Previous studies demonstrated the feasibility of sensory evoked cortical potential (SEP) recordings for electrical stimulation of the human lower urinary tract (LUT) with heterogeneous methodologies and results. The aim of the current study was to evaluate the impact of different stimulation parameters (i.e. frequency and number of stimuli) on mean amplitudes and latencies of LUT SEPs in order to refine the methodology for efficient evaluation of viscero-sensory afferent pathways of the LUT. We hypothesized that higher stimulation frequencies would lead to smaller amplitudes, while no changes in latencies would be expected. The amplitudes were expected to decrease in the course of a stimulation cycle.

After local ethics committee approval, 40 healthy subjects (age: 23.3±3.3 years) were included. Electrical stimulation of different frequencies (0.5Hz, 1.1Hz, 1.6Hz) was randomly applied at the bladder dome (group: 10 females, 10 males) or proximal urethra (group: 10 females, 10 males) using a 14Ch custom-made catheter. Prior to each measurement, the bladder was filled with 60mL of contrast medium and stimulation intensity was increased as far as tolerable without being painful. Five consecutive runs, each with 100 electrical stimuli, were applied. SEPs were recorded from surface electrodes at Cz referenced to Fz and filtered using bandpass (0.5Hz-70Hz) and notch filter. All measurements were repeated in a second visit using the same order of frequency assessments. Linear mixed models with within-subjects factors frequency and visit and between-subjects factors location and gender were adjusted for stimulation intensity to avoid potential confounding. To compare the individual runs, only subjects with stable SEPs in all five runs were included. A SEP was considered stable when the SEP odd and even runs were in parallel and the components P1, N1 and P2 were clearly identifiable visually.

Across 500 stimuli, stable LUT SEPs with 100% responder rate and the three main components P1, N1, and P2 were recorded for all frequencies, locations, and visits. Linear mixed model revealed significant influence of stimulation frequency on P1N1 (F(2,96)=5.32, p=0.006) and P2N1 (F(2,89)=7.83, p<0.001) amplitudes, but not on the latencies of P1, N1, and P2. Mean amplitudes decreased by increasing stimulation frequency (P1N1- 0.5Hz: 5.1±2.8μV, 1.1Hz: 3.8±2.1μV, 1.6Hz: 3.3±2.0μV; P2N1- 0.5Hz: 9.8±5.0μV, 1.1Hz: 7.3±4.0μV, 1.6Hz: 6.4±3.7μV). No significant effect was found for stimulation intensity, location and visit. Considering runs separately, decreasing amplitudes were observed from run 1 to 5 (P1N1:F(4,481)=14.47, p<0.001; P2N1:F(4,481)=21.09, p<0.001) accompanied by decreasing responder rate. However, by summation of runs, the responder rate could be further increased compared to the first run.

SEPs could be recorded in the LUT with all three frequencies. Higher frequencies resulted in reduced SEP amplitudes, indicating that the choice of the stimulation parameters is crucial. Lower stimulation frequencies such as 0.5Hz might lead to larger amplitudes because of a better susceptibility of the slow fibers in the LUT to these frequencies compared to higher frequencies. The gradual decrease in amplitude and responder rate across runs suggests that the total number of runs (=number of stimuli) can be reduced in order to achieve reliable LUT SEPs and at the same time reducing acquisition time. We assume that habituation respectively rapidly changing bladder volumes leading to electrode dislocation may cause this observed decrease in amplitudes. The size of the peak-to-peak amplitudes is important since LUT SEPs with bigger amplitudes are better detectable and thereby marker setting of P1, N1, P2 gets frequently easier.

We could successfully record SEPs from different locations in the LUT with a very high responder rate and systematically evaluate the impact of several stimulation parameters on LUT SEP outcome. The choice of the stimulation parameters is very relevant for implementation of LUT SEPs into daily clinical practice. Based on the current results, we would recommend applying the slowest stimulation frequency 0.5Hz, because of the biggest amplitudes, but one could reduce the number of electrical stimuli to 200 (2 runs of 100 stimuli) to achieve a faster acquisition of reliable SEPs. This constitutes a good compromise between the duration of a stimulation cycle and peak-to-peak amplitudes of the SEP. Further studies including patients are needed.