Letter by Pischon et al Regarding Article, “Adiponectin and Coronary Heart Disease: A Prospective Study and Meta-Analysis”
To the Editor:
We read with interest the report by Sattar and colleagues1 that shows no significant relationship of plasma adiponectin levels with risk of coronary heart disease in the British Regional Heart Study. In a summary of published studies, Sattar et al1 also report only a modest inverse association. However, we have concerns about their pooling of only a few studies when several different methodologies, populations, and analysis methods were used. For example, the Health Professionals Follow-Up Study was the only study that had blood specimens stored in liquid nitrogen at temperatures <−130°C.2 Adiponectin was significantly inversely associated with coronary heart disease risk in the Health Professionals Follow-Up Study.2 In the British Regional Heart Study, in which serum was stored for approximately 16 years at −20°C, reported adiponectin levels were considerably lower, and correlations of adiponectin with high-density lipoprotein cholesterol (r=0.33) and triglycerides (r=−0.25) were substantially weaker as compared with the Health Professionals Follow-Up Study (r=0.44; r=−0.39). The body mass index levels between the 2 studies were similar, but the correlations with body mass index were also weaker, even though body mass index was measured in the British Regional Heart Study (r=−0.21) but self-reported in the Health Professionals Follow-Up Study (r=−0.27). Therefore, additional data are required to better understand the impact of blood storage methods and characteristics on levels of adiponectin.
We were surprised to read that Wolk et al3 and Zoccali et al4 had measured adiponectin using a radioimmunoassay by Linco (St Charles, Mo) in fresh samples at baseline in 1997 to 1998, although—to our knowledge—this assay had not been introduced to the market at this time. Further, the publication by Wolk et al3 does not contain information about adiponectin. It would be informative to know how the authors identified these unpublished data.
Meta-analyses should be used to provide greater insight across the quantitative distribution of biomarkers instead of ignoring these differences across studies. It is also unclear to us why the authors have chosen tertiles for comparison, because the size of their study would have allowed more informative comparisons. Previous studies, including their own prior analysis, have reported relative risks for quartiles or quintiles.2,5 Our observed relative risk of 0.56 (95% CI: 0.32 to 0.99) for quintiles2 was reduced to a relative risk of 0.65 (0.44 to 0.98) in the meta-analysis by the use of tertiles.1 Interestingly, this estimate of effect is very similar to the results these authors have reported in their previously published study among women (relative risk of 0.63; 0.36 to 1.08).5 Further, although not documented in the meta-analysis, the studies that were included had different degrees of adjustment. Clearly, much more data are needed before meta-analyses can clarify the potential and important inverse association between adiponectin and clinically diagnosed coronary heart disease.
Sattar N, Wannamethee G, Sarwar N, Tchernova J, Cherry L, Wallace AM, Danesh J, Whincup PH. Adiponectin and coronary heart disease: a prospective study and meta-analysis. Circulation. 2006; 114: 623–629.
Zoccali C, Mallamaci F, Tripepi G, Benedetto FA, Cutrupi S, Parlongo S, Malatino LS, Bonanno G, Seminara G, Rapisarda F, Fatuzzo P, Buemi M, Nicocia G, Tanaka S, Ouchi N, Kihara S, Funahashi T, Matsuzawa Y. Adiponectin, metabolic risk factors, and cardiovascular events among patients with end-stage renal disease. J Am Soc Nephrol. 2002; 13: 134–141.