A type II diabetic that progresses to need insulin still has enough endogenous insulin production that, while they can get very high sugars, they don't develop ketoacidosis because they can still utilize that sugar to some degree. You also need to realize the limit of glucose control. Nobody has ever shown a reduction in cardiovascular events from a glucose lowering trial. The ACCORD study just halted the glucose arm of the study a month ago because the intervention (tighter sugar control with HbA1c < 6.0) had more adverse events / mortality than the "standard" treatment (in this case a HbA1c between 7 and 8) the investigators have switched to targeting everyone to HbA1c 7 to 8 for the remainder of the trial. This is even higher than the usual HbA1c < 7 target in conventional practice.
There is a significant mortality / morbidity benefit to BP control in a diabetic. There is a similar benefit to statin use, although I personally feel that the number needed to treat an otherwise well controlled diabetic may not make it worth it for that individual to take the drug. You are right that # LDL particles may be a better measure than LDL itself (and I commend your depth of knowledge) but I would only say that was true in the setting of following patients on treatment. As you point out, contrary to what most people believe, achieved LDL levels on treatment do not correlate with outcome, only dose of drug does. Number of particles can be assessed by measuring ApoB levels (essentially one apoB particle per LDL molecule even though lipid volume of the particle can vary). On treatment changes in ApoB do seem to correlate with outcomes. As a baseline (pre-treatment) assessment of risk however, ApoB is no better a predicter than the more conventional lipid ratio (TCOL / HDL).
Market Data 
Markets 
AI
