Background: Whether psychosis in people living with HIV reflects a coexisting primary disorder or the effects of HIV, its complications, or treatment remains a central clinical uncertainty. We mapped evidence on frequency, presentation, secondary causes, and treatment, distinguishing psychosis outcomes from contextual evidence. Methods: We conducted a systematic scoping review using three structured Scopus searches and a supplementary PubMed/MEDLINE record set. Google Scholar and preprint servers (medRxiv, bioRxiv, PsyArXiv, and Research Square) supported discovery and citation checking. Findings were synthesised narratively. Risk of bias and GRADE-informed certainty were assessed for principal quantitative questions; case reports were not graded independently. Results: We mapped 128 reports, including 37 identified through a reproducible PubMed sensitivity search. Of these, 108 addressed psychosis-related questions and 20 provided contextual evidence on cerebrospinal fluid escape or antiretroviral safety. Fifty-one reports predated 2010, including 22 published before 1997. Nine population-level reports represented seven independent data sources; most indicated a higher recorded burden of psychotic disorders in people with HIV, although outcomes and comparators differed. We identified no study estimating new-onset psychosis during documented, durable plasma suppression. In selected high-prevalence first-presentation settings, 21-40% of patients with psychosis had HIV. A Johannesburg emergency cohort recorded medical aetiologies in 84.2% of HIV-positive versus 35.4% of HIV-negative presentations, but largely represented untreated or advanced HIV. Treatment evidence comprised one small randomised trial and uncontrolled series. Certainty was low for the population-level association and very low for other directly assessed questions. Conclusions: Population-level evidence links HIV with psychosis but cannot establish causation in individual patients. The evidence gap during durable plasma suppression limits its applicability to contemporary care; management in this setting was not directly tested. These findings suggest that HIV status, antiretroviral exposure, or an undetectable viral load should not alone determine attribution or preclude clinically indicated targeted investigation for secondary causes.
Distinguishing primary from secondary psychosis in people living with HIV: A systematic scoping review
Messina, Antonino
;Ceccarelli, Manuela;Maggi, Paolo
2026-01-01
Abstract
Background: Whether psychosis in people living with HIV reflects a coexisting primary disorder or the effects of HIV, its complications, or treatment remains a central clinical uncertainty. We mapped evidence on frequency, presentation, secondary causes, and treatment, distinguishing psychosis outcomes from contextual evidence. Methods: We conducted a systematic scoping review using three structured Scopus searches and a supplementary PubMed/MEDLINE record set. Google Scholar and preprint servers (medRxiv, bioRxiv, PsyArXiv, and Research Square) supported discovery and citation checking. Findings were synthesised narratively. Risk of bias and GRADE-informed certainty were assessed for principal quantitative questions; case reports were not graded independently. Results: We mapped 128 reports, including 37 identified through a reproducible PubMed sensitivity search. Of these, 108 addressed psychosis-related questions and 20 provided contextual evidence on cerebrospinal fluid escape or antiretroviral safety. Fifty-one reports predated 2010, including 22 published before 1997. Nine population-level reports represented seven independent data sources; most indicated a higher recorded burden of psychotic disorders in people with HIV, although outcomes and comparators differed. We identified no study estimating new-onset psychosis during documented, durable plasma suppression. In selected high-prevalence first-presentation settings, 21-40% of patients with psychosis had HIV. A Johannesburg emergency cohort recorded medical aetiologies in 84.2% of HIV-positive versus 35.4% of HIV-negative presentations, but largely represented untreated or advanced HIV. Treatment evidence comprised one small randomised trial and uncontrolled series. Certainty was low for the population-level association and very low for other directly assessed questions. Conclusions: Population-level evidence links HIV with psychosis but cannot establish causation in individual patients. The evidence gap during durable plasma suppression limits its applicability to contemporary care; management in this setting was not directly tested. These findings suggest that HIV status, antiretroviral exposure, or an undetectable viral load should not alone determine attribution or preclude clinically indicated targeted investigation for secondary causes.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


