Machine learning, deep learning and hernia surgery. Are we pushing the limits of abdominal core health? A qualitative systematic review.

INTRODUCTION: This systematic review aims to evaluate the use of machine learning and artificial intelligence in hernia surgery. METHODS: The PRISMA guidelines were followed throughout this systematic review. The ROBINS-I and Rob 2 tools were used to perform qualitative assessment of all studies included in this review. Recommendations were then summarized for the following pre-defined key items: protocol, research question, search strategy, study eligibility, data extraction, study design, risk of bias, publication bias, and statistical analysis. RESULTS: A total of 13 articles were ultimately included for this review, describing the use of machine learning and deep learning for hernia surgery. All studies were published from 2020 to 2023. Articles varied regarding the population studied, type of machine learning or Deep Learning Model (DLM) used, and hernia type. Of the thirteen included studies, all included either inguinal, ventral, or incisional hernias. Four studies evaluated recognition of surgical steps during inguinal hernia repair videos. Two studies predicted outcomes using image-based DMLs. Seven studies developed and validated deep learning algorithms to predict outcomes and identify factors associated with postoperative complications. CONCLUSION: The use of ML for abdominal wall reconstruction has been shown to be a promising tool for predicting outcomes and identifying factors that could lead to postoperative complications.

Preservation of deep epigastric perforators during anterior component separation technique (ACST) results in equivalent wound complications compared to transversus abdominis release (TAR).

PURPOSE: The use of component separation results in myofascial release and increased rates of fascial closure in abdominal wall reconstruction(AWR). These complex dissections have been associated with increased rates of wound complications with anterior component separation having the greatest wound morbidity. The aim of this paper was to compare the wound complication rate between perforator sparing anterior component separation(PS-ACST) and transversus abdominus release(TAR). METHODS: Patients were identified from a prospective, single institution hernia center database who underwent PS-ACST and TAR from 2015 to 2021. The primary outcome was wound complication rate. Standard statistical methods were used, univariate analysis and multivariable logistic regression were performed. RESULTS: A total of 172 patients met criteria, 39 had PS-ACST and 133 had TAR performed. The PS-ACST and TAR groups were similar in terms of diabetes (15.4% vs 28.6%, p = 0.097), but the PS-ACST group had a greater percentage of smokers (46.2% vs 14.3%, p < 0.001). The PS-ACST group had a larger hernia defect size (375.2 ± 156.7 vs 234.4 ± 126.9cm2, p < 0.001) and more patients who underwent preoperative Botulinum toxin A (BTA) injections (43.6% vs 6.0%, p < 0.001). The overall wound complication rate was not significantly different (23.1% vs 36.1%, p = 0.129) nor was the mesh infection rate (0% vs 1.6%, p = 0.438). Using logistic regression, none of the factors that were significantly different in the univariate analysis were associated with wound complication rate (all p > 0.05). CONCLUSION: PS-ACST and TAR are comparable in terms of wound complication rates. PS-ACST can be used for large hernia defects and promote fascial closure with low overall wound morbidity and perioperative complications.

National epidemiologic trends (2008-2018) in the United States for the incidence and expenditures associated with incisional hernia in relation to abdominal surgery.

PURPOSE: It is unknown whether the trend of rising incisional hernia (IH) repair (IHR) incidence and costs until 2011 currently persists. We aimed to evaluate how the IHR procedure incidence, cost and patient risk-profile have changed over the last decade relative to all abdominal surgeries (AS). METHODS: Repeated cross-sectional analysis of 38,512,737 patients undergoing inpatient 4AS including IHR within the 2008-2018 National Inpatient Sample. Yearly incidence (procedures/1,000,000 people [PMP]), hospital costs, surgical and patient characteristics were compared between IHR and AS using generalized linear and multinomial regression. RESULTS: Between 2008-2018, 3.1% of AS were IHR (1,200,568/38,512,737). There was a steeper decrease in the incidence of AS (356.5 PMP/year) compared to IHR procedures (12.0 PMP/year) which resulted in the IHR burden relative to AS (2008-2018: 12,576.3 to 9,113.4 PMP; trend difference P < 0.01). National costs averaged $47.9 and 1.7 billion/year for AS and IHR, respectively. From 2008-2018, procedure costs increased significantly for AS (68.2%) and IHR (74.6%; trends P < 0.01). Open IHR downtrended (42.2%), whereas laparoscopic (511.1%) and robotic (19,301%) uptrended significantly (trends P < 0.01). For both AS and IHR, the proportion of older (65-85y), Black and Hispanic, publicly-insured, and low-income patients, with higher comorbidity burden, undergoing elective procedures at small- and medium-sized hospitals uptrended significantly (all P < 0.01). CONCLUSION: IH persists as a healthcare burden as demonstrated by the increased proportion of IHR relative to all AS, disproportionate presence of high-risk patients that undergo these procedures, and increased costs. Targeted efforts for IH prevention have the potential of decreasing $17 M/year in costs for every 1% reduction.

Characterizing hernia centers in the United States: what defines a hernia center?

PURPOSE: A universal definition for what constitutes a hernia center does not exist. The purpose of this study was to characterize hernia centers in the United States by analyzing hernia centers and their non-hernia center counterparts. METHODS: A web-based search was conducted to identify defining features of hernia centers including faculty demographics and composition, research output, research funding, clinical trials, and website content. Hernia centers and non-hernia centers were compared. RESULTS: Most hernia centers (n = 36) are in urban areas (89%) and distributed evenly across regions of the United States. Hernia centers are associated with University program types (p = 0.001) while non-hernia centers are associated with University-Affiliate (p = 0.001) and Community (p = 0.02) program types. Hernia centers are associated with Abdominal Core Health Quality Collaborative participation (p = 0.01) and Center of Excellence by the Surgical Review Corporation certification (p = 0.005). Hernia centers are associated with presence of active clinical trials (p < 0.001) and number of clinical trials (p < 0.001). Hernia centers are associated with industry-sponsored trials (p < 0.001) but are not associated with NIH-sponsored trials. Fifty percent of hernia centers have PRS faculty. The vast majority of hernia center websites describe hernias treated (92%) and repair techniques (89%). The majority of hernia center mission statements emphasize an individualized care plan (61%) and multidisciplinary care (57%). Only 39% of websites and 17% of mission statements mention research. CONCLUSION: In the United States, hernia centers are clinically oriented, multidisciplinary surgical teams at predominantly urban, University programs that may use this title to attract patient referrals and industry sponsorship of clinical trials.

Perioperative optimization in complex abdominal wall hernias: Delphi consensus statement.

BACKGROUND: The incidence of incisional hernia after major abdominal surgery via a midline laparotomy is 20-41 per cent with short-term follow-up, and over 50 per cent in those surviving an abdominal catastrophe. Abdominal wall reconstruction (AWR) requires complex operations, often involving mesh resection, management of scarred skin, fistula takedown, component separation or flap reconstruction. Patients tend to have more complex conditions, with multiple co-morbidities predisposing them to a vicious cycle of complications and, subsequently, hernia recurrence. Currently there appears to be variance in perioperative practice and minimal guidance globally. The aim of this Delphi consensus was to provide a clear benchmark of care for the preoperative assessment and perioperative optimization of patients undergoing AWR. METHODS: The Delphi method was used to achieve consensus from invited experts in the field of AWR. Thirty-two hernia surgeons from recognized hernia societies globally took part. The process included two rounds of anonymous web-based voting with response analysis and formal feedback, concluding with a live round of voting followed by discussion at an international conference. Consensus for a strong recommendation was achieved with 80 per cent agreement, and a weak recommendation with 75 per cent agreement. RESULTS: Consensus was obtained on 52 statements including surgical assessment, preoperative assessment, perioperative optimization, multidisciplinary team and decision-making, and quality-of-life assessment. Forty-six achieved over 80 per cent agreement; 14 statements achieved over 95 per cent agreement. CONCLUSION: Clear consensus recommendations from a global group of experts in the AWR field are presented in this study. These should be used as a baseline for surgeons and centres managing abdominal wall hernias and performing complex AWR.

Robotic Sugarbaker parastomal hernia repair: technique and outcomes.

PURPOSE: To present a novel technique for the repair of parastomal hernias. METHODS: A total of 15 patients underwent parastomal hernia repair. A robotic Sugarbaker technique was utilized for repair. The fascial defect was closed prior to robotic intraperitoneal placement of the mesh. Baseline demographics of the patients were obtained, and intra-operative and post-operative outcomes were tracked. RESULTS: The etiology of the ostomies was oncologic in all but three patients. Five of the stomas were urostomies (33.3%). Patient characteristics were as follows: age 64.9.1 ± 9.3 years, BMI 30.1 ± 4.7 kg/m2, smoking history 60.0%, and diabetes 6.7%. The mean size of the hernia defect was 46.0 ± 40.1 cm2 with a mesh size of 372.0 ± 101.2 cm2. The mean operative time was 182.0 ± 51.9 min. In seven patients, an inferolateral preperitoneal flap was created for mesh placement. Intraoperatively, only one enterotomy was made during dissection, which was repaired without complication. The mean length of stay was 4.2 ± 1.9 days. There was only one hernia recurrence (6.7%). There were no wound complications, surgical site infections, or mesh infections. A mean follow-up time of 14.2 ± 9.4 months was achieved. CONCLUSIONS: Robotic Sugarbaker parastomal hernia repair is a safe and effective technique. The results demonstrate the feasibility of fascial closure with this technique and a low recurrence rate. The authors propose this technique should be widely considered for parastomal hernia repair.

A systematic review of CT chest in COVID-19 diagnosis and its potential application in a surgical setting.

AIM: The aim of this work was to investigate the sensitivity and utility of CT of the chest in diagnosing active SARS-Cov-2 (COVID-19) infection, and its potential application to the surgical setting. METHOD: A literature review was conducted using Google Scholar® and MEDLINE®/PubMed® to identify current available evidence regarding the sensitivity of CT chest compared with RT-PCR for the diagnosis of COVID-19-positive patients. GRADE criteria and the QUADAS 2 tool were used to assess the level of evidence. RESULTS: A total of 20 articles were identified that addressed the question of sensitivity of CT for diagnosis of symptomatic and asymptomatic COVID-19-positive patients. Overall sensitivity of CT scan ranged from 57%-100% for symptomatic and 46%-100% for asymptomatic COVID-19 patients, while that of RT-PCR ranged from 39%-89%. CT chest was a better diagnostic modality and capable of detecting active infection earlier in the time course of infection than RT-PCR in symptomatic patients. In asymptomatic patients, disease prevalence seems to play a role in the positive predictive value. Minimal evidence exists regarding the sensitivity of CT in patients who are asymptomatic. CONCLUSIONS: In surgical patients, CT chest should be considered as an important adjunct for detection of COVID-19 infection in patients who are symptomatic with negative RT-PCR prior to any operation. For surgical patients who are asymptomatic, there is insufficient evidence to recommend routine preoperative CT chest for COVID-19 screening.

Long-term patient-reported outcomes and quality of the evidence in ventral hernia mesh repair: a systematic review.

PURPOSE: Patient-reported outcome measures (PROMs) have been increasingly accepted to evaluate the quality of surgery. The impact of a hernia on PROMs and the indication for elective ventral hernia mesh repair are poorly researched. The primary objective in this systematic review was to provide evidence for PROM changes at least 3 months after elective ventral hernia mesh repair. Secondarily, a critical appraisal of the study quality was undertaken. METHODS: Ovid MEDLINE, PubMed, Ovid Embase, and CENTRAL were searched (year 2000-May 12, 2019) for studies reporting any of 21 predefined PROMs pre- and ≥ 3 months postoperatively following adult ventral hernia mesh repair. A pre-study defined analysis method was used to assess pre- vs. postoperative PROM changes. Quality assessment was guided by criteria formulated by the National Institutes of Health. RESULTS: The search yielded 11,438 potentially eligible studies of which 24 met the inclusion criteria. Most studies were of poor or moderate quality and one study was of high quality. There was no evidence for a clinically relevant postoperative improvement in any PROM following umbilical hernia repair or medium-sized incisional hernia repair. A clinically relevant postoperative improvement of pain, physical impairment, and social involvement was seen in patients with a large-sized incisional hernia repair and stoma-related complaints in patients with a medium-sized parastomal hernia repair. CONCLUSION: This analysis suggested that a minor subset of PROMs improved in patients undergoing large-sized incisional and medium-sized parastomal hernia repair. High-quality studies are imperative in this extremely often conducted surgical procedure. TRIAL REGISTRATION: The review was registered at PROSPERO (CRD42018096671, https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=96671 ).

The impact of preoperative anxiety, depression, and chronic pain on outcomes in abdominal wall reconstruction.

PURPOSE: An association of anxiety with surgical outcomes has been suggested, including with open ventral hernia repair (OVHR). This study examines the interaction of multiple comorbidities, including anxiety, depression, chronic pain, and hernia characteristics with outcomes after OVHR. METHODS: Patients with anxiety were identified in an existing, prospectively collected, data set of OVHR with preoperative work-up including CT scans (2007-2018). A patient with a diagnosis or prescription for anxiolytics, anti-depressants, or narcotics was considered to have anxiety, depression, or chronic pain, respectively. Hernia characteristics were analyzed using 3D volumetric software. Univariate and multivariate analyses were performed to assess for the impact of anxiety on surgical outcomes. RESULTS: A total of 1178 OVHRs were identified. The diagnosis of anxiety (23.9%) was associated with female gender (29.1% females vs. 16.9% males, p = 0.002), depression (56.7 vs. 18.8%, p < 0.0001), preoperative chronic pain (43.6 vs. 26.9%, p < 0.0001), COPD, arrhythmia, history of MRSA, and sleep apnea (p ≤ 0.05 all values). Patients with anxiety had larger hernia volume and defect size, and were more likely to undergo component separation, with higher rates of wound complication and intervention for pain (p ≤ 0.05 all values). After multivariate analysis controlling for multiple potentially confounding factors, the comorbidities of anxiety, depression, and preoperative chronic pain were not found to be significantly associated with adverse outcomes. CONCLUSIONS: The diagnosis of anxiety is associated with preoperative comorbidity, surgical complexity, and adverse outcomes after OVHR. However, when comorbidities are controlled for, the diagnosis of anxiety, depression or preoperative pain does not independently predict adverse outcomes. In this context, anxiety may be considered a marker of patient comorbidity in a complex patient population.

Chronic immunosuppressant use in colorectal cancer patients worsens postoperative morbidity and mortality through septic complications in a propensity-matched analysis.

AIM: Chronic immunosuppressant use increases the risk of septic complications after colectomy; however, adverse effects on other organ systems remain poorly understood. The aim of this study was to evaluate the multisystem organ effect(s) of chronic immunosuppressant(s) in colorectal cancer patients. METHODS: This was a retrospective study. The American College of Surgeons National Surgical Quality Improvement database (2005-2012) was queried. The primary end-points were 30-day mortality and 30-day morbidity after colectomy in patients on chronic immunosuppressant(s) compared to a non-immunosuppressant cohort. RESULTS: In total, 50 766 patients were identified, with 1203 (2.4%) taking chronic immunosuppressant(s). After propensity matching, 1197 patients in each cohort were evaluated with no differences seen in age, body mass index, male sex, wound classification, emergency case status, the presence of preoperative sepsis or operative time. On outcome analysis, 30-day mortality (5.7% vs 3.4%, P < 0.001) and 30-day overall morbidity (35.4% vs 29.0%, P = 0.001) were higher in patients on chronic immunosuppressant(s). Septic complications (10.6% vs 7.9%, P = 0.02) and surgical site infections (15.3% vs 12.3%, P = 0.03) were elevated with chronic immunosuppressant(s). There were no differences in cardiovascular, pulmonary, renal or neurological complications. Chronic immunosuppressant patients demonstrated longer total hospital stay (11.4 ± 11.7 vs 9.5 ± 9.4 days, P < 0.001) and postoperative length of stay (9.4 ± 9.2 vs 8.1 ± 7.6 days, P < 0.001). The limitation was that this was a retrospective study using a clinical dataset. CONCLUSION: In this study, immunosuppressant use is associated with worsened infective complications, without contributing to organ-specific complications following colectomy. Significant thought should be given to anastomosis vs stoma creation to possibly prevent worsened morbidity and mortality. Future study is required to determine specific pathways for risk reduction.

MELD-Na score associated with postoperative complications in hernia repair in non-cirrhotic patients.

PURPOSE: In patients with cirrhosis, the Model for End-Stage Liver Disease Sodium (MELD-Na) score is a validated predictor of outcomes after transplant and non-transplant surgical procedures. This study investigates the association of MELD-Na score with complications following elective ventral hernia repair in non-cirrhotic patients. METHODS: The ACS NSQIP database was queried (2005-2016) for all elective laparoscopic and open ventral hernia procedures in patients without ascites or esophageal varices. Postoperative outcomes were compared by MELD-Na score using Chi-square tests. Multivariate logistic regression was used to control for potentially confounding variables. RESULTS: A total of 48,955 elective hernia repairs were identified; 68.7% were open repairs. The overall complication rate (Clavien-Dindo ≥ 1) was 14.3%, with a wound complication rate of 5.5%, and major complication rate (Clavien-Dindo ≥ 3) of 4.3%. A preoperative MELD-Na score ≥ 10 was present in 29.4%. Incremental increases in MELD-Na score (10-14, 15-19, and ≥ 20) were associated with increased overall complications (OR 1.25, CI 1.31-1.37; OR 1.53, CI 1.30-1.80; OR 1.70, CI 1.24-2.31, respectively), major complications (OR 1.42, CI 1.20-1.69; OR 1.85, CI 1.43-2.39; OR 2.13, CI 1.35-3.38, respectively), 30-day mortality (OR 1.58, CI 1.05-2.37; OR 2.34, CI 1.39-3.96; OR 3.16, CI 1.37-7.28, respectively), and return to the operating room (OR 1.19, CI 1.01-1.41; OR 1.38, CI 1.05-1.81; OR 1.78, CI 1.10-2.90, respectively). CONCLUSION: MELD-Na score is independently associated with postoperative complications in ventral hernia repair. As an objective and simple predictive model, it may be useful in preoperative risk calculations for complex patients.

Quality of life and outcomes for femoral hernia repair: does laparoscopy have an advantage?

BACKGROUND: Due to their relative scarcity and to limit single-center bias, multi-center data are needed to study femoral hernias. The aim of this study was to evaluate outcomes and quality of life (QOL) following laparoscopic vs. open repair of femoral hernias. METHODS: The International Hernia Mesh Registry was queried for femoral hernia repairs. Laparoscopic vs. open techniques were assessed for outcomes and QOL, as quantified by the Carolinas Comfort Scale (CCS), preoperatively and at 1, 6, 12, and 24 months postoperatively. Outcomes were evaluated using the standard statistical analysis. RESULTS: A total of 80 femoral hernia repairs were performed in 73 patients: 37 laparoscopic and 43 open. There was no difference in mean age (54.7 ± 14.6 years), body mass index (24.2 ± 3.8 kg/m2), gender (60.3 % female), or comorbidities (p > 0.05). The hernias were recurrent in 21 % of the cases with an average of 1.23 ± 0.6 prior repairs (p > 0.1). Preoperative CCS scores were similar for both groups and indicated that 59.7 % of patients reported pain and 46.4 % had movement limitations (p > 0.05). Operative time was equivalent (47.2 ± 21.2 vs. 45.9 ± 14.8 min, p = 0.82). There was no difference in postoperative complications, with an overall 8.2 % abdominal wall complications rate (p > 0.05). The length of stay was shorter in the laparoscopic group (0.5 ± 0.6 vs. 1.3 ± 1.6 days, p = 0.02). Follow-up was somewhat longer in the open group (23.8 ± 10.2 vs. 17.3 ± 10.9 months, p = 0.02). There was one recurrence, which was in the laparoscopic group (3.1 vs. 0 %, p = 0.4). QOL outcomes at all time points demonstrated no difference for pain, movement limitation, or mesh sensation. Postoperative QOL scores improved for both groups when compared to preoperative scores. CONCLUSION: In this prospective international multi-institution study of 80 femoral hernia repairs, no difference was found for operative times, long-term outcomes, or QOL in the treatment of femoral hernias when comparing laparoscopic vs. open techniques. After repair, QOL at all time-points postoperatively improved compared to QOL scores preoperatively for laparoscopic and open femoral hernia repair. While international data supports improved outcomes with laparoscopic approach for femoral hernia repair, no data had existed prior to this study on the difference of approach impacting QOL. In the setting where recurrence and complication rates are equal after femoral hernia repair for either approach, surgeons should perform the technique with which they are most confident, as the operative approach does not appear to change QOL outcomes after femoral hernia repair.

Intraoperative indocyanine green fluorescence angiography to predict wound complications in complex ventral hernia repair.

INTRODUCTION: Complex ventral hernia repair (VHR) is associated with a greater than 30% wound complication rate. Perfusion mapping using indocyanine green fluorescence angiography (ICG-FA) has been demonstrated to predict skin and soft tissue necrosis in many reconstructive procedures; however, it has yet to be evaluated in VHR. METHODS: Patients undergoing complex VHR involving component separation and/or extensive subcutaneous advancement flaps were included in a prospective, blinded study. Patients with active infection were excluded. ICG-FA was performed prior to incision and prior to closure, but the surgeon was not allowed to view it. An additional blinded surgeon documented wound complications and evaluated postoperative photographs. The operative ICG-FA was reviewed blinded, and investigators were then unblinded to determine its ability to predict wound complications. RESULTS: Fifteen consecutive patients were enrolled with mean age of 56.1 years and average BMI of 34.9, of which 60% were female. Most (73.3%) had prior hernia repairs (average of 1.8 prior repairs). Mean defect area was 210.4 cm2, mean OR time was 206 min, 66.6% of patients underwent concomitant panniculectomy, and 40% had component separation. Mean follow-up was 7 months. Two patients developed wound breakdown requiring reoperation, while 1 had significant fat necrosis and another a wound infection, requiring operative intervention. ICG-FA was objectively reviewed and predicted all 4 wound complications. Of the 12 patients without complications, 1 had an area of low perfusion on ICG-FA. This study found a sensitivity of 100% and specificity of 90.9% for predicting wound complications using ICG-FA. CONCLUSION: In complex VHR patients, subcutaneous perfusion mapping with ICG-FA is very sensitive and has the potential to reduce cost and improve patient quality of life by reducing wound complications and reoperation.

Laparoscopic appendectomy and cholecystectomy versus open: a study in 1999 pregnant patients.

BACKGROUND: When pregnant patients require surgery, whether to perform an operation open or laparoscopic is often debated. We evaluated the impact of laparoscopy for common general surgical problems in pregnancy to determine safety and trends in operative approach over time. METHODS: Pregnant patients undergoing appendectomy or cholecystectomy were identified using the National Surgical Quality Improvement Program (NSQIP) database. We analyzed demographics, operative characteristics, and outcomes. Univariate comparison and multivariate regression analysis (MVA) were performed adjusting for confounding factors: age, body mass index (BMI), diabetes, and smoking, and an additional MVA was performed for perforated cases. RESULTS: A total of 1999 pregnant patients between 2005 and 2012 were evaluated. Of 1335 appendectomies, 894 were performed laparoscopically (LA) and 441 open (OA). For 664 cholecystectomies, 606 were laparoscopic (LC) and 58 open (OC). There were no deaths. For LA versus OA, patient characteristics were not different {age: 27.7 vs. 28.2 years, p = 0.19; diabetes: 1.8 vs. 0.9%, p = 0.24; smoking: 19 vs. 16.1%, p = 0.2} except for BMI (27.9 vs. 28.4 kg/m(2); p = 0.03). LA had shorter operative times (ORT), length of stay (LOS), and fewer postoperative complications compared to OA. In MVA, difference between approaches remained statistically significant for ORT (<0.0001), LOS (<0.01), and wound complications (<0.01). MVA was performed for perforated cases alone: LA had equal ORT (p = 0.19) yet shorter LOS (p = <0.001). The majority of LA were performed in the last 4 years versus the first 4 years (61 vs. 39%, p < 0.001). For LC versus OC, patient characteristics were not different: age (28.3 vs. 28.7 years; p = 0.33), BMI (31.4 vs. 33.2 kg/m(2), p = 0.25), diabetes (2.8 vs. 3.5%, p = 0.68), and smoking (21.1 vs. 25.9%, p = 0.4). LC had a shorter ORT, LOS, and fewer postoperative complications than OC. In MVA, the difference between approaches remained statistically significant for ORT (<0.0001), LOS (<0.0001), and minor complications (<0.01). In MVA for cholecystitis with perforation, no difference was seen for LOS, ORT, or postoperative complications (p > 0.05). The percentage of LC cases appeared to increase over time (89 vs. 93%, p = 0.06). CONCLUSION: While fetal events are unknown, LA and LC in pregnant patients demonstrated shorter ORT, LOS, and reduced complications and were performed more frequently over time. Even in perforated cases, laparoscopy appears safe in pregnant patients.

Risk factors for postoperative sepsis in laparoscopic gastric bypass.

INTRODUCTION: Postoperative sepsis is a rare but serious complication following elective surgery. The purpose of this study was to identify the rate of postoperative sepsis following elective laparoscopic gastric bypass (LGBP) and to identify patients' modifiable, preoperative risk factors. METHODS: The American College of Surgeons National Surgical Quality Improvement Program database was queried from 2005 to 2013 for factors associated with the development of postoperative sepsis following elective LGBP. Patients who developed sepsis were compared to those who did not. Results were analyzed using the Chi-square test for categorical variables and Wilcoxon two-sample test for continuous variables. A multivariate logistic regression analysis was utilized to calculate adjusted odds ratios for factors contributing to sepsis. RESULTS: During the study period, 66,838 patients underwent LGBP. Of those, 546 patients developed postoperative sepsis (0.82%). The development of sepsis was associated with increased operative time (161 ± 77.8 vs. 135.10 ± 56.5 min; p < 0.0001) and a greater number of preoperative comorbidities, including diabetes (39.6 vs. 30.6%; p < 0.0001), hypertension requiring medication (65.2 vs. 54%; p < 0.0001), current tobacco use (16.7 vs. 11.5%; p = 0.0002), and increased pack-year history of smoking (8.6 ± 18.3 vs. 5.6 ± 14.2; p = 0.0006), and the Charlson Comorbidity Index (0.51 ± 0.74 vs. 0.35 ± 0.57, p < 0.0001). Sepsis resulted in an increased length of stay (10.1 ± 14.4 vs. 2.4 ± 4.8 days; p < 0.0001) and a 30 times greater chance of 30-day mortality (4.03 vs. 0.11%, p < 0.0001). Multivariate logistic regression analysis showed that current smokers had a 63% greater chance of developing sepsis compared to non-smokers, controlling for age, race, gender, BMI, and CCI score (OR 1.63, 95% CI 1.23-2.14; p = 0.0006). CONCLUSIONS: Laparoscopic gastric bypass is uncommonly associated with postoperative sepsis. When it occurs, it portends a 30 times increased risk of death. A patient history of diabetes, hypertension, and increasing pack-years of smoking portend an increased risk of sepsis. Current smoking status, a preoperative modifiable risk factor, is independently associated with the chance of postoperative sepsis. Preoperative patient optimization and risk reduction should be a priority for elective surgery, and patients should be encouraged to stop smoking prior to gastric bypass.