Asthma affects over 300 million patients worldwide and is characterized by intermittent reversible airflow obstruction and airway inflammation. Pneumocystis, an opportunistic fungus, has recently been implicated in asthma pathogenesis. In preclinical models, Pneumocystis exposure causes goblet cell hyperplasia, bronchial hyperreactivity, and airway inflammation.1,2 From a clinical perspective, patients with severe asthma have elevated anti-Pneumocystis antibody titers compared to healthy controls and higher levels of anti-Pneumocystis IgG are correlated with worsened lung function.1 Furthermore, evaluation of the lung mycobiome identified that Pneumocystis jirovecii 18S rRNA was more abundant in the bronchoalveolar lavage fluid of patients with severe asthma compared to non-asthmatic controls.3 Cumulatively, these studies suggest that Pneumocystis may represent an unrecognized contributor to asthma pathophysiology.
In this study, we sought to determine if Pneumocystis treatment improved lung function or exacerbation frequency in patients with asthma. We compared children who received trimethoprim-sulfamethoxazole (TMP-SMX), a standard therapeutic for Pneumocystis, with asthmatic patients who received clindamycin, an antibiotic without activity against Pneumocystis that is commonly used for similar indications (e.g. SSTI) as a control.4 The electronic health record was queried for all patients from 2010-2018 who had an order/prescription for TMP-SMX or clindamycin and at least one spirometry. Each record was then manually reviewed for a diagnosis of asthma and for exclusion criteria (cystic fibrosis, ciliopathy, transplantation, malignancy, bronchiectasis, restrictive lung disease, prematurity <32 weeks, bronchopulmonary dysplasia, ventilator dependence, congenital heart disease, immunodeficiency, sickle cell disease, collagen vascular disorders and one-time doses of antibiotic). This study was approved by the Institutional Review Board (STUDY19010136).
Between 2010-2018, 79,047 orders of TMP-SMX and 86,033 orders of clindamycin were placed for 14,109 and 20,283 unique patients, respectively. Similarly, there were 46,215 spirometries performed on 16,444 unique patients. Among patients with spirometry data, there were 1,089 patients with an order for TMP-SMX and 937 patients with an order for clindamycin. After chart review and exclusion, 144 patients with asthma in the TMP-SMX group and 202 patients with asthma in the clindamycin group were included in the cross-sectional analysis (Table 1, right). Of those, 24 patients in the TMP-SMX group and 41 patients in the clindamycin group had both baseline and follow-up spirometry (+/− 12 months of antibiotic course) and were included in the longitudinal analysis (Table 1, left). Demographics between the TMP-SMX and clindamycin groups were similar, although TMP-SMX was prescribed more commonly for urologic indications (Table 1).
Table 1.
Trimethoprim-sulfamethoxazole use in children with asthma is associated with improved FEV1 and reduced ED visits.
| Baseline/Follow up PFT cohort | Total cohort | ||||||
|---|---|---|---|---|---|---|---|
| TMP-SMX (n=24) | Clindamycin (n=41) | p-value | TMP-SMX (n=144) | Clindamycin (n=202) | p-value | ||
| Demographics | |||||||
| Age | |||||||
| Mean in years (+/− SEM) | 10.7 (0.86) | 11.2 (0.63) | 0.59 | 10.6 (0.44) | 10.7 (0.33) | 0.73 | |
| Biologic sex, n (%) | |||||||
| Male | 10 (42%) | 25 (61%) | 0.20 | 68 (47%) | 107 (53%) | 0.33 | |
| Female | 14 (58%) | 16 (39%) | 76 (53%) | 95 (47%) | |||
| Race, n (%) | |||||||
| Caucasian | 18 (75%) | 21 (51%) | 0.16 | 101 (70%) | 122 (60%) | 0.09 | |
| African American | 5 (21%) | 18 (44%) | 40 (28%) | 76 (38%) | |||
| Other/declined | 1 (4%) | 2 (5%) | 3 (2%) | 4 (2%) | |||
| Indication, n (%) | |||||||
| ENT | 7 (29%) | 7 (17%) | 0.02 | 26 (18%) | 51 (25%) | 0.03 | |
| Urologic | 6 (25%) | 1 (2%) | 44 (31%) | 1 (<1%) | |||
| SSTI | 9 (37%) | 24 (59%) | 71 (49%) | 134 (66%) | |||
| Respiratory | 2 (8%) | 9 (22%) | 3 (2%) | 16 (8%) | |||
| Asthma therapy, n (%) | |||||||
| No controller | 3 (13%) | 6 (15%) | 0.99* | 75 (52%) | 87 (43%) | 0.09 | |
| Inhaled corticosteroid | |||||||
| Low-dose ICS | 9 (37%) | 11 (27%) | 0.49 | 28 (19%) | 47 (23%) | 0.51 | |
| Medium ICS | 3 (13%) | 7 (17%) | 4 (3%) | 14 (6%) | |||
| High-dose ICS | 9 (37%) | 17 (41%) | 39 (27%) | 54 (27%) | |||
| Systemic steroid | 1 (4%) | 2 (5%) | 0.99* | 1 (<1%) | 2 (<1%) | 0.99* | |
| Biologic | 1 (4%) | 3 (7%) | 0.99* | 1 (<1%) | 3 (1%) | 0.64* | |
| Spirometry | Emergency Department Visits | ||||||
| Days from baseline spirometry to antibiotics Mean (+/−SE) | 177 (+/−43) | 194 (+/−26) | 0.71* | Total ED visits | |||
| 12 months prior | 60 | 71 | 0.04 | ||||
| 12 months after | 32 | 66 | |||||
| Days from antibiotics to follow up spirometry Mean (+/−SE) | 164 (+/−33) | 131 (+/−19) | 0.36* | Change in ED visits after antibiotics | |||
| Fewer | 22 (15.1%) | 28 (13.9%) | 0.05 | ||||
| Escalation in therapy n (%) | 2 (8%) | 6 (15%) | 0.70 | Same | 6 (4.2%) | 10 (5.0%) | |
| More | 9 (6.3%) | 32 (15.8%) | |||||
| De-escalation in therapy n (%) | 1 (4%) | 3 (7%) | 0.99 | None | 107 (74.3%) | 132 (65.3%) | |
| FEV1 Mean (95% CI) | |||||||
| Baseline | 90.3 (82.8-97.9) | 93.6 (89.3-98.5) | 0.42 | ||||
| Follow-up | 96.4 (89.5-103) | 93.3 (88.6-98.1) | 0.44 | ||||
| Change | 6.04 (2.4-9.7) | −0.56 (−2.8-1.7) | 0.002, 0.001‡ | ||||
| FVC Mean (95% CI) | |||||||
| Baseline | 97.8 (91.0-104.5) | 102.3 (97.4-107.1) | 0.29 | ||||
| Follow-up | 102.5 (96.6-108.3) | 101.4 (96.2-106.5) | 0.78 | ||||
| Change | 4.71 (0.96-8.5) | −0.93 (−3.1-1.3) | 0.006, 0.001‡ | ||||
| FEV/FVC Mean (95% CI) | |||||||
| Baseline | 81.3 (77.5-85.1) | 80.4 (77.8-83.0) | 0.68 | ||||
| Follow-up | 82.6 (79.2-86.0) | 80.7 (78.1-83.4) | 0.78 | ||||
| Change | 1.33 (−1.4-4.1) | 0.34 (−1.3-1.9) | 0.49, 0.335‡ | ||||
| FEF25-75 Mean (95% CI) | |||||||
| Baseline | 79.8 (61.9-97.7) | 83.2 (72.4-94.0) | 0.46 | ||||
| Follow-up | 83.1 (69.5-96.8) | 75.4 (67.7-83.1) | 0.38 | ||||
| Change | 4.35 (−8.8-17.5) | −7.80 (−17.4-1.8) | 0.13, 0.065‡ | ||||
Categorical variables analyzed via Chi-square test or Fisher’s exact test (for analyses with n<5, denoted by *).
Continuous variables analyzed by unpaired t-test unless noted otherwise.
denotes multivariable linear regression analysis.
Baseline spirometry showed no differences in FEV1, FVC, FEV1/FVC, or FEF25-75 between antibiotic groups (Table 1). Similarly, there were no significant differences in the interval between baseline spirometry to antibiotics or antibiotics to follow-up spirometry. There were no differences in the proportion of patients requiring either an escalation or de-escalation in therapy between measurement of spirometry.
In patients that received TMP-SMX, there was a significant increase in FEV1 between baseline and follow-up spirometry (p=0.0023 by paired t-test, Table 1); in contrast, there was no significant FEV1 change in the clindamycin group (p=0.78). Of note, 79% of asthma patients receiving TMP-SMX had an increased FEV1 at the follow-up, compared to 44% in the clindamycin group (p=0.01). Similarly, there was a significant increase in FVC with TMP-SMX (p=0.016) but not clindamycin (p=0.54). After adjusting for age, sex, race, antibiotic indication, ICS dose, and the time intervals between spirometry and antibiotics, the improvements in FEV1 and FVC were again significantly higher in the TMP-SMX group than the clindamycin group. Furthermore, patients who received TMP-SMX had a significant reduction in the proportion and total number of ED visits for asthma exacerbations the 12 months following antibiotics (Table 1, right).
To our knowledge, this is the first association between the use of an antibiotic active against Pneumocystis and the improvement of lung function and asthma control. The role of antibiotics in asthma pathogenesis has been evaluated previously. Any antibiotic use, including TMP-SMX, within the first 6 months of life is associated with development of atopic diseases including asthma.5 Therapeutically, there is limited evidence for lung function improvement following antibiotic treatment in the setting of an exacerbation.6 As a long-term therapy used outside of exacerbations, however, macrolide antibiotics have been shown to increase peak flow measurements and correlate with improved symptoms without changes in lung function measures.7 None of these studies, however, evaluated antibiotics with activity against Pneumocystis or assessed changes in pulmonary function or exacerbation rate after the antibiotic course.
The only known natural reservoir of Pneumocystis jirovecii is humans and near ubiquitous Pneumocystis prevalence has been demonstrated in infants on autopsy specimens. 8,9 The presence of Pneumocystis has been associated with increased MUC5AC expression, a mucin implicated in asthma pathogenesis.9,10 The detection of subclinical Pneumocystis infection or Pneumocystis colonization is limited by the lack of a non-invasive biomarker. Future studies characterizing the natural history of Pneumocystis burden in the lungs of patients with asthma could be highly informative.
There are several limitations to the current study. Despite being at a large tertiary care pediatric hospital, the cohort of patients with baseline and follow up spirometry in the time frame of antibiotic exposure was relatively small. Second, while our hypothesis is that TMP-SMX would improve lung function due to activity against Pneumocystis, TMP-SMX is active against other human commensals and pathogens, and the impact of TMP-SMX on the human microbiota (e.g. lung, gut) cannot be excluded. Additionally, the retrospective nature of the current pilot investigation cannot demonstrate a causal or direct relationship between TMP-SMX and changes in lung function. Finally, although quite similar, the dose and duration of antibiotic exposure was not uniform across the patient population and warrants further investigation.
Despite the limitations of this retrospective study, the current study demonstrates an association between TMP-SMX use and improved lung function and reduced exacerbation rate in pediatric asthma. Further, the mounting animal model and clinical data linking Pneumocystis to asthma suggests that further study using a randomized, double-blind placebo-controlled trial evaluating the effects of TMP-SMX treatment in patient with asthma is warranted.
Acknowledgements:
We would like to thank Dr. Jay Kolls and Dr. Judith Martin for assistance in conceptualization of this study. We would like to thank Dr. Alyce Anderson for serving as a statistical advisor on this study.
Funding sources: Dr. Forno’s contribution was in part funded by grant HL149693 from the U.S. National Institutes of Health (NIH). Dr. Campfield’s contribution was in part funded by grant K08HL128809 from the U.S. National Institutes of Health (NIH). The funder/sponsor did not participate in this work.
Abbreviations:
- TMP-SMX
Trimethoprim-sulfamethoxazole
- PFT
pulmonary function test
- EHR
electronic health record
- FEV1
forced expiratory volume in the first second
- FVC
forced vital capacity
- FEF25-75
mid-expiratory flow rate
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of interest: The authors have no conflicts of interest to disclose.
Bibliography
- 1.Eddens T, Campfield BT, Serody K, et al. A Novel CD4+ T Cell-Dependent Murine Model of Pneumocystis-driven Asthma-like Pathology. Am J Respir Crit Care Med. 2016;194(7):807–820. doi: 10.1164/rccm.201511-2205OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Iturra PA, Rojas DA, Pérez FJ, et al. Progression of Type 2 Helper T Cell-Type Inflammation and Airway Remodeling in a Rodent Model of Naturally Acquired Subclinical Primary Pneumocystis Infection. Am J Pathol. 2018;188(2):417–431. doi: 10.1016/j.ajpath.2017.10.019 [DOI] [PubMed] [Google Scholar]
- 3.Goldman DL, Chen Z, Shankar V, Tyberg M, Vicencio A, Burk R. Lower airway microbiota and mycobiota in children with severe asthma. J Allergy Clin Immunol. 2018;141(2):808–811.e7. doi: 10.1016/j.jaci.2017.09.018 [DOI] [PubMed] [Google Scholar]
- 4.Queener SF, Bartlett MS, Richardson JD, Durkin MM, Jay MA, Smith JW. Activity of clindamycin with primaquine against Pneumocystis carinii in vitro and in vivo. Antimicrob Agents Chemother. 1988;32(6):807–813. doi: 10.1128/aac.32.6.807 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zven SE, Susi A, Mitre E, Nylund CM. Association between use of multiple classes of antibiotic in infancy and allergic disease in childhood. JAMA Pediatr. December 2019. doi: 10.1001/jamapediatrics.2019.4794 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Normansell R, Sayer B, Waterson S, Dennett EJ, Del Forno M, Dunleavy A. Antibiotics for exacerbations of asthma. Cochrane Database Syst Rev. 2018;6:CD002741. doi: 10.1002/14651858.CD002741.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Reiter J, Demirel N, Mendy A, et al. Macrolides for the long-term management of asthma--a meta-analysis of randomized clinical trials. Allergy. 2013;68(8):1040–1049. doi: 10.1111/all.12199 [DOI] [PubMed] [Google Scholar]
- 8.Chabé M, Aliouat-Denis C-M, Delhaes L, Aliouat EM, Viscogliosi E, Dei-Cas E. Pneumocystis: from a doubtful unique entity to a group of highly diversified fungal species. FEMS Yeast Res. 2011;11(1):2–17. doi: 10.1111/j.1567-1364.2010.00698.x [DOI] [PubMed] [Google Scholar]
- 9.Vargas SL, Ponce CA, Gallo M, et al. Near-universal prevalence of Pneumocystis and associated increase in mucus in the lungs of infants with sudden unexpected death. Clin Infect Dis. 2013;56(2):171–179. doi: 10.1093/cid/cis870 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Evans CM, Kim K, Tuvim MJ, Dickey BF. Mucus hypersecretion in asthma: causes and effects. Curr Opin Pulm Med. 2009;15(1):4–11. doi: 10.1097/MCP.0b013e32831da8d3 [DOI] [PMC free article] [PubMed] [Google Scholar]