AIM: To examine whether there is an independent association between body mass index (BMI) and estimated glomerular filtration rate (GFR) in a large primary care population. METHODS: Anonymous data were sequentially extracted from primary care records between 2006 and 2009 in a primary care population of approximately 220,000 people in Kent, South East UK. Using GFR, BMI, age, gender and comorbidities we examined the association between BMI and GFR. Univariate and multivariate analysis was performed using SPSS(®) (SPSS Inc., Chicago). RESULTS: Sixty-one thousand six-hundred thirty seven people fulfilled the inclusion criteria. There was no correlation between BMI and GFR on univariate analysis. When stratified by BMI, ANOVA demonstrated a statistically significant difference in GFR across BMI strata (p < 0.001). However the absolute differences in BMI between groups were very small. There was a small association between BMI and GFR on multivariate analysis, much of which was lost on adjustment for confounding variables. CONCLUSION: These findings suggest that elevated BMI is not a biologically significant predictor of diminished GFR and therefore may be an insufficiently accurate measure of risk for the metabolic syndrome and CKD.
AIM: To examine whether there is an independent association between body mass index (BMI) and estimated glomerular filtration rate (GFR) in a large primary care population. METHODS: Anonymous data were sequentially extracted from primary care records between 2006 and 2009 in a primary care population of approximately 220,000 people in Kent, South East UK. Using GFR, BMI, age, gender and comorbidities we examined the association between BMI and GFR. Univariate and multivariate analysis was performed using SPSS(®) (SPSS Inc., Chicago). RESULTS: Sixty-one thousand six-hundred thirty seven people fulfilled the inclusion criteria. There was no correlation between BMI and GFR on univariate analysis. When stratified by BMI, ANOVA demonstrated a statistically significant difference in GFR across BMI strata (p < 0.001). However the absolute differences in BMI between groups were very small. There was a small association between BMI and GFR on multivariate analysis, much of which was lost on adjustment for confounding variables. CONCLUSION: These findings suggest that elevated BMI is not a biologically significant predictor of diminished GFR and therefore may be an insufficiently accurate measure of risk for the metabolic syndrome and CKD.