EA for CIPN prevention, stating that further EA studies should focus only on treatment for established CIPN symptoms.

A subsequent study by Zhang et al compared manual acupuncture to EA for prevention of CIPN during chemotherapy.⁶ In this study, no exacerbation of pain was observed in the EA group, and both the manual acupuncture and EA groups improved in both pain and CIPN symptoms, with EA being superior to manual acupuncture.

CIPN Treatment Studies

In the 4 studies on EA treatment of established CIPN, 2 studies reported beneficial changes pre- and post-treatment with EA and between EA and controls.^8,9 In the other 2 studies, 1 found no difference between EA and sham EA control, while the other study reported that sham EA control was superior to verum EA for reducing pain. The researchers expressed concern that EA may have maintained neuropathic pain, and recommended further studies should use manual rather than EA.^7,10

DISCUSSION

Is There Sufficient Evidence to Justify a Contraindication for EA for CIPN?: The Evidence

There are 3 trials related to EA for CIPN that could be regarded as having negative outcomes: 1 finding no difference between verum EA and sham, 1 questioning if verum EA may have prolonged neuropathic pain, and 1 reporting a worsening of neuropathic pain, compared to sham EA.^5,7,10

In Rostock et al., a very small pilot study (EA group: n = 14), EA showed no difference from 3 controls—hydroelectric baths, vitamin B/vitamin B₆, and placebo capsules.⁷ The researchers acknowledged that other studies on acupuncture for CIPN had positive findings and that several factors, including inadequate treatment dose, may have been confounding factors in this study. They cited a Schröder et al. 2007 study that advised that it can take up to 10 weeks of acupuncture to see measurable results in nerve-conduction studies.¹¹ In Rostock et al. study, participants received 8 + 1 treatments over 3 weeks.⁷

Rostock et al. also stated, given that the goal of stable success for treating CIPN involves inducing neuroregeneration and that this can take from “3 to 6 months,” the effects of treatment would take longer than “in other indications of acupuncture like pain or vegetative imbalance.”⁷ These researchers also emphasized that “our results only indicate that our particular standardized acupuncture protocol might not be effective in the treatment of CIPN, but the results cannot be generalized to other acupuncture concepts.” ⁷ This warning makes clear that, while this study would be appropriate to include in a review investigating different EA protocols for treating CIPN, these results should not be included in systematic reviews investigating the larger question of acupuncture's effectiveness for treating CIPN. The electrostimulation parameters may also have been problematic. This study used 15 minutes of EA (50Hz) “consisting of a combination of rectangular currents and high amplitude waves.” ⁷

Hammond et al, in another very small pilot trial (EA: n = 10) found that the Streitberger sham group was superior to the EA group for reducing pain.¹⁰ The declarative title of this study was “Electro Acupuncture Is Not Recommended for Managing Chronic Neuropathic Pain in Chemotherapy Induced Peripheral Neuropathy: A Double Blind Randomized Controlled Trial.”¹⁰ Curiously, this statement is not based on the outcomes of the study, which clearly showed improvements in both groups, so the EA group did not suffer harm.

While the researchers cited their small sample size as a limitation of the study, unlike Rostock,⁷ they did not cite their low treatment dosage as a limitation.¹⁰ In addition, while the researchers did point out that their potentially negative findings were based on true EA “at these stimulation and treatment parameters,” one would not glean that limitation from the study's title. Again, with such a small sample size, only 6 weekly treatments, and a noninert sham, it is difficult to see how this study would justify such a negative treatment recommendation. ”Not recommended” would generally suggest not merely a relative lack of effectiveness but actual harms, when none were reported.

Greenlee et al. used a larger sample size (EA: n = 36), although the researchers described the work as a pilot study.⁵ The EA stimulation parameters (2 Hz for 30 minutes) were appropriate and the total treatment sessions (12) may have been adequate, although weekly treatments may not have been enough to achieve EA's maximum potential benefit. The goal of the study was to stay ahead of the developing CIPN as the chemotherapy course progressed. Both groups had increased worst-pain scores at week 12, suggesting that perhaps future studies might try more-frequent treatments. What raised concerns was that, at week 16 (4 weeks post-treatment), the sham group's worst pain-scores had reduced but the EA group's worst-pain scores had increased.

Clinicans suggest, when treating CIPN, there may be a phase of neuroregeneration when pain sensitivity increases before settling back down like a foot that has “fallen asleep” starting to reawaken. An exacerbation of pain at week 16 may have been part of the normal course of nerve regeneration, and hence should not be necessarily interpreted as a negative outcome. When assessing the possibility of nerve regeneration in CIPN, it is important that pain is interpreted within the context of the normal course of nerve regeneration. This includes follow-up timeframes that capture the whole picture of neuroregeneration over 3–6 months.

Problems With Sham/Placebo Acupuncture Protocols

In Hammond et al.'s study, the sham protocol (described as “placebo acupuncture”) involved Streitberger “placebo needle” devices placed at the same acupuncture points as used in the verum EA group, but attached to an inactivated electrostimulator.¹⁰ Greenlee et al. used the Park sham device but on nonpoints.⁵ It has been been shown that “placebo needle” devices, such as the Streitberger, Park, and Takakura, are not inert interventions.¹² Hammond et al. appear to have made the assumption that Streitberger device was inert, hence, all changes in the control group were attributed to a placebo response.¹⁰ “The evidence suggests that EA does not provide superior analgesia compared to placebo acupuncture and may reduce the placebo response.”¹⁰ Given that the Streiberger needle device has been shown to be an active intervention, not inert, this study was essentially a comparison between EA and nonpenetrating manual acupuncture at the same acupuncture points.

EA Stimulation Parameters and Treatment Dose

EA frequencies used in trials are generally based on previous research that has shown that specific frequencies and stimulation durations are associated with production and release of specific endogenous opioids. Namely, met-enkephalin and endomorphins are produced after 3–7 minutes of low-frequency stimulation (2–5 Hz), ß-endorphin is produced after 20 minutes of low frequency stimulation (2–5 Hz), while dynorphins are produced after 20 minutes of high-frequency stimulation (100+ Hz).¹³ Using the optimal frequency and duration of stimulation could be essential for producing the desired clinical effects.

In 2012, Lu et al. used 30 minutes of EA at 20–25 Hz, which would still have had some effect on the low-frequency opioids, but was not optimal.⁴ Rostock et al. used 15 minutes of EA (50 Hz) “consisting of a combination of rectangular currents and high amplitude waves.”⁷ This was a suboptimal frequency, and stimulation was too short to produce either ß-endorphin or dynorphins. Neither of these studies produced any significant differences apart from the social function score in Lu et al.⁴ In contrast, the 2 studies on treatment of CIPN that reported beneficial outcomes both used low-frequency stimulation (2–10 Hz), and stimulation duration was a minimum of 20 minutes in both studies.^8,9

CONCLUSIONS

Table 1 summarizes the results of the studies (see Table 1). Outcomes from studies to date suggest that, in order to optimize production and release of endogenous opioids, EA for CIPN should use either low-frequency (2–10 Hz) or cycling from low- to high-frequency (2–10 Hz/100+ Hz)—the so-called “dense–disperse” waveform. Electrostimulation should be delivered for a minimum of 20 minutes, with 30 minutes probably preferred.

Table 1.

Summary of Electroacupuncture trials for Chemotherapy-Induced Peripheral Neuropathy

Study 1st author, yr, and ref.	Condition	Study type	EA	Treatment dose	# of subjects	Outcomes
Lu 20124	CIPN (ovarian CA) (carboplatin & paclitaxel)	Substudy: Effects of acup/EA on QoL during active chemo Acup & EA vs sham (nonpoints)	20–25 Hz 30 min	2–3x/wk 4 wks Total: 10	n = 21 11:10	EORTC-QLQ-C30—social function improvement in Acup/EA group QLQ-OV28—no significant differences
Rostock 20137	CIPN (mixed CA types & chemo types)	4-arm RCT (pilot): EA vs HBs vs vit B1/B6 vs placebo capsules	50 Hz 15 min	3 wks Total: 8 + 1	n = 59 14:14:15:17	NRS, NCS (median & sural nerves), NCI-CTC, EORTC QLQ-C30 EA, HB, vit B1/B6—similar effects to placebo
Greenlee 20165	CIPN (breast CA) (taxane)	RCT (pilot): EA vs Park sham on nonpoints	2 Hz 30 min	1x/wk 12 wks Total 12	n = 63 31:32	BPI-SF–worst pain score increased at wk 12: EA 2.6 vs sham 2.8; at wk 16: EA 3.40 vs sham 1.7 FACT-NTX, FACT-TAX, biothesiometer & grooved pegboard tests—no differences
Zhang 20176	CIPN (malignant tumors; chemo not detailed)	RCT: Manual acupuncture vs EA	2–100 Hz 30 min	7x/wk 1 wk for 2 chemo courses Total 14	n = 40 20:20	KPS, TCM S&S, NK cells, CD3+, CD4+, CD8+, & CD4+/CD8+—EA & manual acupuncture both improved KPS but no effect on immune cells; EA superior to manual acupuncture for improving QoL, KPS, CIPN
Lu 20199	CIPN (breast CA) (taxane)	RCT (pilot): EA vs usual care Manual acup first week then EA	2–10 Hz 30 min	8 wks Total 9	n = 40 20:20	PNQ sensory score—(−1.0 ± 0.9 vs −0.3 ± 0.6; p = 0.01) FACT-NTX summary score—(8.7+ 8.9 vs 1.2 ± 5.4; p = 0.002) BPI-SF pain severity score—(−1.1 ± 1.7 vs 0.3 ± 1.5; p = 0.03)
Bao 20208	CIPN (breast & colorectal CA)	RCT (pilot): EA vs sham vs usual care Manual body acup + ear acup + EA vs sham (nonpoints)	2–5 Hz 20 min	6 wks	n = 75 27:24:24	NRS week 8—verum −1.75; sham −0.91; usual −0.19 NRS week 12—verum −1.74; sham −0.34
Hammond 202010	CIPN in primary breast CA (taxane)	EA vs Streitberger sham: With inactivated EA on same points	2 Hz 30 min	1x/wk 6 weeks Total 6	n = 19 10:9	Primary: NPRS, S-LANSS Secondary: QST EA—NPRS 5.0 (4.75-6.0) to 4.25 (3.25-5.0) sham—NPRS 5.5 (3.5-7.75) to 2.5 (2.0-3.0) S-LANSS, QST—no change

yr, year(s); EA, electroacupunture; CIPN, chemotherapy-induced peripheral neuropathy; CA, cancer; QoL, quality of life; chemo, chemotherapy; acup, acupuncture; min, minutes; wk(s), week(s); EORTC-QLQ-C30; European Organization for Research and Treatment of Cancer-Quality-of-Life Questionnaire-Core 30 Item; QLQ-OV28, Quality of Life Questionnaire-Ovarian Cancer Module-28 Item; RCT, randomized controlled trial; HB, hydroelectric bath; vit, vitamin; NRS, numeric rating scale; NCS, nerve-conduction studies; NCI-CTC, National Cancer Institutz common toxicity criteria; BPI-SF, Brief Pain Inventory—Short Form; FACT-NTX, Functional Assessment of Cancer Therapy—Neurotoxicity subscale; FACT-TAX, Functional Assessment of Cancer Therapy—Taxane neurotoxicity; KPS; Karnofsky Performance Status; TCM S&S, Traditonal Chinese Medicine signs and symptoms; NK, natural killler; PNQ, Patient Neurotoxicity Questionnaire; NPRS, numeric pain-rating scale; S-LANSS, Self-report version of the Leeds Assessment for Neuropathic Symptoms and Signs; QST, quantitative sensory testing.

Optimal treatment dose of EA for CIPN has not yet been defined, however, there is some limited evidence to date that suggests that adequate frequency and duration of treatments may have an important influence on treatment outcomes in general.¹⁴ Successful studies on CIPN indicate that treatments should probably be more than once (perhaps 2–3 times) weekly, if given during active chemotherapy. For established CIPN symptoms, that same treatment frequency should be carried out over at least 10 weeks of time.

Follow-up times of 3 and 6 months are recommended to assess nerve regeneration realistically. This could be practically difficult, as some participants may receive subsequent cycles of chemotherapy during this follow-up time.

Further comparative studies comparing EA with manual acupuncture are needed to assess the relative merits of each intervention.

Future research into possible interactions between endogenous opioids and neurotrophins (which are essential for nerve regeneration) may provide further insights into the mechanisms of EA and manual acupuncture for treatment of CIPN and other forms of peripheral neuropathy.

No exacerbations of pain have been reported in the numerous studies of manual acupuncture for CIPN, and, in the EA studies on CIPN, only 1 study reported worse pain at week 16.¹⁰ This finding may have been reflecting the normal course of nerve regeneration.

It is unfortunate that the authors of a single, small pilot study declared that EA is “not recommended” for CIPN, contrary to the findings within this study and contradicting the findings of other studies that found significant benefits without harms. The inclusion of such studies in systematic reviews has the potential to introduce a negative bias. Consequently, that study should be excluded from systematic reviews on the grounds that the study's title does not reflect the findings of the study accurately.

AUTHORs' CONTRIBUTIONS

Drs. McDonald and Bauer contributed equally to writing this review.

AUTHOR DISCLOSURE STATEMENT

No financial conflicts of interest exist.

FUNDING INFORMATION

No funding was provided for work on this article.