Division of Bariatric, Department of Surgery, Foregut and Advanced Gastrointestinal Surgery, Stony Brook University Medical Center, 100 Nichols Road, HSC T19, Stony Brook, NY, 11794, USA.
Department of Family, Population and Preventive Medicine, Stony Brook University Medical Center, Stony Brook, NY, USA.
Surg Endosc. 2018 May;32(5):2355-2364. doi: 10.1007/s00464-017-5931-0. Epub 2017 Nov 3.
There is a growing debate regarding outcomes following complex hepato-pancreato-biliary (HPB) procedures. The purpose of our study is to examine if facility type has any impact on complications, readmission rates, emergency department (ED) visit rates, and length of stay (LOS) for patients undergoing HPB surgery.
The SPARCS administrative database was used to identify patients undergoing complex HPB procedures between 2012 and 2014 in New York. Univariate generalized linear mixed models were fit to estimate the marginal association between outcomes such as overall/severe complication rates, 30-day and 1-year readmission rates, 30-day and 1-year ED-visit rates, and potential risk factors. Univariate linear mixed models were used to estimate the marginal association between possible risk factors and LOS. Facility type, as well as any variables found to be significant in our univariate analysis (p = 0.05), was further included in the multivariable regression models.
There were 4122 complex HPB procedures performed. Academic facilities were more likely to have a higher hospital volume (p < 0001). Surgery at academic facilities were less likely to have coexisting comorbidities; however, they were more likely to have metastatic cancer and/or liver disease (p = 0.0114, < 0. 0001, and = 0.0299, respectively). Postoperatively, patients at non-academic facilities experienced higher overall complication rates, and higher severe complication rates, when compared to those at academic facilities (p < 0.0001 and = 0.0018, respectively). Further analysis via adjustment for possible confounding factors, however, revealed no significant difference in the risk of severe complications between the two facility types. Such adjustment also demonstrated higher 30-day readmission risk in patients who underwent their surgery at an academic facility.
No significant difference was found when comparing the outcomes of academic and non-academic facilities, after adjusting for age, gender, race, region, insurance, and hospital volume. Patients from academic facilities were more likely to be readmitted within the first 30-days after surgery.
对于复杂肝胆胰（HPB）手术的结果，存在着越来越多的争论。我们的研究目的是检验医疗机构类型是否会对并发症、再入院率、急诊就诊率和住院时间（LOS）产生影响，这些因素都与接受 HPB 手术的患者有关。
我们使用 SPARCS 行政数据库，在 2012 年至 2014 年间，确定了在纽约进行复杂 HPB 手术的患者。采用单变量广义线性混合模型来估计结局（如总体/严重并发症发生率、30 天和 1 年再入院率、30 天和 1 年急诊就诊率）之间的边际关联，并对潜在的风险因素进行评估。采用单变量线性混合模型来估计可能的风险因素与 LOS 之间的边际关联。我们还将医疗机构类型以及单变量分析中发现的任何具有统计学意义的变量（p = 0.05）纳入多变量回归模型中。
共有 4122 例复杂的 HPB 手术。学术机构的医院容量更大（p < 0001）。学术机构进行的手术合并症更少；然而，它们更有可能患有转移性癌症和/或肝脏疾病（p = 0.0114，< 0.0001，和 = 0.0299，分别）。与学术机构相比，非学术机构的患者术后总体并发症发生率和严重并发症发生率更高（p < 0.0001 和 = 0.0018，分别）。然而，通过调整可能的混杂因素进行进一步分析后，并未发现两种医疗机构类型之间严重并发症风险存在显著差异。这种调整还表明，在学术机构接受手术的患者在术后 30 天内再次入院的风险更高。
在调整年龄、性别、种族、地区、保险和医院容量等因素后，比较学术和非学术机构的结果时，并未发现显著差异。来自学术机构的患者在手术后 30 天内更有可能再次入院。
