Ma Qin, Liu Chunyu, Zhao Guozhen, Guo Shiqi, Li Hancong, Zhang Bo, Li Bo, Cai Zhaolun
Division of Gastrointestinal Surgery, Department of General Surgery, West China Hospital, Sichuan University, Chengdu, China.
Evidence-Based Pharmacy Center, West China Second University Hospital, Sichuan University, Chengdu, China.
Cochrane Database Syst Rev. 2025 Dec 5;12(12):CD015177. doi: 10.1002/14651858.CD015177.pub2.
Insomnia is a common issue affecting people with cancer. Although acupuncture is widely used as a treatment option for insomnia, its effects on cancer patients require a rigorous and up-to-date evaluation.
To evaluate the benefits and harms of acupuncture for insomnia in people with cancer.
We searched CENTRAL, MEDLINE, Embase, PsycINFO, and five other databases or trial registries in January 2024.
We included randomised controlled trials (RCTs) with a minimum duration of four weeks that evaluated acupuncture (defined as needle insertion at specific acupoints) for treating insomnia in patients with cancer.
Our outcomes were insomnia severity measured by the Insomnia Severity Index (ISI), sleep quality measured by the Pittsburgh Sleep Quality Index (PSQI), adverse events, and sleep diary outcomes including sleep onset latency (SOL), wake after sleep onset (WASO), total sleep time (TST), and sleep efficiency (SE).
We assessed the risk of bias using the RoB 2 tool.
We performed random-effects meta-analysis to calculate risk ratios (RR) for dichotomous outcomes and mean differences (MD) for continuous outcomes, with 95% confidence intervals (CIs). We assessed the certainty of evidence with GRADE and interpreted findings for continuous outcomes against minimally important differences (MIDs).
We included five studies with 402 participants. The participants were predominantly females with breast cancer, and most were people following primary cancer treatment.
We identified three comparisons with outcomes assessed at the end of the interventions. We rated the certainty of the evidence as very low-to-moderate, mainly due to risk of bias and the imprecision of effect estimates from the small studies. Acupuncture versus sham acupuncture We are very uncertain about all results due to very low-certainty evidence. Compared to sham acupuncture, acupuncture may have little to no effect on post-intervention ISI scores (MD -3.17, 95% CI -10.39 to 4.05; MID -4.7 points; 2 studies, 152 participants; very low-certainty evidence) and PSQI scores (MD -1.16, 95% CI -3.53 to 1.22; MID -3 points; 2 studies, 152 participants; very low-certainty evidence). Acupuncture may increase the risk of adverse events (RR 2.60, 95% CI 0.98 to 6.90; 1 study, 138 participants; very low-certainty evidence), but this result is very uncertain. Regarding sleep diary outcomes, acupuncture compared with sham acupuncture may improve post-intervention SOL (MD -10.02 min, 95% CI -19.09 to -0.94; MID 20 minutes; 2 studies, 152 participants; very low-certainty evidence) and SE (MD 4.90%, 95% CI 1.98 to 7.82; MID 10%; 2 studies, 152 participants; very low-certainty evidence) very slightly. It may have a large effect on TST (MD 45.94 min, 95% CI -0.93 to 92.80; MID 15 minutes; 2 studies; 152 participants; very low-certainty evidence), but this result is very uncertain. Data on WASO were unavailable. No outcome both exceeded its MID and was statistically significant. Acupuncture versus inactive control We are very uncertain about all results due to very low-certainty evidence. Compared to an inactive control, acupuncture may reduce post-intervention ISI scores (MD -3.88, 95% CI -7.25 to -0.52; MID -4.7 points; 2 studies, 46 participants; very low-certainty evidence) and PSQI scores (-2.20, 95% CI -3.35 to -1.04; MID -3 points; 3 studies, 98 participants; very low-certainty evidence) slightly, but may increase the risk of adverse events (RR 15.49, 95% CI 2.12 to 113.10; 2 studies, 76 participants; very low-certainty evidence). With respect to sleep diary outcomes, acupuncture may slightly improve post-intervention SOL (MD -15.61 min, 95% CI -29.23 to -1.99; MID 20 minutes; 2 studies, 46 participants; very low-certainty evidence), TST (MD 34.61 min, 95% CI 12.54 to 56.69; MID 15 minutes; 2 studies, 46 participants; very low-certainty evidence) and SE slightly (MD 5.65, 95% CI 0.99 to 10.32; MID 10%; 2 studies, 46 participants; very low-certainty evidence). However, it may result in little to no difference in post-intervention WASO (MD 5.70 min, 95% CI -17.25 to 28.65; 1 study, 30 participants; very low-certainty evidence). Only the TST improvement surpassed the MID. Acupuncture versus cognitive behavioural therapy for insomnia (CBT-I) Compared to CBT-I, acupuncture probably results in slightly higher (worse) post-intervention ISI scores (MD 2.60, 95% CI 1.13 to 4.07; 1 study, 160 participants; moderate-certainty evidence) and PSQI scores (MD 1.51, 95% CI 0.51 to 2.51; 1 study, 160 participants; moderate-certainty evidence). However, it may have little to no effect on adverse events (RR 1.68, 95% CI 0.59 to 4.79; 1 study; 160 participants; low-certainty evidence). Regarding sleep diary outcomes, acupuncture compared with CBT-I probably slightly worsens post-intervention SOL (MD 16.33 min, 95% CI 8.22 to 24.44; MID 10 minutes; 1 study, 160 participants; moderate-certainty evidence) and SE (MD -5.00%, 95% CI -8.48 to -1.52; MID 5%; 1 study, 160 participants; moderate-certainty evidence) but probably increases TST (MD 26.80 min, 95% CI 3.87 to 49.73; MID 15 minutes; 1 study, 160 participants; moderate-certainty evidence). It probably has little to no effect on WASO (MD 8.94 min, 95% CI -1.47 to 19.35; MID 15 minutes; 1 study, 160 participants; moderate-certainty evidence). The effects on SOL, TST, and SE reached the MIDs.
AUTHORS' CONCLUSIONS: Based on very low-certainty evidence, acupuncture may have little to no effect on insomnia severity or sleep quality compared to sham acupuncture, though it may offer slight improvements in some sleep diary metrics. In contrast, when compared to an inactive control, acupuncture may alleviate insomnia severity and improve sleep quality and most sleep diary metrics, but adverse events should be taken into consideration. These findings are derived primarily from studies of female adults with breast cancer. Based on low- to moderate-certainty evidence, when compared with CBT-I, acupuncture is likely less effective at reducing insomnia severity, improving sleep quality, SOL and SE. Conversely, acupuncture probably improves TST. Larger, methodologically robust, long-term trials that include diverse cancer populations are required to provide definitive conclusions.
This work was funded by the Postdoctor Research Fund of West China Hospital, Sichuan University (2025HXBH063) and the Fundamental Research Fund of China Academy of Chinese Medical Sciences (No. ZZ17-XRZ-113).
Protocol available via doi.org/10.1002/14651858.CD015177.
失眠是影响癌症患者的常见问题。尽管针灸被广泛用作失眠的治疗选择，但其对癌症患者的疗效需要进行严格且最新的评估。
评估针灸治疗癌症患者失眠的益处和危害。
我们于2024年1月检索了Cochrane系统评价数据库、医学期刊数据库、荷兰医学文摘数据库、心理学文摘数据库以及其他五个数据库或试验注册库。
我们纳入了至少为期四周的随机对照试验（RCT），这些试验评估了针灸（定义为在特定穴位进针）治疗癌症患者失眠的效果。
我们的结局指标包括通过失眠严重程度指数（ISI）测量的失眠严重程度、通过匹兹堡睡眠质量指数（PSQI）测量的睡眠质量、不良事件以及睡眠日记结局，包括入睡潜伏期（SOL）、睡眠中觉醒时间（WASO）、总睡眠时间（TST）和睡眠效率（SE）。
我们使用RoB 2工具评估偏倚风险。
我们进行随机效应荟萃分析，以计算二分结局的风险比（RR）和连续结局的平均差（MD），并给出95%置信区间（CI）。我们使用GRADE评估证据的确定性，并根据最小重要差异（MID）解释连续结局的研究结果。
我们纳入了五项研究，共402名参与者。参与者主要为患有乳腺癌的女性，且大多数为接受原发性癌症治疗后的患者。
我们确定了三项在干预结束时评估结局的比较。我们将证据的确定性评为非常低到中等，主要是由于偏倚风险以及小型研究中效应估计的不精确性。针灸与假针灸相比 由于证据确定性非常低，我们对所有结果都非常不确定。与假针灸相比，针灸对干预后ISI评分（MD -3.17，95% CI -10.39至4.05；MID -4.7分；2项研究，152名参与者；证据确定性非常低）和PSQI评分（MD -1.16，95% CI -3.53至1.22；MID -3分；2项研究，152名参与者；证据确定性非常低）可能几乎没有影响。针灸可能会增加不良事件的风险（RR 2.60，95% CI 0.98至6.90；1项研究，138名参与者；证据确定性非常低），但这个结果非常不确定。关于睡眠日记结局，与假针灸相比，针灸可能会非常轻微地改善干预后的SOL（MD -10.02分钟，95% CI -19.09至-0.94；MID 20分钟；2项研究，152名参与者；证据确定性非常低）和SE（MD 4.90%，95% CI 1.98至7.82；MID 10%；2项研究，152名参与者；证据确定性非常低）。它可能对TST有较大影响（MD 45.94分钟，95% CI -0.93至92.80；MID 15分钟；2项研究；152名参与者；证据确定性非常低），但这个结果非常不确定。关于WASO的数据不可用。没有任何结局既超过其MID又具有统计学意义。针灸与非活性对照相比 由于证据确定性非常低，我们对所有结果都非常不确定。与非活性对照相比，针灸可能会轻微降低干预后的ISI评分（MD -3.88，95% CI -7.25至-0.52；MID -4.7分；2项研究，46名参与者；证据确定性非常低）和PSQI评分（-2.20，95% CI -3.35至-1.04；MID -3分；3项研究，98名参与者；证据确定性非常低），但可能会增加不良事件的风险（RR 15.49，95% CI 2.12至113.10；2项研究，76名参与者；证据确定性非常低）。关于睡眠日记结局，针灸可能会轻微改善干预后的SOL（MD -15.61分钟，95% CI -29.23至-1.99；MID 20分钟；2项研究，46名参与者；证据确定性非常低）、TST（MD 34.61分钟，95% CI 12.54至56.69；MID 15分钟；2项研究，46名参与者；证据确定性非常低）和SE（MD 5.65，95% CI 0.99至10.32；MID 10%；2项研究，46名参与者；证据确定性非常低）。然而，它可能导致干预后WASO几乎没有差异（MD 5.70分钟，95% CI -17.25至28.65；1项研究，30名参与者；证据确定性非常低）。只有TST的改善超过了MID。针灸与失眠认知行为疗法（CBT-I）相比 与CBT-I相比，针灸可能会导致干预后ISI评分（MD 2.60，95% CI 1.13至4.07；1项研究，160名参与者；中等确定性证据）和PSQI评分（MD 1.51，95% CI 0.51至2.51；1项研究，160名参与者；中等确定性证据）略高（更差）。然而，它可能对不良事件几乎没有影响（RR 1.68，95% CI 0.59至4.79；1项研究；160名参与者；低确定性证据）。关于睡眠日记结局，与CBT-I相比，针灸可能会使干预后的SOL（MD 16.33分钟，95% CI 8.22至24.44；MID 10分钟；1项研究，160名参与者；中等确定性证据）和SE（MD -5.00%，95% CI -8.48至-1.52；MID 5%；1项研究，160名参与者；中等确定性证据）略有恶化，但可能会增加TST（MD 26.80分钟，95% CI 3.87至49.73；MID 15分钟；1项研究，160名参与者；中等确定性证据）。它可能对WASO几乎没有影响（MD 8.94分钟，95% CI -1.47至19.35；MID 15分钟；1项研究，160名参与者；中等确定性证据）。对SOL、TST和SE的影响达到了MID。
基于非常低确定性的证据，与假针灸相比，针灸对失眠严重程度或睡眠质量可能几乎没有影响，尽管它可能会在一些睡眠日记指标上略有改善。相比之下，与非活性对照相比，针灸可能会减轻失眠严重程度，改善睡眠质量和大多数睡眠日记指标，但应考虑不良事件。这些发现主要来自对成年乳腺癌女性的研究。基于低到中等确定性的证据，与CBT-I相比，针灸在降低失眠严重程度、改善睡眠质量、SOL和SE方面可能效果较差。相反，针灸可能会改善TST。需要开展更大规模、方法学严谨的长期试验，纳入不同的癌症人群，以得出明确的结论。
本研究由四川大学华西医院博士后科研基金（2025HXBH063）和中国中医科学院基本科研业务费（编号ZZ17-XRZ-113）资助。
方案可通过doi.org/10.1002/14651858.CD015177获取。
