A recent Acute Medicine digest,
as sent to subscribers

Study Type: Retrospective population-based cohort study

Key Question: What is the plausible range of chronic traumatic encephalopathy (CTE) prevalence at death among former NFL players, and how does CTE severity relate to dementia risk?

Key Findings:

• Among 1712 deceased NFL players, 338 (19.7%) had brain donation/autopsy; 315 (93.2%) of these had pathological CTE, yielding an estimated population prevalence range of 18.5–98.7% (24.5–97.7% for 2016–21, the highest-donation period).
• Stage IV (severe) CTE was associated with clinician-adjudicated dementia (risk ratio 1.44, 95% CI 1.16–1.78; p<0.001); dementia onset averaged 63.4 years.
• Only 40.6% of donors with diagnosed dementia had neurodegenerative disease recorded as a cause of death, indicating substantial underrepresentation in death certification.

Clinical Relevance: Highlights that neurodegenerative and dementia burden linked to repetitive head impact exposure is likely underestimated by routine mortality data, relevant to UK clinicians managing contact-sport athletes and counselling on long-term neurological risk.

Limitations: Selection bias from voluntary brain donation limits precise prevalence estimation despite bias-adjustment methods.

Study Type: Narrative review

Key Question: What is the current state of understanding and management of schizophrenia, and how might precision medicine advance future care?

Key Findings:

• Schizophrenia arises from combined genetic and environmental factors with neurodevelopmental origins; onset typically in late adolescence/young adulthood.
• Current management combines antipsychotic drugs (targeting positive symptoms, relapse prevention) with psychosocial/cognitive interventions for functional recovery; neuromodulation and digital health tools remain investigational.
• Progress is limited by clinical (rather than biological) diagnostic criteria, causing biological heterogeneity; identifying objective biomarkers is proposed as the key step towards stratified, mechanism-based treatment.

Clinical Relevance: Highlights the diagnostic and therapeutic limitations UK clinicians face in acute psychiatric presentations of schizophrenia, and signals a future shift towards biomarker-guided, personalised treatment strategies relevant to NHS mental health and acute liaison services.

Limitations: As a narrative review, it lacks systematic methodology and primary data, offering expert synthesis rather than quantifiable evidence.

Study Type: Systematic review with cross-disease meta-analysis and translational modelling of randomised, placebo-controlled trials.

Key Question: Do Type-2-targeted biologics reduce oral corticosteroid burden and exacerbations in severe asthma versus COPD, and what downstream harm might this prevent?

Key Findings:

• Biologics nearly tripled the odds of achieving ≥50% maintenance OCS reduction in severe asthma (OR 2.99, 95% CI 2.11–4.25; 4 trials, n=615).
• Exacerbation reduction was significantly greater in asthma (RR 0.38) than COPD (RR 0.78; p=0.0001).
• Applying a 42% cumulative OCS reduction (BOREAS/NOTUS COPD subgroup) to pharmacoepidemiologic dose-response data suggested lowest NNTs for preventing serious infection, pneumonia, and osteoporotic fracture in COPD.

Clinical Relevance: Supports biologic use as an OCS-sparing strategy in severe asthma and highlights a plausible, quantifiable harm-reduction rationale for expanding biologic access in exacerbation-prone COPD within NHS respiratory pathways.

Limitations: The translational NNT estimates rely on indirect extrapolation from observational cohorts rather than direct trial-level harm outcomes, limiting causal certainty.

Study Type: National retrospective cohort study (administrative hospital data)

Key Question: What is long-term mortality after first emergency admission for chronic liver disease (CLD) in England, and what factors drive variation?

Key Findings:

• Of 82,402 CLD patients, 59.9% died within 5 years; median survival was 2.72 years (95%CI 2.66–2.79).
• Mortality was lower for metabolic (adjHR 0.76), combined metabolic/alcohol (adjHR 0.80), and autoimmune aetiologies (adjHR 0.82) versus alcohol-related disease; encephalopathy (adjHR 1.70), ascites (adjHR 1.64), and GI bleeding (adjHR 1.10) predicted higher mortality.
• All eight regions outside London had higher adjusted mortality (adjHR 1.06–1.19), suggesting geographic inequity.

Clinical Relevance: Highlights exceptionally poor prognosis after emergency CLD admissions in England and identifies high-risk subgroups and regional disparities relevant to NHS service planning and resource allocation.

Limitations: Retrospective administrative data lack granular clinical detail (e.g., liver function scores, treatment received), limiting causal inference regarding regional and aetiological differences.

Study Type: Narrative review (ethical and legal analysis)

Key Question: What ethical and legal challenges arise from integrating AI/ML into medical practice, and how should these be addressed by clinicians, regulators, and developers?

Key Findings:

• AI adoption raises core ethical tensions around autonomy, beneficence, non-maleficence, and justice, particularly regarding algorithmic bias and equitable access.
• Legal frameworks (liability, data protection, medical device regulation) are still adapting to AI-specific issues such as informed consent for automated decision-making and transparency of "black-box" algorithms.
• Regulatory approaches vary across jurisdictions, with existing frameworks being retrofitted rather than purpose-built for AI in healthcare.

Clinical Relevance: As NHS trusts increasingly deploy AI tools for diagnostics and decision support, clinicians must understand evolving liability, consent, and governance implications to ensure safe, equitable, and legally defensible use.

Limitations: As a narrative review, it lacks systematic methodology and empirical data, offering conceptual synthesis rather than quantifiable evidence.

Study Type: Registry-based retrospective cohort study (RIETE registry)

Key Question: Do DOACs outperform VKAs for recurrent VTE and bleeding outcomes in real-world DVT patients, and does this benefit hold across trial-eligible versus trial-ineligible populations?

Key Findings:

• Among 24,728 patients, DOACs were associated with lower recurrent VTE (HR 0.71, 95% CI 0.58–0.87) and major bleeding (HR 0.78, 95% CI 0.63–0.97), but higher CRNMB (HR 1.35, 95% CI 1.17–1.55) versus VKAs.
• Median time to recurrence/major bleeding/CRNMB was 105/88/81 days respectively.
• Efficacy and safety advantages of DOACs were significant only in patients resembling pivotal RCT populations; no benefit was seen in trial-ineligible patients.

Clinical Relevance: Supports current UK practice favouring DOACs as first-line VTE treatment but highlights that benefits may not extend to complex, multimorbid patients typically excluded from RCTs and common in NHS acute medicine settings.

Limitations: Observational design with potential residual confounding despite inverse probability weighting, limiting causal inference.

Study Type: Prospective cohort study

Key Question: What clinical and biological factors predict long-term thrombotic cardiovascular events in women with SLE over a 10-year follow-up?

Key Findings:

• 44% had carotid plaque at baseline; 11.3% experienced thrombotic events (8 arterial) over ~9.7 years.
• Metabolic syndrome independently predicted events (HR 4.3, 95% CI 1.2–15.2), alongside age, higher carotid IMT, cumulative organ damage, and reduced renal function.
• No independent prognostic value was found for inflammatory or adipokine biomarkers; arterial event rates were numerically but not significantly higher than age-matched controls.

Clinical Relevance: Reinforces that in SLE patients seen acutely, traditional cardiovascular risk factors (age, metabolic syndrome, renal impairment) rather than lupus-specific inflammatory markers should guide cardiovascular risk stratification and preventive management.

Limitations: Single-centre study with modest sample size (n=97) and only 11 events, limiting statistical power for multivariable analysis and biomarker assessment.

Study Type: Retrospective cohort analysis (pooled data from three global AF registries)

Key Question: In AF patients at high combined risk of stroke and bleeding, do distinct clinical phenotypes exist, and does adherence to the ABC (Atrial fibrillation Better Care) pathway improve one-year outcomes?

Key Findings:

• Hierarchical clustering identified three phenotypes: Cluster I (prior thromboembolic/bleeding events, low comorbidity), Cluster II (oldest, highest dyslipidaemia/heart failure/PAD burden), Cluster III (youngest, highest BMI/alcohol use).
• Compared with Cluster I, Clusters II and III had significantly higher all-cause mortality (aOR 1.93, 95% CI 1.37–2.78; and aOR 1.65, 95% CI 1.10–2.50, respectively).
• ABC pathway adherence was associated with markedly lower 1-year MACE (OR 0.39, 95% CI 0.15–0.82).

Clinical Relevance: Supports UK acute medicine practice of embedding structured ABC pathway management (anticoagulation, symptom control, comorbidity/lifestyle optimisation) even in complex, high-bleeding-risk AF patients, while highlighting the need for phenotype-tailored risk stratification.

Limitations: Observational design with incomplete ABC adherence data limits causal inference and generalisability.