November-December 2017

Synopsis: There are no reproducible technical/technological maneuvers (e.g., reinforced stapler, fibrin glue, oversewing main pancreatic duct, etc.) that mitigate the incidence of clinically relevant post-operative pancreatic fistula (CR-POPF) following distal pancreatectomy (DP). The DISCOVER trial aimed to assess whether coverage of the pancreatic stump with a teres ligament patch would reduce CR-POPF. 152 patients undergoing DP were randomized into teres ligament coverage (n=76) vs. no coverage (n=76); the primary outcome was incidence of CR-POPF. In the coverage cohort, rate of reoperation (p=0.009) and readmission (p=0.011) were significantly lower. Notably, CR-POPF rates were NOT significantly different between groups; however, multivariable analysis demonstrated teres ligament coverage to be a protective factor for CR-POPF (p=0.0146).This study has several limitations: 1) Despite it's intent-to-treat design, 8 pts (3 study, 5 control) crossed over into opposite arms and were excluded from the final analysis; 2) gland texture, a well-documented contributor to CR-POPF, was not accounted for in this study, raising the possibility that several other pertinent confounding variables were unaccounted for in its design; 3) method of pancreatic transection (stapler vs. scalpel/oversewing of MPD) was not standardized in the study, and surgeon discretion in this regard may have influenced its outcomes; 4) the primary outcome, rate of CR-POPF, was non-significantly different-- which would suggest a negative study. 

Synopsis: Statins are the most commonly prescribed medication but evidence and guidelines regarding who benefits from statin use in the early perioperative period for noncardiac surgery are limited. The authors conducted a retrospective propensity-score matched cohort study amongst veterans (using the VA Surgical Quality Improvement Project-- VASQIP database) to determine whether exposure to statin on the day of or the day after surgery was associated with 30-day all-cause mortality or nonfatal complications. Results recapitulated findings from the prospective non-randomized VISION study (Berwanger et al., Eur Heart J 2016), and demonstrate that early perioperative exposure to statins was associated with decreased 30-day all-cause mortality (RR 0.82) and decreased rates of any nonfatal complication (RR 0.82) and particularly decreased rates of cardiac complications (RR 0.73). The major limitations of the study include the retrospective nature of the study (demonstrating association, not causation), and the non-representative surgical population utilized in the study (96% male, 4% female; increased comorbidity in VA patients). 

Synopsis: The optimal renal replacement strategy for HIV-infected ESRD patients is unknown as these patients have both increased mortality on dialysis and increased morbidity after kidney transplantation. HIV-infected patients who were waitlisted for renal transplant were identified by linking a national transplant database (which does not include HIV status) to pharmacy records of prescriptions for HIV medications. HIV-infected renal transplant recipients had a 79% reduction in risk of mortality (aRR of 0.21) compared to patients who remained on the waitlist, with statistically significant survival benefit being achieved by 194 days post-transplant. The survival advantage of renal transplantation was more pronounced in patients co-infected with HIV and HCV, who had 91% reduction in mortality. This study uses innovative two-database design to examine a question beyond the scope of traditional transplant database and makes a strong case transplanting HIV-infected ESRD patients. However, many secondary questions remain including: 1) is there differential benefit with different HIV treatment regimens; 2) does pre-transplant viral load/CD4 count affect post-transplant outcomes; and--most importantly--3) in light of the recent HIV Organ Policy Equity Act, is the same benefit achieved with a kidney from an HIV-infected donor. 

Synopsis: The ACS-NSQIP surgical risk calculator is a freely available online resource for estimating the postoperative risks by procedure type, patient demographics and current health state according to national registry data. To date, their impact on physicians’ clinical assessment and decision-making has not been studied. This study surveyed a national sample of general surgeons, who were asked to review 4 urgent inpatient clinical scenarios (mesenteric ischemia, GI bleeding, SBO, appendicitis) and supply the risks and benefits of operative and non-op management, as well as how likely they would be to recommend surgery. Surgeons were randomized to a control group (vignettes only) or to those using the NSQIP risk calculator (RC) along with the vignette. 779 surgeons participated. Despite having risk calculator data, both groups varied considerably in their judgments of risk, but the RC group clustered around the risk provided by the calculator. The risks provided by the NSQIP RC were significantly lower than those in the control group. Having the RC data, however, did NOT influence the likelihood of recommending an operation. There are a few major limitations--one being hypothetical vignettes instead of real cases and a narrow scope of scenarios. This is an interesting paper that lends itself to future studies of surgeon decision-making and patient expectation with the use of a registry-based risk calculator.