There is currently much interest in searching for discriminatory urinary biomarkers for overactive bladder (OAB), as well as to understand more about how this condition is related to the release of transmitters from the bladder wall.  Urinary ATP is well placed to be such a biomarker as it is released from the urothelium in response to bladder wall stretch, and a large fraction of that released will end up in urine. 
Urinary ATP concentrations have been measured in a number of studies to assess its usefulness as a biomarker.  However, little is known about how confounding factors such as subject hydration status and time since the previous void affect its concentration even in healthy controls without bladder problems. To be able to fully understand any changes in the amount or release patterns of ATP in OAB it is an essential prerequisite to understand how its levels vary in healthy people with normal bladder function.
The aim of this study was twofold: firstly to determine the relation between urinary ATP concentration and the time since a previous void; and secondly to determine whether normalising ATP levels to markers of urine dilution will normalise any variation in hydration status in the subject.

Urine samples were collected at 15, 30, 60, 120 and 240 minutes following a previous void from 5 male and 7 female volunteers without LUTS. Urine aliquots were immediately frozen at -20°C for later measurement of ATP (luciferin-luciferase assay) and creatinine concentration (Cayman colorimetric kit). Data are mean±SEM, data sets were compared with ANOVA and post hoc unpaired t-tests. Void volume was used to calculate bladder filling rate.

ATP concentration was higher in women than men at all the void times studied. The mean values from all time points being 5.9±0.7 nM in females compared to 2.6±0.4 nM in males (p<0.001) and shown in Figure 1a. Interestingly, the concentration of ATP in urine was also higher at shorter times between voids than than at longer times in both men and women (Figure 1b). This difference being significant in women (10.4±2.5 nM (15 mins) vs 3.3±0.4 nM (120 mins) and 3.6±0.6 nM (240 mins)  (p=0.01). 
If urinary ATP concentration is dependant on urine dilution it would be predicted that the ATP concentration should correlate with markers of urine dilution. However, urinary ATP concentration did not correlate with either creatinine concentration or rate of filling in either gender. This is shown in Figure 1c where ATP concentration is plotted as a function of the time to produce each ml of urine calculated from the void time divided by the void volume.

These results show that the total quantity of ATP in urine increases with time since the previous void, however the concentration decreases as bladder volume increases.  Assuming a constant rate of urine production between voids  these data indicate that ATP quantity increases at a lower rate than the increase of urine volume.  This may be due to an increasingly lower rate of ATP production by the bladder wall as it stretches and/or that ATP is continually being broken down whilst in urine, by hydrolysis in solution and augmented by any ATPases on the urothelium surface.

The results presented here indicate that urinary ATP concentration is dependent on the time since the previous void and not urine dilution. This suggests that normalising urinary ATP concentration to creatinine concentration is misleading and caution must be applied when comparing ATP concentrations in urine from OAB patients with raised urinary frequency, in comparison to those from control subjects.