Cardiometabolic Clinical Trial Ledger: Surrogate Biomarkers vs. Hard Mortality NNT
For decades, cardiometabolic drug marketing equated laboratory improvements with saved lives. Yet history repeatedly demonstrated the danger of unvalidated surrogates: CETP inhibitors (torcetrapib) raised HDL cholesterol by 72% while increasing mortality, and intensive glucose lowering with rosiglitazone or tight ACCORD targets lowered HbA1c without reducing macrovascular death.
This ledger compares contemporary cardiometabolic drug classes strictly by their hard clinical outcome trials, per-1,000 absolute event reductions, and NNT.
In cardiovascular and diabetes medicine, lowering a laboratory surrogate (such as LDL cholesterol or HbA1c) does not guarantee fewer heart attacks or longer life. Clinical decisions should rest on hard-endpoint Randomized Controlled Trials measuring 3-point MACE (cardiovascular death, nonfatal myocardial infarction, nonfatal stroke) and all-cause mortality, expressed as 5-year Absolute Risk Reduction (ARR) and Number Needed to Treat (NNT).
How Many People Must Take a Statin, SGLT2 Inhibitor, or GLP-1 Agonist to Prevent One Event?
Side-by-side synthesis of landmark cardiovascular and renal outcome trials, separating primary prevention (no prior heart disease) from secondary prevention (established atherosclerotic disease or heart failure).
| Therapy & Patient Cohort | Surrogate Biomarker Shift | Hard Primary Endpoint & Horizon | Absolute Reduction (per 1,000) & NNT | Key Harms / NNH |
|---|---|---|---|---|
| Statins (Primary Prevention) 10-yr ASCVD risk 7.5%-15% (CTT / USPSTF) |
LDL-C reduced by 35-50 mg/dL | Major vascular events over 5 years | ARR ~1.0% to 1.2% (10-12 / 1,000) NNT = 83 to 100 (MACE) NNT = 250+ (all-cause death) |
New-onset type 2 diabetes: NNH ~100 to 200; severe myopathy <1 per 1,000. |
| Statins (Secondary Prevention) Prior MI, stroke, or coronary stent (CTT) |
LDL-C reduced by 40-60 mg/dL | Major vascular events & CV death over 5 years | ARR ~2.6% to 3.5% (26-35 / 1,000) NNT = 28 to 39 (MACE) NNT = 83 (all-cause death) |
High net benefit; stopping secondary statin therapy after MI significantly increases recurrent events. |
| SGLT2 Inhibitors (Dapagliflozin / Empagliflozin) HFrEF Heart Failure (DAPA-HF, 18.2 mo) |
HbA1c -0.2% (works even in non-diabetics!) | Worsening HF hospitalization or CV death (1.5 years) | ARR 4.9% (49 per 1,000) (16.3% vs 21.2%) NNT = 21 over 18 months |
Genital mycotic infection NNH ~25; rare euglycemic ketoacidosis in insulin-deficient patients. |
| GLP-1 Receptor Agonist (Semaglutide 2.4 mg) SELECT Trial: BMI >=27 + prior CVD, no diabetes (39.8 mo) |
Body weight -9.4%; hsCRP -38% | 3-point MACE (CV death, MI, stroke) over 3.3 years | ARR 1.5% (15 per 1,000) (6.5% vs 8.0%) NNT = 67 over 3.3 years |
Permanent discontinuation due to GI adverse events: 16.6% vs 8.2% (NNH = 12). |
| Fibrates / Niacin Added to Statin ACCORD-Lipid / AIM-HIGH Trials |
HDL-C +15% to +25%; Triglycerides -25% | 3-point MACE over 3 to 5 years | ARR 0.0% (0 per 1,000) NNT = ∞ (No hard benefit) |
Classic proof that fixing a lipid number on a lab report does not automatically prevent heart attacks. |
Primary sources: NEJM 2019 (DAPA-HF Trial) · NEJM 2023 (SELECT Semaglutide Trial) · JAMA 2022 (USPSTF Statin Systematic Review) · NEJM 2011 (AIM-HIGH Trial).
Preserved MedRead Cardiology & Nephrology Case Series (2019-2021)
Between 2019 and 2021, the MedRead Journal of Cardiology Research and MedRead Journal of Nephrology published clinical case reports and cross-sectional echocardiographic studies that remain referenced in academic repositories. Below is the methodological synthesis of the preserved series:
- Right Ventricular Systolic Function in COPD (30-Case Cross-Sectional Study): Evaluated tricuspid annular plane systolic excursion (TAPSE) and pulmonary arterial systolic pressure in chronic obstructive pulmonary disease. Methodological appraisal: Cross-sectional echocardiography identifies right-heart strain prevalence but requires prospective cohort follow-up to establish prognostic thresholds.
- Vascular Events & Phosphodiesterase-5 Inhibitor Case Reports: Archived autopsy and clinical case reports examining rare temporal associations between sildenafil use and acute aortic dissection or myocardial infarction. Methodological appraisal: Single case reports (Level V evidence) generate pharmacovigilance hypotheses but cannot establish causal incidence without denominator-controlled cohort data.
- Metformin in Chronic Kidney Disease ("Metformin: The Great Forgotten"): Reviewed lactic acidosis thresholds across eGFR strata, aligning with modern KDIGO guidelines that safely permit metformin at eGFR 30-45 mL/min/1.73m² with dose halving.
Frequently Asked Methodological & Clinical Questions
Plain-language answers grounded in CONSORT, GRADE, WHO guidelines, and per-1,000 absolute outcome arithmetic.
Why is the NNT for statins so much lower in secondary prevention than primary prevention?
Because Number Needed to Treat (NNT) depends directly on baseline event risk. A patient who has already survived a heart attack or stroke has a high 5-year risk of a repeat vascular event (10% to 15%), so a 25% relative reduction yields a 3% absolute drop (NNT ~30). A healthy adult with mild cholesterol elevation may have a 5-year risk of only 3% to 4%, yielding a 1% absolute drop (NNT ~90 to 100).
Source: The Lancet (Cholesterol Treatment Trialists Collaboration)
How can SGLT2 inhibitors reduce heart failure deaths even in patients without diabetes?
Although SGLT2 inhibitors (dapagliflozin, empagliflozin) were originally developed to lower blood glucose, trials like DAPA-HF and EMPEROR-Reduced proved that their primary cardiovascular benefit comes from natriuresis, reduced intraglomerular pressure, and improved myocardial energetics rather than HbA1c reduction.
Source: NEJM 2019 (DAPA-HF Trial)
Does lowering HbA1c as low as possible always prevent heart attacks in type 2 diabetes?
No. The landmark ACCORD trial showed that intensively driving HbA1c below 6.0% using high-dose insulin and sulfonylureas actually increased all-cause mortality compared with standard targets (7.0% to 7.9%) due to severe hypoglycemia and weight gain. Choice of cardioprotective drug class matters more than aggressive HbA1c suppression alone.
Source: NEJM 2008 (ACCORD Study Group)