Study analysis · European Heart Journal · 2023
Six months of ultra-low cholesterol after a heart attack was linked to years of protection — but the evidence is far from bulletproof.
In a post hoc analysis, heart-attack patients who briefly got LDL cholesterol below 15 mg/dL with alirocumab plus statin had about 2 fewer heart attacks, strokes, or heart deaths per 100 people over about 3 years compared with similar statin-only patients.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
In simple terms
This study looked back at data from a big trial. It found that people who briefly had very low cholesterol while taking a strong drug plus statin had fewer heart problems later. But because the groups weren't randomly assigned for this specific question, we can't say the drug caused the benefit—only that it was linked.
What’s the bottom line?
In a large trial, some people with a recent acute coronary syndrome took alirocumab plus a statin and got their LDL cholesterol below 15 mg/dL. Then alirocumab was stopped and they stayed on statin alone. Researchers compared them with similar people who had been on placebo plus statin.
How strong is this study?
The study used data from a large, well-run trial and tried to match patients carefully. However, it's a 'look back' analysis, so other differences between patients could still explain the results. That means we should be cautious about trusting the findings as proof.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
56 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control group+15/15
- Sample size (n=2190)+20/20
- Follow-up+10/10
100 / 100
100 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registration+15/15
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 565 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. Post hoc, non-randomized comparison of a selected subgroup (patients who achieved very low LDL-C and underwent placebo substitution) with propensity score-matched placebo patients. Despite matching, residual confounding, selection bias, and confounding by indication remain possible. The exposure is defined by post-randomization LDL-C response and protocol-driven placebo substitution, not randomized treatment allocation.
COI Unknown
Could not determine conflict of interest status
No conflicts of interest or funding statements are included in the provided text, but the study evaluates a proprietary drug (alirocumab), so potential industry ties are not disclosed.
The absence of a conflict of interest and funding statement is notable. The trial is likely industry-funded (Sanofi/Regeneron) based on the drug tested, but this is not stated in the provided text. This lack of disclosure is a potential concern.
Key takeaways
- 01
Over about 2.8 years, MACE happened in 6.4% (47 of 730) of the alirocumab-then-placebo group vs 8.4% (122 of 1,460) of matched placebo patients.
- 02
That is a 28% lower relative risk (HR 0.72; 95% CI 0.51–0.997).
- 03
The absolute difference was about 2 fewer events per 100 people over 2.8 years (NNT about 50).
- 04
The absolute numbers are 6.4% vs 8.4%, meaning about 2 fewer MACE per 100 people over a median 2.8 years, or an NNT of about 50.
- 05
But the matched placebo group itself had lower risk than the whole placebo group in the trial (4-year MACE 10.7% vs 14.5%), so this is not a comparison against all post-ACS patients.
- 06
Also, this study has published corrections/errata; check the correction notices for updated information.
Surprising findings
- Benefit persisted despite LDL-C rising after alirocumab withdrawal.It contradicts the simple model that cardiovascular protection requires continuously lowered LDL-C. Here, a short period of very low LDL-C was associated with lower MACE over years.
- The subgroup HR was numerically lower than the overall trial HR despite shorter treatment and a lower-risk comparator.You might expect a brief treatment period in lower-risk patients to show a weaker signal, not a numerically stronger one. This could reflect selection bias or a true legacy effect.
- The matched placebo group had substantially lower MACE risk than the overall placebo group: 4-year estimates 10.7% vs 14.5%.It shows the very-low-LDL achievers were not representative of the whole trial. Their comparators were also lower-risk, limiting generalizability.
- Only 47 MACE events occurred in the 730-patient subgroup.That is a small number of events for a cardiovascular outcomes claim. It makes the result fragile and sensitive to small changes in event counting or follow-up.
Practical takeaways
If you have had acute coronary syndrome, discuss LDL-C targets and treatment intensification with your clinician rather than stopping or changing therapy on your own.
This post hoc analysis does not prove that a brief PCSK9 inhibitor course is sufficient or that stopping therapy is safe. It is not clinical guidance.
low confidenceWhen you see a 28% relative risk reduction, also ask for the absolute difference: here it was about 2 fewer MACE per 100 people over 2.8 years, NNT about 50.
The absolute benefit is modest and the confidence interval is wide. The matched placebo group was lower-risk than the overall placebo group.
medium confidenceContent creators should frame this as a hypothesis-generating post hoc finding, not a proven legacy effect.
Published corrections/errata exist for this study; check correction notices for updated information.
high confidenceWatch for prospective randomized trials testing short-term intensive LDL lowering followed by statin alone.
No such trial has yet established this as standard care. Current practice remains continuous lipid-lowering therapy for high-risk patients.
medium confidenceWhy this study matters
Short exposure, long tail?
In the ODYSSEY OUTCOMES post hoc analysis, 730 patients achieved two consecutive LDL-C levels <0.39 mmol/L (15 mg/dL) on alirocumab plus statin, then had blinded placebo substitution after a median 8.3 months. They had spent a median 6.0 months below that very low LDL-C threshold. Over median 2.8 years, MACE occurred in 47/730 (6.4%) vs 122/1,460 (8.4%) propensity-matched placebo-plus-statin patients: HR 0.72 (95% CI 0.51–0.997; P=0.047). That is a 28% lower RELATIVE risk, and an ABSOLUTE difference of about 2 fewer MACE per 100 people over 2.8 years (NNT about 50).
It challenges the idea that cholesterol-lowering benefit requires continuous, indefinite treatment intensity. A short burst of very low LDL-C was associated with lower risk that persisted even after LDL-C rose again.
But this is post hoc and fragile
This was not a randomized comparison of stopping alirocumab. The subgroup was defined by post-randomization LDL-C levels, which can introduce selection effects. The subgroup was small: 730 patients and only 47 MACE events. The confidence interval was wide, P=0.047 was borderline, and there was no adjustment for multiple testing. Residual confounding cannot be excluded despite propensity matching.
It is a masterclass in why a statistically significant result is not automatically practice-changing. The same data can support excitement and skepticism.
The matched placebo group was already lower risk
Propensity-matched placebo patients had a substantially lower 4-year Kaplan–Meier MACE risk than the entire placebo group: 10.7% vs 14.5%. Both matched groups also had lower baseline LDL-C and lipoprotein(a) and better study medication adherence than the overall cohort.
The patients who achieved very low LDL-C were inherently lower-risk to begin with. That makes the comparison less generalizable to all post-ACS patients.
Overall trial still supports alirocumab
In the full ODYSSEY OUTCOMES trial, alirocumab vs placebo was associated with lower MACE (HR 0.85; 95% CI 0.78–0.93; P<0.001), lower all-cause death (HR 0.85; 95% CI 0.73–0.98; P=0.03), and lower ischemia-driven revascularization (HR 0.88; 95% CI 0.79–0.97; P=0.01).
This gives context: the post hoc subgroup finding is not the main trial result. The overall randomized evidence remains stronger than the subgroup analysis.
LDL rose after stopping, but benefit didn't vanish
Time-averaged LDL-C was 51 mg/dL in the alirocumab-then-placebo group vs 82 mg/dL in matched placebo patients. After placebo substitution, LDL-C rose to a level not significantly different from the matched placebo group. Yet the cumulative MACE curves continued to diverge.
If benefit tracked only with current LDL-C, you would expect protection to fade when LDL-C rose. The persistence suggests possible plaque remodeling or a legacy effect.
Who achieved ultra-low LDL?
Patients who achieved consecutive LDL-C <0.39 mmol/L on alirocumab were more likely to be male, have diabetes, receive high-intensity statin treatment, have better study medication adherence, and be enrolled in Asia or South America, and less likely to be enrolled in Europe. They also had lower baseline LDL-C and lipoprotein(a).
Response to therapy is not random. Adherence, geography, baseline lipids, and clinical profile all shape who reaches very low LDL-C.
Want the whole report?
Detailed mode opens the full scientific breakdown — every score component, the methodology, conflicts of interest, the evidence analysis behind each claim, and the raw study data.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
In a large trial, some people with a recent acute coronary syndrome took alirocumab plus a statin and got their LDL cholesterol below 15 mg/dL. Then alirocumab was stopped and they stayed on statin alone. Researchers compared them with similar people who had been on placebo plus statin.
Research results
Over about 2.8 years, MACE happened in 6.4% (47 of 730) of the alirocumab-then-placebo group vs 8.4% (122 of 1,460) of matched placebo patients. That is a 28% lower relative risk (HR 0.72; 95% CI 0.51–0.997). The absolute difference was about 2 fewer events per 100 people over 2.8 years (NNT about 50).
What this means - more context
The absolute numbers are 6.4% vs 8.4%, meaning about 2 fewer MACE per 100 people over a median 2.8 years, or an NNT of about 50. But the matched placebo group itself had lower risk than the whole placebo group in the trial (4-year MACE 10.7% vs 14.5%), so this is not a comparison against all post-ACS patients. Also, this study has published corrections/errata; check the correction notices for updated information.
To assess whether a short period of very low LDL-C (<0.39 mmol/L or 15 mg/dL) achieved with a statin plus the PCSK9 inhibitor alirocumab, followed by blinded placebo substitution and continued statin alone, is associated with a legacy effect of lower cardiovascular risk in patients with recent acute coronary syndrome. This is a post hoc analysis of the ODYSSEY OUTCOMES trial. Note: this study has published corrections/errata; readers should check the correction notices for updated information.
In this post hoc analysis of ODYSSEY OUTCOMES, 730 post-ACS patients who achieved two consecutive LDL-C levels <0.39 mmol/L on alirocumab plus statin underwent blinded placebo substitution at a median 8.3 months, after a median 6.0 months below that threshold. Compared with 1,460 propensity-matched placebo-plus-statin patients, MACE occurred in 47/730 (6.4%) vs 122/1,460 (8.4%) over median 2.8 years: HR 0.72 (95% CI 0.51–0.997; P=0.047), a 28% lower relative risk and about 2 fewer MACE per 100 people over 2.8 years (NNT about 50). The matched placebo comparator had lower risk than the overall placebo group (4-year MACE 10.7% vs 14.5%). The study has published corrections/errata; check correction notices for updated information.
Methods Used
Post hoc analysis of the ODYSSEY OUTCOMES randomized trial (NCT01663402). 18,924 patients with recent acute coronary syndrome on high-intensity or maximum-tolerated statin were randomized to alirocumab or placebo. Alirocumab was blindly substituted with placebo if two consecutive LDL-C measurements were <0.39 mmol/L (15 mg/dL). 730 alirocumab patients had blinded placebo substitution at median 8.3 months and were matched 1:2 by propensity score to 1,460 placebo patients with similar baseline characteristics and medication adherence. Outcomes over median 2.8 years were compared with Cox models and log-rank tests.
Main Finding
In the post hoc comparison, MACE occurred in 47/730 (6.4%) alirocumab patients with limited-duration very low LDL-C vs 122/1,460 (8.4%) propensity-matched placebo patients: HR 0.72 (95% CI 0.51–0.997; P=0.047), a 28% lower relative risk. Absolute risk difference was about 2 percentage points lower, corresponding to about 2 fewer MACE per 100 people over median 2.8 years (NNT about 50). In the overall trial, alirocumab vs placebo was associated with lower MACE (HR 0.85; 95% CI 0.78–0.93; P<0.001), lower all-cause death (HR 0.85; 95% CI 0.73–0.98; P=0.03), and lower ischemia-driven revascularization (HR 0.88; 95% CI 0.79–0.97; P=0.01).
Confidence Level
Low to moderate. This is a post hoc exploratory analysis using post-randomization LDL-C levels to define the subgroup; residual confounding cannot be excluded despite propensity matching. The subgroup was small (730 patients, 47 MACE events), the confidence interval was wide, the P value was borderline (0.047), and there was no adjustment for multiple testing. The propensity-matched placebo group had substantially lower risk than the overall placebo group (4-year MACE 10.7% vs 14.5%). Published corrections/errata exist; check notices.
Study Flags
Red Flags
- •Post hoc exploratory analysis using post-randomization LDL-C to define the subgroup; residual confounding cannot be excluded despite propensity matching.
- •Small subgroup (730 patients, 47 MACE events), wide confidence interval, borderline P=0.047, and no adjustment for multiple testing.
- •Propensity-matched placebo comparator had substantially lower 4-year MACE risk than the overall placebo group (10.7% vs 14.5%), limiting generalizability; published corrections/errata also exist and should be checked.
Surprising Findings
Benefit persisted despite LDL-C rising after alirocumab withdrawal.
It contradicts the simple model that cardiovascular protection requires continuously lowered LDL-C. Here, a short period of very low LDL-C was associated with lower MACE over years.
Practical Takeaways
If you have had acute coronary syndrome, discuss LDL-C targets and treatment intensification with your clinician rather than stopping or changing therapy on your own.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 565 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study looked back at data from a big trial. It found that people who briefly had very low cholesterol while taking a strong drug plus statin had fewer heart problems later. But because the groups weren't randomly assigned for this specific question, we can't say the drug caused the benefit—only that it was linked.
Strengths
- Large parent randomized controlled trial with double-blind design and blinded placebo substitution
- Objective and adjudicated cardiovascular endpoints
- Propensity score matching on baseline characteristics and medication adherence
Weaknesses
- Post hoc, non-randomized comparison
- Exposure defined by post-randomization biomarker response
- Small number of events (47 vs 122)
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
In a large trial, some people with a recent acute coronary syndrome took alirocumab plus a statin and got their LDL cholesterol below 15 mg/dL. Then alirocumab was stopped and they stayed on statin alone. Researchers compared them with similar people who had been on placebo plus statin.
Research results
Over about 2.8 years, MACE happened in 6.4% (47 of 730) of the alirocumab-then-placebo group vs 8.4% (122 of 1,460) of matched placebo patients. That is a 28% lower relative risk (HR 0.72; 95% CI 0.51–0.997). The absolute difference was about 2 fewer events per 100 people over 2.8 years (NNT about 50).
What this means - more context
The absolute numbers are 6.4% vs 8.4%, meaning about 2 fewer MACE per 100 people over a median 2.8 years, or an NNT of about 50. But the matched placebo group itself had lower risk than the whole placebo group in the trial (4-year MACE 10.7% vs 14.5%), so this is not a comparison against all post-ACS patients. Also, this study has published corrections/errata; check the correction notices for updated information.
To assess whether a short period of very low LDL-C (<0.39 mmol/L or 15 mg/dL) achieved with a statin plus the PCSK9 inhibitor alirocumab, followed by blinded placebo substitution and continued statin alone, is associated with a legacy effect of lower cardiovascular risk in patients with recent acute coronary syndrome. This is a post hoc analysis of the ODYSSEY OUTCOMES trial. Note: this study has published corrections/errata; readers should check the correction notices for updated information.
In this post hoc analysis of ODYSSEY OUTCOMES, 730 post-ACS patients who achieved two consecutive LDL-C levels <0.39 mmol/L on alirocumab plus statin underwent blinded placebo substitution at a median 8.3 months, after a median 6.0 months below that threshold. Compared with 1,460 propensity-matched placebo-plus-statin patients, MACE occurred in 47/730 (6.4%) vs 122/1,460 (8.4%) over median 2.8 years: HR 0.72 (95% CI 0.51–0.997; P=0.047), a 28% lower relative risk and about 2 fewer MACE per 100 people over 2.8 years (NNT about 50). The matched placebo comparator had lower risk than the overall placebo group (4-year MACE 10.7% vs 14.5%). The study has published corrections/errata; check correction notices for updated information.
Methods Used
Post hoc analysis of the ODYSSEY OUTCOMES randomized trial (NCT01663402). 18,924 patients with recent acute coronary syndrome on high-intensity or maximum-tolerated statin were randomized to alirocumab or placebo. Alirocumab was blindly substituted with placebo if two consecutive LDL-C measurements were <0.39 mmol/L (15 mg/dL). 730 alirocumab patients had blinded placebo substitution at median 8.3 months and were matched 1:2 by propensity score to 1,460 placebo patients with similar baseline characteristics and medication adherence. Outcomes over median 2.8 years were compared with Cox models and log-rank tests.
Main Finding
In the post hoc comparison, MACE occurred in 47/730 (6.4%) alirocumab patients with limited-duration very low LDL-C vs 122/1,460 (8.4%) propensity-matched placebo patients: HR 0.72 (95% CI 0.51–0.997; P=0.047), a 28% lower relative risk. Absolute risk difference was about 2 percentage points lower, corresponding to about 2 fewer MACE per 100 people over median 2.8 years (NNT about 50). In the overall trial, alirocumab vs placebo was associated with lower MACE (HR 0.85; 95% CI 0.78–0.93; P<0.001), lower all-cause death (HR 0.85; 95% CI 0.73–0.98; P=0.03), and lower ischemia-driven revascularization (HR 0.88; 95% CI 0.79–0.97; P=0.01).
Confidence Level
Low to moderate. This is a post hoc exploratory analysis using post-randomization LDL-C levels to define the subgroup; residual confounding cannot be excluded despite propensity matching. The subgroup was small (730 patients, 47 MACE events), the confidence interval was wide, the P value was borderline (0.047), and there was no adjustment for multiple testing. The propensity-matched placebo group had substantially lower risk than the overall placebo group (4-year MACE 10.7% vs 14.5%). Published corrections/errata exist; check notices.
Study Flags
Red Flags
- •Post hoc exploratory analysis using post-randomization LDL-C to define the subgroup; residual confounding cannot be excluded despite propensity matching.
- •Small subgroup (730 patients, 47 MACE events), wide confidence interval, borderline P=0.047, and no adjustment for multiple testing.
- •Propensity-matched placebo comparator had substantially lower 4-year MACE risk than the overall placebo group (10.7% vs 14.5%), limiting generalizability; published corrections/errata also exist and should be checked.
Surprising Findings
Benefit persisted despite LDL-C rising after alirocumab withdrawal.
It contradicts the simple model that cardiovascular protection requires continuously lowered LDL-C. Here, a short period of very low LDL-C was associated with lower MACE over years.
Practical Takeaways
If you have had acute coronary syndrome, discuss LDL-C targets and treatment intensification with your clinician rather than stopping or changing therapy on your own.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 565 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study looked back at data from a big trial. It found that people who briefly had very low cholesterol while taking a strong drug plus statin had fewer heart problems later. But because the groups weren't randomly assigned for this specific question, we can't say the drug caused the benefit—only that it was linked.
Strengths
- Large parent randomized controlled trial with double-blind design and blinded placebo substitution
- Objective and adjudicated cardiovascular endpoints
- Propensity score matching on baseline characteristics and medication adherence
Weaknesses
- Post hoc, non-randomized comparison
- Exposure defined by post-randomization biomarker response
- Small number of events (47 vs 122)
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
The study used data from a large, well-run trial and tried to match patients carefully. However, it's a 'look back' analysis, so other differences between patients could still explain the results. That means we should be cautious about trusting the findings as proof.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
56 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control group+15/15
- Sample size (n=2190)+20/20
- Follow-up+10/10
100 / 100
100 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registration+15/15
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 565 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. Post hoc, non-randomized comparison of a selected subgroup (patients who achieved very low LDL-C and underwent placebo substitution) with propensity score-matched placebo patients. Despite matching, residual confounding, selection bias, and confounding by indication remain possible. The exposure is defined by post-randomization LDL-C response and protocol-driven placebo substitution, not randomized treatment allocation.
COI Unknown
Could not determine conflict of interest status
No conflicts of interest or funding statements are included in the provided text, but the study evaluates a proprietary drug (alirocumab), so potential industry ties are not disclosed.
The absence of a conflict of interest and funding statement is notable. The trial is likely industry-funded (Sanofi/Regeneron) based on the drug tested, but this is not stated in the provided text. This lack of disclosure is a potential concern.
Standing
Who’s using this study?
The videos and claims on this site that lean on this study, and the researchers who wrote it.
1 video from Dr Brad Stanfield cite this study, drawing 1 claim from it.
- Very strong evidence
Randomized or controlled trials support this claim, alongside consistent supporting evidence.
Evidence
Authored by
17 researchersIf this is your work, this is how we attribute it on Fit Body Science. Gregory G. Schwartz is listed as the lead author.