Impact of Hospital Volume on Postoperative Morbidity and Mortality Following a Colectomy for Ulcerative Colitis
2008; Elsevier BV; Volume: 134; Issue: 3 Linguagem: Inglês
10.1053/j.gastro.2008.01.004
ISSN1528-0012
AutoresGilaad G. Kaplan, Ellen P. McCarthy, John Z. Ayanian, Joshua R. Korzenik, Richard A. Hodin, Bruce E. Sands,
Tópico(s)Colorectal Cancer Screening and Detection
ResumoBackground & Aims: Postoperative morbidity and mortality following a colectomy for ulcerative colitis (UC) has been primarily reported from tertiary care referral centers that perform a high volume of operations; however, the postoperative outcomes among nonselected hospitals are not known. We set out to evaluate postoperative morbidity and mortality using a nationally representative database and to determine the factors that influenced outcomes. Methods: We analyzed the 1995–2005 Nationwide Inpatient Sample to identify 7108 discharges for UC patients who underwent a total abdominal colectomy. The effects of hospital volume on postoperative morbidity and mortality were evaluated in logistic regression models adjusting for demographic and clinical factors. Results: Postoperative mortality and morbidity rates were 2.3% and 30.8%, respectively. Most operations were performed in low-volume hospitals that had an increased risk of death (adjusted odds ratio [aOR], 2.42; 95% confidence interval [CI]: 1.26–4.63). In-hospital mortality was increased in patients who were admitted emergently (aOR, 5.40; 95% CI: 3.48–8.40), aged 60–80 years (aOR, 8.70; 95% CI: 3.30–22.92), and those with Medicaid (aOR, 4.29; 95% CI: 2.13–8.66). Emergently admitted UC patients whose surgery was performed 6 days after their admission had significantly increased likelihood of in-hospital death (aOR, 2.12; 95% CI: 1.13–3.97). Conclusions: Postoperative mortality was lowest in hospitals that performed the highest volume of operations. Increasing the proportion of total colectomies performed in high-volume hospitals may improve clinical outcomes for patients with UC. Background & Aims: Postoperative morbidity and mortality following a colectomy for ulcerative colitis (UC) has been primarily reported from tertiary care referral centers that perform a high volume of operations; however, the postoperative outcomes among nonselected hospitals are not known. We set out to evaluate postoperative morbidity and mortality using a nationally representative database and to determine the factors that influenced outcomes. Methods: We analyzed the 1995–2005 Nationwide Inpatient Sample to identify 7108 discharges for UC patients who underwent a total abdominal colectomy. The effects of hospital volume on postoperative morbidity and mortality were evaluated in logistic regression models adjusting for demographic and clinical factors. Results: Postoperative mortality and morbidity rates were 2.3% and 30.8%, respectively. Most operations were performed in low-volume hospitals that had an increased risk of death (adjusted odds ratio [aOR], 2.42; 95% confidence interval [CI]: 1.26–4.63). In-hospital mortality was increased in patients who were admitted emergently (aOR, 5.40; 95% CI: 3.48–8.40), aged 60–80 years (aOR, 8.70; 95% CI: 3.30–22.92), and those with Medicaid (aOR, 4.29; 95% CI: 2.13–8.66). Emergently admitted UC patients whose surgery was performed 6 days after their admission had significantly increased likelihood of in-hospital death (aOR, 2.12; 95% CI: 1.13–3.97). Conclusions: Postoperative mortality was lowest in hospitals that performed the highest volume of operations. Increasing the proportion of total colectomies performed in high-volume hospitals may improve clinical outcomes for patients with UC. Ulcerative colitis (UC) is a serious lifelong condition characterized by periods of colonic inflammation and remission. Most patients are prescribed intermittent or chronic medications to control their symptoms, and a significant proportion will require a proctocolectomy for definitive management.1Hoie O. Wolters F.L. Riis L. et al.Low colectomy rates in ulcerative colitis in an unselected European cohort followed for 10 years.Gastroenterology. 2007; 132: 507-515Google Scholar, 2Farmer R.G. Easley K.A. Rankin G.B. Clinical patterns, natural history, and progression of ulcerative colitis A long-term follow-up of 1116 patients.Dig Dis Sci. 1993; 38: 1137-1146Google Scholar, 3Leijonmarck C.E. Persson P.G. Hellers G. Factors affecting colectomy rate in ulcerative colitis: an epidemiologic study.Gut. 1990; 31: 329-333Google Scholar The decision for surgery is difficult, but most commonly occurs when patients with severe colitis fail to respond to medical management or if they develop dysplasia or colorectal cancer. Most commonly, this decision is made among the patient, gastroenterologist, and surgeon as an outpatient, and the patient is electively admitted to the hospital for a planned proctocolectomy. Alternatively, some patients admitted emergently to the hospital with severe or fulminant colitis will undergo proctocolectomy if they fail medical management in-hospital or if they develop a complication such as a toxic megacolon.4Travis S. Review article: saving the colon in severe colitis—the case for medical therapy.Aliment Pharmacol Ther. 2006; 24: 68-73Google Scholar The 2 most common elective operations performed on UC patients are a 1 or multiple stage total proctocolectomy with ileal pouch anal anastomosis (IPAA) and a total proctocolectomy with a permanent ileostomy. Total abdominal colectomy with an ileorectal anastomosis is rarely performed in UC because of the consistent involvement of the rectum with inflammation. Patients operated emergently most often undergo a total abdominal colectomy with a Hartmann pouch (rectal stump) and temporary end ileostomy. When the patient has stabilized, subsequent operations are performed to form an IPAA, permanent ileostomy, or ileorectal anastomosis. A 2-stage IPAA, however, is sometimes performed even in urgent settings.5Nguyen G.C. Laveist T.A. Gearhart S. et al.Racial and geographic variations in colectomy rates among hospitalized ulcerative colitis patients.Clin Gastroenterol Hepatol. 2006; 4: 1507-1513Google Scholar, 6Cima R.R. Pemberton J.H. Medical and surgical management of chronic ulcerative colitis.Arch Surg. 2005; 140: 300-310Google Scholar Among patients who undergo an elective IPAA, early postoperative mortality rates are reported to be low, ranging between 0.0% and 1.0%7Fazio V.W. Ziv Y. Church J.M. et al.Ileal pouch-anal anastomoses complications and function in 1005 patients.Ann Surg. 1995; 222: 120-127Google Scholar, 8Meagher A.P. Farouk R. Dozois R.R. et al.J ileal pouch-anal anastomosis for chronic ulcerative colitis: complications and long-term outcome in 1310 patients.Br J Surg. 1998; 85: 800-803Google Scholar, 9Romanos J. Samarasekera D.N. Stebbing J.F. et al.Outcome of 200 restorative proctocolectomy operations: the John Radcliffe Hospital experience.Br J Surg. 1997; 84: 814-818Google Scholar, 10Neilly P. Neill M.E. Hill G.L. Restorative proctocolectomy with ileal pouch-anal anastomosis in 203 patients: the Auckland experience.Aust N Z J Surg. 1999; 69: 22-27Google Scholar, 11Young C.J. Solomon M.J. Eyers A.A. et al.Evolution of the pelvic pouch procedure at one institution: the first 100 cases.Aust N Z J Surg. 1999; 69: 438-442Google Scholar, 12Gorfine S.R. Gelernt I.M. Bauer J.J. et al.Restorative proctocolectomy without diverting ileostomy.Dis Colon Rectum. 1995; 38: 188-194Google Scholar; the largest series have reported mortality rates of 0.02% to 0.04%.7Fazio V.W. Ziv Y. Church J.M. et al.Ileal pouch-anal anastomoses complications and function in 1005 patients.Ann Surg. 1995; 222: 120-127Google Scholar, 8Meagher A.P. Farouk R. Dozois R.R. et al.J ileal pouch-anal anastomosis for chronic ulcerative colitis: complications and long-term outcome in 1310 patients.Br J Surg. 1998; 85: 800-803Google Scholar Early postoperative morbidity rates have varied widely between 16.6% and 57.7%.7Fazio V.W. Ziv Y. Church J.M. et al.Ileal pouch-anal anastomoses complications and function in 1005 patients.Ann Surg. 1995; 222: 120-127Google Scholar, 9Romanos J. Samarasekera D.N. Stebbing J.F. et al.Outcome of 200 restorative proctocolectomy operations: the John Radcliffe Hospital experience.Br J Surg. 1997; 84: 814-818Google Scholar, 10Neilly P. Neill M.E. Hill G.L. Restorative proctocolectomy with ileal pouch-anal anastomosis in 203 patients: the Auckland experience.Aust N Z J Surg. 1999; 69: 22-27Google Scholar, 12Gorfine S.R. Gelernt I.M. Bauer J.J. et al.Restorative proctocolectomy without diverting ileostomy.Dis Colon Rectum. 1995; 38: 188-194Google Scholar, 13Jarvinen H.J. Luukkonen P. Experience with restorative proctocolectomy in 201 patients.Ann Chir Gynaecol. 1993; 82: 159-164Google Scholar, 14Belliveau P. Trudel J. Vasilevsky C.A. et al.Ileoanal anastomosis with reservoirs: complications and long-term results.Can J Surg. 1999; 42: 345-352Google Scholar Short-term complications for IPAA may include pouch leak, intraabdominal abscess, anastomotic stricture, and small bowel obstruction.6Cima R.R. Pemberton J.H. Medical and surgical management of chronic ulcerative colitis.Arch Surg. 2005; 140: 300-310Google Scholar Approximately 27% of emergently admitted hospitalized patients will undergo colectomy because they fail in-hospital medical management or they develop a complication of UC.15Turner D. Walsh C.M. Steinhart A.H. et al.Response to corticosteroids in severe ulcerative colitis: a systematic review of the literature and a meta-regression.Clin Gastroenterol Hepatol. 2007; 5: 103-110Google Scholar Historically, mortality rates in this patient population have been high.16Block G.E. Moossa A.R. Simonowitz D. et al.Emergency colectomy for inflammatory bowel disease.Surgery. 1977; 82: 531-536Google Scholar, 17Kaplan H.P. Portnoy B. Binder H.J. et al.A controlled evaluation of intravenous adrenocorticotropic hormone and hydrocortisone in the treatment of acute colitis.Gastroenterology. 1975; 69: 91-95Google Scholar However, more recent studies of emergently admitted UC patients who underwent a colectomy have reported lower postoperative mortality rates that varied between 0.6% and 6.9%.18Alves A. Panis Y. Bouhnik Y. et al.Subtotal colectomy for severe acute colitis: a 20-year experience of a tertiary care center with an aggressive and early surgical policy.J Am Coll Surg. 2003; 197: 379-385Google Scholar, 19Pal S. Sahni P. Pande G.K. et al.Outcome following emergency surgery for refractory severe ulcerative colitis in a tertiary care centre in India.BMC Gastroenterol. 2005; 5: 39Google Scholar, 20Mikkola K.A. Jarvinen H.J. Management of fulminating ulcerative colitis.Ann Chir Gynaecol. 1992; 81: 37-41Google Scholar, 21Hyman N.H. Cataldo P. Osler T. Urgent subtotal colectomy for severe inflammatory bowel disease.Dis Colon Rectum. 2005; 48: 70-73Google Scholar, 22Carter F.M. McLeod R.S. Cohen Z. Subtotal colectomy for ulcerative colitis: complications related to the rectal remnant.Dis Colon Rectum. 1991; 34: 1005-1009Google Scholar The majority of outcome data for colectomies in UC patients has been published from highly specialized tertiary care referral centers. These studies introduced selection bias because centers that published the largest case series7Fazio V.W. Ziv Y. Church J.M. et al.Ileal pouch-anal anastomoses complications and function in 1005 patients.Ann Surg. 1995; 222: 120-127Google Scholar, 8Meagher A.P. Farouk R. Dozois R.R. et al.J ileal pouch-anal anastomosis for chronic ulcerative colitis: complications and long-term outcome in 1310 patients.Br J Surg. 1998; 85: 800-803Google Scholar also have the most experienced clinicians. If a significant proportion of operations are performed in low-volume hospitals, and surgical outcomes are different in those settings, then postoperative morbidity and mortality may be substantially different from those previously reported. Multiple studies have demonstrated that hospital volume is an important predictor of postoperative outcomes in both routine and complex surgeries.23Chowdhury M.M. Dagash H. Pierro A. A systematic review of the impact of volume of surgery and specialization on patient outcome.Br J Surg. 2007; 94: 145-161Google Scholar Because a total abdominal colectomy, particularly in emergently admitted UC patients, is a complex operation, surgical experience may be an important predictor of postoperative morbidity and mortality. Therefore, we used a nationally representative database of hospital discharges in the United States to evaluate how often hospitals performed total abdominal colectomies in UC patients and to determine whether hospital volume is an important predictor of postoperative morbidity and mortality in UC patients admitted electively and emergently. The Nationwide Inpatient Sample (NIS) database contains hospital discharge abstracts from ∼20% of nonfederal acute care hospitals in the United States. Stratified random sampling is done to ensure that the database is representative of the US population and that it accounts for ∼90% of all hospitalizations. The NIS contains information on demographic characteristics, up to 15 diagnostic and procedure codes based on the International Classification of Diseases, 9th Version, Clinical Modification (ICD-9-CM), outcomes, and hospital characteristics. Quality control and validation of the NIS are performed by the Agency for Healthcare Research and Quality (AHRQ; Rockville, MD).24Agency for Healthcare Quality and ResearchNIS database documentation.http://www.hcup-us.ahrq.gov/nisoverview.jspGoogle Scholar The database was provided with deidentified patient information and thus was deemed exempt from institutional review by the Partners Human Research Committee. We used ICD-9-CM diagnosis and procedure codes to identify 7806 discharges of patients with UC (556.X) who underwent a total abdominal colectomy (45.8) between 1995 and 2005. To minimize misclassification with Crohn’s disease, we excluded 194 discharges of patients that were coded both with UC and Crohn’s disease (555.X) or if they had a code for an ileal-to-ileal anastomosis (45.91). To minimize misclassification with ischemic colitis, we excluded 363 discharges of patients over age 80 years and those with a concurrent code for ischemic colitis (557.0, 557.1, or 557.9). Finally, we excluded 141 discharges of patients with evidence suggesting a prior colectomy, such as an ostomy complication (569.6 or v55.2) or ostomy closure (46.5). Our final sample consisted of 7108 patient discharges with UC who underwent a total abdominal colectomy. Our primary factor of interest was the annual hospital volume of total abdominal colectomies performed on UC patients for each unique hospital identifier. Hospitals were divided into terciles based on the number of colectomies performed on UC patients per year, and each patient was assigned a score of low (3 or fewer colectomies per year), medium (4–11 colectomies per year), or high (12 or more colectomies per year) hospital volume. Covariates included patient age categorized as 11 colectomies/year). Percentages and 95% confidence intervals were weighted to reflect national estimates.Low (95% CI) n = 2594Medium (95% CI) n = 2272High (95% CI) n = 2242Age (%)aP value <.05. <4040.6% (38.8–42.4)32.3% (30.1–34.5)41.9% (39.5–44.4)48.9% (46.3–51.6) 40–5935.3% (34.1–36.5)33.6% (31.7–35.5)37.1% (35.0–39.2)35.5% (33.4–37.7) ≥6024.1% (22.6–25.7)34.1% (32.1–36.3)21.0% (19.1–23.1)15.6% (13.9–17.4)Sex Female44.2% (43.0–45.4)43.1% (41.2–45.1)44.5% (42.2–46.8)45.1% (43.3–46.9) Male55.8% (54.6–57.0)56.9% (54.9–58.8)55.5% (53.2–57.8)54.9% (53.1–56.7)Race/ethnicityaP value <.05. White87.8% (86.4–89.1)85.0% (83.2–86.7)88.0% (85.5–90.2)91.1% (88.3–93.2) Non-white12.2% (10.9–13.6)15.0% (13.3–16.8)12.0% (9.8–14.5)8.9% (6.8–11.7)Health insuranceaP value <.05. Medicare17.5% (16.1–19.1)26.5% (24.6–28.5)14.7% (12.9–16.6)10.0% (8.2–12.2) Medicaid5.1% (4.5–5.8)6.2% (5.3–7.3)5.1% (4.2–6.3)3.7% (2.9–4.8) Private71.9% (69.9–73.7)60.4% (58.2–62.7)75.4% (72.9–77.7)81.6% (78.8–84.0) Other5.5% (4.8–6.3)6.8% (5.8–8.0)4.8% (3.7–6.2)4.7% (3.4–6.5)Comorbidity scoreaP value <.05. 0–168.5% (66.9–70.0)59.8% (57.6–61.9)69.7% (67.3–72.0)77.4% (75.3–79.3) 2 or more31.5% (30.0–33.1)40.2% (38.1–42.4)30.3% (28.0–32.7)22.6% (20.7–24.7)Colorectal canceraP value <.05. Colorectal cancer7.6% (7.0–8.3)9.1% (8.1–10.4)8.0% (7.0–9.2)5.5% (4.6–6.5) No colorectal cancer92.4% (91.7–93.0)90.9% (89.6–91.9)92.0% (90.8–93.0)94.5% (93.5–95.4)Admission typeaP value <.05. Emergent or urgent35.0% (32.9–37.2)42.8% (40.4–45.3)33.9% (31.2–36.6)27.5% (23.8–31.6) Elective65.0% (62.8–67.1)57.2% (54.7–59.6)66.1% (63.4–68.8)72.5% (68.4–76.2)a P value <.05.b Hospital volume was defined as low (≤3 colectomies/year), medium (4–11 colectomies/year), and high (>11 colectomies/year). Percentages and 95% confidence intervals were weighted to reflect national estimates.c The overall sample included 7108 discharges of patients. Number of observations with missing data: age (n = 2), sex (n = 6), race/ethnicity (n = 1876), health insurance (n = 27), admission type (n = 857). Open table in a new tab Most operations were performed in low-volume hospitals. More than 1000 hospitals performed only 1 annual colectomy on UC patients; in contrast, the hospital with the highest annual volume performed 52 colectomies over the course of a year. At the patient level of hospital volume, nearly half (48.0%) of UC patients underwent a colectomy in hospitals that performed 5 or fewer procedures. One third of patients underwent surgery in a hospital that performed 3 or fewer annual surgeries, whereas only one third of patients underwent a colectomy in hospitals that performed 12 or more annual colectomies. Pouches were formed in 30.2% of UC patients who underwent a total abdominal colectomy; high volume hospitals performed relatively more (46.3%) pouch operations compared with low-volume hospitals (17.2%) (Table 2). Although a permanent ileostomy was commonly performed (33.5%), it was primarily conducted in the lowest volume hospitals (46.2%). Total abdominal colectomies with or without proctectomy and temporary ileostomy were performed on one quarter of patients and were distributed similarly across hospital volume (Table 2).Table 2Type of Colectomy Performed on Ulcerative Colitis Patients by Hospital VolumeAll hospitals, n = 7108 (95% CI)Hospital volumeaHospital volume was defined as low (≤3 colectomies/year), medium (4–11 colectomies/year), and high (>11 colectomies/year). Percentages and 95% confidence intervals were weighted to reflect national estimates.Low, n = 2594 (95% CI)Medium, n = 2272 (95% CI)High, n = 2242 (95% CI)Pouch30.2% (27.0–33.6)17.2% (15.3–19.2)29.4% (26.6–32.3)46.3% (39.6–53.0)Permanent ileostomy33.5% (31.1–36.0)46.2% (43.8–48.5)32.8% (29.8–35.9)19.5% (16.6–22.9)Proctocolectomy and temporary ileostomy14.2% (12.7–15.9)12.7% (11.4–14.2)16.5% (14.6–18.6)13.6% (10.2–17.9)Ileorectal anastomosis3.4% (3.0–4.0)4.6% (3.8–5.5)3.0% (2.4–3.9)2.5% (1.8–3.5)Total abdominal colectomy, temporary ileostomy, and rectal stump13.6% (12.1–15.2)12.6% (11.3–14.1)13.8% (11.9–16.0)14.4% (11.1–18.5)Nonclassified5.0% (4.3–5.9)6.7% (5.7–7.8)4.5% (3.4–5.9)3.7% (2.5–5.5)a Hospital volume was defined as low (≤3 colectomies/year), medium (4–11 colectomies/year), and high (>11 colectomies/year). Percentages and 95% confidence intervals were weighted to reflect national estimates. Open table in a new tab The overall mortality rate for UC patients undergoing a total abdominal colectomy was 2.3%, 5.4% following emergent or urgent admission, and 0.7% following an elective admission. In univariate analysis, the mortality rate varied by hospital volume such that mortality decreased as hospital volume increased from low (4%), medium (2%), and high (0.7%) (P < .0001). Pouch operations had the lowest mortality rate (0.15%). Total abdominal colectomies with temporary ileostomy and rectal stump accounted for the highest mortality (5.5%), which primarily occurred in patients admitted emergently (7.4%). The mortality rate was 6.8% among patients admitted emergently who underwent a permanent ileostomy (Table 3).Table 3Mortality Rate by Type of Colectomy Performed Among Patients With Ulcerative Colitis Who Were Admitted Emergently and ElectivelyColectomy typeMortality, % (95% CI)Total, n = 7108Emergent or urgent, n = 2186aEight hundred sixty-two observations were missing data on admission type. Percentages and 95% confidence intervals were weighted to reflect national estimates.Elective, n = 4060aEight hundred sixty-two observations were missing data on admission type. Percentages and 95% confidence intervals were weighted to reflect national estimates.Pouch0.15 (0.05–0.45), n = 21440.0 (0.0–0.0), n = 3840.2 (0.1–0.6), n = 1475Permanent ileostomy3.6 (2.9–4.4), n = 23706.8 (5.3–8.6), n = 9171.0 (0.5–1.8), n = 1150Proctocolectomy and temporary ileostomy0.3 (0.1–1.1), n = 10101.6 (0.5–4.9), n = 2200.0 (0.0–0.0), n = 684Ileorectal anastomosis1.6 (0.6–4.0), n = 2465.7 (2.2–14.2), n = 670.0 (0.0–0.0) n = 147Total abdominal colectomy, ileostomy, and rectal stump5.5 (4.1–7.4), n = 9757.4 (5.2–10.4), n = 4832.8 (1.5–5.0), n = 406Nonclassified4.7 (2.9–7.6), n = 35810.6 (6.2–17.8), n = 1151.5 (0.5–4.8), n = 198All surgeries2.3 (2.0–2.8)5.4 (4.4–6.5)0.7 (0.5–1.0)a Eight hundred sixty-two observations were missing data on admission type. Percentages and 95% confidence intervals were weighted to reflect national estimates. Open table in a new tab After adjustment, hospital volume was significantly associated with in-hospital mortality. Compared with high-volume hospitals, low-volume (adjusted odds ratio [aOR], 2.42; 95% CI: 1.26–4.63) and medium-volume hospitals (aOR, 2.02; 95% CI: 1.02–4.01) had significantly higher postoperative mortality (Table 4). Other factors significantly associated with increased likelihood of in-hospital death following adjustment included older age, emergent or urgent admission, having Medicare or Medicaid insurance, and having 2 or more comorbid conditions. UC patients who underwent surgery for colorectal cancer had a lower likelihood of in-hospital death following colectomy (Table 4).Table 4Effect of Hospital Volume on Adjusted Postoperative Mortality and Morbidity for Ulcerative Colitis PatientsMortality adjusted OR (95% CI)aMultivariable logistic regression models adjusted for age, sex, emergency admission, comorbidities, health insurance, colorectal cancer, calendar year (continuous from 1995 to 2005) and clustering by hospital using generalized estimating equations. Model n = 6217 because of missing data on 1 or more covariates.Morbidity adjusted OR (95% CI)aMultivariable logistic regression models adjusted for age, sex, emergency admission, comorbidities, health insurance, colorectal cancer, calendar year (continuous from 1995 to 2005) and clustering by hospital using generalized estimating equations. Model n = 6217 because of missing data on 1 or more covariates.Hospital volumebHospital volume was defined as low (≤3 colectomies/year), medium (4–11 colectomies/year), and high (>11 colectomies/year). Low2.42 (1.26–4.63)1.18 (0.99–1.41) Medium2.02 (1.02–4.01)1.07 (0.89–1.30) High1.001.00Age, yr 11 colectomies/year). Open table in a new tab One or more postoperative complications were identified in 30.8% (95% CI: 29.1%–32.5%) of patients; however, high-volume hospitals reported fewer complications (25.6%; 95% CI: 22.4%–29.0%) compared with low-volume hospitals (35.4%; 95% CI: 33.5%–37.4%). The most common postoperative complications were gastrointestinal (14.9%), which did not vary by hospital volume. High-volume hospitals reported significantly lower complication rates compared with low-volume hospitals for the following complications: infection (6.0% vs 12.5%, respectively, P < .001), respiratory (3.1% vs 7.4%, respectively, P < .001), and cardiovascular (3.6% vs 5.9%, respectively, P < .003). In the adjusted analysis, postoperative complications were not more common in low-volume hospitals compared with high-volume hospitals (aOR, 1.18; 95% CI: 0.99–1.41) (Table 4). Patients who were elderly, males, emergently admitted, insured with Medicare, and had multiple preoperative comorbidities were significantly more likely to have one or more postoperative complications (Table 4). In the subset of patients admitted emergently/urgently the median number of days to operation was 6 (interquartile range [IQR], 2–11). The median preoperative days did not vary by hospital volume: low (n = 954), 6 (IQR, 2–12); medium (n = 670), 5.5 (IQR, 1–11); and high (n = 562), 5 (IQR, 2–10). In this subgroup, hospital volume was predictive of postoperative complications (aOR, 1.38; 95% CI: 1.04–1.82) but not for postoperative mortality (aOR, 1.87; 95% CI: 0.87–4.01). Compared with patients who underwent surgery within 3 days of admission, those who u
Referência(s)