National Centre for Immunisation Research and Surveillance, The Children's Hospital at Westmead, NSW, Australia.
Vaccine. 2012 May 2;30(21):3209-22. doi: 10.1016/j.vaccine.2012.02.048. Epub 2012 Mar 2.
Cross-protection by seasonal trivalent influenza vaccines (TIVs) against pandemic influenza A H1N1 2009 (now known as A[H1N1]pdm09) infection is controversial; and the vaccine effectiveness (VE) of A(H1N1)pdm09 vaccines has important health-policy implications. Systematic reviews and meta-analyses are needed to assess the impacts of both seasonal TIVs and A(H1N1)pdm09 vaccines against A(H1N1)pdm09.We did a systematic literature search to identify observational and/or interventional studies reporting cross-protection of TIV and A(H1N1)pdm09 VE from when the pandemic started (2009) until July 2011. The studies fulfilling inclusion criteria were meta-analysed. For cross-protection and VE, respectively, we stratified by vaccine type, study design and endpoint. Seventeen studies (104,781 subjects) and 10 studies (2,906,860 subjects), respectively, reported cross-protection of seasonal TIV and VE of A(H1N1)pdm09 vaccines; six studies (17,229 subjects) reported on both. Thirteen studies (95,903 subjects) of cross-protection, eight studies (859,461 subjects) of VE, and five studies (9,643 subjects) of both were meta-analysed and revealed: (1) cross-protection for confirmed illness was 19% (95% confident interval=13-42%) based on 13 case-control studies with notable heterogeneity. A higher cross-protection of 34% (9-52%) was found in sensitivity analysis (excluding five studies with moderate/high risk of bias). Further exclusion of studies that recruited early in the pandemic (when non-recipients of TIV were more likely to have had non-pandemic influenza infection that may have been cross-protective) dramatically reduced heterogeneity. One RCT reported cross-protection of 38% (19-53%) for confirmed illness. One case-control study reported cross-protection of 50% (40-59%) against hospitalisation. (2) VE of A(H1N1)pdm09 for confirmed illness was 86% (73-93%) based on 11 case-control studies and 79% (22-94%) based on two cohort studies; VE against medically-attended ILI was 32% (8-50%) in one cohort study. TIVs provided moderate cross-protection against both laboratory-confirmed A(H1N1)pdm09 illness (based on eight case-control studies with low risk of bias and one RCT) and also hospitalisation. A finding of increased risk from seasonal vaccine was limited to cases recruited early in the pandemic. A(H1N1)pdm09 vaccines were highly effective against confirmed A(H1N1)pdm09 illness. Although cross-protection was less than the direct effect of strain-specific vaccination against A(H1N1)pdm09, TIV was generally beneficial before A(H1N1)pdm09 vaccine was available.
季节性三价流感疫苗（TIV）对 2009 年大流行性甲型 H1N1 流感（现称为 A[H1N1]pdm09）感染的交叉保护作用存在争议；A（H1N1）pdm09 疫苗的疫苗效力（VE）对卫生政策具有重要意义。需要系统评价和荟萃分析来评估季节性 TIV 和 A（H1N1）pdm09 疫苗对 A（H1N1）pdm09 的影响。我们进行了系统的文献检索，以确定从大流行开始（2009 年）到 2011 年 7 月报告 TIV 和 A（H1N1）pdm09 VE 交叉保护作用的观察性和/或干预性研究。符合纳入标准的研究进行了荟萃分析。分别针对交叉保护和 VE，我们按疫苗类型、研究设计和终点进行分层。分别有 17 项研究（104781 例受试者）和 10 项研究（2906860 例受试者）报告了季节性 TIV 的交叉保护作用和 A（H1N1）pdm09 疫苗的 VE；有 6 项研究（17229 例受试者）报告了这两者。对 13 项研究（95903 例受试者）的交叉保护作用、8 项研究（859461 例受试者）的 VE 和 5 项研究（9643 例受试者）进行了荟萃分析，结果显示：（1）基于 13 项病例对照研究（具有显著异质性），确诊病例的交叉保护作用为 19%（95%可信区间为 13-42%）。敏感性分析（排除五项偏倚风险中度/高的研究）发现交叉保护作用更高，为 34%（9-52%）。进一步排除在大流行早期招募的研究（此时 TIV 的未接种者更有可能患有非大流行性流感感染，可能具有交叉保护作用），则大大降低了异质性。一项 RCT 报告确诊病例的交叉保护作用为 38%（19-53%）。一项病例对照研究报告了对住院的 50%（40-59%）的交叉保护作用。（2）基于 11 项病例对照研究和 2 项队列研究，A（H1N1）pdm09 的 VE 对确诊病例为 86%（73-93%），对有医疗记录的 ILI 为 32%（8-50%），一项队列研究报告了这一点。基于低偏倚风险的八项病例对照研究和一项 RCT，TIV 对实验室确诊的 A（H1N1）pdm09 疾病（基于低偏倚风险的八项病例对照研究和一项 RCT）和住院均提供了适度的交叉保护作用。季节性疫苗增加风险的发现仅限于大流行早期招募的病例。A（H1N1）pdm09 疫苗对确诊的 A（H1N1）pdm09 疾病非常有效。尽管交叉保护作用小于针对 A（H1N1）pdm09 的特定菌株疫苗的直接作用，但在 A（H1N1）pdm09 疫苗可用之前，TIV 通常是有益的。
