三环类抗抑郁药、5-羟色胺-去甲肾上腺素再摄取抑制剂和抗惊厥药治疗糖尿病痛性神经病变的比较有效性:系统评价与荟萃分析。
完整摘要 · 中文译文
引言:糖尿病痛性神经病变(DPN)是糖尿病最常见且最令人衰弱的并发症之一,严重损害患者的生活质量。在缺乏改善疾病的疗法的情况下,治疗仍集中于使用非阿片类药物的对症疼痛控制;然而,指导最佳药物选择的比较性证据仍然有限。本研究采用系统评价、配对荟萃分析和网络荟萃分析,比较当前基于非阿片类药物的治疗(包括阿米替林、度洛西汀、加巴喷丁和普瑞巴林)在DPN患者中减轻疼痛的有效性。方法:系统检索PubMed和Google Scholar,以获取2010年至2025年间发表的评估这些药物作为DPN单药治疗的随机试验和临床试验。共有n = 11项研究符合纳入标准。使用数字评定量表(NRS)、视觉模拟量表(VAS)或简明疼痛量表(BPI)测量的疼痛结局被标准化为共同指标。进行随机效应配对荟萃分析以估计药物内治疗效果,随后进行频率学网络荟萃分析以通过直接和间接证据比较疗法。疼痛严重程度的平均变化是主要结局。我们在不同的相关性假设下进行了异质性、网络一致性和敏感性分析。结果:共纳入11项研究中的1397名参与者。配对荟萃分析显示度洛西汀可大幅减轻疼痛(平均变化[MC] 27.8;95 % CI 18.5-37.1)、加巴喷丁(MC 34.8;95% CI 9.3-60.3)和普瑞巴林(MC 19.7;95% CI 8.9-30.5)。尽管阿米替林试验各自报告了改善,但汇总估计未达到统计学显著性。网络荟萃分析表明,加巴喷丁、度洛西汀和阿米替林与普瑞巴林相比,估计的镇痛获益更好;加巴喷丁、度洛西汀和阿米替林之间的差异不大且无统计学显著性。敏感性分析支持结果的稳健性,未检测到显著的网络不一致性。结论:常用于糖尿病周围神经病变(DPN)的非阿片类药物治疗在镇痛获益上大体相当,加巴喷丁在网络分析中显示出相对更优的疼痛严重程度减轻。鉴于各药物之间差异不大且个体反应差异显著,治疗决策应根据患者合并症、耐受性和治疗目标进行个体化。需要更大规模、方法学标准化的对照试验来加强证据基础,并支持对DPN进行更精确、基于证据的管理。
完整摘要 · 英文原文
Introduction: Diabetic painful neuropathy (DPN) is one of the most common and debilitating complications of diabetes, substantially impairing the quality of life of patients. In the absence of disease-modifying therapies, treatment remains focused on symptomatic pain control with non-opioid pharmacologic agents; however, comparative evidence guiding optimal medication selection remains limited. Using a systematic review, pairwise meta-analysis, and network meta-analysis, this study compares the effectiveness of current non-opioid-based treatments, including amitriptyline, duloxetine, gabapentin, and pregabalin, for pain reduction in patients with DPN. Methods: PubMed and Google Scholar were systematically searched for randomized and clinical trials published between 2010 and 2025 evaluating these medications as monotherapy for DPN. A total of n = 11 studies met the inclusion criteria. Pain outcomes measured using the Numeric Rating Scale (NRS), Visual Analog Scale (VAS), or Brief Pain Inventory (BPI) were standardized to a common metric. Random-effects pairwise meta-analyses were performed to estimate within-drug treatment effects, followed by a frequentist network meta-analysis to compare therapies through direct and indirect evidence. Mean change in pain severity was the primary outcome. We performed heterogeneity, network consistency, and sensitivity analyses under varying correlation assumptions. Results: A total of 1397 participants across 11 studies were included. Pairwise meta-analysis demonstrated substantial pain reduction with duloxetine (mean change [MC] 27.8; 95% CI 18.5-37.1), gabapentin (MC 34.8; 95% CI 9.3-60.3), and pregabalin (MC 19.7; 95% CI 8.9-30.5). Although amitriptyline trials individually reported improvement, the pooled estimate did not reach statistical significance. Network meta-analysis indicated that gabapentin, duloxetine, and amitriptyline were associated with better estimated analgesic benefit than pregabalin; the differences among gabapentin, duloxetine, and amitriptyline were modest and not statistically significant. Sensitivity analyses supported the robustness of the findings, and no significant network inconsistency was detected. Conclusions: Commonly used non-opioid pharmacologic therapies for DPN provide broadly comparable analgesic benefit, with gabapentin showing comparatively preferable reduction in pain severity within the network analysis. Given the modest differences between agents and substantial variability in individual response, treatment decisions should be individualized according to patient comorbidities, tolerability, and therapeutic goals. Larger, methodologically standardized comparative trials are needed to strengthen the evidence base and support more precise, evidence-based management of DPN.