关节内注射富血小板血浆治疗膝骨关节炎的Meta分析

Intra-articular injections of platelet-rich plasma for treatment of knee osteoarthritis: systemic review of randomized controlled trials

目的评估关节内注射富血小板血浆（PRP）治疗膝骨关节炎的疗效和安全性。方法以“血小板”、“膝”、“骨关节炎”和“随机对照试验”为任意词进行系统检索，英文数据库包括Pubmed、Ovid、ScienceDirect和Cochranelibrary，中文数据库包括中国期刊全文数据库、万方数据库和维普数据库。制定文献纳入和排除标准，并由2名研究员独立对文献进行数据摘录和质量评价。根据Cochrane系统评价员手册，对纳入文献进行分组，并做定性描述和定量Meta分析。采用逆方差方统计学方法，比较膝关节评分在PRP治疗前、后的变化和在PRP与其它保守治疗方法之间的差异。针对PRP与其它保守治疗方法之间发生不良反应的风险比的比较则采用Mantel—Haenszel统计学方法。结果共纳入10项研究，959例患者，随访时间从治疗后1—12个月。PRP治疗后1、2、3、4、6和12个月的膝关节评分均显著低于治疗前，差异有统计学意义（1个月：SMD=0．85，P〈0．05；2个月：SMD=0．97，P〈0．05；3个月：SMD=1．54，P〈0．01；4个月：SMD=1．21，P〈0．01；6个月：SMD=1．46，P〈0．01；12个月：SMD=1．47，P〈0．01）。仅有2项研究以生理盐水为安慰剂对照，在治疗后6个月，PRP组的膝关节评分均优于安慰剂组（SMD=1．59），差异有统计学意义（P=0．01），而PRP组的不良反应风险比与安慰剂组的差异无统计学意义（RR=2．99，P〉0．05）。7项研究比较了PRP与透明质酸治疗膝骨关节炎的疗效，在治疗后6个月PRP组的膝关节评分优于透明质酸，差异具有统计学意义（SMD=0．85，P〈0．05）。除1项研究认为PRP组不良反应高于透明质酸组外，其他4项研究合并分析的结果是两组的不良反应风险比的差异不具有统计学意义（RR=1．07，P〉0．05）。结论关节内注射PRP治疗膝骨关节炎是一种安全、有效的治疗方法。

Objective To evaluate the efficacy and safety of intra-articular injections of platelet- rich plasma （PRP） in the treatment of knee osteoarthritis. Methods A literature search was performed using ＂platelet＂, ＂knee＂, ＂osteoarthritis＂ and ＂randomized controlled trials （RCTs）＂ as the text words. The English database included Pubmed, Ovid, SeienceDirect and the Cochrane library. The Chinese database included China National Knowledge Infrastructure, Wanfang database and VIP database. Literatures were screened according to the inclusion and exclusion criteria. Data extraction and quality evaluation of the studies included for the final analysis were independently performed by two researchers. With the guidance of the Cochrane Handbook, the studies were stratified and then synthesized for qualitative and quantitative analysis. A meta-analysis using the inverse variance method was performed to compare the knee scores between the pre-treatment and post-treatment of PRP and between the PRP treatment and other conservative methods. The Mantel-Haenszel statistical method was adopted to compare the risk ratio of adverse events between PRP injections and other conservative methods. Results A total of 959 patients in 10 RCTs were included, the follow-up time ranging from one month to 12 months. Intra- articular injections of PRP obtained significantly favorable knee scores at the post-treatment 1^st month （SMD=0.85, P〈0.05）, 2^nd month （SMD=0.97, P〈0.05）, 3^rd month （SMD=11.54, P〈0.01）, 4^th month （SMD = 1.21, P 〈0. 01）, 6^th month （SMD = 1.46, P 〈0. 01 ） and 12^th month （SMD = 1.47, P 〈 0. 01 ）. Two studies comparing PRP with a placebo group treated with saline found that PRP was significantly superior to the placebo in terms of knee scores at 6 months （SMD = 1.59, P = 0. 01 ） , whereas no significant difference was found in the post-injection adverse events between the PRP and placebo groups （RR = 2. 99, P 〉 0. 05）. Seven trials compared the efficacy and safety of PRP with hyaluronic acid. The post-injection knee scores were significantly better in the PRP-treated patients than those in the hyaluronic acid group in 6^th month （SMD = 0. 85, P 〈 0. 05）. In terms of the adverse events, no significant difference was found among the four trials （ RR = 1. 07, P 〉 0. 05 ）. Another trial favoring hyaluronic acid was excluded, due to its reporting form of adverse events different from others. Conclusion Intra-articular injection of PRP is an effective and safe method to treat knee osteoarthritis.