Diagnostic accuracy
Can needle biopsies replace a full autopsy to find cause of death?
Open access · cc by · source: Europe PMC
A quick needle-based autopsy agreed with a full autopsy on cause of death in about three out of four adults, and did best for infections and cancers.
Study at a glance
- Design
- Cross-sectional — Diagnostic accuracy study: paired minimally invasive and complete autopsies on the same bodies, read blind by the same expert panel after a washout
- N
- N=112 · 112 adult in-hospital deaths with paired MIA and complete autopsy
- Population
- Adults (over 15, excluding maternal deaths) who died at Maputo Central Hospital, Mozambique; about two-thirds HIV-positive
- Outcome
- Concordance of MIA cause-of-death diagnosis with complete diagnostic autopsy (ICD-10 codes, kappa, sensitivity/specificity)
Structured fields used in claim comparison tables when every cited study has a complete layer.
Key findings
The MIA reached a diagnosis in 89.2% of cases and agreed with the full autopsy in 75.9% (kappa 0.732, 'substantial' agreement). Concordance was 78.8% for infectious deaths and 81.3% for tumours but only 56.2% for other diseases such as cardiovascular and kidney conditions. Infections caused 71.4% of deaths, tuberculosis being the leading cause, and the MIA identified the same microorganism as the full autopsy in 83.8% of infectious deaths with a known agent.
Methodology
At a large hospital in Maputo, Mozambique, researchers performed a minimally invasive autopsy (MIA) using biopsy needles to sample blood, spinal fluid and key organs, followed within an hour by a complete diagnostic autopsy (CDA) by a different pathologist, in 112 adults. An expert panel assigned the MIA cause of death without any clinical information, then after a washout of at least three months assigned the CDA cause using the full autopsy and clinical records. Agreement was judged by matching ICD-10 codes and a kappa statistic.
Limitations
All deaths were in a single referral hospital, so causes may not reflect deaths at home or in rural communities, and bodies were sampled within 24 hours, which may not be possible in the field. More than half the patients had HIV with disseminated infections, which likely made needle sampling easier; focal disease and non-infectious conditions were missed more often. The same experts read both autopsies, which supports internal consistency but not generalisability to other pathologists. Children and maternal deaths were not included.
How this study connects
Role on claims
Each row is a claim on a concept or method page where this paper supports, challenges, or qualifies the statement. Roles are hand-checked — not a model guess.
Not yet placed on a claim. This paper has study layers, but no concept page yet cites it as support, challenge, or qualifier.
Related papers in this topic
Same topic cluster — not a recommendation engine.