Preoperative Transfusion Research Articles

Application of machine learning to the prediction of postoperative sepsis after appendectomy

BackgroundWe applied various machine learning algorithms to a large national dataset to model the risk of postoperative sepsis after appendectomy to evaluate utility of such methods and identify factors associated with postoperative sepsis in these patients. MethodsThe National Surgery Quality Improvement Program database was used to identify patients undergoing appendectomy between 2005 and 2017. Logistic regression, support vector machines, random forest decision trees, and extreme gradient boosting machines were used to model the occurrence of postoperative sepsis. ResultsIn the study, 223,214 appendectomies were identified; 2,143 (0.96%) were indicated as having postoperative sepsis. Logistic regression (area under the curve 0.70; 95% confidence interval, 0.68−0.73), random forest decision trees (area under the curve 0.70; 95% confidence interval, 0.68−0.73), and extreme gradient boosting (area under the curve 0.70; 95% confidence interval, 0.68−0.73) afforded similar performance, while support vector machines (area under the curve 0.51; 95% confidence interval, 0.50−0.52) had worse performance. Variable importance analyses identified preoperative congestive heart failure, transfusion, and acute renal failure as predictors of postoperative sepsis. ConclusionMachine learning methods can be used to predict the development of sepsis after appendectomy with moderate accuracy. Such predictive modeling has potential to ultimately allow for preoperative recognition of patients at risk for developing postoperative sepsis after appendectomy thus facilitating early intervention and reducing morbidity.

Preoperative plasma fibrinogen level and transfusion in cardiac surgery: a biphasic correlation.

Coagulopathic bleeding is a serious complication of cardiac surgery. A very low preoperative plasma fibrinogen level (PFL) has been previously described as a risk factor for red blood cell (RBC) transfusion in the perioperative setting. Nevertheless, contradictory results have been published concerning preoperative fibrinogen infusion to decrease RBC transfusion. These results highlight the need for a better description of the relation between preoperative PFL and RBC transfusion. In this single-centre retrospective study, we showed that both high and low PFLs are associated with an increase in RBC transfusion. A pro-inflammatory preoperative status could be involved in the association between high PFL and the increase in RBC transfusion. This non-linear relation between PFL and RBC transfusion could explain the difficulty of decreasing postoperative bleeding using only an algorithm aimed at increasing preoperative PFL.

Risk factors for complications after abdominal surgery in children with sickle cell disease

BackgroundAbdominal surgery in children with sickle cell disease (SCD) carries an increased risk of postoperative complications. Preoperative transfusions are frequently given to decrease the risk of vasoocclusive events. However, risk factors for postoperative complications are not well-defined in the pediatric population. MethodsPediatric patients with SCD undergoing common abdominal operations were identified from the National Surgical Quality Improvement Program–Pediatric (NSQIP-P) database from 2012 to 2018. Outcomes of interest were the incidence rates of 1) any complication or readmission, and 2) serious SCD-related complications (stroke, new onset seizure, ventilator support >24 h postoperatively, or readmission with SCD crisis within 30 days of surgery).Patients were categorized by transfusion approach (transfusion within 48 h before surgery vs. no transfusion) and preoperative hematocrit (<21.0, 21.0–23.9, 24.0–26.9, 27.0–29.9, ≥30.0).Stratified bivariate analyses and multivariable logistic regression were used to identify independent risk factors for complications. ResultsA total of 813 patients met inclusion criteria. There were 470 cholecystectomy, 251 splenectomy, 39 appendectomy, and 53 combination procedures; 13% of cases were urgent or emergent. Preoperative hematocrit levels were <21.0 in 3%, 21.0–23.9 in 10%, 24.0–26.9 in 17%, 27.0–29.9in 30%, and ≥30.0 in 41% of patients; 52% received perioperative transfusion. The 30-day incidences of any complication/readmission and SCD-related complications were 12% and 4%, respectively. On bivariate analyses, urgent/emergent case status was the only significant predictor of complications, carrying risk of 20% and 8% for overall and SCD-related complications, respectively; this finding persisted on multivariable logistic regression (OR 1.83, 95% CI 1.0.2–3.29, p = 0.04). Neither preoperative transfusion nor preoperative hematocrit level was associated with complication risk, although there was a trend toward higher SCD-related complications in patients with preoperative hematocrit <21.0 (p = 0.07). ConclusionIn this large cohort of pediatric SCD patients undergoing abdominal surgery, there was no clear association between postoperative complications and the transfusion approach or the preoperative hematocrit level within the range above 21.0. Urgent/emergent surgical procedures carried a nearly two-fold higher complication risk compared to elective procedures.Future studies should prospectively evaluate preoperative transfusion approaches and compare immediate and delayed operative management to nonoperative management in this population. Level of evidenceIII Retrospective review.

Prescriptive analytics for reducing 30-day hospital readmissions after general surgery.

New financial incentives, such as reduced Medicare reimbursements, have led hospitals to closely monitor their readmission rates and initiate efforts aimed at reducing them. In this context, many surgical departments participate in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP), which collects detailed demographic, laboratory, clinical, procedure and perioperative occurrence data. The availability of such data enables the development of data science methods which predict readmissions and, as done in this paper, offer specific recommendations aimed at preventing readmissions. This study leverages NSQIP data for 722,101 surgeries to develop predictive and prescriptive models, predicting readmissions and offering real-time, personalized treatment recommendations for surgical patients during their hospital stay, aimed at reducing the risk of a 30-day readmission. We applied a variety of classification methods to predict 30-day readmissions and developed two prescriptive methods to recommend pre-operative blood transfusions to increase the patient's hematocrit with the objective of preventing readmissions. The effect of these interventions was evaluated using several predictive models. Predictions of 30-day readmissions based on the entire collection of NSQIP variables achieve an out-of-sample accuracy of 87% (Area Under the Curve-AUC). Predictions based only on pre-operative variables have an accuracy of 74% AUC, out-of-sample. Personalized interventions, in the form of pre-operative blood transfusions identified by the prescriptive methods, reduce readmissions by 12%, on average, for patients considered as candidates for pre-operative transfusion (pre-operative hematoctic <30). The prediction accuracy of the proposed models exceeds results in the literature. This study is among the first to develop a methodology for making specific, data-driven, personalized treatment recommendations to reduce the 30-day readmission rate. The reported predicted reduction in readmissions can lead to more than $20 million in savings in the U.S. annually.

P116. Comparing short term complications of inpatient versus outpatient elective vertebral augmentation for osteoporotic vertebral compression fractures

BACKGROUND CONTEXT For elderly patients with unremitting back pain secondary to osteoporotic vertebral compression fractures (VCF), vertebral augmentation remains the most utilized surgical intervention. Previous studies have examined perioperative complications, with 30-day readmission and mortality rates of up to 10% and 2%, respectively. Most such studies have included patients with pathologic fractures and those that had been inpatients prior to their procedures. Such shortcomings make it difficult for surgeons in office settings to accurately counsel patients regarding their risks for electively scheduled augmentation. PURPOSE To determine the 30-day postoperative complication rates for inpatient versus outpatient elective vertebral augmentation procedure performed for osteoporotic VCFs. STUDY DESIGN/SETTING Retrospective analysis of prospectively collected database. PATIENT SAMPLE Patients that underwent kyphoplasty or vertebroplasty were identified using CPT codes. To identify elective cases, only patients designated with “elective” status and those with ASA score 3 or under were selected. Patients that were inpatient status prior to the procedure, patients that were transferred from outside facility, patients that required preoperative blood transfusion, and those who were dependent on ventilator were excluded. To select only the osteoporotic VCFs, patients with disseminated cancer and those that underwent chemotherapy or radiotherapy were excluded. OUTCOME MEASURES Outcomes investigated were mortality, sepsis, cardiovascular complications, re-intubation, pneumonia, renal complications, stroke, wound complications, UTI, DVT, PE and transfusion. Also assessed were unplanned reoperation, unplanned readmission and discharge destination. METHODS The NSQIP database 2011-2014 was queried. Rates of 30-day postoperative complications, reoperations and readmissions were compared between the inpatient and the outpatient groups using univariate analyses. Multivariable analysis was proposed, however, was not performed since univariate analyses did not reveal meaningful differences. RESULTS A total of 1,539 patients were identified. Of those, 555 patients underwent inpatient surgery while 984 patients underwent outpatient surgery. Mean ages for inpatient and outpatient groups were 76.0 and 73.7, respectively. The 30-day postoperative complication rates were low for both groups. The mortality rates for the inpatient and outpatient groups were 0.19% and 0.31%, respectively, with no significant difference. UTI and sepsis were the only complications significantly different between the two groups. UTI occurred in 2.28% in the inpatient group and 0.94% in the outpatient group. Sepsis occurred in 0.95% of patients that underwent inpatient surgery, while no patient that underwent outpatient procedure sustained sepsis. Readmission (7.98% vs 5.74%) and reoperation (3.23% vs 2.92%) were both higher in inpatient group, however, the differences were not significant. CONCLUSIONS The only statistically significant differences in short-term morbidity from elective vertebral augmentation performed for osteoporotic VCFs were sepsis and UTI, with higher rates in inpatient group. The analysis suggests that outpatient vertebral augmentation may be safely performed in appropriate patient cohort. Compared with published studies that included pathologic fractures and patients that had been previously admitted, our study reported substantially lower 30-day mortality and readmission rates for electively scheduled augmentation of osteoporotic VCFs. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.

Analysis of Myomectomy during Cesarean Section: A Tertiary Center Experience

Purpose This study analyzed the safety of myomectomy during the cesarean section (CS). Methods Pregnant women who underwent myomectomy during CS in a tertiary center between January 2015 and November 2019 were included in the study in Group A, and pregnant women who did not have myoma and who underwent only CS were included in the study in Group B. The following information was obtained from patient files in hospital archives and was then recorded and compared: age, gravidity, parity, gestational week, characteristics of the myomas (i.e., location, size, number, and type), duration of surgery, perioperative complications, need for blood transfusion, preoperative and postoperative hemoglobin (Hb) values, duration of surgery, and hospital stay duration. Results A total of 83 patients underwent CS plus myomectomy (Group A), and 80 patients (without myoma) underwent only CS (Group B).There were no statistically significant differences between the groups in terms of preoperative and postoperative Hb values or blood transfusion rates (p > 0.05). Hospitalization and surgery duration were significantly higher in the group that underwent CS myomectomy (p = 0.001 and p = 0.001, respectively). The mean myoma size was 8.3 ± 4.1 cm in Group A. There was a statistically significant and inverse correlation between the size of the myoma and the delivery week (p = 0.035). There was a statistically significant and positive correlation between the myoma size and hospital stay (p = 0.01). Conclusion Myomectomy during CS is safe and can be applied regardless of the location, size, type, and number of myomas. However, to make myomectomy routine during CS, multi-center studies that include more cases are needed.

How do I…Incorporate a two-sample blood type verification in a pediatric hospital?

The AABB (American Association of Blood Banks) and the College of American Pathologists (CAP) regulations call on blood banks to address the risk of misidentification of a patient's blood type, which can result in transfusion of a mismatched product. Transfusion of mismatched blood product is potentially fatal due to acute hemolytic transfusion reaction and is considered a preventable event. CAP regulations outline options to reduce risk of mistransfusion by either documenting the ABO group of the intended recipient on a second sample collected at a separate phlebotomy, or utilizing a mechanical barrier system or electronic identification verification system that ensures the patient from whom the pretransfusion specimen was collected is the same patient who is about to be transfused. An electronic or barrier system was not available for implementation at our institution, therefore we developed a protocol for a two-sample verification system. The first determination is performed on a current sample and the second by one of the following methods: (a) comparison with previous laboratory records, (b) testing a second sample collected at a time different from the first sample (i.e., laboratory specimen available with a different timestamp, or a new blood sample). We improved our transfusion process and implemented a policy to require a second sample to confirm a patient's blood type. We also implemented workflows to obtain blood type confirmation from history of a second blood type result from previous laboratory records, including a policy to accept previous blood type records from an outside laboratory. We describe a practice change for two-sample verification for type and screen in a large-scale pediatric hospital. We outline specific workflows for pre-operative and emergency transfusion scenarios, and pediatric-specific challenges.

Preoperative Risk Factors for Postoperative Urinary Tract Infection After Primary Total Hip and Knee Arthroplasties

BackgroundEstablishing clear risk factors for complications such as urinary tract infection (UTI) after arthroplasty procedures helps guide clinical practice and provides more information to both surgeons and patients. This study aims to assess selected preoperative patient characteristics as risk factors for postoperative UTI after primary total hip and knee arthroplasties (THA and TKA). MethodsThis was a retrospective analysis using current procedural terminology codes to investigate the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database for patients who underwent THA or TKA from 2010 to 2017. Patients were classified for UTI by NSQIP guidelines. Patient samples with all possible covariates were included for multivariate logistic regression analysis and assessed for independent associations. ResultsIn a cohort of 983 identified patients (983 of 119,096; 0.83%): ages 57+ years, preoperative red blood cell (RBC) transfusion, perioperative RBC transfusion, bleeding disorders, operative time 110+ minutes, preoperative steroid use, diabetes, pulmonary comorbidities, body mass index 30+ kg/m2 were independent risk factors for postoperative UTI after THA. In a cohort of 1503 identified patients (1503 of 189,327; 0.8%): ages 60+ years, preoperative RBC transfusion, perioperative RBC transfusion, anemia, platelets less than 150k, preoperative steroid use, diabetes, and body mass index 30+ kg/m2 were independent risk factors for postoperative UTI after TKA. Male sex was associated with a decreased risk of UTI in both THA and TKA. ConclusionThis study provides novel evidence on risk factors associated with the development of UTI after THA or TKA. Clinicians should be aware of risk factors in the manifestation of postoperative UTI after primary THA or TKA procedures.

Effects of intravenous iron monotherapy for patients with iron deficient anemia undergoing total knee arthroplasty

BackgroundUnnecessary costs and complications can be reduced by increasing hemoglobin (Hb) levels and minimizing blood transfusions in patients who underwent total knee arthroplasty (TKA). This study aimed to determine the effects of intravenous iron monotherapy before TKA on preoperative iron deficient anemia and postoperative transfusion rates.MethodsThis prospective cohort study included 45 patients scheduled for TKA in the experimental group (Group I) and 221 patients who underwent TKA in 2015 and 2018 in the control group (Group C). One thousand milligrams of ferric carboxymaltose was administrated 1 month before TKA in group I. Intergroup comparison of the rate and total volume of transfusion, perioperative changes in Hb and analysis of iron metabolism variables in group I were performed. Subgroup analysis of Group I was conducted according to the response to iron monotherapy.ResultsAlthough Hb levels increased after intravenous iron monotherapy in Group I, postoperative transfusion rates in Groups I and C were 17.8% and 18.6%, respectively, without significant intergroup differences in the rate and total units of transfusion. Ferritin level and transferrin saturation were corrected in both subgroups of Group I. Only 17 patients (37.78%) showed response to iron monotherapy, with an Hb increase of 1.0 g/dL or more. Subgroup analysis showed lower proportions of coexisting chronic diseases, including chronic kidney disease in responders.ConclusionIV iron monotherapy was shown to be insufficient in successfully treating preoperative iron-deficient anemia to reduce postoperative allogenic blood transfusion in patients who underwent TKA. As preoperative anemia should be managed due to the high rates of postoperative transfusion for this surgery, clinicians should consider the complex interplay among the causal factors of anemia, apart from ID, in patients with preoperative anemia who are scheduled for TKA.

Right Kidney Mini-Invasive Living Donor Nephrectomy: A Safe and Efficient Alternative

Introduction: Right kidney living donor transplantation is considered more difficult and associated with more complications. The objective was to evaluate donor safety and graft function of right hand-assisted laparoscopic donor nephrectomy (HALDN). Methods: A total of 270 consecutive HALDN procedures have been performed in our institution up to April 2017. We retrospectively compared the outcomes of right-sided nephrectomy (R-HALDN) to left-sided nephrectomy (L-HALDN) to evaluate donor safety and graft function of R-HALDN. Results: Sixty-seven right kidneys were removed for functional asymmetry in favour of left kidney (35/67) or left kidney multiple arteries (28/67). Among the donors, neither conversion to open surgery nor preoperative blood transfusion was necessary. There was no significant difference in operative time, compared to L-HALDN group (170 ± 37 min vs. 171 ± 32 min; p value = 0.182). Warm ischaemia time was significantly longer for R-HALDN (4.0 ± 1.6 min vs. 3.0 ± 1.7 min; p < 0.001). There was no significant difference in terms of post-operative complications and serum Cr levels. Among the recipients, there were no graft venous thrombosis. There was no significant difference in delayed graft function (3 for R-HALDN group and 8 for L-HALDN group; p value = 0.847), serum Cr levels, and graft survival. Conclusion: R-HALDN is a safe procedure for kidney donors, with excellent graft function for the recipients, compared to L-HALDN.

A pre-operative platelet transfusion algorithm for patients with cirrhosis and hepatocellular carcinoma undergoing laparoscopic microwave ablation.

Thrombocytopenia is a common finding in patients with chronic liver disease. It is associated with poor clinical outcomes due to increased risk of bleeding after even minor procedures. We sought to determine an algorithm for pre-operative platelet transfusion in patients with cirrhosis and hepatocellular carcinoma (HCC) undergoing laparoscopic microwave ablation (MIS-MWA). A retrospective review identified all patients with cirrhosis and HCC who underwent MIS-MWA at a single tertiary institution between 2007 and 2019. Demographics, pre-operative and post-operative laboratory values, transfusion requirements, and bleeding events were collected. The analyzed outcome of bleeding risk included any transfusion received intra-operatively or a transfusion or surgical intervention post-operatively. Logistic regression models were created to predict bleeding risk and identify patients who would benefit from pre-operative transfusion. There were 433 patients with cirrhosis and HCC who underwent MIS-MWA identified; of these, 353 patients had complete laboratory values and were included. Bleeding risk was evaluated through bivariate analysis of statistically and clinically significant variables. The accuracy of both models was substantiated through bootstrap validation for 500 iterations (model 1: ROC 0.8684, Brier score 0.0238; model 2: ROC 0.8363, Brier score 0.0252). The first model captured patients with both thrombocytopenia and anemia: platelet count < 60 × 109 / L (OR 7.75, p 0.012, CI 1.58-38.06) and hemoglobin < 10 gm/dL (OR 5.76, p 0.032, CI 1.16-28.63). The second model captured patients with thrombocytopenia without anemia: platelet count < 30 × 109/L (OR 8.41, p 0.05, CI 0.96-73.50) and hemoglobin > 10 gm/dL (OR 0.16, p 0.026, CI 0.031-0.80). The prediction of patients with cirrhosis and HCC requiring pre-operative platelet transfusions may help to avoid bleeding complications after invasive procedures. This study needs to be prospectively validated and ultimately may be beneficial in assessment of novel therapies for platelet-based clinical treatment in liver disease.

Preoperative blood transfusions for sickle cell disease.

Preoperative blood transfusions for sickle cell disease.

Risk Factors for Mortality in Acute Aortic Dissection Type A: A Centre Experience Over 15 Years.

Acute aortic dissection Type A (AADA) is still associated with a high mortality rate and frequent postoperative complications. This study was designed to evaluate the risk factors for mortality in AADA patients. This retrospective analysis included 344 consecutive patients who underwent surgery for AADA in moderate hypothermic circulatory arrest (20-24°C nasopharyngeal) between 2001 and 2016. The 30-day mortality rate was 18%. Nonsurvivors were significantly older (65.7 ± 12.0 years vs. 62.0 ± 12.5 years; p = 0.034) with significantly higher Euro-score II [15.4% (6.6; 23.0) vs. 4.63% (2.78; 9.88); p < 0.001)]. Intraoperatively, survivors had statistically shorter cardiopulmonary bypass times [163 (134; 206) vs. 198 min (150; 245); p = 0.001]. However, the hypothermic circulatory arrest time was similar between both groups. Postoperatively, the incidence of acute kidney injury (AKI) (55.9 vs. 15.2%; p < 0.001), stroke (27.9 vs. 12.1%; p = 0.002) and sepsis (18.0 vs. 2.1%; p < 0.001) were significantly higher among nonsurvivors. The multi-variable logistic regression confirmed that older age, previous cardiac surgery, preoperative cardiopulmonary resuscitation (CPR), blood transfusion and postoperative acute kidney injury (AKI) were independent risk factors for mortality. Our analysis suggested that the reason for mortality was multifactorial, especially age, previous cardiac surgery, CPR, transfusion, as well as postoperative AKI were considered risk factors for mortality.

Myomectomy associated blood transfusion risk and morbidity after surgery

To evaluate blood transfusion risks and the associated 30-day postoperative morbidity after myomectomy. Retrospective cohort study. Not applicable. Women who underwent myomectomies for symptomatic uterine fibroids (N = 3,407). Blood transfusion during or within 72 hours after myomectomy. The primary outcomes were rate of blood transfusion with myomectomy and risk factors associated with receiving a transfusion. The secondary outcome was 30-day morbidity after myomectomy. The overall rate of blood transfusion was 10% (hysteroscopy, 6.7%; laparoscopy, 2.7%; open/abdominal procedures, 16.4%). Independent risk factors for transfusion included as follows: black race (adjusted odds ratio [aOR] 2.27, 95% confidence interval [CI] 1.62-3.17) and other race (aOR 1.77, 95% CI 1.20-2.63) compared with white race; preoperative hematocrit <30% compared to ≥30% (aOR 6.41, 95% CI 4.45-9.23); preoperative blood transfusion (aOR 2.81, 95% CI 1.46-5.40); high fibroid burden (aOR 1.91, 95% CI 1.45-2.51); prolonged surgical time (fourth quartile vs. first quartile aOR 11.55, 95% CI 7.05-18.93); and open/abdominal approach (open/abdominal vs. laparoscopic aOR 9.06, 95% CI 6.10-13.47). Even after adjusting for confounders, women who required blood transfusions had an approximately threefold increased risk for experiencing a major postoperative complication (aOR 2.69, 95% CI 1.58-4.57). Analysis of a large multicenter database suggests that the overall risk of blood transfusion with myomectomy is 10% and is associated with an increased 30-day postoperative morbidity. Preoperative screening of women at high risk for transfusion is prudent as perioperative transfusion itself leads to increased major postoperative complications.

Bacteriological Profile of Patients with Intra-Abdominal Sepsis and Superficial Surgical Site Infection Following Emergency Abdominal Surgery—Is It Concordant?

The rates of surgical site infections (SSI) continue to be reported with great variability, even for the same operation, from different geographical areas. Hence, a hospital-based surveillance of SSI can help in designing effective preventive strategies. This study was done to assess the risk factors, characteristics, and incidence of SSI in patients undergoing emergency gastrointestinal surgery. The study was a case series analysis, enrolling all consecutive patients undergoing emergency gastrointestinal surgery. The patients were examined daily for SSI using ASEPSIS score and during the post-operative period up to 30 days. For patients with SSI, wound swab was taken for culture and sensitivity. Patient and operative characteristics were analyzed for identifying risk factors, and the bacteriological profile and sensitivity pattern of intra-abdominal specimens were analyzed and compared with microbiological profile of SSI. A total of 100 patients were studied. The incidence of SSI was 33%. Age (44.33 ± 18.277 vs. 39.43 ± 16.158; p 0.015) and pre-operative blood transfusion (61.1 vs. 38.9; p 0.012) were found to be significant risk factors for SSI. On multivariate analysis, blood transfusion (p 0.027) and the duration of operation (p 0.005) were found to be independent risk factors. In cases where the isolated organisms were Escherichia coli, Enterococcus faecalis, and Klebsiella pneumoniae, concordance was noted between the intra-abdominal pathogens and the organisms isolated from the SSI. Risk factors for SSI in patients undergoing emergency abdominal surgeries include increasing age, pre-operative blood transfusion, prolonged operating time, intra-operative blood loss, and operative procedures involving bowel resection. Concordance exists between intra-abdominal and SSI pathogens.

Outcomes of a preoperative risk-based transfusion assignment protocol in sickle cell disease patients: a single-center retrospective study from Saudi Arabia

Many patients with sickle cell disease (SCD) need surgical management during their lifetime. The best approach for preoperative transfusion in SCD is still to be determined. In this single-center retrospective study, we included HBSS/HBS-Beta0-thalassemia patients younger than 16 years of age who underwent surgery between January 2008 and July 2019. Preoperative transfusion assignment (PTA) was based on SCD severity and surgical risk. Patients were assigned to no transfusion, simple transfusion, or exchange transfusion. A total of 284 patients were identified and 66 (23%) underwent 78 procedures. Mean age at the time of procedure was 8 (5–11) years, mean baseline hemoglobin was 8.5 (7.8–9.3) g/dl, and mean hemoglobin F was 18.4 ± 8.2%. SCD severity was low-risk in 57 (73%) and high-risk in 21 (27%) patients. Surgical risk was low-risk in 20 (25.6%) and medium-risk in 58 (74.4%) procedures. PTA was no transfusion in 17 (22%), simple transfusion in 40 (51%), and exchange transfusion in 21 (27%) procedures. Postoperative complications occurred in five (6.4%) of procedures only in the simple transfusion group (three acute chest syndrome, one hemolytic anemia, one pain crisis) undergoing medium-risk surgery. Preoperative risk-based transfusion assignment is feasible. Despite a high baseline hemoglobin level in the no transfusion group, none of the patients developed postoperative complications. It is possible that the high baseline hemoglobin F phenotype was protective and indicates the need to study the risk/benefit of interventions used in this phenotype.

Preoperative Transfusion for Anemia in Patients Undergoing Abdominal Surgery for Malignancy

BackgroundTransfusion guidelines have been established for severe anemia, but limited data is available regarding the utility of preoperative transfusion. This study evaluates the predictive factors and relative value of preoperative transfusion in oncologic patients with moderate anemia undergoing abdominal surgery. MethodsUsing the American College of Surgeons National Surgical Quality Improvement Program database, adult patients with moderate anemia (hematocrit 21–27%) who underwent non-emergent abdominal oncologic resection from 2005 to 2017 were identified. Preoperative transfusion and non-transfused patients were propensity score matched based on baseline covariates. Outcomes were compared using univariate and Poisson regression analysis. ResultsOf 6222 patients, preoperative transfused (N = 1000, 16.1%) patients were more likely to have bleeding disorders (12.1% vs 6.7%, p < 0.0001) and baseline thrombocytopenia (12% vs 7.3%, p < 0.0001) and had shorter operative length (< 180 min: 69.4% vs 59.8%, p < 0.0001). After matching (N = 987/group), preoperative transfusion was associated with higher rates of intraoperative/postoperative transfusion (odds ratio 1.24, p 0.017) and surgical site infections (odds ratio 1.67, p 0.004) and longer length of stay (incidence rate ratio 1.06, p < 0.0001). ConclusionsPreoperative transfusion is associated with increased surgical site infections and longer hospital stay and should be carefully considered in oncologic patients given the absence of improvement in outcomes.

Significant morbidity is associated with proximal fecal diversion among high-risk patients who undergo colectomy: A NSQIP analysis

BackgroundThe value of proximal fecal diversion for patients undergoing colectomies is an ongoing debate. Previous studies have shown a benefit in decreased anastomotic leak rates and mitigation of the morbidity of a leak, especially in high-risk populations. However, more recent data suggests increased morbidity with fecal diversion, creating a complication with an unknown degree of anastomotic leak reduction. Therefore, we aimed to determine the impact on morbidity of a diverting loop ileostomy (DLI) in patients with a high risk of anastomotic leak. MethodsThe ACS-NSQIP database was queried (via CPT code) for adult patients (age ≥18 years) who underwent a colectomy only or colectomy with ileostomy (CWI) between Jan 2013 and Dec 2016. We compared thirty-day outcomes between a 3:1 propensity-matched colectomy only group to patients who had a CWI. We used risk factors for anastomotic leak as a basis of our propensity match which included preoperative smoking, steroid use, preoperative weight loss, preoperative transfusion, hypoalbuminemia, and leukocytosis; intraoperative match variables included indication for surgery, wound class, duration of operation, primary CPT code, elective vs. emergent, and inpatient vs. outpatient surgery. ResultsWe identified 39,588 patients from the NSQIP database who had a colectomy only or a CWI. The colectomy only group was older (age 63 vs 52 years p < 0.001), overweight (BMI 34 vs 26.7, p < 0.001), more likely to be diabetic (16% vs 9.5%, p < 0.001) and hypertensive (49.3% vs 31.4%). However, the CWI group had higher steroid use (36.8% vs 10%, p < 0.001), preoperative sepsis (13.2% vs 2.5%, p < 0.001), smoking rate (25.7% vs 15.4%, p < 0.001), and preoperative weight loss (12.5% vs 4.9%, p < 0.001). Our propensity analysis matched 2274 colectomy only patients and 758 CWI patients. Baseline demographics were similar between groups. While the mortality rate was similar between groups (1.5% vs 1.8%, p = 0.8), CWI patients had longer length of stay (median 8 vs 7 days, p < 0.001), higher renal injury rates (3.2% vs 0.9%, p < 0.001), higher readmission rates (18.8% vs 11%, p < 0.001) and higher overall NSQIP morbidity (44.5% vs 37.6%, p = 0.001). The anastomotic leak rate was 3.8% in the CWI group and 5.1% in the colectomy only group (p = 0.09). ConclusionsSignificant thirty-day morbidity exists with a diverting ileostomy among high-risk colectomy patients with minimal benefit in anastomotic leak rates.

Contemporary mortality after emergent open repair of complex abdominal aortic aneurysms

ObjectiveMortality after open repair for emergent complex abdominal aortic aneurysm (AAA) is poorly defined. This study evaluated the 30-day mortality of open complex AAA repair performed for rupture or other emergent indication using a national surgical registry. We subsequently identified factors associated with mortality. MethodsThe targeted vascular module from the American College of Surgeons National Surgical Quality Improvement Program was queried to identify patients undergoing open repair for juxtarenal and suprarenal AAAs or type IV thoracoabdominal aneurysms (TAAAs) for rupture or other emergent indication from 2011 to 2017. Univariate analyses were performed using the Fisher's exact test for categorical variables and the Wilcoxon rank sum test for continuous variables. Multivariable logistic regression was performed to identify factors independently associated with mortality. ResultsWe included 374 patients who underwent an emergent complex open AAA repair during the study period. There were 142 (38%) cases performed for rupture with hypotension, 141 (38%) for rupture without hypotension, 40 (11%) for symptomatic AAA, and 51 (14%) for another indication. The distribution by aneurysm type was 224 juxtarenal AAAs, 122 suprarenal AAAs, and 28 type IV TAAAs. Overall, there was a 30-day mortality of 32% (118 deaths). For those with juxtarenal AAA repair, 67 (30%) patients died within 30 days; there were 38 (31%) deaths within 30 days in those with suprarenal AAA, and 13 (46%) deaths within 30 days in those with type IV TAAA. On univariate analysis, preoperative variables associated with death were increasing age, use of a transperitoneal surgical approach, lower preoperative estimated glomerular filtration rate, low baseline albumin concentration (<3.5 g/dL), need for preoperative transfusion, low body mass index (<18.5 kg/m2), and hypotension at presentation. Intraoperative variables associated with mortality were supraceliac clamp location and concurrent renal revascularization. On multivariable analysis, factors independently associated with death included rupture with associated hypotension (reference: other emergent indication; adjusted odds ratio [AOR], 3.28; confidence interval [CI], 1.75-5.41; P < .001), age >60 years (reference: <60 years; AOR, 1.59; CI, 1.18-2.13; P = .002), longitudinal laparotomy incision (reference: retroperitoneal; AOR, 3.28; CI, 1.75-6.16; P < .001), and supraceliac cross-clamp (reference: clamp above one renal artery; AOR, 2.14; CI, 1.31-3.50; P = .003). ConclusionsNearly one-third of patients die within 30 days of emergent open complex AAA repair. Mortality is particularly high for patients with type IV TAAAs, approaching 50%. Predictors of 30-day mortality include rupture with associated hypotension, increasing age, supraceliac clamp location, and longitudinal transperitoneal repair approach. These results will help inform surgical decisions preoperatively and intraoperatively.

Risk Factors for Blood Transfusion in Revision Total Hip Arthroplasty for Mechanical Failure

AbstractRevision total hip arthroplasty (rTHA) is associated with significant intraoperative blood loss requiring transfusion. Given that blood transfusions increase the risk of numerous complications, efforts should be made to determine which patients are at high risk of transfusion and optimize them appropriately. This study aimed to identify independent risk factors associated with blood transfusion in patients undergoing rTHA for mechanical failure. A multi-institutional surgical registry was queried for all patients who underwent rTHA for mechanical failure from 2005 to 2016. Patients were divided into transfused and nontransfused cohorts and underwent univariate analysis for differences in demographic features, preoperative comorbidities, and operative variables. Multivariate regression models were generated to elucidate independent risk factors associated with perioperative blood transfusions. A total of 8,469 rTHA patients met inclusion and exclusion criteria, of which 1,863 (22.0%) received blood transfusions within 72 hours postoperatively. On multivariate logistic regression analysis, age ≥ 65 years (odds ratio [OR] 1.313; p &lt; 0.001), female gender (OR 1.465; p &lt; 0.001), insulin-dependent diabetes mellitus (OR 1.503; p = 0.003), chronic obstructive pulmonary disease (COPD; OR 1.307; p = 0.030), hypertension (OR 1.359; p &lt; 0.001), hematologic disorders (OR 1.565; p &lt; 0.001), preoperative blood transfusions (OR 3.851; p &lt; 0.001), systemic sepsis (OR 2.277; p = 0.003), functional dependence (OR 1.416; p = 0.003), open wounds/wound infections (OR 2.878; p &lt; 0.001), and operations lasting at least 175 minutes (OR 4.841; p &lt; 0.001) were all significant independent risk factors for blood transfusions. Obesity (body mass index [BMI] ≥ 30 kg/m2; OR 0.818; p = 0.001), smoking history (OR 0.800; p = 0.014), and isolated acetabular component revision (OR 0.730; p &lt; 0.001) were associated with a decreased risk of blood transfusion. Older age, female gender, lower BMI, absence of smoking history, functional dependence, insulin-dependent diabetes mellitus, hypertension, revision of the femoral component, preoperative transfusion, hematologic disorders, sepsis, open wounds/wound infections, COPD, and longer operative time were all independent risk factors for blood transfusions in rTHA performed for mechanical failure. By identifying these risk factors, clinicians can identify and optimize high-risk patients to minimize blood transfusion requirements, thereby reducing the risk of transfusion-associated complications in this patient population.