Personalized Surgical Transfusion Risk Prediction Using Machine Learning to Guide Preoperative Type and Screen Orders.

BACKGROUND: Accurate estimation of surgical transfusion risk is essential for efficient allocation of blood bank resources and for other aspects of anesthetic planning. This study hypothesized that a machine learning model incorporating both surgery- and patient-specific variables would outperform the traditional approach that uses only procedure-specific information, allowing for more efficient allocation of preoperative type and screen orders. METHODS: The American College of Surgeons National Surgical Quality Improvement Program Participant Use File was used to train four machine learning models to predict the likelihood of red cell transfusion using surgery-specific and patient-specific variables. A baseline model using only procedure-specific information was created for comparison. The models were trained on surgical encounters that occurred at 722 hospitals in 2016 through 2018. The models were internally validated on surgical cases that occurred at 719 hospitals in 2019. Generalizability of the best-performing model was assessed by external validation on surgical cases occurring at a single institution in 2020. RESULTS: Transfusion prevalence was 2.4% (73,313 of 3,049,617), 2.2% (23,205 of 1,076,441), and 6.7% (1,104 of 16,053) across the training, internal validation, and external validation cohorts, respectively. The gradient boosting machine outperformed the baseline model and was the best- performing model. At a fixed 96% sensitivity, this model had a positive predictive value of 0.06 and 0.21 and recommended type and screens for 36% and 30% of the patients in internal and external validation, respectively. By comparison, the baseline model at the same sensitivity had a positive predictive value of 0.04 and 0.144 and recommended type and screens for 57% and 45% of the patients in internal and external validation, respectively. The most important predictor variables were overall procedure-specific transfusion rate and preoperative hematocrit. CONCLUSIONS: A personalized transfusion risk prediction model was created using both surgery- and patient-specific variables to guide preoperative type and screen orders and showed better performance compared to the traditional procedure-centric approach.

Antireflux Surgery at National Surgical Quality Improvement Program-Pediatric Hospitals.

PURPOSE: This study aims to examine contemporary practice patterns and compare short-term outcomes for vesicoureteral reflux procedures (ureteral reimplant/endoscopic injection) using National Surgical Quality Improvement Program-Pediatric data. MATERIALS AND METHODS: Procedure-specific variables for antireflux surgery were developed to capture data not typically collected in National Surgical Quality Improvement Program-Pediatric (eg vesicoureteral reflux grade, urine cultures, 31-60-day followup). Descriptive statistics were performed, and logistic regression assessed associations between patient/procedural factors and outcomes (urinary tract infection, readmissions, unplanned procedures). RESULTS: In total, 2,842 patients (median age 4 years; 76% female; 68% open reimplant, 6% minimally invasive reimplant, 25% endoscopic injection) had procedure-specific variables collected from July 2016 through June 2018. Among 88 hospitals, a median of 24.5 procedures/study period were performed (range 1-148); 95% performed ≥1 open reimplant, 30% ≥1 minimally invasive reimplant, and 70% ≥1 endoscopic injection, with variability by hospital. Two-thirds of patients had urine cultures sent preoperatively, and 76% were discharged on antibiotics. Outcomes at 30 days included emergency department visits (10%), readmissions (4%), urinary tract infections (3%), and unplanned procedures (2%). Over half of patients (55%) had optional 31-60-day followup, with additional outcomes (particularly urinary tract infections) noted. Patients undergoing reimplant were younger, had higher reflux grades, and more postoperative occurrences than patients undergoing endoscopic injections. CONCLUSIONS: Contemporary data indicate that open reimplant is still the most common antireflux procedure, but procedure distribution varies by hospital. Emergency department visits are common, but unplanned procedures are rare, particularly for endoscopic injection. These data provide basis for comparing short-term complications and developing standardized perioperative pathways for antireflux surgery.

Association of Patient-reported Experiences and Surgical Outcomes Among Group Practices: Retrospective Cohort Study.

OBJECTIVE: The aim of the study was to determine the association of patient-reported experiences (PREs) and risk-adjusted surgical outcomes among group practices. BACKGROUND: The Centers for Medicare and Medicaid Services required large group practices to submit PREs data for successful participation in the Physician Quality Reporting System (PQRS) using the Consumer Assessment of Healthcare Providers and Systems for PQRS survey. Whether these PREs data correlate with perioperative outcomes remains ill defined. METHODS: Operations between January 1, 2014 and December 31, 2016 in the American College of Surgeons' National Surgical Quality Improvement Program registry were merged with 2015 Consumer Assessment of Healthcare Providers and Systems for PQRS survey data. Hierarchical logistic models were constructed to estimate associations between 7 subscales and 1 composite score of PREs and 30-day morbidity, unplanned readmission, and unplanned reoperation, separately, while adjusting for patient- and procedure mix. RESULTS: Among 328 group practices identified, patients reported their experiences with clinician communication the highest (mean ±â€Šstandard deviation, 82.66 ±â€Š3.10), and with attention to medication cost the lowest (25.96 ±â€Š5.14). The mean composite score was 61.08 (±6.66). On multivariable analyses, better PREs scores regarding medication cost, between-visit communication, and the composite score of experience were each associated with 4% decreased odds of morbidity [odds ratio (OR) 0.96, 95% confidence interval (CI) 0.92-0.99], readmission (OR 0.96, 95% CI 0.93-0.99), and reoperation (OR 0.96, 95% CI 0.93-0.99), respectively. In sensitivity analyses, better between-visit communication remained significantly associated with fewer readmissions. CONCLUSIONS: In these data, patients' report of better between-visit communication was associated with fewer readmissions. More sensitive, surgery-specific PRE assessments may reveal additional unique insights for improving the quality of surgical care.

Beyond death and graft survival-Variation in outcomes after kidney transplantation. Results from the NSQIP Transplant beta phase.

The National Surgical Quality Program (NSQIP) Transplant was designed by transplant surgeons from the ground up to track posttransplant outcomes beyond basic recipient and graft survival. After an initial pilot phase, the program has expanded to 29 participating sites and enrolled more than 4300 recipient-donor pairs into the database, including 2876 complete kidney transplant cases. In this analysis, surgical site infection (SSI), urinary tract infection (UTI), and reoperation/intervention were evaluated for kidney transplant recipients. We observed impressive variation in the crude incidence between sites for SSI (0%-17%), UTI (0%-14%), and reoperation/intervention (0%-25%). After adjustment for donor and recipient factors, 2 sites were outliers with respect to their incidence of UTI. For the first time, the field of transplantation has data that demonstrate variation in kidney recipient surgical outcomes between sites. More importantly, NSQIP Transplant provides a powerful platform to improve care beyond basic patient and graft survival.

Beyond death and graft survival-Variation in outcomes after liver transplant. Results from the NSQIP transplant beta phase.

The National Surgical Quality Program (NSQIP) Transplant program was designed by transplant surgeons from the ground up to track posttransplant outcomes beyond basic recipient and graft survival. After an initial pilot phase, the program has expanded to 29 participating sites and enrolled more than 4300 recipient-donor pairs into the database, including 1444 completed liver transplant cases. In this analysis, surgical site infection (SSI), urinary tract infection (UTI), and unplanned reoperation/intervention after liver transplantation were evaluated. We observed impressive variation in the crude incidence between sites for SSI (0%-29%), UTI (0%-10%), and reoperation/intervention (0%-57%). After adjustment for donor and recipient factors, at least 1 site was identified as an outlier for each of the analyzed outcomes. For the first time, the field of transplantation has data that demonstrate variation in liver recipient outcomes beyond death and graft survival between sites. More importantly, NSQIP Transplant provides a powerful platform to improve care beyond basic patient and graft survival.

Prevalence and Risk Factors for Bariatric Surgery Readmissions: Findings From 130,007 Admissions in the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program.

OBJECTIVE: To evaluate readmissions following laparoscopic adjustable gastric banding (LAGB), laparoscopic sleeve gastrectomy (LSG), and laparoscopic Roux-en-Y gastric bypass (LRYGB). BACKGROUND: Few studies have evaluated national readmission rates for primary bariatric surgery with national, bariatric-specific data. METHODS: Patients undergoing primary LAGB, LSG, or LRYGB from January 1, 2014 to December 31, 2014, at 698 centers were identified based upon Current Procedural Terminology codes. The primary outcome was 30-day readmission from date of initial operation. RESULTS: A total of 130,007 patients who underwent primary bariatric surgery were identified: 7378 LAGB (5.7%), 80,646 LSG (62.0%), and 41,983 LRYGB (32.3%). A total of 5663 (4.4%) patients were readmitted within 30 days for all causes. Patients undergoing LAGB had the lowest related readmission rate of 1.4%, followed by LSG (2.8%), and LRYGB (4.9%). Of patients who had a complication, 17.9% (n = 785) were readmitted, whereas those without readmission had a complication 1.9% of the time (P < 0.001). The most common cause of a related readmission was nausea, vomiting, fluid, electrolyte, and nutritional depletion (35.4%), followed by abdominal pain (13.5%), anastomotic leak (6.4%), and bleeding (5.8%), accounting for more than 61% of readmissions. When compared with LAGB, LSG, and LRYGB had significantly higher rates of readmission (LSG: odds ratio 1.89; 95% confidence interval 1.52-2.33; LRYGB: odds ratio 3.06; 95% confidence interval 2.46-3.81). CONCLUSIONS: National bariatric readmissions after primary procedures were closely associated with complications, varied based on the type of procedure, and were most commonly due to nausea, vomiting, electrolyte, and nutritional depletion.

Hospital variation in rates of concurrent fundoplication during gastrostomy enteral access procedures.

BACKGROUND: This study aimed to determine whether (1) the propensity for concurrent fundoplication during gastrostomy varies among hospitals, and (2) postoperative morbidity differs among institutions performing fundoplication more or less frequently. METHODS: Children who underwent gastrostomy with or without concurrent fundoplication were identified in the American College of Surgeons National Surgical Quality Improvement Program Pediatric (ACS-NSQIP-P). A hierarchical multivariate regression modeled the excess effects that hospitals exerted over propensity for concurrent fundoplication adjusting for preoperative clinical variables. Hospitals were designated as low outliers (significantly lower-adjusted odds of concurrent fundoplication than the average hospital with similar patient mix), average hospitals, and high outliers based on their risk-adjusted concurrent fundoplication practice. The postoperative morbidity rates were compared among low-outlier, average, and high-outlier hospitals. RESULTS: Between 2011 and 2013, 3775 children underwent gastrostomy at one of 54 ACS-NSQIP-P participating hospitals. The mean hospital concurrent fundoplication rate was 11.7% (range 0-64%). There was no significant difference in unadjusted morbidity rate in children with concurrent fundoplication, 11.0% compared to 9.7% in children without concurrent fundoplication. After controlling for clinical variables, 8 hospitals were identified as low outliers (fundoplication rate of 0.4%) and 16 hospitals were identified as high outliers (fundoplication rate of 34.6%). The average unadjusted morbidity rate among hospitals with low, average, and high odds of concurrent fundoplication were 9.6, 10.6, and 8.4%, respectively. CONCLUSION: Hospitals vary significantly in propensity for concurrent fundoplication during gastrostomy yet postoperative morbidity does not differ significantly among institutions performing fundoplication more or less frequently.

Outcomes of Concurrent Operations: Results From the American College of Surgeons' National Surgical Quality Improvement Program.

OBJECTIVE: To determine whether concurrently performed operations are associated with an increased risk for adverse events. BACKGROUND: Concurrent operations occur when a surgeon is simultaneously responsible for critical portions of 2 or more operations. How this practice affects patient outcomes is unknown. METHODS: Using American College of Surgeons' National Surgical Quality Improvement Program data from 2014 to 2015, operations were considered concurrent if they overlapped by ≥60 minutes or in their entirety. Propensity-score-matched cohorts were constructed to compare death or serious morbidity (DSM), unplanned reoperation, and unplanned readmission in concurrent versus non-concurrent operations. Multilevel hierarchical regression was used to account for the clustered nature of the data while controlling for procedure and case mix. RESULTS: There were 1430 (32.3%) surgeons from 390 (77.7%) hospitals who performed 12,010 (2.3%) concurrent operations. Plastic surgery (n = 393 [13.7%]), otolaryngology (n = 470 [11.2%]), and neurosurgery (n = 2067 [8.4%]) were specialties with the highest proportion of concurrent operations. Spine procedures were the most frequent concurrent procedures overall (n = 2059/12,010 [17.1%]). Unadjusted rates of DSM (9.0% vs 7.1%; P < 0.001), reoperation (3.6% vs 2.7%; P < 0.001), and readmission (6.9% vs 5.1%; P < 0.001) were greater in the concurrent operation cohort versus the non-concurrent. After propensity score matching and risk-adjustment, there was no significant association of concurrence with DSM (odds ratio [OR] 1.08; 95% confidence interval [CI] 0.96-1.21), reoperation (OR 1.16; 95% CI 0.96-1.40), or readmission (OR 1.14; 95% CI 0.99-1.29). CONCLUSIONS: In these analyses, concurrent operations were not detected to increase the risk for adverse outcomes. These results do not lessen the need for further studies, continuous self-regulation and proactive disclosure to patients.

Pancreatoduodenectomy with venous or arterial resection: a NSQIP propensity score analysis.

INTRODUCTION: Vascular resection during pancreatoduodenectomy (PD) is being performed more frequently. Our aim was to analyze the outcomes of PD with and without vascular resection in a large, multicenter cohort. METHODS: Patient data were gathered from 43 institutions as part of the American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) Pancreatectomy Demonstration Project. Over a 14-month period, 1414 patients underwent PD without (82.2%) or with major venous (PD + V; 13.7%) or arterial (PD + A; 4.0%) vascular resection. RESULTS: Postoperative morbidity and mortality following PD + A (51.0% and 3.6%) was comparable to PD + V (46.9% and 3.6%) and PD (44.3 and 1.5%, p = 0.50 and 0.43). A propensity score matched analysis revealed that vascular resection was associated with significant increases (p ≤ 0.05) in operative time (7:37 vs 6:11), need for blood transfusion (42.2% vs 18.1%), deep venous thromboembolism (6.9% vs 0.9%), postoperative septic shock (6.9% vs 1.7%), and length of stay (12.2 vs 10 days) while overall morbidity (45.7% vs 46.6) and mortality (1.0% vs 0%) were comparable. CONCLUSIONS: Compared to PD alone, PD + VR was associated with increased operative time, perioperative transfusions, deep venous thrombosis, septic shock, as well as length of stay, but overall morbidity and mortality were not increased.

Minimally invasive hepatopancreatobiliary surgery in North America: an ACS-NSQIP analysis of predictors of conversion for laparoscopic and robotic pancreatectomy and hepatectomy.

BACKGROUND: Procedural conversion rates represent an important aspect of the feasibility of minimally invasive surgical (MIS) approaches. This study aimed to outline the rates and predictors of procedural completion/conversion for MIS hepatectomy and pancreatectomy. METHODS: All 2014 ACS-NSQIP laparoscopic and robotic hepatectomy and pancreatectomy procedures were identified and grouped into pure, open assist, or unplanned conversion to open. Risk adjusted multinomial logistic regression models were generated with completion (Pure) set as the primary outcome. RESULTS: 1667 (laparoscopic = 1360, robotic = 307) resections were captured. After risk adjustment, robotic DP was associated with similar open assist (relative risk ratio -1.9%, P = 0.602), but lower unplanned conversion (-8.2%, P = 0.004) and open assist + unplanned conversion (-10.1%, P = 0.015) compared to laparoscopic DP; while robotic PD was associated with lower open assist (-22.2%, P < 0.001), unplanned conversions (-15%, P = 0.006) and open assist + unplanned conversions (-37.2, P < 0.001) compared to laparoscopic PD. The robotic and laparoscopic approaches to hepatectomy were not associated with differences in pure MIS completion rates (P = NS) after risk adjustment. CONCLUSIONS: The robotic approach to pancreatectomy was associated with higher rates of pure MIS completion compared to laparoscopy, whereas no difference in MIS completion rates was noted for robotic versus laparoscopic hepatectomy.

On-demand Reporting of Risk-adjusted and Smoothed Rates for Quality Profiling in ACS NSQIP.

BACKGROUND: Surgical quality improvement depends on hospitals having accurate and timely information about comparative performance. Profiling accuracy is improved by risk adjustment and shrinkage adjustment to stabilize estimates. These adjustments are included in ACS NSQIP reports, where hospital odds ratios (OR) are estimated using hierarchical models built on contemporaneous data. However, the timeliness of feedback remains an issue. STUDY DESIGN: We describe an alternative, nonhierarchical approach, which yields risk- and shrinkage-adjusted rates. In contrast to our "Traditional" NSQIP method, this approach uses preexisting equations, built on historical data, which permits hospitals to have near immediate access to profiling results. We compared our traditional method to this new "on-demand" approach with respect to outlier determinations, kappa statistics, and correlations between logged OR and standardized rates, for 12 models (4 surgical groups by 3 outcomes). RESULTS: When both methods used the same contemporaneous data, there were similar numbers of hospital outliers and correlations between logged OR and standardized rates were high. However, larger differences were observed when the effect of contemporaneous versus historical data was added to differences in statistical methodology. CONCLUSIONS: The on-demand, nonhierarchical approach provides results similar to the traditional hierarchical method and offers immediacy, an "over-time" perspective, application to a broader range of models and data subsets, and reporting of more easily understood rates. Although the nonhierarchical method results are now available "on-demand" in a web-based application, the hierarchical approach has advantages, which support its continued periodic publication as the gold standard for hospital profiling in the program.

Improved Surgical Outcomes for ACS NSQIP Hospitals Over Time: Evaluation of Hospital Cohorts With up to 8 Years of Participation.

BACKGROUND: The American College of Surgeons, National Surgical Quality Improvement Program (ACS NSQIP) surgical quality feedback models are recalibrated every 6 months, and each hospital is given risk-adjusted, hierarchical model, odds ratios that permit comparison to an estimated average NSQIP hospital at a particular point in time. This approach is appropriate for "relative" benchmarking, and for targeting quality improvement efforts, but does not permit evaluation of hospital or program-wide changes in quality over time. We report on long-term improvement in surgical outcomes associated with participation in ACS NSQIP. STUDY DESIGN: ACS NSQIP data (2006-2013) were used to create prediction models for mortality, morbidity (any of several distinct adverse outcomes), and surgical site infection (SSI). For each model, for each hospital, and for year of first participation (hospital cohort), hierarchical model observed/expected (O/E) ratios were computed. The primary performance metric was the within-hospital trend in logged O/E ratios over time (slope) for mortality, morbidity, and SSI. RESULTS: Hospital-averaged log O/E ratio slopes were generally negative, indicating improving performance over time. For all hospitals, 62%, 70%, and 65% of hospitals had negative slopes for mortality, morbidity, and any SSI, respectively. For hospitals currently in the program for at least 3 years, 69%, 79%, and 71% showed improvement in mortality, morbidity, and SSI, respectively. For these hospitals, we estimate 0.8%, 3.1%, and 2.6% annual reductions (with respect to prior year's rates) for mortality, morbidity, and SSI, respectively. CONCLUSIONS: Participation in ACS NSQIP is associated with reductions in adverse events after surgery. The magnitude of quality improvement increases with time in the program.

Evaluation of the ProPublica Surgeon Scorecard "Adjusted Complication Rate" Measure Specifications.

OBJECTIVES: The ProPublica Surgeon Scorecard is the first nationwide, multispecialty public reporting of individual surgeon outcomes. However, ProPublica's use of a previously undescribed outcome measure (composite of in-hospital mortality or 30-day related readmission) and inclusion of only inpatients have been questioned. Our objectives were to (1) determine the proportion of cases excluded by ProPublica's specifications, (2) assess the proportion of inpatient complications excluded from ProPublica's measure, and (3) examine the validity of ProPublica's outcome measure by comparing performance on the measure to well-established postoperative outcome measures. METHODS: Using ACS-NSQIP data (2012-2014) for 8 ProPublica procedures and for All Operations, the proportion of cases meeting all ProPublica inclusion criteria was determined. We assessed the proportion of complications occurring inpatient, and thus not considered by ProPublica's measure. Finally, we compared risk-adjusted performance based on ProPublica's measure specifications to established ACS-NSQIP outcome measure performance (eg, death/serious morbidity, mortality). RESULTS: ProPublica's inclusion criteria resulted in elimination of 82% of all operations from assessment (range: 42% for total knee arthroplasty to 96% for laparoscopic cholecystectomy). For all ProPublica operations combined, 84% of complications occur during inpatient hospitalization (range: 61% for TURP to 88% for total hip arthroplasty), and are thus missed by the ProPublica measure. Hospital-level performance on the ProPublica measure correlated weakly with established complication measures, but correlated strongly with readmission (R = 0.834, P < 0.001). CONCLUSIONS: ProPublica's outcome measure specifications exclude 82% of cases, miss 84% of postoperative complications, and correlate poorly with well-established postoperative outcomes. Thus, the validity of the ProPublica Surgeon Scorecard is questionable.

Hospital-level Variation in Secondary Complications After Surgery.

OBJECTIVES: To assess whether hospital rates of secondary complications could serve as a performance benchmark and examine associations with mortality. BACKGROUND: Failure to rescue (death after postoperative complication) is a challenging target for quality improvement. Secondary complications (complications after a first or "index" complication) are intermediate outcomes in the rescue process that may provide specific improvement targets and give us insight into how rescue fails. METHODS: We used American College of Surgeons' National Surgical Quality Improvement Program data (2008-2012) to define hospital rates of secondary complications after 5 common index complications: pneumonia, surgical site infection (SSI), urinary tract infection, transfusion/bleed events, and acute myocardial infarction (MI). Hospitals were divided into quintiles on the basis of risk- and reliability-adjusted rates of secondary complications, and these rates were compared along with mortality. RESULTS: A total of 524,860 patients were identified undergoing one of the 62 elective, inpatient operations. After index pneumonia, secondary complication rates varied from 57.99% in the highest quintile to 22.93% in the lowest [adjusted odds ratio (OR), 4.64; confidence interval (CI), 3.95-5.45). Wide variation was seen after index SSI (58.98% vs 14.81%; OR, 8.53; CI, 7.41-9.83), urinary tract infection (38.41% vs 8.60%; OR, 7.81; CI, 6.48-9.40), transfusion/bleeding events (27.14% vs 12.88%; OR, 2.54; CI, 2.31-2.81), and acute MI (64.45% vs 23.86%, OR, 6.87; CI, 5.20-9.07). Hospitals in the highest quintile had significantly greater mortality after index pneumonia (10.41% vs 6.20%; OR, 2.17; CI, 1.6-2.94), index MI (18.25% vs 9.65%; OR, 2.67; CI, 1.80-3.94), and index SSI (2.75% vs 0.82%; OR, 3.93; CI, 2.26-6.81). CONCLUSIONS: Hospital-level rates of secondary complications (failure to arrest complications) vary widely, are associated with mortality, and may be useful for quality improvement and benchmarking.

Relative Contributions of Complications and Failure to Rescue on Mortality in Older Patients Undergoing Pancreatectomy.

BACKGROUND: For pancreatectomy patients, mortality increases with increasing age. Our study evaluated the relative contribution of overall postoperative complications and failure to rescue rates on the observed increased mortality in older patients undergoing pancreatic resection at specialized centers. METHODS: We identified 2694 patients who underwent pancreatic resection from the American College of Surgeons' National Surgical Quality Improvement Pancreatectomy Demonstration Project at 37 high-volume centers. Overall morbidity and in-hospital mortality were determined in patients younger than 80 years (N = 2496) and 80 years or older (N = 198). Failure to rescue was the number of deaths in patients with complications divided by the total number of patients with postoperative complications. RESULTS: No significant differences were observed between patients younger than 80 years and those 80 years or older in the rates of overall complications (41.4% vs 39.4%, P = 0.58). In-hospital mortality increased in patients 80 years or older compared to patients younger than 80 years (3.0% vs 1.1%, P = 0.02). Failures to rescue rates were higher in patients 80 years or older (7.7% vs 2.7%, P = 0.01). Across 37 high-volume centers, unadjusted complication rates ranged from 25.0% to 72.2% and failure to rescue rates ranged from 0.0% to 25.0%. Among patients with postoperative complications, comorbidities associated with failure to rescue were ascites, chronic obstructive pulmonary disease, and diabetes. Complications associated with failure to rescue included acute renal failure, septic shock, and postoperative pulmonary complications. CONCLUSIONS: In experienced hands, the rates of complications after pancreatectomy in patients 80 years or older compared to patients younger than 80 years were similar. However, when complications occurred, older patients were more likely to die. Interventions to identify and aggressively treat complications are necessary to decrease mortality in vulnerable older patients.

Variation in intraoperative and postoperative red blood cell transfusion in pediatric surgery.

BACKGROUND: Intraoperative and postoperative red blood cell (RBC) transfusions are relatively frequent events tracked in the American College of Surgeons' National Surgical Quality Improvement Program-Pediatric (ACS-NSQIP-P). This study sought to quantify variation in RBC transfusion practices among hospitals. STUDY DESIGN AND METHODS: This is an observational study of children older than 28 days who underwent a general, neurologic, urologic, otolaryngologic, plastic, or orthopedic operation at 50 hospitals in participating in the ACS-NSQIP-P during 2011 to 2012. The primary outcome was whether or not a RBC transfusion was administered from incision time to 72 hours postoperatively. Transfusions of fresh-frozen plasma, cryoprecipitate, and platelets were excluded from data abstraction due the rarity of their administration. A multivariate hierarchical risk-adjustment model estimated the risk-adjusted hospital RBC transfusion odds ratio (OR) and designated hospitals by transfusion practice. RESULTS: The mean RBC transfusion rate was 1.5%. Five preoperative variables were associated with greater than threefold increased odds of having an intraoperative or postoperative RBC transfusion; young age; 29 days to 1 year (OR, 5.9; p < 0.001) and 1 to 2 years (OR, 3.4; p < 0.001); American Society of Anesthesiologists Class IV (OR, 3.2; p < 0.001); procedure linear risk (OR, 3.1; p < 0.001); preoperative septic shock (OR, 14.5; p < 0.001); and preoperative cardiopulmonary resuscitation (OR, 8.1; p < 0.001). Twenty-five hospitals had RBC transfusion practices significantly different than risk-adjusted mean (17 higher and eight lower). CONCLUSION: Intraoperative and postoperative RBC transfusion practices vary widely among hospitals after controlling for patient and procedural characteristics.

Patterns of hepatic resections in North America: use of concurrent partial resections and ablations.

BACKGROUND: National registries have not adequately captured concurrent partial hepatic resections or ablations. Therefore, the aim of this analysis was to describe the patterns of concurrent partial resections and ablations in North America. METHODS: Patients undergoing a hepatic resection were identified using the American College of Surgeons-National Surgical Quality Improvement Program Targeted Hepatectomy database. Perioperative outcomes were compared for patients undergoing concurrent "wedge" resections and/or ablations and other subsets. RESULTS: A total of 2714 patients were identified who met inclusion criteria. Major hepatectomy was performed in 1037 patients (38.2%) while partial lobectomy was performed in 1677 (61.8%) patients. Concurrent "wedge" hepatic resections and ablations were undertaken in 56.0% and 14.2% of patients, respectively, and were more frequently performed among patients undergoing a partial lobectomy and among patients undergoing surgery for colorectal liver metastasis (both p < 0.001). While associated with a decreased incidence of postoperative complications (p = 0.027) and liver failure (p = 0.031) among patients undergoing a major hepatectomy, concurrent therapies were associated with comparable 30-day outcomes for patients undergoing partial lobectomy. CONCLUSION: Concurrent "wedge" hepatic resections and ablations are performed in 56.0% and 14.2%, respectively of patients undergoing hepatectomy. Concurrent procedures were not associated with worse clinical outcomes.