Endometriosis is a benign gynecologic lesion found mostly in reproductive‐age women with a prevalence approximating 10% or higher.1 It may cause dysmenorrhea, dyspareunia, chronic pelvic pain, and subfertility.2 Confirmation of endometriosis is usually dependent upon laparoscopy or laparotomy. However, since Barbieri et al3 first demonstrated the association between elevated serum CA‐125 concentration and the presence of moderate‐to‐severe endometriosis, preoperative CA‐125 measurement has been increasingly used for the diagnosis of endometriosis, especially in infertile patients. Several studies4–7 and a meta‐analysis8 have assessed the performance of serum CA‐125 assay in the detection of endometriosis, and the results show a specificity of 85% and sensitivity between 20% and 50%. However, these studies focused on infertile women, and the number of patients was small. Furthermore, only the diagnostic usefulness of CA‐125 has been stressed in most papers.
Surgical treatment of endometriosis either conserves fertility or relieves symptoms through hysterectomy with or without adnexectomy in women who have completed childbearing and have severe pain. In these cases, severe adhesions between rectum, bowel wall, and pouch of Douglas are commonly found, making surgery at high risk for bowel trauma. Operative laparoscopic surgery for the treatment of endometriosis offers patients the advantage of reduced hospital stay and cost.9 However, bowel injury during laparoscopic surgery has been reported, especially with severe adhesions.10 Furthermore, the risk of organ injury may be more pronounced when laparoscopic‐assisted vaginal hysterectomy is performed.11 In this study, we attempt to investigate the factors that are associated with an elevated level of CA‐125 in endometriosis, and to study whether preoperative CA‐125 assay is useful to identify women at high risk who would require bowel preparation preoperatively.
MATERIALS AND METHODS
From July 1988 to June 1999, a total of 803 consecutive patients who underwent surgery for endometriosis at the National Cheng Kung University Hospital (Tainan, Taiwan) were included. The diagnosis was made on the basis of pathological examination. Informed consent from the patients was obtained, and the study was approved by the institutional review board of National Cheng Kung University Hospital. We reviewed the medical records including the obstetric and gynecologic history and relevant medical history, clinical symptoms, preoperative serum CA‐125 levels, types of operation, and operation findings in detail. We described the pelvic condition in each patient by the following parameters including adhesion in the peritoneum, omentum, ovary, fallopian tube, colon, and cul‐de‐sac, size and rupture of endometrioma, and the corresponding American Society of Reproductive Medicine stage. The assay for serum CA‐125 was performed 1 day before surgery, and the concentrations were measured with a two‐site immuno‐radiometric assay (CIS Bio International, Group ORIS, Cedex, France). Surgery method, including either laparotomy or laparoscopic surgery, was according to surgeon preference. All complications during or after surgery were recorded. Major operative complications were defined as bowel, bladder, ureter, vascular injuries, ileus that required prolonged hospital stay, or abdominal wall or intraperitoneal bleeding.
Based on the two‐sample Student t test, the sample size required for detecting a difference in mean serum CA‐125 levels of one‐half of one standard deviation with a statistical power of 90% when the sample size ratio of two groups was 1:50 was 675 with a significance level of 5%. F test was used for the comparison of serum CA‐125 levels among different groups. χ2 statistics based on likelihood ratio principle were employed to analyze categorical data. Multiple regression was performed to identify significant correlates of elevated serum CA‐125. A P value of <.05 was considered significant. Receiver operating characteristic (ROC) curve analysis was performed to assess the clinical utility of serum CA‐125 in distinguishing high‐risk from low‐risk patients. The cutoff values derived from ROC curves were evaluated in terms of sensitivity, specificity, and positive and negative predictive values.
From July 1988 to June 1999, 803 consecutive patients underwent surgical procedures for endometriosis that was confirmed by pathologic examination. However, 118 patients were excluded because of incomplete medical records or lack of data for CA‐125. Thus, a total of 685 patients were eligible for analysis. The age of patients ranged from 14 to 68 years, with a mean age of 33.7 ± 7.0 years. The clinical symptoms, parity, and types of surgical procedures are summarized in Table 1. A total of 202 (29.5%) patients were operated on for infertility, and the remainder experienced pelvic tumor or pain. Accordingly, in contrast to studies involving mainly infertile patients, 219 patients underwent hysterectomy, and nearly 45% of the surgical procedures were performed by laparotomy.
The distribution of serum CA‐125 levels in endometriosis patients with regard to their pelvic conditions and American Society of Reproductive Medicine stages are summarized in Table 2. Forty‐seven patients had the serum CA‐125 examination during menstruation. The mean serum CA‐125 levels were 18.8 ± 0.9 IU/mL for stage I endometriosis, 40.3 ± 2.8 IU/mL for stage II disease, 77.1 ± 3.5 IU/mL for stage III, and 182.4 ± 14.0 IU/mL for stage IV endometriosis, respectively. The value of CA‐125 increased as the stage of endometriosis increased, and the difference between stages was statistically significant (P < .001, F test). Analyses as shown in Table 2 indicated that serum CA‐125 levels were significantly higher in patients with more extensive adhesions to the peritoneum, omentum, ovary, fallopian tube, colon, and cul‐de‐sac, or with the presence of ruptured endometrioma at surgery. The mean size for ovarian endometrioma was 4.19 ± 2.91 cm, and no significant correlation was observed between the serum CA‐125 level and sizes of ovarian endometrioma (P > .05 by Pearson correlation analysis). Multiple regression showed that stages III (P = .001) and IV (P < .001), omentum adhesion (P < .001), and ruptured endometrioma (P < .001) were highly significantly associated with elevated serum CA‐125. There was a substantial overlap of serum CA‐125 levels between patients with and without complete cul‐de‐sac adhesion (Figure 1A), which might account for its lack of significant association with elevated serum CA‐125 in multivariate analysis. However, we thought that there was clinical importance for complete cul‐de‐sac adhesion in bowel preparation, and therefore included it for further analysis. As for omentum adhesion and rupture of endometrioma, the discrimination by CA‐125 was better, but the case number was limited to 18 (Figure 1B,C).
We then classified endometriosis patients to either high risk, in which at least one of these three factors, (ie, complete cul‐de‐sac adhesion, omentum adhesion, or rupture of endometrioma) was present, or low risk in which none of these three predictors was present. Accordingly, a total of 121 patients were assigned to the high‐risk group, and the mean serum CA‐125 level was 179.6 ± 15.5 IU/mL. In the low‐risk group, there were 564 patients, and the mean serum CA‐125 value was 59.8 ± 2.6 IU/mL, a significant difference from the high‐risk group (P < .001, F test). The demographic descriptions of high‐ and low‐risk groups are shown in Table 3. Patients in high‐risk groups were older, and were more likely to receive laparotomy when compared with patients in low‐risk groups (P < .01, χ2 statistics). The ROC curve analyses showed that the area under the ROC curve was 78.9%, which indicated that the model could accurately predict the subgroup allocations in 78.9% of patients. From the ROC curve, we selected a cutoff point of 65 IU/mL, which gave a sensitivity of 76% (95% confidence interval 68.4%, 83.6%), a specificity of 71% (95% confidence interval 67.3%, 74.7%), a positive predictive value of 76%, and a negative predictive value of 93.2%. The fraction of CA‐125 values among the high‐ and low‐risk groups is depicted in Figure 2. For any value of CA‐125 along the abscissa, the fraction equal to or greater than a cutoff value was higher in the high‐risk group than in the low‐risk group. For example, approximately 80% of the CA‐125 in the high‐risk group was equal to or greater than 65 IU/mL, whereas only 30% was equal to or greater than 65 IU/mL in the low‐risk group.
This study demonstrates that preoperative CA‐125 assay is useful to distinguish high‐from low‐risk endometriosis patients. Patients with preoperative CA‐125 levels higher than 65 IU/mL are at high risk for advanced stages or severe pelvic adhesion such as omentum adhesion or ruptured endometrioma, which may warrant preoperative bowel preparation. Our results indicate a new use for preoperative serum CA‐125 examination in addition to the diagnosis of endometriosis. Our study was not limited to infertility patients. Furthermore, we intended to emphasize that thorough preoperative planning is necessary for complex surgical procedures such as resection of endometriosis or hysterectomy.
Our results confirm previous reports that preoperative CA‐125 levels increase with the stages of endometriosis.3,4,6 It further extends the observations that omentum adhesion and rupture of endometrioma are also the leading causes of elevated CA‐125 levels. Such pelvic pathology will increase the risk of bowel injury. However, bowel preparation is unpleasant for the patient and increases medical expenses. Thus, it is useful to identify these high‐risk patients who require preoperative bowel preparation. Unfortunately, notable overlap of serum CA‐125 levels among patients with and without complete cul‐de‐sac adhesion and the limited number of patients with omentum adhesion or ruptured endometrioma render the differentiation unlikely. Although cul‐de‐sac adhesion was not significantly associated with elevated serum CA‐125 in multivariate analysis, we included complete cul‐de‐sac adhesion because of its clinical importance in bowel preparation. As shown in Figure 2, its inclusion gave reasonable sensitivity and specificity. Therefore, we subgroup all eligible patients into either high‐ risk or low‐risk groups. The ROC curve analyses demonstrate that we are able to predict which women should receive preoperative bowel preparation in nearly 80% of endometriosis patients. Analysis by distribution curves of CA‐125 levels as illustrated in Figure 2 further indicates that CA‐125 is a valid marker in distinguishing between high‐ and low‐risk patients. A cutoff point of CA‐125 at 65 IU/mL gives a sensitivity of 76% and a specificity of 71%. A shift of cutoff values of CA‐125 from 65 IU/mL to 35 IU/mL results in an increase in sensitivity from 76% to 88.4% but a notable decrease in specificity from 71% to 40.8%.
A limitation of the present analysis is that we are unable to predict the necessity of preoperative bowel preparation in 20% of patients recruited. In addition, with the specificity of 71% for the cutoff value CA‐125 at 65 IU/mL, 29% of predicted high‐risk patients will receive an unnecessary bowel preparation. Koninckx et al2 report that nodularities at clinical examination during menstruation or follicular phase CA‐125 concentrations over 35 IU/mL are useful to decide that bowel preparation should be given, achieving a sensitivity of 87% and a specificity of 83%. However, the finding of painful nodularities at clinical examination is quite subjective and may require analgesia and experience of the surgeons to obtain satisfactory results. Furthermore, women in certain geographic areas such as in Taiwan are reluctant to undergo pelvic examination during menstruation. The mechanism for high CA‐125 level and severe pelvic adhesion in endometriosis is not fully understood. The expression of intercellular adhesion molecule‐1 and interferon‐γ is reported to be associated with endometri‐osis.12–14 Further studies involving the combined use of CA‐125 and cytokines are currently under investigation.
The use of operative laparoscopy in the treatment of minimal and mild endometriosis has been widely accepted.8 More complex gynecologic procedures including treatment for advanced endometriosis are being performed laparoscopically, which may account for an increased rate of injuries to organs such as the bladder, ureter, intestine, and blood vessels.15 Indeed, in this report, nine women had complications, five of which were major, and all of them occurred in patients undergoing laparoscopic surgery without preoperative bowel preparation. Among them, three patients with colon injuries had preoperative CA‐125 values over 65 IU/mL and complete obliteration of the pouch of Douglas. Surgical repair was not attempted because none had received preoperative bowel preparation. They underwent colostomy a few days after laparoscopy. In contrast, 66 patients with CA‐125 values over 65 IU/mL undergoing preoperative bowel preparation did not have major complications after surgery, suggesting that preoperative bowel preparation may help to reduce the consequences of intraoperative injury. It is noteworthy that preoperative use of gonadotropin‐releasing hormone agonist has the benefit of decreasing the size of ovarian endometrioma and the inflammatory reaction surrounding the endometrioma, which would improve surgical performance.16,17
1. Olive DL, Schwartz LB. Endometriosis. N Engl J Med 1993;328:1759–69.
2. Koninckx PR, Meuleman C, Demeyere S, Lesaffre E, Cornillie FJ. Suggestive evidence that pelvic endometriosis is a progressive disease, whereas deeply infiltrating endometriosis is associated with pelvic pain. Fertil Steril 1991; 55:759–65.
3. Barbieri RL, Niloff JM, Bast RC Jr, Schaetzl E, Kistner RW, Knapp RC. Elevated serum concentrations of CA-125 in patients with advanced endometriosis. Fertil Steril 1986;45:630–4.
4. Pittaway DE, Fayez JA. The use of CA-125 in the diagnosis and management of endometriosis. Fertil Steril 1986;46:790–5.
5. Moretuzzo RW, DiLauro S, Jenison E, Chen SL, Reindollar RH, McDonough PG. Serum and peritoneal lavage fluid CA-125 levels in endometriosis. Fertil Steril 1988;50:430–3.
6. Franchi M, Beretta P, Zanaboni F, Donadello N, Ghezzi F. Use of serum CA-125 measurement in patients with endometriosis. Br J Obstet Gynaecol 1993;4:149–52.
7. Barbati A, Cosmi EV, Spaziani R, Ventura R, Montanino G. Serum and peritoneal fluid CA-125 levels in patients with endometriosis. Fertil Steril 1994;61:438–42.
8. Mol BW, Bayram N, Lijmer JG, Wiegerinck MA, Bongers MY, van der Veen F, et al. The performance of CA-125 measurement in the detection of endometriosis: A meta-analysis. Fertil Steril 1998;70:1101–8.
9. Gomel V, James C. Intraoperative management of ureteral injury during operative laparoscopy. Fertil Steril 1991;55:416–9.
10. Soderstrom RM. Bowel injury litigation after laparoscopy. J Am Assoc Gynecol Laparosc 1993;1:74–7.
11. Mirhashemi R, Harlow BL, Ginsburg ES, Signorello LB, Berkowitz R, Feldman S. Predicting risk of complications with gynecologic laparoscopic surgery. Obstet Gynecol 1998;92:327–31.
12. Tabibzadeh S, Kong QF, Babaknia A. Expression of adhesion molecules in human endometrial vasculature throughout the menstrual cycle. J Clin Endocrinol Metab 1994;79:1024–32.
13. Creyghton WM, de Waard-Siebinga I, Danen EH, Luyten GP, van Muijen GN, Jager MJ. Cytokine-mediated modulation of integrin, ICAM-1 and CD44 expression on human uveal melanoma cells in vitro. Melanoma Res 1995;5:235–42.
14. Wu MH, Yang BC, Hsu CC, Lee YC, Huang KE. The expression of soluble intercellular adhesion molecule-1 in endometriosis. Fertil Steril 1998;70:1139–42.
15. Harkki-Siren P, Sjoberg J, Kurki T. Major complications of laparoscopy: A follow-up Finnish study. Obstet Gynecol 1999;94:94–8.
16. Donnez J, Nisolle M, Gillet M, Smets M, Bassil S, Casanas-Roux F. Large ovarian endometriomas. Hum Reprod 1996;11:641–6.
© 2002 The American College of Obstetricians and Gynecologists
17. Donnez J, Nisolle M, Gillerot S, Anaf V, Clerckx-Braun F, Casanas-Roux F. Ovarian endometrial cysts: The role of gonadotropin-releasing hormone agonist and/or drainage. Fertil Steril 1994;62:63–6.