Discussion
This overall sample result in this study would appear to suggest that steroids positively alter FEV1pp trajectory when there is initial failure to progress during APE when we examine all eligible encounters (Table 1). This assumes that all CF admissions were 12-14 days in length and that persons not showing improved FEV1pp after 5-7 days of treatment do not return to baseline lung function after two weeks of conventional treatment. Those assumptions are incorrect and highlight the limits of retrospective cohort studies. To address this limitation, we then used propensity score matching within our data set. This allows the construction of a non-steroid treated control group against the steroid treated group by matching characteristics such as gender, lung function at baseline, or CFTR mutation (Model 1, Table 2). This PS-matching approach has been called a “retrospective randomization” by some authors as it seeks to reduce assignment bias and mimic randomization13. Thus in Models 1 and 2, we found no significance difference in FEV1pp at baseline or discharge. Moreover, extended analysis including FEV1pp at follow-up visit and time to next exacerbation were not significantly different.
Although our data showed a non-significant trend in time to next exacerbation requiring antimicrobial therapy, this is likely due to size of sample. The median time to next antibiotics is 99 (IQR: 51, 123.5) in the steroid group versus 70.5 (IQR: 37-152). The Kaplan-Meier graph (Figure 3) shows that for the first 80 days, the no-steroid group is more likely to need antimicrobial treatment, however beyond 80 days, the steroid group is more likely. The factors such as low number of patients (n=34) in the PS-matched sample, as well as the retrospective nature of the study and non-standardized follow up appointments could contribute to the non-significance.
When FEV1pp or clinical health fails to improve as expected during APE treatment, providers and patients alike look for alterations in the treatment plan to improve lung function to baseline and overall well-being. At our center, approaches might include changing the antimicrobial regimen, increasing the total days of IV therapy, increasing airway clearance therapy frequency from 4 to 5 session per day, or adding corticosteroids. These options are presented to the patient and family. Our center generally utilizes a “rescue” dose of oral prednisone 2 mg/kg/day up to 60 mg given for 5-7 days. Considerations for steroid therapy include: positive response on previous use, suspected asthma, and physical exam finding including wheezing. Although not reviewed in the study, management of hyperglycemia can be a concern even in short term rescue use. Currently, there is no published guidance or recommendation concerning steroid use during CF exacerbations.
There has been previous interest in oral steroid therapy in CF. Long-term (12 week) prednisolone therapy in 24 clinically stable children with CF demonstrated decreased concentration of inflammatory markers including IL-1α, sIL-2R, and IgG 14. This study protocol dosed prednisolone at 2 mg/kg/day for the first week, then tapered to 1 mg/kg every other day for the following 10 weeks. FEV1 benefit was associated with the steroid group at day 14, however significance was not maintained at week 12. While there is some data that alternate-day steroid use for three weeks to four years improves pulmonary function 14-16, adverse events such as poor glucose tolerance, insufficient linear growth, and Pseudomonas aeruginosa colonization are noted after long-term administration of steroids in CF.
Short-term use of systemic high dose steroids appears to have even less evidence for use. CF pulmonary exacerbation treatment guidelines report that there is insufficient evidence for use of oral corticosteroids during APEs 2. Three studies show that short term therapy for three to ten days show modest efficacy in improving pulmonary function. In a pilot placebo-controlled study of oral prednisone therapy (2 mg/kg, max 60 mg) administration for the first five days of hospitalization in patients with APE, FEV1pp was consistently higher in the steroid-treated group, however the FEV1 difference did not reach statistical significance in this small study of 24 patients 17. A case report of four young patients under age six with no clinical improvement after IV antimicrobial therapy for APE showed dramatic improvement in respiratory distress and oxygen requirements after IV methylprednisolone burst (1 g/1.73 m2 per day for 3 days) 18. A study of 20 infants with APE demonstrated a statistically significant increase in forced expiratory flows after treatment with hydrocortisone in addition to standard therapy, with no significant increase in placebo treated group 19. Additionally, several of the infants given placebo in this study, and none of the steroid-treated infants, had a recurrence in respiratory symptoms between discharge and outpatient follow up. Overall, studies remain poorly powered and larger clinical trials would be needed to better elucidate the role of steroids used during treatment of pulmonary exacerbations.
An alternate approach when failure to improve is noted during APE is to increase the number of treatment days. However, in our experience most families and their children are opposed to longer hospital stays. Antimicrobial therapy in CF exacerbation ranges from 10-21 days with most individuals treated between 10 and 14 days. 20Studies examining length of IV therapy suggest diminishing returns after 14 days of therapy 21. Treatment durations of 10 versus 14 versus 21 days of conventional therapy has also been recently been studied by the CF Foundation and is also uncertain benefit (NCT02781610) [Goss]. The longer length of stay noted in our overall study cohort was primarily due to initiating steroid after poor response to 10- 14 days of therapy. As our conventional therapy duration is 10-14 days, the providers’ intent was to reassess lung function prior to steroid therapy. This often resulted in at least five more days of hospital care and additional spirometry at the end of the steroid treatment while continuing other treatments including antimicrobial therapy. Despite anecdotal reports of improvement, our Models 1 and 2 suggest no benefit overall to acute rescue steroids. Waiting until day 12-14 to start steroid treatment may only add unnecessary hospital days.
To our knowledge, there are no trials showing the effectiveness of switching antimicrobial therapy when using respiratory culture-based guided therapy when patients fail to progress. As summarized by Chmiel et al current thinking of CF airway pathogens is based on the recovery of a known cohort on surveillance cultures 22. Extended culturing techniques have demonstrated previously undocumented species in the CF microbiome including anaerobic species, however, these techniques are not in use in routine care. While diverse in youth, the CF microbiome diversity narrows as the patient ages23, suggesting the ability to tightly narrow antimicrobial coverage. Importantly, considering agents against MRSA orPseudomonas aeruginosa when not treated might be an alternate when confronted with failure to progress.
There is conflicting evidence that corticosteroid use is a risk factor for nontuberculous mycobacteria (NTM) colonization. Review of the literature notes that isolation of Aspergillus spp or Allergic Bronchopulmonary Aspergillosis (ABPA) diagnosis more common in NTM positive patients. Corticosteroid therapy is the mainstay of ABPA treatment 24. In a study of 139 patients with CF in Israel, six developed NTM lung disease 25. Five of the six patients had prolonged steroid treatments, four of which were treatment for an ABPA diagnosis prior to NTM acquisition. Of the 133 patients without NTM, only one had ABPA and prolonged steroids (p<0.001), therefore suggesting steroids as a risk factor for NTM. Conversely, a nested-cohort study found there was less steroid exposure days patients with NTM disease meeting ATS criteria than those who were NTM negative (p<0.05) 26. This study included 159 patients. Sixty were NTM positive, 22 of which had NTM disease meeting ATS criteria. A multicenter study in Israel reviewed 186 patients, 42 of which had NTM isolation 27. A multivariate analysis found an increased odds of Aspergillus sppin the NTM positive cohort (odds ratio 5.14, 95% CI 1.87–14.11). In a retrospective database review in the US, NTM was noted in 166 individuals of the 1216 28. Aspergillus fumigatus was more frequently found in NTM positive patients 13.9% vs. 7.2%, respectively, p<0.01. It appears that the Aspergillus colonization, not the steroids themselves may be the association with NTM positivity. Of the studies concerning for steroids, they include prolonged use, thus short “rescue” dosing regimens of steroids may not warrant alarm for development of NTM infection.
Our analysis is limited by sample size and retrospective nature, these data suggest a prospective trial with clear criteria for starting “rescue” treatment should be undertaken. There continues to be interest in the use of steroids in CF APE and a clear understanding of their use is warranted. A current randomized, double blind, placebo controlled trial of prednisone for patients failing to recover their FEV1 baseline at 7 days into IV antibiotic treatment for APE underway (NCT03070522) [Waters].
1. Bhatt JM. 2013. Treatment of pulmonary exacerbations in cystic fibrosis. Eur Respir Rev. 22(129):205-216.
2. Flume PA, Mogayzel PJ, Jr., Robinson KA, Goss CH, Rosenblatt RL, Kuhn RJ, Marshall BC. 2009. Cystic fibrosis pulmonary guidelines: Treatment of pulmonary exacerbations. American journal of respiratory and critical care medicine. 180(9):802-808.
3. Sanders DB, Hoffman LR, Emerson J, Gibson RL, Rosenfeld M, Redding GJ, Goss CH. 2010. Return of fev1 after pulmonary exacerbation in children with cystic fibrosis. Pediatric pulmonology. 45(2):127-134.
4. Flume PA, O’Sullivan BP, Robinson KA, Goss CH, Mogayzel PJ, Jr., Willey-Courand DB, Bujan J, Finder J, Lester M, Quittell L et al. 2007. Cystic fibrosis pulmonary guidelines: Chronic medications for maintenance of lung health. American journal of respiratory and critical care medicine. 176(10):957-969.
5. Cogen JD, Oron AP, Gibson RL, Hoffman LR, Kronman MP, Ong T, Rosenfeld M. 2017. Characterization of inpatient cystic fibrosis pulmonary exacerbations. Pediatrics. 139(2).
6. West NE, Beckett VV, Jain R, Sanders DB, Nick JA, Heltshe SL, Dasenbrook EC, VanDevanter DR, Solomon GM, Goss CH et al. 2017. Standardized treatment of pulmonary exacerbations (stop) study: Physician treatment practices and outcomes for individuals with cystic fibrosis with pulmonary exacerbations. Journal of cystic fibrosis : official journal of the European Cystic Fibrosis Society. 16(5):600-606.
7. Flynn TN. 2001. Design and analysis of cluster randomization trials in health research.: Allan donner and neil klar. London: Arnold, 2000, pp.178, £35.00. Isbn: 0-340-69153-0. International Journal of Epidemiology. 30(2):407-408.
8. Newson R. 2006. Confidence intervals for rank statistics: Somers’ d and extensions. The Stata Journal. 6(3):309-334.
9. MacDonald KD, Vesco KK, Funk KL, Donovan J, Nguyen T, Chen Z, Lapidus JA, Stevens VJ, McEvoy CT. 2016. Maternal body mass index before pregnancy is associated with increased bronchodilator dispensing in early childhood: A cross-sectional study. Pediatric pulmonology. 51(8):803-811.
10. Bellera CA, MacGrogan G, Debled M, de Lara CT, Brouste V, Mathoulin-Pélissier S. 2010. Variables with time-varying effects and the cox model: Some statistical concepts illustrated with a prognostic factor study in breast cancer. BMC Med Res Methodol. 10:20-20.
11. Adelson JL, McCoach DB, Rogers HJ, Adelson JA, Sauer TM. 2017. Developing and applying the propensity score to make causal inferences: Variable selection and stratification. Front Psychol. 8:1413.
12. Arpino B, Cannas M. 2016. Propensity score matching with clustered data. An application to the estimation of the impact of caesarean section on the apgar score. Stat Med. 35(12):2074-2091.
13. Jupiter DC. 2017. Propensity score matching: Retrospective randomization? J Foot Ankle Surg. 56(2):417-420.
14. Greally P, Hussain MJ, Vergani D, Price JF. 1994. Interleukin-1 alpha, soluble interleukin-2 receptor, and IgG concentrations in cystic fibrosis treated with prednisolone. Archives of disease in childhood. 71(1):35-39.
15. Auerbach HS, Williams M, Kirkpatrick JA, Colten HR. 1985. Alternate-day prednisone reduces morbidity and improves pulmonary function in cystic fibrosis. Lancet (London, England). 2(8457):686-688.
16. Eigen H, Rosenstein BJ, FitzSimmons S, Schidlow DV. 1995. A multicenter study of alternate-day prednisone therapy in patients with cystic fibrosis. Cystic fibrosis foundation prednisone trial group. The Journal of pediatrics. 126(4):515-523.
17. Dovey M, Aitken ML, Emerson J, McNamara S, Waltz DA, Gibson RL. 2007. Oral corticosteroid therapy in cystic fibrosis patients hospitalized for pulmonary exacerbation: A pilot study. Chest. 132(4):1212-1218.
18. Ghdifan S, Couderc L, Michelet I, Leguillon C, Masseline B, Marguet C. 2010. Bolus methylprednisolone efficacy for uncontrolled exacerbation of cystic fibrosis in children. Pediatrics. 125(5):e1259-1264.
19. Tepper RS, Eigen H, Stevens J, Angelicchio C, Kisling J, Ambrosius W, Heilman D. 1997. Lower respiratory illness in infants and young children with cystic fibrosis: Evaluation of treatment with intravenous hydrocortisone. Pediatric pulmonology. 24(1):48-51.
20. Plummer A, Wildman M, Gleeson T. 2016. Duration of intravenous antibiotic therapy in people with cystic fibrosis. Cochrane Database Syst Rev. 9:CD006682.
21. Collaco JM, Green DM, Cutting GR, Naughton KM, Mogayzel PJ, Jr. 2010. Location and duration of treatment of cystic fibrosis respiratory exacerbations do not affect outcomes. American journal of respiratory and critical care medicine. 182(9):1137-1143.
22. Chmiel JF, Aksamit TR, Chotirmall SH, Dasenbrook EC, Elborn JS, LiPuma JJ, Ranganathan SC, Waters VJ, Ratjen FA. 2014. Antibiotic management of lung infections in cystic fibrosis. I. The microbiome, methicillin-resistant staphylococcus aureus, gram-negative bacteria, and multiple infections. Ann Am Thorac Soc. 11(7):1120-1129.
23. Stokell JR, Gharaibeh RZ, Hamp TJ, Zapata MJ, Fodor AA, Steck TR. 2015. Analysis of changes in diversity and abundance of the microbial community in a cystic fibrosis patient over a multiyear period. Journal of Clinical Microbiology. 53(1):237-247.
24. Janahi IA, Rehman A, Al-Naimi AR. 2017. Allergic bronchopulmonary aspergillosis in patients with cystic fibrosis. Ann Thorac Med. 12(2):74-82.
25. Mussaffi H, Rivlin J, Shalit I, Ephros M, Blau H. 2005. Nontuberculous mycobacteria in cystic fibrosis associated with allergic bronchopulmonary aspergillosis and steroid therapy. Eur Respir J. 25(2):324-328.
26. Olivier KN, Weber DJ, Lee JH, Handler A, Tudor G, Molina PL, Tomashefski J, Knowles MR. 2003. Nontuberculous mycobacteria. Ii: Nested-cohort study of impact on cystic fibrosis lung disease. American journal of respiratory and critical care medicine. 167(6):835-840.
27. Levy I, Grisaru-Soen G, Lerner-Geva L, Kerem E, Blau H, Bentur L, Aviram M, Rivlin J, Picard E, Lavy A et al. 2008. Multicenter cross-sectional study of nontuberculous mycobacterial infections among cystic fibrosis patients, israel. Emerg Infect Dis. 14(3):378-384.
28. Esther CR, Jr., Esserman DA, Gilligan P, Kerr A, Noone PG. 2010. Chronic mycobacterium abscessus infection and lung function decline in cystic fibrosis. Journal of cystic fibrosis : official journal of the European Cystic Fibrosis Society. 9(2):117-123.