Adding cefixime to azithromycin did not improve typhoid cure, South Asian trial finds

A large trial in South Asia has tested a plausible idea — that two antibiotics might clear typhoid fever better than one — and found no benefit. Adding cefixime to azithromycin did not lower the treatment-failure rate, according to the ACT-South Asia trial published in The Lancet Infectious Diseases on August 20 [s1]. For a disease increasingly hard to treat, the result argues for using less, not more.

The problem the trial addressed

WHO recommends azithromycin for uncomplicated typhoid fever, and cefixime is also commonly used, but both drugs have reported failure rates of 10% or more [s1]. The investigators reasoned that combining azithromycin's activity inside cells with cefixime's activity outside them might close each drug's pharmacokinetic gaps and sterilise tissue more completely [s1]. The trial was designed to find out whether the combination would cut failure rates compared with azithromycin alone [s1].

How it was done

ACT-South Asia was a double-blind, parallel-group, randomised, placebo-controlled, phase 4 trial run in Nepal, Bangladesh, and Pakistan [s1]. It enrolled adults and children aged 2 to 65 years attending emergency and outpatient clinics with acute undifferentiated fever lasting 3 to 14 days, a C-reactive protein of at least 10 mg/L, and negative tests for dengue, scrub typhus, malaria, and COVID-19 [s1].

Participants were randomly assigned 1:1 to receive oral azithromycin at 20 mg/kg once daily (maximum 1 g) plus cefixime at 10 mg/kg twice daily (maximum 400 mg), or azithromycin plus a matching placebo, each for 7 days [s1]. Allocation was concealed from all study personnel and participants throughout [s1].

The primary outcome was a composite of treatment failure: a fever clearance time of 7 days or more, microbiological failure at day 7, the need for rescue treatment, or a complication or relapse within 28 days [s1].

What they found

Between May 9, 2021, and September 30, 2025, the team screened 46,947 people and randomly assigned 1,847 — 926 to azithromycin-cefixime and 921 to azithromycin-placebo [s1]. The modified intention-to-treat population comprised 1,831 participants, of whom 350 had blood culture-confirmed typhoid [s1].

Treatment failed in 44 participants (5.1%) in each group — an absolute risk difference of essentially zero (-0.00 percentage points, 95% CI -2.08 to 2.08; p=1.00) [s1]. Among the culture-confirmed cases, 19 of 179 (11.2%) failed in the combination group versus 26 of 171 (16.0%) on azithromycin alone, a difference of 4.48 percentage points (95% CI -2.52 to 12.21) that did not reach significance (p=0.20) [s1].

Adverse events were similar: they occurred in 142 of 918 participants (16%) on the combination and 144 of 917 (16%) on azithromycin alone [s1]. Serious adverse events requiring hospitalisation occurred in 21 participants (2%) and 17 participants (2%), respectively [s1].

Why a null result matters here

Typhoid remains a heavy burden. WHO's fact sheet, drawing on 2019 estimates, records about 9 million cases of typhoid fever annually, resulting in around 110,000 deaths per year [s2]. The disease is spread through contaminated food and water, and drug resistance has narrowed the options for treating it.

Against that backdrop, a clean negative result is useful precisely because it settles a question. There was no signal that a second antibiotic helps, and the trial's own reading is that its findings support the existing WHO recommendation of oral azithromycin alone — with important implications for antimicrobial stewardship [s1]. Every unnecessary antibiotic given is selection pressure for further resistance, so evidence that the add-on offers no benefit is evidence against using it.

The caveats

The composite failure rate rested on relatively few culture-confirmed cases: 350 of the 1,831 analysed had blood culture confirmation, so most of the primary analysis drew on clinically suspected rather than laboratory-confirmed typhoid [s1]. The confidence interval around the culture-confirmed comparison was wide, spanning both a meaningful benefit and a meaningful harm, so the trial cannot entirely rule out a smaller effect in that subgroup [s1]. Enrolment also ran across the COVID-19 years, and participants who tested positive for the virus at baseline were excluded from the primary analysis [s1].

The trial was funded by the UK Department of Health and Social Care, the Foreign, Commonwealth & Development Office, the Medical Research Council, and Wellcome Trust, and is registered as NCT04349826 [s1].

What to watch

Whether treatment guidelines and prescribing habits in high-burden settings hold to single-drug therapy, and whether the wider push against typhoid leans harder on prevention — clean water, sanitation, and conjugate vaccination — rather than on ever more complex antibiotic regimens.

This article is informational and is not medical advice.