Effect of total cholesterol and statin therapy on mortality in ARDS patients: a secondary analysis of the SAILS and HARP-2 trials

Table 2 60-day mortality varies with baseline cholesterol level and treatment assignment in SAILS and HARP-2

Subgroup	Statin	Placebo	OR (Fisher’s)	Adjusted OR¹	Interaction¹	Fully Adjusted OR²	Fully Adjusted Interaction²
SAILS (rosuvastatin)
Low Cholesterol	37.8%	21.3%	2.23 (95% CI 1.06–4.77, p = 0.02)	2.28 (95% CI 1.14–4.68, p = 0.02)	p = 0.02	3.35 (95% CI 1.55–7.59, p = 0.003)	p = 0.004
Not-low Cholesterol	26.2%	27.0%	0.96 (p = 0.92)	0.95 (p = 0.82)	p = 0.02	0.93 (p = 0.75)	p = 0.004
HARP-2 (simvastatin)
Low Cholesterol	31.9%	52.1%	0.44 (95% CI 0.17–1.07, p = 0.06)	0.48 (95% CI 0.20–1.14, p = 0.10)	p = 0.22	0.24 (95% CI 0.08–0.67, p = 0.009)	p = 0.045
Not-low Cholesterol	26.3%	26.9%	0.97 (p = 1.00)	0.93 (p = 0.80)	p = 0.22	0.71 (p = 0.28)	p = 0.045

SAILS cohort N = 678 and HARP-2 N = 384 septic patients
Adjusted OR is for statin versus placebo group in binary logistic regression, after adjustment for age, gender and body mass index
¹The cholesterol-statin interaction denotes the significance of the logistic regression interaction term between low cholesterol and randomization to statin therapy after adjustment for age, gender, and BMI
²This cholesterol-interaction indicates the significance of the logistic regression interaction term between low cholesterol and randomization to statin therapy after adjustment for age, gender, BMI, and APACHE II/III score. Inclusion of APACHE II/III score was not a prespecified analysis

ISSN: 1364-8535