Gastrointestinal Bleeding in Mechanical Cardiac Support.

BACKGROUND: Mechanical cardiac support (MCS) is a lifesaving therapy option in patients with heart failure and other medical disorders. However, there is an associated risk of gastrointestinal bleeding (GIB). The goal of this study was to determine GIB incidence and associated risk factors. METHODS: All patients at one institution from 2009 to 2018 under durable and nondurable support were retrospectively reviewed for GIB during their MCS period. Clinical records were evaluated for patient demographics, GIB characteristics, and interventions. Univariate and multivariate analyses were performed to compare patient groups. RESULTS: A total of 427 patients were reviewed, with 111 (25.9%) patients representing 218 episodes of GIB during our study period. The incidence rate from support initiation to GIB was 44.9% by 6 months and 60.6% in 12 months, occurring at a mean of 216.7 days. Higher rates of bleeding were found in patients with hypertension (82% vs 71.5%; P = .03) and diabetes mellitus (62.2% vs 38.3%; P < .0001), as well as pulmonary (48.7% vs 35.4%; P = .014), hepatic (21.6% vs 10.4%; P = .003), and renal disease (48.7% vs 37.3%; P = .037). Endoscopy revealed an upper GI source in 56% (n = 123) of bleeds. The most common etiology of bleeding included angiodysplasia/vascular malformation (35.7%). Therapeutic intervention was performed in 109 (50%) cases, with only 1 surgical intervention. DISCUSSION: Overall, GIB can be a significant adverse event in patients under mechanical cardiac support, so proper management of anticoagulation and early endoscopy evaluation remains of great importance.

Acute Diverticulitis Outcomes in the Acute Care Surgery Model.

BACKGROUND: The acute care surgery (ACS) model has been widely implemented, with single institution studies demonstrating improved outcomes but multicenter studies questioning the efficiency. Acute care surgery programs care for sicker and more economically disadvantaged patients. This study compares outcomes between ACS and traditional models in the management of diverticulitis across an entire state. STUDY DESIGN: The Virginia Health Information administrative database for adults discharged with diverticulitis from January 2008 through September 2015, was reviewed. Patient characteristics were analyzed and compared between ACS and traditional models. Outcome differences were compared using logistic regression. RESULTS: We reviewed 23,943 admissions, with 2,330 (9.7%) patients cared for in ACS programs. The ACS patients were more likely to be uninsured (10.6% vs 6.8%, p < 0.0001) or covered by Medicaid (5.5% vs 3.4%, p < 0.0001), and the ACS hospitals cared for a higher percentage of minority patients than in the traditional programs (30.4% vs 19.8%, p < 0.0001). Operative rates were higher in ACS hospitals (14.7% vs 11.8%, p < 0.0001), as were rates of complicated diverticulitis (24.5% vs 20.3%, p < 0.0001). The ACS patients had significantly higher rates of comorbidities. After adjusting for patient comorbidities and demographics, ACS patients had a higher rate of complications (odds ratio [OR] 1.36, p = 0.0017). However, there was no difference in mortality, length of stay, or costs. When comparing only operative patients, there were no outcome differences after adjusting for patient factors. CONCLUSIONS: Acute care surgery patients present to the hospital with more severe disease, higher rates of medical comorbidities, and lower socioeconomic status. Once patient factors are accounted for, outcomes are equivalent for operative patients in either model. Acute care surgery hospitals provide high quality and efficient care to sicker and more complex patients than traditional programs.

Statewide assessment of surgical outcomes and the acute care surgery model.

BACKGROUND: The acute care surgery (ACS) model has been widely implemented with single institution studies demonstrating improved outcomes. Recent multicenter studies have raised questions about the economics and efficacy of ACS. This study compares traditional and ACS outcomes across an entire state. METHODS: A retrospective review of Virginia's Health Information administrative database was completed. Adults admitted with appendicitis or cholecystitis between 2008 and 2014 were included. Hospital administration was contacted to determine surgical model. To compare patient characteristics, t-test and chi-square analyses were used. Total charges and length of stay (LOS) differences between ACS and traditional were examined using generalized linear models, whereas logistic regression was used for the presence of complications and 30-day mortality. RESULTS: Overall, the ACS model showed an increased proportion of uninsured patients with a higher rate of comorbidities. In the appendicitis subgroup, (n = 22,011; ACS n = 1993), ACS patients had higher total charges ($30,060 versus $28,460, P = 0.013), longer LOS (3.31 versus 2.92 d, P < 0.001), and higher chance of complications (odds ratio [OR] = 1.2, P = 0.016) and mortality (OR = 2.4, P = 0.029). After adjustment for comorbidities and insurance, mortality was no longer significantly different. In the cholecystitis group (n = 6936; ACS n = 777), ACS patients had a longer LOS (4.55 versus 4.13 d; P = 0.009) without significant differences in mortality, complications, or cost. There were no significant differences after adjustment for patient characteristics. CONCLUSIONS: ACS patients in Virginia have a higher rate of medical comorbidities and uninsured status, with slightly worse outcomes than the traditional model for appendicitis. Further studies to determine which patients benefit the most from ACS are warranted.