Secondary Logo

Journal Logo

Original Research

The Impact of Trocar-Cannula Design and Simulated Operative Manipulation on Incisional Characteristics: A Randomized Trial

Munro, Malcolm G. MD; Tarnay, Christopher M. MD

Author Information
doi: 10.1097/01.AOG.0000120144.85187.61
  • Free

The use of laparoscopically guided gynecologic surgical techniques has allowed many complex procedures, previously performed by laparotomy, to be accomplished in a minimally invasive fashion. The initial and critically important surgical step required for such procedures is the positioning of peritoneal access devices or trocar-cannula systems in the anterior abdominal wall. There are inherent risks associated with placement of laparoscopic trocar-cannula system that include bleeding from the abdominal wall, injury to the great vessels of the pelvis, and damage to intraperitoneal viscera including the bowel and urinary tract.1–4 More recently, wound dehiscence and hernia have been associated with laparoscopic port insertion.5 Although originally considered to be rare occurrences, these complications are more frequent than previously thought, occurring in 0.2–3.0% of cases.6 The increasing risk of incisional dehiscence and hernia is likely related both to the larger diameter of the cannulas used in contemporary operative laparoscopy and to the increasing number of ports used in a given case.5 Although reported with the use of 5-mm trocar-cannula systems,7 dehiscence with such small-caliber devices is rare and is far more commonly reported in association with cannulas 10 mm or more in diameter.6 Consequently, fascial closure has generally been recommended for incisions created with access systems with a 10-mm or greater internal diameter. Nevertheless, dehiscence and hernia have been reported rather frequently despite attempted fascial incisional closure, an observation that may reflect the paradoxical difficulty of consistently attaining adequate approximation of the wound6 and which further justifies continuing attempts to reduce the size of wound defects.

In previous studies, we demonstrated substantial differences in both insertion force and defined wound parameters based on the design of the trocar-cannula system.8–11 For those systems that use a blunt dilating obturator, we and others have shown that facial wounds are significantly smaller in area than those wounds resultant from a cutting or bladed tip.8,12 The potential clinical significance of such observations is reflected in a nonrandomized comparative study, where incisional hernia risk was found to be more than 10 times greater when a disposable pyramidal device was used than when access was gained with reusable conical trocar-cannula system (1.83% versus 0.17%).13

There may be inherent host wound qualities in addition to trocar-cannula system design that affect the risk of hernia and dehiscence, and there may be other factors that have an impact on the characteristics of the wounds made by laparoscopic access systems. For example, it is possible that damage to the muscular layers of the anterior abdominal wall may contribute to the risk of wound dehiscence and hernia. Consequently, in a previous study, we devised, analyzed, and reported a scoring system that was designed to reflect the degree of damage to the muscular layers underlying the fascia.10 Another factor that we have postulated may increase wound metrics is the manipulation of the ports themselves, caused by motion of the laparoscopic instruments during a surgical case. We devised a group of standardized cannula manipulations and applied them to a study comparing pyramidal- and threaded-access cannulas that did not show significant changes in the wound metrics. However, in the previous study, a surgical incision was required for the threaded-access system, a feature that could have impacted the response of the incision to simulated surgical manipulations.11

The next steps in our research plan were to determine if there exists a trocar-cannula system design of greater than 10-mm internal diameter that may not require fascial closure to prevent dehiscence and herniation, or at least one that would be associated with minimal risk of such complications. Given the relative paucity of reported wound dehiscence and hernia associated with pyramidal trocar-cannula system of less than 10-mm internal diameter, we performed some preliminary work that suggested that the wound metrics associated with an 8-mm pyramidal devices were similar to those created by a 12-mm blunt conical system. Consequently, we hypothesized that 12-mm blunt conical peritoneal access devices would be comparable to 8-mm pyramidal devices with respect to the described wound metrics; incisional area, wound length, and muscle injury. If this hypothesis could be substantiated, 12-mm blunt conical devices may not require routine fascial closure. We also hypothesized that application of our standardized operative manipulations of the cannulas would result in differential expansion of the wounds made by the pyramidal devices.


The 2 devices selected were a 12-mm-internal-diameter blunt conical design, the ConMed Trogard (ConMed Inc, Utica, NY) and an 8-mm-internal-diameter pyramidal design (United States Surgical Corporation, Stamford CT).

For our study model, 8 white swine animals, each weighing 60–65 kg, were selected. The white swine are large animals and possess a double-layer lower abdominal fascia similar to the layers of muscular aponeuroses in the human. The protocol was reviewed and approved by the University of California, Los Angeles Office for the Protection of Research Subjects. Animals were anesthetized with ketamine and butorphanol and maintained with inhaled halothane, titrated to effect. The abdomens were shaved, an insufflation needle was placed in the midline, supraumbilically, and carbon dioxide gas was insufflated to maintain intraperitoneal pressure between 13 and 15 mm Hg. A 12-mm cannula was positioned in a separate midline site in the supraumbilical area after the creation of a suitable skin incision, and a laparoscope was inserted and attached to a standard surgical camera and television monitor.

Previous studies determined that more than 6 trocar-cannula system insertions began to compromise maintenance of target intraperitoneal pressure. Consequently, each subject animal was limited to 6 total insertion sites. Eight animals were used, allowing a total of 48 study insertions. Twelve trocar-cannula system insertions were performed for each group. Group A comprised the 12-mm-diameter blunt conical system, with wound metrics obtained immediately after insertion. Group B was also a set of blunt conical insertions, but wound measurements were made after the application of standardized cannula manipulations. Group C comprised the 8-mm pyramidal trocar-cannula system with wound measures recorded immediately after insertion. For group D, the 8-mm devices were inserted but the incision metrics were obtained after cannula manipulation. Groups A through D were created through the use of a randomized block design where each group was randomly assigned twice to each site (block). Trocar-cannula system assignments were recorded on cards that were placed into sealed and numbered envelopes.

For each subject, 1 envelope was opened and each designated trocar-cannula system (from 1 of group A, B, C, or D) was inserted into the assigned location. Insertions were performed under direct vision (by using the television monitor) through 1 of 6 transverse 1.8-cm paramedian incisions that were made in a uniform manner below the level of the umbilicus and separated by 8–10 cm. Each of the trocar-cannula system was introduced by the same experienced operator (M.G.M.), and inserted at a 90-degree angle to the abdominal wall.

After trocar-cannula system insertions were completed in a given animal, the devices that were randomized to the movement group (groups B and D) were identified and subjected to a set of manipulations reported in previously published work and designed to simulate those encountered in an operative laparoscopic surgical procedure.11 To summarize, a 10-mm laparoscopic hand instrument was inserted in these cannulas and then withdrawn a total of 10 times. Then the hand instrument was moved in an arc of approximately 150 degrees in a transverse plane and then cephalad to caudad in a sagittal plane, thus covering four directions, a total of 50 times. These standardized movements were designed to mimic instrument manipulation at surgery.

After trocar-cannula system insertions and any required simulated surgical manipulations were completed, all cannulas were removed and the intraperitoneal gas allowed to escape. Measurement of incision length followed previously described methodology8 summarized here. After exposure of the fascial wounds with careful dissection, metric calipers were used to measure the maximal length of exposed underlying muscle. Wound area was determined by tracing the fascial borders around the exposed area of muscle on a transparency of 2-mm grid squares placed over the fascial defect. The sum of the grid box quartiles was tabulated and the area calculated in square millimeters at a remote time without reference to the device assignment for each wound. After fascial parameters were obtained, the fascia was carefully dissected from each wound and muscle damage for each wound was graded based on a previously described scale10 comprising a 0–3 scoring system, where 0 is no damage, 1 is minimal damage, 2 is damage less than the fascial incision, and 3 is greater than or equal than the fascial incision. At the end of all testing, the animals were euthanized according to institution protocol using Eutha-6 (Western Medical Supply, Arcadia, CA).

This study has a balanced randomized block design blocking on site. A 2-way repeated-measure analysis of variance model was used with the subject animal as a random effect and trocar-cannula system group, site, and their interaction as fixed effects. SAS Procedure MIXED (SAS Inc. Cary, NC) was used to perform the analyses. In addition to the parametric (mean) comparisons made with repeated-measures analysis of variance, muscle scores between devices was also compared nonparametrically by using the Friedman test, blocking on site. The Shapiro-Wilkes test and “normal probability plots” were computed to assess whether the residual errors are well approximated by a Gaussian distribution.


The marginal means for each group across all sites are reported along with the corresponding standard error of the mean. Results are summarized in Figure 1 with all errors reported as standard errors of the mean. Mean fascial wound area associated with the 12-mm blunt conical device without manipulation (group A) was 8.58 ± 0.68 mm2. For the 8-mm pyramidal tipped trocar-cannula system (group C, also without motion), the area was 9.83 ± 0.68 mm2. After the standardized movements, the mean fascial wound areas for the 12-mm devices (group B) and group D, the 8-mm systems were 9.71 ± 0.74 mm2 and 9.63 ± 0.68 mm2, respectively. None of the differences were statistically significant (Figure 1). The analysis of variance showed no significant site-by-group interaction (P = .89). The residual errors were well approximated by a Gaussian distribution (Shapiro-Wilkes W = 0.99, P = .88).

Figure 1.
Figure 1.:
Mean wound area in square millimeters is shown for the 4 groups. The 12-mm blunt conical device is shown without (group A) and with (group B) standardized cannula manipulation or motion. Similarly, the 8-mm device is depicted without (group C) and with (group D) cannula motion. Error bars reflect standard error of the mean.Munro. Laparoscopic Access Systems. Obstet Gynecol 2004.

The mean maximal incision length for the 12-mm blunt conical device was 9.16 ± 0.77 mm without cannula manipulation (group A) and 9.61 ± 0.83 mm after standardized movement (group B). For the 8-mm pyramidal device, the premovement (group C) and postmovement (group D) mean maximal incision lengths were 9.14 ± 0.77 mm and 8.52 ± 0.77 mm, respectively. There were no statistically significant differences between groups (Figure 2). The analysis of variance showed no significant site-by-group interaction (P = .19). The residual errors were well approximated by a Gaussian distribution (Shapiro-Wilkes W = 0.993, P = .99).

Figure 2.
Figure 2.:
Mean wound length in millimeters is shown for all groups. The 12-mm blunt conical device is shown without (group A) and with (group B) standardized cannula manipulation or motion. The 8-mm device is depicted both without (group C) and with (group D) cannula motion. Error bars reflect standard error of the mean.Munro. Laparoscopic Access Systems. Obstet Gynecol 2004.

Mean muscle damage scores (Figure 3) were 1.0 for the 12-mm blunt conical device (group A), without movement, and 1.1 after cannula manipulation (group B). For the 8-mm pyramidal device without movement (group C), the mean score was 1.5 and 1.0 for the same device after standardized movements. None of these differences seemed clinically significant and none of the differences were statistically significant either under the parametric analysis of variance model or by the nonparametric Friedman tests. There was no significant site by group interaction (P = .19).

Figure 3.
Figure 3.:
Mean muscle damage score is shown for the 4 groups. The 12-mm blunt conical device is shown without (group A) and with (group B) standardized cannula manipulation or motion. Similarly, the 8-mm device is depicted without (group C) and with (group D) cannula motion. There were no significant differences between scores.Munro. Laparoscopic Access Systems. Obstet Gynecol 2004.


By using our white swine model, the measured wound metrics resulting from a 12-mm (internal diameter) laparoscopic peritoneal access system with a blunt conical trocar were similar to those of an 8-mm device with a pyramidal tip, both immediately after insertion and after subjecting the cannulas to simulated operative manipulation. Muscle damage scores were also similar, with no significant difference between systems regardless of cannula manipulation. The explanation for these findings is thought to reside in the different mechanisms by which blunt conical and pyramidal devices traverse the abdominal wall. Our previously published comparative trial showed that blunt conical systems created a fascial wound that was substantially smaller in area than those associated with pyramidal devices of similar caliber. We have hypothesized that, during introduction into the peritoneal cavity, blunt conical access devices divide tissue along fibromuscular lines by finding a path of lesser resistance.14 This is in contrast to pyramidal devices, which cut and cleave fascia and muscle, generally leaving larger defects. (Figures 1A and 1B)

In addition to objective measures, it was clear during examination of the fascial wounds that there were characteristic features generated by a specific device (Figure 1B). The asymmetrical nature of most characteristics of the wounds appearance can limit the accuracy of even carefully taken measurements. Our method for wound area calculation, which uses exact tracing of wound margins, attempts to compensate for these irregularities. The relative importance of wound area and maximal wound length on the subsequent risk of wound dehiscence and hernia remains unclear.

Somewhat surprisingly, cannula movements designed to simulate operative manipulations did not alter any of our measured parameters including wound area, length, and underlying muscle damage. Clinically, slippage of laparoscopic cannulas seems to occur with greater ease as a case progresses, a feature that led us to speculate that such a finding would be associated with measurable wound area or muscle damage. It is possible that differences in wound parameters adequate to allow slippage are not measurable with our current set of wound metrics. We are currently working on a computerized system to characterize cannula slippage that will be the subject of future reports.

Whereas this report suggests that the risk of incisional dehiscence and hernia with a 12-mm blunt conical device may be similar to that of an 8-mm pyramidal device, the risk of other complications of laparoscopic surgery, such as vascular injury, relative to the design differences of the devices was not addressed in this study. However, there is evidence from animal models that the pyramidal trocar design may increase the risk of major vascular injury when compared with conical designs.15

This comparison supports our hypothesis that a 12-mm blunt conical peritoneal access device and an 8-mm pyramidal device would have equivalent wound metrics in our animal model. Surprisingly, there were no absolute or relative changes in measured wound parameters after simulated surgical manipulation.

Although we do believe that the current work lends credence to the notion that it may be unnecessary to close defects that are the result of 12-mm blunt conical access devices, our data fall short of being able to prove such a contention. Whereas our study examines the performance of trocar-cannula system in a biologic model with quantifiable measures, there is always concern about translating the laboratory animal study to the clinical setting.

Unfortunately, it would be difficult to design a clinical trial to evaluate the relative risks of hernia in the 2 types of access system tested. Because the incidence of dehiscence and hernia is so low, the sample sizes required to design a study with adequate power would be very large. As a result, such a clinical trial may never be done. Nevertheless, we hope that these data provide information that will be useful to clinicians and the designers and manufacturers of access systems that may reduce wound-related complications related to laparoscopic surgery.

Financial Disclosure

This work was performed under an unrestricted grant from ConMed Inc, Utica, NY. The grant support covered the cost of acquiring and maintaining the animals, laboratory staff, and our laboratory facility in addition to the services of our statistical consultant. There were no funds supporting the investigators’ salaries. Neither of the investigators has stock in the company nor do they have stock options.


1. Nordestgaard AG, Bodily KC, Osborne RWJr, Buttorff JD. Major vascular injuries during laparoscopic procedures. Am J Surg 1995;169:543–5.
2. Catarci M, Carlini M, Gentileschi P, Santoro E. Major and minor injuries during the creation of pneumoperitoneum: a multicenter study on 12,919 cases. Surg Endosc 2001;15:566–9.
3. Champault G, Cazacu F, Taffinder N. Serious trocar accidents in laparoscopic surgery: a French survey of 103,852 operations. Surg Laparosc Endosc 1996;6:367–70.
4. Bhoyrul S, Vierra MA, Nezhat CR, Krummel TM, Way LW. Trocar injuries in laparoscopic surgery. J Am Coll Surg 2001;192:677–83.
5. Plaus WJ. Laparoscopic trocar site hernias. J Laparoendosc Surg 1993;3:567–70.
6. Montz FJ, Holschneider CH, Munro MG. Incisional hernia following laparoscopy: a survey of the American Association of Gynecologic Laparoscopists. Obstet Gynecol 1994;84:881–4.
7. Kadar N, Reich H, Liu CY, Manko GF, Gimpelson R. Incisional hernias after major laparoscopic gynecologic procedures. Am J Obstet Gynecol 1993;168:1493–5.
8. Tarnay CM, Glass KB, Munro MG. Incision characteristics associated with six laparoscopic trocar-cannula systems: a randomized, observer-blinded comparison. Obstet Gynecol 1999;94:89–93.
9. Tarnay CM, Glass KB, Munro MG. Entry force and intra-abdominal pressure associated with six laparoscopic trocar-cannula systems: a randomized comparison. Obstet Gynecol 1999;94:83–8.
10. Glass KB, Tarnay CM, Munro MG. Intraabdominal pressure and incision parameters associated with a pyramidal laparoscopic trocar-cannula system and the EndoTIP cannula. J Am Assoc Gynecol Laparosc 2002;9:508–13.
11. Glass KB, Tarnay CM, Munro MG. Randomized comparison of the effect of manipulation on incisional parameters associated with a pyramidal laparoscopic trocar-cannula system and the EndoTIP cannula. J Am Assoc Gynecol Laparosc 2003;10:412–4.
12. Bohm B, Knigge M, Kraft M, Grundel K, Boenick U. Influence of different trocar tips on abdominal wall penetration during laparoscopy. Surg Endosc 1998;12:1434–8.
13. Leibl BJ, Schmedt CG, Schwarz J, Kraft K, Bittner R. Laparoscopic surgery complications associated with trocar tip design: review of literature and own results. J Laparoendosc Adv Surg Tech A 1999;9:135–40.
14. Munro MG. Laparoscopic access: complications, technologies, and techniques. Curr Opin Obstet Gynecol 2002;14:365–74.
15. Hurd WW, Wang L, Schemmel MT. A comparison of the relative risk of vessel injury with conical versus pyramidal laparoscopic trocars in a rabbit model. Am J Obstet Gynecol 1995;173:1731–3.
© 2004 The American College of Obstetricians and Gynecologists