Medical Research · Supplementary · from Medical Research
Case 2: Understanding Number Needed to Treat
Patient Presentation
Demographics: 58-year-old male accountant
Chief Complaint: "My cardiologist wants to add another medication and I want to understand if it's really worth it."
History of Present Illness: A 58-year-old male with a history of a non-ST elevation myocardial infarction (NSTEMI) 8 months ago, treated with percutaneous coronary intervention (PCI) and drug-eluting stent to the left anterior descending artery, presents for a second opinion regarding his medication regimen. His cardiologist has recommended adding low-dose rivaroxaban (2.5 mg BID) to his existing dual antiplatelet therapy (DAPT), citing the COMPASS trial, which demonstrated a reduction in the composite of cardiovascular death, stroke, or MI (HR 0.76, p<0.001).
The patient is an accountant and is analytically minded. He wants to understand the absolute benefit of adding rivaroxaban versus the bleeding risk. He is particularly concerned because he developed a GI bleed requiring hospitalization while on DAPT alone 3 months post-MI, which resolved after Helicobacter pylori eradication and PPI therapy.
His cardiologist provided the following data from the COMPASS trial: the composite MACE endpoint occurred in 5.4% of the rivaroxaban + aspirin group versus 7.0% of the aspirin-alone group over a mean follow-up of 23 months. Major bleeding occurred in 3.1% of the combination group versus 1.9% of the aspirin-alone group.
Past Medical History:
- NSTEMI (8 months ago, PCI with DES to LAD)
- Upper GI bleed (3 months post-MI, H. pylori-related)
- Type 2 diabetes mellitus
- Hyperlipidemia
- Peripheral arterial disease (ABI 0.82 bilaterally)
Medications:
- Aspirin 81 mg daily
- Clopidogrel 75 mg daily (planned to continue for 12 months total post-PCI)
- Atorvastatin 80 mg daily
- Metformin 1000 mg BID
- Pantoprazole 40 mg daily
- Metoprolol succinate 50 mg daily
- Lisinopril 10 mg daily
Social History:
- Accountant, works full-time
- Non-smoker (quit 8 months ago at time of MI, 25 pack-year history)
- No alcohol use (stopped after GI bleed)
- Married, two adult children
Family History:
- Father: coronary artery disease, CABG at age 63
- Mother: type 2 diabetes, alive at 82
Physical Examination
- Vital Signs: BP 122/74 mmHg, HR 62 bpm, RR 14/min, Temp 36.8°C, SpO2 98% on room air
- General: Well-appearing male, mildly anxious
- Cardiovascular: Regular rate and rhythm, S1/S2 normal, no murmurs or gallops
- Lungs: Clear bilaterally
- Abdomen: Soft, non-tender, no organomegaly, healed midline scar from prior EGD
- Extremities: Diminished dorsalis pedis pulses bilaterally, no ulcers, hair loss on bilateral lower legs
Workup and Results
Laboratory Studies:
| Test | Result | Reference Range |
|---|---|---|
| Hemoglobin | 13.1 g/dL | 13.5-17.5 g/dL |
| Platelets | 224,000/µL | 150,000-400,000/µL |
| INR | 1.0 | 0.9-1.1 |
| Creatinine | 1.0 mg/dL | 0.7-1.3 mg/dL |
| eGFR | 78 mL/min | >60 mL/min |
| HbA1c | 7.1% | <7.0% |
| LDL | 58 mg/dL | <70 mg/dL (post-MI) |
| Fecal occult blood | Negative | Negative |
| H. pylori stool antigen | Negative | Negative |
NNT/NNH Analysis of COMPASS Trial Data:
| Metric | Calculation | Result |
|---|---|---|
| Absolute risk reduction (MACE) | 7.0% - 5.4% | 1.6% |
| NNT to prevent one MACE event | 1/0.016 | 63 over 23 months |
| Absolute risk increase (major bleeding) | 3.1% - 1.9% | 1.2% |
| NNH for one major bleed | 1/0.012 | 84 over 23 months |
| Net clinical benefit | NNT vs NNH ratio | Marginal (63 vs 84) |
Clinical Image
Educational diagram comparing Number Needed to Treat (NNT) and Number Needed to Harm (NNH) using icon arrays to visually represent the absolute benefit and risk of adding anticoagulation to antiplatelet therapy in secondary cardiovascular prevention. Source: Educational illustration.
Diagnosis
Post-MI Patient on DAPT with Peripheral Arterial Disease, History of GI Bleeding; Considering Addition of Low-Dose Rivaroxaban Based on COMPASS Trial Evidence
Key Diagnostic Criteria:
- Established atherosclerotic cardiovascular disease (post-MI + PAD)
- COMPASS trial-eligible population
- History of GI bleeding increases individual bleeding risk above trial average
- NNT of 63 vs NNH of 84 suggests marginal net clinical benefit at population level
- Individual risk-benefit ratio likely unfavorable given GI bleed history
Treatment Plan
- Decision: Defer addition of rivaroxaban at this time given history of GI bleeding, which places patient at higher individual bleeding risk than trial population
- Continue DAPT (aspirin + clopidogrel) for total 12 months post-PCI, then transition to aspirin monotherapy
- Continue PPI therapy indefinitely given GI bleed history and ongoing antiplatelet use
- Reassess rivaroxaban addition after completing DAPT transition to aspirin alone (net bleeding risk may be lower)
- Optimize other modifiable risk factors: HbA1c to <7%, continued smoking cessation, PAD management with supervised exercise program
- Annual fecal occult blood testing and low threshold for repeat EGD if symptoms recur
- Shared decision-making discussion documented in medical record
Key Learning Points
- NNT (Number Needed to Treat) represents how many patients must be treated for one additional patient to benefit; lower NNT indicates greater treatment effect
- NNH (Number Needed to Harm) represents how many patients must be treated before one additional patient is harmed; higher NNH indicates better safety
- Comparing NNT and NNH provides a framework for evaluating the net clinical benefit of a therapy (ideal: NNT much lower than NNH)
- Individual patient risk factors (such as prior GI bleed) may shift the personal risk-benefit ratio away from population-level trial results
- Presenting absolute risk data and NNT/NNH facilitates shared decision-making and respects patient autonomy