Presentation Authors: Samuel Antoine*, Geoffrey Rosen, Jeffrey Shaman, Gen Li, New York, NY, John Asplin, Chicago, IL, Ojas Shah, New York, NY
Introduction: Kidney stone incidence demonstrates regional and seasonal variation, much of which is correlated with higher temperatures. Other studies have looked at the impact of temperature, humidity and season on 24-hour urine collections with conflicting results. We hypothesize that these patterns may differ when daily temperature and humidity anomalies as well as the entire US is considered.
Methods: All 24-hour urine collections from 2005 to 2015 were procured from Litholink. Local average daily temperature (T) and specific humidity (SH) were compiled from the National Oceanic and Atmospheric Association for each collection date as well as the 6 days preceding it. T and SH anomalies were continuous variables defined as the difference between the observed value and the 30-year average for that date (Tanomaly = Tobserved - Taverage). The first collection from each patient with weight recorded and 24-hour urine creatinine/kg within normal limits were kept for analysis using R (287289 subjects, P < 0.0001). Univariate regressions were performed for urine volume, Na, Ca and supersaturation of Ca-oxalate (SSCaOx) using patient weight, age, gender, T and SH as predictors. Repeat regressions were performed using T and SH averaged over 2, 4 and 7 days leading up to the urine collection.
Results: Results for regressions using T and SH on the day of the collection are summarized in TABLE 1. In general, the sign and significance for the coefficients in each model were consistent whether 1, 2, 4 or 7 days of T and SH were used. Notably, temperature had a significant coefficient in predicting all response variables except Ca. This was true for models using absolute and anomalous T and SH.
Conclusions: In general subjects increased urine volume in response to increased absolute temperature but did not do so when temperature was anomalously high and vice versa. Negative coefficients for Na correspond to an appropriate response to maintain intravascular volume. Interestingly, the coefficients for absolute and anomalous T were not significant in predicting Ca, but absolute SH was negative and significant for Ca. Summer T anomalies had a significant coefficient for volume, but not winter T anomalies, possibly due to increased exposure to ambient temperature during summer months.