There is now overwhelming evidence that the total cholesterol in the blood (TC) is not a reliable risk factor for heart disease. In Blog 38 I pointed out that the death rate due to heart disease in Belfast is 4 times that of Toulouse even though the TC levels are almost the same. In Blog 39 those who consumed a Mediterranean diet reduced the heart disease death rate by almost 70% when compared with the control group which did not alter its’ diet. However the TC levels of both groups were virtually identical. Therefore efforts to monitor and reduce TC are largely ineffective and a huge waste of resources.

Despite this consideration of the different types of cholesterol is proving to be useful. This has led to studies on the LDL(which stands for Low Density Lipoprotein) cholesterol and HDL(High Density Lipoprotein) cholesterol. The LDL C has been labelled “bad” while the HDL C is considered to be “good”. Here is some of the background.

In the Framingham Study, monitoring of 2815 men and women aged 49 to 82 years and free of CHD commenced between 1969 and 1971. Over the next 4 years CHD developed in 79 of the 1,025 men and in 63 of the 1,445 women. It was found that the major potent risk factor was the HDL C which had an inverse association with the incidence of heart disease (ie. the higher the HDL C the lower the risk). There was a weaker direct association for the LDL cholesterol (LDL C). At these ages there was no relation between TC and the risk of CHD (1).

After a follow-up period of 12 years a further paper was published which confirmed the inverse relationship between HDL C and the risk of experiencing a myocardial infarction(MI) or heart attack(Table 1) This was especially pronounced for women as those with the highest HDL C had less than one quarter of the risk compared with those who had the lowest HDL C values. When this information is broken down by the different quartiles of TC it is clear that the decline in incident rate as the HDL C increases applies to all TC values (Table 2). This was particularly marked for the lowest TC in men and for all TC values in women. These results emphasise that at any given TC level the incident rate can vary enormously depending on the HDL C value and demonstrate the limitations of relying entirely on screening for TC alone as a means of identifying those at risk of a heart attack(2).

TABLE 1.  HDL Cholesterol and Myocardial Infarction (Rate/100)

HDL Quartile, mg/dL Total No of events Age adjusted rate
1,12-36 239 36 15.3
2.37-44 275 41 14.8
3,45-52 231 34 14.9
4,53-129 262 25 9.4
Total 1007 136 13.5
1,23-46 356 38 10.6
2,47-55 354 26 7.4
3,56-66 353 23 6.5
4,47-139 355 5 1.4
Total 1418 92 6.5


TABLE 2. Variation of Myocardial Infarction (Rate/100) with HDL Cholesterol and Total Cholesterol

                     Total Cholesterol   Quartile, mmol/L
HDL Quartile 1, 3.0-5.0 2, 5.0-5.6 3, 5.6-6.3 4, 6.3-9.7
1, 17.1 17.0 12.9 14.3
2, 9.5 18.0 18.3 14.2
3, 7.0 12.5 20.2 19.7
4, 2.2 13.1 8.7 10.9
  1, 3.2-5.5 2, 5.5-6.1 3, 6.1-6.9 4, 6.9-10.9
1, 9.4 7.6 12.6 13.1
2, 5.4 7.8 5.6 10.7
3, 6.0 5.9 5.3 8.9
4, 1.3 0.0 2.3 2.0


In another report based on essentially the same raw data, information was provided on all-cause mortality as well as CHD (3). The results in Table 3 confirm that there is an inverse association for both men and women between the HDL C and the risk of CHD. For cancers the death rate in men is similar for all HDL C levels but nevertheless for all-cause mortality an inverse relationship with HDL C is still evident. The results for cancer do not show any consistent pattern but for all-cause mortality the inverse relationship with HDL C is evident. Although heart disease is responsible for a high proportion of the deaths which occur in the USA and countries in Northern Europe it is crucial to take account of the other causes of death which are not necessarily influenced by the same factors. The information in Table 3 confirms that the death rate due to CHD and to cardiovascular diseases (CVD) generally falls as the HDL C increases. However the death rate for cancer appears to be independent of the HDL C level.

TABLE 3. Variation in the 12-year relative rates of death due to various causes.

                MEN                     Causes of Death
HDL Quintile, mg/dL All causes CVD CHD Cancers
1,12-34 1.92 3.58 4.09 1.17
2,35-40 1.28 1.46 1.99 1.10
3,41-46 1.27 1.87 2.76 1.10
4,47-54 1.17 2.37 2.77 0.80
5,55-129 1.00 1.00 1.00 1.00
1,23-44 1.47 1.55 3.07 1.08
2,45-51 1.17 1.82 3.94 1.09
3,52-58 0.99 1.27 2.60 0.49
4,59-69 0.99 1.27 2.60 0.49
5,70-139 1.00 1.00 1.00 1.00


The recognition that the level of HDL C is inversely associated with the risk of developing heart disease coupled with the many reports that there is an increase in the all-cause mortality at low levels of TC has eventually resulted in a shift of emphasis to the different forms of cholesterol.  It is now common practice to monitor the HDL C levels and this is taken into consideration in assessing the risk of developing heart disease. Nevertheless the need to reduce TC has been maintained even though the rationale for this approach can no longer be sustained.


  1. T Gordon et al (1977) The American Journal of Medicine 62 (5) pp 707-714
  2. R D Abbott et el (1988) Arteriosclerosis, Thrombosis and Vascular Biology 8 (3) pp 207-211
  3. P W Wilson et al (1988) Arteriosclerosis, Thrombosis and Vascular Biology 8 (6) pp 737-741.