0022-5347/04/1724-1318/0 THE JOURNAL OF UROLOGY® Copyright © 2004 by AMERICAN UROLOGICAL ASSOCIATION

Vol. 172, 1318 –1322, October 2004 Printed in U.S.A.

DOI: 10.1097/01.ju.0000139883.08934.86

NERVE SPARING RADICAL PROSTATECTOMY: EFFECTS OF HEMOSTATIC ENERGY SOURCES ON THE RECOVERY OF CAVERNOUS NERVE FUNCTION IN A CANINE MODEL ALBERT M. ONG,* LI-MING SU, IOANNIS VARKARAKIS, TAKESHI INAGAKI, RICHARD E. LINK, SAM B. BHAYANI, ALEX PATRICIU, BARBARA CRAIN AND PATRICK C. WALSH From the James Buchanan Brady Urological Institute and Department of Pathology (BC), Johns Hopkins Medical Institutions, Baltimore, Maryland

ABSTRACT

Purpose: To preserve sexual function following radical prostatectomy, one must avoid injury to the neurovascular bundles (NVBs). In the conventional open surgical technique, the use of energy sources for hemostasis is avoided to prevent damage to the cavernous nerves. In contrast, during laparoscopic radical prostatectomy, electrosurgical and ultrasonic energy sources are frequently used for hemostasis during dissection of the prostate. In this study, we evaluated the acute and chronic physiological effects of various hemostatic energy sources on cavernous nerve function in the canine model. Materials and Methods: A total of 12 dogs were divided into 4 groups based on the type of energy source used for hemostasis during unilateral dissection of the NVB. The groups included conventional dissection with suture ligatures (group 1), monopolar (group 2) or bipolar (group 3) electrosurgery and ultrasonic shears (group 4). The contralateral NVB was left undissected as a control. Erectile function was assessed acutely and after 2 weeks of survival by measuring peak intracavernous pressures in response to cavernous nerve stimulation. Results: Following conventional techniques of nerve sparing, the erectile response to nerve stimulation was unaffected. In contrast, the use of energy sources in proximity to the NVB during nerve preservation was associated with a substantial decrease in erectile response both acutely (74% to 91% decrease compared to controls) and after 2 weeks (93% to 96% decrease). Conclusions: In the chronic canine model, use of hemostatic energy sources in proximity to the prostate during dissection of the neurovascular bundle is associated with a significantly decreased erectile response to cavernous nerve stimulation. KEY WORDS: prostate, dogs, hemostasis, penile erection, prostatectomy

The conventional technique of cavernous nerve preservation during radical retropubic prostatectomy relies on the use of a fine right-angled dissector to develop gently the plane between the neurovascular bundle (NVB) and prostate. Vascular branches from the neurovascular bundle to the prostate are carefully defined and ligated during dissection, and the use of electrosurgical sources in proximity to the prostate neurovascular bundle is scrupulously avoided. This technique is associated with the preservation of sexual function.1– 4 The technique of laparoscopic radical prostatectomy (LRP) has undergone numerous refinements since its description by Schuessler et al in 1997.5 Although a variety of approaches, port configurations, and instrumentation are used to perform this procedure, the majority of surgical techniques rely on energy sources, including ultrasonic shears, monopolar or bipolar electrosurgery, for hemostasis during dissection of the prostate.6, 7 Only a small minority of surgeons duplicate the conventional open method of dissection in the laparoscopic field.8 Although monopolar and bipolar electrosurgical techniques were introduced in the 1920s and 1940s, no systematic study of their effects on nerve function has been performed until recently.9 –11 These studies have focused on

myelinated nerves similar to those encountered in head and neck procedures. However, the data from these studies may not be directly applicable to radical prostatectomy because the parasympathetic fibers responsible for spontaneous erections are unmyelinated and may be affected differently by thermal and electrical injury. The current study was designed to evaluate both the physiological and histological effects of various energy sources commonly used during laparoscopic prostatectomy for dissection around the neurovascular bundle. In this canine model, laparoscopic energy sources identical to those used during laparoscopic radical prostatectomy in clinical series were used to assist with dissection of the NVB from the prostate. With the exception of an open incision, the surgical technique of NVB dissection during nerve sparing LRP was duplicated as closely as possible. The functional and histological outcomes of these dissection techniques were correlated with the type of energy source used. MATERIALS AND METHODS

After approval of the experimental protocol by the institutional animal care and use committee, 12 adult male beagles weighing 15 to 20 kg underwent laparotomy. Six-hour fasting animals were pretreated with 10 to 20 mg/kg zolazepam intramuscularly and given 25 mg/kg cefazolin. The animals were intubated and given 1% to 2% halothane for anesthesia. The abdomen was prepared and draped, and the skin was

Accepted for publication May 21, 2004. Study received institutional animal care and use committee approval. * Correspondence: James Buchanan Brady Urological Institute, Johns Hopkins Medical Institutions, 600 North Wolfe St., Baltimore, Maryland 21287 (e-mail: [email protected]). 1318

NERVE SPARING RADICAL PROSTATECTOMY

incised. The bladder was drained with a syringe to improve exposure. A separate incision at the penoscrotal junction was made to expose the corpora cavernosa, and 21 gauge intravenous catheters were introduced into both corpora for intracavernous pressure (ICP) measurements, as well as into the femoral artery to record systemic arterial pressure. The catheters were connected to pressure transducers (Part 46086-21, Abbot Laboratories, Abbot Park, Illinois), which were flushed with normal saline containing 1,000 U heparin per l. The pressure transducers were connected to a Marquette 7010 monitor (GE Medical Systems, Waukesha, Wisconsin) and the analog output was recorded on a computer. To eliminate variability, all procedures were performed by the same surgeon under the guidance of a skilled laparoscopic surgeon experienced with laparoscopic prostatectomy. The bladder was retracted anteriorly and the cavernous nerve bundles were identified posterolateral to the prostate. The nerve bundles were stimulated proximally at the prostatovesical junction using a CaverMap with a 0.2 cm tip (Blue Torch Medical Products, Ashland, Massachusetts). Stimulation of the neurovascular bundle was started at 2 mA and doubled every minute to a maximum of 8 mA. The erectile response to stimulation of each NVB was recorded. Unilateral dissection of a neurovascular bundle from the prostate was then performed using 1 of the 4 described methods for NVB dissection and hemostasis. In group 1, 3-zero silk ligatures were used for hemostasis prior to tissue division. In group 2, a 5 mm laparoscopic hook (Aesculap, Center Valley, Pennsylvania) was used to apply energy for hemostasis and transect tissue during dissection of the NVB from the prostate with the right-angled clamp. In group 3, a 5 mm Aesculap microbipolar laparoscopic instrument was used for both hemostasis and tissue dissection. In group 4, 5 mm laparoscopic ultrasonic shears (Ethicon, Somerville, New Jersey) were used for hemostasis and tissue dissection. Figure 1 shows the instruments used to apply the energy sources. The contralateral NVB was not dissected and it served as an internal control. Due to the difference in exposure between open and laparoscopic techniques, initial dissection was performed with a fine right-angled clamp to define tissue planes. With the exception of the monopolar electrosurgery group, the inactivated laparoscopic instruments were then used to continue dissection in order to duplicate the standard laparoscopic technique. In the monopolar electrosurgery group, dissection was continued with the right-angled clamp with the monopolar energy source activated between the tines to divide tissue. Energy sources were applied between the NVB and prostate, and activated for no longer than 1 second to minimize collateral tissue injury. The dessicated tissue was then sharply divided with Jameson scissors when necessary. This was continued until the NVB was freed from the prostate unilaterally. For the electrosurgical groups, a ValleyLab Force-2 electrosurgery generator (Tyco, Inc., Boulder, Colorado) set at 10 W was used to power the electrical devices. For the harmonic scalpel group, the power source was set at 5, and the minimal coagulation mode (cutting mode) was used to transect simultaneously tissue and achieve hemostasis. The inactivated blade of the device was placed between the prostate and bundle. The instrument was then closed, and the instrument was lifted away from the neurovascular bundle and subsequently activated for no longer than 1 second. In all groups, contact of activated instruments with the cavernous nerve was carefully avoided. Figure 2 shows the method of dissection. After dissection, the neurovascular bundles were again identified at the prostatovesical junction and the erectile response to nerve stimulation was recorded. Peak intracavernous pressures were normalized against mean arterial pressure (MAP) and reported as the percent MAP. The laparotomy incisions were closed and the animals were allowed to recover. The animals

1319

FIG. 1. Instruments used to apply hemostatic energy were 5 mm Ethicon laparoscopic harmonic scalpel (top), 5 mm Aesculap laparoscopic monopolar hook electrode (5 mm) (middle) and 5 mm Aesculap microbipolar laparoscopic forceps (bottom).

FIG. 2. Dissection. Arrows indicate dissection plane with energy sources. Dashed line indicates position of periprostatic NVB under superficial layer of fascia.

were transferred to the housing area and allowed to survive for 2 weeks. Postoperative pain medication consisted of 2 mcg/kg buprenorphine every 8 hours for 3 days. After 2 weeks, the animals were taken back to the operating room and were reexplored. The abdomen was re-opened and the bladder was exposed. Intracavernous and femoral

1320

NERVE SPARING RADICAL PROSTATECTOMY

arterial pressures were again recorded. The neurovascular bundles were located as before, and the erectile response to stimulation with the Cavermap was recorded. The animals were then sacrificed, and the prostate and adjacent periprostatic nerve bundles harvested en bloc and placed in formalin. Histological assessment. Specimens were embedded in paraffin, sectioned, and stained with hematoxylin and eosin with Luxol-fast blue counterstaining for myelin. Transverse sections from the apex, middle, and base were submitted for examination. Masson trichrome staining and immunostaining using SMI-311 for neurofilaments were also performed. The histological findings were evaluated by a single pathologist who was blinded to the various study groups, and the neurovascular bundles were evaluated for the presence of fibrosis, vascular injury and overt neural damage. RESULTS

The periprostatic neurovascular bundles were clearly visualized before and after NVB dissection. All animals demonstrated equivalent erectile responses bilaterally prior to undergoing NVB dissection. Despite the use of a large number of silk ligatures (10 per animal), the conventional technique for release of the NVB (group 1) resulted in no significant reduction in the erectile response. In contrast, the use of an energy source (groups 2 to 4) to aid in hemostasis was associated with a significant decrease in erectile response to nerve stimulation compared to both controls and conventional dissection (see table, fig. 3). Acutely, this effect was less severe for the group of animals undergoing dissection with the ultrasonic shears (ICP 26% of control) than for the groups undergoing dissection with the electrosurgical energy sources (ICP 8.5% to 16.1% of control), but it was not statistically significant (Student’s t test p ⫽ 0.15). However, at 2 weeks, the functional outcome for all groups undergoing dissection with energy sources for hemostasis was equivalent (ICP 4.6% to 6.3% of control). In comparison, the ICP in the conventional group was equal to that in controls (acute 97.4% and chronic 106.3%, p not significant). The difference in erectile responses between the use of energy sources and conventional dissection was statistically significant (p ⬍0.005). The acute erectile response to cavernous nerve stimulation after dissection correlated well with the erectile response to nerve stimulation after 2 weeks of survival. The histological results were collated by group, and the presence of fibrosis, nerve injury, and vascular thrombosis was noted (see table). No attempt was made to grade the severity of the findings within the neurovascular bundle or perform statistical analysis due to the small sample numbers. There were more histological findings suggestive of injury in the animals from groups 2 to 4 than from group 1. There was a suture granuloma adjacent to a vessel within the

FIG. 3. Functional results. All pressures were normalized against MAP. Vertical bars indicate 1 SD above and below mean ICP.

NVB in 1 specimen from an animal in the conventional group. DISCUSSION

Traditionally, the use of energy sources for hemostasis during radical retropubic prostatectomy was avoided due to the potential for vascular or neural injury.4 The current techniques of laparoscopic prostatectomy rely commonly on a variety of energy sources for hemostasis to improve visualization of the operative field and increase operative efficiency by minimizing the number of instrument changes. To address the question of the optimal technique of hemostasis and NVB dissection, we developed a study based on an animal model first described by Lue et al.12 To our knowledge, our study is the first to examine systematically the effect of energy sources on unmyelinated autonomic nerves. Although data from the neurosurgical literature suggest that these nerves may be more susceptible to injury than their myelinated counterparts, the majority of nerve injury studies used a myelinated rodent sciatic nerve model due to its low cost, convenience and ease of histological analysis.10, 13 Despite the differences between the experimental models, our findings corroborate those of other groups who examined the effect of thermal and electrical energy on rodent sciatic nerves.10, 13 The use of energy sources in proximity to the periprostatic NVB substantially decreased the stimulated erectile response compared to controls and the conventional method of NVB dissection performed without energy. The use of energy sources was associated with an increased number of abnormal histological findings. Although not re-

Results Mean % MAP ⫾ SD* Experimental Group

Control ICP

Conventional: Acute 57.7 ⫾ 17.0 Chronic 59.3 ⫾ 15.9 Monopolar: Acute 65.8 ⫾ 25.3 Chronic 57.9 ⫾ 37.7 Bipolar: Acute 81.8 ⫾ 43.0 Chronic 63.8 ⫾ 21.3 Harmonic: Acute 73.6 ⫾ 15.6 Chronic 58.4 ⫾ 15.5 Three dogs in each chronic group. * (Maximum ICP/MAP) ⫻ 100.

No. Histological Evaluation

Dissected ICP

Dissected ICP % Control

p Value

56.2 ⫾ 25.9 63.0 ⫾ 14.6

97.4 106.3

5.6 ⫾ 7.6 2.7 ⫾ 2.0

Fibrosis

Vascular Injury

Nerve Injury

0.8215 0.9571

0

1 (granuloma)

0

8.5 4.6

⬍0.0001 0.0049

1

0

1

13.2 ⫾ 10.3 4.0 ⫾ 5.2

16.1 6.3

⬍0.0001 0.0008

1

2

1

19.2 ⫾ 18.0 2.3 ⫾ 2.6

26.0 3.9

⬍0.0001 0.0003

2

0

0

1321

NERVE SPARING RADICAL PROSTATECTOMY

lated directly to nerve function, the presence of adjacent fibrosis indicated local damage. Traditional histological changes suggestive of nerve injury, such as loss of axons and thinning of myelin sheaths, were not visible because the autonomic nerves are small and not myelinated. Ultimately, the best indicator of neurovascular bundle injury is the erectile response to cavernous nerve stimulation. This study also demonstrates the prognostic value of intraoperative nerve stimulation. The acute erectile response to stimulation correlated well with the chronic response to cavernous nerve stimulation. This has important clinical implications. In future studies of alternate techniques for obtaining hemostasis, successful stimulation of the erectile response at the end of the case should provide an early reliable surrogate end point for successful preservation of the NVB. In monopolar electrosurgery, the patient becomes part of the electrical circuit. Electrical current from the active electrode seeks the shortest path back to the return electrode. This concentrates current along conduits with less electrical resistance such as nerves or vascular structures. The path that current takes may have no correlation with anatomical distance.9 Because of this undesirable property, the bipolar electrode was introduced by Greenwood in 1942 to eliminate the patient from the circuit.14 The active and return electrodes are in close proximity to each other, thus, minimizing potential injury. However, as tissue is dessicated, its resistance increases. Current may spread to surrounding tissue with lower electrical resistance and widen the area of damage. Additionally, current leakage from the device may cause the bipolar electrocautery to behave like a monopolar electrical source.9 Prior studies examining electrosurgery have identified heat as the cause of neural injury, which may occur with temperatures as low as 41C.10 Because of this, irrigation is routinely used to cool tissue during use of the bipolar cautery in head and neck surgery. This practice is corroborated by studies that found a clear neuroprotective effect with the use of irrigation during bipolar cautery of the rat sciatic nerve.10 Alternative energy sources such as the ultrasonic shears have been developed to obviate the need for electrical current and they may be more suitable for use around neural structures. Many groups have used this instrument to perform laparoscopic radical prostatectomy due to its excellent hemostatic properties.7 However, Owaki et al found that the blade of the ultrasonic shears becomes hot after use, increasing to 63C after 3 seconds and 150C after 30 seconds.15 They suggested that contact of the blade with neural structures immediately following use caused recurrent laryngeal nerve injury in their series of patients undergoing endoscopic parathyroid surgery. This is important to note since the surgeon has no indication of the temperature of the instrument tips while performing laparoscopic surgery, and there is relatively little space for the dissipation of heat. The majority of the published techniques of laparoscopic prostatectomy rely on bipolar, monopolar, or ultrasonic energy sources for hemostasis. Successful preservation of potency has recently been reported from centers specializing in LRP with results ranging from 23% to 82%.16 –18 This wide variability is likely multifactorial but it may be in part due to the method of nerve dissection and use of hemostatic energy sources. It is possible that cavernous nerve regeneration may explain the gradual recovery of potency observed in many clinical series. Evidence from nerve grafting studies first performed by Quinlan et al in the rat model suggest that the cavernous nerves can regenerate.19 Their findings were corroborated by Kim et al in subsequent clinical series.20 Our current study is limited by the short duration of survival (2 weeks) after the initial surgery. Longer term survival studies are certainly required to determine whether there is any

recoverability of cavernous nerve function, although the detrimental effect of all energy sources on cavernous nerve function observed in this short-term survival study highlights the vulnerability of the cavernous nerves to both heat and electrical current. Based on the results of this study, refraining from the use of energy sources during nerve dissection may be associated with a more rapid and certain return of potency. If power sources are desired during other portions of the procedure, irrigation during the use of the bipolar forceps may protect against inadvertent neural injury by cooling the tissues.10 However, this may be problematic because it can obscure the laparoscopic field. We have developed an alternate method for the use of the ultrasonic shears in conjunction with a fine right-angled clamp, which keeps the active element away from critical structures. Preliminary results with this alternate method suggests the preservation of potency when the instrument is used in this manner. CONCLUSIONS

The use of energy sources to achieve hemostasis during cavernous nerve dissection in nerve sparing radical prostatectomy should be scrupulously avoided, regardless of the surgical approach used. In the canine model, use of these devices in proximity to the NVB is associated with the loss of erectile response to cavernous nerve stimulation both acutely and after 2 weeks of survival. The use of a nerve stimulator in conjunction with continuous intracorporeal pressure assessments may aid in the mapping and preservation of the cavernous nerves during nerve sparing radical prostatectomy. REFERENCES

1. Catalona, W. J., Carvalhal, G. F., Mager, D. E. and Smith, D. S.: Potency, continence and complication rates in 1,870 consecutive radical retropubic prostatectomies. J Urol, 162: 433, 1999 2. Rabbani, F., Stapleton, A. M., Kattan, M. W., Wheeler, T. M. and Scardino, P. T.: Factors predicting recovery of erections after radical prostatectomy. J Urol, 164: 1929, 2000 3. Walsh, P. C., Marschke, P., Ricker, D. and Burnett, A. L.: Patient-reported urinary continence and sexual function after anatomic radical prostatectomy. Urology, 55: 58, 2000 4. Walsh, P. C.: Anatomic radical retropubic prostatectomy. In: Campbell’s Urology, 8th ed. Edited by P. C. Walsh, A. B. Retik, E. D. Vaughan, Jr. and A. J. Wein. Philadelphia: W. B. Saunders Co., vol. 4, sect. XI, chapt. 90, p. 3107, 2002 5. Schuessler, W. W., Schulam, P. G., Clayman, R. V. and Kavoussi, L. R.: Laparoscopic radical prostatectomy: initial short-term experience. Urology, 50: 854, 1997 6. Guillonneau, B. and Vallancien, G.: Laparoscopic radical prostatectomy: the Montsouris technique. J Urol, 163: 1643, 2000 7. Dahl, D. M., L’esperance, J. O., Trainer, A. F., Jiang, Z., Gallagher, K., Litwin, D. E. et al: Laparoscopic radical prostatectomy: initial 70 cases at a U.S. university medical center. Urology, 60: 859, 2002 8. Su, L., Link, R. E., Bhayani, S. B. and Pavlovich, C. P.: Nerve sparing laparoscopic radical prostatectomy: techniques to aid in the neurovascular bundle dissection. Presented at the World Congress on Endourology and Shockwave, Montreal, Canada, September 2003 9. Smith, T. L. and Smith, J. M.: Electrosurgery in otolaryngologyhead and neck surgery: principles, advances, and complications. Laryngoscope, 111: 769, 2001 10. Donzelli, J., Leonetti, J. P., Wurster, R. D., Lee, J. M. and Young, M. R.: Neuroprotection due to irrigation during bipolar cautery. Arch Otolaryngol Head Neck Surg, 126: 149, 2000 11. Kumar, K. and Crawford, A. H.: Role of “Bovie” in spinal surgery: historical and analytical perspective. Spine, 27: 1000, 2002 12. Lue, T. F., Takamura, T., Schmidt, R. A. and Tanagho, E. A.: Potential preservation of potency after radical prostatectomy. Urology, 22: 165, 1983 13. Hnatuk, L. A., Li, K. T., Carvalho, A. J., Freeman, J. L., Bilbao, J. M. and McKee, N. H.: The effect of bipolar electrocautery on peripheral nerves. Plast Reconstr Surg 101: 1867, 1998 14. Greenwood, J., Jr.: Two point coagulation: a follow-up report of a

1322

NERVE SPARING RADICAL PROSTATECTOMY

new technic and instrument for electrocoagulation in neurosurgery. Arch Phys Ther, 23: 552, 1942 15. Owaki, T., Nakano, S., Arimura, K. and Aikou, T.: The ultrasonic coagulating and cutting system injuries nerve function. Endoscopy, 34: 575, 2002 16. Guillonneau, B., Cathelineau, X., Doublet, J. D., Baumert, H. and Vallancien, G.: Laparoscopic radical prostatectomy: assessment after 550 procedures. Crit Rev Oncol Hematol, 43: 123, 2002 17. Katz, R., Salomon, L., Hoznek, A., de la Taille, A., Vordos, D., Cicco, A. et al: Patient reported sexual function following lapa-

roscopic radical prostatectomy. J Urol, 168: 2078, 2002 18. Tewari, A. and Menon, M.: Vattikuti Institute prostatectomy: surgical technique and current results. Curr Urol Rep, 4: 119, 2003 19. Quinlan, D. M., Nelson, R. J. and Walsh, P. C.: Cavernous nerve grafts restore erectile function in denervated rats. J Urol, 145: 380, 1991 20. Kim, E. D., Nath, R., Slawin, K. M., Kadmon, D., Miles, B. J. and Scardino, P. T.: Bilateral nerve grafting during radical retropubic prostatectomy: extended follow-up. Urology, 58: 983, 2001