Gross and histologic characteristics of laparoscopic injuries with four different energy sources

FERTILITY AND STERILITY VOL. 75, NO. 4, APRIL 2001 Copyright 2001 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Gross and histologic characteristics of laparoscopic injuries with four different energy sources Paul K. Tulikangas, M.D., a Tamara Smith, M.D., b Tommaso Falcone, M.D., a Navdeep Boparai, M.S., c and Mark D. Walters, M.D. a The Cleveland Clinic Foundation, Cleveland, Ohio Received August 9, 2000; revised and accepted October 2, 2000. Supported by the Cleveland Clinic Foundation s Minimally Invasive Surgery Center and the Research Programs Council, grant 6291. Presented at the Society for Gynecologic Surgeons, New Orleans, Louisiana, March 29, 2000. Reprint requests: Paul K. Tulikangas, M.D., Minimally Invasive Surgery Center, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44195 (FAX: 216-444-8551; E-mail: tulikap@ccf.org). a Minimally Invasive Surgery Center. b Department of Anatomic Pathology. c Department of Biostatistics. 0015-0282/01/$20.00 PII S0015-0282(00)01785-4 Objective: To compare the gross and histologic effects of bipolar and monopolar cautery, ultrasonic scalpel, and CO 2 laser on porcine ureter, bladder, and rectum. Design: Experimental prospective study. Setting: Cleveland Clinic Foundation Animal Research Laboratory, Cleveland, Ohio. Animal(s): Nonpregnant adult female pigs. Intervention(s): The rectum, bladder, and ureters of 12 female pigs were injured with four different laparoscopic energy sources. Main Outcome Measure(s): Gross measurements of injured tissue and histologic analysis of the depth of the tissue injury. Result(s): Gross assessment results were that monopolar injuries of the bowel and bladder were significantly longer than ultrasonic injuries (P 0.01). Injuries were generally manifest as coagulative denaturation of collagen bundles. This resulted in an eosinophilic homogenization of tissue. Nuclei were retained in the injured tissue, although in most cases they had a pyknotic, streamed appearance. The CO 2 laser caused no deep-tissue injury. Conclusion(s): Laparoscopic energy sources injure tissue differently. Monopolar cautery appears to have the most lateral spread of thermal energy. The CO 2 laser appears to cause the least deep-tissue injury. (Fertil Steril 2001;75:806 10. 2001 by American Society for Reproductive Medicine.) Key Words: Intraoperative complications, ureter, bowel, bladder, laparoscopy The number of gynecologic laparoscopic surgeries performed each year is increasing (1). With increasingly complex surgeries being performed, there are more complications (2). Many of these complications involve inadvertent contact of energy sources with internal organs in the pelvis. There are a variety of energy sources available for use in laparoscopic surgery. Injuries to internal organs are different with different energy sources (3 6). By directly comparing different energy sources effects on tissue, we could determine the best energy source for use in particular laparoscopic surgeries. We performed a pilot study investigating the gross and histologic characteristics of different energy sources on the bladder, bowel, and ureter in a porcine model. MATERIALS AND METHODS This study was approved by the Cleveland Clinic Foundation Animal Review Committee and the Research Institute s Research Programs Council. Twelve nonpregnant adult female pigs, a crossbreed of Landrace and Yorkshire and weighing 40 50 kg each, were used in this study. Animals were placed in the dorsal recumbent position. After induction of general anesthesia, the animals were prepared and draped. A 10-mm laparoscopic trocar was inserted into the peritoneal cavity through a 10-mm incision in the midline on the dorsal surface. A laparoscope with attached camera was cannulated through this port, and pneumoperitoneum of 12 mmhg was obtained using 806

CO 2 gas. Five-millimeter trocars were advanced into the peritoneal cavity on the right and left sides, caudal to the camera port. When necessary, a third 5-mm trocar was placed on the animal s right side at the level of the camera port. The ureters were identified, and their course was traced from the pelvic brim caudad. The ureter was dissected off the pelvic sidewall, and the infundibulopelvic blood vessels were isolated from the ureter at the level of the pelvic brim. The infundibulopelvic blood vessels were coagulated bilaterally with each energy source. The energy source was applied until the pedicle was hemostatic, and it was then transected to confirm hemostasis. The time taken to coagulate each infundibulopelvic blood vessel was measured and the average calculated for each animal. Each ureter was then injured. The energy source was applied to the ureter 4 cm distal to where the ureter crossed the iliac vessels. The energy source was applied for the average time it took to coagulate the infundibulopelvic blood vessels in that animal. Both ureters were injured identically. A segment of ureter 4 cm proximal to the iliac bifurcation was used as the control (uninjured) site. Next, the rectum and bladder were injured. Like in the ureteral injuries, the energy source was applied for the average time it took to coagulate the infundibulopelvic blood vessels in that animal. The energy source was applied to the anterior surface of the colon at the level of the sacral promontory. A segment of bowel 8 cm proximal to the site of injury was used as the control site. The bladder was injured on the posterior surface above the trigone. A section of bladder from the anterior surface was used as the control site. Three animals were used for each of the four energy sources: [1] bipolar Kleppinger forceps at 40 W; [2] monopolar graspers at 40 W; [3] ultrasonic scalpel scissors at a variable setting of 3.5; and [4] the CO 2 laser at superpulse one. The bipolar Kleppinger forceps have a cautery surface that is 13 4 mm. The monopolar graspers have a cutting surface that is 13 2 mm. The harmonic scalpel cuttings instrument used has a blade surface of 12 1 mm and a backstop that is 10 3 mm. The CO 2 laser was set at a width of 400 nm and was focused over a 1-cm portion of tissue. A midline abdominal incision was then made, and measurements of the maximum length and width of the grossly evident injuries were recorded in millimeters with a calibrated measuring tape. All injured tissues were resected, with a grossly clear margin of normal tissue around the organ injury. Normal control tissue was excised at sites described above, and all specimens were placed in 10% neutral buffered formalin. For the bladder and bowel specimens, serial sections of the entire grossly injured area were submitted for histologic examination. Ureter specimens were cross-sectioned and FIGURE 1 Full-thickness rectal injury with monopolar cautery (original magnification, 5). A indicates epithelial layer; B, submucosal layer; C, muscularis. submitted in their entirety for histologic review. The hematoxylin and eosin stained histologic sections were examined for evidence of thermal injury. Normal, grossly uninjured bladder, bowel, and ureter specimens were also submitted for histologic review. Tissue was considered thermally injured if there was clear-cut coagulative denaturation of collagen bundles, characterized by eosinophilic homogenization and coagulation of collagen. The rectal epithelium and bladder urothelium were considered thermally injured if there was any identifiable streaming of nuclei. For the bladder and bowel specimens, the amount of thermal injury was calculated by measuring the maximum longitudinal measurement of identifiable injury on the histologic slide. For the ureter specimens, thermal injury was evaluated on the serosal and mucosal surfaces as well. Thermal injury was FERTILITY & STERILITY 807

FIGURE 2 Rectal epithelium injury with monopolar cautery (original magnification, 20). documented as either being present or absent, and the total length of the injury was inferred by multiplying the number of cross-sections involved by 3 mm (the estimate of the average thickness of each piece of tissue submitted). Comparison of the extent of injury with each energy source for each organ was done by the Kruskal-Wallis test. If significant differences were found in these, then multiple comparisons were done using the Dunn s test. The Wilcoxon signed rank test was used to detect differences in assessment methods. To adjust for multiple comparisons, P values 0.017 were considered significant. RESULTS Gross Assessment of Injuries The average length of injury, measured grossly, for the bipolar cautery was 0.4 0.2 cm on the ureter, 1.3 0.2 cm for the bladder, and 1.3 0.2 cm for the rectum. The average length of injury for the monopolar cautery injuries was 0.6 0.2 cm for the ureter, 2.1 0.4 cm for the bladder, and 1.8 0.3 cm for the rectum. For the ultrasonic scalpel, the average length of injury was 0.5 0.2 cm for the ureter, 0.9 0.2 cm for the bladder, and 0.7 0.2 cm for the rectum. Monopolar energy appears to cause a larger collateral injury than the ultrasonic scalpel. Gross assessment of monopolar bladder and bowel injuries indicated that they were significantly longer than injuries with the ultrasonic scalpel (P 0.01 and 0.04, respectively). For the other bladder, bowel, and ureteral injuries, the monopolar injuries were on average longer than the bipolar and ultrasonic injuries; however, these differences were not statistically significant. It was not possible to reliably determine the size of the laser injuries because they varied with the movement of the laser beam. Histologic specimens shrank when stored in formalin, so the histologic size of the injuries could not be compared with the gross measurements. General Histologic Features of Injuries The histologic features of thermal injury were for the most part similar, regardless of tissue type or source of energy used. Injuries were generally manifest as a coagulative denaturation of collagen bundles. This resulted in an eosinophilic homogenization of tissue. Nuclei were generally retained in the injured tissue, although in most cases they had a pyknotic, streamed appearance. Denatured collagen bundles were identifiable primarily in the serosal, intermuscular, and lamina propria locations (Figure 1). Injury effects were generally difficult to appreciate in the muscularis propria layers. Epithelial or urothelial injury was manifested as clear-cut streaming of glandular or urothelial nuclei. The degree of streaming varied, from polarization with relative maintenance of nuclear contours to condensation and smudging of nuclear features, such that individual cell outlines were blurred (Figure 2). Some morphologic differences were noted when compar- 808 Tulikangas et al. Characteristics of laparoscopic injuries Vol. 75, No. 4, April 2001

ing energy sources. For bladder and bowel specimens injured with the harmonic scalpel, in general, a distinct divet-shaped defect was identifiable on the serosal surface, often penetrating through the muscularis propria, connecting the serosal surface with the lamina propria. When this occurred, the lamina propria was involved with the injury (Figure 3). A striking tissue basophilia was identified in the area of the serosal defect. Specimens injured by the CO 2 laser also demonstrated a striking tissue basophilia; however, a serosal divot-shaped defect was not identified. Involvement of the lamina propria was not identified, except in the one case in which the serosal injury was extensive enough to compromise the muscularis propria layer, allowing direct communication between the serosa and the lamina propria. Serosal defects were not identified in any of the ureter specimens, most likely because of the manner in which the specimens were dissected. Striking tissue basophilia, however, was identified in ureter specimens injured with the harmonic scalpel and the CO 2 laser. Histologic Assessment of Injuries Histologic assessment of the injuries with bipolar cautery found an average ureter injury length of 1.5 1.4 cm, an average bladder injury length of 0.9 0.1 cm, and an average rectum injury length of 1.1 0.3 cm. Histologic assessment of the monopolar injuries found an average ureter injury length of 2.3 0.4 cm, an average bladder injury length of 1.0 0.6 cm, and an average rectum injury length of 1.1 0.4 cm. In one of the bladder specimens, the length of the serosal injury was at least 1.0 cm; however, the injured tissue extended to one edge of the histologic section. In many specimens, there was evidence of epithelial or urothelial injury. When bipolar cautery caused the injury, 5/6 specimens showed urothelial injury in the ureter; 2/3, in the bladder; and 3/3, in the rectum. When monopolar cautery was used, 5/6 specimens showed evidence of urothelial injury in the ureter; 1/3, in the bladder; and 3/3, in the rectum. When the ultrasonic scalpel was used, 4/6 showed evidence of urothelial injury in the ureter; 0/3, in the bladder; and 1/3, in the rectum. In all 12 of the specimens injured with the CO 2 laser, none showed urothelial or epithelial injury. In one segment of ureter that was excised as a control for a monopolar injury, there was evidence of cautery injury. The segment of control ureter was taken approximately 8 cm proximal to the site of injury. DISCUSSION FIGURE 3 Rectal injury with ultrasonic scalpel (original magnification, 5). Arrow points to divot on serosal surface. We have completed a pilot study to evaluate the effect of different laparoscopic energy source injuries on pelvic organs. As in any pilot study, the major limitation is the number of subjects. Although some of our results are statistically significant, they should be interpreted cautiously until follow-up studies are completed. Monopolar injuries appeared to have more collateral spread than the bipolar and ultrasonic scalpel injuries. Baggish and Tucker demonstrated more collateral spread with monopolar electrocautery scissors than bipolar electrocautery scissors in rabbits uterine horns (6). Our study supports these findings. In one segment of ureter, there appeared to be a cautery effect as far as 8 cm from the site of injury. This finding is plausible; cautery effect has been demonstrated after an electrical burn in bowel, as far as 5 cm away from the site of injury (7). Saye et al. demonstrated extensive collateral spread of monopolar electrothermal energy on a loop of porcine small bowel (8). They hypothesized that distal burn injuries would occur in ductlike structures because of the increased electrical resistance of this tissue away from the operative site. Because of the risk of distal burn injuries, they FERTILITY & STERILITY 809

concluded, the aware surgeon should avoid application of monopolar electricity to appendix and duct-like structures. Our findings support this conclusion. The size of the organ injuries measured grossly is very different than the size of the organ injuries measured histologically. One reason for this is that the specimens shrank when preserved in formalin before processing. Another reason for this is the definition we used for tissue injury. When assessing tissue grossly, we examined the serosal surface area that appeared injured. In the fixed slides, the extent of serosal injury could have been altered by the handling of the tissue. In these specimens, the extent of injury was determined by histologic changes such as coagulative denaturation of collagen bundles and changes in nuclear morphology. With regard to the depth of injury, the CO 2 laser showed the least tissue penetration. If a CO 2 laser injury to the serosa of the bowel or bladder is noted, it might be reasonable to follow the injury clinically and not resect the tissue. In the ureter, placement of a ureteral stent and close follow-up might be a reasonable management choice. Chronic studies following these injuries with a larger number of subjects will be necessary before any definitive conclusions about appropriate management are made. This is the first study to directly compare the histologic findings of organ injury with four different energy sources. Coagulative denaturation of collagen bundles and an eosinophilic homogenization of tissue was present with all energy sources. The CO 2 laser and ultrasonic scalpel produced a tissue basophilia in the injured serosal tissue, suggesting a different mechanism of injury with these two energy sources. It is not clear whether there is any clinical significance to this finding. The bladder and bowel injured with the ultrasonic scalpel demonstrated a divot in the tissue. This is likely from the active cutting blade compressing the tissue against the static backstop. When this divet penetrated into the lamina propria, there was evidence of tissue injury at that level. Injury at the level of the lamina propria is an important finding because the blood supply for the epithelium is in this area. Deeptissue injury with this instrument may be dependent on the surgeon s amount of force when squeezing tissue. This pilot study provides us with some interesting data with regard to laparoscopic energy source injuries. As in all pilot studies, there are significant limitations with regard to statistical power. If one wished to determine whether there was a 3-mm difference between monopolar and bipolar injuries to the bowel, one would need 16 subjects in each group (80% power, alpha level of 0.05). It seems feasible to conduct such studies in the future. To determine the clinical significance of these injuries, chronic studies must be performed. References 1. Levy BS, Hulka JF, Peterson HB, Phillips JM. Operative laparoscopy: American Association of Gynecologic Laparoscopists, 1993 Membership Survey. J Am Assoc Gynecol Laparosc 1994;1:301 5. 2. Harkki-Siren P, Kurki T. A nationwide analysis of laparoscopic complications. Obstet Gynecol 1997;89:108 12. 3. Levy BS, Soderstrom RM, Dail DH. Bowel injuries during laparoscopy, gross anatomy and histology. J Reprod Med 1985;30:168 72. 4. Gale P, Adeyemi B, Ferrer K, Ong A, Brill AI, Scoccia B. Histologic characteristics of laparoscopic argon beam coagulation. J Am Assoc Gynecol Laparosc 1998;5:19 22. 5. Amaral JF. The experimental development of an ultrasonically activated scalpel for laparoscopic use. Surg Laparosc Endosc 1994;4:92 9. 6. Baggish MS, Tucker RD. Tissue actions of bipolar scissors compared with monopolar devices. Fertil Steril 1995;63:422 6. 7. Soderstrom RM, Levinson C, Levy B. Complications of operative laparoscopy. In: Soderstrom RM, ed. Operative laparoscopy. New York: Raven Press, 1993:187 97. 8. Saye WB, Miller W, Hertzmann P. Electrosurgery thermal injury: myth of misconception. Surg Laparosc Endosc 1991;1:223 8. 810 Tulikangas et al. Characteristics of laparoscopic injuries Vol. 75, No. 4, April 2001