Factors that increase the risk of leakage during surgical removal of benign cystic teratomas

Magdy P. Milad1,3 and Elissa Olson2

1 Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Northwestern University Medical School and 2 Northwestern University Medical School, Chicago, Illinois, USA


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The contents of mature cystic teratomas can be a potent irritant resulting in chemical peritonitis. Using a retrospective cohort, we examined the various risk factors for leakage of benign cystic teratomas during laparoscopy and laparotomy. Cyst leakage of the benign cystic teratoma contents was the primary endpoint. In all, 158 women underwent surgery for a total of 178 ovarian benign cystic teratomas. Statistical analysis was performed using {chi}2, Mann–Whitney U and multivariate logistic regression analysis. A total of 115 benign cystic teratomas was successfully removed without intra-operative leakage and 63 underwent intra-operative leakage either at laparoscopy or laparotomy. The likelihood of success of removing the benign cystic teratoma intact was unrelated to age, pre-operative size or surgical technique. There was no difference among cystectomies performed by laparotomy in surgeon experience or the presence of adhesions. However, surgeons with more laparoscopic experience (>35 laparoscopies/year) were less likely to have intra-operative leakage (relative risk: 0.5, 95% confidence interval: 0.2, 1.2) compared to surgeons with less experience (<20/year) at cystectomy (26.1 versus 51.2% respectively). Oophorectomy significantly reduced the frequency of intra-operative leakage at both laparoscopy and laparotomy (14.7%). These findings suggest that laparoscopic experience can reduce the risk of leakage at cystectomy. At laparotomy, lack of surgeon postgraduate years of experience was not a risk factor for leakage.

Key words: dermoids/laparoscopy/mature cystic teratomas/ovarian cysts/surgeon experience


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The management of mature cystic teratomas of the ovary has changed dramatically following advances in endoscopic equipment and techniques. As a result, benign cystic teratomas are often managed laparoscopically with shorter recovery periods and smaller incisions (Yuen et al., 1997Go). However, laparoscopy has reduced tactile feedback and requires specimen removal through small incisions that may increase the leakage at surgery. This is a matter for concern, since the contents of benign cystic teratomas may cause significant peritonitis (Pantoja et al., 1975Go; Fiedler et al., 1996Go).

This study was undertaken to identify the risk factors associated with the intra-operative leakage of benign cystic teratomas. Specifically, we sought to determine whether surgeons' experience, surgical technique, presence of adhesions and cyst size predisposed to intra-operative leakage.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Our computerized surgical server was used to screen consecutive cases seen at the Northwestern Memorial Hospital from January 1993 through May 1998. The diagnosis of benign cystic teratoma was made following pathological examination of the surgical specimen. Initially, 192 potential cases were identified. After an initial chart review, 14 cases were excluded for the following reasons: benign cystic teratoma not confirmed by pathology (n = 5) or the complete record was not available (n = 9). This left 178 cysts in 158 women available for analysis. Pertinent information regarding each patient's age, gravidity, parity, previous abdominal surgeries, previous benign cystic teratoma cysts and subsequent admissions was recorded. Site of the cyst, presence or absence of adhesions and mean diameter by ultrasound and at surgery as estimated with a calibrated instrument were also recorded. Surgical procedure, including laparoscopy versus laparotomy, method of cyst opening and dissection and the performance of concurrent oophorectomy were noted. Benign cystic teratoma leakage was confirmed if the cyst leaked prior to placement into an endoscopic bag at laparoscopy or at any time when the cyst fluid was visible prior to specimen removal. The phase of surgery when leakage occurred was also recorded. The surgical operating time was measured from skin incision to closure. The estimated blood loss was measured at laparoscopy by comparing the volume of irrigation fluid used with the volume of fluid collected after the procedure was completed. At laparotomy, the blood loss was estimated by anaesthesia personnel and corroborated by the circulating nurse.

Each surgeon's experience was estimated by calculating the average number of laparoscopic cases per year performed from 1993 to 1997. Number of years as a physician was determined based on the year of graduation from medical school as reported to the American Medical Association. Patients under the resident service were cared for by physicians undergoing training under the supervision of an experienced surgeon. Statistical analysis was performed using SPSS Version 6.1.1 (SPSS Inc., Chicago, IL, USA). Dichotomous variables were compared by {chi}2 test. Continuous variables were compared with two-sided sample t-test and for non-parametric analysis the Mann–Whitney U-test. Multiple logistic regression was also performed.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
There were no significant differences in age, gravidity, history of previous benign cystic teratoma, pelvic surgery or presentation to the resident service between women with and without intra-operative leakage (Table IGo). There was no difference between cysts that leaked and those that did not in location (left or right), mean diameter at pre-operative ultrasound, or diameter as estimated at surgery (Table IIGo). The presence of adhesions did not increase the risk of intra-operative leakage. Of the 178 cysts in the study, 34 were removed by oophorectomy and most (29) did not undergo intra-operative leakage (85.3%). Of those that did (14.7%), three occurred at laparotomy and two at laparoscopy. Leakage occurred while performing adhesiolysis (n = 2), during specimen removal (n = 2), or was not specified (laparotomy). In all, 144 cysts were removed by cystectomy; 80 at laparotomy and 64 by laparoscopy. The overall incidence of leakage at cystectomy was not statistically different between patients undergoing laparotomy (38.7%) and laparoscopy (42.2%). There was no difference in leakage among cystectomies performed by the resident service or by the presence of adhesions. Among patients undergoing cystectomy by laparotomy, surgeon postgraduate years and pre-operative size were not associated with intra-operative leakage (Table IIIGo).


View this table:
[in this window]
[in a new window]
 
Table I. Patients undergoing laparoscopy or laparotomy for benign cystic teratoma grouped according to whether or not intra-operative leakage occurred
 

View this table:
[in this window]
[in a new window]
 
Table II. Factors associated with benign cystic teratoma leakage at oophorectomy and cystectomy, with cysts grouped according to whether or not leakage occurred
 

View this table:
[in this window]
[in a new window]
 
Table III. Factors associated with leakage at benign cystic teratoma cystectomy
 
The most commonly performed technique employed for laparoscopic cystectomy was using three or four punctures using a 10 mm umbilical port and a second 10 mm trocar for specimen removal. Monopolar needle electrosurgery was used to score the ovarian capsule followed by enucleation without intentionally puncturing the cyst. Copious irrigation was used following cystectomy (1–4 l). Success or failure of cystectomy was not influenced by surgical techniques such as use of aquadissection, blunt or sharp dissection (data not shown).

Surgeon laparoscopic experience was distributed in a bimodal pattern, with 39 surgeons performing <20 laparoscopic cases per year and five surgeons performing >35 cases per year. No surgeon averaged between 20 and 35 laparoscopic cases per year. Patients who underwent laparoscopic cystectomy by experienced laparoscopic surgeons (>35 laparoscopies/year) were less likely to have intra-operative leakage (relative risk: 0.5, 95% confidence interval: 0.2–1.2) (Table IIIGo). Among patients undergoing unilateral cystectomy without concurrent procedures, estimated blood loss and operating time were not different between women (leakage, no leakage, laparoscopy, laparotomy; data not shown). The phase at which leakage occurred during cystectomy did not statistically differ for laparoscopy and laparotomy, including respectively at ovarian capsule incision (14.1 versus 11.3%), cyst dissection (17.2 versus 22.5%) and during cyst removal (9.4 versus 3.8%). A backward likelihood ratio logistic regression analysis was performed on the entire dataset to assess further factors related to intra-operative benign cystic teratoma leakage including surgeon years of experience, surgeon laparoscopic experience, preoperative size, laparoscopy, and presence of adhesions. In the final logistic model, only laparoscopic experience was found to protect against intra-operative leakage (P = 0.058).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Leblanc in 1831 first used the term `kyste dermoid' in referring to a tumour at the base of a horse's skull (see Commerci et al., 1994). Since mature cystic teratomas are composed of all three germ cell layers, the term `dermoid' is not considered correct. Mature cystic teratomas are thought to arise from primordial germ cells (Pantoja et al., 1975Go). This theory is supported by the distribution of these tumours along the lines of migration from the yolk sac to the primitive gonad. The majority of these tumours occur during the reproductive years lending further support to the germ cell theory.

With the introduction of high-resolution transvaginal ultrasonography, the diagnosis of a benign cystic teratoma can be made with greater accuracy. Cystic teratomas are suspected if any one of the following three sonographic findings is present in a woman of reproductive age: a densely echogenic tubercle associated with a cystic echo pattern, a thin, echogenic, band-like echo or a dense echo pattern with or without a cystic component (Cohen and Sabbagha, 1993Go). In a prospective series, a diagnostic accuracy of 97% was noted in predicting cystic teratomas (Caspi et al., 1996Go). Unfortunately, malignancy complicates 0.17–3% of cases (Commerci et al., 1994Go) and cannot be completely excluded on the basis of ultrasound findings, even with colour Doppler flow imaging (Luxman et al., 1996Go). Also, benign cystic teratomas may be at risk for chronic leakage (1%), acute rupture (1–3%) or torsion (3%), and as a result surgical intervention is often recommended (Commerci et al., 1994Go). Careful observation (Caspi et al., 1997Go) or cystectomy should be considered in patients who desire future fertility.

The rate of bilateral mature cystic teratomas was 11% in this study, consistent with other studies (Commerci et al., 1994Go). In the past, sampling of the contralateral ovary was recommended to rule out a second mature cystic teratoma. Unfortunately, indiscriminate removal of ovarian tissue rarely leads to the identification of a teratoma. In addition, potential complications of ovarian biopsy include haemorrhage, infection, adhesion formation and possible reduction of follicles (Toaff et al., 1976Go). For these reasons, careful inspection of the contralateral ovary is recommended with diagnostic cyst aspiration followed by cystectomy if indicated.

Although several studies have emphasized the importance of experience (Howard, 1995Go; Nitke et al., 1996Go), this is the first paper to study the effects of surgeon experience on leakage from mature cystic teratomas. The current study suggests that cystectomy by laparotomy is highly insensitive to surgeon experience as measured in years. Postgraduate years offer no relative benefit to preventing benign cystic teratoma leakage at laparotomy. However, laparoscopic experience is highly predictive of success at laparoscopic cystectomy. Those surgeons performing <20 cases per year are more likely to have inadvertent leakage at laparoscopic cystectomy. In light of the concerns of chemical peritonitis from benign cystic teratoma cyst fluid, less experienced laparoscopic surgeons are advised to involve more experienced surgeons during endoscopic management, or to consider laparotomy. This finding also helps to validate the Society for Reproductive Surgeons' recommendation that laparoscopic cystectomy be considered a level II procedure (Keye, 1994Go).

In the only randomized prospective study reporting leakage rates (Yuen et al., 1997Go) during the management of benign ovarian cysts, the overall frequency of inadvertent leakage was similar for laparoscopy (27%) and laparotomy (30%). Unfortunately, only 20 benign cystic teratomas were managed by cystectomy. In that study, the benign cystic teratoma rupture rate during cystectomy at laparotomy was 44% compared to 18% in the laparoscopy group. This was similar to the current findings, with cystectomy leakage at laparotomy (39%) occurring more frequently at laparoscopy than by experienced surgeons (26%). Other observational studies have reported leakage rates at laparoscopic resection of anywhere between 13 and 100% (Nezhat et al., 1989Go; Bollen et al., 1992Go; Reich et al., 1992Go; Chapron et al., 1994Go; Howard, 1995Go; Lin et al., 1995Go; Teng et al., 1996Go; Ulrich et al., 1996Go; Shalev et al., 1998Go).

Granulomatous chemical peritonitis has been previously reported in at least 100 patients (Wolfe et al., 1984Go). Peritonitis associated with the leakage of benign cystic teratomas is characterized by multiple small yellow–white implants and dense adhesions (Stern et al., 1981Go). Microscopically, these lesions are granulomas with lipoid-laden macrophages, lymphocytes, plasma cells and foreign-body giant cells present (Kistner et al., 1952Go; Stuart and Smith, 1983Go). Chronic peritoneal irritation secondary to leakage can mimic pelvic inflammatory disease, carcinomatosis or tuberculous peritonitis. Keratin and sebum are considered potent irritants (Kurrein and Fothergill, 1961Go).

At laparotomy, intraperitoneal leakage during ovarian cystectomy can be minimized by exteriorizing the ovary and placing packs prior to ovarian capsule incision. A similar approach can be taken at laparoscopy with placement of the ovary into a bag prior to cystectomy. Some authors argue that copious irrigation will reduce or prevent the risk of peritonitis at laparoscopic cystectomy. For this reason, benign cystic teratoma cysts have been managed laparoscopically without regard to cyst leakage (Hessami et al., 1995Go; Howard, 1995Go). Although we agree that irrigation is an important component of any surgery, it is not a substitute for experience. As demonstrated by this study, laparoscopic experience can reduce benign cystic teratoma leakage to a minimum. Also, benign cystic teratoma fluid is commonly viscous and may contain other potentially inflammatory products, including hair, sebum, bone, cartilage and fat. If these contents are allowed to escape, they may be difficult to remove manually or with irrigation. Irrigation in the Trendelenberg position may also displace these products into the upper abdomen or other areas not easily visualized. Also, lavage failed to reduce the level of inflammation and adhesion formation down to control levels as assessed histologically in the rabbit model (Fiedler et al., 1996Go).

The laparoscopic approach to large benign cystic teratoma cysts (>10 cm diameter) may provide some technical difficulties to specimen removal. Commercially available endoscopic bags often do not open to beyond 6–10 cm in diameter. To facilitate removal of large specimens, some authors suggest the use of laparoscopic assisted vaginal resection (Pardi et al., 1995Go; Teng et al., 1996Go), extending the abdominal incision (Shalev et al., 1998Go) or aspirating and irrigating the cyst fluid until the fluid clears, followed by cystotomy closure and subsequent cystectomy (Ulrich et al., 1996Go). In our series, none of the benign cystic teratomas exceeded 10 cm in diameter.

This is the largest observational study to date comparing the management of benign cystic teratomas by laparoscopy and laparotomy. We were able to examine the role of cyst size, adhesions and experience on leakage rate. As with all observational studies, we are limited in our conclusions. It is possible that a residual confounding factor was unmeasured. For example, we did not subdivide the number of advanced laparoscopic cases performed by each surgeon when estimating laparoscopic experience. We could have also attempted to quantify surgeon experience using the average number of cystectomies of all types for laparotomy and laparoscopy. This would have required reviewing thousands of charts. Despite these limitations, we were able to demonstrate a trend with overall laparoscopic experience.

In conclusion, benign cystic teratoma leakage at surgery can be significantly reduced if an oophorectomy is performed. However, oophorectomy should only be considered in patients who are not interested in preserving ovarian function. The method of access (laparoscopy versus laparotomy) does not influence the overall rate of benign cystic teratoma leakage at cystectomy. However, if cystectomy is attempted by laparoscopy, benign cystic teratomas are less likely to leak when the operation is performed by an experienced endoscopic surgeon.


    Notes
 
3 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Bollen, N., Camus, M., Tournaye, H. et al. (1992) Laparoscopic removal of benign mature teratoma. Hum. Reprod., 7, 1429–1432.[Abstract]

Caspi, B., Appelman, Z., Rabinerson, D. et al. (1996) Pathognomonic echo patterns of benign cystic teratomas of the ovary: classification, incidence and accuracy rate of sonographic diagnosis. Ultrasound Obstet. Gynecol., 7, 275–280.[ISI][Medline]

Caspi, B., Appelman, Z., Rabinerson, D. et al. (1997) The growth pattern of ovarian dermoid cysts: a prospective study in premenopausal and postmenopausal women. Fertil. Steril., 68, 501–505.[ISI][Medline]

Chapron, C., Dubuisson, J.B., Samouh, N. et al (1994) Treatment of ovarian dermoid cysts. Place and modalities of operative laparoscopy. Surg. Endosc., 8, 1092–1095.[ISI][Medline]

Cohen, L. and Sabbagha, R. (1993) Echo patterns of benign cystic teratomas by transvaginal ultrasound. Ultrasound Obstet. Gynecol., 3, 120–123.[ISI][Medline]

Commerci, J.T., Licciardi, F., Bergh, P. et al. (1994) Mature cystic teratoma: a clinicopathologic evaluation of 517 cases and review of the literature. Obstet. Gynecol., 84, 22–28.[Abstract]

Fiedler, E., Guzick, D., Guido, R. et al. (1996) Adhesion formation from release of dermoid contents in the peritoneal cavity and effect of copious lavage: a prospective, randomized, blinded, controlled study in a rabbit model. Fertil. Steril., 65, 852–859.[ISI][Medline]

Hessami, S., Kohanim, B. and Grazi, R. (1995) Laparoscopic excision of benign dermoid cysts with controlled intra-operative spillage. J. Am. Assoc. Gynecol. Laparosc., 2, 479–481.[ISI][Medline]

Howard, F. (1995) Surgical management of benign cystic teratoma. J. Reprod. Med., 40, 495–499.[ISI][Medline]

Keye, W.R. (1994) Hitting a moving target: credentialing the endoscopic surgeon. Fertil. Steril., 62, 1115–1119.[ISI][Medline]

Kistner, R.W. (1952) Intraperitoneal rupture of benign cystic teratomas. Review of the literature with a report of two cases. Obstet. Gynecol. Surv., 7, 603–617.[Medline]

Kurrein, F. and Fothergill, R.J. (1961) Oleo-keratin granuloma in peritoneum: a rare complication of ovarian dermoid. J. Obstet. Gynaecol. Br. Commonw., 68, 124–127.[ISI]

Lin, P., Falcone, T. and Tulandi, T. (1995) Excision of ovarian dermoid cyst by laparoscopy and by laparotomy. Am. J. Obstet. Gynecol., 173, 769–771.[ISI][Medline]

Luxman, D., Cohen, J.R. and David, M.P. (1996) Laparoscopic conservative removal of ovarian dermoid cysts. J. Am. Assoc. Gynecol. Laparosc., 3, 409–411.[ISI][Medline]

Nezhat, C., Weiner, W.K. and Nezhat, F. (1989) Laparoscopic removal of dermoid cysts. Obstet. Gynecol., 73, 278–280.[Abstract]

Nitke, S., Goldman, G.A., Fisch, B. et al. (1996) The management of dermoid cysts – a comparative study of laparoscopy and laparotomy. Isr. J. Med. Sci., 32, 1177–1179.[ISI][Medline]

Pantoja, I., Noy, M.A., Axtmayer, R.W. et al. (1975) Ovarian dermoids and their complications. Comprehensive historical review. Obstet. Gynecol., 30, 1–20.

Pardi, G., Carminati, R., Ferrari, M.M. et al. (1995) Laparoscopically assisted vaginal removal of ovarian dermoid cysts. Obstet. Gynecol., 85, 129–132.[Abstract/Free Full Text]

Reich, H., McGlynn, F., Sekel, L. et al. (1992) Laparoscopic management of ovarian dermoid cysts. J. Reprod. Med., 37, 640–647.[ISI][Medline]

Shalev, E., Bustan, M., Romano, S. et al. (1998) Laparoscopic resection of ovarian benign cystic teratomas: experience with 84 cases. Hum. Reprod., 13, 1810–1812.[Abstract]

Stern, J.L., Buscema, J., Rusenshein, N.B. et al. (1981) Spontaneous rupture of benign cystic teratomas. Obstet. Gynecol., 57, 363–366.[Abstract]

Stuart, G.C. and Smith, J.P. (1983) Ruptured benign cystic teratomas mimicking gynecologic malignancy. Gynecol. Oncol., 16, 139–143.[ISI][Medline]

Teng, F., Muzsnai, D., Perez, R. et al. (1996) A comparative study of laparoscopy and colpotomy for the removal of ovarian dermoid cysts. Obstet. Gynecol., 87, 1009–1013.[Abstract/Free Full Text]

Toaff, R., Toaff, M.E. and Peyser, M.R. (1976) Infertility following wedge resection of the ovaries. Am. J. Obstet. Gynecol., 48, 9S–35S.

Ulrich, U., Keckstein, J., Paulus, W. et al. (1996) Endoscopic surgery for mature teratoma of the ovary. Surg. Endosc., 10, 900–903.[ISI][Medline]

Wolfe, R.D., Weinreb, I. and Silver, C. (1984) Small-bowel changes secondary to intraperitoneal leakage from dermoid cyst of the ovary. Mt Sinai J. Med., 51, 382–326.[ISI][Medline]

Yuen, P., Yu, K., Yip, S., et al. (1997) A randomized prospective study of laparoscopy and laparotomy in the management of benign ovarian masses. Am. J. Obstet. Gynecol., 177, 109–114.[ISI][Medline]

Submitted on January 26, 1999; accepted on May 28, 1999.





This Article
Abstract
FREE Full Text (PDF )
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (3)
Request Permissions
Google Scholar
Articles by Milad, M. P.
Articles by Olson, E.
PubMed
PubMed Citation
Articles by Milad, M. P.
Articles by Olson, E.