February-April 2018

Synopsis: Liver transplantation is the most effective treatment for end-stage liver failure, but the supply of organs falls short of demand. Increased utilization of marginal quality grafts, particularly those from donation after cardiac death, offers a way to reduce this shortage but is limited partially because such grafts tolerate static cold storage (SCS) poorly. Normothermic machine perfusion (NMP) has been proposed as an alternative preservation process to reduce ischemic injury and increase marginal graft utilization. In this multi-center study, 335 donor liver allografts were randomized to SCS or NMP resulting in 220 transplants. Compared to SCS, NMP was associated with a 50% reduction in AST release and a >50% reduction in early allograft dysfunction along with a 50% reduction in discard rates and a 54% longer mean preservation time. Although not powered to examine patient and graft survival explicitly, this study suggests that NMP offers a way to both improve post-transplant outcomes and increase organ utilization. The major limitation of this technology is its higher cost relative to SCS.

Synopsis: Rectal cancer outcomes have improved greatly over the past decades due to improvements in the quality of surgery and multidisciplinary treatment. In particular, the avoidance of circumferential resection margin (CRM) positivity has been recognized as one of the most important determinants of local and distant recurrence as well as overall survival. Anticipated CRM positivity ranges from 8-12% in the hands of experienced specialists. The aim of this study was to identify risk factors associated with CRM positivity using the Binational Colorectal Cancer Registry (BCCA) of the Colorectal Surgical Society of Australia and New Zealand. Among 3367 patients between 2007 and 2016, the overall CRM positivity was 7.75%. This is in contrast to NCDB findings from 2010 to 2011 of 16,619 patients with a CRM positivity of 17.2%. On univariate analysis, low socioeconomic status (9.8% vs 6.9%; p=0.033), public patients (9.4% vs 6.1%; p<0.001), hospital location by state (p=0.025), urgent cases (17.3% vs 7.4%; p<0.001) and open surgery (10% vs 3.9%; p<0.001) were associated with positive CRM. Additionally, low tumors, low to minimal response to neoadjuvant chemoradiotherapy, and T3/T4 or N1/N2 tumors were more likely to have a positive CRM. Hierarchical logistic regression analysis identified 6 independent risk factors associated with a positive CRM: urgent operation (OR=1.88), open approach (OR=1.61), abdominoperineal resection (OR=1.14), tumor height <8cm (OR=1.81), T3/T4 (OR=7.62), and positive lymph nodes (OR=2.02). These can be attributed to the technical challenges as the mesorectum narrows in the distal rectum, in addition to the narrowing pelvis. In conclusion, there are a number of factors associated with CRM, are care should be taken to avoid positive CRM to improve outcomes. 

Synopsis: Guidelines regarding perioperative antibiotic administration have been previously published by many groups, including the Centers for Disease Control (CDC) and the Surgical Care Improvement Project (SCIP). These guidelines, however, are not applicable to patients less than one year of age, and great variability in antibiotic treatment practices exists for this age group. In this retrospective review of 275 neonates with congenital surgical conditions who underwent surgical correction in the first month of life, Walker et al. discuss infectious outcomes before and after the implementation of an antibiotic stewardship program at their institution. This antibiotic protocol focused on eliminating empiric postnatal antibiotics in the absence of infection and limiting perioperative antibiotics to within 72 hours of surgery. The primary outcome was surgical site infection (SSI) at 30 days, with secondary outcomes including the development of hospital-acquired infections (HAIs) and infection and/or colonization with multidrug-resistant organisms (MDROs). Pre-protocol patients (1/2009-6/2012, n = 127) were compared to post-protocol patients (7/2012-3/2016, n = 148). Overall, compliance with the protocol guidelines was relatively high at 89% and the median duration of antibiotics given postnatally as well perioperatively was reduced after implementation of the protocol. SSI, HAI, and MDRO all decreased in the post-protocol group compared to the pre-protocol group, although these differences were not statistically significant. The authors conclude that elimination of empiric postnatal antibiotics and limitation of perioperative antibiotic prophylaxis to within 72 hours of surgery did not increase the rate of SSI, HAI, or MDRO. While this study is limited by its retrospective nature and single institutional experience, the results suggest that more judicious use of antibiotics in surgical neonates is safe and may be beneficial. 

Synopsis: Administration of supplemental oxygen is nearly ubiquitous in critically ill patients, irrespective of the presence of hypoxemia. Many providers view this practice as harmless but there are potential adverse effects of hyperoxemia, such as reduced cardiac output, cerebral and coronary vasoconstriction, inflammatory cytokine production, central nervous system toxicity and absorption atelectasis. Furthermore, there is little high-quality evidence to support this practice. The Improving Oxygen Therapy in Acute-illness (IOTA) systematic review and meta-analysis was designed to address this issue. The IOTA trial compiled data from 25 randomized controlled trials, containing 16,037 patients with critical illness, sepsis, stroke, trauma, myocardial infarction, cardiac arrest or which had emergency surgery. Of note, 43% of the patients admitted with critical illness and sepsis were admitted for a surgical diagnosis. Studies in patients <18 years old were excluded and also studies that were limited to patients with chronic respiratory disease, psychiatric disease, patients receiving extracorporeal life support, treated with hyperbaric oxygen or who underwent elective surgery. Within the included trials, liberal oxygen supplementation provided a median FiO2 of 0.52 (range 0.28-1.00: IQR 0.39-0.85) for a median duration of 8 hours (range 1-144 hours; IQR 4-24) compared with conservative oxygen supplementation (median FiO2 0.21, range 0.21-0.50; IQR 0.21-0.25). The baseline median SpO2 in the liberal oxygen arm was 96.4% (range 94.0-99.0%). Mortality data was compiled and demonstrated that a liberal oxygen strategy increased the risk of in mortality compared to a conservative strategy while in-hospital (n=15,071, RR 1.21 [95% CI 1.0-1.43], p=0.020, I2=0), at 30 days (n=15,053, RR 1.14 [1.01-1.28], p=0.033, I2=0) and at longest reported follow-up (median 3 months, n=15,755, RR 1.10 [1.00-1.20], p=0.044, I2=0). This equates to an absolute risk increase of in-hospital mortality of 1.1% (95% CI 0.2-2.2) and 30-day mortality of 1.4% (0.1-2.7). In addition, meta-regression analysis demonstrated that as SpO2 increased, liberal oxygen therapy was associated with a higher RR of in-hospital mortality (slope 1.25 [95% CI 1.00-1.57], p=0.0080). There was no significant between-group differences in risk of hospital-acquired infection, hospital-acquired pneumonia or length of hospital stay. In summary, the IOTA systematic review and meta-analysis provides high-quality evidence that hyperoxia, when applied to a heterogeneous population of critically ill patients, is harmful and leads to a clinically and statistically significant increase in mortality. Further investigation is needed to determine upper limits for safe administration of oxygen therapy.

Synopsis: Social determinants of health, including socioeconomic status, are known to impact health outcomes, but the effect of acute negative wealth shock (large sudden loss of net worth, defined as >75% loss within 2 years in this study) on long-term mortality is unknown. This paper analyzed data from the Health and Retirement Study, a nationally representative prospective cohort of US adults aged 51-61 at study entry in 1994 through 2014, to assess whether negative wealth shock affected all-cause mortality as compared to populations experiencing positive wealth in the same time period. Marginal structural survival methods were used to adjust for change in health status as the triggering event for both negative wealth shock and mortality. Results demonstrated that, projected to the US population, 26.2% of middle-aged adults experienced negative wealth shock with a median net worth change of -92.4% and adjusted HR of death of 1.50 (1.36-1.67) compared to the population experiencing positive wealth. The HR of death for negative wealth shock was similar to the HR of death for baseline asset poverty (0 or negative total net worth, known to increase risk of mortality) in the same time period (1.67, 95% CI: 1.44-1.94). Limitations include 1) is not generalizable to other age groups experiencing negative wealth shock, 2) individual causes of negative wealth shock are unknown, and 3) unknown impact of less acute forms of negative wealth shock. This study’s findings illuminate the large proportion of the population affected and opportunities for targeted interventions to improve overall health outcomes.