Introduction

Percutaneous drainage is a safe and effective alternative to surgery in the management of intra-abdominal and pelvic abscesses. The approach is a challenge for deep pelvic abscesses because of many anatomical obstacles (pelvic bones, iliac vessels, bladder, bowel, uterus, vagina and nerves). A transgluteal computed tomography (CT)-guided approach is preferred when the abscess is inaccessible with a conventional anterior route^1-4.

The aim of this study is to analyze the safety and efficacy of the transgluteal approach in the management of deep pelvis abscesses.

Material and methods

We conducted a retrospective search from 2011 to 2019 and found 10 patients requiring transgluteal drainage. The variables analyzed were patients’ age, origin or cause of abscesses, results of laboratory cultures, complications (categorized using the Clavien-Dindo classification), duration of drainage expressed in days and other variables (as particular characteristics of the patients, among others). A visual analogue scale was used to categorize tolerance to the procedure and postoperative pain.

Transgluteal percutaneous drainages were performed by 3 general surgeons from the Department of Surgery with experience in percutaneous procedures using CT guidance. The diagnosis of deep pelvic abscess was confirmed on an initial CT scan in 8 cases and by magnetic resonance imaging in the remaining patients. All the patients underwent coagulation screen (platelet count, PT and aPTT) before the procedure to identify and correct any coagulation abnormality (low platelet count or coagulopathy).

The Seldinger technique was used in all the cases. The patient is placed in the ventral decubitus position. A first CT scan is performed; a black marker is used to mark the axes for the needle entry point, and the exact depth from the center of the collection to the skin is measured. Local anesthesia (2% lidocaine) was administered at the puncture site. An 18-gauge needle is inserted in the previously marked site and once placed, the collection is aspired, and a sample is taken for bacteriological analysis. A 0.035-inch guidewire is advanced, and a dilator is inserted. The size of the dilator depends on the size of the catheter chosen for the procedure. Then, a multipurpose catheter (Cook or Boston) is placed over the wire. The catheter is fixed to the skin, the purulent fluid is aspirated and then the catheter is attached to a collection bag. Finally, adequate catheter placement is confirmed by CT control (Fig. 1).

Figure 1 Procedure used in our department describing the technique of the infrapiriformis approach. A, the axes are marked for the needle entry point. B, local anesthesia is administered at the site of the trajectory of the drain. C, an 18-gauge needle is inserted. D, CT image confirming the needle placed in the collection. E, the collection is aspired, and the material obtained is sent for culture. F, a multipurpose catheter is inserted using the Seldinger technique. G. the purulent fluid is evacuated through the catheter. H, adequate catheter placement is confirmed by CT control. 

A rigorous daily catheter care was essential to detect any complications (skin puncture site inflammation, pain, bleeding or accidental removal). In addition to the drainage, empiric antibiotics were initially given and adapted to the bacteriological data obtained. Catheter withdrawal was decided according to clinical, biochemical and radiological success criteria. Drainage was considered successful if the abscess had regressed and had not recurred. Conversely, failed drainage was considered in case of abscess recurrence, persistent sepsis or the need for another new procedure.

Results

The variables analyzed are summarized in Table 1. Mean age was 49.2 years (SD: 9.70). Escherichia coli was the most common germ found (50%). In 50% of the cases, the abscesses occurred postoperatively. Mean duration of drainage was 9.2 days (SD: 2.95). The size of the catheters was 8 Fr in 7 procedures and 10 Fr in the remaining 3 patients. Transgluteal drainage was successful in all the cases as there was no need for any additional procedure.

Table 1 

The procedure was well tolerated in 8 patients (80%). Two patients developed moderate pain that was relieved with analgesics. There were no major complications (as bleeding or nerve injury) during or after the procedure.

Discussion

The transgluteal approach requires precise anatomical knowledge of the region to plan a safe approach and thus avoid known complications (Fig. 2). The greater sciatic foramen is an oval space in the posterolateral aspect of the pelvis below the sacroiliac joint, bordered by the sacrum posteriorly, the sacrospinous ligament inferiorly, the ischium anteriorly, and the ilium superiorly. The piriformis muscle originates on the ventral surface of the sacrum and passes through the center of the greater sciatic foramen to insert on the greater trochanter. The sacral plexus rests on the anterior surface of the piriformis muscle and continues inferiorly as the sciatic nerve. The superior and inferior gluteal arteries and veins also cross the foramen in the more cephalic aspect. The sacrospinous ligament, which extends from the sacrum to its insertion on the ischial spine, is an important landmark, as it marks the inferior aspect of the greater sciatic foramen and all the major vascular and neural structures are located cephalad to this ligament. It is important to translate the described anatomic principles to the axial CT images to plan a safe access and thus avoiding major anatomic structures. Typically, for the transgluteal approach the patient is positioned in the prone, prone oblique, or lateral decubitus position. The ideal approach for transgluteal access is to insert the catheter as close to the sacrum as possible, at the level of the sacrospinous ligament. At this level, as previously described, the sciatic nerve is situated more laterally and can be easily avoided. Although the preference is to have an infrapiriformis approach to avoid the gluteal vessels and sacral plexus lying anterior to the piriformis muscle, this is not always possible. In such situations, the sciatic nerve, sacral plexus, and gluteal vessels can be avoided by planning the entry as close to the sacrum as possible³. The Seldinger technique was used in all the cases. We only used local anesthesia, and there was no need for any other type of anesthesia; we asked the patient to inform us in case of excessive pain or pain radiating to the leg (suggestive of sciatic nerve injury), which allowed us to reposition the needle and avoid such injury.

Figure 2 Most relevant anatomical landmarks. A. 1, Iliac; 2, Greater sciatic foramen; 3, Ischial spine; 4, Sacrospinous ligament; 5, Sacrotuberous ligament; 6, Lesser sciatic foramen. B. 1, Piriformis muscle; 2, Ischial spine; 3, Obturator internus muscle; 4, Coccygeus muscle; 5, Levator ani muscle; Arrow, Pudendal nerve. C. 1, Piriformis muscle; 2, Coccygeus muscle; 3, Sacrotuberous ligament; 4, Sacral plexus; Arrow contour, Internal iliac artery; Long arrow, Superior gluteal artery; Short arrow, Inferior gluteal artery. Extracted from Mukesh G. Harisinghani, Debra A. Gervais, Peter F. Hahn, Chie Hee Cho, Kartik Jhaveri, Jose Varghese. CT-guided Transgluteal Drainage of Deep Pelvic Abscesses: Indications, Technique, Procedurerelated Complications and Clinical Outcome. RadioGraphics 2002; 22(6):1353-1367 (published with authorization of the copyright owner). 

As we have previously mentioned in case of deep pelvic abscesses, an anterior approach is not always feasible because of many anatomical obstacles already described and the transgluteal approach is an interesting and safe alternative⁴. Since the initial report by Butch et al. in 1986, few studies have been published about the transgluteal approach for drainage of deep pelvic abscesses⁵. The largest evaluation on this topic was described by Harisinghani et al. who used the transgluteal drainage approach to treat 154 deep pelvic abscesses in 140 patients. Most collections occurred after surgery (115) and 96% of the patients had complete resolution of the abscess. The authors insisted on the fact that the infrapiriform approach was associated with a significant decrease in the incidence of post-procedural pain⁶. Our study confirmed the clinical efficacy of the transgluteal approach with 100% complete resolution as reported in the literature. In our series, the incidence of pain (10%) was similar to the one described in the literature, even with the single use of lidocaine^1,2,7,8. Major complications are uncommon with this approach. There were no cases of bleeding or significant nerve injury in our study. Harisinghani et al. reported only three cases (2%) of pelvic hemorrhage after the transpirifom approach. Two patients with pseudoaneurysms of the inferior gluteal artery were treated with selective arterial embolization and one patient presented a large pelvic hematoma that spontaneously resolved during follow-up⁶. In our cases, we always used catheters with small diameter, as opposed to what Harisimgghani et al. reported, which could have also contributed to the lower rate of complications, as Robert et al. have described¹.

Ultrasound-guided transrectal or transvaginal drainage is an alternative to the transgluteal approach. Nevertheless, this type of approach has some disadvantages, such as the fact that presacral abscesses cannot be accessed transvaginally, and catheter fixation can be very difficult and can cause discomfort for the patient, especially if drainage duration is prolonged^9,10. Although we have performed drainage through this route, we have seldom left catheters in place due to the aforementioned disadvantages, and we consider transgluteal drainage more comfortable and safer for draining pelvic collections.

Finally, the therapeutic use of endoscopic ultrasound (EUS) has gained great popularity as it is safe and effective for draining peripancreatic fluid collections through the stomach or duodenum; thus, EUS-guided drainage of deep pelvic abscesses may provide an alternative in selected patients. Since this procedure was introduced, high success rate has been reported in case series with no major complications. This can be an alternative approach to surgery and percutaneous transgluteal drainage^11,12, although in our environment it is more expensive and more difficult to access.

In conclusion, CT-guided percutaneous transgluteal drainage is a safe and well-tolerated approach for the treatment of deep pelvic abscesses; in our cases we only used local anesthesia and small diameter catheters with a very low rate of complications and high efficiency.