Doxycycline vs levofloxacin for macrolide-unresponsive Mycoplasma pneumoniae pneumonia in children: a real-world study from China.

BACKGROUND

Macrolide-unresponsive Mycoplasma pneumoniae pneumonia (MUMPP) is a growing concern in paediatrics, yet comparative data on second-line antibiotics remain limited. This study aimed to evaluate the efficacy and safety of doxycycline and levofloxacin in treating MUMPP in children.

METHODS

This single-centre, real-world observational cohort study was conducted in a tertiary paediatric hospital in China. We included children hospitalised with M. pneumoniae pneumonia between July and December 2023. Patients were classified as having macrolide-sensitive M. pneumoniae pneumonia (MSMPP) or MUMPP according to their clinical response to initial azithromycin therapy. MUMPP patients were further allocated to the doxycycline, levofloxacin, or azithromycin group, based on the second-line antibiotic prescribed. Clinical characteristics, treatment outcomes, and adverse events were compared across groups. The primary outcome measures were the duration of pulmonary signs and the duration of fever after the change of antibiotic. The secondary outcome measure was the incidence of adverse events.

RESULTS

MUMPP accounted for 66.6% (365/548) of cases. Compared with MSMPP, children with MUMPP had higher rates of tachypnoea (17.5% vs 1.6%, p < 0.001) and high-grade fever (82.2% vs 60.7%, p < 0.001).The median fever duration after second-line antibiotic treatment was 0 days (IQR 0–1) in both the doxycycline (n = 190) and levofloxacin (n = 130) groups, with no significant difference between them (p = 0.088), but both were significantly shorter than in the azithromycin group (n = 45; median 3 days, IQR 2–3; both p < 0.001). IPTW-adjusted and age-stratified (3–6 years) analyses, in which baseline characteristics were fully balanced, yielded consistent results, and showed no significant difference between the doxycycline and levofloxacin groups in defervescence rates across the four time intervals (within 24 h, 24–48 h, 48–72 h, and > 72 h), and no differences in pulmonary sign resolution among any of the three groups. No cases of tooth discolouration or musculoskeletal complications were observed, and no gastrointestinal events of GSRS grade ≥5 or other CTCAE grade ≥3 adverse events occurred in any treatment group during the treatment period.

CONCLUSIONS

This study demonstrated that both doxycycline and levofloxacin are effective and safe treatments for paediatric MUMPP. Their judicious use may also help optimise antibiotic prescribing practices and mitigate antimicrobial resistance.