Mycoplasma pneumoniae carriage in children with recurrent respiratory tract infections is associated with a less diverse and altered microbiota

Abstract

Background: Mycoplasma pneumoniae is a common cause of community-acquired pneumonia in school-aged children and can be preceded by asymptomatic carriage. However, its role in recurrent respiratory tract infections is unclear. We studied the prevalence of M.pneumoniae carriage in children with recurrent respiratory infections and identified associated factors.

Methods: We tested M.pneumoniae carriage by qPCR in children with recurrent infections and their healthy family members in a cross-sectional study. Serum and mucosal total and M.pneumoniae-specific antibody levels were measured by ELISA and nasopharyngeal microbiota composition was characterized by 16S-rRNA sequencing.

Findings: Prevalence of M.pneumoniae carriage was higher in children with recurrent infections (68%) than their family members without infections (47% in siblings and 27% in parents). M.pneumoniae carriage among family members appeared to be associated with transmission within the household, likely originating from the affected child. In logistic regression corrected for age and multiple comparisons, IgA (OR 0.16 [0.06-0.37]) and total IgG deficiency (OR 0.15 [0.02-0.74]) were less prevalent in M.pneumoniae carriers (n = 78) compared to non-carriers (n = 36). In multivariable analysis, the nasopharyngeal microbiota of M.pneumoniae carriers had lower alpha diversity (OR 0.27 [0.09-0.67]) and a higher abundance of Haemophilus influenzae (OR 45.01 [2.74-1608.11]) compared to non-carriers.

Interpretation: M.pneumoniae carriage is highly prevalent in children with recurrent infections and carriers have a less diverse microbiota with an overrepresentation of disease-associated microbiota members compared to non-carriers. Given the high prevalence of M.pneumoniae carriage and the strong association with H. influenzae, we recommend appropriate antibiotic coverage of M.pneumoniae and H. influenzae in case of suspected pneumonia in children with recurrent respiratory tract infections or their family members.

Funding: Wilhelmina Children's Hospital Research Fund, 'Christine Bader Stichting Irene KinderZiekenhuis', Sophia Scientific Research Foundation, ESPID Fellowship funded by Seqirus, Hypatia Fellowship funded by Radboudumc and The Netherlands Organisation for Health Research and Development (ZonMW VENI grant to LM Verhagen).

Keywords: Children; Haemophilus influenzae; Immunoglobulin A; Microbiota; Mycoplasma pneumoniae; Recurrent respiratory tract infections.

Fig. 1

Flowchart of children with recurrent respiratory tract infections and their family members included in the study.

Fig. 2

Antibody levels in children with recurrent respiratory tract infections. (A–D) Antibody levels in children with rRTIs. (A) Serum total IgA levels in M.pneumoniae carriers and non-carriers. (B) Serum total IgG levels in M.pneumoniae carriers and non-carriers. (C) Nasopharyngeal swab total IgA levels in M.pneumoniae carriers and non-carriers. (D) Nasopharyngeal swab M.pneumoniae-specific IgA levels in M.pneumoniae carriers and non-carriers. Lines represent medians and error bars show interquartile ranges. ∗p < 0.05, ∗∗p < 0.01 (logistic regression analysis corrected for age and multiple comparison).

Fig. 3

Nasopharyngeal microbiota alpha diversity in M.pneumoniae carriers versusnon-carriers. Distribution of alpha diversity (A, Shannon) and richness (B, number of observed species) of the respiratory microbiota in both M.pneumoniae carriers and non-carriers with rRTIs. (C) Spearman correlation of alpha diversity with age for M.pneumoniae carriers (Spearman correlation coefficient −0.064, p = 0.61) and non-carriers (correlation coefficient 0.318, p = 0.07). (A–B) Lines represent medians and error bars show interquartile ranges. ∗∗p < 0.01 (logistic regression analysis corrected for age).

Fig. 4

Microbiota composition of Mycoplasma pneumoniae carriers compared to non-carriers with rRTIs. (A) nMDS plot comparing beta diversity based on Bray–Curtis dissimilarity between M.pneumoniae carriers and non-carriers (Stress 0.192). (B) Stacked barplot of highest ranking ASVs (based on mean relative abundance) in M.pneumoniae carriers and non-carriers on genus level. (C) Stacked barplot of the highest ranking (based on mean relative abundance) on 99%-similarity species level in M.pneumoniae carriers and non-carriers. (D) Hierarchical clustering of nasopharyngeal microbiota on 99%-similarity species.