32
Management
Figure 4. Logic for Defining the Absolute Estimated 10-Year
ASCVD Risk for Consideration of LLT at ≥3% in
Primary Prevention
The graphs plot the number-needed-to-treat to prevent 1 ASCVD event (NNT-benefit; y-axis) as a
function of the observed 10-year ASCVD event rate (x-axis) from primary prevention RCTs of statin
therapy. The red curves show the number-needed-to-treat to prevent 1 ASCVD event (NNT-benefit)
with initiation of (A) moderate-intensity or (B) high-intensity statin, assuming a 35% RRR with
moderate-intensity and 45% RRR with high-intensity statin, as has been observed in meta-
regression of statin trials. The horizontal black lines represent the number-needed-to-treat to cause
1 case of incident diabetes (NND) over 10 years with (A) moderate-intensity or (B) high-intensity
statin. The risk of incident diabetes is higher for high-intensity than lower-intensity statins.
Individuals with baseline normoglycemia rarely develop elevations in glucose sufficient to cause a
new diagnosis of diabetes. Individuals who develop statin-attributed diabetes almost universally
have documented prediabetes and are already near the threshold for diagnosis of diabetes.
Number
Needed
to
Treat
(NNT-benefit)
120
100
80
60
40
20
0
Panel A:
Moderate-intensity statin assumptions
RRR for ASCVD: 35%; NND in 10 y: 100
10-y CVD Event Rate
0.0 5.0 10.0 15.0 20.0 25.0
NNT to prevent 1
ASCVD event across
risk spectrum
NND
=100
At 10-y risk levels ≥3%,
more likely to prevent ASCVD
event than cause diabetes
with moderate-intensity statin
