- Email: [email protected]
Comparison of Novel Tissue Apposing Device and Standard Anastomotic Technique for Vesicourethral Anastomoses
Basic Science Comparison of Novel Tissue Apposing Device and Standard Anastomotic Technique for Vesicourethral Anastomoses Gregory Hruby, Kyle J. Weld, Franzo Marruffo, Sean Collins, Evren Durak, Rob Mitchell, Alan Herron, and Jaime Landman OBJECTIVES METHODS
RESULTS
CONCLUSIONS
To evaluate a novel sutureless tissue apposing vesicourethral anastomosis (VUA) device in a porcine model and compare it with standard laparoscopically sutured VUA. Thirty domestic pigs were divided into six groups. In groups 1, 2, and 3, a standard laparoscopic sutured running VUA was performed. In groups 4, 5, and 6, a novel device VUA was performed. In all cases, cystography was completed immediately after completion of the anastomosis and when each pig was killed. At necropsy, the gross findings of the VUA were documented, and each anastomosis was excised en bloc for histopathologic evaluation of healing parameters. In the 30 pigs, 29 (97%) device or sutured VUAs were successfully performed laparoscopically without conversion to an open approach. The mean operative time for the standard and device groups was 87 and 68 minutes, respectively (P ⫽ 0.04). The anastomotic time for the standard and device groups was 41 and 12 minutes, respectively (P ⬍0.01). Histopathologic evaluation of the groups at 1 week of follow-up revealed significantly lower fibrosis scores for the novel anastomosis device VUA compared with the standard sutured VUA (median score 1 and 3, respectively; P ⫽ 0.04). The evaluation of groups 2 and 5 (3-week survival) and groups 3 and 6 (7-week survival) revealed no significant differences in any of the histopathologic parameters evaluated. The novel device requires little technical skill to deploy and is expeditious, requiring less time than a standard sutured anastomosis. At the 1 week follow-up point, histopathologic examination revealed that the novel device was superior regarding fibrotic reactions. UROLOGY 70: 190 –195, 2007. © 2007 Elsevier Inc.
T
he vesicourethral anastomosis (VUA) is one of the more technically demanding and time-consuming components of laparoscopic prostatectomy. Many of the complications associated with open or laparoscopic radical prostatectomy are related to the VUA. With the contemporary sutured VUA, the surgeon must overcome significant challenges to successfully perform a high-quality VUA, including tissue exposure in the confined area of the pelvis, fragile tissue, and physician fatigue at this point of the operation.1 Even after the learning curve has been overcome, the VUA can still pose a problem for the surgeon.2 Another problem with contemporary anastomosis is the application of absorbable suture. Absorbable suture, by definition, engenders a local inflammatory reaction, potentially compromising the healing process.
From the Department of Urology and Institute of Comparative Medicine, Columbia University Medical Center, New York, New York; Department of Urology, Wilford Hall Medical Center, Lackland Air Force Base, Texas Reprint requests: Jaime Landman, M.D., Department of Urology, Columbia University Medical Center, 161 Fort Washington Avenue, Room 1111, New York, NY 10032. E-mail: [email protected] Submitted: May 12, 2006; accepted (with revisions): February 27, 2007
190
© 2007 Elsevier Inc. All Rights Reserved
We evaluated a novel device (American Medical Systems, Minnetonka, Minn) that incorporates retractable tines that approximate the bladder and urethral tissues and simultaneously allow for bladder decompression with a drainage lumen and balloon. The 22F device with an inner lumen is equivalent to that of a 16F Foley catheter and consists of two opposing sets of circumferential nitinol tines, six bladder and six urethral tines, and a standard Foley balloon. The tines are 9.5 mm in length and have a diameter of 0.2 mm. The bladder tines are extended from the body of the catheter just below the Foley balloon, and the urethral tines are at a fixed axial distance from the bladder tines, ensuring urethral tine deployment within the membranous urethra. The VUA device was compared with standard sutured anastomosis for surgical efficiency and tissue healing.
MATERIAL AND METHODS The Institutional Animal Care and Use Committee approved the study. Thirty female domestic pigs were divided into six groups to compare a standard running anastomosis and a VUA 0090-4295/07/$32.00 doi:10.1016/j.urology.2007.02.061
Table 1. Animal groups
Group
Anastomosis Type
Animals (n)
Length of Foley Decompression (wk)
1 2 3 4 5 6
Running sutured Running sutured Running sutured VUA device VUA device VUA device
6 6 3 6 6 3
1 3 1 1 3 1
Survival Time (wk) 1 3 7 1 3 7
VUA ⫽ vesicourethral anastomosis.
created with the novel device. The study design is summarized in Table 1. The animals were initially anesthetized with ketamine 5 to 10 mg/kg and then intubated. The anesthetic was maintained with 2% to 5% Isofuorane. In a sterile manner, a Foley catheter was placed in the bladder, and the pig was prepared and draped for transperitoneal access, which was gained using three trocars: one at the umbilicus and two inferior to the umbilicus and lateral to the rectus abdominus muscles. The bladder was mobilized, and the bladder neck was dissected. The ureters were identified, and the urethra was incised caudal to the ureteral orifices. In all cases, the bladder neck was incised anteriorly, and a “tennis racket” closure was performed posteriorly with interrupted 2-0 Vicryl suture on an SH needle (Ethicon, Somerville, NJ) to reconstruct the bladder neck and prevent ureteral orifice occlusion. In the control groups (groups 1 to 3), a standard Van Velthoven running anastomosis was then performed using a 2-0 Vicryl suture on an SH needle.3 In groups 4 to 6, the VUA was completed using the novel device. The device was placed into the bladder as one would place a regular urethral catheter, the 15-mL balloon was inflated, and the bladder tines were deployed using the housing control unit. Monitored under laparoscopic vision, gentle caudal traction on the device approximated the bladder neck to the urethral stump. The urethral tines were then extended from the body of the device, and the anastomosis was completed (Fig. 1). No sutures were placed. Device removal consisted of first complete urethral and then complete bladder tine retraction into the device. The device was then removed from the animal in a fashion similar to standard Foley catheter removal. The operative and VUA times were recorded. After each anastomosis, all animals underwent a visual inspection of the anastomosis using a 250-mL cystogram (Conray, Mallinckrodt Institute, St. Louis, Mo) performed by gravity infusion at a pressure of 40 cm H2O. Fluoroscopy was used to grade the extravasation of the anastomosis (0, none; 1, minimal; 2, moderate; and 3, severe). After cystography, the bladder was allowed to drain to completion by gravity drainage, and the fluid returned was collected and measured. All catheters were secured to the pig with 2-0 nonabsorbable Prolene suture (Ethicon) every 2 cm. Groups 1 and 4, 2 and 5, and 3 and 6 were killed at 1, 3, and 7 weeks postoperatively, respectively. Beforehand, cystography was again performed in the same manner. After death, the anastomotic tissue was harvested en bloc and placed in 5% formalin for fixation and later histopathologic examination. Gross observation of circumferential healing was documented at the time of death. Each anastomosis was divided into three sections, placed in tissue cassettes for permanent sectioning, and stained with hematoxylin-eosin in a standard manner. UROLOGY 70 (1), 2007
Figure 1. Novel anastomotic device (A,E) inserted through urethra and placed inside bladder. (B) Foley balloon inflated with 15 mL sterile water. (C) Bladder tines deployed using housing control unit. Slight caudal retraction used to approximate bladder neck to urethral stump. (D) Urethral tines deployed using housing control unit, and anastomosis is completed. Three slides from each cassette were evaluated and graded by a pathologist (A.H.) unaware of the treatment groups on a grading scale of 0 to 3 (0, none; 1, minimal; 2, moderate; and 3, severe). The parameters evaluated included the inflammatory reactions (mural, lamina propria, edema, and perianastomotic fat inflammation), fibrotic reactions (muscle, lamina propria, granulation, perianastomotic fat, and myofibroblasts), and other categories (foreign body reaction, squamous metaplasia, mucinous metaplasia, hemosiderin deposits, perianastomotic hemorrhage, necrotic debris, and intact urothelium). A two-sample mean comparison test was used to evaluate the operative times, fluid returned on the postoperative and just before death cystograms, and animal weight. We used a multi191
Table 2. Comparison of median histopathologic changes in groups 1 and 4 Group Histologic Finding
Device
Complete gross circumferential healing (%) Intact urothelium (%) Inflammatory reactions Mural inflammation Lamina propria inflammation Edema Perianastomotic fat inflammation Fibrotic reaction Muscle layer fibrosis Lamina propria fibrosis Granulation tissue Perianastomotic fat fibrosis Presence of myofibroblasts Other Focal foreign body reaction Squamous metaplasia Mucinous metaplasia Hemosiderin deposits Perianastomotic hemorrhage Necrotic debris
83
Control 20
P Value 0.08
66
0
0.18
2.5 2 1.5 2
3 2.5 2 2
0.56 0.43 0.23 0.92
1 0 0.5 0.5 0
3 3 2 1.5 0
0.04 0.03 0.05 0.08 NS
0 0 0 0.25 0.5 1
1 0 0 0 1 0.5
0.23 NS NS NS NS NS
NS ⫽ not significantly different between groups; 0 ⫽ normal; 1 ⫽ slight changes; 2 ⫽ moderate changes; 3 ⫽ severe changes.
variate comparison test to compare the subjective histologic rankings.
RESULTS All 15 novel device VUAs were successfully completed. In the control group, 14 (93%) of the 15 standard VUAs were completed. One pig in group 1 died of an accessrelated surgical complication. The mean operative time for the control and device groups was 87 and 68 minutes, respectively (P ⫽ 0.04). The anastomotic time for the control and device groups was 41 and 12 minutes, respectively (P ⬍0.01). For the postoperative cystogram, the median extravasation score for the standard and device animals was 0 and 0, respectively (P ⫽ 0.36). The mean volume of contrast returned from the control and device groups was 238.57 and 228.33 mL, respectively (P ⫽ 0.38). The median score for extravasation on the just before death cystogram for the control and device groups was 0 and 0, respectively (P ⫽ 0.65). One animal in the standard anastomosis group experienced severe leakage immediately postoperatively, and one animal in the novel device, 3-week group had severe extravasation on the second cystogram. The average fluid returned for the control and device groups was 229.00 and 240.71 mL, respectively (P ⫽ 0.53). The novel devices were successfully deployed in a single attempt in 13 (87%) of 15 cases. In 2 cases in group 5 (13%), the surgeon noted leakage at the anastomotic site. The novel anastomotic device catheter was easily disengaged and then redeployed for a watertight anastomosis. The total anastomosis time for these 2 cases was 15 and 20 minutes (including the redeployment). In both cases, the redeployment was easily performed in an atraumatic manner. 192
All pigs in groups 1, 3, and 6 lived to the desired follow-up time with the catheters in place. One pig in group 4 avulsed the novel VUA device. On the fourth postoperative day, the device was found intact; the tines were properly deployed, and the balloon was inflated. In group 2, one standard Foley catheter was removed by us on postoperative day 16 because of chronic rectal prolapse believed to be caused by the indwelling catheter. Of the 6 pigs in group 5, 4 (67%) avulsed the intact device (tines deployed and balloon intact). In this group, the mean indwell time of the device was 12 days. One device was removed by us on postoperative day 11 to prevent a traumatic event because the device was no longer secured by the anchoring sutures placed at the procedure, and avulsion was likely. Tables 2 through 4 present the mean histopathologic grades for the 1-week (groups 1 and 4), 3-week (groups 2 and 5), and 7-week (groups 3 and 6) survival groups, respectively. At 1 week, significant differences were found in the healing demonstrated by histopathologic evaluation (Fig. 2A,B). The novel device produced significantly less fibrosis in the muscle layer, fibrosis in the lamina propria, and tissue granulation. At 3 and 7 weeks, few differences in the histopathologic findings were noted (Fig. 2C-F). Mural inflammation and perianastomotic fat fibrosis were significantly less for the novel device at 3 weeks postoperatively.
COMMENT After laparoscopic radical prostatectomy, performing the sutured VUA is time-consuming and requires advanced laparoscopic skill and techniques.1–3 Minimizing or eliminating knot tying could help the laparoscopic surgeon UROLOGY 70 (1), 2007
Table 3. Comparison of median histopathologic changes in groups 2 and 5 Group Histologic Finding
Device
Control
Complete gross circumferential healing (%) Intact urothelium (%) Inflammatory reactions Mural inflammation Lamina propria inflammation Edema Perianastomotic fat inflammation Fibrotic reaction Muscle layer fibrosis Lamina propria fibrosis Granulation tissue Perianastomotic fat fibrosis Presence of myofibroblasts Other Focal foreign body reaction Squamous metaplasia Mucinous metaplasia Hemosiderin deposits Perianastomotic hemorrhage Necrotic debris
100
100
P Value NS
100
100
NS
1 1 1 1
2.5 0.5 1 0
0.05 0.56 0.74 0.07
2 1.5 1.5 1 0
3 0.5 2 0 0
0.12 0.56 0.56 0.02 NS
0.5 0 0 0 0 0
3 0 0 0 0 0
0.23 NS NS NS NS NS
Abbreviations as in Table 2.
Table 4. Comparison of median histopathologic changes in groups 3 and 6 Histologic Finding Complete gross circumferential healing (%) Intact urothelium (%) Inflammatory reactions Mural inflammation Lamina propria inflammation Edema Perianastomotic fat inflammation Fibrotic reaction Muscle layer fibrosis Lamina propria fibrosis Granulation tissue Perianastomotic fat fibrosis Presence of myofibroblasts Other Focal foreign body reaction Squamous metaplasia Mucinous metaplasia Hemosiderin deposits Perianastomotic hemorrhage Necrotic debris
Group Device Control 100
100
100
100
1 1 1 2
0 0 0 0
1.5 0 0 2 0
1 1 0 1 0
0 0 0 0 2 1
0 0 1 0 0 0
Abbreviations as in Table 2. All P values nonsignificant.
overcome the challenges of the VUA.3,4 Even with the elimination of knot tying, the most challenging component of the VUA is needle orientation and mastery of the forehand and backhand stitches required to successfully perform the VUA in both laparoscopic and roboticassisted prostatectomy.1,3,5 Even though the learning curve for robotic-assisted radical prostatectomy is shorter UROLOGY 70 (1), 2007
than that for laparoscopic radical prostatectomy, the robotic VUA can still be challenging and cumbersome.2,6,7 Despite our historical dependence on suturing techniques for all types of anastomosis, several theoretical disadvantages exist in the use of suture to approximate tissues. Passing needles and tying knots theoretically traumatize tissue. Absorbable suture, by its very definition, relies on an inflammatory process for the hydroxylation of the suture matrix.8 These factors might be responsible for some of the complications associated with a standard sutured VUA. The novel device was significantly better than a standard running VUA in surgical efficiency. The device reduced the anastomosis time by a mean of 29 minutes, with comparable results to a standard anastomosis with regard to extravasation. At 1 week postoperatively, the device VUAs demonstrated superior healing. Tissue fibrosis and tissue granulation were significantly less in the muscle layer and lamina propria in the 1-week device group than in the 1-week control group. At 3 and 7 weeks, minimal differences were found in the healing parameters. Because most surgeons remove the Foley catheter approximately 1 week after a standard anastomosis, the novel device could have clinical impact on long-term healing. The superior histopathologic findings with the VUA device might have resulted from the immediate watertight anastomosis. Additionally, trauma to the tissues by the tines appeared to be minimal, and no inflammatory reaction such as is seen with suture material was noted. Deployment of the urethral tines was monitored under laparoscopic vision. No tension or pulling was seen on the urethra by the urethral tines. Urethral mucosal or 193
Figure 2. Histopathologic findings of all groups. (A) Histologic findings of group 1 (1-week control group) showing severe fibrotic change (arrow). (B) In contrast, in 1-week VUA device group, minimal tissue granulation occurred. (C) Multinucleated giant cells (arrow) were common in 3-week control group, resulting from foreign body reaction commonly seen with suture material. (D) Relatively healthy tissue samples from 3-week VUA device group. As follow-up extended to 7 weeks both (E) control and (F) VUA device groups showed minimal histopathologic changes.
muscular damage from the tines was a minimal concern because of the pliable nature of the tines. Also, the tension applied to the anastomotic site through the device during urethral tine deployment was minimal. The tension applied by the surgeon was manually palpable by gently pulling on the device, and the surgeons involved believed that the anastomotic tension was easier to gauge 194
in this manner compared with the standard gauging of anastomotic tension by the tying down of suture material. As such, the potential for increased tension at the anastomotic site is unlikely and probably results in the same, if not less, tension at the anastomotic site with sutures. The superior healing parameters noted at 1 week postoperatively in this animal model suggest that the critical UROLOGY 70 (1), 2007
issues of anastomotic tension and mucosa-to-mucosa anastomosis are both well managed using this device. Because the novel device functions by tissue apposition, which is initiated by the Foley balloon, the bladder neck must be relatively small to allow the balloon to engage the cystotomy. In cases in which a larger bladder neck is created, some bladder neck suture reconstruction could be needed before device deployment to ensure proper mucosa-to-mucosa alignment. Another unique observation was the ease with which the device could be repositioned if the surgeon was unsatisfied with the anastomosis created by the device. The device was repositioned during two different procedures to achieve a superior anastomosis without significantly increasing the operating time. The avulsion of the catheters in the 3-week group, although concerning, is unlikely to be a clinical problem. Each anastomotic device was found intact, with the balloon inflated and the tines deployed. In our extensive experimental experience with the porcine model, it is very common for pigs to remove even the most wellanchored external drains and catheters. Although it is clearly clinically feasible to remove a Foley catheter with the balloon inflated, this is a rare occurrence in the awake and alert patient. We do not believe this device will be removed more commonly than a standard Foley catheter, because they both depend on the same balloonanchoring mechanism. We believe that the significant extravasation documented in 1 animal in group 5 (3-week group) at death was not a device failure, but rather an iatrogenic injury. In this pig, the device catheter was avulsed on postoperative day 18. At death, cystoscopy was performed to place a Foley catheter for the cystogram. This cystoscopy was performed in a traumatic manner, and the defect at the
UROLOGY 70 (1), 2007
VUA noted at death appeared grossly fresh. Overall, our initial animal experimental experience with this novel tissue anastomotic device has been favorable.
CONCLUSIONS In this animal model, the novel VUA device tested resulted in a very efficient and technically simple VUA. The device yielded efficacy equal to that of a standard running anastomosis for leakage. The histopathologic data demonstrated that the VUA device resulted in superior tissue healing at the anastomotic site. Clinical evaluation of the device is recommended. References 1. Nadu A, Olsson LE, and Abbou CC: Simple model for the training in the laparoscopic vesicourethral running anastomosis. J Endourol 17: 481– 484, 2003. 2. Ferguson GG, Ames CD, Weld KJ, et al: Prospective evaluation of learning curve for laparoscopic radical prostatectomy: identification of factors improving operative times. Urology 66: 840 – 844, 2005. 3. Velthoven RF, Ahlering TE, Peltier A, et al: Technique for laparoscopic running urethrovesical anastomosis: the single knot method. Urology 61: 699 –702, 2003. 4. Orvieto MA, Chijen GW, Laven B, et al: Eliminating knot tying during warm ischemia time for laparoscopic partial nephrectomy. J Urol 172: 2292–2295, 2004. 5. Ball AJ, Bordeau KP, Davis JW, et al: Modified running vesicourethral anastomosis after robotically assisted laparoscopic radical prostatectomy: use of solitary Lapra-Ty to secure posterior approximation. Urology 66: 16 –18, 2005. 6. Patel VR, Tully AS, Holmes R, et al: Robotic radical prostatectomy in the community setting—the learning curve and beyond: initial 200 cases. J Urol 174: 269 –272, 2005. 7. Herrell SD, and Smith JA: Robotic-assisted laparoscopic prostatectomy: what is the learning curve? Urology 66: 105–107, 2005. 8. Setzen G, and Williams EF: Tissue response to suture materials implanted subcutaneously in a rabbit model. Plast Reconstr Surg 100: 1788 –1795, 1997.
195
RESULTS
CONCLUSIONS
To evaluate a novel sutureless tissue apposing vesicourethral anastomosis (VUA) device in a porcine model and compare it with standard laparoscopically sutured VUA. Thirty domestic pigs were divided into six groups. In groups 1, 2, and 3, a standard laparoscopic sutured running VUA was performed. In groups 4, 5, and 6, a novel device VUA was performed. In all cases, cystography was completed immediately after completion of the anastomosis and when each pig was killed. At necropsy, the gross findings of the VUA were documented, and each anastomosis was excised en bloc for histopathologic evaluation of healing parameters. In the 30 pigs, 29 (97%) device or sutured VUAs were successfully performed laparoscopically without conversion to an open approach. The mean operative time for the standard and device groups was 87 and 68 minutes, respectively (P ⫽ 0.04). The anastomotic time for the standard and device groups was 41 and 12 minutes, respectively (P ⬍0.01). Histopathologic evaluation of the groups at 1 week of follow-up revealed significantly lower fibrosis scores for the novel anastomosis device VUA compared with the standard sutured VUA (median score 1 and 3, respectively; P ⫽ 0.04). The evaluation of groups 2 and 5 (3-week survival) and groups 3 and 6 (7-week survival) revealed no significant differences in any of the histopathologic parameters evaluated. The novel device requires little technical skill to deploy and is expeditious, requiring less time than a standard sutured anastomosis. At the 1 week follow-up point, histopathologic examination revealed that the novel device was superior regarding fibrotic reactions. UROLOGY 70: 190 –195, 2007. © 2007 Elsevier Inc.
T
he vesicourethral anastomosis (VUA) is one of the more technically demanding and time-consuming components of laparoscopic prostatectomy. Many of the complications associated with open or laparoscopic radical prostatectomy are related to the VUA. With the contemporary sutured VUA, the surgeon must overcome significant challenges to successfully perform a high-quality VUA, including tissue exposure in the confined area of the pelvis, fragile tissue, and physician fatigue at this point of the operation.1 Even after the learning curve has been overcome, the VUA can still pose a problem for the surgeon.2 Another problem with contemporary anastomosis is the application of absorbable suture. Absorbable suture, by definition, engenders a local inflammatory reaction, potentially compromising the healing process.
From the Department of Urology and Institute of Comparative Medicine, Columbia University Medical Center, New York, New York; Department of Urology, Wilford Hall Medical Center, Lackland Air Force Base, Texas Reprint requests: Jaime Landman, M.D., Department of Urology, Columbia University Medical Center, 161 Fort Washington Avenue, Room 1111, New York, NY 10032. E-mail: [email protected] Submitted: May 12, 2006; accepted (with revisions): February 27, 2007
190
© 2007 Elsevier Inc. All Rights Reserved
We evaluated a novel device (American Medical Systems, Minnetonka, Minn) that incorporates retractable tines that approximate the bladder and urethral tissues and simultaneously allow for bladder decompression with a drainage lumen and balloon. The 22F device with an inner lumen is equivalent to that of a 16F Foley catheter and consists of two opposing sets of circumferential nitinol tines, six bladder and six urethral tines, and a standard Foley balloon. The tines are 9.5 mm in length and have a diameter of 0.2 mm. The bladder tines are extended from the body of the catheter just below the Foley balloon, and the urethral tines are at a fixed axial distance from the bladder tines, ensuring urethral tine deployment within the membranous urethra. The VUA device was compared with standard sutured anastomosis for surgical efficiency and tissue healing.
MATERIAL AND METHODS The Institutional Animal Care and Use Committee approved the study. Thirty female domestic pigs were divided into six groups to compare a standard running anastomosis and a VUA 0090-4295/07/$32.00 doi:10.1016/j.urology.2007.02.061
Table 1. Animal groups
Group
Anastomosis Type
Animals (n)
Length of Foley Decompression (wk)
1 2 3 4 5 6
Running sutured Running sutured Running sutured VUA device VUA device VUA device
6 6 3 6 6 3
1 3 1 1 3 1
Survival Time (wk) 1 3 7 1 3 7
VUA ⫽ vesicourethral anastomosis.
created with the novel device. The study design is summarized in Table 1. The animals were initially anesthetized with ketamine 5 to 10 mg/kg and then intubated. The anesthetic was maintained with 2% to 5% Isofuorane. In a sterile manner, a Foley catheter was placed in the bladder, and the pig was prepared and draped for transperitoneal access, which was gained using three trocars: one at the umbilicus and two inferior to the umbilicus and lateral to the rectus abdominus muscles. The bladder was mobilized, and the bladder neck was dissected. The ureters were identified, and the urethra was incised caudal to the ureteral orifices. In all cases, the bladder neck was incised anteriorly, and a “tennis racket” closure was performed posteriorly with interrupted 2-0 Vicryl suture on an SH needle (Ethicon, Somerville, NJ) to reconstruct the bladder neck and prevent ureteral orifice occlusion. In the control groups (groups 1 to 3), a standard Van Velthoven running anastomosis was then performed using a 2-0 Vicryl suture on an SH needle.3 In groups 4 to 6, the VUA was completed using the novel device. The device was placed into the bladder as one would place a regular urethral catheter, the 15-mL balloon was inflated, and the bladder tines were deployed using the housing control unit. Monitored under laparoscopic vision, gentle caudal traction on the device approximated the bladder neck to the urethral stump. The urethral tines were then extended from the body of the device, and the anastomosis was completed (Fig. 1). No sutures were placed. Device removal consisted of first complete urethral and then complete bladder tine retraction into the device. The device was then removed from the animal in a fashion similar to standard Foley catheter removal. The operative and VUA times were recorded. After each anastomosis, all animals underwent a visual inspection of the anastomosis using a 250-mL cystogram (Conray, Mallinckrodt Institute, St. Louis, Mo) performed by gravity infusion at a pressure of 40 cm H2O. Fluoroscopy was used to grade the extravasation of the anastomosis (0, none; 1, minimal; 2, moderate; and 3, severe). After cystography, the bladder was allowed to drain to completion by gravity drainage, and the fluid returned was collected and measured. All catheters were secured to the pig with 2-0 nonabsorbable Prolene suture (Ethicon) every 2 cm. Groups 1 and 4, 2 and 5, and 3 and 6 were killed at 1, 3, and 7 weeks postoperatively, respectively. Beforehand, cystography was again performed in the same manner. After death, the anastomotic tissue was harvested en bloc and placed in 5% formalin for fixation and later histopathologic examination. Gross observation of circumferential healing was documented at the time of death. Each anastomosis was divided into three sections, placed in tissue cassettes for permanent sectioning, and stained with hematoxylin-eosin in a standard manner. UROLOGY 70 (1), 2007
Figure 1. Novel anastomotic device (A,E) inserted through urethra and placed inside bladder. (B) Foley balloon inflated with 15 mL sterile water. (C) Bladder tines deployed using housing control unit. Slight caudal retraction used to approximate bladder neck to urethral stump. (D) Urethral tines deployed using housing control unit, and anastomosis is completed. Three slides from each cassette were evaluated and graded by a pathologist (A.H.) unaware of the treatment groups on a grading scale of 0 to 3 (0, none; 1, minimal; 2, moderate; and 3, severe). The parameters evaluated included the inflammatory reactions (mural, lamina propria, edema, and perianastomotic fat inflammation), fibrotic reactions (muscle, lamina propria, granulation, perianastomotic fat, and myofibroblasts), and other categories (foreign body reaction, squamous metaplasia, mucinous metaplasia, hemosiderin deposits, perianastomotic hemorrhage, necrotic debris, and intact urothelium). A two-sample mean comparison test was used to evaluate the operative times, fluid returned on the postoperative and just before death cystograms, and animal weight. We used a multi191
Table 2. Comparison of median histopathologic changes in groups 1 and 4 Group Histologic Finding
Device
Complete gross circumferential healing (%) Intact urothelium (%) Inflammatory reactions Mural inflammation Lamina propria inflammation Edema Perianastomotic fat inflammation Fibrotic reaction Muscle layer fibrosis Lamina propria fibrosis Granulation tissue Perianastomotic fat fibrosis Presence of myofibroblasts Other Focal foreign body reaction Squamous metaplasia Mucinous metaplasia Hemosiderin deposits Perianastomotic hemorrhage Necrotic debris
83
Control 20
P Value 0.08
66
0
0.18
2.5 2 1.5 2
3 2.5 2 2
0.56 0.43 0.23 0.92
1 0 0.5 0.5 0
3 3 2 1.5 0
0.04 0.03 0.05 0.08 NS
0 0 0 0.25 0.5 1
1 0 0 0 1 0.5
0.23 NS NS NS NS NS
NS ⫽ not significantly different between groups; 0 ⫽ normal; 1 ⫽ slight changes; 2 ⫽ moderate changes; 3 ⫽ severe changes.
variate comparison test to compare the subjective histologic rankings.
RESULTS All 15 novel device VUAs were successfully completed. In the control group, 14 (93%) of the 15 standard VUAs were completed. One pig in group 1 died of an accessrelated surgical complication. The mean operative time for the control and device groups was 87 and 68 minutes, respectively (P ⫽ 0.04). The anastomotic time for the control and device groups was 41 and 12 minutes, respectively (P ⬍0.01). For the postoperative cystogram, the median extravasation score for the standard and device animals was 0 and 0, respectively (P ⫽ 0.36). The mean volume of contrast returned from the control and device groups was 238.57 and 228.33 mL, respectively (P ⫽ 0.38). The median score for extravasation on the just before death cystogram for the control and device groups was 0 and 0, respectively (P ⫽ 0.65). One animal in the standard anastomosis group experienced severe leakage immediately postoperatively, and one animal in the novel device, 3-week group had severe extravasation on the second cystogram. The average fluid returned for the control and device groups was 229.00 and 240.71 mL, respectively (P ⫽ 0.53). The novel devices were successfully deployed in a single attempt in 13 (87%) of 15 cases. In 2 cases in group 5 (13%), the surgeon noted leakage at the anastomotic site. The novel anastomotic device catheter was easily disengaged and then redeployed for a watertight anastomosis. The total anastomosis time for these 2 cases was 15 and 20 minutes (including the redeployment). In both cases, the redeployment was easily performed in an atraumatic manner. 192
All pigs in groups 1, 3, and 6 lived to the desired follow-up time with the catheters in place. One pig in group 4 avulsed the novel VUA device. On the fourth postoperative day, the device was found intact; the tines were properly deployed, and the balloon was inflated. In group 2, one standard Foley catheter was removed by us on postoperative day 16 because of chronic rectal prolapse believed to be caused by the indwelling catheter. Of the 6 pigs in group 5, 4 (67%) avulsed the intact device (tines deployed and balloon intact). In this group, the mean indwell time of the device was 12 days. One device was removed by us on postoperative day 11 to prevent a traumatic event because the device was no longer secured by the anchoring sutures placed at the procedure, and avulsion was likely. Tables 2 through 4 present the mean histopathologic grades for the 1-week (groups 1 and 4), 3-week (groups 2 and 5), and 7-week (groups 3 and 6) survival groups, respectively. At 1 week, significant differences were found in the healing demonstrated by histopathologic evaluation (Fig. 2A,B). The novel device produced significantly less fibrosis in the muscle layer, fibrosis in the lamina propria, and tissue granulation. At 3 and 7 weeks, few differences in the histopathologic findings were noted (Fig. 2C-F). Mural inflammation and perianastomotic fat fibrosis were significantly less for the novel device at 3 weeks postoperatively.
COMMENT After laparoscopic radical prostatectomy, performing the sutured VUA is time-consuming and requires advanced laparoscopic skill and techniques.1–3 Minimizing or eliminating knot tying could help the laparoscopic surgeon UROLOGY 70 (1), 2007
Table 3. Comparison of median histopathologic changes in groups 2 and 5 Group Histologic Finding
Device
Control
Complete gross circumferential healing (%) Intact urothelium (%) Inflammatory reactions Mural inflammation Lamina propria inflammation Edema Perianastomotic fat inflammation Fibrotic reaction Muscle layer fibrosis Lamina propria fibrosis Granulation tissue Perianastomotic fat fibrosis Presence of myofibroblasts Other Focal foreign body reaction Squamous metaplasia Mucinous metaplasia Hemosiderin deposits Perianastomotic hemorrhage Necrotic debris
100
100
P Value NS
100
100
NS
1 1 1 1
2.5 0.5 1 0
0.05 0.56 0.74 0.07
2 1.5 1.5 1 0
3 0.5 2 0 0
0.12 0.56 0.56 0.02 NS
0.5 0 0 0 0 0
3 0 0 0 0 0
0.23 NS NS NS NS NS
Abbreviations as in Table 2.
Table 4. Comparison of median histopathologic changes in groups 3 and 6 Histologic Finding Complete gross circumferential healing (%) Intact urothelium (%) Inflammatory reactions Mural inflammation Lamina propria inflammation Edema Perianastomotic fat inflammation Fibrotic reaction Muscle layer fibrosis Lamina propria fibrosis Granulation tissue Perianastomotic fat fibrosis Presence of myofibroblasts Other Focal foreign body reaction Squamous metaplasia Mucinous metaplasia Hemosiderin deposits Perianastomotic hemorrhage Necrotic debris
Group Device Control 100
100
100
100
1 1 1 2
0 0 0 0
1.5 0 0 2 0
1 1 0 1 0
0 0 0 0 2 1
0 0 1 0 0 0
Abbreviations as in Table 2. All P values nonsignificant.
overcome the challenges of the VUA.3,4 Even with the elimination of knot tying, the most challenging component of the VUA is needle orientation and mastery of the forehand and backhand stitches required to successfully perform the VUA in both laparoscopic and roboticassisted prostatectomy.1,3,5 Even though the learning curve for robotic-assisted radical prostatectomy is shorter UROLOGY 70 (1), 2007
than that for laparoscopic radical prostatectomy, the robotic VUA can still be challenging and cumbersome.2,6,7 Despite our historical dependence on suturing techniques for all types of anastomosis, several theoretical disadvantages exist in the use of suture to approximate tissues. Passing needles and tying knots theoretically traumatize tissue. Absorbable suture, by its very definition, relies on an inflammatory process for the hydroxylation of the suture matrix.8 These factors might be responsible for some of the complications associated with a standard sutured VUA. The novel device was significantly better than a standard running VUA in surgical efficiency. The device reduced the anastomosis time by a mean of 29 minutes, with comparable results to a standard anastomosis with regard to extravasation. At 1 week postoperatively, the device VUAs demonstrated superior healing. Tissue fibrosis and tissue granulation were significantly less in the muscle layer and lamina propria in the 1-week device group than in the 1-week control group. At 3 and 7 weeks, minimal differences were found in the healing parameters. Because most surgeons remove the Foley catheter approximately 1 week after a standard anastomosis, the novel device could have clinical impact on long-term healing. The superior histopathologic findings with the VUA device might have resulted from the immediate watertight anastomosis. Additionally, trauma to the tissues by the tines appeared to be minimal, and no inflammatory reaction such as is seen with suture material was noted. Deployment of the urethral tines was monitored under laparoscopic vision. No tension or pulling was seen on the urethra by the urethral tines. Urethral mucosal or 193
Figure 2. Histopathologic findings of all groups. (A) Histologic findings of group 1 (1-week control group) showing severe fibrotic change (arrow). (B) In contrast, in 1-week VUA device group, minimal tissue granulation occurred. (C) Multinucleated giant cells (arrow) were common in 3-week control group, resulting from foreign body reaction commonly seen with suture material. (D) Relatively healthy tissue samples from 3-week VUA device group. As follow-up extended to 7 weeks both (E) control and (F) VUA device groups showed minimal histopathologic changes.
muscular damage from the tines was a minimal concern because of the pliable nature of the tines. Also, the tension applied to the anastomotic site through the device during urethral tine deployment was minimal. The tension applied by the surgeon was manually palpable by gently pulling on the device, and the surgeons involved believed that the anastomotic tension was easier to gauge 194
in this manner compared with the standard gauging of anastomotic tension by the tying down of suture material. As such, the potential for increased tension at the anastomotic site is unlikely and probably results in the same, if not less, tension at the anastomotic site with sutures. The superior healing parameters noted at 1 week postoperatively in this animal model suggest that the critical UROLOGY 70 (1), 2007
issues of anastomotic tension and mucosa-to-mucosa anastomosis are both well managed using this device. Because the novel device functions by tissue apposition, which is initiated by the Foley balloon, the bladder neck must be relatively small to allow the balloon to engage the cystotomy. In cases in which a larger bladder neck is created, some bladder neck suture reconstruction could be needed before device deployment to ensure proper mucosa-to-mucosa alignment. Another unique observation was the ease with which the device could be repositioned if the surgeon was unsatisfied with the anastomosis created by the device. The device was repositioned during two different procedures to achieve a superior anastomosis without significantly increasing the operating time. The avulsion of the catheters in the 3-week group, although concerning, is unlikely to be a clinical problem. Each anastomotic device was found intact, with the balloon inflated and the tines deployed. In our extensive experimental experience with the porcine model, it is very common for pigs to remove even the most wellanchored external drains and catheters. Although it is clearly clinically feasible to remove a Foley catheter with the balloon inflated, this is a rare occurrence in the awake and alert patient. We do not believe this device will be removed more commonly than a standard Foley catheter, because they both depend on the same balloonanchoring mechanism. We believe that the significant extravasation documented in 1 animal in group 5 (3-week group) at death was not a device failure, but rather an iatrogenic injury. In this pig, the device catheter was avulsed on postoperative day 18. At death, cystoscopy was performed to place a Foley catheter for the cystogram. This cystoscopy was performed in a traumatic manner, and the defect at the
UROLOGY 70 (1), 2007
VUA noted at death appeared grossly fresh. Overall, our initial animal experimental experience with this novel tissue anastomotic device has been favorable.
CONCLUSIONS In this animal model, the novel VUA device tested resulted in a very efficient and technically simple VUA. The device yielded efficacy equal to that of a standard running anastomosis for leakage. The histopathologic data demonstrated that the VUA device resulted in superior tissue healing at the anastomotic site. Clinical evaluation of the device is recommended. References 1. Nadu A, Olsson LE, and Abbou CC: Simple model for the training in the laparoscopic vesicourethral running anastomosis. J Endourol 17: 481– 484, 2003. 2. Ferguson GG, Ames CD, Weld KJ, et al: Prospective evaluation of learning curve for laparoscopic radical prostatectomy: identification of factors improving operative times. Urology 66: 840 – 844, 2005. 3. Velthoven RF, Ahlering TE, Peltier A, et al: Technique for laparoscopic running urethrovesical anastomosis: the single knot method. Urology 61: 699 –702, 2003. 4. Orvieto MA, Chijen GW, Laven B, et al: Eliminating knot tying during warm ischemia time for laparoscopic partial nephrectomy. J Urol 172: 2292–2295, 2004. 5. Ball AJ, Bordeau KP, Davis JW, et al: Modified running vesicourethral anastomosis after robotically assisted laparoscopic radical prostatectomy: use of solitary Lapra-Ty to secure posterior approximation. Urology 66: 16 –18, 2005. 6. Patel VR, Tully AS, Holmes R, et al: Robotic radical prostatectomy in the community setting—the learning curve and beyond: initial 200 cases. J Urol 174: 269 –272, 2005. 7. Herrell SD, and Smith JA: Robotic-assisted laparoscopic prostatectomy: what is the learning curve? Urology 66: 105–107, 2005. 8. Setzen G, and Williams EF: Tissue response to suture materials implanted subcutaneously in a rabbit model. Plast Reconstr Surg 100: 1788 –1795, 1997.
195