To the Editor:
Every year, 10% to 20% of patients with idiopathic pulmonary fibrosis have an acute exacerbation (AE-IPF).1 American Thoracic Society1 guidelines make a weak recommendation for corticosteroid use to treat AE-IPF; however, the risks, benefits, and optimal dose of corticosteroids in AE-IPF remain unclear. Corticosteroid use is common,2 but dosing for AE-IPF can vary widely, from doses similar to COPD exacerbations (eg, 60-120 mg methylprednisolone) to higher pulse doses (eg, 500-1,000 mg methylprednisolone), with varying pharmacokinetic effects.3 Three-fourths of pulmonologists self-report using pulse doses for AE-IPF treatment in multinational surveys,4 but this has not been confirmed by observational studies. Understanding use patterns of pulse dose corticosteroids can help benchmark care and facilitate future studies identifying optimal treatment strategies.
Methods
We used the PINC AI Healthcare Database (2016-2022), an enhanced claims-based dataset capturing approximately 25% of US hospitalizations; hospitals contributing to PINC AI have an overall profile similar to hospitals in the American Hospital Association database. We identified adults aged ≥ 50 years hospitalized from January 2016 to September 2022 with AE-IPF (primary diagnosis of idiopathic pulmonary fibrosis with acute respiratory failure, pneumonia, or ARDS secondary diagnosis or primary diagnosis of acute respiratory failure, pneumonia, or ARDS with idiopathic pulmonary fibrosis secondary diagnosis).5 To increase specificity, we excluded patients with alternate interstitial lung disease or acute respiratory diagnoses, admitted for major surgery, and with prior or admission for lung transplant. To increase reliability and stability of our estimates, we excluded patients transferred from another facility, who did not receive corticosteroids or had unquantifiable steroid doses, and hospitals with ≤ 10 patients meeting inclusion criteria. See e-Table 1 for inclusion and exclusion details.
Corticosteroid doses were converted to methylprednisolone equivalents (ME) for analysis (e-Table 2). Pulse dosing was defined per prior survey data as receipt of ≥ 500 mg ME on a single day.4 Using hierarchical multivariable logistic regression models, we assessed factors associated with and between hospital variation in (1) use of pulse dose corticosteroids at any point in admission and (2) empirical use of pulse dose corticosteroids (ie, within the first 2 days of hospitalization). Models were adjusted for patient-level and hospital-level covariables as fixed effects (e-Table 3) and hospital of admission as a random intercept. If patients had > 1 encounter in the data set, a random hospital admission was chosen for analysis. From each model, we estimated (1) the intraclass correlation coefficient (ICC) variance attributable to between-hospital variance and (2) each hospital’s risk-adjusted outcome rate, with CIs obtained through bootstrapping. An ICC ≥ 15% was a priori considered high variation between hospitals based on prior health services literature.2 In sensitivity analysis, analogous models were built for patients (1) with only IPF as primary hospital diagnosis, (2) admitted to an ICU or stepdown unit, and (3) without sepsis on admission. R (V 4.1.2; R Development Core Team) was used for analysis. Hypothesis tests were 2-sided and alpha = .05. This study was deemed exempt by Boston University institutional review board (No. H-41991).
Results
We identified 6,885 patients meeting inclusion (e-Fig 1) for AE-IPF treated with corticosteroids (median age, 75 years; interquartile range [IQR], 68-81; 4,077 male patients [59.2%]) (Table 1).6,7 The median day of first corticosteroid administration was day 1 (IQR, day 1-2). The median dose of corticosteroids among all patients on the first full day of dosing was 120 mg ME (IQR, 41.6-250 mg). Of these, 1,423 patients (20.7%) received pulse dose corticosteroids at any point during hospitalization (median first pulse day, 2; IQR, days 2-4; median dose on first full day, 516 mg ME; IQR, 500-750 mg); 724 patients (10.5%) received pulse dose corticosteroids within the first 2 days of admission.
Table 1.
Baseline Patient and Hospital Characteristics and Factors Associated With Receipt of Pulse Dose Corticosteroids (N = 6,885)
| Characteristic | Value | Adjusted OR (95% CI) |
|---|---|---|
| Age, y | 75 (68-81) | 0.75 (0.69-0.81) |
| Sex | ||
| Male (reference) | 4,077 (59.2) | NA |
| Female | 2,808 (40.8) | 0.81 (0.71-0.93) |
| Race | ||
| White (reference) | 5,720 (83.1) | NA |
| Black | 454 (6.6) | 0.68 (0.51-0.90) |
| Asian | 143 (2.1) | 0.97 (0.61-1.54) |
| Other | 408 (5.9) | 0.76 (0.55-1.05) |
| Unknown | 160 (2.3) | 1.18 (0.77-1.82) |
| Ethnicity | ||
| Hispanic (reference) | 589 (8.6) | NA |
| Not Hispanic | 5,158 (74.9) | 1.32 (0.97-1.78) |
| Unknown | 1,138 (16.5) | 1.50 (1.06-2.13) |
| Hospital type | ||
| Urban (reference) | 6,239 (90.6) | NA |
| Rural | 646 (9.4) | 0.84 (0.55-1.26) |
| Hospital teaching status | ||
| Teaching hospital (reference) | 3273 (47.5) | NA |
| Nonteaching | 3612 (52.5) | 1.00 (0.75-1.32) |
| Hospital bed size | ||
| ≥ 500 (reference) | 2,542 (36.9) | NA |
| 400-499 | 926 (13.4) | 0.89 (0.61-1.32) |
| 300-399 | 1,422 (20.7) | 0.73 (0.52-1.04) |
| 200-299 | 1,189 (17.3) | 0.78 (0.54-1.12) |
| 100-199 | 661 (9.6) | 0.84 (0.55-1.30) |
| < 100 | 145 (2.1) | 0.75 (0.36-1.58) |
| Hospital region | ||
| South (reference) | 3,601 (52.3) | NA |
| Midwest | 1,442 (20.9) | 1.11 (0.82-1.52) |
| Northeast | 892 (13.0) | 0.77 (0.52-1.12) |
| West | 950 (13.8) | 1.53 (1.07-2.19) |
| Safety-net hospital | 1,487 (21.6) | 0.96 (0.72-1.29) |
| Insurance status | ||
| Private (reference) | 731 (10.6) | NA |
| Medicare | 5,629 (81.8) | 1.15 (0.93-1.45) |
| Medicaid | 296 (4.3) | 0.89 (0.62-1.27) |
| Uninsured, self-pay, charity | 61 (0.9) | 0.77 (0.37-1.58) |
| Othera | 168 (2.4) | 1.33 (0.86-2.05) |
| Admission floor type | ||
| Intensive care unit | 1,060 (15.4) | 1.56 (1.27-1.92) |
| Stepdown unit | 2,149 (31.2) | 1.26 (1.07-1.49) |
| Acute organ dysfunctionb | ||
| Cardiovascular | 382 (5.5) | 0.93 (0.69-1.25) |
| Neurologic | 272 (4.0) | 0.78 (0.54-1.14) |
| Hematologic | 306 (4.4) | 1.20 (0.71-2.03) |
| Hepatic | 14 (0.2) | 0.75 (0.19-2.97) |
| Renal | 944 (13.7) | 1.26 (1.02-1.56) |
| Vasopressors | 123 (1.8) | 0.45 (0.25-0.80) |
| Sepsisb | 1,687 (24.5) | 1.08 (0.90-1.28) |
| Ventilatory support | ||
| Noninvasive | 1,384 (20.1) | 1.28 (1.09-1.51) |
| Invasive | 362 (5.3) | 1.36 (1.01-1.84) |
| Comorbiditiesc | ||
| Alcohol abuse | 122 (1.8) | 0.68 (0.42-1.13) |
| Any tumor | 389 (5.7) | 1.30 (0.99-1.71) |
| Cardiac arrythmia | 2,277 (33.1) | 1.00 (0.87-1.16) |
| Chronic pulmonary disease | 4,693 (68.2) | 0.81 (0.69-0.95) |
| Coagulopathy | 430 (6.2) | 1.28 (0.81-2.01) |
| Complicated diabetes | 1,355 (19.7) | 0.94 (0.79-1.12) |
| Congestive heart failure | 2,735 (39.7) | 0.94 (0.78-1.12) |
| Anemia | 1,055 (15.3) | 1.19 (1.00-1.43) |
| Dementia | 353 (5.1) | 0.69 (0.49-0.98) |
| Fluid and electrolyte disorders | 2,323 (33.7) | 1.07 (0.94-1.23) |
| Hypertension | 2,586 (37.6) | 0.95 (0.77-1.16) |
| Liver disease | 263 (3.8) | 1.02 (0.73-1.41) |
| Metastatic cancer | 121 (1.8) | 0.81 (0.49-1.36) |
| Peripheral vascular disease | 593 (8.6) | 0.84 (0.65-1.07) |
| Psychosis | 1,160 (16.8) | 0.99 (0.82-1.17) |
| Pulmonary circulation disorders | 2,379 (34.6) | 1.20 (1.02-1.40) |
| Renal failure | 1,363 (19.8) | 0.86 (0.69-1.07) |
| Weight loss | 427 (6.2) | 1.05 (0.81-1.37) |
| Obesity | 1,025 (14.9) | 0.99 (0.82-1.19) |
| Do-not-resuscitate order | 1,603 (23.3) | 1.11 (0.95-1.30) |
All characteristics measured on point of admission. Values are presented as No. (%), median (interquartile range), or as otherwise indicated. Bolded results have confidence intervals that do not cross 1. HIPAA = US Health Insurance Portability and Accountability Act; NA = not applicable.
Some race designations have been rolled into "Other" to ensure that the data set conforms to HIPAA and other regulatory requirements.8
As described by Bosch et al.7
As described by Sun et al.6
Patients were more likely to receive a pulse dose of corticosteroids if they were admitted to an ICU or stepdown unit, required invasive or noninvasive ventilation on admission, or had acute renal dysfunction or chronic pulmonary circulation disease at baseline (Table 1). The ICC for between-hospital variation in pulse dose corticosteroids dosing at any point was 17.6% (95% CI, 15.7%-19.4%). The median hospital risk-adjusted probability of receiving pulse dose corticosteroids at any point was 16.1% (IQR, 11.5%-24.7%) (Fig 1). The ICC for between-hospital variation in pulse dose steroid dosing within 2 days of admission was 19.5% (95% CI, 17.6%-21.4%). The median hospital risk-adjusted estimate of the probability of receiving pulse dose corticosteroids within 2 days of admission was 7.5% (IQR, 5.3%-12.2%). Sensitivity analysis among patients with a primary diagnosis of IPF, admitted to an ICU or stepdown unit, and without sepsis on admission yielded similar results to primary analysis (e-Table 4).
Figure 1.
A, B, Risk-adjusted predicted probability of outcomes by hospital of admission. Predicted rates of receiving a pulse dose steroid at each hospital at (A) any point in hospitalization (green dots) and (B) empirically within the first 2 d of admission (blue dots) were ranked from lowest to highest and plotted. Each point along the x-axis represents a hospital in the data set with the predicted probability of using a pulse dose calculated from adjusted hierarchical multivariable logistic regression models plotted along the y-axis. Error bars (gray) indicate the 95% CI for each point estimate and the horizontal red dotted lines represent the overall median risk-adjusted estimate for the probability of receiving pulse dose corticosteroids among all hospitals.
Discussion
We examined variation in use of pulse dose corticosteroids for patients with AE-IPF in the United States from 2016 to 2022. Only one-fifth of patients received pulse dose corticosteroids, with high between-hospital variation in use of pulse dose corticosteroids at any point in admission. This contrasts with multinational survey data of pulmonologists,4 which may be due to our database capturing practices encompassing nonpulmonary providers in a geographically and socioeconomically diverse US population.
There are limited data available to guide the optimal dose of corticosteroids for AE-IPF; most studies are retrospective, small, single-center studies that have yielded contrasting findings, from suggestions of harm to benefit to no difference with higher-dose vs lower-dose corticosteroids.9, 10, 11 Experts have noted that the management of AE-IPF remains an area of major unmet medical need.1 Our results highlight the high degree of clinical uncertainty in practice, and our findings characterizing usual practice can be leveraged to design future pragmatic randomized trials and/or observational target trial emulations focused on corticosteroid efficacy and effectiveness.
Our study has limitations. As in studies using administrative data, misclassification of AE-IPF is possible. Given this risk, we followed validated algorithms for identifying AE-IPF and comorbidities5, 6, 7 and further excluded confounding conditions both from a diagnostic and treatment standpoint (heart failure, asthma, COPD).1 We additionally performed sensitivity analyses to address the possibility of misclassification of AE-IPF with broader definitions and found similar results. Furthermore, we could not accurately capture the type of provider to assess whether provider specialty may underlie practice variation.
Conclusions
In a cohort of US patients hospitalized with AE-IPF, we observed high variation in use of pulse dose corticosteroids, highlighting the need for additional data to guide practice. Characterizing current practices is the first step to benchmarking care and designing future studies to identify optimal treatment strategies for patients with AE-IPF.
Funding/Suppport
D. A. S. was supported by the National Heart, Lung, and Blood Institute (Grant 1F32HL168959-01).
Financial/Nonfinancial Disclosures
None declared.
Acknowledgments
Author contributions: D. A. S., F. J. H., K.-D. A., K. C. W., A. J. W., N. A. B., and A. C. L. were involved in the planning of this study and in writing the manuscript for this study. D. A. S., A. J. W., N. A. B., and A. C. L. were involved in the data collection and data analysis of this study. D. A. S. submitted the study.
Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or the preparation of the manuscript.
Additional information: The e-Figure and e-Tables are available online under “Supplementary Data.”
Supplementary Data
References
- 1.Collard H.R., Ryerson C.J., Corte T.J., et al. Acute exacerbation of idiopathic pulmonary fibrosis. An International Working Group Report. Am J Respir Crit Care Med. 2016;194(3):265–275. doi: 10.1164/rccm.201604-0801CI. [DOI] [PubMed] [Google Scholar]
- 2.Shankar D.A., Walkey A.J., Hawkins F.J., Bosch N.A., Peterson D., Law A.C. Hospital-level variation in practices and outcomes for patients with severe acute exacerbations of idiopathic pulmonary fibrosis: a retrospective multicentre cohort study. BMJ Open Respir Res. 2023;10(1) doi: 10.1136/bmjresp-2022-001593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Edel Y., Avni T., Shepshelovich D., et al. The safety of pulse corticosteroid therapy- systematic review and meta-analysis. Semin Arthritis Rheum. 2020;50(3):534–545. doi: 10.1016/j.semarthrit.2019.11.006. [DOI] [PubMed] [Google Scholar]
- 4.Kreuter M., Polke M., Walsh S.L.F., et al. Acute exacerbation of idiopathic pulmonary fibrosis: international survey and call for harmonisation. Eur Respir J. 2020;55(4) doi: 10.1183/13993003.01760-2019. [DOI] [PubMed] [Google Scholar]
- 5.Herberts M.B., Teague T.T., Thao V., et al. Idiopathic pulmonary fibrosis in the United States: time to diagnosis and treatment. BMC Pulm Med. 2023;23(1):1–10. doi: 10.1186/s12890-023-02565-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sun J.W., Rogers J.R., Her Q., et al. Adaptation and validation of the combined comorbidity score for ICD-10-CM. Med Care. 2017;55(12):1046–1051. doi: 10.1097/MLR.0000000000000824. [DOI] [PubMed] [Google Scholar]
- 7.Bosch N.A., Law A.C., Rucci J.M., Peterson D., Walkey A.J. Predictive validity of the sequential organ failure assessment score versus claims-based scores among critically ill patients. Ann Am Thorac Soc. 2022;19(6):1072–1076. doi: 10.1513/AnnalsATS.202111-1251RL. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.PINC AI PHD Data Dictionary V2.2-A. PINC AI Healthcare. February 15, 2024. [Google Scholar]
- 9.Farrand E., Vittinghoff E., Ley B., Butte A.J., Collard H.R. Corticosteroid use is not associated with improved outcomes in acute exacerbation of IPF. Respirology. 2020;25(6):629–635. doi: 10.1111/resp.13753. [DOI] [PubMed] [Google Scholar]
- 10.Arai T., Tachibana K., Sugimoto C., et al. High-dose prednisolone after intravenous methylprednisolone improves prognosis of acute exacerbation in idiopathic interstitial pneumonias. Respirology. 2017;22(7):1363–1370. doi: 10.1111/resp.13065. [DOI] [PubMed] [Google Scholar]
- 11.Hyung K., Lee J.H., Kim J.-Y., Choi S.M., Park J. Pulse versus non-pulse corticosteroid therapy in patients with acute exacerbation of idiopathic pulmonary fibrosis. Respirology. 2024;29(3):235–242. doi: 10.1111/resp.14643. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.