- Email: [email protected]
Prostate Heating Patterns Comparing Electrosurgical Transurethral Resection and Vaporization: A Prospective Randomized Study
0022-534719711571-0169$03.00/0
THEJOURNAL OF UROLOGY Copyright 0 1997 by AMERICAN UROLOCICAL ASSOCIATION, INC.
Vol. 157. 169-172,January 1997 Printed in U S A .
PROSTATE HEATING PATTERNS COMPARING ELECTROSURGICAL TRANSURETHRAL RESECTION AND VAPORIZATION: A PROSPECTIVE RANDOMIZED STUDY ANUP PATEL, GERHARD J. FUCHS, JORGE GUTIERREZ-ACEVES AND THOMAS P. RYAN From the UCLA Department of Urology, Los Angeles, California, Uniuersidad de Guadalajara, Guadalajara, Mexico, and Research and Deuelopment Department, Valleylab, Boulder, Colorado
ABSTRACT
Purpose: A prospective study was performed to determine if transurethral electrosurgical vaporization of the prostate is associated with unseen heat damage to vital periprostatic structures compared to conventional loop resection. In addition, energy consumption and its relationship to observed tissue temperature at the prostate periphery were evaluated for each treatment. Materials and Methods: Patients with moderate to severe symptoms of benign prostatic bladder outflow obstruction and objective evidence of diminished flow or acute urinary retention were randomized to undergo either transurethral loop resection or eledrosurgical vaporization &r stratification for gland volume. Instrumentation was standardized for both groups except for the monopolar electrode used. The radiofrequency power source in the study was a new computer controlled generator with a constant power delivery feature. Regional tissue heating patterns were evaluated with optical fiber probes in real time. Four stationary sites were chosen for temperature measurements, namely the lateral lobe of the prostate, neurovascular bundle beside the prostatic apex at the level of the external sphincter, and anterior rectal wall at the level of the prostatic base and apex. A pull back technique was used to search for hot points in the long axis of the probe (3-dimensional temperature mapping) in 2 patients from each group. Incident generator panel power settings for the electrosurgical vaporization treatments were equivalent to those commonly used for loop resection (150 watts) and were adjusted up or down as needed. Results: Prostate electrosurgical vaporization was possible at generator panel settings that were nearly equivalent to those for transurethral resection of the prostate (130 to 190 watts). NO si&cant rectal or sphincteric heating was detected with either procedure. Conductive cooling of the neurovascular bundles was observed in 2 patients in each group toward the end of the operation. More energy was used per minute of treatment during electrosurgicalvaporization than with regular loop resection (p (0.004) but this was not associated with unwanted tissue heating. Conclusions: Neither conventional loop resection nor electrosurgical vaporization of the prostate appeared to be unsafe treatments with respect to unseen deep heating effects to vital periprostatic structures when performed at equivalent low incident power. The extra energy used during electrosurgical vaporization provided the benefit of improved coagulative hemostasis concurrently with shallow tissue vaporization using pure cutting current alone, without compromising treatment safety. KEY WORDS:prostate, prostatectomy, electrosurgery, heating, bladder neck obstruction Transurethral electrosurgical resection of the prostate re- on electrosurgical units.2 Since to our knowledge a real-time mains the gold standard treatment for symptomatic bladder study of prostate and periprostatic heating during electrosuroutflow obstruction from benign prostatic hyperplasia. The gical treatment (transurethral resection of the prostate or procedure has withstood the test of time by virtue of the fact transurethral electrosurgical vaporization) has never been that it is associated with rapid, often dramatic subjective and done, it is unclear whether an element of unseen heat damobjective success rates of 85 to 90%. However, disadvantages age might occur to explain any of the observed complications of morbidity, including bleeding, intraoperative fluid absorp- relating to potency and continence. In our study the hypothtion, infection, retrograde ejaculation, stricture formation, esis that unseen regional thermal heating of vital periprosurinary incontinence and erectile impotence, are all well tatic structures may occur was tested in a prospectively randocumented in the urological literature.' The etiology of this domized fashion during conventional transurethral resection last complication is still poorly understood but, with the of the prostate and transurethral electrosurgical vaporizaadvent of the new treatment of transurethral electrosurgical tion with the VaporTrode* grooved bar electrode. vaporization of the prostate, the possibility of unseen therPATIENTS AND METHODS ma1 injury to the erectile nerves has been highlighted, particularly since high panel power settings have been required Patients with symptomatic bladder outlet obstruction were evaluated by self-administered international prostate sympAccepted for publication July 26, 1996. tom scores (I-PSS),urine culture and uroflowmetry. Prostate Editor's Note: This article is the third of 6 p ~ b l k h d in this volume as measured by transrectal ultrasound wm calcuissue for which category 1 CME credits can be earned. InStructions for obtaining credits are given with the questions on pages 336 and 837. 169
* Circon ACMI, Stamford, Connecticut.
170 PROSTATE HEATING PATTERNS OF ELECTROSURGICAL TRANSURETHRAL RESECTION AND VAPORIZATION lated using the prolate sphere formula (0.523 X AF' X width x length). Six patients with moderate or severe symptoms of bladder outlet obstruction and 6 with acute urinary retention were selected for surgery provided there was no coexistent neurological cause of voiding dysfunction and no urinary infection. The study protocol was approved by the local ethical committee and all study patients were required to provide informed consent. ARer stratification for prostate size (larger or smaller than 50 ~ m . ~all) ,patients were randomized to undergo either loop resection or electrosurgical vaporization with the VaporTrode grooved bar electrode. All treatments were performed with a 25F continuous flow resectoscope and water irrigation. A new (Force FX*)computer controlled tissue response radiofrequency generator was used as the energy source throughout the study, since this device has been shown by the manufacturer to maintain constant power for a wider range of tissue impedance than conventional radiofrequency generators. Incident panel power settings in pure cut (150 watts) and spray coagulation (40 watts) modes, which had a range similar to conventional settings for transurethral resection of the prostate, were selected at the start of each treatment and adjusted up or down as indicated by the observed tissue effect. In this manner we hoped to obtain a true comparison of differences in tissue heating related to active electrode configuration as the main variable tested. Generator power delivery parameters were also measured during each procedure using a passive feed-through measurement system connected to the radiofrequency output of the generator, transparent to the surgeon. The device was insulated to eliminate any patient hazards during treatment. This system monitored current and voltage, and calculated phase, power, energy, impedance and derivatives of all of these measurements in real time. Data were acquired at 10 Hz. during each activation of cutting current and stored every 10 minutes for later off-line analysis. Patients were prepared and draped in sterile fashion in the lithotomy position before treatment. The perineum was shaved and prepared along with the external genitalia. A linear array 7 MHz. transrectal ultrasound probe was introduced through the anal verge into the rectum, and used to image the external sphincter, prostate and bladder neck. Using real-time transrectal ultrasound guidance, transperineal puncture was performed with a sharp-tipped closed ended 18 gauge polytetrafluoroethylene (Teflon)catheter and tungsten trocar. The first of 2 such catheters was guided into the trapezoid area of the external urethral sphincter, and its tip was positioned adjacent to the right neurovascular bundle beside the prostatic apex. The second catheter tip was placed within the left lateral lobe 15 to 20 mm. parallel to the prostatic urethra at the mid gland level. The separation of the tip of this second catheter from the urethra was measured using sector transrectal ultrasound scans. The trocars of both catheters were replaced by optical thermosensor fibers pre-calibrated in a constant temperature water bath at the table side to ensure 0.1Caccuracy and secured in position with tape markers. Using the polytetrafluoroethylene catheters the phosphor sensor tips were kept dry and free of false readings from adherent blood contamination. Sensor positions were confirmed with the ultrasound probe. Finally, 2 additional sensors mounted on the surface of a specially designed rectal probe were placed 3 and 8 cm. from the anal verge against the anterior rectal wall to detect undesirable rectal heating at the level of the bladder neck and prostatic apex or pelvic floor. Probe placement is shown in figure 1. All 4 fiberoptic sensors were connected via a fiberoptic extension to a thermometry mainframe device. Each probe was able to measure up to a 120C limit. Prostate temperatures were sampled every 2 seconds in real time after equilibrium had
* Valleylab. Boulder, Colorado.
Optical Thermosenson 1 = lateral lobe sensor 2 = sphincter-NV bundle sensor 3 = proximal rectal sensor 4 =distal rectal sensor
Bladder
Rectal probe
P
FIG.1. Placement of optical fiber thermometry probes. NV, neurovascular.
been attained at baseline and throughout each treatment. A laptop computer was used for data acquisition. Data were subsequently used to reconstruct treatment thermometry profiles from each point of acquisition. Toward the end of the transurethral resection or vaporization, a stepwise sequential pull back technique was used for the lateral lobe and apical sensor in 2 patients in each group to search for any previously undetected hot spots. Differences in total energy use, energy used per minute and operating time were compared using a nonparametric 2-sample Wilcoxon rank sum test. For transurethral resection of the prostate the resected tissue was weighed before formalin fixation so that energy use could be calculated for resected weight. RESULTS
Patient characteristics with respect to age, symptoms and objective voiding parameters were well matched for each group. Patients with acute urinary retention were equally divided between the 2 groups based on stratification for prostate size before randomization. After insertion of the resectoscope sheath with continuous irrigation inflow at room temperature, there was no change in recorded temperatures from the baseline values at the 4 measurement points. Representative thermometry profiles during transurethral resection of the prostate and transurethral electrosurgical vaporization are shown in figure 2. Temperature elevation 3C greater than baseline was recorded at the distal rectal probe position in 1patient treated with regular loop resection, and of 1.5C in 1treated with electrosurgical vaporization. No change was noted at the proximal rectal site in any patient. Temperature elevations at the lateral lobe site were noted in 4 and 5 patients undergoing transurethral resection and electrosurgical vaporization, respectively, while no heating occurred at any site in the remaining 3 patients. The range of lateral lobe heating was 1.5to 26 and 2 to 20C, respectively. All changes greater than baseline were of short duration (lasting less than 1 minute each and no longer than 2 to 6 minutes cumulatively) and occurred only after the active electrode had been used to clear sufficient adenomatous tissue, so that it closely approximated the sensor tip a few millimeters away. The probe in the neurovascular bundle a t the external sphincter recorded a temperature decrease of 2 to 1OC in 2 patients undergoing transurethral resection of the prostate and 2 to 4C in 2 undergoing transurethral electrosurgical vaporization (fig. 2). These changes were observed only when the apical tissue had been adequately excavated. No hidden hot spots were detected when probes were slowly withdram
PROSTATE HEATING PATTERNS OF ELECTROSURGICAL TRANSURETHRAL RESECTION AND VAPORIZATION 171 -M M 55 55
0-
I
50
. - -. *?a*
-
-
-
50
-~ *-*
< -I
'5
CF&"
Po-'.
-
-,*.bm
i-dl.WRn.l P m I C b .
'
f-
-&.mi
My
,
Wm.mUIncUl
?5.
35 yi yi
25
15 0
A
3
10
I5
20
25
Time (rnln)
yi
35
m
0
n m (mini
FIG. 2. Thermometry profile in real time. A, during transurethral loop resection. B , during transurethral electrovaporization with VaporTrode grooved bar. min, minutes. TABLE2. Characteristics of voiding dysfunction in loop resection in a stepwise fashion during the final phase of treatment for and electrosuraical vaDorizatwn ErOUDS 3-dimensional temperature mapping. Mean (range) In the transurethral resection group generator panel power settings of 120 to 170 watts were necessary in the pure cut mode. A similar range of settings was also effective for transurethral electrosurgical vaporization (130to 190 watts). These panel settings represented the maximum power deliv- Pt.age (yrs.) 65.8 (59-71) 67 (6045) ery (kept relatively constant at any given time for a range of I-PSS (pts. without retention) 23.3 (17-29) 29.6 (28-31) 7.5 (5.1-11) 10 (7.3-13.1) tissue impedances by virtue of computer controlled technol- Peak flow (mlJsec.) 64.6 (31.5-119) 54 (25-90) vol. ogy in the radiofrequency generator). No sigmficant differ- Prostate 64.3 (40-120) Operative time (mins.) 66 (27-95) ences were found in prostate volume (p <0.57) or operating No. requiring postop. imgation 3 6 times between the 2 procedures (p <0.33).Total energy use Hospital stay (days) 2.6 (24) 1.8 (1-2) 2.6 (2-4) 2 ( 1-3) was significantly greater for transurethral electrosurgical Catheterization (days) 3 mos. postop. 3.2 (1-5) 3.5 (24) vaporization (p <0.025, table 1).When corrected for proce- I-PSS Peak flow 3 mos. postop. (mllsec.) 22.6 (19.3-25.2) 21.4(17.2-25.3) dure length, the energy used per minute for transurethral Mean number of patients in each group waa 6,of whom 3 had retention. resection of the prostate ranged from 555 to 977 J. (median There were no postoperative complications. No patient required transfusions. 753),while the equivalent range for transurethral electrosurgical vaporization was 1,286 to 2,010 J. (median 1,657,p c0.004).The greater energy use during the latter procedure vaporization has been the need for high curwas not associated with any undesirable remote heating of electrosurgical rent density, which provides the rapid tissue heating necesperiprostatic tissues. sary for vaporization, and lower current density areas at the There was no loss of full preoperative potency during the electrode contact surface, which provides broad coagulative first 6 months of followup in 11 of the 12 patients studied, hemostasis. To achieve both effects concurrently using the while 1 had partial erections that did not deteriorate after VaporTrode grooved bar, panel power settings of 200 to 300 transurethral resection of the prostate. The subjective and watts in monopolar pure cut mode have been used on a objective improvements in voiding status were comparable variety of existing radiofrequency generators.2.3 These high but hospital stay and catheter times were shorter after trans- panel settings have led to hypothetical concerns of unseen urethral electrosurgical vaporization (table 2). heat damage to vital periprostatic structures (neurovascular bundles and external sphincter). DISCUSSION The effects of internal organ architecture, vascularity and Electrosurgical vaporization of the prostate is the latest of tissue inhomogeneity on radiofrequency power deposition in several new treatments that have been sought to challenge the adenomatous human prostate are poorly understood and the supremacy of transurethral resection of the prostate. dBicult to quantify in a meaningful dynamic manner in the Through evolution from electrodesiccation by virtue of mod- individual patient prostate, and cannot be simulated accuified electrode design, the technique has gained rapid popu- rately in tissue phantoms. The use of a continuous flow larity despite the shortfall of clinical data from large random- irrigation resectoscope can offset some of the risks of tissue ized series. Key to the successful achievement of bloodless heating by acting as a heat venting system. Furthermore, an apparent protective effect around blood vessels from circulatory cooling has been reported in the rabbit liver model.4 Studies of thermometry in the canine model during treatTABLE1. Parameters of conventional loop resection and ment with an electrode similar in tip design to the Vaportransurethral electrosurgical vaporization compared with the 2"rode coupled to a Valleylab Force 4B generatoe have consample Wilcoxon mnk sum test firmed the heat sink effect of continuous irrigation and Median/Mean suggested minimal deep tissue heating. Temperature elevaTransurethral Trans~rethral p Value tions of 2 to 4.3C were found 5 to 7 mm. from the active Electrosurgical electrode, and there was no heating of the rectal serosa or Resection Vaporization area analogous to the human neurovascular bundle. hostate vol. (cc) 52.75164.67 52.954 0.57 Our prospective comparison of transurethral resection of Operating time (mins.) 44/66 62.5'64.3 0.34 the prostate and transurethral electrosurgical vaporization Enerm: with stratification for prostate size demonstrated that both 36,52563,468 109,827/102,910 0.025 To% J. J./min. 753/770 1,657/1,620 0.004 electrosurgical treatments were free of serious complications 15.314.52 31.JY35.75 0.016 JJmin./ccprostate at wwer levels of less than 190 watts with a mo-. incident ~.~~~
172
PROSTATE HEATING PATTERNS OF ELECTROSURGICAL TRANSURETHRAL RESECTION AND VAPORIZATION
nopolar cutting wave form current. Neither treatment was associated with unwanted heat damage to vital tissues at the perimeter of the prostate. Furthermore, no significant differences in prostate heating patterns in man were apparent as a result of electrode configuration comparing the loop to the more bulky VaporTrode grooved bar. Although there was greater overall energy used during transurethral electrosurgical vaporization, as one might expect from the greater contact surface area of the electrode, there was no correlation between energy used and observed temperatures at the perimeter of the gland, since these values remained within normal limits at all times, Since the undampened low peak voltage current used for electrosurgical cutting inherently has a shallow depth of tissue penetration, it is likely that any heat generated locally during tissue vaporization can be easily dissipated by the continuous flow of irrigant, and by hyperemia within the prostate and its surrounding tissue. The modest equivalent short-lived temperature increases detected in the lateral lobe sensors, as the active electrode approached the sensor tip during transurethral resection of the prostate and transurethral electrosurgical vaporization, attested to the shallow tissue penetration of heat from transurethral applications of high density cutting radiofrequency current. Because the cavity was extended to the prostatic apex, the neurovascular bundles and sphincter were not heated but were occasionally cooled despite the slower movement (and greater energy deposition) of the VaporTrode electrode. It is possible that this protective effect may have been related to the i d o w of cool fresh irrigant or reactive vasodilatation and increased flow velocity in the large veins that intimately accompany the erectile nerves and drain the pelvic floor musculature at this site. These possibilities merit further study in the future. Therefore, if radiofrequency energy applied to the prostate during either procedure is ultimately responsible for impaired activity in the erectile nerves (to account for historical observations of postoperative impotence) or transient urinary incontinence, the exclusion of a thermal etiology by our study suggests another causative mechanism that has not been elucidated. It is theoretically possible that demodulated low frequency, low voltage currents may be induced in an inconsistent and unpredictable manner during nonlinear arcing at the active electrode or by the phenomenon of capacitance in the metallic resectoscope sheath, particularly when high incident powers are used.* Stray currents of this type could then discharge to the dispersive electrode through small localized points of contact between the resectoscope sheath and urethra at the prostatic apex or sphincter, where the effects of a n insulating low conductivity lubricant gel may be absent.? It is unknown whether this latter phenomenon can impede erectile nerve function. However, if this postulate is valid a greater incidence of impotence might be expected in patients subjected to high power electrosurgical vaporization compared to those treated with equal efficacy at a lower power. This effect should be carefully documented from the studies currently in progress as longer followup is accrued. Alternatively, the use of excessive coagulating current at the prostatic apex may itself induce neural damage, since this pulsed wave form has a greater peak voltage and, therefore, a greater depth of tissue penetration than the pure cut mode used for vaporization. Wary of this possibility, a low power setting of 40 watts in spray coagulation mode was used sparingly in short application bursts to achieve effective coagulative hemostasis at the prostatic apex a t the end of the vaporization procedure. In addition, the temptation to use high dial settings or frequent application of coagulating current alongside high cut settings in the monopolar mode should be resisted, particularly if imtative symptoms from slowly separating tissue slough (as with laser coagulation of the prostate) are to be avoided in the early postoperative period.
The use of a computer controlled constant power output radiofrequency generator that responded instantly to changes in tissue impedance permitted effective treatment during both techniques a t near equivalent dial power settings, typically of 130 to 170 watts. This device was chosen as the energy source in our study due to its reliability in maintaining constant power output for a wide range of tissue impedance. Thus, it differs from the majority of radiofrequency generators, which reach a peak power output a t low tissue impedance and whose output decreases dramatically with higher impedance tissue so that a large discrepancy exists between actual power output and the chosen panel setting on the unit. Since accurate quantification of the tissue removed by electrosurgical vaporization with the Vapor"rode was not possible in practical terms (unlike transurethral resection of the prostate, where the tissue can be weighed), it is impossible to be certain whether more energy was needed per gram tissue removed during transurethral electrosurgical vaporization. Although this event seems likely (from the greater energy used per minute of operative time, table l),the majority of the extra energy was probably invested in decreasing intraoperative bleeding through low power density contact areas of the electrode surface, since no unwanted remote heating was apparent. CONCLUSIONS
Transurethral resection of the prostate and transurethral electrosurgical vaporization have allowed for electrosurgical removal of obstructive prostate tissue in our study group without any serious complications. Furthermore, inadvertent heating of vital structures at the prostate perimeter, such as the neurovascular bundles and external sphincter, did not occur with the cutting wave form regardless of electrode design (loop or VaporTrode) at the lower panel power settings used in our study. Transurethral electrosurgical vaporization is still a treatment modality in evolution with the prospect of additional electrode tip designs and innovations in radiofrequency generator technology. If its early promise is fulfilled in acceptable longer clinical outcomes from controlled randomized prospective studies with larger numbers of patients, this method may eventually supersede transurethral resection of the prostate as the dominant surgical treatment for symptomatic bladder outflow obstruction from benign prostatic hyperplasia. Dr. Frederick Dorey assisted in the statistical analysis. REFERENCES
1. Benign Prostatic Hyperplasia: Diagnosis and Treatment. US. Department of Health and Human Services, Rockville, Maryland, 1994. 2. Kaplan, S.A. and Te, A.: A comparative study of transurethral resection of the prostate using a modified electro-vaporizing loop and transurethral laser vaporization of the prostate. J. Urol., 154: 1785,1995. 3. Stewart, S.C.,Benjamin, D., Ruckle, H., Lui, P. and Hadley, R.: Electrovaporization of the prostate: new technique for treatment of symptomatic benign hyperplasia. J. Endourol., 9 413,
1995. 4. Lounsberry,W., Goldschmidt, V., Linke, C. A,, Walder, H. J. and Chrzan, D.: The early histologic changes following electrow agulation. J. Urol., 86 321, 1961. 5. Perlmutter, A. P., Muschter, R. and Razvi, H. A,: Electrosurgkal vaporization of the prostate in the canine model. Urology, 46: 518, 1995. 6. Tucker, R. D., Schmitt, 0. H., Sievert, C. E. and Silvis, E.: Demodulated low frequency currents from electrosurgical pm cedures. Surg., Gynec. & Obst., 159 39,1984. 7. Flachenecker, G. and Fastenmeier, K.: High frequency current effects during transurethral resection. J. Urol., 122: 336. 1979.
THEJOURNAL OF UROLOGY Copyright 0 1997 by AMERICAN UROLOCICAL ASSOCIATION, INC.
Vol. 157. 169-172,January 1997 Printed in U S A .
PROSTATE HEATING PATTERNS COMPARING ELECTROSURGICAL TRANSURETHRAL RESECTION AND VAPORIZATION: A PROSPECTIVE RANDOMIZED STUDY ANUP PATEL, GERHARD J. FUCHS, JORGE GUTIERREZ-ACEVES AND THOMAS P. RYAN From the UCLA Department of Urology, Los Angeles, California, Uniuersidad de Guadalajara, Guadalajara, Mexico, and Research and Deuelopment Department, Valleylab, Boulder, Colorado
ABSTRACT
Purpose: A prospective study was performed to determine if transurethral electrosurgical vaporization of the prostate is associated with unseen heat damage to vital periprostatic structures compared to conventional loop resection. In addition, energy consumption and its relationship to observed tissue temperature at the prostate periphery were evaluated for each treatment. Materials and Methods: Patients with moderate to severe symptoms of benign prostatic bladder outflow obstruction and objective evidence of diminished flow or acute urinary retention were randomized to undergo either transurethral loop resection or eledrosurgical vaporization &r stratification for gland volume. Instrumentation was standardized for both groups except for the monopolar electrode used. The radiofrequency power source in the study was a new computer controlled generator with a constant power delivery feature. Regional tissue heating patterns were evaluated with optical fiber probes in real time. Four stationary sites were chosen for temperature measurements, namely the lateral lobe of the prostate, neurovascular bundle beside the prostatic apex at the level of the external sphincter, and anterior rectal wall at the level of the prostatic base and apex. A pull back technique was used to search for hot points in the long axis of the probe (3-dimensional temperature mapping) in 2 patients from each group. Incident generator panel power settings for the electrosurgical vaporization treatments were equivalent to those commonly used for loop resection (150 watts) and were adjusted up or down as needed. Results: Prostate electrosurgical vaporization was possible at generator panel settings that were nearly equivalent to those for transurethral resection of the prostate (130 to 190 watts). NO si&cant rectal or sphincteric heating was detected with either procedure. Conductive cooling of the neurovascular bundles was observed in 2 patients in each group toward the end of the operation. More energy was used per minute of treatment during electrosurgicalvaporization than with regular loop resection (p (0.004) but this was not associated with unwanted tissue heating. Conclusions: Neither conventional loop resection nor electrosurgical vaporization of the prostate appeared to be unsafe treatments with respect to unseen deep heating effects to vital periprostatic structures when performed at equivalent low incident power. The extra energy used during electrosurgical vaporization provided the benefit of improved coagulative hemostasis concurrently with shallow tissue vaporization using pure cutting current alone, without compromising treatment safety. KEY WORDS:prostate, prostatectomy, electrosurgery, heating, bladder neck obstruction Transurethral electrosurgical resection of the prostate re- on electrosurgical units.2 Since to our knowledge a real-time mains the gold standard treatment for symptomatic bladder study of prostate and periprostatic heating during electrosuroutflow obstruction from benign prostatic hyperplasia. The gical treatment (transurethral resection of the prostate or procedure has withstood the test of time by virtue of the fact transurethral electrosurgical vaporization) has never been that it is associated with rapid, often dramatic subjective and done, it is unclear whether an element of unseen heat damobjective success rates of 85 to 90%. However, disadvantages age might occur to explain any of the observed complications of morbidity, including bleeding, intraoperative fluid absorp- relating to potency and continence. In our study the hypothtion, infection, retrograde ejaculation, stricture formation, esis that unseen regional thermal heating of vital periprosurinary incontinence and erectile impotence, are all well tatic structures may occur was tested in a prospectively randocumented in the urological literature.' The etiology of this domized fashion during conventional transurethral resection last complication is still poorly understood but, with the of the prostate and transurethral electrosurgical vaporizaadvent of the new treatment of transurethral electrosurgical tion with the VaporTrode* grooved bar electrode. vaporization of the prostate, the possibility of unseen therPATIENTS AND METHODS ma1 injury to the erectile nerves has been highlighted, particularly since high panel power settings have been required Patients with symptomatic bladder outlet obstruction were evaluated by self-administered international prostate sympAccepted for publication July 26, 1996. tom scores (I-PSS),urine culture and uroflowmetry. Prostate Editor's Note: This article is the third of 6 p ~ b l k h d in this volume as measured by transrectal ultrasound wm calcuissue for which category 1 CME credits can be earned. InStructions for obtaining credits are given with the questions on pages 336 and 837. 169
* Circon ACMI, Stamford, Connecticut.
170 PROSTATE HEATING PATTERNS OF ELECTROSURGICAL TRANSURETHRAL RESECTION AND VAPORIZATION lated using the prolate sphere formula (0.523 X AF' X width x length). Six patients with moderate or severe symptoms of bladder outlet obstruction and 6 with acute urinary retention were selected for surgery provided there was no coexistent neurological cause of voiding dysfunction and no urinary infection. The study protocol was approved by the local ethical committee and all study patients were required to provide informed consent. ARer stratification for prostate size (larger or smaller than 50 ~ m . ~all) ,patients were randomized to undergo either loop resection or electrosurgical vaporization with the VaporTrode grooved bar electrode. All treatments were performed with a 25F continuous flow resectoscope and water irrigation. A new (Force FX*)computer controlled tissue response radiofrequency generator was used as the energy source throughout the study, since this device has been shown by the manufacturer to maintain constant power for a wider range of tissue impedance than conventional radiofrequency generators. Incident panel power settings in pure cut (150 watts) and spray coagulation (40 watts) modes, which had a range similar to conventional settings for transurethral resection of the prostate, were selected at the start of each treatment and adjusted up or down as indicated by the observed tissue effect. In this manner we hoped to obtain a true comparison of differences in tissue heating related to active electrode configuration as the main variable tested. Generator power delivery parameters were also measured during each procedure using a passive feed-through measurement system connected to the radiofrequency output of the generator, transparent to the surgeon. The device was insulated to eliminate any patient hazards during treatment. This system monitored current and voltage, and calculated phase, power, energy, impedance and derivatives of all of these measurements in real time. Data were acquired at 10 Hz. during each activation of cutting current and stored every 10 minutes for later off-line analysis. Patients were prepared and draped in sterile fashion in the lithotomy position before treatment. The perineum was shaved and prepared along with the external genitalia. A linear array 7 MHz. transrectal ultrasound probe was introduced through the anal verge into the rectum, and used to image the external sphincter, prostate and bladder neck. Using real-time transrectal ultrasound guidance, transperineal puncture was performed with a sharp-tipped closed ended 18 gauge polytetrafluoroethylene (Teflon)catheter and tungsten trocar. The first of 2 such catheters was guided into the trapezoid area of the external urethral sphincter, and its tip was positioned adjacent to the right neurovascular bundle beside the prostatic apex. The second catheter tip was placed within the left lateral lobe 15 to 20 mm. parallel to the prostatic urethra at the mid gland level. The separation of the tip of this second catheter from the urethra was measured using sector transrectal ultrasound scans. The trocars of both catheters were replaced by optical thermosensor fibers pre-calibrated in a constant temperature water bath at the table side to ensure 0.1Caccuracy and secured in position with tape markers. Using the polytetrafluoroethylene catheters the phosphor sensor tips were kept dry and free of false readings from adherent blood contamination. Sensor positions were confirmed with the ultrasound probe. Finally, 2 additional sensors mounted on the surface of a specially designed rectal probe were placed 3 and 8 cm. from the anal verge against the anterior rectal wall to detect undesirable rectal heating at the level of the bladder neck and prostatic apex or pelvic floor. Probe placement is shown in figure 1. All 4 fiberoptic sensors were connected via a fiberoptic extension to a thermometry mainframe device. Each probe was able to measure up to a 120C limit. Prostate temperatures were sampled every 2 seconds in real time after equilibrium had
* Valleylab. Boulder, Colorado.
Optical Thermosenson 1 = lateral lobe sensor 2 = sphincter-NV bundle sensor 3 = proximal rectal sensor 4 =distal rectal sensor
Bladder
Rectal probe
P
FIG.1. Placement of optical fiber thermometry probes. NV, neurovascular.
been attained at baseline and throughout each treatment. A laptop computer was used for data acquisition. Data were subsequently used to reconstruct treatment thermometry profiles from each point of acquisition. Toward the end of the transurethral resection or vaporization, a stepwise sequential pull back technique was used for the lateral lobe and apical sensor in 2 patients in each group to search for any previously undetected hot spots. Differences in total energy use, energy used per minute and operating time were compared using a nonparametric 2-sample Wilcoxon rank sum test. For transurethral resection of the prostate the resected tissue was weighed before formalin fixation so that energy use could be calculated for resected weight. RESULTS
Patient characteristics with respect to age, symptoms and objective voiding parameters were well matched for each group. Patients with acute urinary retention were equally divided between the 2 groups based on stratification for prostate size before randomization. After insertion of the resectoscope sheath with continuous irrigation inflow at room temperature, there was no change in recorded temperatures from the baseline values at the 4 measurement points. Representative thermometry profiles during transurethral resection of the prostate and transurethral electrosurgical vaporization are shown in figure 2. Temperature elevation 3C greater than baseline was recorded at the distal rectal probe position in 1patient treated with regular loop resection, and of 1.5C in 1treated with electrosurgical vaporization. No change was noted at the proximal rectal site in any patient. Temperature elevations at the lateral lobe site were noted in 4 and 5 patients undergoing transurethral resection and electrosurgical vaporization, respectively, while no heating occurred at any site in the remaining 3 patients. The range of lateral lobe heating was 1.5to 26 and 2 to 20C, respectively. All changes greater than baseline were of short duration (lasting less than 1 minute each and no longer than 2 to 6 minutes cumulatively) and occurred only after the active electrode had been used to clear sufficient adenomatous tissue, so that it closely approximated the sensor tip a few millimeters away. The probe in the neurovascular bundle a t the external sphincter recorded a temperature decrease of 2 to 1OC in 2 patients undergoing transurethral resection of the prostate and 2 to 4C in 2 undergoing transurethral electrosurgical vaporization (fig. 2). These changes were observed only when the apical tissue had been adequately excavated. No hidden hot spots were detected when probes were slowly withdram
PROSTATE HEATING PATTERNS OF ELECTROSURGICAL TRANSURETHRAL RESECTION AND VAPORIZATION 171 -M M 55 55
0-
I
50
. - -. *?a*
-
-
-
50
-~ *-*
< -I
'5
CF&"
Po-'.
-
-,*.bm
i-dl.WRn.l P m I C b .
'
f-
-&.mi
My
,
Wm.mUIncUl
?5.
35 yi yi
25
15 0
A
3
10
I5
20
25
Time (rnln)
yi
35
m
0
n m (mini
FIG. 2. Thermometry profile in real time. A, during transurethral loop resection. B , during transurethral electrovaporization with VaporTrode grooved bar. min, minutes. TABLE2. Characteristics of voiding dysfunction in loop resection in a stepwise fashion during the final phase of treatment for and electrosuraical vaDorizatwn ErOUDS 3-dimensional temperature mapping. Mean (range) In the transurethral resection group generator panel power settings of 120 to 170 watts were necessary in the pure cut mode. A similar range of settings was also effective for transurethral electrosurgical vaporization (130to 190 watts). These panel settings represented the maximum power deliv- Pt.age (yrs.) 65.8 (59-71) 67 (6045) ery (kept relatively constant at any given time for a range of I-PSS (pts. without retention) 23.3 (17-29) 29.6 (28-31) 7.5 (5.1-11) 10 (7.3-13.1) tissue impedances by virtue of computer controlled technol- Peak flow (mlJsec.) 64.6 (31.5-119) 54 (25-90) vol. ogy in the radiofrequency generator). No sigmficant differ- Prostate 64.3 (40-120) Operative time (mins.) 66 (27-95) ences were found in prostate volume (p <0.57) or operating No. requiring postop. imgation 3 6 times between the 2 procedures (p <0.33).Total energy use Hospital stay (days) 2.6 (24) 1.8 (1-2) 2.6 (2-4) 2 ( 1-3) was significantly greater for transurethral electrosurgical Catheterization (days) 3 mos. postop. 3.2 (1-5) 3.5 (24) vaporization (p <0.025, table 1).When corrected for proce- I-PSS Peak flow 3 mos. postop. (mllsec.) 22.6 (19.3-25.2) 21.4(17.2-25.3) dure length, the energy used per minute for transurethral Mean number of patients in each group waa 6,of whom 3 had retention. resection of the prostate ranged from 555 to 977 J. (median There were no postoperative complications. No patient required transfusions. 753),while the equivalent range for transurethral electrosurgical vaporization was 1,286 to 2,010 J. (median 1,657,p c0.004).The greater energy use during the latter procedure vaporization has been the need for high curwas not associated with any undesirable remote heating of electrosurgical rent density, which provides the rapid tissue heating necesperiprostatic tissues. sary for vaporization, and lower current density areas at the There was no loss of full preoperative potency during the electrode contact surface, which provides broad coagulative first 6 months of followup in 11 of the 12 patients studied, hemostasis. To achieve both effects concurrently using the while 1 had partial erections that did not deteriorate after VaporTrode grooved bar, panel power settings of 200 to 300 transurethral resection of the prostate. The subjective and watts in monopolar pure cut mode have been used on a objective improvements in voiding status were comparable variety of existing radiofrequency generators.2.3 These high but hospital stay and catheter times were shorter after trans- panel settings have led to hypothetical concerns of unseen urethral electrosurgical vaporization (table 2). heat damage to vital periprostatic structures (neurovascular bundles and external sphincter). DISCUSSION The effects of internal organ architecture, vascularity and Electrosurgical vaporization of the prostate is the latest of tissue inhomogeneity on radiofrequency power deposition in several new treatments that have been sought to challenge the adenomatous human prostate are poorly understood and the supremacy of transurethral resection of the prostate. dBicult to quantify in a meaningful dynamic manner in the Through evolution from electrodesiccation by virtue of mod- individual patient prostate, and cannot be simulated accuified electrode design, the technique has gained rapid popu- rately in tissue phantoms. The use of a continuous flow larity despite the shortfall of clinical data from large random- irrigation resectoscope can offset some of the risks of tissue ized series. Key to the successful achievement of bloodless heating by acting as a heat venting system. Furthermore, an apparent protective effect around blood vessels from circulatory cooling has been reported in the rabbit liver model.4 Studies of thermometry in the canine model during treatTABLE1. Parameters of conventional loop resection and ment with an electrode similar in tip design to the Vaportransurethral electrosurgical vaporization compared with the 2"rode coupled to a Valleylab Force 4B generatoe have consample Wilcoxon mnk sum test firmed the heat sink effect of continuous irrigation and Median/Mean suggested minimal deep tissue heating. Temperature elevaTransurethral Trans~rethral p Value tions of 2 to 4.3C were found 5 to 7 mm. from the active Electrosurgical electrode, and there was no heating of the rectal serosa or Resection Vaporization area analogous to the human neurovascular bundle. hostate vol. (cc) 52.75164.67 52.954 0.57 Our prospective comparison of transurethral resection of Operating time (mins.) 44/66 62.5'64.3 0.34 the prostate and transurethral electrosurgical vaporization Enerm: with stratification for prostate size demonstrated that both 36,52563,468 109,827/102,910 0.025 To% J. J./min. 753/770 1,657/1,620 0.004 electrosurgical treatments were free of serious complications 15.314.52 31.JY35.75 0.016 JJmin./ccprostate at wwer levels of less than 190 watts with a mo-. incident ~.~~~
172
PROSTATE HEATING PATTERNS OF ELECTROSURGICAL TRANSURETHRAL RESECTION AND VAPORIZATION
nopolar cutting wave form current. Neither treatment was associated with unwanted heat damage to vital tissues at the perimeter of the prostate. Furthermore, no significant differences in prostate heating patterns in man were apparent as a result of electrode configuration comparing the loop to the more bulky VaporTrode grooved bar. Although there was greater overall energy used during transurethral electrosurgical vaporization, as one might expect from the greater contact surface area of the electrode, there was no correlation between energy used and observed temperatures at the perimeter of the gland, since these values remained within normal limits at all times, Since the undampened low peak voltage current used for electrosurgical cutting inherently has a shallow depth of tissue penetration, it is likely that any heat generated locally during tissue vaporization can be easily dissipated by the continuous flow of irrigant, and by hyperemia within the prostate and its surrounding tissue. The modest equivalent short-lived temperature increases detected in the lateral lobe sensors, as the active electrode approached the sensor tip during transurethral resection of the prostate and transurethral electrosurgical vaporization, attested to the shallow tissue penetration of heat from transurethral applications of high density cutting radiofrequency current. Because the cavity was extended to the prostatic apex, the neurovascular bundles and sphincter were not heated but were occasionally cooled despite the slower movement (and greater energy deposition) of the VaporTrode electrode. It is possible that this protective effect may have been related to the i d o w of cool fresh irrigant or reactive vasodilatation and increased flow velocity in the large veins that intimately accompany the erectile nerves and drain the pelvic floor musculature at this site. These possibilities merit further study in the future. Therefore, if radiofrequency energy applied to the prostate during either procedure is ultimately responsible for impaired activity in the erectile nerves (to account for historical observations of postoperative impotence) or transient urinary incontinence, the exclusion of a thermal etiology by our study suggests another causative mechanism that has not been elucidated. It is theoretically possible that demodulated low frequency, low voltage currents may be induced in an inconsistent and unpredictable manner during nonlinear arcing at the active electrode or by the phenomenon of capacitance in the metallic resectoscope sheath, particularly when high incident powers are used.* Stray currents of this type could then discharge to the dispersive electrode through small localized points of contact between the resectoscope sheath and urethra at the prostatic apex or sphincter, where the effects of a n insulating low conductivity lubricant gel may be absent.? It is unknown whether this latter phenomenon can impede erectile nerve function. However, if this postulate is valid a greater incidence of impotence might be expected in patients subjected to high power electrosurgical vaporization compared to those treated with equal efficacy at a lower power. This effect should be carefully documented from the studies currently in progress as longer followup is accrued. Alternatively, the use of excessive coagulating current at the prostatic apex may itself induce neural damage, since this pulsed wave form has a greater peak voltage and, therefore, a greater depth of tissue penetration than the pure cut mode used for vaporization. Wary of this possibility, a low power setting of 40 watts in spray coagulation mode was used sparingly in short application bursts to achieve effective coagulative hemostasis at the prostatic apex a t the end of the vaporization procedure. In addition, the temptation to use high dial settings or frequent application of coagulating current alongside high cut settings in the monopolar mode should be resisted, particularly if imtative symptoms from slowly separating tissue slough (as with laser coagulation of the prostate) are to be avoided in the early postoperative period.
The use of a computer controlled constant power output radiofrequency generator that responded instantly to changes in tissue impedance permitted effective treatment during both techniques a t near equivalent dial power settings, typically of 130 to 170 watts. This device was chosen as the energy source in our study due to its reliability in maintaining constant power output for a wide range of tissue impedance. Thus, it differs from the majority of radiofrequency generators, which reach a peak power output a t low tissue impedance and whose output decreases dramatically with higher impedance tissue so that a large discrepancy exists between actual power output and the chosen panel setting on the unit. Since accurate quantification of the tissue removed by electrosurgical vaporization with the Vapor"rode was not possible in practical terms (unlike transurethral resection of the prostate, where the tissue can be weighed), it is impossible to be certain whether more energy was needed per gram tissue removed during transurethral electrosurgical vaporization. Although this event seems likely (from the greater energy used per minute of operative time, table l),the majority of the extra energy was probably invested in decreasing intraoperative bleeding through low power density contact areas of the electrode surface, since no unwanted remote heating was apparent. CONCLUSIONS
Transurethral resection of the prostate and transurethral electrosurgical vaporization have allowed for electrosurgical removal of obstructive prostate tissue in our study group without any serious complications. Furthermore, inadvertent heating of vital structures at the prostate perimeter, such as the neurovascular bundles and external sphincter, did not occur with the cutting wave form regardless of electrode design (loop or VaporTrode) at the lower panel power settings used in our study. Transurethral electrosurgical vaporization is still a treatment modality in evolution with the prospect of additional electrode tip designs and innovations in radiofrequency generator technology. If its early promise is fulfilled in acceptable longer clinical outcomes from controlled randomized prospective studies with larger numbers of patients, this method may eventually supersede transurethral resection of the prostate as the dominant surgical treatment for symptomatic bladder outflow obstruction from benign prostatic hyperplasia. Dr. Frederick Dorey assisted in the statistical analysis. REFERENCES
1. Benign Prostatic Hyperplasia: Diagnosis and Treatment. US. Department of Health and Human Services, Rockville, Maryland, 1994. 2. Kaplan, S.A. and Te, A.: A comparative study of transurethral resection of the prostate using a modified electro-vaporizing loop and transurethral laser vaporization of the prostate. J. Urol., 154: 1785,1995. 3. Stewart, S.C.,Benjamin, D., Ruckle, H., Lui, P. and Hadley, R.: Electrovaporization of the prostate: new technique for treatment of symptomatic benign hyperplasia. J. Endourol., 9 413,
1995. 4. Lounsberry,W., Goldschmidt, V., Linke, C. A,, Walder, H. J. and Chrzan, D.: The early histologic changes following electrow agulation. J. Urol., 86 321, 1961. 5. Perlmutter, A. P., Muschter, R. and Razvi, H. A,: Electrosurgkal vaporization of the prostate in the canine model. Urology, 46: 518, 1995. 6. Tucker, R. D., Schmitt, 0. H., Sievert, C. E. and Silvis, E.: Demodulated low frequency currents from electrosurgical pm cedures. Surg., Gynec. & Obst., 159 39,1984. 7. Flachenecker, G. and Fastenmeier, K.: High frequency current effects during transurethral resection. J. Urol., 122: 336. 1979.