Table 1_Time-dependent comparative efficacy of non-surgical treatments for pain relief in lateral epicondylitis: a systematic review and network meta-analysis.docx

BackgroundLateral epicondylitis (LE) is a common tendinopathy, but the relative pain-relieving efficacy of different non-surgical interventions across follow-up stages remains unclear. This study used a Bayesian network meta-analysis to compare the time-dependent effects of multiple non-surgical treatments for pain relief in LE.
MethodsPubMed, Web of Science, and the Cochrane Central Register of Controlled Trials were systematically searched from database inception to March 5, 2024, with an updated search conducted to February 2, 2026. Randomized controlled trials involving adults with LE who received non-surgical interventions were included. Pain intensity measured by the visual analog scale (VAS) was the primary outcome, and all VAS scores were standardized to a 0–10 scale, with lower scores indicating less pain. Post-treatment VAS scores were synthesized in three predefined time windows: short-term (1–4 weeks; the result closest to 4 weeks), intermediate-term (4–12 weeks; the result closest to 12 weeks), and long-term (>12 weeks; the longest follow-up beyond 12 weeks). A Bayesian random-effects network meta-analysis was performed. Treatment effects were expressed as mean differences (MDs) with 95% credible intervals (CrIs), and ranking probabilities were summarized using the surface under the cumulative ranking curve (SUCRA).
ResultsA total of 27 randomized controlled trials were included. In the short term, kinesio taping (KT), corticosteroid injection (CSI), brace, and laser therapy (LA) showed superior pain relief compared with placebo; the MDs (95%CrIs) for KT and CSI were -4.10 (-6.14 to -2.11) and -3.57 (-5.71 to -1.47), respectively. In the intermediate term, CSI, extracorporeal shock wave therapy (ESWT), glycosaminoglycan polysulfate (GAGPS), KT, physical therapy (PT), pulsed ultrasound (PU), and ultrasound (US) were superior to placebo; the MDs (95%CrIs) for KT and CSI were -2.58 (-3.92 to -1.32) and -1.60 (-2.77 to -0.44), respectively. In the long term, no intervention showed a statistically significant advantage over placebo. Although KT and CSI ranked relatively high in the short term, and GAGPS and KT ranked relatively high in the intermediate term, ranking results should be interpreted cautiously in light of interval width, direct evidence, and network consistency.
ConclusionThe pain-relieving effects of non-surgical interventions for LE appear to be time-dependent. Some treatments may be more favorable for short- or intermediate-term pain relief, but evidence for long-term superiority remains insufficient. Clinical interpretation should not rely on treatment ranking alone, but should instead integrate the follow-up stage, effect estimates, and evidence certainty. More high-quality randomized controlled trials with long-term follow-up are needed to clarify the long-term value of different non-surgical treatment strategies.

背景 肱骨外上髁炎（Lateral epicondylitis, LE）是一种常见肌腱病，但不同非手术干预措施在各随访阶段的相对镇痛疗效仍不明确。本研究采用贝叶斯网络meta分析（Bayesian network meta-analysis），比较多种非手术治疗方案用于肱骨外上髁炎患者镇痛的时间依赖性效应。
方法 系统检索PubMed、Web of Science及Cochrane对照试验中心注册库（Cochrane Central Register of Controlled Trials）自建库至2024年3月5日的文献，并于2026年2月2日完成更新检索。纳入针对接受非手术干预的成人肱骨外上髁炎患者的随机对照试验（randomized controlled trial, RCT）。以视觉模拟评分法（visual analog scale, VAS）评估的疼痛强度为主要结局指标，所有VAS评分均标准化为0~10分制，评分越低代表疼痛程度越轻。按预设的三个时间窗对治疗后VAS评分进行汇总：短期（1~4周，取最接近4周的随访结果）、中期（4~12周，取最接近12周的随访结果）及长期（>12周，取超过12周的最长随访结果）。采用贝叶斯随机效应网络meta分析进行统计分析。治疗效应以均差（mean differences, MDs）及95%可信区间（95% credible intervals, CrIs）表示，排序概率采用累积排序曲线下面积（surface under the cumulative ranking curve, SUCRA）进行汇总分析。
结果 共纳入27项随机对照试验。短期随访时，肌内效贴布（kinesio taping, KT）、糖皮质激素注射（corticosteroid injection, CSI）、支具及激光疗法（laser therapy, LA）的镇痛效果均优于安慰剂；肌内效贴布与糖皮质激素注射的均差（95%CrIs）分别为-4.10（-6.14~-2.11）与-3.57（-5.71~-1.47）。中期随访时，糖皮质激素注射、体外冲击波疗法（extracorporeal shock wave therapy, ESWT）、硫酸氨基多糖（glycosaminoglycan polysulfate, GAGPS）、肌内效贴布、物理治疗（physical therapy, PT）、脉冲超声（pulsed ultrasound, PU）及超声疗法（ultrasound, US）的镇痛效果均优于安慰剂；肌内效贴布与糖皮质激素注射的均差（95%CrIs）分别为-2.58（-3.92~-1.32）与-1.60（-2.77~-0.44）。长期随访时，无任何干预措施的镇痛效果具有统计学意义上的优势。尽管肌内效贴布与糖皮质激素注射在短期随访中排名相对靠前，硫酸氨基多糖与肌内效贴布在中期随访中排名相对靠前，但鉴于区间宽度、直接证据及网络一致性，对排名结果的解读应谨慎。
结论 肱骨外上髁炎非手术干预的镇痛效应呈现时间依赖性。部分治疗方案更适用于短期或中期镇痛，但长期疗效优于安慰剂的证据仍不充分。临床解读不应仅依赖治疗排名，而应综合考虑随访阶段、效应估计值及证据确定性。未来需开展更多高质量、带长期随访的随机对照试验，以明确不同非手术治疗策略的长期价值。