The global burden of sepsis is roughly double that of conventional estimates, amounting to an estimated 48.7 million cases and 22 million deaths worldwide in 2017, or about 20% of all deaths (Lancet 2020;395:200-11). Sepsis also disproportionately affects low- and middle-income countries, where 85% of incident cases occurred in 2017. Even though the estimated incidence of sepsis fell by 37% and mortality decreased by 52.8% between 1990 and 2017, sepsis incidence and deaths varied significantly across regions, most impacting Southeast Asia, sub-Saharan Africa, Oceania, and other parts of Asia.

Women are more likely to experience sepsis than men, with an estimated age-standardized incidence in 2017 of 716.5 cases per 100,000 versus 642.8 cases per 100,000 respectively. Children also bear the brunt of sepsis, with more than 40% of all cases occurring in those younger than age 5. The leading underlying cause of sepsis-related death in every year from 1990 to 2017 was lower respiratory infection. Diarrheal diseases and neonatal disorders were the other two most common underlying causes of sepsis-related death in 2017.

Researchers developed these updated sepsis figures by accessing data from the Global Burden of Disease, Injuries, and Risk Factors Study 2017, which contains more than 1 billion datapoints for 282 underlying causes of death in 195 countries. Prior estimates considered only hospital administrative databases (thereby missing individuals never admitted to hospitals) and national or subnational locations from only selected middle- or high-income countries.

The findings “highlight the need for greater prevention and treatment of sepsis,” according to the investigators.

One-quarter of Those With Atherosclerotic Cardiovascular Disease and Elevated Triglycerides Might Benefit From Emerging Therapies

Up to 25% of individuals with atherosclerotic cardiovascular disease (ASCVD) and low-density lipoprotein cholesterol (LDL-C) controlled by statin medications but who still have elevated triglyceride (TG) levels might be eligible for and benefit from emerging therapies such as ethyl eicosapentanoic acid (EPA) (Eur Heart J 2020;41:86-94).

This finding comes from CANHEART, an observational cohort study of adults in Ontario, Canada, based on a dataset that links 17 different individual-level data sources. Of 2.4 million people in CANHEART with recorded lipid panel results, the researchers identified 196,717 who had established ASCVD. The investigators sought to assess the prevalence of hypertriglyceridemia and ASCVD events in this population, estimate the percentage who might qualify for emerging therapies, and see how well this population’s experience tracked with the REDUCE-IT trial. REDUCE-IT found that subjects with ASCVD and plasma TG ≥135-499 mg/dL who were on statin therapy and daily EPA therapy had 25% lower risk of major ASCVD outcomes compared with those who took placebo.

CANHEART researchers observed 24,097 composite ASCVD events over a mean of 2.9 years’ follow-up. They found an increasing adjusted hazard ratio for composite ASCVD based on TG level, ranging from 1.0 for TG <89 mg/dL to 1.52 for TG ≥354 mg/dL.

While these findings suggest the potential of EPA therapy to lower risk of ASCVD events, an accompanying editorial cautions that the estimate of 25% of eligible individuals deriving benefit from this approach is optimistic given the “broad definition” of hypertriglyceridemia the investigators used (Eur Heart J 2020;41:95-8). The more usual threshold, >200 mg/dL, “would identify substantially fewer candidates for this intervention.” The editorialist also suggested that because TG affects the “whole spectrum” of apolipoprotein B-containing lipoproteins, “we should not expect interpretation of the relationship between lipid reduction and decreased ASCVD risk to be as straightforward as it has been for [LDL-C].”

4 Blood Biomarkers Show Potential in Identifying Student Athletes With Sport-related Concussion

A quartet of candidate blood biomarkers has a “modest ability” in the acute post-injury period to differentiate collegiate athletes with sport-related concussion (SRC) from control group student athletes (JAMA Netw Open 2020;3:e1919771). This finding supports the use of these analytes as research tools to better understand the underlying mechanisms of concussion and eventually, if optimized and validated, as clinical measurements, according to the researchers.

This multicenter, prospective, case-control study of 264 student athletes with concussion, 138 contact sport control athletes, and 102 non-contact sport control athletes was conducted by the National Collegiate Athletic Association and the U.S. Department of Defense Concussion Assessment, Research, and Education Consortium. Participants had preseason baseline measurements taken, and in the acute post-injury period, 24-48 hours after injury, at the point of being asymptomatic, and 7 days after return to play. The researchers matched athletes with concussion against non-concussed contact sport athletes based on sports-related and concussion history criteria.

Athletes with concussion versus controls had, at the point of acute post-injury in comparison with preseason baseline levels, significantly elevated concentrations of glial fibrillary acidic protein (GFAP), ubiquitin C-terminal hydrolase-L1 (UCH-L1), and total tau but not neurofilament light chain (NF-L). GFAP levels remained elevated in comparison with baseline levels at all time points. The area under the receiver operating characteristic for GFAP, UCH-L1, NF-L, and total tau in differentiating athletes with concussion from sports-playing controls in the post-injury period was 0.72.

Athletes with concussion who experienced loss of consciousness or post-traumatic amnesia had significantly higher levels of GFAP than their concussed peers who did not lose consciousness or have amnesia.

The authors of a related commentary suggested that the authors would have advanced the understanding of these analytes’ clinical relevancy in sports concussion if they had reported absolute and relative changes in the biomarkers between baseline and the acute post-injury period (JAMA Netw Open 2020;3:e1919799). They also noted that the median time for acquiring acute phase samples was 3.42 hours after injury, generally too late to have much diagnostic importance.