Notes Washed Up in a Bottle

Don. W. MacCorquodale M.D. M.S.P.H.

Periodic Notes From the Field on Philosophy and Science.

ANTIBIOTICS AND HEART ATTACKS

In a recent editorial Dr. David Taylor-Robinson and Jens Boman noted that two recent trials of antibiotics in patients with coronary heart disease, one group with stable coronary artery disease and one with acute coronary heart disease, antibiotic therapy failed to prevent further serious cardiovascular events (The failure of antibiotics to prevent heart attacks.   BMJ 2005; 331:   361-362).  

          They pointed out that three facts regarding the relationship between Chlamydia pneumoniae are clearly established.   Firstly, C. pneumoniae DNA and/or antigen have been detected in 40% of atherosclerotic plaques of patients in various countries.   Secondly, mice and rabbits inoculated with C. pneumoniae have developed inflammatory lesions in arteries.   Lastly, antibiotics do not reduce the incidence of further coronary events in patients, who are already sick.

          Why does antibiotic therapy offer no protective effect in such patients?   Atherosclerosis in humans begins at an early age, and taking antibiotics too late in the inflammatory process is unlikely to have an effect.  

          The editorialists asked if this is the end of the road.   “It is unless a different tack is taken.”   C. pneumoniae or at least its DNA is probably carried from the respiratory tract to atherosclerotic lesions by peripheral blood mononuclear cells.   It is still important to determine whether or not antibiotic therapy would reduce the carriage of C. pneumoniae by peripheral mononuclear cells.   If so, a finding that carriage can be greatly reduced or abolished would provide a logical basis for trials of antibiotic treatment.   Vaccination against C. pneumoniae is another approach that might be taken.

          COMMENT:   A fascinating editorial!   I am not hopeful however that it will be feasible to prevent atherosclerosis by either approach in the near future.

PREECLAMPSIA AND BODY MASS INDEX

          In the developing world, between 20% and 80% of all maternal mortality is attributable to eclampsia.   In the developed world, the perinatal mortality rate among preeclamptic pregnancies is five times greater than that in non-preeclamptic pregnancies (Bodnar L M et al.   The risk of preeclampsia rises with increasing prepregnancy body mass index.   Annals Epidemiology 2005; 15:   475-482).  

          Maternal fatness has long been associated with preeclampsia, however unmeasured confounding may have biased these results.   The authors conducted a cohort study to quantify the independent effect of prepregnancy body mass index (BMI) on the risk of preeclampsia.   Women with singleton pregnancies were identified at outpatient clinics in Pittsburgh private obstetric practices, and of slightly more than 4,000 women eligible to participate, 77% elected to do so.  

The clinics primarily served a population of low income, uninsured, unmarried, biracial women while the practices served middle and upper class women.   Nulliparous women were selected to take part in this study.   Almost 1,500 women were excluded because of pre-existing hypertension, diabetes, or HIV infection.   Some were lost to follow-up.   A total of 1,179 women were included in the final analysis.   Prepregnancy height and weight were self-reported.

Multiple logistic regression was used to estimate the total effect of prepregnancy BMI on the risk of preeclampsia, independent of maternal race and smoking status the year before the index pregnancy.   Other potential confounders were excluded from the model if they did not effect a change in the coefficient of > 8%.

The association between prepregnancy BMI and the risk of preeclampsia is show below:  

BMI                                          Adjusted OR (95% CI)

The authors concluded that “maternal prepregnancy adiposity is a strong, independent risk factor for preeclampsia.

COMMENT:   The powerful gradient shown in the table certainly suggests that the authors' conclusion is correct.   I find it remarkable however that the authors did not collect data on cigarette smoking during pregnancy.   A number of studies conducted in the United States , Canada , Italy , and Latin America and the Caribbean have consistently shown that cigarette smoking during pregnancy is protective against preeclampsia.

          The role of socioeconomic status as a possible risk factor for preeclampsia is unclear.   Unhappily, the data from this study could not be used to evaluate a possible association of social class with preeclampsia.   Fully 90% of the women in this sample were unmarried, and hence, they were overwhelmingly of lower socioeconomic status.   With such a large proportion of women from the same social class, it is not feasible to assess the possible association of social class and preeclampsia.  

RISK FACTORS FOR CARDIOVASCULAR DISEASE IN WOMEN

          It has recently been suggested that risk prediction of cardiovascular disease might be achieved by measuring apolipoproteins B 100 and A-1, non-HDL-C (high-density lipoprotein cholesterol); and lipid ratios such as total cholesterol to HDL-C; LDL-C to HDL-C; apolipoprotein B 100 to HDL-C.   The authors used baseline levels of each of these lipid biomarkers as predictors of first cardiovascular event in a large cohort of initially health women, who were followed prospectively over a ten-year period (Women's Health Study).  

          The mean age at baseline for the 15,600 women (rounded) was 54 years, and the mean body mass index was 26.  

          Strong correlations (Spearman) were found between LDC-C and apolipoprotein B 100 (r = 0.81), between LDL-C and non-HDL-C (r = 0.92), between LDL-C and total cholesterol (r = 0.91), between HDL-C and apolipoprotein A-1 (r = 0.80), between total cholesterol and non-HDL-C (r = 0.94, and most importantly, between non-HDL-C and apolipoprotein (r = 0.87).  

          Over an average of 1- years of follow-up, rates of completed follow-up exceeded 97% for morbidity and 99% for mortality.   During this period, 464 participants developed a first-ever confirmed cardiovascular event (131 myocardial infarction, 122 ischemic stroke, 274 coronary revascularizations, and 76 cardiovascular deaths.  

          Proportional hazards modeling was employed to control for age, blood pressure, BMI, diabetes, and current smoking status.   The strongest association was found between non-HDL-C and apolipoprotein B 100 .   The adjusted hazard ratio (HR) for non-HDL-C was 2.51.   The comparable value for apolipoprotein B 100 was 2.50.   Both non-HDL-C and apolipoprotein B 100 showed stronger association than either total cholesterol or LDL-C.  

          C-reactive protein (CRP) levels were also associated with future CVD events.   The adjust HR for CRP in the highest quintile compared with the lowest quintile was 2.98.  

          The authors commented that they observed “the magnitude of the association was greater for apolipoprotein B 100 than for either total cholesterol or LDL-C.”   They added that the easily calculated ratio of total cholesterol to HDL-C   proved to be at least as strongly associated with CVD events as the ratio of apolipoprotein B 100 to apolipoprotein A-1.   They concluded that their data do not support the use of apolipoprotein B 100 in primary risk detection because non-HDL-C can be directly calculated by subtracting HDL-C from total cholesterol at no incremental cost beyond the usual lipid evaluation.

          COMMENT:   The authors pointed out very importantly that they evaluated plasma levels only once.   I think it is also important to remember that the women in this study were not representative of the entire population of women in our country since they were overwhelmingly from the middle class.