Currently the apolipoprotein B:AI ratio integrates information about the potential for cardiovascular disease (CVD) risk reduction better than any other lipid or lipoprotein index. Certainly it could, with benefit, replace serum cholesterol and HDL cholesterol in the estimation of CVD risk. Defining the therapeutic target of statin therapy in terms of serum apolipoprotein B (apo B) rather than LDL cholesterol could also help to optimize statin treatment. Deciding whether a therapeutic response is adequate also requires knowledge of whether there is persisting hypertriglyceridaemia, because this gives an indication of whether small dense LDL is likely to have been satisfactorily reduced. Raising low levels of HDL, probably best measured as apo AI, may also prove to be an important aim of treatment. This is, however, a more complex issue and also depends on the mechanism by which a particular therapy alters HDL levels and on whether the capacity of HDL to perform its anti-inflammatory and antioxidative functions is restored. A meta-analysis of randomized clinical trials of statins in which apo B and apo AI have been reported could provide valuable information.