To the Editor:
Tellez-Plaza et al1 conducted a long-term follow-up study of 3348 adults 45 to 74 years of age at baseline to estimate how well urinary cadmium predicts incident cardiovascular disease (CVD) and CVD mortality.
Cadmium exposure was a risk factor for incident CVD and CVD mortality, but several potential risk factors for CVD are also informative for understanding the contribution of cadmium exposure to CVD prognosis. The authors presented data stratified by participants’ characteristics in the eAppendix. Unfortunately, this information does not include the risk of age, sex, education, body mass index, smoking, hyperlipidemia, hypertension, diabetes, and glomerular filtration rate simultaneously in combination with urinary cadmium. As a first query, the contribution of each risk factor is needed to understand the importance of urinary cadmium for predicting CVD incidence or mortality.
Second, sex differences should be cautiously interpreted.1 In their Table 1, the percentage of men linearly decreased as urinary cadmium concentration increased, whereas the percentage of current smoking linearly increased. From these data, a higher prevalence of smoking in women compared with men would be expected if the level of urinary cadmium was mainly caused by smoking. However, I previously reported that the effect of smoking on urinary cadmium was clear only in men. An increase in urinary cadmium by creatinine adjustment was predominant in women,2 which would be caused by the sex difference in the amount of muscle. Considering these associations, we recommend analysis stratified by sex (as was partly presented in the supplemental materials), rather than adjustment for sex.
Third, the geometric mean of urinary was 0.94 μg/g creatinine. I previously reported a renal effect of low level of occupational exposure to cadmium, with geometric mean of urinary cadmium as 1.02 μg/g creatinine.3 In our study, urinary excretions of N-acetyl-beta-D-glucosaminidase was positively associated with urinary cadmium, although a clinically significant increase in urinary N-acetyl-beta-D-glucosaminidase (reflecting the damage of proximal tubules of the kidney) was not observed. From my experience, in combination with past reports, the effect of cadmium exposure on incidence and mortality of CVD should be specified by considering sex differences.
Department of Hygiene and Public Health
Nippon Medical School
Bunkyo-Ku, Tokyo, Japan
1. Tellez-Plaza M, Guallar E, Howard BV, et al. Cadmium exposure and incident cardiovascular disease. Epidemiology. 2013;24:421–429
2. Kawada T, Shinmyo RR, Suzuki S. Urinary cadmium and N-acetyl-beta-D-glucosaminidase excretion of inhabitants living in a cadmium-polluted area. Int Arch Occup Environ Health. 1992;63:541–546
3. Kawada T, Tohyama C, Suzuki S. Significance of the excretion of urinary indicator proteins for a low level of occupational exposure to cadmium. Int Arch Occup Environ Health. 1990;62:95–100