Efficacy of initial high- versus low-dose intravenous corticosteroid therapy in patients with acute exacerbation of idiopathic interstitial pneumonia: A nationwide observational study

ABSTRACT

BACKGROUND

Acute exacerbation of idiopathic interstitial pneumonias (AE-IIPs) has a high mortality. However, there is no established treatment for AE-IIPs. Therefore, we aimed to compare the efficacy of high- and low-dose corticosteroid therapies in AE-IIPs patients.

METHODS

Data were retrospectively collected from the Japanese Diagnosis Procedure Combination database from July 2010 to March 2018. Adult patients with AE-IIPs who received high-dose (methylprednisolone at a dose of 500–1000 mg/day for 3 days starting within 4 days after admission) or low-dose (methylprednisolone at a dose of 100–200 mg/day for at least 5 days starting within 4 days after admission) corticosteroid therapy were identified. Eligible patients (n = 17,317) were divided into the high-dose (n = 16,998) and low-dose (n = 319) groups. A stabilized inverse probability of treatment weighting using propensity scores was performed to compare outcomes between the groups.

RESULTS

The primary outcome was in-hospital mortality, and the secondary outcomes were 28-day mortality, infections during hospitalization, length of hospitalization, duration of steroid use, and discharge to home. The in-hospital mortality rates of the high- and low-dose corticosteroid groups were 50.6% and 47.0%, respectively. In-hospital mortality did not significantly differ between the two groups after stabilized inverse probability of treatment weighting, and the odds ratio in the low-dose corticosteroid group was 0.86 (95% confidence interval: 0.64–1.16; p = 0.33). The secondary outcomes also did not significantly differ between the groups.

CONCLUSIONS

There was no significant difference in outcomes between patients with AE-IIPs who received high- and low-dose corticosteroid therapies.

INTRODUCTION

Idiopathic pulmonary fibrosis (IPF) is an interstitial lung disease that is characterized by chronic fibrosis and has a poor prognosis, with an average survival time of 3–4 years [1]. A previous study showed that acute exacerbation (AE) of IPF was the leading cause of death among patients with IPF and that it was correlated with a high mortality, with a mean survival time of <1 year and a 90-day mortality rate of approximately 50% after AE of IPF [2].

Interstitial pneumonia of unknown causes is called idiopathic interstitial pneumonias (IIPs) and is classified into major IIPs, rare IIPs, and unclassifiable IIPs [3]. IPF is the most frequent among IIPs and is categorized as major IIPs [3]. Patients with IIPs other than IPF, including non-specific interstitial pneumonia and unclassifiable interstitial lung disease, can also develop a progressive phenotype during the clinical course [3–9]. So far, there is no official definition of AE of IIPs (AE-IIPs). However, as clinical, radiological, and pathological findings are similar to those of AE of IPF, rapid progressive deterioration of respiratory condition in patients with IIPs is considered to be AE [3, 6–9]. A large observational study by Suzuki et al. showed that short-term mortality in AE-IIPs was as poor as that in AE of IPF (90-day mortality 38% vs. 47%) [6]. Other studies have also reported poor prognosis in various types of IIPs developing AE, similar to the recognized poor prognosis in AE of IPF [4, 5, 7–9].

There is no established treatment for AE of IPF. Based on the Japanese guidelines for the treatment of IPF, the therapeutic options include immunosuppressive agents and corticosteroids such as high-dose steroids (pulse therapy), but none have been proven to be effective based on a randomized controlled trial [10]. In addition, the international evidence-based guidelines on the management of IPF had a weak recommendation for the use of corticosteroids, including methylprednisolone pulse therapy, for AE of IPF [1, 11]. By contrast, as there are no treatment guidelines for AE-IIPs, the treatment of AE-IIPs is often performed in accordance with that of AE of IPF [6–9]. Thus, corticosteroids are often administered to patients with AE-IIPs in daily practice. In Japan, high-dose corticosteroid therapy such as methylprednisolone 500–1000 mg/day for 3 days and subsequent corticosteroid treatment maintained at 0.5–1 mg/kg/day are commonly used for the empirical treatment of AE-IIPs [5–8, 12, 13].

Conversely, long-term low-dose corticosteroid therapy has improved survival in patients with acute respiratory distress syndrome (ARDS), a condition similar to AE-IIPs [14, 15]. Based on this result, low-dose corticosteroid therapy is occasionally administered to patients with AE of IPF [16]. However, to date, its efficacy has also not been validated. In addition, numerous side effects including infections have been correlated with corticosteroids. Thus, they should be used with caution particularly if their efficacy has not been proven yet.

The present study aimed to compare the efficacy between high- and low-dose corticosteroid therapies in patients with AE-IIPs using data collected from a Japanese national inpatient database. In addition, the incidence of nosocomial infections, length of hospital stay, duration of steroid use, and discharge to home were compared between the two treatment groups.

METHODS

DATA SOURCE

Inpatient data were extracted from the Japanese Diagnosis Procedure Combination database, which is a nationwide administrative claims database with discharge abstracts. Other details of the database have been reported elsewhere [17, 18]. More than 1,000 hospitals voluntarily contribute to the database, representing approximately 50% of all discharges from acute care hospitals in Japan. We collected data including those of sex and age; dates of hospitalization and discharge; weight and height; severity of dyspnea based on the Hugh–Jones dyspnea scale [19]; level of consciousness on admission; smoking index; activities of daily living (ADL); intensive care unit (ICU) or emergency ward admission during hospitalization; main diagnoses, pre-existing comorbidities on admission and complications after hospitalization as recorded by the attending physicians with the International Classification of Diseases, 10th revision (ICD-10) codes accompanied by text in Japanese; procedures and their dates; dates and doses of drugs administered during hospitalization; and discharge status.

ETHICS

The Institutional Review Board of the University of Tokyo approved this study (Approval number: 3501; May 26, 2017) and also provided waiver for informed consent because anonymized data were used.

PATIENT SELECTION

This study used data collected from July 1, 2010 to March 31, 2018. The inclusion criteria were patients aged ≥15 years, those diagnosed with interstitial pneumonia (ICD-10 codes J84.1, J84.8 and J84.9), those who underwent computed tomography within 1 day after admission and those who received treatment with intravenous methylprednisolone at a dose of 500–1000 mg/day for 3 days, which was started within 4 days after admission [20], or 100–200 mg/day for at least 5 days, which was started within 4 days after admission (Fig. 1) [15, 21]. Patients with IIPs were selected as follows. First, we excluded patients with the following secondary interstitial lung diseases identified using ICD-10 codes: hypersensitivity pneumonitis (J67), connective tissue disease associated with interstitial lung disease (M05, M06 and M30–35), sarcoidosis (D86), amyloidosis (E85), drug-induced lung disease (J70), radiation pneumonitis (J70), Pneumocystis jirovecii pneumonia (B59), pneumoconiosis (J60–65), pulmonary alveolar proteinosis (J84.0), eosinophilic pneumonia (J82), Langerhans cell histiocytosis (C96) and lymphangioleiomyomatosis (D21.9). We then excluded patients who received medications including furosemide, azosemide, carperitide, landiolol hydrochloride, digoxin, deslanoside and tolvaptan for acute heart failure within 1 day after admission [20] and those who received intra-aortic balloon pump therapy during hospitalization. The remaining patients were assumed to have AE-IIPs. We divided the patients who received intravenous methylprednisolone at a dose of 500–1000 mg/day for 3 days starting within 4 days after admission into the high-dose corticosteroid group and those who received intravenous methylprednisolone at a dose of 100–200 mg/day for at least 5 days starting within 4 days after admission into the low-dose corticosteroid group. The dosage of corticosteroids was determined as follows: the former was the most commonly used dose of methylprednisolone pulse therapy [5, 6, 10, 20], and the latter was calculated with reference to the prolonged low-dose methylprednisolone treatment for ARDS reported by Meduri et al. [14, 15, 21] and the ideal body weight of Japanese patients. According to their studies, the dosage of corticosteroids, especially for patients with unresolving ARDS, was methylprednisolone at a dose of 2 mg/kg/day. Ideal body weight was 55–63 kg, calculated from the average height of Japanese. Thus, intravenous methylprednisolone at a dose of 100–200 mg/day was defined as low-dose corticosteroid therapy. The dosage of corticosteroids was defined by the amount of drug ordered, not administered.

Fig. 1 High- and low-dose corticosteroid therapies

We defined the high-dose corticosteroid therapy as 500–1000 mg/day of intravenous methylprednisolone for 3 days (starting within 4 days after admission) and low-dose corticosteroid therapy as 100–200 mg/day of intravenous methylprednisolone for at least 5 days (starting within 4 days after admission).

Two meta-analyses have reported that administration of low-dose corticosteroids improved the prognosis of patients with sepsis [22, 23]. Hence, to accurately evaluate the effect of corticosteroids against AE-IIPs, we excluded patients with sepsis (ICD-10 codes A40 and A41). However, we included patients with pulmonary infections without sepsis. Moreover, patients with missing data on treatment year, hospitalization procedure, or level of consciousness were excluded, whereas patients with missing data on age, BMI, Hugh-Jones dyspnea score, smoking index, and ADL were not excluded and were included as missing categories in each variable. Finally, those who died within 6 days after admission were excluded to prevent immortal time bias.

PATIENT CHARACTERISTICS

The patient characteristics evaluated in this study were sex, age, treatment year, body mass index (BMI), Hugh–Jones dyspnea scores on admission, level of consciousness on admission, comorbidities, Charlson comorbidity index [24, 25], smoking index, ADL scale (Barthel index) on admission, and hospitalization procedure. We also examined procedures and treatments, including mechanical ventilation, continuous renal replacement therapy, high-flow nasal cannula oxygen therapy, transfusion, nasal tube feeding, and use of medications for IIPs within 3 days after admission. The level of consciousness on admission was evaluated using the Japan Coma Scale [26, 27], which is widely used in Japan and has been well correlated with the Glasgow Coma Scale [28]. The following comorbidities were identified using ICD-10 codes (Table S1): bronchial asthma, chronic obstructive pulmonary disease (COPD), pneumonia, mycotic infection, pulmonary embolism, bronchiectasis, pneumothorax, cor pulmonale, lung and other types of cancer, disseminated intravascular coagulation, chronic heart failure, tachycardia, acute coronary syndrome, diabetes mellitus, stroke, dementia, renal failure, liver dysfunction, and gastroesophageal reflux disease. The Charlson comorbidity index scores were classified as follows: 0, 1, 2 and ≥3.

OUTCOME

The primary outcome was all-cause in-hospital mortality. The secondary outcomes were 28-day mortality (in-hospital 28-days all cause of mortality), infections during hospitalization, length of hospital stay, duration of steroid use, and discharge to home. The following infections were assessed using ICD-10 codes: ventilator-associated pneumonia (J958), pneumonia (A481, J100, J110, J12, J13, J14, J15, J16, J170, J178, J18, J85, J86), sinusitis (J014, J019, J324, J329), catheter-related infection (T814, T827), urinary tract infection (N390, T835), peritonitis (K65), bacteremia (A39, A49), candidemia (B377), Clostridium difficile colitis (A047), acalculous cholecystitis (K810) and wound infection (T793, T941) [29].

STATISTICAL ANALYSIS

Dichotomous and categorical variables were presented as numbers with percentages and continuous variables as median and interquartile range.

Stabilized inverse probability of treatment weighting (IPTW) analyses using propensity scores were performed to compare the outcomes between the high- and low-dose corticosteroid groups to account for the differences in baseline characteristics, including comorbidities and treatments. Stabilized IPTW uses propensity scores and adjusts for measured potential confounding while preserving sample size [30, 31]. To control for imbalance in covariates, the specific stabilized weights were generated using propensity scores, which predict the probability of receiving high-dose corticosteroid therapy. To estimate the propensity score, a logistic regression model for receiving high-dose corticosteroid therapy was used with the following independent variables: sex, age, treatment year, BMI, Hugh–Jones dyspnea scale score, level of consciousness on admission, Charlson comorbidity index, smoking index, Barthel index on admission, frequency of hospitalization before admission, academic hospital, unscheduled hospitalization, emergency, or ICU hospitalization within 3 days after hospitalization, comorbidities and procedures (mechanical ventilation, hemodialysis, high-flow nasal cannula oxygen therapy, transfusion and enteral tube feeding) and drugs for AE-IIPs (noradrenaline, azithromycin, cyclophosphamide, cyclosporin, tacrolimus, azathioprine, pirfenidone, nintedanib, sivelestat sodium hydrate, and thrombomodulin alfa). The balance of covariates was assessed using a standardized mean difference; <0.15 indicated acceptable balancing of covariates between the two groups. Stabilized IPTW analysis can preserve sample size and appropriately estimate average treatment effects over the marginal distribution of measured covariates in a study cohort.

We used generalized linear models with cluster-robust standard errors treating each hospital as a cluster for comparisons of the primary and secondary outcomes. A logistic regression analysis of in-hospital mortality, 28-day mortality, and infections during hospitalization was conducted. Then, odds ratios and their 95% confidence intervals (CIs) were calculated. The length of hospital stay and duration of steroid use between the two groups were compared via Poisson regression analyses, and the incidence rate ratios and their 95% CIs were calculated. To address the issue of competing outcomes, infections during hospitalization, length of hospital stay, and duration of steroid use were evaluated among the survivors alone and all patients. Discharge to home was compared between the high- and low-dose corticosteroid groups after stabilized IPTW using Cox regression analysis, and hazard ratio (HR) and 95% CI were calculated.

We performed a sensitivity analysis 1 excluding patients with COPD (ICD-10 codes J44) or bronchial asthma (ICD-10 codes J45 and J46). Furthermore, we performed a sensitivity analysis 2 excluding patients with cryptogenic organizing pneumonia (COP) using the diagnoses in Japanese because COP is more responsive to corticosteroids than other IIPs [3]. Odds ratios, incidence rate ratios, and their corresponding 95% CIs were calculated.

A two-tailed significance level of 0.05 was used in all statistical analyses. All tests were performed using STATA/MP version 16 software (STATA Corp., College Station, TX, USA).

RESULTS

The procedure of patient selection is depicted in Fig. 2. During the study period, 33,922 patients underwent computed tomography within 1 day after admission and received high- or low-dose corticosteroid therapy within 4 days. Among them, 17,317 were eligible for this study. The patients were divided into the high-dose (n = 16,998) and low-dose (n = 319) corticosteroid groups.

Fig. 2 Flow chart of patient selection

IP, interstitial pneumonia; CT, computed tomography; IIPs, idiopathic interstitial pneumonia; COPD, chronic obstructive pulmonary disease; COP, cryptogenic organizing pneumonia

Tables 1 and 2 show the baseline characteristics of the patients, comorbidities, and treatments before and after stabilized IPTW using propensity scores. The proportion of patients older than 80 years was higher in the low-dose than in the high-dose corticosteroid group, and the Charlson comorbidity index was not balanced between the two groups. The proportion of patients with bronchial asthma and COPD was higher in the low-dose than in the high-dose corticosteroid group. Meanwhile, a higher proportion of patients received azithromycin, sivelestat sodium hydrate, and mechanical ventilation in the high-dose than in the low-dose corticosteroid group. After stabilized IPTW using propensity scores, the baseline characteristics of the patients were well balanced between the two groups.

Table 1 Baseline characteristics of patients before and after stabilized IPTW using propensity scores

Characteristics	All patients					Patients after IPTW estimation
	High-dose corticosteroid group (n = 16998)	%	Low-dose corticosteroid group (n = 319)	%	SMD	High-dose corticosteroid group (n = 16998)	%	Low-dose corticosteroid group (n = 311)	%	SMD
Male sex	11918	70.1	226	70.8	1.6	11921	70.1	223	71.7	2.3
Age, years
15–70	4758	28.0	82	25.7	−5.2	4751	28.0	84	27.0	−2.6
71–80	6949	41.0	113	35.4	−11.3	6932	40.8	141	45.4	8.7
≥80	5165	30.4	121	37.9	16.0	5188	30.5	86	27.7	−6.5
Missing	126	0.7	3	0.9	2.2	127	0.7	2	0.6	−2.4
Treatment year
2010–2011	2715	16.0	42	13.2	−8.0	2707	15.9	52	16.7	1.8
2012–2013	3755	22.0	62	19.4	−6.5	3747	22.0	66	21.3	−2.1
2014–2015	4758	28.0	92	28.8	1.9	4761	28.0	82	26.4	−4.1
2016–2017	5770	34.0	123	38.6	9.6	5785	34.0	113	36.3	4.3
BMI (kg/m2)
<23	8863	52.1	176	55.2	6.1	8873	52.2	163	52.4	−0.4
23–25	3074	18.1	52	16.3	−4.7	3069	18.1	62	19.8	4.1
≥25	3202	18.8	63	19.7	2.3	3205	18.9	57	18.3	−1.7
Missing	1859	10.9	28	8.8	−7.2	1852	10.9	32	10.2	−2.4
Hugh–Jones dyspnea score on admission
1–2	2207	13.0	41	12.9	−0.4	2207	13.0	37	12.0	−3.3
3	1393	8.2	26	8.2	−0.2	1393	8.2	27	8.5	1.0
4	3112	18.3	58	18.2	−0.3	3111	18.3	68	21.9	8.7
5	6720	39.5	122	38.2	−2.6	6716	39.5	104	33.5	−13.0
Missing	3566	21.0	72	22.6	3.9	3572	21.0	77	24.8	8.5
Japan Coma Scale score on admission
0- or 1-digit (alert or dull)	16559	97.4	307	96.2	−6.7	16556	97.4	305	98.1	0.4
2-digit (somnolence)	246	1.5	3	0.9	−4.7	244	1.4	4	1.2	−1.9
3-digit (coma)	193	1.1	9	2.8	12.1	198	1.2	4	1.3	1.5
Charlson comorbidity index
0	8839	52.0	114	35.7	−33.2	8788	51.7	169	54.4	4.7
1	2212	13.0	98	30.7	43.9	2268	13.3	37	12.1	−4.1
2	3651	21.5	56	17.6	−9.9	3639	21.4	60	19.4	−5.4
≥3	2296	13.5	51	16.0	7.0	2304	13.6	46	14.9	3.5
Smoking index, pack-years
0	7631	44.9	144	45.1	0.5	7632	44.9	140	45.0	−0.5
1–39	3261	19.2	52	16.3	−7.6	3252	19.1	60	19.3	0.1
≥40	4043	23.8	90	28.2	10.1	4058	23.9	74	23.9	−0.2
Missing	2063	12.1	33	10.3	−5.7	2057	12.1	39	12.5	0.8
ADL on admission (Barthel index score)
100	4904	28.9	96	30.1	2.7	4908	28.9	93	29.8	1.5
≤95	8991	52.9	185	58.0	10.3	9007	53.0	170	54.6	2.4
Missing	3103	18.3	38	11.9	−17.8	3083	18.1	51	16.3	−5.0
Frequency of hospitalization before admission
0	9773	57.5	178	55.8	−3.4	9768	57.5	166	53.5	−8.7
1–2	5215	30.7	91	28.5	−4.7	5209	30.6	108	34.6	7.9
≥3	2010	11.8	50	15.7	11.2	2022	11.9	39	12.6	1.9
Academic hospital	13717	80.7	236	74.0	−16.1	13696	80.6	252	81.0	−0.2
Unscheduled hospitalization	13564	79.8	261	81.8	5.1	13570	79.8	244	78.4	−4.8
Emergency ward admission	1863	11.0	20	6.3	−16.8	1848	10.9	36	11.6	2.3
ICU admission	954	5.6	15	4.7	−4.1	951	5.6	25	7.9	10.1

Data are presented as n (%).

BMI, body mass index; ADL, activities of daily living; ICU, intensive care unit; IPTW, inverse probability of treatment weighting; SMD, standardized mean difference

Table 2 Comorbidities and treatments before and after stabilized IPTW using propensity scores

Variables	All patients					Patients after IPTW estimation
	High-dose corticosteroid group (n = 16998)	%	Low-dose corticosteroid group (n = 319)	%	SMD	High-dose corticosteroid group (n = 16998)	%	Low-dose corticosteroid group (n = 311)	%	SMD
Comorbidity
Bronchial asthma	1011	5.9	54	16.9	35.0	1046	6.2	19	6.1	−0.4
COPD	825	4.9	53	16.6	38.7	863	5.1	17	5.3	0.8
Pneumonia	2754	16.2	58	18.2	5.2	2760	16.2	54	17.3	2.4
Mycotic infection	196	1.2	5	1.6	3.6	197	1.2	4	1.3	0.9
Pulmonary embolism	92	0.5	1	0.3	−3.5	91	0.5	0	0.1	−6.2
Bronchiectasis	708	4.2	16	5.0	4.1	710	4.2	9	2.8	−6.7
Pneumothorax	113	0.7	6	1.9	10.9	117	0.7	2	0.5	−1.7
Cor pulmonale	155	0.9	5	1.6	5.9	157	0.9	5	1.6	6.3
Lung cancer	2019	11.9	41	12.9	3.0	2022	11.9	39	12.6	1.9
Other types of cancer*	1615	9.5	26	8.2	−4.8	1611	9.5	32	10.2	2.3
Disseminated intravascular coagulation	1073	6.3	20	6.3	−0.2	1073	6.3	20	6.3	−0.2
Chronic heart failure	1780	10.5	37	11.6	3.6	1784	10.5	31	9.9	−2.1
Tachycardia	814	4.8	14	4.4	−1.9	813	4.9	22	7.1	10.9
Acute coronary syndrome	1110	6.5	25	7.8	5.1	1114	6.6	23	7.5	3.6
Diabetes mellitus	4272	25.1	88	27.6	5.6	4280	25.2	93	29.8	10.0
Stroke	898	5.3	22	6.9	6.7	903	5.3	17	5.4	0
Dementia	432	2.5	9	2.8	1.7	433	2.5	10	3.2	3.8
Renal failure	933	5.5	15	4.7	−3.6	931	5.5	20	6.4	3.8
Liver dysfunction	790	4.6	18	5.6	4.5	793	4.7	15	4.8	0.4
Gastroesophageal reflux disease	3411	20.1	47	14.7	−14.1	3394	20.0	56	18.1	−5.2
Treatment within 3 days after hospitalization
Noradrenaline	540	3.2	16	5.0	9.3	546	3.2	19	6.0	13.9
Azithromycin	2313	13.6	23	7.2	−21.1	2293	13.5	36	11.6	−6.4
Cyclophosphamide (intravenous)	213	1.3	1	0.3	−10.7	210	1.2	3	1.1	−2.1
Cyclosporin	470	2.8	5	1.6	−8.2	466	2.7	11	3.5	4.9
Tacrolimus	83	0.5	3	0.9	5.4	84	0.5	1	0.3	−2.7
Azathioprine	81	0.5	2	0.6	2.0	81	0.5	1	0.3	−2.3
Pirfenidone	297	1.7	4	1.3	−4.1	296	1.7	6	1.9	1.3
Nintedanib	112	0.7	1	0.3	−5.0	111	0.7	1	0.3	−5.0
Sivelestat sodium hydrate	1545	9.1	9	2.8	−26.7	1525	9.0	26	8.3	−3.1
Thrombomodulin alfa	430	2.5	6	1.9	−4.4	428	2.5	5	1.7	−5.5
Mechanical ventilation	2823	16.6	36	11.3	−15.4	2807	16.5	57	18.4	5.2
Hemodialysis	232	1.4	2	0.6	−7.4	230	1.4	5	1.5	1.6
High-flow nasal cannula oxygen therapy	578	3.4	7	2.2	−7.3	574	3.4	9	2.9	−2.7
Transfusion	325	1.9	13	4.1	12.7	332	2.0	8	2.6	3.8
Enteral tube feeding	542	3.2	9	2.8	−2.1	541	3.2	16	5.2	11.6

Data are presented as n (%).

COPD, chronic obstructive pulmonary disease; IPTW, inverse probability of treatment weighting; SMD, standardized mean difference

* Detailed information in Table S1.

Table 3 presents the outcomes before and after stabilized IPTW. The in-hospital mortality rates of the high- and low-dose corticosteroid groups were 50.6% (8593/16998) and 47.0% (150/319), respectively. Table 4 shows the comparison of outcomes between the high- and low-dose corticosteroid groups before and after stabilized IPTW. In-hospital mortality did not significantly differ between the two groups after stabilized IPTW, and the odds ratio in the low-dose corticosteroid group was 0.86 (95% CI: 0.64–1.16; p = 0.33). Similarly, after stabilized IPTW, the odds ratio of 28-day mortality in the low-dose corticosteroid group was 0.88 (95% CI: 0.64–1.23; p = 0.46).

Table 3 Outcomes in the high- and low-dose corticosteroid groups before and after stabilized IPTW

	Patients before IPTW estimation		Patients after IPTW estimation
	High-dose corticosteroid group	Low-dose corticosteroid group	High-dose corticosteroid group	Low-dose corticosteroid group
All patients, (n)	16998	319	16998	311
In-hospital mortality, n (%)	8593 (50.6)	150 (47.0)	8593 (50.6)	147 (47.3)
28-day mortality, n (%)	5468 (35.2)	87 (27.3)	5463 (32.1)	92 (29.6)
Infection during hospitalization, n (%)	1018 (6.0)	25 (7.8)	1018 (6.0)	31 (10.0)
Length of hospital stay (days), median (IQR)	26 (16–42)	25 (11–45)	26 (16–42)	25 (16–42)
Duration of steroid use (days), median (IQR)*	23 (13–40)	21 (12–40)	23 (13–40)	20 (12–39)
Discharge to home, n (%)	6169 (36.3)	121 (37.9)	6168 (36.3)	117 (37.6)
Survivors, (n)	8405	169	8405	166
Infection during hospitalization, n (%)	325 (3.9)	10 (5.9)	325 (3.9)	11 (6.6)
Length of hospital stay (days), median (IQR)	30 (19–46)	25 (15–45)	30 (19–46)	27 (16–47)
Duration of steroid use (days), median (IQR)*	28 (16–43)	21 (9–43)	28 (16–43)	21 (10–44)

IPTW, inverse probability of treatment weighting; IQR, interquartile range

* Steroids include methylprednisolone and prednisolone.

Table 4 Comparison of outcomes between the high- and low-dose corticosteroid groups before and after stabilized IPTW

Logistic regression analyses for patients in the low-dose corticosteroid group compared with the high-dose corticosteroid group
	Before IPTW estimation			After IPTW estimation
	Odds ratio	95% CI	p value	Odds ratio	95% CI	p value
All patients
In-hospital mortality	0.87	0.68–1.10	0.25	0.86	0.64–1.16	0.33
28-day mortality	0.79	0.62–1.02	0.066	0.88	0.64–1.23	0.46
Infection during hospitalization	1.33	0.90–1.97	0.15	1.70	0.99–2.94	0.056
Survivors
Infection during hospitalization	1.56	0.81–3.00	0.18	1.79	0.68–4.70	0.23
Incidence rate ratios of length of hospital stay and duration of steroid use in low-dose corticosteroid group compared with high-dose corticosteroid group
	Before IPTW estimation			After IPTW estimation
	Incidence rate ratio	95% CI	p value	Incidence rate ratio	95% CI	p value
All patients
Length of hospital stay	1.00	0.90–1.11	0.99	1.00	0.85–1.17	0.99
Duration of steroid use*	0.95	0.84–1.07	0.37	0.97	0.81–1.16	0.73
Survivors
Length of hospital stay	0.97	0.84–1.13	0.73	1.04	0.82–1.34	0.73
Duration of steroid use*	0.92	0.77–1.09	0.31	1.02	0.77–1.34	0.91

IPTW, inverse probability of treatment weighting; CI, confidence interval

* Steroids include methylprednisolone and prednisolone.

The proportions of patients with infections during hospitalization were similar between the groups after stabilized IPTW, and the odds ratio in the low-dose corticosteroid group was 1.70 (95% CI: 0.99–2.94; p = 0.056) compared with that in the high-dose corticosteroid group. In the low-dose corticosteroid group, the incidence rate ratios of length of hospital stay and duration of steroid use were 1.00 (95% CI: 0.85–1.17; p = 0.99) and 0.97 (95% CI: 0.81–1.16; p = 0.73), respectively, compared with those in the high-dose corticosteroid group after stabilized IPTW. Moreover, there were no significant differences in terms of these secondary outcomes between the survivors alone of the two groups after stabilized IPTW. Cox regression analysis showed that there was no significant difference in discharge to home between the two groups after stabilized IPTW (HR: 1.04; 95% CI: 0.87–1.24).

The results of the sensitivity analysis 1 restricted to patients without COPD or bronchial asthma were compatible with those of the main analyses (Table S2–S5). Moreover, the results of the sensitivity analysis 2, which excluded patients with COP, were identical to those of the main analyses.

DISCUSSION

Using data from a nationwide inpatient database in Japan, we examined the efficacy of high- versus low-dose corticosteroid therapy in patients with AE-IIPs. The results revealed no significant difference in terms of in-hospital mortality between the high- and low-dose corticosteroid groups. Similarly, there was no significant difference in terms of 28-day mortality, frequency of infections during hospitalization, length of hospital stay, or duration of steroid treatment between the two groups. In addition, Cox regression analysis revealed that there was no significant difference in discharge to home between the two groups. The results of the sensitivity analyses supported these findings.

AE-IIPs are a potentially fatal condition that is characterized by rapid progression of respiratory failure, and its pathological feature is diffuse alveolar damage [1, 7, 8]. The pathogenesis of ARDS is similar to that of AE-IIPs, and its pathological feature is also diffuse alveolar damage [32]. In addition, in both AE-IIPs and ARDS, the levels of tumor necrosis factor-α, interleukin-1β, and interleukin-8 levels are elevated in the serum and bronchoalveolar lavage fluid [21, 33–35]. Corticosteroids have an anti-inflammatory effect, and they can reduce the levels of these cytokines [11, 34]. In fact, several trials have shown that low-dose corticosteroid therapy is associated with low mortality rates among patients with ARDS [14, 15]. In contrast, the use of short-term high-dose corticosteroid therapy for ARDS is not recommended because of the high incidence of adverse events and mortality [36–38]. In the present study, we compared the high- and low-dose corticosteroid therapies in patients with AE-IIPs. In contrast to patients with ARDS, there were no differences in efficacy or side effects depending on the dose of corticosteroid administered to patients with AE-IIPs. In Japan, patients with AE-IIPs are mostly treated with high-dose corticosteroid therapy. However, low-dose corticosteroid therapy could be a potential treatment option when the corticosteroid dose should be reduced as much as possible due to concerns about the side effects. In contrast, since methylprednisolone has non-genomic effects, high-dose corticosteroid therapy beyond nuclear receptor saturation may be expected to have a greater effect [39]. In clinical practice, there are many situations where high-dose corticosteroid therapy has to be administered despite side-effects because AE-IIPs are life-threatening,

In Japan, patients with AE-IIPs are frequently treated empirically with high-dose corticosteroid therapy followed by prednisolone therapy [5, 6]. In fact, the number of patients in the high-dose corticosteroid group was much larger than that in the low-dose corticosteroid group. One of the reasons for this high number was thought to be that high-dose corticosteroid therapy was referred to as the treatment for AE of IPF in the Japanese and international guidelines [10, 11], whereas, there is also a report showing the efficacy of low-dose prolonged methylprednisolone therapy for such patients [16]. In the present study, there was no significant difference in terms of in-hospital mortality between the high- and low-dose corticosteroid groups (the odds ratio in the low-dose corticosteroid group was 0.86 [95% CI: 0.64–1.16; p = 0.33]), and the mortality rate was in accordance with that of previous reports [1, 2]. The 28-day mortality, discharge to home, and length of hospital stay did not significantly differ between the two groups, indicating that these treatments had a similar efficacy. Further studies should be performed to compare the prognosis of AE-IIPs with and without corticosteroid therapy.

Corticosteroids have several side effects, and patients receiving corticosteroid therapy develop a variety of infections. High-dose corticosteroid therapy was found to be associated with a high incidence of infection [40]. In contrast, in a meta-analysis of the side effects of low-dose corticosteroid treatment for ARDS, there was no significant difference in the incidence of pneumonia between the steroid and control groups [41, 42]. A previous study on ARDS found no difference in terms of the incidence of infections between patients treated with high-dose corticosteroid treatment and those who received placebo [36]. Meanwhile, other studies revealed a significantly higher incidence of infections in the treatment group than in the placebo group [37, 43]. In the present study, the incidence of infections during hospitalization did not significantly differ between the high- and low-dose corticosteroid groups (the odds ratio in the low-dose corticosteroid group was 1.70 [95% CI: 0.99–2.94; p = 0.056]), and the incidence was relatively low. Although the side effects of corticosteroid therapy for AE-IIPs should be considered, this treatment could be safely used.

The present study had several limitations. First, because the database did not include data about laboratory examinations, pulmonary function tests, performance status, and radiological findings, the severity of IIPs at the onset of AE could not be accurately evaluated. Second, since IIP was diagnosed by a physician, it was not confirmed via radiological and pathological examinations. To accurately classify IIPs, all cases of secondary interstitial pneumonia were excluded based on ICD-10 codes because the specificity of diagnoses in the database is generally high [44]. However, because this study was retrospective in nature, unmeasured confounding cannot be excluded. Moreover, it may still be possible that a proportion of patients in the low-dose corticosteroid group of our study were treated for the exacerbation of COPD or asthma attack. The proportion of patients with comorbid COPD or asthma was only 5%–6% in the two groups after IPTW estimation. However, contamination of these patients who were likely to have better outcomes may have resulted in an underestimation of the effect of high-dose corticosteroid therapy. To address this issue, we conducted a sensitivity analysis 1 including only patients without COPD or bronchial asthma, and the results were similar to those of the main analyses. Furthermore, because the response to treatments for AE differs according to the type of IIPs [6], we performed sensitivity analysis 2, which excluded patients with COP, and the results were identical to those of the main analyses. Third, because the date of onset for each diagnosis during hospitalization was not recorded in the DPC database, we were unable to calculate incident rate ratios for each infection. Finally, patients who did not receive corticosteroid treatment for AE-IIPs could not be evaluated because corticosteroid treatment is the first-line treatment for AE-IIPs in Japan, albeit with limited evidence. Therefore, prospective studies must be conducted to compare the outcome and side effects of corticosteroids in patients with AE-IIPs.

CONCLUSIONS

In conclusion, there was no significant difference in terms of in-hospital mortality between patients with AE-IIPs who received high- and low-dose corticosteroid therapies. Further studies are needed to investigate effective treatments for AE-IIPs.

CONFLICT OF INTEREST

T.J. received research funds from Tsumura. H.Y. received research grants from the Ministry of Health, Labour and Welfare, Japan (21AA2007 and 20AA2005) and a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan (20H03907). The funding bodies had no role in the design of the study; collection, analysis, or interpretation of the data; or writing of the manuscript.

AVAILABILITY OF DATA AND MATERIALS

The information such as the study protocol, raw data, and programming code used during the current study are available from the corresponding author on reasonable request.

REFERENCES

• 1.   Raghu  G,  Collard  HR,  Egan  JJ,  Martinez  FJ,  Behr  J,  Brown  KK, et al. An official ATS/ERS/JRS/ALAT statement: Idiopathic pulmonary fibrosis: Evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med 2011;183:788–824.
• 2.   Natsuizaka  M,  Chiba  H,  Kuronuma  K,  Otsuka  M,  Kudo  K,  Mori  M, et al. Epidemiologic survey of Japanese patients with idiopathic pulmonary fibrosis and investigation of ethnic differences. Am J Respir Crit Care Med 2014;190:773–779.
• 3.   Travis  WD,  Costabel  U,  Hansell  DM,  King  TE,  Lynch  DA,  Nicholson  AG, et al. An official American Thoracic Society/European Respiratory Society statement: Update of the international multidisciplinary classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med 2013;188:733–748.
• 4.   Miyamoto  A,  Sharma  A,  Nishino  M,  Mino-Kenudson  M,  Matsubara  O,  Mark  EJ. Expanded acceptance of acute exacerbation of nonspecific interstitial pneumonia, including 7 additional cases with detailed clinical pathologic correlation. Pathol Int 2018;68:401–408.
• 5.   Enomoto  N,  Naoi  H,  Aono  Y,  Katsumata  M,  Horiike  Y,  Yasui  H, et al. Acute exacerbation of unclassifiable idiopathic interstitial pneumonia: comparison with idiopathic pulmonary fibrosis. Ther Adv Respir Dis 2020;14:1753466620935774.
• 6.   Suzuki  A,  Kondoh  Y,  Brown  KK,  Johkoh  T,  Kataoka  K,  Fukuoka  J, et al. Acute exacerbations of fibrotic interstitial lung diseases. Respirology 2020;25:525–534.
• 7.   Kolb  M,  Bondue  B,  Pesci  A,  Miyazaki  Y,  Song  JW,  Bhatt  NY, et al. Acute exacerbations of progressive-fibrosing interstitial lung diseases. Eur Respir Rev 2018;27:180071.
• 8.   Leuschner  G,  Behr  J. Acute exacerbation in interstitial lung disease. Front Med 2017;4:176.
• 9.   Park  IN,  Kim  DS,  Shim  TS,  Lim  CM,  Lee  SD,  Koh  Y, et al. Acute exacerbation of interstitial pneumonia other than idiopathic pulmonary fibrosis. Chest 2007;132:214–220.
• 10.   Homma  S,  Bando  M,  Azuma  A,  Sakamoto  S,  Sugino  K,  Ishii  Y, et al. Japanese guideline for the treatment of idiopathic pulmonary fibrosis. Respir Investig 2018;56:268–291.
• 11.   Collard  HR,  Ryerson  CJ,  Corte  TJ,  Jenkins  G,  Kondoh  Y,  Lederer  DJ, et al. Acute exacerbation of idiopathic pulmonary fibrosis. An international working group report. Am J Respir Crit Care Med 2016;194:265–275.
• 12.   Kondoh  Y,  Taniguchi  H,  Kawabata  Y,  Yokoi  T,  Suzuki  K,  Takagi  K. Acute exacerbation in idiopathic pulmonary fibrosis: Analysis of clinical and pathologic findings in three cases. Chest 1993;103:1808–1812.
• 13.   Akira  M,  Hamada  H,  Sakatani  M,  Kobayashi  C,  Nishioka  M,  Yamamoto  S. CT findings during phase of accelerated deterioration in patients with idiopathic pulmonary fibrosis. Am J Roentgenol 1997;168:79–83.
• 14.   Meduri  GU,  Headley  AS,  Golden  E,  Carson  SJ,  Umberger  RA,  Kelso  T, et al. Effect of prolonged methylprednisolone therapy in unresolving acute respiratory distress syndrome: A randomized controlled trial. JAMA 1998;280:159–165.
• 15.   Meduri  GU,  Siemieniuk  RAC,  Ness  RA,  Seyler  SJ. Prolonged low-dose methylprednisolone treatment is highly effective in reducing duration of mechanical ventilation and mortality in patients with ARDS. J Intensive Care 2018;6:53.
• 16.   Nishiyama  O,  Shimizu  M,  Ito  Y,  Kume  H,  Suzuki  R,  Yokoi  T, et al. Effect of prolonged low-dose methylprednisolone therapy in acute exacerbation of idiopathic pulmonary fibrosis. Respir Care 2001;46:698–701.
• 17.   Yasunaga  H. Real world data in Japan: Chapter II the diagnosis procedure combination database. Ann Clin Epidemiol 2019;1:76–79.
• 18.   Matsuda  S,  Fujimori  K,  Fushimi  K. Development of casemix based evaluation system in Japan. Asian Pac J Dis Manag 2010;4:55–66.
• 19.   Hugh-Jones  P,  Lambert  AV. A simple standard exercise test and its use for measuring exertion dyspnoea. Br Med J 1952;1:65–71.
• 20.   Aso  S,  Matsui  H,  Fushimi  K,  Yasunaga  H. Effect of cyclosporine A on mortality after acute exacerbation of idiopathic pulmonary fibrosis. J Thorac Dis 2018;10:5275–5282.
• 21.   Meduri  GU,  Tolley  EA,  Chrousos  GP,  Stentz  F. Prolonged methylprednisolone treatment suppresses systemic inflammation in patients with unresolving acute respiratory distress syndrome. Am J Respir Crit Care Med 2002;165:983–991.
• 22.   Annane  D,  Bellissant  E,  Bollaert  PE,  Briegel  J,  Keh  D,  Kupfer  Y. Corticosteroids for severe sepsis and septic shock: A systematic review and meta-analysis. Br Med J 2004;329:480–484.
• 23.   Minneci  PC,  Deans  KJ,  Banks  SM,  Eichacker  PQ,  Natanson  C. Meta-analysis: The effect of steroids on survival and shock during sepsis depends on the dose. Ann Intern Med 2004;141:47–56.
• 24.   Quan  H,  Sundararajan  V,  Halfon  P,  Fong  A,  Burnand  B,  Luthi  JC, et al. Coding algorithms for defining comorbidities in ICD-9- CM and ICD-10 administrative data. Med Care 2005;43:1130–1139.
• 25.   Quan  H,  Li  B,  Couris  CM,  Fushimi  K,  Graham  P,  Hider  P, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol 2011;173:676–682.
• 26.   Ohta  T,  Waga  S,  Handa  W,  Saito  I,  Takeuchi  K. New grading of level of disordered consciousness (author’s translation). No Shinkei Geka 1974;2:623–627.
• 27.   Shigematsu  K,  Nakano  H,  Watanabe  Y. The eye response test alone is sufficient to predict stroke outcome reintroduction of Japan Coma Scale: A cohort study. BMJ Open 2013;3:e002736.
• 28.   Ono  K,  Wada  K,  Takahara  T,  Shirotani  T. Indications for computed tomography in patients with mild head injury. Neurol Med Chir 2007;47:291–297.
• 29.   Meduri  GU,  Mauldin  GL,  Wunderink  RG,  Leeper  KV,  Jones  CB,  Tolley  E, et al. Causes of fever and pulmonary densities in patients with clinical manifestations of ventilator-associated pneumonia. Chest 1994;106:221–235.
• 30.   Graham  DJ,  Reichman  ME,  Wernecke  M,  Hsueh  YH,  Izem  R,  Southworth  MR, et al. Stroke, bleeding, and mortality risks in elderly medicare beneficiaries treated with dabigatran or rivaroxaban for nonvalvular atrial fibrillation. JAMA Intern Med 2016;176:1662–1671.
• 31.   Yasunaga  H. Propensity score analysis. Ann Clin Epidemiol 2020;2:33–37.
• 32.   Tomashefski  JF. Pulmonary pathology of acute respiratory distress syndrome. Clin Chest Med 2000;21:435–466.
• 33.   Chadda  K,  Annane  D. The use of corticosteroids in severe sepsis and acute respiratory distress syndrome. Ann Med 2002;34:582–589.
• 34.   Emura  I,  Usuda  H. Acute exacerbation of IPF has systemic consequences with multiple organ injury, with SRA+ and TNF-α+ cells in the systemic circulation playing central roles in multiple organ injury. BMC Pulm Med 2016;16:1–7.
• 35.   Schupp  JC,  Binder  H,  Jäger  B,  Cillis  G,  Zissel  G,  Müller-Quernheim  J, et al. Macrophage activation in acute exacerbation of idiopathic pulmonary fibrosis. PLoS One 2015;10:1–11.
• 36.   Bernard  GR,  Luce  JM,  Sprung  CL,  Rinaldo  JE,  Tate  RM,  Sibbald  WJ, et al. High-dose corticosteroids in patients with the adult respiratory distress syndrome. N Engl J Med 1987;317:1565–1570.
• 37.   Bone  RC,  Fisher  CJ,  Clemmer  TP,  Slotman  GJ,  Metz  CA. Early methylprednisolone treatment for septic syndrome and the adult respiratory distress syndrome. Chest 1987;92:1032–1036.
• 38.   Luce  JM,  Montgomery  AB,  Marks  JD,  Turner  J,  Metz  CA,  Murray  JF. Ineffectiveness of high-dose methylprednisolone in preventing parenchymal lung injury and improving mortality in patients with septic shock. Am Rev Respir Dis 1988;138:62–68.
• 39.   Wilkenfeld  SR,  Lin  C,  Frigo  DE. Communication between genomic and non-genomic signaling events coordinate steroid hormone actions. Steroids 2018;133:2–7.
• 40.   Lefering  R,  Neugebauer  EAM. Steroid controversy in sepsis and septic shock: A meta-analysis. Crit Care Med 1995;23:1294–1303.
• 41.   Tang  BMP,  Craig  JC,  Eslick  GD,  Seppelt  I,  McLean  AS. Use of corticosteroids in acute lung injury and acute respiratory distress syndrome: A systematic review and meta-analysis. Crit Care Med 2009;37:1594–1603.
• 42.   Annane  D,  Sébille  V,  Bellissant  E. Effect of low doses of corticosteroids in septic shock patients with or without early acute respiratory distress syndrome. Crit Care Med 2006;34:22–30.
• 43.   Lamontagne  F,  Briel  M,  Guyatt  GH,  Cook  DJ,  Bhatnagar  N,  Meade  M. Corticosteroid therapy for acute lung injury, acute respiratory distress syndrome, and severe pneumonia: A meta-analysis of randomized controlled trials. J Crit Care 2010;25:420–435.
• 44.   Yamana  H,  Moriwaki  M,  Horiguchi  H,  Kodan  M,  Fushimi  K,  Yasunaga  H. Validity of diagnoses, procedures, and laboratory data in Japanese administrative data. J Epidemiol 2017;27:476–482.