During the evaluation of urodynamic studies,  rhythmic contractions can be seen in patients with neurogenic bladders and in many instances these contractions are not counted as detrusor overactivity (DO) because they do not meet the predefined criteria for detrusor overactivity and are subsequently ignored.  Recent Work by Colhoun et al. have identified low amplitude rhythmic contractions (LARC) in patients with OAB in muscle specimens, as well as in bladder pressure studies using fast fourier transform (FFT) analysis.  FFT is a form of signal analysis which is capable of taking data that has a cyclic nature and identifying hidden frequencies or periodicity in the data.  
We hypothesized that in many of our neurogenic patients we would be able to find rhythmic patterns using FFT on the raw urodynamic data that is readily available to the urodynamicists that would match the low amplitude contractions that were seen on the urodynamic studies.

The study was designed as a pilot study to evaluate the hypothesis that LARC commonly seen in neurogenic patients urodynamic studies could be identified with FFT based spectral analysis and these would coincide with each other.  IRB approval was obtained for the use of deidentified urodynamic data.  There were 39 urodynamic studies available for review from 21 patients.  Some patients had 2 studies on the same day, while 7 patients had pre and post botulinum A injection studies, others had studies after starting new therapy or an increase in medication doses.   Urodynamic data was exported from a Laborie machine into a text delimited format and then run through MATLAB where a spectral analysis was performed using FFT over the whole time span of the urodynamic procedure.  Approximate entropy  (ApEn) was calculated and displayed using a MATLAB.   Urodynamic curves were compared to the results from the FFT analysis. Results were recorded if there were LARC visible on the abdominal, and detrusor channels and the frequency of the 3 highest  peaks for each channel were recorded.  The slope of the ApEn was compared to the urodynamic curve and confirmation was made whether the entropy rose or declined with increasing bladder volume.  Statistics were performed using Excel (Microsoft, Redmond Washington)  and Xlstat (Addinsoft, Paris France).

39 Urodynamic studies were analyzed from 21 patients (9 male and 12 female) with a mean age of 10.8 (sd=5.1). Spectral density graphs were produced for each urodynamic study and the peaks for each urodynamic parameter were plotted and the first 3 highest peaks if present were recorded and tabulated for the Abdominal and Detrusor curves. (Figure 1).  Invariable a spike was noted in the range of 1 cycle/min or a period of 60 secs in both the abdominal and Detrusor spectral analysis.  In 35 of 39 plots phasic detrusor contractions were identified in the the second spectral band (median 2.7 cycles/min).  In 1 of the 4 who did not have a spectral peak this patient had an underactive bladder, while the other 3 had received botulinum toxin A injections 2- 3 months earlier.  17 of 39 studies had a drop in approximate entropy and 3 of the 4 without phasic detrusor contractions showed no decrease in the ApEn. The patient with underactive bladder did not have a decrease in ApEn nor did he have phasic contractions.  Of the seven patients that underwent Botulinum toxin A injections 6/7 had no decrease in entropy and the one that did had had an injection more than 6 months earlier.   A normality test was run on the data and the data was determined to not follow a normal distribution.

Standardization of urodynamics by utilizing objective criteria is a critical need in urology and especially when medications are being assessed for efficacy.  The present criteria we have at this time relies on a large amount of subjectivity and can leave things open to interpretation.  One common problem in pediatric patients with neurogenic detrusor overactivity (NDO) is that in many cases low level phasic contractions are ignored as not being significant.  It is critical in this NDO population that these contractions be eliminated since continued contractions can be detrimental to the patients over the long-term.  Utilizing FFT analysis of the pressure curves allows for identification of signals that may indicate bladder overactivity and in some cases we have seen spectral evidence of contractions that were not apparent but were ameliorated after Botulinum toxin A  injections.  We also saw that the ApEn decreased in almost all patients after Botulinum toxin A  injections indicating that there may be a beneficial effect by Botulinum toxin A  in eliminating LARC in NDO bladders.  As the ApEn decreases we see an increase in order which is associated with an increase in the phasic contractions amplitude on the urodynamic tracings.  This is exactly what we would expect to see; the increase in order leads to concerted contractions throughout the detrusor thereby increasing the amplitude of the contraction.  Increasing ApEn would lead to chaotic contractions and thereby not lead to phasic contractions.  Our findings diverge minimally from the work by Colhoun et al.,  we observed a frequency was an average of 2.7 cycles/min which has a period of 22 sec between contractions, while they observed a frequency of 2.34 cycles/min or a  period of  25 sec between contractions.  Whether these differences are significant or just differences in technique it is reassuring that the numbers are relatively close together indicating that the technique is reproducible.

We have seen that use of FFT based spectral analysis  can produce results that give important data that can substantiate the effects of medical treatment in this pilot study.  Additionally our data matches the results noted by colhoun et al in their series of patients with detrusor overactivity due to overactive bladder.  This indicates that there is a similarity and a regularity to the overactive detrusor contractions seen in both OAB and NDO patients.  Further work needs to be done to further improve the predictive power of this technique.

Low amplitude rhythmic contraction frequency in human detrusor strips correlates with phasic intravesical pressure waves.  Colhoun AF, Speich JE, Cooley LF, Bell ED 3rd, Barbee RW, Guruli G, Ratz PH, Klausner AP.World J Urol. 2017 Aug;35(8):1255-1260. doi: 10.1007/s00345-016-1994-0. Epub 2016 Dec 26.