Heart Protection Study Results in Detail: Numbers, Subgroups, and Time Course

At a glance

• Trial name: Heart Protection Study (HPS)
• N: 20,536 randomized (10,269 simvastatin; 10,267 placebo)
• Intervention: Simvastatin 40 mg daily
• Comparator: Matching placebo
• Duration: Mean 5.0 years of follow-up
• Primary endpoint: First major vascular event (coronary death, non-fatal MI, stroke, or revascularization)
• Key result: 24% relative risk reduction in major vascular events (p <0.00001)
• Sponsor: MRC/BHF (UK Medical Research Council and British Heart Foundation)

Why This Trial Changed Clinical Practice

Before 2002, statin prescribing was guided primarily by LDL thresholds. Physicians debated whether patients with "normal" cholesterol, diabetes without prior coronary disease, peripheral arterial disease, or cerebrovascular disease alone would benefit from lipid-lowering therapy. The Heart Protection Study enrolled the broadest population of any statin trial to that date and answered those questions with a single, definitive dataset.

The trial's importance was not just the top-line result. It was the sheer breadth of subgroups that showed consistent benefit, forcing guideline committees worldwide to shift from cholesterol-target-based prescribing toward risk-based prescribing.

Primary Endpoint: Major Vascular Events

The composite of major coronary events, strokes, and revascularizations occurred in 2,033 patients (19.8%) assigned simvastatin versus 2,585 patients (25.2%) assigned placebo. That translates to:

• Absolute risk reduction (ARR): 5.4 percentage points
• Relative risk reduction: 24% (event rate ratio 0.76; 95% CI 0.72-0.81)
• p-value: <0.00001
• Number needed to treat (NNT): ~19 over 5 years

Each component of the composite endpoint showed a statistically significant reduction, a rarity in cardiovascular megatrials where one element often drives the overall signal.

HPS Primary and Secondary Endpoint Breakdown

| Endpoint | Simvastatin n (%) | Placebo n (%) | Rate Ratio (95% CI) | p-value | |---|---|---|---|---| | Major vascular events (composite) | 2,033 (19.8%) | 2,585 (25.2%) | 0.76 (0.72-0.81) | <0.00001 | | Major coronary events | 898 (8.7%) | 1,212 (11.8%) | 0.73 (0.67-0.79) | <0.00001 | | Coronary death | 587 (5.7%) | 707 (6.9%) | 0.82 (0.74-0.92) | 0.0005 | | Non-fatal MI | 357 (3.5%) | 574 (5.6%) | 0.62 (0.54-0.70) | <0.00001 | | All stroke | 444 (4.3%) | 585 (5.7%) | 0.75 (0.66-0.85) | <0.00001 | | Ischemic stroke | 339 (3.3%) | 488 (4.8%) | 0.69 (0.60-0.79) | <0.00001 | | Revascularization (any) | 939 (9.1%) | 1,205 (11.7%) | 0.76 (0.70-0.83) | <0.00001 | | All-cause mortality | 1,328 (12.9%) | 1,507 (14.7%) | 0.87 (0.81-0.94) | 0.0003 | | Vascular mortality | 781 (7.6%) | 937 (9.1%) | 0.83 (0.75-0.91) | <0.00001 |

All-cause mortality dropped by 13%. This is one of a limited number of statin trials to demonstrate a significant all-cause mortality benefit, driven primarily by the reduction in vascular deaths. Non-vascular mortality did not differ between groups (547 vs 570), putting to rest early concerns about statin-related non-cardiovascular death.

LDL Reduction and the Lipid Response

Mean baseline LDL cholesterol was approximately 3.4 mmol/L (131 mg/dL). Simvastatin 40 mg produced:

• Average LDL reduction: approximately 1.0 mmol/L (39 mg/dL), roughly a 29-33% decrease from baseline
• Compliance-adjusted reduction: approximately 1.3 mmol/L in the fully-adherent subset (the study used intention-to-treat, and non-compliance diluted the on-treatment effect)
• Non-compliance rate: roughly 18% of the simvastatin group had stopped taking study medication and approximately 17% of the placebo group had started a non-study statin by year five

The compliance crossover is clinically important. The intention-to-treat analysis underestimates the true pharmacologic effect. When the investigators modeled the compliance-adjusted benefit, the per-1.0 mmol/L LDL reduction corresponded to roughly a one-third reduction in major vascular events, consistent with meta-analytic data from the Cholesterol Treatment Trialists' Collaboration.

Time-Course Pattern: When Benefits Appeared

The Kaplan-Meier curves for major vascular events began separating at approximately 12 months and continued to diverge throughout the five-year follow-up period. The investigators noted:

• Year 1: Small, non-significant separation. LDL reduction was immediate, but event reduction lagged.
• Year 2: Statistically detectable separation. The annual event rate ratio in year two was already near 0.76.
• Years 3-5: Continued divergence with no plateau. The proportional reduction remained roughly constant at 24-25% per year.

This pattern is consistent with other statin trials and the understanding that atherosclerotic plaque stabilization requires sustained LDL exposure reduction over months. The absence of a plateau through five years suggests that longer treatment durations would produce additional absolute benefit, a point confirmed by extended follow-up data in other statin trial cohorts.

Subgroup Analyses: The Trial's Most Important Contribution

What set HPS apart from earlier statin trials (4S, CARE, LIPID, WOSCOPS) was its pre-specified subgroup design. The primary publication showed consistent benefit across groups that had been excluded from or underrepresented in previous trials.

Results by Baseline LDL Category

| Baseline LDL | n | Rate Ratio (95% CI) | |---|---|---| | <3.0 mmol/L (<116 mg/dL) | ~6,800 | 0.76 (0.69-0.85) | | 3.0-3.5 mmol/L (116-135 mg/dL) | ~6,800 | 0.76 (0.68-0.85) | | >3.5 mmol/L (>135 mg/dL) | ~6,900 | 0.76 (0.69-0.84) |

The rate ratios were virtually identical across all three LDL strata (interaction p = 0.9). This was the single most practice-changing finding: patients with LDL below 3.0 mmol/L (116 mg/dL) benefited just as much proportionally as those with elevated LDL. This result directly shaped the 2004 NCEP ATP III update and later the 2013 ACC/AHA cholesterol guideline shift toward risk-based rather than target-based statin prescribing.

Results by Qualifying Condition

| Subgroup | n | Events Simvastatin | Events Placebo | Rate Ratio | |---|---|---|---|---| | Prior MI | ~8,500 |, |, | 0.74 | | Other coronary disease | ~4,900 |, |, | 0.76 | | No prior coronary disease | ~7,100 |, |, | 0.77 | | Diabetes (no prior CHD) | ~3,900 |, |, | 0.75 | | Cerebrovascular disease | ~3,200 |, |, | 0.76 | | Peripheral arterial disease | ~6,700 |, |, | 0.78 | | Women | ~5,000 |, |, | 0.75 | | Age ≥75 at entry | ~1,200 |, |, | 0.74 |

The diabetes-without-prior-CHD subgroup finding was particularly impactful. HPS provided some of the first large-scale RCT evidence that diabetic patients benefit from statin therapy regardless of baseline LDL or coronary history. This evidence now underpins the near-universal recommendation for statin use in diabetic adults found in ADA Standards of Care.

Safety and Adverse Events

Over five years of follow-up, simvastatin 40 mg showed a reassuring safety profile.

• Myopathy (CK >10x ULN with muscle symptoms): 11 cases in the simvastatin group vs 4 in the placebo group (excess risk approximately 0.01% per year)
• Rhabdomyolysis: 5 cases (simvastatin) vs 3 cases (placebo)
• Hepatic transaminase elevation (>4x ULN): 0.8% simvastatin vs 0.6% placebo (not significantly different)
• Cancer incidence: No difference (7.9% vs 7.8%)
• Hemorrhagic stroke: No significant excess (51 vs 53 events)

The lack of hemorrhagic stroke excess was important because LDL lowering had been linked to hemorrhagic stroke risk in observational data. HPS was large enough to test this concern directly. Ischemic stroke fell by 31% while hemorrhagic stroke rates were identical, yielding a clear net neurological benefit.

What the Trial Could Not Show

The investigators acknowledged several limitations in the original report and subsequent analyses.

Single dose tested. HPS used simvastatin 40 mg only. It could not determine whether higher-intensity statin therapy (such as atorvastatin 80 mg, later tested in TNT and IDEAL) would produce greater benefit.

UK-dominant population. The cohort was predominantly white and British. Ethnic generalizability was limited, though the biological mechanism of LDL reduction is not population-specific.

2x2 factorial with antioxidant vitamins. The trial also randomized participants to antioxidant vitamins (vitamin E, vitamin C, beta-carotene) versus placebo. The vitamin arm showed no benefit, which increased confidence that the vascular event reductions were attributable to simvastatin alone. Still, the factorial design introduced a layer of complexity in interpretation.

Non-compliance dilution. With roughly 18% discontinuation in the statin arm and 17% crossover to open-label statins in the placebo arm by year five, the observed 24% reduction understates the true on-treatment effect. The investigators estimated the compliance-adjusted reduction in major vascular events at approximately one-third per 1.0 mmol/L LDL reduction.

No ezetimibe or PCSK9 comparator. Tested in 2002, HPS predated modern combination lipid-lowering strategies. It cannot speak to the incremental benefit of adding ezetimibe or PCSK9 inhibitors to statin therapy.

Lasting Influence on Guidelines

HPS was cited in every major lipid guideline revision between 2003 and 2018. Its specific contributions include:

1. Establishing that statin benefit is proportional to absolute risk reduction per unit of LDL lowering, regardless of baseline LDL
2. Providing the evidence base for statin use in diabetic patients without established cardiovascular disease
3. Confirming vascular benefit in patients with peripheral arterial disease and cerebrovascular disease
4. Demonstrating all-cause mortality benefit in a broad population, not just post-MI patients

The 2018 ACC/AHA multi-society cholesterol guideline directly references HPS when recommending moderate-intensity statin therapy for adults aged 40-75 with diabetes, regardless of estimated 10-year ASCVD risk.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

1. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet. 2002;360(9326):7-22. PubMed
2. Cholesterol Treatment Trialists' (CTT) Collaborators. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet. 2005;366(9493):1267-1278. PubMed
3. Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol. J Am Coll Cardiol. 2019;73(24):e285-e350. PubMed
4. Stone NJ, Robinson JG, Lichtenstein AH, et al. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults. J Am Coll Cardiol. 2014;63(25 Pt B):2889-2934. PubMed
5. LaRosa JC, Grundy SM, Waters DD, et al. Intensive lipid lowering with atorvastatin in patients with stable coronary disease (TNT trial). N Engl J Med. 2005;352(14):1425-1435. PubMed
6. Cannon CP, Blazing MA, Giugliano RP, et al. Ezetimibe added to statin therapy after acute coronary syndromes (IMPROVE-IT). N Engl J Med. 2015;372(25):2387-2397. PubMed