Radical Perineal Prostatectomy

Radical perineal prostatectomy, first performed in 1869 by Buchler and popularized in the United States by Young in 1905, remained the primary surgical approach to carcinoma of the prostate until the mid-1970s. With the recognition of the importance of assessing pelvic lymph nodes preoperatively and the advantage that retropubic prostatectomy offered with concomitant pelvic node dissection, perineal prostatectomy declined in popularity for the treatment of prostate cancer. The perineal approach, however, has seen a resurgence in the 1990s for several reasons: (a) the trend toward minimally invasive surgery with a focus on reducing the morbidity and therefore the hospital stay of patients, (b) the advent of laparoscopic surgery for lymph node assessment, (c) the introduction of PSA for screening for prostate cancer with reduction in the numbers of patients with node-positive disease, and (d) algorithms that may predict patients at high risk for positive lymph nodes. The procedure is also associated with reduced blood loss, low morbidity, and can be modified to incorporate the neurovascular sparing techniques for preservation of potency.


All patients who are potential candidates for radical perineal prostatectomy should undergo preoperative staging to assure that the patients are operable candidates. Methods of differentiating local from advanced disease include digital rectal examination, transrectal ultrasonography, radionuclide bone scan, assessment of pelvic lymph nodes, as well as pathologic indicators of progression such as Gleason sum and other markers.

Since the late 1980s, PSA has made a significant impact on the preoperative stage of patients with prostate cancer. Patients presenting for surgery are generally younger, healthier, and more likely to have organ- confined prostate cancer than the population treated only a decade earlier, and this in many ways contributes to the large increase in the number of radical prostatectomies done in the United States in the past decade. Other contributing factors are the modifications in technique that have reduced morbidity, including the nerve-sparing technique described by Walsh and alternative methods of restoring potency.

Digital rectal examination has a limited role in the clinical staging of prostate cancer. Its primary capability is to crudely estimate the volume of the cancer. Transrectal ultrasonography is another modality that also has limitations in assessing local disease but, combined with digital rectal examination, at least gives some gross assessment of likelihood of extracapsular disease. Other modalities such as transrectal MRI, CT scan, and pelvic MRI have been shown to have limited usefulness. Radionuclide bone scans are useful in assessing advanced bony disease but generally are not positive in patients with PSA below 20 ng/ml and no other sign of advanced disease.

For the past 11 years we have prospectively applied an algorithm to the preoperative assessment of patients with prostate cancer based on the evaluation of over 400 patients who had undergone pelvic node dissection at our institution. The current algorithm includes patients with a Gleason 7 or less, with low-volume cancer (T1b–c, T2a), normal acid phosphatase, and PSA of less than 20. Patients meeting all of these criteria have a less than 5% chance of positive lymph nodes, and, therefore, we do not perform pelvic lymph node dissections. Patients exceeding any one of the above criteria are considered to be in the high-risk group and have undergone pelvic lymph node dissections.4 With this method of assessment, our PSA recurrence rate from 1988 through 1994 was 27%, which compares favorably to our own and other series of retropubic prostatectomies, which have shown PSA recurrence rates between 24% and 29%.4,7,9 Because 75% of the patients who did not develop PSA recurrences had PSA <10 ng/ml, we have now modified our algorithm to place patients in the low-risk group if the serum PSA is <10 ng/ml.


Patients who are candidates for radical prostatectomy must have clinically organ-confined prostate cancer (T1–2). In addition to having organ-confined disease, other factors that need to be taken into consideration are the patient’s life expectancy, other comorbidities, or any other factors that may affect the patient’s choice. We generally would not offer a radical prostatectomy to patients who have a life expectancy less than 10 years. Over the age of 70, we would offer a radical prostatectomy only in selected cases in which we feel that the benefits that can be obtained from radical prostatectomy outweigh the potential risks, particularly when compared to alternative therapies.


Alternatives to radical prostatectomy include observation, hormonal deprivation, and radiation therapy. We do not consider either observation or hormonal deprivation to be curative, and the patients for whom this is a good option are those patients with less than 5 years of life expectancy, patients who are over 70 years old with a well-differentiated cancer, or patients who are a high risk for surgery and refuse radiation. Overall, observation is associated with a 75% mortality over 10 years. Radiation therapy, however, may be definitive and has an equivalent 5- and 10-year survival. The recurrence rates with radiation therapy are bimodal, with initial recurrences within 1 to 2 years of treatment and a delayed peak at 5 to 7 years after treatment. In our institution, 359 patients who received brachytherapy from 1972 to 1984 were compared to a contemporaneous series of 161 patients undergoing radical prostatectomy. The 5-year recurrence rates and survival were similar in both groups, which were equivalent in preoperative stage, age, Gleason sum, and other demographics. By 7 years, however, the recurrence rate among the brachytherapy group was 48% (9% local recurrence, 39% distant recurrence) versus 27% in the radical prostatectomy patients (8% local and 19% distant recurrence). If the patient is young and has a 15-year or longer outlook, we feel that our results would favor radical prostatectomy.


Before the patient is put in position, the legs are wrapped with Ace bandages. The patient is placed in an exaggerated lithotomy position. It is important that the patient’s perineum be parallel to the floor in that this directly affects exposure. We use a standard operating room table with folded sheets under the patient’s sacrum supporting the patient’s entire weight. We do not use shoulder braces, and if a patient tends to slide off the sheets, we will place the table in a slight reverse Trendelenburg position. The patient’s legs are stabilized using candy cane stirrups, again taking precautions to prevent stretching the hamstring or causing pressure on the legs.

Four instruments are significant in assisting the surgeon for this operation. These include the Lowsley curved tractor, the Young straight prostatic tractor, a halogen headlamp, and an Omnitract miniwishbone retractor system. The curved Lowsley tractor is used to bring the prostate up into the perineum to allow the dissection against the prostate while separating the rectum from the prostate. The straight Young tractor is used to manipulate the prostate laterally as well as cephalad and caudad after the membranous urethra has been divided. The halogen headlamp is important in that it allows the surgeon to aim a strong light into the operative field, which may be deep and narrow, preventing standard operating lights from adequately illuminating the structures. The Omnitract miniwishbone allows virtually unlimited retraction in any direction.

It should be noted that, in manipulation of the prostate from the perineum, the pelvis should be viewed as a cone with its apex at the site of the incision. To get better visualization at times, it may be necessary to actually push the prostate further into the pelvis. Also note that traction is not placed directly on the bulb or membranous urethra, as this will decrease the likelihood of restoration of potency and potentially affect the patient’s continence postoperatively.

The incision is made from the ischial tuberosity, crossing the midline at the juncture between the squamous epithelium and mucocutaneous border of the rectum. The incision extends posteriorly to a line equal to the posterior portion of the anus. By use of sharp dissection and electrocautery, the ischiorectal fossae are entered, and the central perineal tendon is identified by blunt dissection and transected with an electrocautery. At this point, we employ the Belt approach and dissect down to the white fascia of the rectum and proceed subsphincterically. Predominantly by blunt dissection with an index finger in the rectum, the rectal sphincter and levator ani can be dissected free of the rectum with minimal bleeding, and the blades from the miniwishbone retractor are then used to retract these muscles anteriorly and laterally. With tension on these muscles and tension on the rectum, the rectourethralis is identified and divided, which allows the surgeon to dissect the rectum free of the apex of the prostate.

If this is to be a nerve-sparing technique, the dissection is carried down to approximately 1.5 to 2 cm from the apex, at which point the posterior layer of Denonvillier’s fascia is divided, and dissection is carried between the two layers of the Denonvillier’s fascia. Care is taken not to damage the neurovascular bundles that course along the lateral posterior prostate on either side.

The distal portion of Denonvillier’s fascia is then incised in the midline with scissors, and the tag is then used to facilitate dissection of the neurovascular bundle from the prostate; the inferior pedicle, if present, is ligated and divided. If the dissection is in any way impaired by fibrosis such that there is a potential for prostatic tissue to be left behind, the neurovascular bundle is sacrificed on that side. It should be noted during this dissection that the neurovascular bundle actually courses across the posterior surface of the prostate at the apex and enters the urogenital diaphragm just posterior to the membranous urethra. This proximal relationship is important in that the vesicourethral anastomosis may incorporate the neurovascular bundle if the sutures are placed too deeply during the anastomosis. Once the neurovascular bundle has been dissected, a Vessi-loop is placed around this to aid in lateral traction.

The retraction of the neurovascular bundle on either side thereby exposes the proximal membranous urethra, allowing a right-angled clamp to be placed around the membranous urethra. There should be little resistance anterior to the urethra to the passage of the tip of the right-angled clamp anterior to the urethra if one stays posterior to the endopelvic fascia. The Lowsley tractor is removed, and the membranous urethra is divided. The Young tractor is then placed into the bladder via the severed prostatic urethra, and the endopelvic fascia is dissected free of the anterior prostate. In most cases, there is insignificant bleeding from the dorsal venous complex, but if there should be communicating veins, they should be ligated using 3-0 Vicryl.

The groove between the prostate and bladder is identified, and, with either sharp or blunt dissection, the prostate and bladder can be separated. If there is any resistance to blunt dissection, the bladder neck should be sharply divided, and biopsies taken of the bladder neck to assure that the resistance is not secondary to bladder neck invasion with the cancer. Patients who have had prior transurethral resections may have an obliteration of the plane between the prostate and bladder, and the blades of the Young tractor can be used to aid in this identification.

The prostate is dissected from the bladder anteriorly to the 5-o’clock and 7-o’clock positions, respectively, on the patient’s left and right; the bladder neck is divided over the Young tractor, and the Young tractor is removed. The bladder is evacuated of any urine, and the posterior bladder neck is divided at its juncture with the prostatic urethra. The prostate is then dissected free from the posterior bladder, allowing identification of the superior pedicles of the prostate as well as the seminal structures. The superior pedicles are isolated, divided, and ligated. The seminal vesicles are dissected to their tips with blunt dissection, and the artery from the seminal vesicle is either cauterized or ligated; the vasa deferentia are generally cauterized. Finally, the Denonvillier’s fascia overlying the seminal structures is divided, allowing removal of the prostate.

After complete hemostasis has been ensured, the bladder neck is reconstructed using 3-0 Vicryl on an SH needle, beginning posteriorly to anteriorly as described by Dees. This direction of the closure, beginning in the posterior bladder, is done to facilitate the closure without injury to the ureters and also to take advantage of the anatomic relationship between the bladder neck and the membranous urethra with the shorter distance being anteriorly. The anastomosis is performed using 3-0 Vicryl simple sutures and an RB-1 controlled-release needle around a 22-Fr 5-cc Foley catheter. Generally, seven or eight interrupted sutures are used for this anastomosis, though, alternatively, the anastomosis can be performed with a running suture. Care should be taken that small portions of the membranous urethra are incorporated in the anastomosis so that the continence mechanism is left undisturbed and the neurovascular bundles that contribute to potency are avoided.

The rectum is then inspected; a Foley balloon is inflated, and a Penrose drain is placed through the left ischiorectal fossa and a separate stab incision. The incision is closed with a 3-0 chromic gut closure. One suture is placed to reapproximate the central tendon, and the remainder of the sutures are used to close the skin in a horizontal mattress.

Postoperatively, the patients have very low requirement for pain medication. Most patients either do not require parenteral pain medication or are off the parenteral medications within 12 to 24 hours. Average time to discharge is approximately 48 hours from the time of surgery. The patient’s catheter is removed on the 12th day. The Penrose drain is removed before discharge.



Perioperative complications include hemorrhage, wound infection, cardiovascular complications, and rectal injury. The incidence of rectal injury is less than 2%, and with current techniques, preoperative bowel preparation (Golightly), and antibiotics, these are closed primarily in two layers without the need to perform a diverting colostomy. Wound infection rates are less than 1%, and cardiovascular complications are approximately 1%. The average blood loss in these patients is approximately 450 cc, and our transfusion rate is less than 5%.

Long-term complications include incontinence and impotence. Incontinence requiring intervention such as pads, clamps, or inflatable devices occurs in 2.8% of our patients. We have found that incontinence generally occurs in patients who are older, obese, or have had prior radiation therapy. Potency following nerve-sparing perineal prostatectomy is dependent on the patient’s age and preoperative status. Patients under the age of 60 who are fully potent and have both neurovascular bundles spared have approximately a 50% potency rate. Patients who are over the age of 60 have a reduced rate of potency, and we have not yet had a patient over the age of 70 who has spontaneously regained his potency. Patients who are having difficulties with potency before surgery and patients in whom the neurovascular bundles could not be spared will likely be impotent. We generally advocate early use of pharmacologic or other means of assistance in these patients to help them regain their potency.


Following radical prostatectomy, recurrence can be measured using PSA, which is exquisitely sensitive. Any patient who undergoes a radical prostatectomy can expect his PSA to fall below detectable levels. Its failure to do so generally means that the patient has significant residual disease, either locally or distantly. Another group of patients will have an initial drop in their PSA to undetectable levels and then a return to measurable levels. These patients may have local and/or distant recurrence of their disease or possibly residual malignancy. Patients who develop recurrence based on PSA will generally manifest a clinical progression within 18 months. Additionally, if the PSA is going to rise, it will do so within 2 years in 90% of patients and within 4 years in virtually every patient. Based upon surrogate markers such as PSA, we can now predict disease recurrence earlier, allowing assessment of the outcome of radical prostatectomy within 5 years as opposed to the older data, which required a 7- to 10-year follow-up.

The primary predictor of recurrent disease after radical prostatectomy is the presence of positive margins. In SEER data, radical prostatectomy has been shown to have a long-term disease-free rate approaching 90% at 10 years with organ-confined disease. Positive margins may reflect capsular penetration, invasion of the periprostatic tissue, or may reflect a pathologic discrepancy. Patients with negative margins will have a 10% risk of recurrence, and the risk of recurrence with positive margins will be 30% to 50%. Before PSA screening, the incidence of positive margins in patients undergoing radical prostatectomy in our institution was 48% with an overall PSA recurrence rate of 24%. Whether the current reports of 20% positive margin rates with serial PSA screenings has a proportional drop in the PSA recurrence rates remains to be seen.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: