Penn Evidence Based Literature Review

Penn Evidence-Based Literature Review (PEBLR)

Summarized highlights from contemporary literature in surgical and allied disciplines for general surgery residents.

Critical Care

Liberal or Restrictive Postoperative Transfusion in Patients at High Cardiac Risk: The TOP Randomized Clinical Trial
Kougias P, Sharath SE, Zhan M, Carson JL, Norman LE, et al, JAMA, 2025 Dec 23;334(24):2197-2207
Contributor: Iulia Barbur

Brief synopsis

This is a parallel, single-blind, randomized clinical trial assessing whether patients at high cardiac risk who underwent a liberal transfusion approach experienced a lower risk of post-operative death or ischemic events compared to those who underwent a restrictive transfusion approach in patients who underwent major vascular or general surgery operation and developed postoperative anemia.  Patients in the liberal transfusion group experienced a similar rate of the primary outcome compared to those in the restrictive transfusion group (9.1% versus 10.1%). However, patients undergoing a liberal transfusion approach experienced a lower rate of non-ischemic adverse cardiac events compared to those in the restrictive group (5.9% versus 9.9%, RR 0.59 [0.36 – 0.98]). 

Summary: This trial randomized 1,428 adults deemed to be at high cardiac risk who underwent major general or vascular surgery between February 2018 to March 2023 across 16 different Veterans Affairs Medical Centers in the United States to at a restrictive (transfusion triggered at <7 g/dL) versus liberal (transfusion triggered at <10 g/dL) post-operative transfusion strategy. Patients were considered to be at elevated cardiac risk if they had a history of ischemic heart disease, myocardial infarction, peripheral arterial disease, stroke, or transient ischemic attack. The primary outcome was either all-cause death or occurrence of a major ischemic event (myocardial infarction, open or percutaneous coronary revascularization, acute kidney failure, or ischemic stroke) within 90 days of randomization. While the primary outcome rate was similar between the two groups (liberal: 9.1% versus restrictive: 10.1%), those undergoing a liberal transfusion strategy did experience a lower rate of new cardiac arrhythmias, heart failure, or non-fatal cardiac arrest (5.9% versus 9.9%, RR 0.59 [0.36 – 0.98]). The rate of protocol deviations was low in both groups, with 9% of patients assigned to the liberal strategy not being transfused per a threshold of 10 g/dL and 4% of patients in the restrictive group transfused when hemoglobin was higher than the threshold of 7g/dL. This led the authors to conclude that a liveral transfusion strategy did not reduce 90-day death or major ischemic outcomes rates compared with a restrictive transfusion strategy.

There are some important limitations of the trial that should be noted. First, this was conducted Veterans Affairs Medical Centers so the population was largely male meaning that some of these conclusions may not apply to females. Additionally, the event rate for the trial was lower than expected which reduced the statistical power of the study. Finally, the trial had a disproportionate ratio of patients who underwent vascular (~90%) versus general surgery (~10%), as this may itself impact the time course and physiology of post-operative anemia. 

Bottom line: Patients undergoing a restrictive versus liberal transfusion strategy following major vascular or general surgery did not experience different rates of death or ischemic events, however those undergoing a liberal transfusion strategy did experience a lower rate of new cardiac arrhythmias, heart failure, or non-fatal cardiac arrest.

Urology

Evaluating Benefits of Peristomal Mesh Placement at the Time of Radical Cystectomy and Ileal Conduit Formation: A Phase 3 Randomized Controlled Trial
Timothy F. Donahue, Melissa Assel, Eugene K. Cha, Alvin Goh, Eugene Pietzak, et al, J Urol 2025 Nov;214(5):487-495
Contributor: Shirley Wang

Brief synopsis

Parastomal hernias are common among patients undergoing radical cystectomy and ileal conduit diversion. This phase 3 randomized controlled trial investigated the impact of prophylactic mesh placement on radiographic parastomal hernia development within two years. Risk of parastomal hernia did not differ between the mesh vs. no mesh group (OR with mesh=1.78 [95% CI 0.89-3.55]) at two years, and there were no significant differences in complication rates between the two groups.

Summary: Parastomal hernias (PHs) occur in up to half of patients undergoing radical cystectomy (RC). Prophylactic mesh placement has been evaluated in colorectal surgery, with most studies showing resultant decrease in rate of PH. Prior studies evaluating prophylactic mesh placement in cystectomy patients have shown mixed results. The purpose of this study was to evaluate the impact of prophylactic mesh placement at the time of radical cystectomy and ileal conduit on risk of developing a radiographic PH. 

This was a randomized controlled study including patients with nonmetastatic urothelial carcinoma undergoing RC with formation of ileal conduit. Patients in the mesh cohort underwent placement of partially absorbable mesh in the sublay position at the time of RC. Patients were evaluated for PH clinically and radiographically for two years. Secondary endpoints included 90-day perioperative complications and stomal complications within the two-year follow-up period.

Eighty-eight patients were randomized to undergo prophylactic mesh placement and 90 randomized to undergo standard surgery. Ultimately, 59 were available for analysis from the mesh group and 58 in the control group. At two years, 32 (47%) patients in the mesh group and 23 (33%) in the control group had radiographic evidence of a PH, corresponding to OR of 1.78 (95% CI 0.89-3.55).  Effects of BMI (<30 vs. ≥30) and surgeon were adjusted for using Cochran-Mantel-Haenszel methods, which also did not yield a significant difference in risk of developing radiographic PH. There was no significant difference in Clavien-Dindo grade 2 to 5 complications (69% vs. 81%, p=0.11) or stomal complications (34% vs 33%, p>0.9) between the control and study groups. Furthermore, there were no reported mesh-related complications in the study group. 

This study is limited in its inclusion of all patients undergoing radical cystectomy, rather than patients at higher risk of developing a PH. While a prior exploratory study conducted by this group noted a 22% decrease in rate of PH, this was conducted in female patients with BMI >30. The present study did not include time to hernia development. Although most PHs develop within two years, longer-term follow-up with time-to-event analysis could clarify temporal trends in PH development. 

Bottom line: In this phase 3 randomized controlled trial, prophylactic mesh placement was not associated with reduced risk of developing parastomal hernia among patients undergoing cystectomy with ileal conduit formation. While this data does not support routine mesh placement, further work may evaluate benefit of prophylactic mesh placement specifically in higher-risk subgroups. 

Vascular Surgery

Medical Management and Revascularization for Asymptomatic Carotid Stenosis
Brott TG, Howard G, Lal BK, et al; CREST-2 Investigators, N Engl J Med 2026 Jan 15;394(3):219-231
Contributor: Dasha Savage

Brief synopsis

CREST-2 is a randomized clinical trial consisting of two parallel observer-blinded studies. It evaluates intensive medical management alone versus intensive medical management plus revascularization for patients with high-grade (≥70%) asymptomatic carotid stenosis. The stenting trial showed a lower 4-year incidence of the primary composite outcome (periprocedural stroke or death plus subsequent ipsilateral ischemic stroke) with stenting (2.8%) compared to medical management alone (6.0%, p=0.02), while the endarterectomy trial found no significant difference (3.7% vs. 5.3%, p=0.24). These findings highlight the potential benefit of stenting in select patients but underscore the foundational role of optimized medical therapy in reducing stroke risk.

Summary: Treatment of high-grade asymptomatic carotid stenosis varies internationally, with many European countries restricting revascularization primarily to symptomatic or high-risk cases, while in the U.S., 75–80% of procedures are for asymptomatic patients. The CREST-1 trial (2004–2010) previously demonstrated that carotid endarterectomy (CEA) was superior to medical therapy alone in symptomatic patients and comparable to medical therapy in asymptomatic patients. However, advancements in medical management, stenting, and endarterectomy necessitated re-evaluation. The CREST-2 trial addresses this by conducting two parallel, multicenter, randomized, observer-blinded trials: one comparing intensive medical management alone to transfemoral carotid artery stenting (TF-CAS) plus medical management, and the other comparing medical management alone to CEA plus medical management.

Patients aged ≥35 years with ≥70% asymptomatic stenosis (confirmed by Doppler ultrasonography [peak systolic velocity ≥230 cm/s], CTA/MRA, or angiography) and no ipsilateral symptoms (within 180 days) were enrolled at 155 centers across five countries from December 2014 to 2025. Exclusions included prior disabling stroke, unstable angina, or atrial fibrillation requiring anticoagulation. Randomization was stratified by center and trial, with intention-to-treat analysis. Intensive medical management targeted systolic blood pressure <130 mm Hg after 2018, LDL cholesterol <70 mg/dL, and addressed other risk factors through health coaching. Follow-up included up to 4 year post-procedural assessments with annual Doppler and MRI/CT for suspected events. The primary outcome was a composite of periprocedural (0–44 days) stroke/death plus postprocedural ipsilateral ischemic stroke. 

A total of 2485 patients were randomized (1245 in stenting trial and 1240 in endarterectomy trial). 17–18% crossover from medical management to revascularization occurred due to new symptoms, clinical progression, or patient preference. In the stenting trial, the 4-year primary event rate was 6.0% (95% CI, 3.8–8.3) with medical management versus 2.8% (95% CI, 1.5–4.3) with stenting (p=0.02), driven by lower postprocedural ipsilateral strokes (annual rate 0.4% vs. 1.7%) despite periprocedural events (1.3% vs. 0%). In the endarterectomy trial, rates were 5.3% (95% CI, 3.3–7.4) versus 3.7% (95% CI, 2.1–5.5; p=0.24), with periprocedural strokes higher in CEA (1.5% vs. 0.5%). 

The Society for Vascular Surgery commentary (AbuRhama Dec 2025) critiqued CREST-2, and emphasized real-world data showing superior outcomes for CEA/TCAR. CREST-2 did not include transcarotid artery revascularization (TCAR), and interventionalists that performed TF-CAS for the study were subject to strict selection. Interventionalists for TF-CAS submitted cases for review with a 50% acceptance rate to the study, compared to surgeons for carotid endarterectomy which had a 90% acceptance rate and were required to have a documented <3% periprocedural stroke/death in 50 cases. There were also concerns regarding non-replicable medical management, as all study participants received health coaching, and after 2018, Alicorumab (PCSK9 inhibitor) was available to all participants free of cost (donation from the company). The study was able to achieve sustained risk factor control with >80% of participants meeting LDL targets by 12 months. Other limitations include 17–18% crossovers diluting differences between the parallel studies, predominantly White/U.S.-based cohort limiting generalizability, median 4-year follow-up (longer-term unknown), COVID-19 impacts on visits, low event rates reducing power, and exclusion of emerging and validated technologies like TCAR.

Bottom Line: CREST-2 supports selective TF-CAS in asymptomatic high-grade carotid stenosis patients with suitable anatomy and experienced operators but shows no clear benefit for CEA over intensive medical management alone. An individualized approach—considering patient risk, anatomy, and all modalities (including TCAR)—is essential, with further research needed on diverse populations, longer-term outcomes, and plaque features to refine guidelines and minimize perioperative risks.

Penn Evidence-Based Literature Review (PEBLR)

Summarized highlights from contemporary literature in surgical and allied disciplines for general surgery residents.

Endocrine Surgery

Beyond the Neck: When Is Thoracic Operation Needed for Thyroidectomy?
Wang R, Song Z, Balachandra S, Gillis A, Fazendin J, Lindeman B, Wei B, and Chen H. J Am Coll Surg 2025 Apr 1;240(4):599-608
Contributor: Amanda Bader

Brief synopsis

This is a single-institution retrospective cohort study evaluating computed tomography (CT)-based predictors to determine the need for intraoperative thoracic surgery assistance during thyroidectomy. The authors found that there are several particular CT measurements that strongly predict the need for thoracic surgery involvement during a thyroidectomy. 

Summary: Intrathoracic (substernal) goiters are a heterogeneous subset of thyroid disease that can present significant technical challenges during thyroidectomy. While most can be removed through a cervical approach, a minority require thoracic surgical assistance, making reliable preoperative predictors essential. Existing criteria, such as posterior mediastinal extension or depth below the aortic arch, remain subjective and inconsistently defined.

This retrospective, single-institution study reviewed 7,370 thyroidectomies performed between 2012 and 2023 to identify patients with intrathoracic thyroid glands, defined on CT as extension below the sternal notch. Of 255 patients meeting this criterion, 34 (13.3%) required intraoperative thoracic surgery assistance. The authors calculated comprehensive and standardized CT measurements, including thyroid dimensions in three planes, mediastinal location, relationship to thoracic inlet dimensions, and craniocaudal extension relative to multiple anatomic landmarks. They then used to receive operating characteristics (ROC) and area under the curve (AUC) analyses to determine if the CT measurements were predictive of thoracic surgery assistance. Multivariable regression analysis was also used to evaluate the independent predicting factors for intraoperative thoracic involvement.

ROC analysis demonstrated that craniocaudal extension measured from the inferior thyroid border to the sternal notch was the most accurate predictor of thoracic assistance (AUC 0.94), outperforming measurements that used the aortic arch or sternal angle. Thresholds providing ≥80% sensitivity for predicting thoracic involvement included thyroid extension ≥5.3 cm below the sternal notch, horizontal dimension ≥5.3 cm, and anteroposterior dimension ≥5.1 cm. On multivariable regression, three independent predictors were identified: thyroid enlargement below the sternal notch (OR 10.1, 95% CI 1.96 to 51.63, p = 0.006), craniocaudal extension ≥5.3 cm (OR 5.3, 95% CI 1.78 to 15.85, p = 0.003), and horizontal dimension ≥5.3 cm (OR 3.8, 95% CI 1.13 to 13.00, p = 0.032). Additionally, patients requiring thoracic assistance had longer operative times and hospital stays but did not experience significantly higher rates of hypocalcemia, recurrent laryngeal nerve injury, or mortality.

Limitations of this study include the single-center setting, variability in surgeon experience and thoracic approach selection over time, and small sample size and therefore, limited power to predict the specific thoracic procedure required. These factors may limit generalizability to other practice environments.

Bottom line: Objective CT measurements may be able to reliably identify patients with intrathoracic goiter who should undergo preoperative thoracic surgery consultation.

Plastic & Reconstructive Surgery

Single-Dose versus 24-Hour Antibiotic Prophylaxis in Reduction Mammaplasty: A Randomized Controlled Trial
Viega F, Garcia ES, Viega-Fialho J, Fialho SV, Fortunato Borges AS, et al. Plast Reconstr Surg. 2025 Dec 1;156(6):835-842
Contributor: Carrie Z. Morales

Brief synopsis

This triple-blind, randomized, noninferiority trial compared single-dose versus 24-hour antibiotic prophylaxis in 146 patients undergoing reduction mammaplasty. The study found no significant difference in surgical-site infection (SSI) rates or wound complications between groups.

Summary: The optimal duration of antibiotic prophylaxis in reduction mammaplasty remains controversial, with many surgeons extending postoperative antibiotic use beyond current guideline recommendations. This randomized controlled trial was conducted at a Brazilian university hospital to compare the efficacy of a single preoperative dose of cefazolin versus 24 hours of postoperative prophylaxis in preventing SSIs following reduction mammaplasty. This study follows a previous randomized controlled trial from the group demonstrating no significant difference in SSI rates between patients given 24 hours of perioperative antibiotics or 7 days of post-operative antibiotics.

A total of 146 women (aged 18-60 years) with symptomatic breast hypertrophy were randomized 1:1 using sequentially numbered, opaque envelopes to receive either a single dose of cefazolin 1 g at anesthesia induction followed by placebo (sodium chloride solution every 6 hours for 24 hours; n=73), or the same initial dose followed by cefazolin 1 g every 6 hours for 24 hours postoperatively (n=73). All medications were prepared in identical vials with obscured labels to maintain triple-blinding. Patients were admitted the evening before surgery, underwent standardized preoperative chlorhexidine showering, and received reduction mammaplasty via superomedial pedicle technique with inverted-T incision by plastic surgery residents under attending supervision. Patients were discharged after 24 hours and followed weekly for 30 days by a surgeon blinded to group allocation.

The primary outcome was SSI occurrence according to CDC criteria, with secondary outcomes including other postoperative complications. The two groups were well-matched at baseline, with median age of 33 years, BMI 25.2 kg/m2, excised tissue weight 925g, and surgery duration 220 minutes. The overall SSI rate was 5.5% with 4 infections in each group (P=1.000). All SSIs were superficial incisional infections. Superficial dehiscence occurred in 30 patients (20.5%): 16 in the placebo group and 14 in the antibiotic group (P=0.682). Other complications including epidermolysis, partial nipple-areola complex necrosis, and seroma showed no significant between group differences. Logistic regression analysis identified increasing age as protective against wound dehiscence (OR 0.938, 95% CI 0.897-0.982, P=0.006), with approximately 5% decreased odds per additional year of age. 

Strengths of this study include the rigorous triple-blind design and standardized protocols. The main limitation is generalizability—strict exclusion criteria (BMI >30 kg/m2, smokers, significant comorbidities) and the 30-day follow-up period may not reflect real-world reduction mammaplasty populations, particularly given that most patients undergoing this procedure have higher BMIs.

Bottom Line: Extending antibiotic prophylaxis beyond a single preoperative dose does not reduce SSI or wound dehiscence rates in reduction mammaplasty in low comorbidity patients, supporting single-dose prophylaxis as a safe antibiotic stewardship practice.

Surgical Education

Practice Readiness of Chief Residents in a National Sample During the First Year of EPA Assessments
Lindeman B, Scott-Smith C, Jones A, Sarosi GA, Minter RM, Jung S, et al. Ann Surg. 2025 Oct 1;282(4):601-607
Contributor: Sarah Landau

Brief synopsis

Synopsis: This is a retrospective analysis of national EPA microassessment data collected for general surgery residents during their final year of training (2023-2024). Authors found that entrustment increased over the year and by the end of training nearly ¾ of chief residents achieved practice-ready entrustment. Operative skills were entrusted at a slower rate than nonoperative skills, particularly for renal replacement therapy, thyroid and parathyroid disease, and benign and malignant colon conditions. As more EPA microassessment data becomes available, future studies will be needed to monitor the distribution of EPA ratings and define resident learning curves over the duration of training to facilitate early identification of trainees who may not reach practice readiness by graduation.  

Summary: Entrustable Professional Activities (EPAs) are workplace-based assessments designed to provide frequent, formative feedback about trainee performance and represent a framework for competency-based assessment. The American Board of Surgery (ABS) developed 18 EPAs for general surgery which were rolled out to residency programs in July 2023 and will be part of the admissibility requirements for ABS general surgery certification beginning in 2028. This study sought to examine EPA data of chief residents across US general surgery programs to better understand the performance of this population and identify potential gaps in preparedness for independent practice. 

Authors retrospectively analyzed de-identified chief resident EPA data that had been prospectively collected from July 1, 2023 – May 31, 2024 across 3 platforms (SIMPL, Firefly, SEPA). For each microassessment, residents were scored 1-4 based on entrustment level (1=limited participation, 2=direct supervision, 3=indirect supervision, 4=practice-ready). Not all residents were evaluated on every EPA. Data were analyzed using descriptive statistics and bivariate analyses.  

A total of 12,611 EPA microassessments were collected, corresponding to 1032 chief residents (67%) from 265 programs (75.7%). The mean number of EPA assessments per resident was 12.2 +/- 19.5 with a mode of 1. The overall proportion of chief residents rated as practice-ready increased significantly over the year from 43.5% to 74.5% (p<0.001). Residents were more likely to receive practice-ready entrustment in nonoperative (nonoperative/preoperative and postoperative) compared with intraoperative phases of care and this gap narrowed but remained significant over the year (nonoperative practice-ready: 59% → 84.9%, intraoperative practice-ready: 36.8% → 70.8%; p<0.001). The proportion of residents rated as practice-ready varied inversely with case complexity but increased for all levels of complexity over the year (straightforward: 54% → 81%, moderate: 34% → 71%, complex: 42% → 1%; p<0.001). At the end of the year, the median entrustment was practice-ready for the nonoperative phases of all EPAs and for the intraoperative phases of all EPAs except for renal replacement therapy, thyroid and parathyroid disease, and management of benign and malignant colon disease. By the end of chief year, a total of 85 EPA microassessments from 51 unique residents were rated as direct supervision or limited participation. 

Limitations of this study include selection bias as the class of 2024 was not required to participate in EPA assessments therefore included residents may not be representative of all chief residents. There also may be selection bias on the individual resident level; EPAs are designed to be used as frequent microassessments but most chief residents in this study only had 1 assessment which may have not been representative of the resident’s true or overall performance. Additionally, because programs progressively began using EPAs and not all included trainees had July 2023 assessment, it is possible that the highest performing trainees are disproportionately represented, resulting in an overestimation of practice-ready residents at the start of their chief year. 

Bottom Line: In this nationwide study of general surgery residents who received EPA microassessments during their chief year, nearly ¾ achieved practice-ready entrustment by the end of training with slower entrustment of operative skills compared with nonoperative skills. These findings illustrate how EPA microassessment data can provide insight into longitudinal trainee performance. Future studies are needed to define individual, program, and national level learning trajectories to facilitate early detection of trainees who may not achieve practice readiness by the end of training. 

Penn Evidence-Based Literature Review (PEBLR)

Summarized highlights from contemporary literature in surgical and allied disciplines for general surgery residents.

Health Policy

Hospitals Acquired By Private Equity Firms: Increased Postoperative Mortality For Common Inpatient Surgeries
Diaz A, Mead M, Rohde S, Kunnath N, Dimick JB, Ibrahim AM,  Health Aff (Millwood) 2025 May;44(5):554-562
Contributor: Russell Simons

Brief synopsis

This retrospective cohort study finds a significant difference in 30-day post-operative mortality – specifically because of failure to rescue (i.e., prevent death among patients with post-op complications) – for common general surgery procedures when comparing hospitals recently acquired by private equity (PE) firms versus non-PE hospitals. 

Summary: Private equity (PE) firms have played an increasing role in American health care over the past decade, with a notable focus on restructuring healthcare organizations to maximize profitability. PE firm acquisitions increasingly concern inpatient surgical care, and there has been limited analysis thus far on the effects of these acquisitions on care quality. In this study, the authors used the Medicare Provider Analysis and Review Files (2011-2020), linked to the American Hospital Association Annual Survey, to review all claims for colectomies, cholecystectomies, appendectomies, and incisional hernia repairs. Exposure was acquisition of a US non-PE hospital by a PE firm. The primary outcome was 30-day post-operative mortality. Secondary outcomes were any/serious complication, failure to rescue, and 30-day readmission.

The results showed that 22,446 surgical episodes (7.5% of total) occurred at PE-acquired hospitals (67/701 total). These hospitals were typically <200 beds and located in urban/suburban areas of the U.S. South and West. Adjusted difference-in-difference analyses showed that PE acquisition was associated with increased 30-day postoperative mortality (+2.7%) and an increased failure to rescue (+3.9%). No significant associations were seen for other secondary outcomes. On subset analysis, these differences were seen primarily in patients who underwent emergent surgery as opposed to elective surgery.

Limitations of the study include generalizability since the study population was Medicare patients. Additionally, there are some limitations in internal validity since there is selection bias from non-random acquisition of hospitals by PE firms as well as coding bias, thus potential confounding by factors not captured in administrative data.

Bottom Line: This study finds a significant increase (+2.7%) in 30-day post-operative mortality for patients undergoing emergent surgical procedures at hospitals acquired by PE firms during the study period, which appears attributable to failure to rescue. This increase in failure to rescue suggests issues in recognizing and managing complications and may be due to an observed relative reduction in registered nurse FTEs at PE hospitals.

Transplantation

Association of delayed graft function with cardiovascular outcomes in kidney transplant recipients
Beaudrey T, Aymes E, Thaunat O, Bedo D, Masset C, et al., Am J Transplant, 2025 Oct;25(10):2182-2193
Contributor: Nic Muñoz

Brief synopsis

This study from France of more than 18,000 kidney transplant recipients evaluates the impact of delayed graft function (DGF, the acute need for dialysis following transplant) on cardiovascular morbidity and mortality. They found that kidney transplant recipients who had DGF were at increased risk of major adverse cardiovascular events, and this association intensified with greater DGF severity. 

Summary: Kidney transplant provides improved overall survival for patients with end stage renal disease over dialysis; however, cardiovascular disease remains highly prevalent among kidney transplant recipients (KTRs). Delayed graft function (DGF) is characterized as acute kidney injury in KTRs and is defined as need for at least 1 dialysis session in the 7 days post-transplant. It has previously been associated with worse transplant outcomes such as graft survival, and all-cause mortality though the mechanisms are not fully understood. In this study, the authors explored the idea that acute kidney injury, such as in KTRs experiencing DGF, may increase cardiovascular risk. They did this using a retrospective multicenter cohort study of 17 transplant centers, and 18,149 KTRs from 2008-2022, facilitated by a national centralized network. The primary outcome was major adverse cardiovascular events (MACEs), including death, coronary artery disease, heart failure and stroke. Secondary outcomes included graft failure, non-cardiovascular death, and all-cause death. Kaplan-Meier analysis was used to estimate the cumulative incidence of MACEs, as well as with competing risk assessment using Gray method. Multivariable cause-specific Cox proportional hazards models were used to evaluate associations between DGF and each of the outcomes, adjusting for center effect and potential confounders. 

Among 18,149 KTRs, 23% experienced DGF, 12.8% experienced at least one MACE, 11.1% had graft failure, and 17.4% died during the follow up period. MACEs were characterized as CAD-related events (26.4%), heart failure episodes (53.6%), stroke (12.3%) and cardiovascular death (9.3%). KTRs with DGF had an increased incidence of MACEs, with rising incidence over time (10.4% at 1 year, 16.7% at 5 years), while those without DGF were significantly less likely to have MACEs (5.2% at 1 year, 9.7% at 5 years). DGF was associated with coronary artery disease, heart failure, and cardiovascular death but not with stroke. Additionally, there was a significant association between DGF and graft failure and non-cardiac death at 10 years. 

In summary, DGF was associated with a 24% increased hazard of MACE, with this being independent of the calculated glomerular filtration rate at 3 months post-transplant. The increased risk was higher in populations with typically low cardiovascular risk, such as women, patients without diabetes or without cardiovascular disease history, and was persistent across time (ie, beyond 1 year, out to 5+ years). The risk of MACE increased with higher severity of DGF (requiring more than 1 dialysis session). Study limitations include retrospective trial design and unaccounted for/residual confounders.

Bottom line: This study investigated the downstream risk to KTRs who experience DGF, showing durable increased risk of MACEs post-transplant, even when there is expedient graft recovery. KTRs remain a high-risk group for cardiovascular morbidity and mortality post-transplant, highlighting the importance preventative medical therapies to minimize risks.

Penn Evidence-Based Literature Review (PEBLR)

Summarized highlights from contemporary literature in surgical and allied disciplines for general surgery residents.

Colon and Rectal Surgery

Chemoradiotherapy and Local Excision vs. Total Mesorectal Excision in T2-T3ab, N0, M0 Rectal Cancer; The TAUTEM Randomized Clinical Trial
Serra-Aracil X, Pericay C, Cidoncha A, Badia-Closa J, Golda T, et al., TAUTEM Collaborative Group, JAMA Surgery May 2025
Contributor: Drew Goldberg

Brief synopsis

This study is a phase III non-inferior randomized clinical trial involving 17 hospitals within Spain, exploring whether neoadjuvant chemoradiotherapy and trans-anal endoscopic excision in T2 or T3 rectal cancers lead to similar clinical outcomes compared to the standard of care. The authors found, in the 5-year modified intention-to-treat analysis, there were no differences in local recurrence, and similar results for distant recurrence and overall survival rates.

Summary: International guidelines recommend total mesorectal excision (TME) as the standard treatment for T2 and early T3 (T3ab), N0, M0 rectal cancer. T3ab tumors are those that penetrate the muscularis propria but with invasion ≤ 5mm without nodal involvement or distal metastases. Local excision is generally not recommended, especially for T3 tumors, due to the risk of regional lymph node involvement. In this setting, TME is typically performed as the oncologic resection, but carries with it a risk of complications as high as 54%, in large part due to the likely need for diverting stoma creation and associated risks. An alternative strategy is neoadjuvant chemoradiotherapy with local excision, with single-center observational studies having published clinical data with inconclusive results. The purpose of this study was to compare standard treatment of stage T2-T3ab, N0, M0 by TME vs. long-course chemoradiotherapy followed by local excision via transanal endoscopic microsurgery (TEM).

The study design was an open-label, prospective, non-inferiority trial at 17 hospitals within Spain. Patients were included if they had a diagnosis of a rectal adenocarcinoma located 10 cm from the anal verge, clinical stage T2-T3ab, N0, M0 confirmed by endorectal ultrasound and rectal MRI, tumors with maximum diameter of 4 cm or less on MRI, and ASA stage 3 or less. Exclusion criteria included presence of distant metastases, synchronous colorectal cancers, undifferentiated rectal adenocarcinomas, or intolerance of chemotherapy or radiotherapy. Treatment in the TEM group received long-course neoadjuvant treatment with combined chemotherapy and radiotherapy (oral capecitabine and daily fractions of 1.8 Gy, 5 days a week). Clinical response was assessed using rectal MRI. Surgery was performed in the eighth week after CRT. Local excision was performed with TEM or transanal minimally invasive surgery (TAMIS) by a colorectal surgeon. The protocol was discontinued in patients who showed no clinical response after CRT and in those whose pathology after excision showed characteristics of poor prognosis. The primary outcome was local recurrence (LR), defined as the presence of adenocarcinoma in the biopsy of the residual scar, anastomosis, tumor resection bed, locoregional pelvic lymphadenopathy, or in the perineal skin after abdominoperineal resection. Secondary outcomes included distal recurrence (DR), overall survival (OS) and disease free survival (DFS).

From 2010 to 2021, 364 patients were evaluated for inclusion with 173 ultimately randomized into the study; 81 patients were included in each group for the modified intention to treat (mITT) analysis. The median follow-up time for mITT and per protocol (PP) analysis was 63 months. Post operative morbidity was 20.7% in the TEM group and 50.6% in the TME group. In the 5-year mITT analysis, LR was 7.4% in the CRT-TEM group and 6.2% in the TME group (difference -1.23%, p=0.76). In the TME group, 17 patients had nodal involvement and underwent adjuvant treatment; of these 41.2% developed distant recurrence versus 10.9% of those with node-negative status. In the mITT analysis, CRT-TEM had a DR of 12.3% versus 17.3% in the TME group (difference 4.94%, CI 15.85% to -5.98%). In the 5-year mITT analysis, the OS was 82.7% for the CRT-TEM group versus 85.2% in the TME group (p=0.62). The DFS was the same in the 2 groups at 88.9%.

Strengths of the study include a notable reduction in the incidence of occult nodal involvement of the TEM group with neoadjuvant treatment (approximately 15% reduction) and good long-term results obtained in the TME control group for the secondary outcomes. Limitations include the long duration of the study, in which changes in diagnostic and treatment modalities may have occurred. Lastly, the study was not formally powered to detect differences between groups on secondary outcomes.

Bottom Line: A Spanish multicenter non-inferiority randomized clinical trial comparing TME (standard treatment) with neoadjuvant chemoradiation with trans-anal endoscopic microsurgical (TEM) excision strategy for T2 or T3ab (confined to ≤ 5mm outside muscularis propria) rectal cancer treatment demonstrated non-inferior local recurrence rates with similar distant recurrence and overall survival. A less invasive approach sparing the rectum with chemoradiation followed by trans-anal endoscopic excision may be suitable for these tumors, allowing patients a less morbid alternative to TME. 

Retrospective Cohort / Surgical Oncology

Prognostic impact of positive peritoneal cytology (POPEC) in gastric cancer: Multi-centre European retrospective cohort study
Richard Owen, Swathikan Chidambaram, Khalid Shamiyah, Nick Maynard, Sheraz Markar et al., on behalf of the POPEC Study Group, Surgical Oncology Insight September 2025
Contributor: Lily Owei

Brief synopsis

This multi-center European and Asian retrospective cohort study evaluated the prognostic significance of positive peritoneal cytology (POPEC) in patients with gastric cancer who had no radiologic or macroscopic evidence of metastatic disease. The investigators found that while positive cytology correlated with higher tumor stage, it was not independently associated with poorer 5-year overall survival after adjustment for pathologic stage and receipt of neoadjuvant therapy. The study also demonstrated substantial variability in cytology sampling techniques across centers, which affected the detection rate of positive cytology. These findings suggest that positive peritoneal cytology alone should not preclude curative-intent surgery but underscore the need for standardized cytologic sampling protocols and prospective validation.

Summary: The POPEC study was a large, multi-institutional retrospective cohort analysis conducted across 16 European and Asian tertiary gastric cancer centers. The study population included 837 adult patients with histologically confirmed gastric adenocarcinoma who underwent staging laparoscopy with peritoneal lavage cytology prior to curative-intent gastrectomy between 2009 and 2023. Patients with radiologic or macroscopic peritoneal metastases (M1 disease) were excluded. The primary objective was to assess the prognostic impact of positive peritoneal cytology on overall survival in patients without overt metastatic disease.

Among the 837 included patients, the median age was 66 years, and 71% were male. Non-distal gastric cancer was most common (47 %), and the majority of tumors were advanced at diagnosis, with 59% staged as pT3–4 and 43% as pN2–3. Positive peritoneal cytology was detected in 66 patients (7.9%). Those with positive cytology (CY⁺) tended to have more advanced T and N stages, consistent with greater disease burden. Patients with CY⁺ disease had a median overall survival of 37 months and not reached for CY⁻ patients (p = 0.06). After multivariable adjustment for confounders—including pathological stage, tumor location, and administration of neoadjuvant therapy—positive cytology was not independently associated with worse 5-year overall survival (HR 1.06, 95% CI 0.61–1.8).

The study also identified significant heterogeneity in cytology sampling techniques across institutions. Specifically, the number of peritoneal quadrants sampled and the lavage methods varied, and while there was no clear trend, there was an association with overall survival. Centers that sampled multiple quadrants detected a higher proportion of positive cytology cases, suggesting that under-sampling may lead to underestimation of true disease burden.

Key limitations included the retrospective nature of the study, which introduces potential for selection bias, and the relatively small number of cytology-positive patients, limiting statistical power for subgroup analyses. The lack of standardization in peritoneal lavage technique across centers further complicates interpretation of the results, and long-term outcomes beyond five years were not evaluated.

Bottom Line: In this large multi-center European and Asian cohort, positive peritoneal cytology in patients with gastric cancer without macroscopic metastases was not independently predictive of worse 5-year overall survival after adjustment for tumor stage and treatment factors. These findings challenge the conventional view that cytology-positive status should automatically exclude patients from curative surgical intervention. Instead, they highlight the importance of integrating cytology results with other clinical and pathological factors when determining surgical eligibility. 

Trauma

Predicting and Preventing Gun Violence: An Experimental Evaluation of READI Chicago
Monica P Bhatt, Sara B Heller, Max Kapustin, Marianne Bertrand, Christopher Blattman, The Quarterly Journal of Economics May 2024
Contributor: Joe Kern

Brief synopsis

This randomized control trial studied the impact of violence, arrests, and death among individuals participating in a community-researcher partnership, Rapid Employment and Development Initiative (READI) Chicago. The trial enrolled nearly 2500 individuals and showed evidence of reduced shooting and homicide arrests in the READI arm compared to controls. Additionally, there was an estimated $182,000 to $916,000 per participant social costs savings.

Summary: This study is a randomized controlled trial of a Chicago-based community violence intervention, Rapid Employment and Development Initiative (READI). READI works in 5 Chicago neighborhoods with the highest rates of violence and offers high risk individuals an 18-month subsidized employment program along with cognitive behavioral therapy. High risk individuals were identified based on a machine learning algorithm with 3 components: administrative arrests and victimization records, referrals from local outreach workers and screening among individuals leaving prison and jail. 2,456 participants in the study were randomized to either participate in READI or a control group that was free to pursue other community-based services. The primary outcome was violent crime associated events, which was measured by shooting and homicide victimization, arrest or other serious violent crime arrests, such as robbery or battery.

For every 100 people in the control group, there were 11 shootings and homicide victimizations among the intervention group. There were no statistically significant differences between groups for the primary outcome, across all three violent crime categories. There was however a 65% relative reduction in suicide and homicide related arrests, though this was not statistically significant. Additionally, participants who were referred to READI by outreach workers (e.g. HVIP specialists), had the greatest decrease in serious violent involvement compared to matched controls (79 vs 43%). The authors estimated cost savings between $182,000 and $916,000 per participant, and a benefit to cost ratio of READI to be between 4:1 and 18:1.

Hospital-based and community-based violence intervention programming often provide individualized services, leading to immense heterogeneity and presenting a challenge to evaluating efficacy. The authors provide a robust study of a community-based violence intervention program, highlighting the challenges in their evaluation. While the authors show no difference in their primary outcome with READI participation, they do demonstrate benefit. This study demonstrates the intensive, multipronged community-based programming that addresses upstream social determinants of health is feasible and cost-effective. It also highlights the positive impact outreach workers, such as HVIP specialists, have on outcomes for survivors of violent injury, noting a reduction in serious violent crime.

Bottom line: Targeted, multi-pronged, and coordinated violence intervention programming, is both feasible and cost-effective and can reduce recurrent involvement in violence.

Penn Evidence-Based Literature Review (PEBLR)

Summarized highlights from contemporary literature in surgical and allied disciplines for general surgery residents.

Critical Care

IMpact of PerioperAtive KeTamine on Enhanced Recovery After abdominal Surgery (IMPAKT ERAS): a pragmatic randomized single-cluster trial
Raymond BL, Allen BFS, Freundlich RE, McEvoy MD, Parrish CG, Vanderbilt Learning Healthcare System Platform Investigators et al. Br J Anaesth. 2025 Sep 2:S0007-0912(25)00508-2
Contributor: Iulia Barbur

Brief synopsis

This is a randomized, double-blind, placebo-controlled clinical trial assessing the impact of ketamine on post-operative hospital length-of stay, total opioid consumption, incidence of post-operative ileus, and incidence of side-effects or adverse events in adult patients undergoing abdominal surgery following an existing ERAS protocol. The use of ketamine was not associated with a shorter hospital length of stay but was associated with a higher rate of adverse events including rapid responses and ICU transfers.  

Summary: This trial randomized 1,522 adult patients undergoing abdominal surgery following an ERAS protocol (not specified beyond “colorectal, ventral hernia, or surgical oncology ERAS protocols”) to receive either ketamine or normal saline via intravenous bolus at the time of anesthesia induction and continuing through 48 hours post-operatively. The primary outcome was hospital length-of-stay, while secondary outcomes included total opioid consumption and incidences of adverse events. Patients transferred directly from the operating room to the ICU still intubated and patients whose operation was aborted were excluded from all analysis. 

The median duration of hospital stay was five days in each group; authors state that regression analysis showed patients receiving ketamine to have an odds ratio of 1.21 for prolonged hospital stay, however the 95% confidence interval included 1.00 and is therefore not statistically significant (1.00—1.47). There was no difference in opioid consumption between ketamine and placebo groups. Patients in the ketamine group did experience higher odds of an ICU transfer, with an odds ratio of 2.03 (95% CI 1.14—3.63) and more often received antiemetics due to significant post-operative nausea and vomiting within 48 hours of surgery (55% versus 47%, p = 0.002). Randomization of patients to the ketamine group was not associated with lower odds of ileus requiring gastric decompression. Of note, the study drug was discontinued due to adverse effects in 32% of the ketamine group versus 13% of the placebo group, and all analysis was completed on an intention-to-treat basis. The high rate of drug discontinuation due to adverse effects, the limited patient population included, and its single center nature are the chief limitations of this trial.

Bottom line: Patients undergoing ERAS pathway abdominal operations randomized to ketamine versus placebo were more likely to experience post-operative ICU transfers and other adverse events, without evidence of shorter hospital stay or lower post-operative opioid consumption.

HPB

Parenchymal-sparing hepatectomy vs single-stage major hepatectomy for bilateral colorectal liver metastases: a higher rate of microscopically positive margins is not associated with increased local recurrence after parenchymal-sparing hepatectomy
Mateo Lendoire, Harufumi Maki, Antony Haddad, Anish J. Jain, Hop S. Tran Cao, et al., J Gastrointest Surg. 2025 Jul;29(7):102070
Contributor: Gracia Vargas

Brief synopsis

This retrospective study assessed surgical and oncologic outcomes of parenchymal-sparing hepatectomies (PSH) versus major hepatectomies (MH) for colorectal liver metastases. Among 206 patients who underwent curative-intent hepatectomy for ≥5 bilaterally distributed colorectal liver metastases at a single-institution, MHs were associated with increased estimated blood loss, length of stay, and rate of postoperative hepatic insufficiency, while PSHs had higher rates of R1 resection. There was no significant difference in overall survival or recurrence-free survival between the two groups. 

Summary: Colorectal liver metastases (CLMs) affect approximately one in four patients with colorectal cancer. Surgical resection is recognized as the gold standard treatment for CLMs, and is associated with a favorable (>50%) overall survival among well-selected patients, including those with bilateral CLMs. Nevertheless, there is not a clear consensus regarding the optimal surgical approach to CLM resection, as parenchymal-sparing hepatectomy (PSH) raises concern for margin status and local recurrence while major hepatectomies (MH) raise concern for major morbidity. This retrospective study aimed to evaluate the surgical and oncologic outcomes of R0-intent single-stage hepatectomies among patients with ≥5 bilaterally distributed CLMs to compare outcomes of PSHs versus MHs. 

Among 206 patients from 1998 to 2022, 90 (44%) underwent PSH, while 116 (56%) underwent MH. 59% of patients were male, and 98% received chemotherapy prior to hepatectomy. There were no significant differences between PSH and MH in major surgical complications (Clavien Dindo ≥ 3), reoperations within 45 days of surgery, and 90-day mortality. MH had significantly higher estimated blood loss (median 350 mL, IQR 250-575 versus 250 mL, IQR 158-400 for PSH; p=0.002), length of stay (median 6 days, IQR 5-8 versus 5 days, IQR 4-7 for PSH; p<0.001), and hepatic insufficiency (12.9% incidence vs 1.1% incidence; p=0.003), and significantly lower rates of R1 resection (30.2% versus 46.7%, p=0.020). Of note, there was no difference in rates of R1 resection between the two groups when stratifying by total number of tumors (p=0.165). 

With a median follow-up of 48.5 months, the median overall survival was 51.0 months (IQR 43.9-58.0) and recurrence-free survival was 8.2 months (IQR 7.1-9.3). 173 patients (84.0%) developed a recurrence, of which 106 (61%) were intrahepatic. Kaplan-Meier analyses found no significant difference between the two groups for overall survival (p=0.072) or recurrence-free survival (p=0.585). Additionally, there was no difference in hepatic-specific disease-free survival (median 11.6 months, IQR 7.9-15.3; p=0.320). Multivariate analysis found no association between resection type and overall survival. 

Study limitations included exclusion of patients who required combined ablation or extrahepatic resection, significant differences in length of follow-up between PSH (median 30.0 months, IQR 24.5-35.5) and MH (median 60.8, IQR 44.2-77.3), and variability in surgical practice – notably in minimally invasive surgery – over the two decades included in the study. Despite these limitations, this study provides compelling evidence to support the use of PSH among patients with bilateral colorectal liver metastases. 

Bottom line: This single-institution retrospective study of patients who underwent hepatic resection for ≥5 bilaterally distributed colorectal liver metastases found that parenchymal-sparing hepatectomy had similar short-term surgical outcomes and long-term oncologic outcomes when compared to major hepatectomy, despite higher rates of R1 resection, supporting the use of parenchymal-sparing hepatectomy as a surgical approach in these patients.

Vascular Surgery

Drug-Gene Interactions and Clinical Outcomes After Vascular Surgery in the Million Veteran Program
Sony Tuteja, William J. O’Brien, Jeffrey P. Ferraro, Scott M. Damrauer, Kamal M. F. Itani, et al. for the Million Veteran Program JAMA Surg 2025;160;(7):804-813.
Contributor: Dasha Savage-Robinson

Brief synopsis

Synopsis: This retrospective cohort study utilizes the VA Million Veteran Program (MVP) and VA Surgical Quality Improvement Program (VASQIP) to examine the impact of drug-gene interactions on post-surgical outcomes in vascular surgery. Patients with two or more drug-gene interactions experienced longer hospital stays, higher 30-day readmission rates, and increased rates of the composite outcome of stroke, myocardial infarction, or myocardial injury after noncardiac surgery, underscoring the clinical relevance of pharmacogenetic interactions in adverse outcomes.

Summary: Adverse drug reactions, influenced by individual patient variations in drug metabolism, contribute to perioperative morbidity and mortality. Pharmacogenetic (PGx) information, which tailors medication based on a patient’s DNA profile, has been translated into clinical practice for many US FDA approved drugs and contributed to improved outcomes. However, the impact of PGx on vascular surgery patients, who are frequently prescribed drugs with known drug-gene interactions during the perioperative period, remains underexplored.

To address this, the study analyzed genetic data from the MVP and surgical outcomes from the VASQIP for veterans undergoing vascular procedures between January 2011 and December 2022. Patients with documented vascular surgeries and genetic data were included, with drug exposure assessed from 30 days before to 7 days after surgery. Outcomes evaluated included length of stay (LOS), 30-day readmission, a composite outcome (stroke, myocardial infarction, or myocardial injury after noncardiac surgery), and 30-day postoperative mortality. Statistical analyses used mean (SD) for continuous variables, logistic regression for binomial variables, and negative binomial regression for count variables, with all models adjusted for the same covariates.

The study included 10,098 veterans (mean [SD] age, 68.8 [8.3] years; 15.7% Black [self-reported]; 97.9% male) after excluding those with multiple procedures, long-term dialysis, missing baseline eGFR, or incomplete genetic data. Of these, 49.7% had drug-gene interactions, primarily involving proton pump inhibitors (associated gene: CYP2C19), statins (SLCO1B1), and clopidogrel (CYP2C19). For each drug studied, the authors identified patients in the cohort with phenotypes that were actionable: poor, intermediate, rapid, or ultrarapid metabolizers. Patients were stratified by the number of drug-gene interactions (0, 1, 2, or 3+). Compared to those with no interactions, patients with any drug-gene interactions had significantly longer LOS: 0 interactions, 3 (1–6) days; 1 interaction, 3 (1–7) days (P < .001); 2 interactions, 3 (1–7) days (P < .001); 3+ interactions, 4 (2–8) days (P < .001). Both 30-day readmission and composite outcome rates were significantly higher in patients with 2 or 3+ interactions: readmission rates were 17.4% (0 interactions), 17.6% (1 interaction, P = .84), 21.2% (2 interactions, P = .004), and 25.1% (3+ interactions, P < .001); composite outcome rates were 3.5% (0 interactions), 4.1% (1 interaction, P = .24), 5.7% (2 interactions, P = .001), and 5.5% (3+ interactions, P = .02).

Limitations of the study include the predominantly male cohort (97.9%), potential underreporting of events due to unavailable Medicare data, and the study’s focus on pharmacokinetic rather than pharmacodynamic responses or drug-drug interactions. Copy number variants were also not assessed.

Bottom Line: Drug-gene interactions in vascular surgery patients are associated with longer hospital stays, higher readmission rates, and increased cardiovascular morbidity.