Background Blood culture time to positivity (TTP) reflects bacterial inoculum and may serve as an early prognostic marker in bacteremia. However, heterogeneous study designs and variable TTP cutpoints have precluded definitive conclusions regarding its clinical utility. Methods We conducted a systematic review and Bayesian meta-analysis following PRISMA 2020 and MOOSE guidelines. PubMed, Embase, CINAHL, Web of Science and the Cochrane Library were searched for studies examining the association between TTP and clinical outcomes in bacteremia. Effect sizes were standardised to log-odds ratios and pooled using Bayesian and frequentist random-effects models. Meta-regression explored sources of heterogeneity. Publication bias was assessed using funnel plots, Egger's test, trim-and-fill, and Bayesian selection models. Results Thirty-seven studies (33 mortality effect sizes, 7 persistent bacteremia effect sizes) were included. Short TTP was associated with significantly increased mortality (Bayesian OR 2.23, 95% CrI: 1.68–3.01; P(OR>1) = 100%), corroborated by the frequentist estimate (OR 2,29 ;95% CI: 1,29–2,74; P<0.001). The association was pathogen-agnostic, with overlapping estimates for Gram-positive (OR 2.17) and Gram-negative (OR 2.06) infections. TTP cutpoint was the only significant moderator (p < 0.001), explaining 27.3% of heterogeneity, with shorter cutpoints yielding larger effect sizes. Persistent bacteremia showed a suggestive but inconclusive association (OR 1.95, 95% CrI: 0.53–6.24). Results were robust across all sensitivity analyses, prior specifications, and publication bias adjustments. Conclusions Short blood culture TTP is a robust, pathogen-agnostic prognostic marker associated with approximately twice the odds of mortality. Standardised cutpoints and prospective validation studies are needed to support clinical implementation.
Introduzione Il time to positivity (TTP) delle emocolture riflette l’inoculo batterico e può fungere da marcatore prognostico precoce nelle batteriemie. Tuttavia, l’eterogeneità dei disegni di studio e la variabilità dei cutpoint di TTP hanno finora impedito conclusioni definitive sulla sua utilità clinica. Metodi È stata condotta una revisione sistematica con meta analisi bayesiana secondo le linee guida PRISMA 2020 e MOOSE. Sono stati consultati PubMed, Embase, CINAHL, Web of Science, e Cochrane Library per identificare studi che valutassero l’associazione tra TTP ed esiti clinici nelle batteriemie. Le misure di effetto sono state standardizzate in log odds ratio e combinate mediante modelli ad effetti casuali bayesiani e frequentisti. Meta regressioni hanno esplorato le fonti di eterogeneità. Il bias di pubblicazione è stato valutato tramite funnel plot, test di Egger, metodo trim and fill e modelli di selezione bayesiani. Risultati Sono stati inclusi 37 studi (33 stime di effetto per la mortalità, 7 per la batteriemia persistente). Un TTP breve è risultato associato a un aumento significativo della mortalità (OR bayesiano 2,23; ICr al 95%: 1,68–3,01; P(OR>1) = 100%), confermato dalla stima frequentista (OR 2,29 ;95% CI: 1,29–2,74; P<0.001). L’associazione è risultata indipendente dal tipo di patogeno, con stime sovrapponibili per infezioni da Gram-positivi (OR 2,17) e Gram-negativi (OR 2,06). Il cutpoint del TTP è emerso come unico moderatore significativo (p < 0,001), spiegando il 27,3% dell’eterogeneità: cutpoint più brevi erano associati a effetti più marcati. Per la batteriemia persistente è emersa un’associazione suggestiva ma non conclusiva (OR 1,95; ICr al 95%: 0,53–6,24). I risultati si sono dimostrati robusti in tutte le analisi di sensibilità, nelle diverse specificazioni di prior e nei metodi di correzione del bias di pubblicazione. Conclusioni Un TTP breve nelle emocolture rappresenta un marcatore prognostico robusto e indipendente dal patogeno, associato a un raddoppio delle probabilità di mortalità. Sono necessari cutpoint standardizzati e studi prospettici di validazione per supportarne l’implementazione clinica.
Il Time To Positivity (TTP) delle Emocolture come Predittore di Fallimento della Terapia Antibiotica: una Meta-Analisi Bayesiana
PANGALLO, FRANCESCO
2025/2026
Abstract
Background Blood culture time to positivity (TTP) reflects bacterial inoculum and may serve as an early prognostic marker in bacteremia. However, heterogeneous study designs and variable TTP cutpoints have precluded definitive conclusions regarding its clinical utility. Methods We conducted a systematic review and Bayesian meta-analysis following PRISMA 2020 and MOOSE guidelines. PubMed, Embase, CINAHL, Web of Science and the Cochrane Library were searched for studies examining the association between TTP and clinical outcomes in bacteremia. Effect sizes were standardised to log-odds ratios and pooled using Bayesian and frequentist random-effects models. Meta-regression explored sources of heterogeneity. Publication bias was assessed using funnel plots, Egger's test, trim-and-fill, and Bayesian selection models. Results Thirty-seven studies (33 mortality effect sizes, 7 persistent bacteremia effect sizes) were included. Short TTP was associated with significantly increased mortality (Bayesian OR 2.23, 95% CrI: 1.68–3.01; P(OR>1) = 100%), corroborated by the frequentist estimate (OR 2,29 ;95% CI: 1,29–2,74; P<0.001). The association was pathogen-agnostic, with overlapping estimates for Gram-positive (OR 2.17) and Gram-negative (OR 2.06) infections. TTP cutpoint was the only significant moderator (p < 0.001), explaining 27.3% of heterogeneity, with shorter cutpoints yielding larger effect sizes. Persistent bacteremia showed a suggestive but inconclusive association (OR 1.95, 95% CrI: 0.53–6.24). Results were robust across all sensitivity analyses, prior specifications, and publication bias adjustments. Conclusions Short blood culture TTP is a robust, pathogen-agnostic prognostic marker associated with approximately twice the odds of mortality. Standardised cutpoints and prospective validation studies are needed to support clinical implementation.| File | Dimensione | Formato | |
|---|---|---|---|
|
Pangallo_Francesco.pdf
accesso aperto
Dimensione
2.2 MB
Formato
Adobe PDF
|
2.2 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/105333