Far be it from a Gynaecologist to presume to invade the turf of his colorectal or gastro-enterological colleagues! Having said that, I do think that pelvic floor ultrasound can contribute substantially to the investigation of anorectal disorders, especially obstructed defecation (that is, the inability to empty completely, and/ or to pass stool without frequent straining), and several recent studies have shown that ultrasound may well be an appropriate replacement for defecation proctography in the initial investigation of defecatory disorders.
The distinction between true rectocele, perineal hypermobility and enterocele has been discussed on the Prolapse page. The depth of a rectocele is easily measured on Valsalva (see Figure 1), and while this manoeuvre (hopefully) bears little resemblance to normal defecation, findings seem to be similar to rectocele depth as measured on defecation proctography. Maximum depth of a rectocele may be reached well before maximal descent of the stool bolus distending the rectocele. Hard stool may in fact reduce defect depth by distending it. Occasionally this effect will be so marked as to obliterate the discontinuity marking the distal margin of the defect. On the other hand, prolapse in other compartments can shield or cover the site of the fascial defect, that is, avoid the negative consequences of a pressure differential between intra- abdominal and atmospheric pressures. In essence, there will be no pressure diferential between the rectal ampulla and the vagina if that part of the vagina is covered by bladder (at intraperitoneal pressure) or uterus (a nondistensible, solid body). Hence, it is not uncommon to find a true rectocele in Theatre in women with major prolapse in whom there was no evidence of a true rectocele on imaging.
Rarely one might find a posterior rectocele (see Figure 2), the clinical significance of which is doubtful. Video 2 shows a rotational volume of a 2nd degree rectocele, measuring about 2 cm in depth. The volume nicely demonstrates spatial relationships. Virtually all rectoceles appear symmetrical or near- symmetrical, suggesting transverse defects of the rectovaginal septum as the underlying aetiology. Figure 3 demonstrates a typical 3rd degree rectocele in the three orthogonal planes and in a rendered volume, showing spatial relationships.
Defects of the rectovaginal septum are found in young nulliparous women (Dietz et al., 2005) but sometimes arise (or enlarge) in childbirth (Dietz et al., 2006). However, other aetiological factors tend to mask the effect of vaginal delivery in older patients. On dissecting the rectovaginal septum in OT (see Figure 4), one has to take great care to strip it off the vaginal muscularis. It is easy to go too deep and miss the septum entirely- this is why some eminent colleagues deny the existence of this structure altogether. Hydrodissection helps to avoid this error. If the septum can be successfully anchored to the top of the vault and the uterosacral ligaments, normal anatomy can be completely reconstituted (see Figure 5).
As opposed to anterior and central compartment prolapse, there is no strong association between the state of the levator and rectocele (Dietz and Simpson, 2008).
The distinction between true rectocele and perineal hypermobility is illustrated in Figure 6. While there is formation of a diverticulum of the rectal ampulla in true rectocele, with a greater or lesser degree of rectal mobility, this is not the case in the latter. There is no diverticulum or pocket, and consequently a lower likelihood of symptoms of obstructed defecation (Dietz and Korda, 2005).
Another entity that can masquerade as a clinical rectocele is enterocele, either in combination with a true rectocele (Video 3) or as an isolated enterocele (Video 4). It is not clear what causes the difference between those three conditions. The location of the Pouch of Douglas, and the cranial extent of the rectovaginal septum is likely to play a role.
Rectal intussusception/ prolapse
On functional pelvic floor ultrasound imaging, rectal intussusception or occult rectal prolapse is occasionally demonstrated in women without any symptoms of evacuatory dysfunction. Such findings are, however, much more common in women with symptoms of obstructed defecation. The method is also capable of demonstrating clinically apparent rectal prolapse. Normally, the anal canal is tubular, with little difference between luminal diameters along most of its length (see e.g. Figure 1 at rest). In less marked cases of rectal intussusception or ‘occult’ prolapse, rectal wall and small bowel enter the proximal anal canal, forcing it open and producing an arrow- shaped distension on Valsalva (see Figure 6 and Video 5), with the anal canal assuming thr shape of a Martini glass. This appearance is pathognomonic and very similar to images obtained on defecation proctography.
Rectal intussusception/ prolapse seems to represent an enterocele that descends into or through the anal canal, rather than into the vagina, and levator ballooning and avulsion are strong risk factors for both (Rodrigo et al., 2011). This explains cases of rectal prolapse after successful vaginal prolapse repair: as descent of the enterocele into the vaginal is barred, the anal canal may become the point of least resistance.
If there is overt rectal prolapse, the enterocele will be seen to ‘flow’ through the anal canal, inverting rectal mucosa, until the prolapse exits through the external anal sphincter (see Video 6). The maximal depth of an intussusception may be measured by connecting the most proximal aspects of the internal anal sphincter and measuring to the apex of the intussusception.
Apart from rectocele and rectal intussusception or prolapse, pelvic floor ultrasound may identify other, less common causes of obstructed defaecation. It appears that an abnormally mobile anteverted uterus may impinge on the rectal ampulla and virtually ‘plug’ it on Valsalva, a situation that is termed a ‘colpocele’ by radiologists (see Figure 7). This may cause the sensation of incomplete emptying and prompt the patient to strain at stool- which only makes matters worse, similar to the situation in rectal intussusception. There may be no other finding that more graphically illustrates how much we still have to learn about pelvic floor function- and how much pelvic floor ultrasound can teach us as well as our patients. Once the situation is demonstrated to the patient on imaging, attempts at behaviour modification may well be much more likely to succeed.
Clearly, much work will have to be done in defining the role of the new method in the evaluation of women with obstructed defecation, in particular in comparison with defaecation proctography. There are a few studies to date that have shown varying agreement between defecation proctography (DP) and pelvic floor ultrasound (see Figure 6), and it seems that, once rectocele or rectal intus-susception/ prolapse are diagnosed on ultrasound, confirmation by DP is probably unnecessary (Perniola et al. 2007). Figures 8 and 9 show comparisons of DP and ultrasound findings in the same patient respectively.
Rectal intussusception in a patient with 2nd degree rectoenterocele and marked levator ballooning.
Video 6: Rectal prolapse. The iso- to hyperechoic mass entering the anal canal represents small bowel.
Figure 1: Quantification of rectocele. The inner contour of the anal canal provides the reference line for depth measurement. From: Dietz et al., Pelvic Floor Ultrasound. Springer London, 2007.
Figure 2: Posterior rectocele on Valsalva. From: Dietz et al., Pelvic Floor Ultrasound. Springer London, 2007
Figure 3: 3D ultrasound of a 3rd degree true rectocele, filling most of the hiatus.
Figure 4: Dissection of the rectovaginal septum (held by forceps), with the defect visible above the septum.
Figure 5: 2D ultrasound findings on maximal Valsalva before (left) and after (right) defect- specific rectocele repair.
Figure 6: Typical findings on rectal intus-susception on maximal Valsalva, with an enterocele entering the anal canal in a typical arrow- shape.
Figure 7: First degree uterine prolapse compressing the anorectal junction on Valsalva, in a patient with symptoms of obstructed defecation.
Figure 8: Rectocele on defecation procto-graphy (rotated for easier comparison) and ultrasond in a patient with symptoms of obstructed defecation.
Figure 9: Rectal intussusception (propelled by sigmoid colon) on defecation proctography and ultrasond in a patient with symptoms of obstructed defecation.
Video 1: Rectocele outlined by intrarectal fluid during urodynamic testing using a fluid-filled catheter. The hyperechogenic area descending into the rectocele is stool.
Video 2: 2nd degree rectocele as seen in a rotational volume of the levator hiatus.
Video 3: Recto- enterocele in patient with marked hiatal ballooning.
Video 4:Isolated enterocele. The rectal ampulla on the far right of the image remains completely immobile.