Effect of preadmission glucocorticoid therapy on 30-day mortality in critically ill patients: a retrospective study of a mixed ICU population in a tertiary hospital

Annals of Intensive Care

Table 4 Subgroup analysis according to the main diagnosis at the time of ICU admission

Variable	30-Day mortality
Variable	Hazard ratio (95% CI)	P-value*
Septic shock (n = 409)
Preadmission GC user	3.21 (1.72, 5.97)	< 0.001
Cardiac disease (n = 8192)
Preadmission GC user	1.49 (0.65, 3.43)	0.345
Neurologic disease (n = 3348)
Preadmission GC user	1.90 (1.67, 5.40)	< 0.001
Kidney failure (n = 1725)
Preadmission GC user	1.07 (0.53, 2.16)	0.860
Respiratory insufficiency or failure (n = 2906)
Preadmission GC user	2.04 (1.47, 2.82)	< 0.001
Drug intoxication (n = 176)
Preadmission GC user	0.00 (0.00–)	0.966
Trauma (n = 191)
Preadmission GC user	0.00 (0.00–)	0.984
GI bleeding, ischemia, or perforation (n = 1108)
Preadmission GC user	0.55 (0.12, 2.59)	0.453
Hemorrhagic shock (n = 107)
Preadmission GC user	0.00 (0.00–)	1.00
Post-cardiac arrest (n = 331)
Preadmission GC user	2.33 (0.87, 6.25)	0.093
Others** (n = 9522)
Preadmission GC user	1.83 (0.52, 3.25)	0.138

Covariates of P < 0.2 in univariate Cox regression analysis in Additional file 1: Table S1 (except the main diagnosis of ICU admission) were included in multivariate Cox regression analysis: age, body mass index, APACHE II, comorbidity at ICU admission (hypertension, cerebrovascular disease, liver disease, chronic kidney disease, anemia, and cancer), admission through emergency department
APACHE Acute Physiology and Chronic Health Evaluation, CI confidence interval, GC glucocorticoid, GI gastrointestinal, ICU intensive care unit
*P < 0.005 was considered statistically significant after Bonferroni correction
Others**include liver failure, urinary tract infection, unknown, peripheral vascular disease, and ICU admission for monitoring