Severely frail elderly patients do not need lipid-lowering drugs

Article Figures & Data

TABLE 1

Studies of lipid-lowering therapy discussed in this paper

Study	Design	Population	Mean follow-up
PROSPER¹³	Randomized controlled trial Pravastatin 40 mg vs placebo	N = 5,804, mean age 75 Primary and secondary prevention Mini-Mental State score ≥ 24 of 30	3.2 years
JUPITER^28,29	Randomized controlled trial Rosuvastatin 20 mg vs placebo	N = 5,695 (elderly subgroup) Median age 74 Primary prevention	1.9 years (stopped prematurely)
CTT³²	Meta-analysis	Not applicable^a	Variable
Afilalo et al³¹	Meta-analysis	N = 19,569 Age range 65–82 Secondary prevention	4.9 years
SPARCL^27,30	Randomized controlled trial Atorvastatin 80 mg vs placebo	N = 2,249 (subgroup age ≥ 65) Mean age of subgroup 72.4 Secondary prevention	4.9 years
GISSI-HF²⁵	Randomized controlled trial Rosuvastatin 10 mg vs placebo	N = 4,574, mean age 68 Heart failure, NYHA class II to IV	3.9 years (median)
CORONA²⁶	Randomized controlled trial Rosuvastatin 10 mg vs placebo	N = 5,011, mean age 73 Heart failure, NYHA class II to IV	2.7 years

↵a The CTT meta-analysis presents the effects of statins on major vascular events per annum, per 1.0 mmol/L reduction in LDL-C according to 5-year risk at baseline. The analysis in patients > 70 years old includes a total of 2,952 events in the statin group and 3,385 events in the control group (or a total of 6,337 events.
CORONA = Controlled Rosuvastatin Multinational Trial in Heart Failure; CTT = Cholesterol Treatment Trialists; GISSI-HF= Gruppo Italiano per lo Studio della Sopravvivenza nell’Insufficienza Cardiaca Heart Failure; JUPITER = Justification for the Use of Statins in Prevention; NYHA = New York Heart Association; PROSPER = Prospective Study of Pravastatin in the Elderly at Risk; SPARCL = Stroke Prevention by Aggressive Reduction in Cholesterol Levels

TABLE 2

Coronary heart disease outcomes

Trial and outcome	Event rate		P value	Absolute risk reduction	Relative risk reduction	Number needed to treat (95% CI)
Trial and outcome	Placebo	Statin	P value	Absolute risk reduction	Relative risk reduction	Number needed to treat (95% CI)
Primary prevention studies
PROSPER¹³ Coronary heart disease death and nonfatal myocardial infarction	8.8%	7.9%	.401	0.9%	9%	NS^a
JUPITER^28,29 yocardial infarction	1.1%	0.6%	.046	0.5%	45%	211 (106–32,924)
Secondary prevention studies
PROSPER¹³ Coronary heart disease death and nonfatal myocardial infarction (includes definite and suspect events)	16.8%	12.7%	.004	4.1%	24%	25 (15–77)
Afilalo et al meta-analysis³¹ Nonfatal myocardial infarction	NA	NA	NA	NA	26%^b	38 (16–118)^b
Combined primary and secondary prevention studies
PROSPER¹³ Nonfatal myocardial infarction (excluding silent and unrecognized events)	4.3%	3.4%	.099	0.9%	20%	NS^a

↵a Not statistically significant (P > .05) and numbers needed to treat (NNT) not calculated on these results. NNT calculated on raw numbers if not provided in publication. See Table 1 for duration of treatment required for NNT.
↵b Afilalo et al provided relative risk reductions and NNTs in their publication. Absolute risk reductions not calculated on meta-analytic outcomes.
CI = confidence interval; NA = not available; NS = not statistically significant

TABLE 3

Stroke outcomes

Trial and outcome	Event rate		P value	Absolute risk reduction	Relative risk reduction^a	Number needed to treat (95% CI)
Trial and outcome	Placebo	Statin	P value	Absolute risk reduction	Relative risk reduction^a	Number needed to treat (95% CI)
Primary prevention studies
PROSPER¹³ Fatal or nonfatal stroke Transient ischemic attack	3.7%	3.8%	.882	0.1% increase	3% increase	NS^b
	2.3%	1.9%	.422	0.4%	18%	NS
JUPITER^28,29 Stroke (percent based on raw number of events)	1.4%	0.8%	.023	0.6%	45%	161 (86–1,192)
Secondary prevention studies
PROSPER¹³ Fatal and nonfatal stroke Transient ischemic attack	5.5%	5.7%	.838	0.2% increase	3% increase	NS
	5.1%	3.6%	.065	1.5%	29%	NS
Afilalo meta-analysis³¹ Stroke (disabling and nondisabling)	NA	NA	NA	NA	25%	58^c (27–177)
SPARCL²⁷ Fatal or nonfatal stroke Nonfatal stroke	13.1%	11.2%	.03	1.9%	15%	52 (26–1,303)
SPARCL²⁷ Fatal or nonfatal stroke Nonfatal stroke	11.8%	10.4%	.11	1.4%	12%	NS
SPARCL subgroup ≥ 65³⁰ Fatal or nonfatal stroke	16.2%	14.7%	.33	1.5%	10%	NS
SPARCL subgroup < 65³⁰ Fatal or nonfatal stroke	10.5%	7.9%	.022	2.6%	24%	39 (21–354)

↵a Numbers for relative risk reduction are rounded to the nearest whole number
↵b Not statistically significant (P >.05) and numbers needed to treat not calculated on these results. Number needed to treat (NNT) calculated on raw numbers if not provided in publication. See Table 1 for duration of treatment required for NNT. CI = confidence interval
↵c Afilalo et al provided relative risk reductions and NNT. Absolute risk reductions not calculated on meta-analytic outcomes.