Three papers. Three practical lessons.
1. Faster troponin can shorten ED stays
ICare-FASTER · Circulation · 19 October 2026 · Full paper reviewed; supplement unavailable
Study: Six New Zealand EDs introduced bedside high-sensitivity troponin I within existing chest-pain pathways; the main comparison included 52,601 presentations.
Finding: Adjusted ED stay fell by 13%, equivalent to 47 minutes (95% CI 33–61). Thirty-day MI/cardiac death after discharge was 0.39% in both groups.
Main caution: Similar event rates do not prove equivalent safety. Also, 8.6% of bedside tests were invalid; backup laboratory testing remained necessary.
Bedside takeaway: Consider the device within a validated pathway when laboratory delay limits decisions. A negative troponin alone is not a discharge decision.
2. NT-proBNP: rule-out and rule-in are different tasks
European Journal of Heart Failure · 17 October 2026 · Abstract-based brief
Study: Meta-analysis of 31 studies involving 26,948 breathless ED patients.
Finding: The 300 ng/L rule-out threshold had 97.9% sensitivity and 97.0% negative predictive value. Age-dependent rule-in thresholds had 82.0% specificity; this fell to 62.6% with CKD.
Main caution: Predictive values depend on the population. Evidence in CKD and obesity was limited; the obesity analysis did not validate the low rule-out threshold.
Bedside takeaway: A low result reduces probability; a high result alone does not confirm heart failure. Interpret it with clinical probability, renal function, and imaging.
3. “Mild” GCS does not make a subdural bleed low risk
Journal of Trauma and Acute Care Surgery · 16 October 2026 · Abstract-based brief
Study: Retrospective, 10-center cohort of 1,157 adults with GCS 13–15 and acute subdural hematoma on CT.
Finding: Death or hospice discharge occurred in 10% of patients aged ≥65 versus 3% of younger adults. Neurosurgical intervention occurred in 11% versus 8%.
Main caution: The 10% figure is not an in-hospital death rate. These were CT-positive patients; 58% were transferred in. Findings do not apply to every minor head injury.
Bedside takeaway: Do not let a reassuring GCS override a confirmed bleed. Consider the examination, CT findings, specialist input, and patient-centered goals of care.
Evidence limits and disclosures
ICare-FASTER: Five rollout sites were randomized; one sentinel site was not. The 47-minute estimate translates the adjusted relative effect into minutes. The reported safety difference was 0.01 percentage points (95% CI −0.14 to +0.15). MI follow-up used coding; deaths were adjudicated. Siemens supplied devices/materials and funded installation/support personnel; authors reported no company role in protocol, data collection, analysis, or writing. Several authors disclosed industry relationships. The supplied full paper was marked “First Proof Only”; clinical cutoffs are omitted because text and figure thresholds differed.
Papers 2 and 3: Full methods, tables, supplements, and funding were unavailable; these are abstract-based briefs, not full critical appraisals. They do not establish new biomarker cutoffs, a repeat-CT schedule, or treatment protocols. The trauma study’s associations do not prove causation; an advance directive is not evidence of biological futility.
ICare-FASTER protocol · NT-proBNP abstract · Subdural hematoma abstract
