October 2022

This Philadelphia-based retrospective cohort study from our Penn Trauma investigators uses geospatial data and network analysis to examine how timely access to trauma care impacts mortality with firearm violence.

Summary: Firearm violence is a worsening public health crisis in the United States; with injury due to interpersonal violence as a major contributor in the urban setting. Delayed access to appropriate trauma center care is associated with worse outcomes and higher mortality; however, associations of access--beyond straight-line distance--to trauma center care in an urban setting and mortality from firearm injury has not been well studied.  The investigators in this study sought to examine these associations.

This retrospective cohort study examined data from the Philadelphia Police Department’s registry of shooting victims. They used shootings classified as criminal and associated geospatial coordinates to explore access to level one and two trauma centers, and associated mortality. The authors examined data from 10,105 Philadelphia shooting victims, aged 15 and older, from 2015 to 2021 and used ArcMap Network Analysis tool to predict fastest transport time to the nearest level one or two trauma center.

Unsurprisingly, the study found that an estimated 23% of fatalities—or 455 deaths--from gun violence could be attributed to increased transport time; with each additional minute of predicted transport time associated with increased odds of mortality. The study utilized a population attributable factor in its analysis; this is an epidemiological measure that assumes a causal relationship, and while this can’t be determined from this observational study it does offer a starting point from which to base further studies and system-wide interventions to improve access to timely trauma care.

The above article is an important contribution to trauma literature, and the methods used can be of particular use to policy makers and hospital systems, beyond the Philadelphia, to find ways to minimize death from firearm injury.  

This retrospective cohort study reports the first set of long-term outcomes data for patients with hepatocellular carcinoma (HCC) who underwent liver transplantation (LT) after downstaging to meet the Milan Criteria (MC). The data validates an American Association for the Study Liver Disease recommendation introduced in 2018 to consider patients downstaged to meet the MC for liver transplantation.

Summary: The Milan criteria stipulates that patients with HCC may be considered for liver transplantation if they have one tumor <5cm in diameter or up to 3 tumors all <3cm, with no angioinvasion and no extrahepatic involvement. Strict adherence to the Milan criteria excludes many patients from liver transplantation, leaving treatment options at surgical resection versus chemoablation. This study evaluates 10-year outcomes for HCC patients who underwent successful liver transplantation after downstaging to MC following locoregional therapy.

Five high-volume US academic centers collected prospective data on all patients undergoing LT for HCC between January 2001 and December 2015. The outcomes of 341 patients whose disease was downstaged to MC were compared to 2122 patients always within MC and 182 patients whose disease remained beyond MC. Primary outcomes of the study were overall survival and HCC recurrence in the MC and downstaged groups. Secondary outcomes were independent variables associated with overall survival or recurrence after transplant, including factors related to downstaging.

Ten-year post-LT survival was 52.1% among downstaged group and 61.5% among the MC group (p<.001). Observed recurrence of HCC (and Kaplan-Meier probability of recurrence at 10 years) was 10.0% (13.3%) for the MC group, 15.8% (20.6%) for the downstaged group, and 35.2% (41.4%) for the group beyond MC. Among those with recurrence, median survival was 14.3 months; 31.9% underwent surgical resection, 26.2% locoregional therapy, 25.9% systemic therapy, and 15.8% opted for supportive care. Those with recurrence who underwent resection had a significantly longer survival (31.6 months) compared to those who underwent locoregional therapy (17.2 months) and systemic treatment (12.5 months), noting that they were more likely to have solitary (69.3%) extrahepatic (89.1%) recurrences.

Patients who underwent liver transplantation for HCC downstaged to MC after locoregional therapy were observed to have 52.1% 10-year overall survival and 20.6% HCC recurrence rate. While these outcomes are inferior compared to outcomes of liver transplants in HCC patients that classically meet MC (61.5% 10-year survival, 10.0% recurrence), they notably outperform the best outcomes for surgical resection or chemoembolization in patients with HCC beyond the MC currently reported in the literature. However, absence of an intention-to-treat analysis and lack of data on waitlist withdrawal further limit the inference of a clear utility benefit for LT prioritization decision making from a retrospective study.

The POISE-3 randomized control trial found that routine administration of tranexamic acid at the beginning and end of non-cardiac surgery was associated with a reduction in adverse bleeding outcomes that was tempered by the study’s inability to demonstrate non-inferiority for composite cardiovascular outcomes.

Summary: 
Tranexamic acid (TXA) is an antifibrinolytic drug that has been shown to reduce the severity of bleeding in trauma as well as patients undergoing cesarean section, cardiac surgery, and potentially orthopedic surgery. Data on the utility of tranexamic acid in other surgical specialties is lacking, including the risk of thrombotic events – a potential downside to augmenting hemostasis. The Perioperative Ischemic Evaluation-3 (POISE-3) Trial set out to determine if treatment with TXA lowers the incidence of perioperative bleeding complications compared to placebo and is non-inferior to placebo in terms of major cardiovascular outcomes.

This multicenter, double-blind, randomized control trial randomized non-cardiac surgery patients to receive a 1g IV bolus of TXA at the beginning and end of surgery (n=4757) or placebo (n=4778). Inclusion criteria included age >45 years old, taking >1 antihypertensive medication and not otherwise requiring TXA as part of their procedure. Notable exclusion criteria included significant heart failure, cranial neurosurgery, and severe renal dysfunction. Primary composite endpoint was life-threatening bleeding, major bleeding, or bleeding into a critical organ at 30 days. The composite safety outcome included myocardial infarction, troponin leak, non-hemorrhagic stroke, peripheral arterial thrombosis, and symptomatic proximal venous thromboembolism. To test non-inferiority of this outcome, the study employed a stringent non-inferiority margin (upper boundary of the of the one-sided 97.5% confidence interval was 1.125)

The POISE-3 study population was on average 70 years, 44% female, and likely to undergo urgent or emergent surgery (79%). The most common specialties included were general (37%), orthopedic (22%), and vascular (15%). Of the 9,535 patients recruited, 4757 were randomized to TXA and 4778 to placebo. The study found a 24% reduction in composite bleeding outcome with administration of TXA compared to placebo, (9.1 vs. 11.7%, HR 0.76; CI 0.67-0.87; p<0.001) driven by a reduction in major bleeding events and complemented by a reduction in transfusions. However, non-inferiority for the composite safety outcome was not met (14.2% vs. 13.9%, HR 1.02, 95% CI 0.92-1.14; p for noninferiority = 0.04) including notable hazard ratios of 1.27 for myocardial infarction (95% CI, 0.89 to 1.82) and 1.51 for non-hemorrhagic stroke (95% CI, 0.8-2.84) suggesting a low probability of a small (0.3%) increased risk.

Of note, this study also featured a 2x2 partial factorial design to evaluate a hypotension-avoidance strategy versus a hypertension-avoidance strategy in this population on one or more hypertensive medications i.e., patients received chronic hypertensive medications with a target MAP of >60 intraoperatively (n=3742) versus only given chronic medications if hypertensive prior to surgery with a target MAP > 80 (n=3748). The authors found no difference in morbidity or mortality between the strategies for chronic hypertensive administration and MAP goals.  

Criticisms of the study include lack of analysis based on renal function given TXA is primarily renally excreted. Further, the study encountered financial challenges during the COVID-19 pandemic, prompting them to revise their power calculation and end the study early.  

POISE-3 demonstrated that perioperative administration of TXA in non-cardiac surgery patients reduced the risk of adverse bleeding events by almost 25%, (-2.6% absolute difference, number needed to treat of 38 patients). This is complemented by the results of previous large cohort study, the Aspirin and Tranexamic Acid for Coronary Artery Surgery (ATACAS), that found TXA was associated with a reduction in the composite outcome of death and thrombotic complications, reoperation, and transfusion. However, these benefits must be weighed against POISE-3’s inability to prove non-inferiority for the primary safety outcome. Although this may have been influenced study design facets such as the chosen non-inferiority margin, it remains a concern especially for patients at risk for major cardiovascular events. The authors recognize this limitation and point to a bleeding-risk calculator that could be used to identify the patients who stand to benefit from routine administration TXA. Future work is needed to isolate those in whom the benefits might not outweigh the risks, as well as the potential impact on costs.

Together, POISE-3 supports a more widespread adoption of TXA across non-cardiac surgery, a practice that could lead to improved patient outcomes and reduced transfusions. However, this benefit may be limited by a low probability increased risk of composite cardiovascular outcome events found in the study – suggesting patient selection may the limiting factor rather that surgical specialty when it comes to the routine administration of TXA.