30-Second Takeaway
- In East African adults with HIV and sepsis, **M. tuberculosis** is the leading and often occult pathogen, outpacing Gram-negatives.
- Most streptococcal bacteremias rarely require follow-up blood cultures unless complicated infection or specific species suggest persistence.
- High-dose meropenem in ICU patients correlated with lower 90-day mortality without apparent renal penalty in observational data.
Week ending March 7, 2026
Concise updates in HIV sepsis, diagnostics, dosing optimization, and viral prevention
Disseminated TB is the dominant cause of HIV-associated sepsis in East Africa
Among 437 adults with HIV and sepsis in Tanzania and Uganda, M. tuberculosis was detected in 52% of patients. TB caused half of all bloodstream infections, emphasizing disseminated TB as a central driver of sepsis in this setting. Urine LF-LAM plus sputum GeneXpert still missed nearly one-third of TB bloodstream infections, revealing substantial diagnostic gaps. Klebsiella and E. coli were the most frequent non-mycobacterial pathogens, with ceftriaxone resistance present in most tested isolates. Longer pre-illness duration, prolonged cough, younger age, and low CD4 counts best predicted TB etiology in the random forest model. These findings support early empiric antituberculous therapy and reconsideration of ceftriaxone-based monotherapy for Gram-negative coverage.
Follow-up blood cultures seldom add value in uncomplicated streptococcal bacteremia
In this retrospective cohort of 454 patients with streptococcal bacteremia, only 10.6% had positive follow-up blood cultures (FUBCs). Positivity ranged from 0% for Streptococcus pyogenes to 18.5% for Streptococcus anginosus, indicating species-specific persistence risk. Endocarditis and osteomyelitis were more common when FUBCs were positive, but mortality and length of stay were unchanged. Positive FUBCs correlated with more relapses and longer antibiotic courses, reflecting underlying complexity rather than benefit from culturing. Foregoing FUBCs appears reasonable for several streptococcal species when endocarditis or osteomyelitis are not clinically suspected.
Single-dose intrapartum azithromycin reduces diverse maternal and infant infections
In Fiji, 2110 birthing parents were randomized to 2 g oral azithromycin or placebo during labour. Infant infections by 3 months were lower with azithromycin (13.6%) than placebo (17.3%; RR 0.79, 95% CI 0.63-0.99). Maternal infections, including SSTIs, were markedly reduced at 1 week postpartum, with benefit waning by 6 months. Antibiotic prescription rates were unchanged, suggesting direct prophylactic benefit rather than altered downstream management. These data support intrapartum azithromycin to prevent multiple infections, while underscoring the need to evaluate resistance and microbiome effects.
High-dose intermittent meropenem associated with lower ICU 90-day mortality
This Austrian observational cohort matched 1144 ICU patients receiving 6 g/day meropenem to 572 receiving 3 g/day. High-dose meropenem was associated with lower adjusted 90-day mortality (31.5% vs 40.9%; adjusted risk difference 9.4%, 95% CI 4.8-14.1). Thirty-day mortality, resistance emergence, ECMO initiation, ARDS, and ICU or hospital length of stay were similar between regimens. High-dose treatment was paradoxically associated with a lower adjusted risk of acute kidney injury (60.6% vs 68.2%). Despite confounding concerns, findings support considering higher doses when aiming for stringent PK/PD targets in severe infections.
References
Numbered in order of appearance. Click any reference to view details.
Additional Reads
Optional additional studies from this edition.