Effects of surgery
Before discussing the results of surgery, the first important issue is about the interest of a bilateral decompression. The benefit – risk ratio must be studied. The results of Shafik were obtained after bilateral decompression [5, 29, 30]. For urinary and anal incontinence, it seems logical to treat both nerves because the sphincters have a bilateral innervation and if one nerve is suffering maybe there is a problem on the other. The EMG exploration is not very sensitive and doesn't study the sensory pathway, which could be very important for continence. For the same reason, bilateral decompression seems logical in the treatment of proctalgia. For unilateral pain, the dilemma is more important. The risk is to induce pain at the "normal" side. Until now we have been performing bilateral operations without such a side effect but a controlled randomised study would be necessary to conclude.
The treatment of pain starts with a holistic approach of the woman (drugs, psychotherapy, relaxation...) with exclusion of other causes of pain: piriformis syndrome, coccygodynia, interstitial cystitis, endometriosis... The other neurological causes must be excluded by a complete electrophysiological study of the perineum (sacral latencies, PNTML, detection EMG and sensory evoked potentials) and imaging of the spinal cord . If the diagnosis is confirmed an infiltration of the Alcock's canal under scanner control can be tried. This infiltration is successful in 57% in the short term but only in 15 % of the cases after one year . It can be repeated maximum 3 times to avoid a nerve irritation. In the treatment of pain the results of this study are similar or better than those obtained in previous studies [32–34].
Even with the transgluteal approach where the "clamp" between the sacro-spinal and the sacro-tuberous ligament is opened by sectioning the sacro-spinal ligament, the cure rate remains around 50%. In the 4 cases of proctalgia fugax the results were better (3 cured and 1 improved). Shafik  had also very good outcomes with this type of pain (100% cured). In this study, the results were worse if the pain was bilateral.
For Robert early diagnosis appears to be the determining factor in improving results . He used the infiltration of the Alcock's canal with lidocaïne-corticoïds as a test before operating. According to him a sufficient pain relieve, lasting during a short period, is a good indication for surgery. Mauillon et al also thinks that complete disappearance of pain for at least two weeks after a nerve block repeated twice before surgery may be the best criterion to predict success . In this study, the patients presenting with perineodynia only (n = 10; Figure 1) had an infiltration before surgery but the number of cases was not sufficient to give a relevant impression about the infiltration test.
For anal incontinence, our results were in the same range as Shafik . In a previous study we also had similar results [36, 37]. The exclusion of the patients who had sphincteroplasty, levatorplasty (possible post-anal repair effect [27, 28]) and/or a cure of rectocele did not change the cure rate.
More interesting was the cure rate in the group of patients with a clear rupture of one or the two anal sphincters. The traumatic rupture of the anal sphincter (delivery, sphincterotomy...) usually induces an immediate anal incontinence. In the patients who remained continent the power of the broken muscles remain sufficient to avoid flatus or faeces leakage. In the long term the continence is probably maintained with the help of the fibrous tissue located between the two edges of the ruptured muscle which acts like a bridge and therefore enables the sphincter to be efficient during many years. The aging process of the muscle and the pudendal neuropathy reduce the power of the muscle (and probably the sensibility in the anal canal) with time and explain the appearance of an anal incontinence. Therefore it is logical to restore continence by improving the conduction in the pudendal nerve. This fact can also explain why the results of sphincteroplasty decrease with time especially in the non diagnosed or treated pudendal neuropathies .
The fact that 5 patients were cured from their anal incontinence only 2 years after surgery emphasized the importance of a long follow up period to obtain relevant cure rates. Surprisingly, the cure rate seems to be not dependant of the degree of anal incontinence but the number of solid incontinences (5 cases; 3 cured and 2 improved) was too small to validate this impression.
The results of the pudendal nerve decompression seem to be equivalent to these of neuromodulation  and the procedure is far less expensive because there is no need for a special material. If this study is confirmed by others, the treatment of the neuropathy should be done before any trial of neuromodulation. In fact it is logical to repair the electric cable before enabling the current to pass.
For urinary incontinence the number of cases is too small to give a relevant cure rate but there were enough cases to suggest that this surgery can treat some patients with stress or urge incontinence.
In a previous study, 3 of the 7 patients presenting a stress urinary incontinence were cured by bilateral pudendal nerve decompression alone [36, 37].
In Shafik's study 6 patients were cured from their stress urinary incontinence, 5 improved and one was a failure . For this author, the efficacy of the pudendal nerve decompression on stress urinary incontinence is due to an increase of the external urethral sphincter EMG activity and to a decrease in the straining urethral reflex latency (time between the expiration involved with the cough and the first deflection of the reflex muscle action potential complex) and PNTML. For Shafik  the increase of urethral pressure during abdominal hyperpressure is not only passive but is induced by an active contraction of the urethral sphincter. After an injection of lidocaïne in the sphincter the urethral hyperpressure was suppressed. Thind & al. clearly demonstrated the role of the pudendal nerve in urinary continence. These authors showed a clear reduction of the maximum urethral pressure and a decrease in the adjunctive urethral closure forces during stress after bilateral pudendal blockade [41, 42]. This is also in agreement with the study of Constantinou which demonstrated that a fast-acting contraction occurs in the distal third of the urethra 240 plus or minus 30 msec before the bladder hyperpressure . Furthermore, Ko and Kim demonstrated that pudendal nerve block with a 7% phenol solution is very effective in the treatment of external urethral sphincter hypertonicity in patients with spinal cord injury .
This study is the first one dealing with a possible effect of the pudendal nerve decompression on urge incontinence. It is probably due to a better control of the urethral sphincter which can reduce urethral instability  and improve the inhibition of the detrusor activity.
One weakness of this study is the rough evaluation of the symptoms. We did not use any scoring system, pad test, quality of life questionnaires or "visual analog pain scale". Furthermore the number of anal and urethro-vesical manometries done before and after PND was too small to give relevant results.
The objective evaluation of PND was done using two neurophysiological tests and the clinical examination. Like Shafik [29, 30] we found a significant increase in anal richness on EMG, a significant reduction of anal and perineal PNTML after surgery and a significant reduction in the frequency of the clinical signs. The skin rolling test was improved as much as the perineal sensibility and the Alcock's canal pain, thus showing its relevant link with the PCS.
Evaluation of the clinical signs and minimal criteria needed for the diagnosis
Shafik described many clinical signs of the PCS [5, 29, 30]. In this study only three signs were studied carefully. Shafik described two of them: abnormal perineal sensibility and pain during the palpation of the Alcock's canal by a rectal examination. The third one is the "skin rolling test" which is well known in the diagnosis of neuralgia in other parts of the body . This study is the first one in which this test was utilized as a clinical sign of PCS.
Compared to patients with negative clinical signs, those having positive clinical signs have a 4,42 ; 5,52 and 6,56 higher likelihood of PCS for "Abnormal sensibility", "Painful Alcock's canal" and "Painful skin rolling test", respectively (Table 10). When patients with all three signs positive are compared to patients with all three signs negative, the odd ratio is 16,97 (Table 11). All the estimated 95% confidence intervals for the odds ratios are significantly higher than 1, indicating that the clinical signs can be considered as valuable signs in the diagnosis of PCS.
The most specific sign was the "Painful skin rolling test" and the most sensitive was the "Painful Alcock's canal". The association of the three positive tests had a very high specificity in the diagnosis of a PCS (89 %). This high specificity was confirmed by the low frequency of this association after the operation (return to the same level as the control group).
Therefore, in some cases the clinical examination should be sufficient to prove the existence of a PCS. For example, a patient presenting anal incontinence, an intact sphincter proven by ultrasound and the three clinical signs positive has almost certainly a PCS. Of course, from a scientific point of view it is still interesting to perform a complete electrophysiological study and a precise neurological examination to exclude a central problem (multiple sclerosis, tumor...) or a polyneuropathy.
However, making the diagnosis of PCS is not usually an easy task. Many times, there is a high degree of suspicion but, as in many illness, all the symptoms or signs are not present. In this study, we decided to operate when at least two of the five clinical and neurophysiological signs described in the methods section were associated with one or more of the 3 classical symptoms (perineodynia, anal incontinence and urinary incontinence). At the beginning of this study, it was usually "increased PNTML" and "painful Alcock's canal". With the introduction of the "skin rolling test" and of the "sensibility test", clinical examination became more important in the decision. The more symptoms (especially anal incontinence and perineodynia) and signs were present, the more confident we were in the diagnosis of PCS. Further studies are necessary to validate this and to define more precisely the minimal criteria needed for the PCS diagnosis.
The pudendal nerve decompression by the perineal route is a blind procedure. The search for the inferior rectal nerve and the opening of the Alcock's canal are done under finger control. In our experience it is not easier with retractors. Therefore it is necessary to have a clear anatomical vision of this area before performing the operation.
Maybe the use of a laparoscope would help  but the procedure will become more expensive and time consuming. To suppress the blind character of the procedure the transgluteal approach proposed by Robert  or the more recent transvaginal approach from Bautrant  could be other ways to treat the PCS. Until now the results on pain are the same as those obtained by the Shafik's approach but with the concurrent sections of one or two ligaments of the pelvis (sacro-spinal and/or sacro-tuberous ligaments). However, we should be aware that the long term effects of these sections on the stability of the pelvic region are until yet unknown. Therefore, if the "clamp" must be open efforts should be done to open it without cutting a ligament. Up to now no data are available about a potential effect of the transgluteal or transvaginal procedures on urinary or anal incontinence.
Despite the blind character of the procedure we only had one heavy bleeding probably coming from the pudendal artery. One patient still presents with a mild intermittent clitoridal pain and a worsening of anal incontinence. Because the nerve of the clitoris leaves the pudendal nerve just before the entrance into the Alcock's canal this problem is probably the result of a too large dissection in the upper part of this canal. The two cases of anal incontinence worsening (gas incontinence becoming liquid incontinence), including the aforementioned patient, are difficult to explain. Maybe the neuropathy increased with the stretching involved in the procedure, the scarring process or a too large dissection. It could also be the result of the changes in the posterior level anatomy induced by concomitant procedures (easier expulsion of gas or faeces). For the 2 patients the EMG data and the clinical examination after the operation did not improve therefore showing that the neuropathy was not healing.
Because the roughly estimate prevalence of PCS is around 20%, this "defect" seems to be a very frequent problem in Perineology. Therefore it should be logical to search for it in each clinical examination of a patient presenting with prolapse, perineodynia, urinary or anal incontinence.