Penn Evidence-Based Literature Review (PEBLR)

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

November 2017 - January 2018

  1. Roux-en-Y or Billroth II Reconstruction After Radical Distal Gastrectomy for Gastric Cancer
  2. Bok-Yan So et al.  Annals of Surgery.  February 2018 (PDF)
    Contributor: Brett Ecker
    Discipline: Surgical Oncology

Synopsis: Following distal gastrectomy, Billroth II (B-II) and Roux-en-Y (R-Y) are the predominant reconstruction techniques used in the West. B-II is a simpler procedure to perform, requiring one fewer anastomosis, but is associated with a high risk of bile reflux. In an effort to further delineate the long-term clinical gastrointestinal symptoms following these reconstruction techniques, 162 patients from 4 centers in Hong Kong and Singapore were randomized to B-II or R-Y in a multicenter prospective RCT. Primary outcomes were QOL and Gastrointestinal symptom severity (epigastric pain, heartburn, biliary vomiting, postprandial bloating, and nausea) assessed at 1 year. There were no differences in perioperative morbidity – including anastomotic leak – although the B-II procedure was performed 21 minutes faster. At 1 year, patients with R-Y were found to have a lower median endoscopic grade for gastritis, although there were no differences in patient-reported QOL or global GI symptom scores. Most patients did not report any symptoms (57% and 62% of B-II and R-Y patients, respectively), and none reported severe symptoms (grade 4). By subgroup analysis of the individual symptom scores, patients of B-II reconstruction reported more frequent heartburn symptoms. In conclusion, pouch gastritis is more common after B-II reconstruction although it is frequently clinically insignificant and not globally different than R-Y reconstruction. The long-term (years+) effect of chronic bile reflux remains to be determined, as prior studies in non-cancer patients who underwent resection for duodenal ulcers demonstrated greater risk for Barrett’s esophagus with B-II reconstruction.


  1. Long-term outcome of renal transplantation from octogenarian donors: A multicenter controlled study
  2. Ruggenenti et al.  American Journal of Transplantation.  August 2017 (PDF)
    Contributor: David Aufhauser
    Discipline: Transplantation

Synopsis: Demand for kidney transplantation far exceeds supply. Expansion of the donor pool offers one way to reduce this discrepancy. In this observational study, 3 Italian transplant centers used clinical and histologic criteria to select allografts from >80 year old donors for either single or duel kidney transplantation and compared the outcomes to reference-recipients of allografts from donors <60 yrs old selected based on clinical criteria alone. In 37 recipients of allografts from >80 years and 198 reference-recipients, the authors observed similar outcomes in terms of graft loss and serum creatinine. The recipients of allografts from >80 yr old donors had dramatically shorter waitlist times. The study is limited by small numbers and a flawed comparison group (all received single kidney transplants and were subjected to less rigorous selection process), but nevertheless provides interesting evidence to support consideration of use of kidney from older donors. The use of histology to prevent unnecessary discard of kidneys deemed low quality based on clinical criteria and the use of duel kidney transplantation to ensure transplantation of adequate nephron mass in donors who would otherwise go unused are both interesting concepts to address organ scarcity.


  1. A Population-Based Study of Complication After colorectal Surgery in Patients Who Have Received Bevacizumab
  2. Baxter et al.  Diseases of the Colon and Rectum.  January 2018 (PDF)
    Contributor: Charles Vining
    Discipline: Colorectal Surgery

Synopsis: Bevacizumab (BV; Avastin) is a monoclonal antibody vascular endothelial growth factor (VEGF) inhibitor used to treat Stage IV colorectal cancer (CRC). Randomized controlled trials have demonstrated its potential to inhibit wound healing, with increased wound complications in patients who received BV preoperatively versus those who were exposed to chemotherapy alone (13% vs 3.4%). The aim of this study was to identify rates of complications in patients who underwent colorectal surgery after receiving BV as well as the temporal relationship between the two. Using an Ontario database, patients with Stage IV CRC who received BV between Jan. 2008 and Dec. 2011 were identified. Of the 2759 patients, 265 (9.6%) underwent a colorectal procedure. Of these patients, 20.8% had surgery ≤28 days following BV, 32.1% had surgery between 29 days and 3 months and 47.2% had surgery >3 months following their last BV treatment. Patients who received BV within 28 days of surgery were more likely to have emergent surgery (92.7%). The overall postoperative complication rate was 36.2% including death (7.9%), readmission (20.4%) and surgical complications (including wound complications – 12.5%) within 30 days. On multivariate analysis, there was no difference in complication rate between groups who received BV within 28 days, 29 days to 3 months and >3 months. Emergency procedure was the only variable on multivariable analysis to be associated with increased complications. BV has a half-life of 20 days and even low levels can inhibit VEGF. As VEGF is key to angiogenesis and therefore wound healing, theoretically BV will have a negative effect on surgical outcomes. Although this study failed to detect a relationship between last BV dose and surgical morbidity, this patient population is at substantial risk for postoperative complications even with long interval since exposure.


  1. Adjunctive Glucocorticoid Therapry in Patients with Septic Shock
  2. Vankatesh et al.  New England Journal of MEdicine.  January 2018 (PDF)
    Contributor: Charlie Vasquez
    Discipline: Critical Care

Synopsis: Sepsis is a highly lethal condition with mortality ranging from 30-45%. Previous randomized trials on the use of hydrocortisone in patients with septic shock showed conflicting effects on mortality, despite both demonstrating an earlier a reversal of shock. The Adjunctive Corticosteroid Treatment in Critically Ill Patients with Septic Shock (ADRENAL) study was designed to further address this contentious topic. Patients enrolled in the ADRENAL study were adults >18 years old, on mechanical ventilation with strong or confirmed evidence of infection, at least two systemic inflammatory response syndrome (SIRS) criteria, and who were currently being treated with vasoactive medications for at least four hours. Patients were excluded if they were likely to receive treatment with systemic glucocorticoidsfor another indication, had received Etomidate during the current admission, or if they were deemed likely to die from a pre-existing co-morbid disease within 90 days after randomization. A total of 3658 patients (from 5501 eligible patients screened) were enrolled from 69 medical-surgical ICUS in five countries. A total of 1826 were randomized to the placebo group and 1832 were randomized to the hydrocortisone group. The hydrocortisone group received a continuous infusion of 200mg IV hydrocortisone per day for a maximum of 7 days or until ICU discharge or death, whichever occurred first. The primary outcome was all-cause mortality at 90 days. Secondary outcomes included all-cause mortality at 28 days, time to resolution of shock, recurrent shock, ICU and hospital length of stay, duration of mechanical ventilation, incidence/duration of renal-replacement therapy, incidence of new-onset bacteremia or fungemia and receipt of blood transfusion. At 90 days, there was no difference in mortality between the hydrocortisone group was 27.9% vs. 28.8% for the placebo group (OR 0.85, 95% CI 0.82 to 1.10; P=0.50). Notable differences in secondary outcomes include shorter time to shock resolution in hydrocortisone group (median 3 days vs. 4 days; hazard ratio, 1.32; 95% CI, 1.23 to 1.41; P<0.001), shorter time to ICU discharge in the hydrocortisone group (median, 10 days vs. 12 days; hazard ratio 1.14; 95% CI, 1.06 to 1.23; P<0.001), shorter duration of initial mechanical ventilation (median, 6 days vs. 7 days; hazard ratio, 1.13; 95% CI, 1.05 to 1.22; P<0.001). Fewer patients in the hydrocortisone group received a blood transfusion (37.0% vs. 41.7%; OR 0.82; 95% CI, 0.72 to 0.94; P=0.004). There were no significant differences in recurrent shock episodes, time to hospital discharge, recurrent need for mechanical ventilation, use of renal-replacement therapy or new onset bacteremia/fungemia. In summary, this double-blind, placebo-controlled RCT showed that in patients with septic shock requiring mechanical ventilation and vasopressor use for at least four hours, there was no mortality benefit at 90 days with administration of continuous IV hydrocortisone (200mg/day). However, the hydrocortisone group did demonstrate clinically and statistically significant improvement in a number of secondary outcomes, including faster time to shock resolution, shorter time to ICU discharge, shorter duration of initial mechanical ventilation and fewer blood transfusions, all without an increased incidence of new onset bactermia/fungemia.


  1. Angiotensin II for the Treatment of Vasodilatory Shock
  2. Khanna et al.  New England Journal of Medicine.  December 2017 (PDF)
    Contributor: Charlie Vasquez
    Discipline: Critical Care

Synopsis: Shock is a clinical syndrome characterized by decreased tissue perfusion. The most common type of shock is vasodilatory shock in the setting of sepsis. Until recently, only two classes of vasopressors were available, catecholamines and vasopressin. The Angiotensin II for the Treatment of High-Output Shock (ATHOS-3) is a multinational, double-blind, randomized-controlled trial (RCT) that tested the efficacy of Angiotensin II (AT-II) as an additional, mechanistically unique vasopressor. Patients enrolled in the ATHOS-3 study were age >18 with vasodilatory shock and high-dose vasopressor requirement. Vasodilatory shock was defined to require the following: cardiac index > 2.3 L/min/m2 or ScvO2 >70% with CVP > 8mmHg, MAP between 55-70mmHg and >0.2 ug/kg/min norepinephrine (or equivalent dose of another vasopressor) for at least six hours but <48 hours. Patients with the following diagnoses were excluded: active bleeding, mesenteric ischemia, abdominal aortic aneurysm, absolute neutrophil count <1000/cm3, liver failure, bronchospasm, burns >20% total-body surface area, acute coronary syndrome, receiving venoarterial extracorporeal membrane oxygenation, or on high-dose glucocorticoids. A total of 344 patients across 75 ICUs in North America, Australasia and Europe were enrolled (from 404 screened) and randomly assigned in 1:1 fashion to receive either saline placebo or AT-II at rate of 20ng/kg/min (titrated to MAP >75mmHg; max dose 300ng/kg/min). The primary end point was response of MAP at hour three, defined as MAP >75mmHg or increase from baseline of >10mmHg, without increase in background vasopressor dose. Secondary endpoints included change in cardiovascular and total SOFA score. The primary endpoint was achieved in significantly more patients in the AT-II group vs. placebo (69.9% vs. 23.4%, P<0.001). As such, the AT-II group showed a significantly greater increase in MAP (12.5mmHg vs. 2.9mmHg, P<0.001). The change in cardiovascular SOFA was significantly different in the AT-II group compared to placebo (-1.75 vs. -1.28, P<0.01), but no significant difference was observed in total SOFA score change between the AT-II group compared to placebo (1.05 vs. 1.04, P=0.49). There were no statistically significant differences in 28-day or all-cause mortality. In summary, this double-blind RCT in patients with vasodilatory shock requiring high-dose vasopressors, demonstrated that significantly more patients receiving AT-II achieved the primary end-point of MAP>75mmHg or increase MAP>10mmHg compared to saline placebo group. Importantly, this study was not powered to detect differences in mortality and larger trials are indicated to address this question. Lastly, Angiotensin-II received FDA approval on 12/21/17 and is expected to become available after March 2018.


  1. Future Supply of Pediatric Surgeons: Analytical Study of the Current and Projected Supply of Pediatric Surgeons in the Context of a Rapidly Changing Process for Specialty and Subspecialty Training
  2. Ricketts TC et al. Annals of Surgery, March 2017 (PDF)
    Contributor: Robert Swendiman
    Discipline: Pediatric Surgery

Synopsis: There is a growing concern the supply of pediatric surgeons in outstripping the demand for infant and children’s surgery. Fellowship slots have increased over the past few decades, and the population has not kept up with this overall “inflation.” If this is the case, the ability to maintain an appropriate case volume to keep up skills is in jeopardy. Ricketts et al. used forecasting software in three scenarios: 24, 34, and 56 new pediatric surgeon entrants per year (corresponding to the number of fellowships offered). Currently, there are approximately 54 fellowship graduates per year. Based on the projected “headcount” of the national pediatric population, if all 54 fellowships continue to graduate trainees, then “growth of the pediatric surgeon headcount would be 5 times greater than the pediatric population growth between 2013 and 2030.” While the model isn’t perfect, this study certainly echoes concerns that we are training too many pediatric surgeons. Of note, this study did not factor in distribution of surgeons, as some argue that the real workforce issue is geographical maldistribution.