读医学网

目的：我们旨在阐明髋骨骨折风险和使用质子泵抑制剂（PPIs）或组胺2受体阻断剂（H2RAs）之间的本质关联。

方法：我们使用英国初级医疗保健数据库——健康改进网络，确定2000～2008年间，有髋骨骨折诊断记录，且年龄在的40～49岁患者。通过回顾计算机记录和向初级保健医生发放调查问卷确认髋骨骨折病例。通过实施巢式病例对照分析的队列研究评价抑酸药的使用情况与髋骨骨折之间的关联。

结果：髋骨骨折的整体发生率为1.31/1000人年（95%置信区间[CI] 1.28～1.33）。调整潜在的混杂因素后，髋骨骨折的风险轻微增加(目前使用PPIs 和 H2RAs的比值比[OR] 分别为：1.09 [95% CI 1.01～1.17] 和1.04 [95% CI 0.90～1.19])。相对于未使用者，可观察到中高剂量PPIs（分别为OR 1.11 [95% CI 1.01～1.22]； OR 1.31 [95% CI 1.06～1.61]）和高剂量H2Ras（OR 2.77，95% CI1.21～6.37）会造成骨折风险增加。并未观察到持续性的反应（目前使用PPI的时间小于1个月和5年及以上的ORs分别为：1.16 [95% CI 0.94～1.43] 和1.02 [95% CI 0.87～1.20]）。

结论：校正了潜在的混杂因素后，接受PPIs治疗的患者髋骨骨折的风险轻度增加。髋骨骨折风险和PPI使用之间的剩余关联可能归因于残余混杂。

OBJECTIVES:

We aimed to clarifythe nature of the association between hip fracture risk and use of proton pumpinhibitors (PPIs) or histamine2 -receptor antagonists (H2 RAs).

METHODS:

We identifiedpatients 40-89 years of age with a recorded hip fracture diagnosis in 2000-2008using The Health Improvement Network, a UK primary care research database.Computerized records were reviewed and questionnaires sent to primary carephysicians to validate hip fracture cases. A cohort study with a nestedcase-control analysis was performed to estimate the association between the useof acid-suppressive drugs and hip fracture.

RESULTS:

Overall incidence ofhip fracture per 1000 person-years was 1.31 (95% confidence interval [CI]1.28-1.33). There was a modest increased risk of hip fracture after adjustmentfor potential confounders (odds ratios [OR] during current use of PPIs and H2RAs: 1.09 [95% CI 1.01-1.17] and 1.04 [95% CI 0.90-1.19], respectively).Relative to nonuse, an increased risk of fracture was observed with medium andhigh doses of PPIs (OR 1.11 [95% CI 1.01-1.22] and OR 1.31 [95% CI 1.06-1.61],respectively) and high doses of H2 RAs (OR 2.77, 95% CI 1.21-6.37). No durationresponse was observed (ORs for current PPI use less than 1 month and 5 years orlonger: 1.16 [95% CI 0.94-1.43] and 1.02 [95% CI 0.87-1.20], respectively).

CONCLUSIONS:

Patients treated withPPIs showed a modest increased risk of hip fracture after adjustment forpotential cofounders. Any remaining association between PPI use and hipfracture risk may be attributable to residual confounding.