| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Department of Radiology, University of Western Ontario, London Health Sciences Centre, 375 South Street, London, ON, Canada N6A 4G5. e-mail: sfmillward@rogers.com
Editor:
We read with interest the article by Dr Harisinghani and colleagues in the September 2003 issue of Radiology (1), in which they report their extensive experience with the transgluteal approach for percutaneous drainage of deep pelvic abscesses in 140 patients. This case series expanded on earlier work by the same authors published in RadioGraphics in 2002 (2).
When achieving transgluteal access through the greater sciatic foramen, the authors advocate an infrapiriformis approach in which the catheter is inserted as close to the sacrum as possible, at the level of the sacrospinous ligament below the piriformis muscle, to avoid injury to the sciatic nerve, sacral plexus, and inferior gluteal vessels (1). To illustrate the technique, the authors provide three figures that show computed tomographic (CT) images with catheters or needles in place. In two of these images, the catheter or needle is too cephalad, too lateral, or both. Despite their extensive experience, the authors report near misses of the sciatic nerve in four (3%) of their 140 patients; fortunately, there were no permanent sciatic nerve injuries.
Furthermore, they report injury with pseudoaneurysm formation in the inferior gluteal artery in an additional three (2%) of their patients. We are concerned that this 5% complication rate could be increased considerably in the hands of operators with less experience, if these operators modeled their technique after reading the two articles of Dr Harisinghani and colleagues (1,2).
To achieve a CT-guided puncture as low as possible, we believe that caudocephalad angulation of the CT gantry should be used. This was mentioned by Dr Harisinghani and colleagues (1,2), but only very briefly. We would like to offer the following modification and/or addition to the technique described by the authors (1,2).
The patient is placed prone (or occasionally prone oblique or lateral decubitus) on the CT table, and preliminary CT scans are obtained with up to 20 degrees of caudocephalad gantry angulation. After choosing an appropriate skin entry position, local anesthetic is injected, and further CT images are obtained with the local anesthetic needle in place to confirm that the needle is indeed close to the sacrum, below the piriformis muscle, and well away from nerves and blood vessels. An 18-gauge needle is then introduced along the same path by using repeat scans to chart its path as necessary. Having entered the abscess, a catheter is then introduced by using the Seldinger technique.
Dr Harisinghani and colleagues appear to prefer use of a trocar technique (1,2). We believe that this may be less precise than the Seldinger technique and would advocate use of the Seldinger technique by operators less experienced with transgluteal drainage. Use of gantry angulation can also facilitate a very low paracoccygeal approach, which may further reduce postprocedural pain and complications (3).
Dr Harisinghani and colleagues stated their preference for the transgluteal approach over the transrectal approach when feasible. At our institution, simple anterior percutaneous drainage is preferred when possible. Otherwise, the transrectal approach is favored over the transgluteal approach unless a long-term indwelling catheter is likely to be required, such as in patients with rectal anastomotic dehiscence or other situations in which a fistula may be present. Fewer vital structures are at risk of injury with the transrectal approach, and transrectal catheters are well tolerated in adults (4) and in children (5). As far as we know, a formal comparison of these two treatment methods has not yet been undertaken, but we believe that the rate of complications from transrectal drainage will generally be lower than that with the transgluteal approach (6).
REFERENCES
Department of Radiology, Massachusetts General Hospital, White 270, 55 Fruit Street, Boston, MA 02114. e-mail: mharisinghani@partners.org
I thank Drs Woo and Millward for their interest in our articles (1,2). Their comments indicate the existence of varying approaches to common problems in intervention.
In response to their comments, although it is true that two of our images show the catheter too cephalad or lateral, the figures were chosen precisely to illustrate these features and were described as such in the figure legends. Operators undertaking transgluteal abscess drainage must be aware of proper and improper catheter deployment and know what to expect when facing this situation. Transient pain is encountered frequently during catheter placement in any location. Our report reflects the performance of trainees under supervision. If appropriate technique is followed, other practices may expect our very acceptable rate of complications, as well. Moreover, the reported complication rate of 2% for pseudoaneurysm formation reflects procedures performed in a busy high-volume academic setting in patients with varying abscess sizes, locations with respect to the percutaneous window, and tissue resistance to catheter placement. Thus, I believe that if the appropriate technique is followed, the complication rate should certainly not be any higher.
The use of gantry angulation may certainly be useful in providing access routes when routine fixed-plane scans appear to be limited. We too have used this technique on numerous occasions. There are some practical challenges that still need to be mentioned. Angling of the gantry and subsequent needle placements require the needle to be aligned with the gantry. This is facilitated by using the laser light in the scanner and aligning the needle parallel to the gantry, leaving a margin for error. Also, such positioning in not always possible for patients who are critically ill, and many cases do not require gantry angulation for successful catheter deployment.
Use of the Seldinger technique brings its own challenges, including coiling a sufficient length of guidewire into a small abscess cavity, keeping the wire sterile, and performing blind catheter deployment through tough gluteal tissue planes. As we stated, we reserve the Seldinger technique for certain indications that include small nearly inaccessible collections and large multiloculated ones (2).
In short, to my knowledge, trocar and Seldinger techniques have never been compared in a randomized trial to assess safety and effectiveness. The choice of one technique or the other is based on operator preference that in turn is generally based on prior experience. I certainly do not advocate that operators who are not comfortable with trocar technique on the basis of lack of experience with it should limit themselves to trocar technique when performing transgluteal drainage.
I agree with Drs Woo and Millward that the transrectal and transgluteal drainage approaches have not been compared in a formal study. The transrectal approach suits only short-term drainages. Particularly if the collection is hemorrhagic or if there is a likelihood of ongoing colonic leak, a longer period of catheter drainage can be anticipated. In such circumstances, transgluteal catheters can be maintained for days or even weeks. For frequently encountered deep pelvic abscesses with no anterior access, I recommend that interventional radiologists become familiar with and consider the transgluteal approach.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
