Department of Surgery, University of Utah, Salt Lake City, UT.
Surgical Services Clinical Program, Intermountain Healthcare, Salt Lake City, UT.
Ann Surg. 2019 Jan;269(1):133-142. doi: 10.1097/SLA.0000000000002404.
To (1) evaluate rates of surgery for clinical stage I-II pancreatic ductal adenocarcinoma (PDAC), (2) identify predictors of not undergoing surgery, (3) quantify the degree to which patient- and hospital-level factors explain differences in hospital surgery rates, and (4) evaluate the association between adjusted hospital-specific surgery rates and overall survival (OS) of patients treated at different hospitals.
Curative-intent surgery for potentially resectable PDAC is underutilized in the United States.
Retrospective cohort study of patients ≤85 years with clinical stage I-II PDAC in the 2004 to 2014 National Cancer Database. Mixed effects multivariable models were used to characterize hospital-level variation across quintiles of hospital surgery rates. Multivariable Cox proportional hazards models were used to estimate the effect of adjusted hospital surgery rates on OS.
Of 58,553 patients without contraindications or refusal of surgery, 63.8% underwent surgery, and the rate decreased from 2299/3528 (65.2%) in 2004 to 4412/7092 (62.2%) in 2014 (P < 0.001). Adjusted hospital rates of surgery varied 6-fold (11.4%-70.9%). Patients treated at hospitals with higher rates of surgery had better unadjusted OS (median OS 10.2, 13.3, 14.2, 16.5, and 18.4 months in quintiles 1-5, respectively, P < 0.001, log-rank). Treatment at hospitals in lower surgery rate quintiles 1-3 was independently associated with mortality [Hazard ratio (HR) 1.10 (1.01, 1.21), HR 1.08 (1.02, 1.15), and HR 1.09 (1.04, 1.14) for quintiles 1-3, respectively, compared with quintile 5] after adjusting for patient factors, hospital type, and hospital volume.
Quality improvement efforts are needed to help hospitals with low rates of surgery ensure that their patients have access to appropriate surgery.
（1）评估临床 I-II 期胰腺导管腺癌（PDAC）患者的手术率，（2）确定未接受手术的预测因素，（3）量化患者和医院水平因素解释医院手术率差异的程度，（4）评估调整后的医院特定手术率与不同医院治疗的患者总体生存率（OS）之间的关联。
在美国，对于潜在可切除的 PDAC 进行治愈性手术的应用不足。
对 2004 年至 2014 年国家癌症数据库中年龄≤85 岁的临床 I-II 期 PDAC 患者进行回顾性队列研究。采用混合效应多变量模型对手术率五分位组的医院水平差异进行描述。采用多变量 Cox 比例风险模型估计调整后的医院手术率对 OS 的影响。
在没有手术禁忌证或拒绝手术的 58553 名患者中，63.8%接受了手术，手术率从 2004 年的 2299/3528（65.2%）下降到 2014 年的 4412/7092（62.2%）（P<0.001）。调整后的医院手术率差异高达 6 倍（11.4%-70.9%）。在接受手术率较高的医院治疗的患者中，未调整的 OS 更好（五分位组 1-5 的中位 OS 分别为 10.2、13.3、14.2、16.5 和 18.4 个月，P<0.001，对数秩检验）。在调整了患者因素、医院类型和医院容量后，在手术率较低的五分位组 1-3 中治疗与死亡率独立相关[风险比（HR）分别为 1.10（1.01，1.21）、HR 1.08（1.02，1.15）和 HR 1.09（1.04，1.14）]。
需要进行质量改进努力，以帮助手术率较低的医院确保其患者能够获得适当的手术。
