Surgical site occurrences, not body mass index, increase the long-term risk of ventral hernia recurrence.

BACKGROUND: Recurrence rates after ventral hernia repair vary widely and evidence about risk factors for recurrence are conflicting. There is little evidence for risk factors for long-term recurrence. METHODS: Patients who underwent ventral hernia repair at our institution and were captured in the American College of Surgeons-National Surgical Quality Improvement Program database between 2002 and 2015 were included. We reviewed all demographic, procedural, and hernia-specific data. RESULTS: Six hundred and thirty patients were included for analysis with a median follow-up of 4.9 years (inter-quartile range, 2-7.3 years). By univariate analysis, index hernia repairs were more likely to recur if defect size was ≥4 cm (P = .019), no mesh was used (P = .026), or if the repair was for a recurrent hernia (P = .001). Five-year cumulative incidence of recurrence and reoperation was 24.3% and 16.0%, respectively. Patients with a perioperative surgical site occurrence, which included superficial, deep-incisional, and organ space surgical site infections as well as wound disruption, had a 5-year cumulative incidence of recurrence of 54.9% compared with 22.6% for those without surgical site occurrence. By multivariable analysis, non-primary hernia repair (hazard ratio 1.7, 95% confidence interval 1.2-2.4, P = .005) and any postoperative surgical site occurrence (hazard ratio 1.9, 95% confidence interval 1.1-3.6, P = .02) were the only risk factors predictive of recurrence. Patient body mass index had no independent effect on recurrence. CONCLUSION: 1 in 4 patients undergoing an open ventral hernia repair will have a recurrence after 5 years, and this risk is doubled among patients who experience any perioperative surgical site occurrence. After controlling for patient comorbidities, including body mass index, hernia size, and mesh position, the most significant risk factor for recurrence after ventral hernia repair was a non-primary hernia and surgical site occurrence.

Predicting venous thrombosis in patients undergoing elective splenectomy.

BACKGROUND: Venous thrombosis (VT) is an ongoing problem for patients undergoing elective splenectomy. There is limited data evaluating risk factors for VTs. An increase in platelet counts is commonly seen after splenectomy; however, there is a paucity of literature evaluating post-operative platelet counts as a risk factor for VTs in this patient cohort. The objective of this study was to determine the incidence of VT events and to use the platelet count as a predictor for VT development. METHODS: A retrospective review was undertaken at Brigham Women's Hospital, evaluating elective splenectomy patients between 1997 and 2018. Descriptive statistics were utilized to determine the incidence of VTs. Receiver operator characteristic (ROC) curves were utilized to identify platelet counts that could predict VTs. RESULTS: Five hundred and twenty splenectomies were included in the study of which 344 were completed in an open manner and 176 were done laparoscopically. The overall incidence of VT events was 6.7% (35/520), 6.1% (21/344) for open, and 8.0% (14/176) for laparoscopic approaches (p = 0.43). ROC curves demonstrated platelet counts to be a good predictor for the development of VTs with an area under the curve (AUC) of 0.77 (95% CI 0.69-0.86; p < 0.001) for all splenectomy patients, 0.70 (95% CI 0.59-0.81; p < 0.001) for those completed in an open manner, and 0.88 (95% CI 0.77-0.99; p < 0.001) for those done laparoscopically. The optimal platelet cutoff was found to be 545 for the overall splenectomy cohort, 457 for the open, and 659 for the laparoscopic cohorts. These platelet counts had a diagnostic accuracy that ranged from 61 to 86% and a negative predictive value (NPV) that ranged from 97 to 99%. CONCLUSION: These results suggest platelet cutoffs that predict VTs. This information can be used to individualize prophylactic strategies.

Laparoscopic versus open splenectomy for splenomegaly: the verdict is unclear.

BACKGROUND: The benefits of laparoscopic splenectomy (LS) over open splenectomy (OS) for normal-sized spleens have been well documented. However, the role of laparoscopy for moderate and massive splenomegaly is debated. METHODS: A retrospective review of patients undergoing elective splenectomy at one institution from 1997 to 2017 was conducted. Moderate and massive splenomegaly was defined as splenic weight of 500-1000 g and greater than 1000 g, respectively. We performed a 1:2 matching of laparoscopic to open splenectomy to control for differences in splenic weight. Differences in perioperative morbidity (infection, thromboembolism, reoperation, readmission), intraoperative factors (blood loss, operative time), length of stay, and mortality were examined. RESULTS: A total of 491 elective splenectomies were identified. 268 cases were for splenic weights greater than 500 g. After a 1:2 matching of LS:OS, we identified 22 LS and 44 matched OS for moderate splenomegaly. The LS group had longer mean operative times (178 vs. 107 min, p < 0.01), with similar length of stay and blood loss. For massive splenomegaly, 26 LS were identified and matched to 52 OS. LS had longer mean operative times (171 vs. 112 min, p < 0.01) and higher readmission rates (27% vs. 6%, p < 0.05). Other factors and outcomes did not differ between LS and OS for moderate or massive splenomegaly. The conversion rate for LS was higher for massive versus moderate splenomegaly, but was not statistically significant (35% vs. 14%, p = 0.09). CONCLUSIONS: LS for moderate and massive splenomegaly is associated with longer operative times. Other perioperative outcomes were comparable to OS, with no demonstrated benefits for LS. Although LS may be a feasible approach to moderate and massive splenomegaly, its benefits require further clarification in this patient population.