The study, led by Masoumeh Beig and co-first author Sara Hajilari, with Khalil Azizian and Ebrahim Kouhsari as equally contributing corresponding authors, followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and included studies published up to 8 February 2026. The team used R software to compute pooled prevalence estimates with 95 percent confidence intervals, quantifying heterogeneity with Cochran’s Q and tau-squared statistics. Their focus was the minimum inhibitory concentration, or MIC, the lowest concentration of an antibiotic that prevents visible bacterial growth. When MICs rise above established breakpoints, an isolate is classified as non-susceptible, signaling that standard dosing may fail at the site of infection.

Macrolides work by binding to the 50S subunit of the bacterial ribosome, blocking the exit tunnel through which newly synthesized proteins must pass. Erythromycin, the prototype 14-membered macrolide, azithromycin, a 15-membered derivative prized for its tissue penetration, and clarithromycin and roxithromycin, both 14-membered agents, all share this target. Josamycin, a 16-membered macrolide, binds somewhat differently and has retained activity where its smaller relatives have faltered. Resistance in mycoplasmas typically arises through point mutations in the 23S ribosomal RNA gene, particularly in domain V, which alter the drug-binding pocket. Because these organisms lack a cell wall, beta-lactam antibiotics such as penicillins are useless against them, leaving macrolides, tetracyclines, and fluoroquinolones as the principal therapeutic options.

The numbers assembled by the Iranian team paint a genus-specific picture. For Ureaplasma species, pooled non-susceptibility was 13.3 percent for erythromycin, 15.3 percent for roxithromycin, 9.9 percent for clarithromycin, 6.4 percent for azithromycin, and just 2.6 percent for josamycin. For Mycoplasma species, the corresponding estimates were dramatically higher: 63.4 percent for roxithromycin, 84.5 percent for clarithromycin, 78.8 percent for azithromycin, and 7.7 percent for josamycin. Erythromycin non-susceptibility in Mycoplasma reached 75.7 percent. The authors caution, however, that Mycoplasma hominis is intrinsically resistant to 14- and 15-membered macrolides, meaning pooled estimates for erythromycin, azithromycin, clarithromycin, and roxithromycin in that species reflect biology as much as acquired resistance. For M. hominis, the clinically meaningful figure is josamycin, where pooled non-susceptibility was only 6.1 percent.

Geography emerged as a significant modifier for four of the five drugs examined. The meta-analysis detected significant geographic variation in non-susceptibility for erythromycin, clarithromycin, azithromycin, and josamycin, all with P values below 0.05, while roxithromycin showed no significant regional difference, with a P value of 0.283. This pattern suggests that local prescribing habits, population structure, and transmission dynamics shape resistance profiles unevenly across the globe. In regions where azithromycin has been deployed widely, including mass drug administration campaigns and routine sexually transmitted infection management, elevated MICs appear more frequently, whereas areas with lighter macrolide pressure retain comparatively vulnerable bacterial populations.

Perhaps the most reassuring finding is what the analysis did not show. Across the entire dataset, no significant temporal trends were observed for any of the evaluated macrolides: the P value was 0.994 for erythromycin, 0.165 for roxithromycin, 0.615 for clarithromycin, 0.326 for azithromycin, and 0.467 for josamycin. In other words, resistance levels appear to have remained roughly stable over the period covered by the included studies rather than climbing year over year. The authors interpret this cautiously, noting that stability is not the same as safety. A plateau at 75 percent erythromycin non-susceptibility in Mycoplasma offers little therapeutic comfort, and the absence of a trend may simply reflect the coarse resolution of pooled cross-sectional data.

The clinical implications are direct. Urogenital mycoplasmas and ureaplasmas are associated with urethritis, cervicitis, pelvic inflammatory disease, adverse pregnancy outcomes, and complications in neonates, and they are among the most difficult sexually associated bacteria to culture, often requiring nucleic acid amplification tests for detection and specialized broth-based systems for susceptibility testing. When macrolides fail, clinicians typically turn to tetracyclines such as doxycycline or to fluoroquinolones, but resistance to those classes has also been documented in these organisms, narrowing the armamentarium further. The new pooled estimates give guideline committees a quantitative baseline for deciding when macrolide therapy remains defensible and when an alternative should be the default.

Josamycin stands out as the quiet winner of the analysis. With pooled non-susceptibility of 2.6 percent in Ureaplasma and 7.7 percent in Mycoplasma overall, and 6.1 percent in M. hominis specifically, the 16-membered macrolide demonstrated consistently low MICs across the included studies. The authors suggest it may represent a promising alternative, particularly where 14- and 15-membered agents have lost efficacy. Josamycin is not licensed in every country, and its availability varies widely, but the data provide an evidence-based argument for reconsidering its place in treatment algorithms for urogenital infections, especially in regions where azithromycin MICs have drifted upward.

For erythromycin and azithromycin, the message is restraint. The combination of high erythromycin resistance and moderate elevation of azithromycin MICs, the authors conclude, warrants cautious use of these agents, reserving them for cases where susceptibility has been confirmed or where no better option exists. Empirical macrolide monotherapy for suspected Mycoplasma-driven disease, once a reflexive choice, now carries a substantial probability of failure in many settings. The researchers emphasize that ongoing global surveillance remains essential to support evidence-based treatment guidelines, since resistance in organisms with rapid replication and frequent horizontal gene exchange can shift within years, and the fragmented geographic data they aggregated highlight how little standardized susceptibility testing exists in much of the world.

The study, published open access on 8 October 2026 and citable under DOI 10.1186/s12879-026-14585-9, was approved by the Ethics Committee of Golestan University of Medical Sciences and conducted in accordance with the Declaration of Helsinki. The authors declared no competing interests and no external funding, and they state that a language-model tool was used solely to polish the manuscript’s prose, not to collect or analyze data. As antimicrobial resistance continues to erode the utility of first-line antibiotics across the infectious disease spectrum, this meta-analysis offers both a warning and a waypoint: the macrolide era for urogenital Mycoplasma is largely over in much of the world, but for Ureaplasma, and for josamycin across both genera, carefully stewarded options remain.

Subject of Research: Global macrolide antibiotic resistance in urogenital Ureaplasma and Mycoplasma species

Article Title: Global macrolide non-susceptibility in urogenital Ureaplasma and Mycoplasma: a systematic review and meta-analysis

Article References: Beig, M., Azizian, K., Hajilari, S., Shakourzadeh, M. Z., & Kouhsari, E. (2026). Global macrolide non-susceptibility in urogenital Ureaplasma and Mycoplasma: a systematic review and meta-analysis. BMC Infectious Diseases. https://doi.org/10.1186/s12879-026-14585-9

Image Credits: AI Generated

DOI: 10.1186/s12879-026-14585-9

Keywords: antimicrobial resistance, macrolides, Ureaplasma, Mycoplasma, erythromycin, azithromycin, josamycin, minimum inhibitory concentration, systematic review, meta-analysis, sexually transmitted infections, BMC Infectious Diseases