This Week in Infectious Disease — Aug 8, 2026

Welcome to This Week in Infectious Disease. This week we're covering 10 notable papers spanning antimicrobial resistance — from new society guidance through to shifting molecular epidemiology and stewardship — plus global infectious disease trials in tuberculosis and malaria, and finally prevention and care delivery, including respiratory syncytial virus immunoprophylaxis and longitudinal human immunodeficiency virus care. Let's dive in.

We start with resistant gram-negative infections, where the anchor publication of the week is the Infectious Diseases Society of America 2026 Guidance on the treatment of antimicrobial-resistant gram-negative infections, published in Clinical Infectious Diseases [1]. This update replaces earlier versions and again covers six problem groups: extended-spectrum beta-lactamase-producing Enterobacterales, AmpC beta-lactamase producers, carbapenem-resistant Enterobacterales, Pseudomonas aeruginosa with difficult-to-treat resistance, carbapenem-resistant Acinetobacter baumannii, and Stenotrophomonas maltophilia. A panel of six specialists sets out preferred and alternative regimens with rationales tied to pathogen identification and in vitro susceptibility, and briefly addresses empiric therapy, transition to oral agents, and treatment duration. Two practical caveats matter: the document is deliberately United States-focused because resistance epidemiology and drug availability vary geographically, and dosing guidance is given only for adults even though the suggestions apply to children as well. It is current as of the first of March 2026 and will be revised periodically, so the web version — not a printed copy — should be your reference point.

Two papers add nuance beneath that guidance. In the Journal of Antimicrobial Chemotherapy, a single-centre retrospective series of 60 patients with bacteraemia due to non-carbapenemase-producing carbapenem-resistant Enterobacterales asked which of these isolates are also meropenem non-susceptible [5]. Patients with meropenem non-susceptible isolates were far more likely to have had prior meropenem exposure — about sixty percent versus thirteen percent — and more likely to have breakthrough bacteraemia. Treatment-emergent resistance arose in roughly one in six patients treated with meropenem and in none of the fifteen treated with ceftazidime-avibactam, though with these numbers this is exploratory and hypothesis-generating rather than practice-defining. Complementing that, the Journal of Infectious Diseases reports whole-genome sequencing of 881 carbapenem-resistant Klebsiella pneumoniae isolates from the same California long-term acute care hospital network, comparing 2014 to 2015 with 2021 to 2023 [6]. A near-monoclonal picture dominated by sequence type 258 — about 94 percent of isolates in the earlier period — gave way to a much more diverse population, with sequence type 258 down to roughly 42 percent and epidemic lineages including sequence types 14, 307, and 17 emerging. Non-carbapenemase-containing isolates rose from under three percent to nearly 13 percent, and critically, ceftazidime-avibactam resistance rose from about one percent to nine percent while colistin, fosfomycin, and meropenem-vaborbactam resistance all fell. The practical message is that your local carbapenem-resistant Klebsiella population may look nothing like it did a decade ago, and empiric assumptions about newer beta-lactam-beta-lactamase inhibitor activity need refreshing against current local data.

Stewardship supplies the next theme. In the International Journal of Antimicrobial Agents, a single-institution Japanese cohort of 1,619 patients receiving piperacillin-tazobactam found that just under six percent developed a new piperacillin-tazobactam-resistant gram-negative rod within 180 days, rising to 12 percent among those treated for 14 days or longer [8]. After accounting for death as a competing risk, exposure of 14 days or more independently increased the hazard of resistance emergence by roughly three-quarters. Most emergent isolates were Pseudomonas aeruginosa or AmpC-producing Enterobacterales — precisely the organisms where prolonged broad-spectrum beta-lactam pressure is most consequential — which is a concrete argument for hard-stopping unnecessarily long courses. From the same journal, a Chinese single-centre study of 423 intensive care unit patients with Acinetobacter baumannii infection developed and temporally validated models to predict carbapenem resistance before susceptibility results return [9]. A logistic regression nomogram achieved an area under the curve of about 0.83 and a gradient-boosting model about 0.88 in the temporal test cohort, with prior carbapenem exposure, Glasgow Coma Scale, duration of mechanical ventilation, and Acute Physiology and Chronic Health Evaluation II score doing most of the predictive work. The authors are appropriately cautious: this needs prospective multicentre external validation before it guides empiric therapy, but the predictors themselves are a useful bedside heuristic today.

Turning to global infectious disease trials, the BMJ reports a pragmatic randomised trial of a patient-centred digital adherence technology for tuberculosis in four public hospitals in Buenos Aires [2]. Among 555 adults and adolescents with newly diagnosed drug-susceptible tuberculosis, the app — which combined daily adherence reporting, two-way messaging with treatment supporters, educational content, and a weekly urine isoniazid test — raised treatment success from 74 percent to 82 percent, an absolute gain of about seven percentage points, with the confidence interval only just clearing no effect. Loss to follow-up fell from 24 percent to 17 percent, roughly a thirty percent relative reduction, with larger benefits in women and those under 35. This is a modest but real effect from a scalable intervention in a programme setting. On the pharmacology side, Antimicrobial Agents and Chemotherapy reports population pharmacokinetic modelling in 78 adults with human immunodeficiency virus-associated tuberculosis, nested in a trial of adjunctive metformin as host-directed therapy [7]. Metformin reduced isoniazid bioavailability by about 16 percent and rifampicin bioavailability by about 24 percent, cutting rifampicin exposure over 24 hours from roughly 36 to 27 milligram-hours per litre, with no significant effect on pyrazinamide. The authors suggest simulations indicate an extra 150 milligrams of rifampicin could offset this — but the immediate point for practice is that metformin is not pharmacokinetically inert in tuberculosis treatment, and that matters for the many patients with diabetes on both drug classes.

Also in the global health space, JAMA published the IRMA trial, which randomised 2,168 six-month-old Malawian infants — 71 percent of whom were already anaemic — to six months of iron syrup plus malaria chemoprevention, iron-containing micronutrient powders plus chemoprevention, chemoprevention alone, or placebo [3]. The primary outcome, the Bayley Cognitive Scale composite at one year, showed no benefit from either iron intervention; nor did chemoprevention alone beat placebo on cognition. Both iron strategies raised ferritin but neither reduced anaemia prevalence beyond chemoprevention alone. The reassuring news is that iron did not increase malaria risk when given with chemoprevention, and chemoprevention alone did reduce malaria incidence. This is a clearly negative trial for the functional-benefit rationale behind universal preventive iron, and it should temper expectations about neurodevelopmental gains from supplementation programmes.

Finally, two papers on prevention and care delivery. Clinical Infectious Diseases reports a target trial emulation across 32 primary care practices, comparing over 7,400 infants who received nirsevimab with those who did not among more than 15,000 infants [4]. Nirsevimab was associated with a 14 percent relative reduction in antibiotic prescribing for outpatient acute respiratory tract infections, a 40 percent reduction for outpatient bronchiolitis encounters, and a large but imprecisely estimated reduction for antibiotic use in respiratory syncytial virus-related hospitalisations. As observational data it carries confounding risk, but it positions respiratory syncytial virus immunoprophylaxis as an antimicrobial stewardship intervention, not just a virology one. And in Open Forum Infectious Diseases, a multistate analysis of 1,110 people newly linked to human immunodeficiency virus care in St Louis, Missouri, found that at one year only about 55 percent were continuously retained, 15 percent had disengaged and reengaged, and 29 percent were disengaged [10]. Viral suppression was 93 percent among the continuously retained but 82 percent among those who reengaged, and younger people and those from historically marginalised communities cycled in and out most. Cross-sectional reporting hides this churn; the implication is that retention programmes should target sustained engagement rather than isolated missed visits.

If you only have time for one paper this week, make it the Infectious Diseases Society of America 2026 antimicrobial resistance guidance in Clinical Infectious Diseases [1]. It is the document you will actually reach for at the bedside, and this week's molecular epidemiology and stewardship papers show exactly why an annually refreshed version matters.

Here are the key takeaways from this week in Infectious Disease. First, use the live web version of the updated Infectious Diseases Society of America resistant gram-negative guidance, remembering it is United States-centric and gives adult dosing only. Second, carbapenem-resistant Klebsiella populations are diversifying and ceftazidime-avibactam resistance is rising in some networks, so check current local susceptibility rather than a decade-old mental model. Third, piperacillin-tazobactam courses of 14 days or longer independently raise the risk of emergent resistance, particularly Pseudomonas and AmpC producers — a concrete stewardship target. Fourth, prior meropenem exposure flags patients whose non-carbapenemase-producing carbapenem-resistant Enterobacterales are likely meropenem non-susceptible. Fifth, a digital adherence tool modestly improved tuberculosis treatment success and cut loss to follow-up, while metformin measurably lowered rifampicin and isoniazid exposure. Sixth, universal iron supplementation did not improve infant cognition in a malaria-endemic setting, though it did not increase malaria risk. And finally, nirsevimab reduces antibiotic prescribing, and longitudinal human immunodeficiency virus care data show care engagement is a cycle, not a binary.

That's your roundup for This Week in Infectious Disease. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.