Sacral neuromodulation (SNM) has proved to be a safe and effective modality for patients with voiding dysfunctions refractory to or intolerant to conservative treatments. These include overactive bladder (OAB), dry and wet, and nonobstructive urinary retention (NOUR) [1].  Currently SNM is a staged procedure involving a “test phase” (first stage) during which the clinical benefits are evaluated and if successful an internal pulse generator (IPG) is implanted (second stage). Although the test phase represents a crucial factor in treatment, optimal duration is not known [2].
This study evaluates whether prolonged test phase with weekly evaluation increases cumulative success rate and related duration of the test to the one-year outcome of implanted patients. Currently it is not known whether patients who are in need for a longer test phase are as successful as those who respond early (the first week) after a one-year follow-up.

During this monocentric prospective study (August 2015 - November 2018) patients eligible for SNM underwent a tined lead procedure. To reflect real life clinical practice in a tertiary centre, no strict exclusion criteria were used. Patients with neurologic disease were included if the disease was stable. 
During test procedure all possible electrode settings were tested upon stimulation of the tined lead (0-3+/0-1+/0-2+/1-0+/1-2+/1-3+/2-0+/2-1+/2-3+/3-0+/3-1+/3-2+). Square wave pulses with pulse width of 210 µsec and 14 Hz frequency were delivered at the respective electrodes with increasing amplitudes up to the sensory threshold (ST), which was noted together with the location of sensation (LoS) on a pelvic chart grid [3]. Optimal stimulation is perceived at the (peri)anal or genital LoS with a ST < 2 V.
Patients were tested with an external pulse generator for three weeks, with weekly follow ups in hospital with the possibility to change the electrode configurations when lack of efficacy. Success was defined as at least 50% improvement in a key voiding diary variable collected over three days. Key variables for NOUR were the post void residual determined by catheterization, for OAB wet the number of incontinence episodes and for OAB dry the number of voids per day. 
After the first week of stimulation voiding diaries were evaluated and compared to baseline. When the 50% improvement criteria were not met, electrode settings were retested and another configuration was programmed. This procedure could be repeated at the end of the second test week if necessary. Patients with more than 50% improvement after three weeks proceeded to second stage and IPG implant.
Success after 12 months was defined as ≥80% satisfaction on a visual analogue scale (VAS) from 0-100% combined with no wish to receive alternative or add-on treatment. The one-year follow-up results were correlated to the clinical results during test period with patient groups defined as early responders (i.e. after the first week) and those in need for reprogramming.

Ninety patients participated in the study. Median age at implantation was 55,0 +/- 16,9 years (range 18-85yr) with 73 (81%) females and 17 (19%) males. 48 patients (53%) were primarily treated for OAB wet, 34 patients (38%) for NOUR and 8 patients (9%) for OAB dry. Most of the cases had an idiopathic origin. Neurogenic disorders were present in 12 patients (13%). Six patients presented with interstitial cystitis/painful bladder, 5 with nocturnal enuresis and 1 with chronic pelvic pain syndrome.
Fifty-six patients (62%) were successful after three weeks test phase. One major adverse event occurred (infection) and the tined lead was removed in a test failure after two reprogrammings. 
Figure 1 shows an overview of the test phase - with a weekly evaluation – and the percentage of definitive implants. 
After the first week, 33/90 (36,7%) patients were successful and didn’t need other parameter settings during the test phase. After three weeks 32 received a definite implant (97%). One patient unexpectedly returned to baseline at the final analysis. 
Fifty-seven patients did not meet the success criteria after one week. The electrode configuration was changed (reprogrammed) and patients proceeded into the second test week.  Of those 57 patients 29 were successful at the end of week two and were therefore not reprogrammed. 15 of these 29 patients (52%) met success criteria at the final analysis after three weeks. 
The 28 patients considered non-responders after the first and second week were reprogrammed for the second time. Nine of them (32%) were successful at final analysis of the test phase after three weeks and received a definitive implant (Figure 1). 
In patients who received a definitive implant, mean amplitude of the sensory threshold was not different between non-reprogramming/early responders (1,4 +/- 0,8 V) and the reprogramming group (1,5 +/- 1,02 V) (MWHU: p= 0,183). There was no difference in location of sensation between both groups (chi2-test p=0,48). As optimal parameters, no significant difference between the reprogramming and non-reprogramming group was found, respectively 62,2% vs 64,4% (chi2 test: p=0,480).
The chance of a definitive implant for patients with no, one or two reprogramming sessions decreased significantly, respectively 97, 52 and 32%. Cumulative success rate increased significantly. (Figure 2).  
Fifty-five patients had a follow-up of at least one year during this study. One patient was lost during follow-up because of a psychiatric admission in another hospital. 
At one-year follow-up 45 patients (82%) were considered successful. Those who responded early (i.e. after one week) had the same success rate as those who needed reprogramming, 84% (27/32) versus 78% (18/23) respectively (Chi2; p= 0,562). There was no difference in age (MWU; p= 0,222), gender (Chi2; p=0,952) or indication (Chi2; p= 0,975) between both groups.

Current data shows that weekly follow ups during a 3-week test phase with reprogramming in case of lack of efficacy increases number of patients that respond successfully upon the test phase and proceed to implantation and chronic SNM treatment.
During this supervised test phase, 24/56 (43%) responders were identified after one or two reprogramming sessions. In a 7-day test trial, they would have been considered failures for SNM. 
One-year data show patients who require additional programming during the test do equally well as those who respond early and without programming.

Test duration and reprogramming (in case of insufficient clinical benefit) during the test phase of three weeks increases the number of patients that benefit in the long term from SNM. Therefore, a supervised 3 weeks test phase is strongly recommended to become standard practice.

Siegel S, Noblett K, Mangel J, et al. Five-Year Followup Results of a Prospective, Multicenter Study of Patients with Overactive Bladder Treated with Sacral Neuromodulation. The Journal of urology 2018;199(1):229-36 doi: 10.1016/j.juro.2017.07.010[published Online First: Epub Date]|.
Goldman HB, Lloyd JC, Noblett KL, et al. International Continence Society best practice statement for use of sacral neuromodulation. Neurourology and urodynamics 2018;37(5):1823-48 doi: 10.1002/nau.23515[published Online First: Epub Date]|.
Vaganee D, Van de Borne S, Fransen E, Voorham J, Voorham-van der Zalm P, De Wachter S. Repeatability of tools to assist in the follow up and troubleshooting of sacral neuromodulation patients using the sensory response. Neurourology and urodynamics 2019;38(2):801-08 doi: 10.1002/nau.23925[published Online First: Epub Date]|.