LAUSR

We thank Dr. Trinchieri for his thoughtful comments [1] regarding our paper on body fatness, diabetes, physical activity and risk of kidney stones [2]. With regard to the first point about the geographic location of the studies, we included all available evidence from prospective cohort studies which met the inclusion criteria. We conducted subgroup analyses by geographic location and the association was significantly positive in both American and Asian studies, but somewhat stronger in American studies with a RR of 1.26 (95% CI 1.12–1.41) per 5 BMI units, than in Asian studies, RR = 1.12 (95% CI 1.05–1.18), however, the p value for heterogeneity between subgroups was not significant (p = 0.20). If there is a real difference by geographical region (which we may not have had enough power to detect) it could be because Asian countries are not as advanced in their obesity epidemic as the US. With more people at the extremes of obesity in the US, American studies will also have more ability to detect an association both because of the wider exposure range and larger number of participants at extreme levels of BMI. One additional study from Europe, the EPIC-Oxford study [3], which was excluded because it reported risk estimates for only 2 categories of BMI (3 or more categories of BMI or a continuous risk estimate is required to run linear dose–response analyses with the methods used, as described in the statistical methods), also reported an increased risk of kidney stones with elevated BMI with a hazard ratio of 1.77 (95% CI 1.24–2.54) for a BMI ≥ 27.5 kg/m2 compared to 20–22.5 kg/m2 which is similar to the remaining studies. Although cut-off points for BMI are not identical across studies the confidence intervals for the HRs are overlapping with all the other available studies for the BMI category most resembling the EPIC-Oxford study (Supplementary Table 2 of our paper). We agree with Dr. Trinchieri that further studies are needed in other regions of the world, including the Mediterranean countries. We also agree that a healthy diet with a high intake of fruit and vegetables and less meat may be important for the prevention of kidney stones [3, 4], and it is quite possible that dietary factors may modify the association between adiposity and risk of kidney stones, but none of the studies that were included in our meta-analysis conducted analyses stratified by dietary factors or other risk factors. Any future studies could further look into this to clarify whether there is important effect modification of the association between adiposity and kidney stones by diet or other risk factors. The study Dr. Trinchieri refers to in his letter is a case series with a comparison to general population data [5]. We think more rigorous data is needed before any conclusions can be made to whether a Mediterranean diet modifies the association between adiposity and kidney stones. Lastly, we also agree with Dr. Trinchieri that it would be interesting to know whether there are differences of the associations by subtypes of kidney stones, however, none of the studies that were included in our meta-analysis reported such data. Some of the concerns Dr. Trinchieri has regarding our article may be better addressed at the initial stage of the research studies being done, rather than at the systematic review and meta-analysis stage, as meta-analyses are * Dagfinn Aune [email protected]