Accuracy of laparoscopy for assessing patients with endometriosis

ABSTRACT CONTEXT AND OBJECTIVE: Diagnoses of endometriosis are based on observation of endometriotic lesions by means of laparoscopy, along with the pathological findings. The aim of this study was to evaluate the sensitivity and specificity of the macroscopic findings in relation to the histopathological findings. More specifically, we aimed to test the efficacy of laparoscopy alone for diagnosing endometriosis and to evaluate the laterality of endometriosis among the study population. DESIGN AND SETTING: Cross-sectional study on women undergoing laparoscopy due to pelvic pain or infertility, in the Gynecology Department of Hospital Santa Cruz in Curitiba, Paraná, Brazil, and Pontifícia Universidade Católica do Paraná. METHODS: A total of 976 patients underwent laparoscopy and biopsy due to pelvic pain and/or infertility. We analyzed the laparoscopic and histopathological findings from patients with pelvic endometriosis (n = 468) and patients without endometriosis (n = 508). RESULTS: In 468 (47.95%) of the cases, the clinical and laparoscopic findings were consistent with endometriosis, and this was confirmed histopathologically in 337 (34.5%). Among the remaining 508 patients, although the laparoscopy was performed for other reasons relating to acute pelvic pain, eight were diagnosed with endometriosis from histopathological examination of the pelvic specimens obtained. Therefore, endometriosis was confirmed in 345 patients (35.3%). In comparison with the histopathology, laparoscopy alone presented 97.68% sensitivity, 79.23% specificity, 72% positive predictive value and 98.42% negative predictive value. CONCLUSION: Laparoscopy should be used in conjunction with histopathology for diagnosing endometriosis.


INTRODUCTION
Endometriosis is described as a benign disease of the female genital system.It is principally characterized by endometrium-like tissue, consisting of glands and/or stroma, found outside the uterine cavity.Although implanted ectopically, this tissue presents histopathological and physiological responses that are similar to the responses of the endometrium. 1he main symptom of endometriosis is pelvic pain, which is often very intense.Dysmenorrhea and other complaints like dyspareunia and infertility are also seen. 2,3he diagnostic hypothesis of endometriosis is based on the clinical history, along with the results from gynecological examinations, laboratory tests and transvaginal ultrasound. 4,5ome clinical characteristics, the physical examination itself, laboratory test results and evidence from imaging examinations may suggest the diagnosis. 6The greatest diffi culty lies in diagnosing minimal and mild lesions.In these cases, the ideal access for confi rmation is always laparoscopic, since the complementary examinations available do not offer the necessary specifi city. 7iagnosis by means of laparoscopy, which is considered the gold standard, may depend to an as yet unknown degree on confi rmation by means of histopathological assessment.However, the defi nitive diagnosis of the disease can only be obtained through histopathological examination of the biopsy sample. 8ssessment of the accuracy of laparoscopy for diagnosing endometriosis has demonstrated that it is highly precise in ruling out the disease, thereby informing therapeutic decisions. 9Recent studies have shown that endometriosis is principally diagnosed by laparoscopy combined with histopathological examination, although a negative result does not rule out the possibility of the disease. 10

OBJECTIVE
The objective of this study was to assess the sensitivity and specificity of the macroscopic findings from laparoscopy, in relation to diagnoses of endometriosis based on the results from histopathological examinations.More specifi cally, we aimed to test the effi cacy of laparoscopy alone for diagnosing endometriosis and to evaluate the laterality of endometriosis among the study population.

MATERIALS AND METHODS
This was a cross-sectional study.We analyzed 976 women who underwent laparoscopy due to pelvic pain and/or infertility at the Obstetrics and Gynecology Department of Hospital Santa Cruz between 1994 and 2004.We analyzed the laparoscopic and histopathological fi ndings from all the patients.Of these 976 patients, 468 presented pelvic endometriosis and 508 patients did not present endometriosis (but had other gynecological conditions).
This study was analyzed and approved by the Ethics Committee of Pontifícia Universidade Católica do Paraná (PUCPR), under Ethics Committee Registration No. 530 and protocol No. 056.476.
The criteria for performing laparoscopy were as follows: the subject needed to be in the menacme and presenting pelvic pain, dyspareunia, dysmenorrhea or infertility; and the results from complementary tests such as CA125 determination and ultrasound needed to reveal pelvic masses or blood in the pelvic cavity.Patients who had not yet reached the menarche or had reached the menopause and cases of laparoscopic reintervention were excluded from the laparoscopy performed due to pelvic pain.
During the laparoscopy, we performed biopsies on anatomical abnormalities that presented the macroscopic appearance of endometriosis, i.e. typical lesions such as "powder burn", of reddish color (light or dark), light color (yellow or brown) or dark color (black or blue), or even on fibrotic lesions.The lesions suggestive of endometriosis were biopsied and histopathologically examined in the Pathological Anatomy Department of Hospital Santa Cruz.The endometriosis was staged in accordance with the 1985 American Fertility Society (AFS) classification system, and the staging was compared with the result from the histopathological analysis on the biopsies. 11earson's chi-squared test and Fisher's exact test were used to assess any proportional differences between the groups with and without endometriosis.Differences between the continuous variables were studied using analysis of variance (ANOVA).The significance level was set at P ≤ 5% for all tests and the power test was 90%.

RESULTS
Out of the 976 patients who underwent laparoscopy, 468 (47.95%) were selected for inclusion in the present study, since they presented clinical and laparoscopic profiles suggestive of endometriosis.Among these patients, the clinical and laparoscopic suspicion of endometriosis was confirmed from histopathological analysis in 337 cases (72.0%) (Figure 1).
In the cases of a further eight patients, a positive histopathological diagnosis was made during surgical procedures that were performed due to other causes.Therefore, a total of 345 patients were diagnosed through histopathology.
The mean age of these patients was 30.85 ± 5.54 years.The frequency of endometriosis of any stage was found to be highest among patients between the ages of 20 and 40 (P = 0.001).
Out of the 345 patients evaluated, 341 (98.84%) presented acute or chronic pelvic pain.Acute pain was more common among the patients presenting the milder stages of endometriosis, whereas chronic pelvic pain was more common in the more severe stages (P = 0.03).
There were 129 patients (37.39%) who complained of dysmenorrhea, 69 (20%) who reported primary infertility and 23 (6.66%) who reported secondary infertility.A tendency towards higher frequency of dysmenorrhea was found among patients with the more severe forms of endometriosis, whereas the frequency of primary or secondary infertility was comparable at all stages of the disease (Table 1).
The histopathological examination confirmed the presence of endometriosis in the right ovary in 77 cases (22.31%) and in the left ovary in 89 cases (25.79%).No statistically significant difference in frequency was observed between the right and left ovaries (22.31% versus 25.79%, P > 0.05).In 29 patients (8.4%), both ovaries were involved.
Endometriosis was identified in the peritoneum in 260 patients (75.36%) and in the rectovaginal septum in 41 (11.88%).
Taking the histopathological findings to be definitive for the diagnosis of endometriosis, the clinical suspicion and laparoscopic findings presented 97.68% sensitivity, 79.23% specificity, 72% positive predictive value, 98.42% negative predictive value, and 85.75% accuracy (Table 2).False positive results were obtained in 27.99% of the tests, compared with false negative results in 1.57% of the tests.

DISCUSSION
Despite the efforts of the scientific community to increase the efficacy of the methods used to diagnose endometriosis, various limitations remain, thus making it difficult to reach a definitive diagnosis.Clinical parameters such as pelvic pain, dysmenorrhea, dyspareunia and infertility are insufficient to confirm the diagnosis.3][14] Combining laparoscopy with histopathological examination yields greater sensitivity for the definitive diagnosis of the disease and also decreases the diagnostic errors. 15mong the 976 laparoscopies performed in this study, the frequency of endometriosis was 35.3%.This result is in accordance with findings from previous studies carried out among smaller population samples. 16,17Furthermore, our findings corroborate data in the literature regarding the mean age of the patients studied and are in keeping with the results from other studies showing that the onset of symptoms usually occurs within seven to twelve years after the menarche. 18n the present study, and in accordance with the 1985 AFS system for staging endometriosis, 12 4.36% of our patients were classified as stage I (minimal), 51.01% as stage II (mild), 32.46% as stage III (moderate) and 12.17% as stage IV (severe).In a study involving 44 patients who underwent laparoscopy due to pelvic pain, Petta reported that 50% presented stage I endometriosis, 12.5% presented stage II, 25% presented stage III and 12.5% presented stage IV. 19These results, together with others found in the literature, are listed in Table 3. 19,[20][21][22] Between these different studies, discrepancies can be observed among the stages found.Although one particular macroscopic mapping method for endometriosis was recommended by the American Society for Reproductive Medicine in 1997, the results from many studies differ according to the background and experience of the professional who performed the laparoscopy.Therefore, comparative assessments are affected. 20The results from the present study demonstrate the difficulties ensuing from macroscopic assessments made by various observers.The diversity of the results justifies the use of histopathological analysis for diagnostic confirmation of endometriosis. 23he diagnosis of histopathology-confirmed endometriosis presented a statistically significant association with chronic pelvic pain.However, according to the findings of Wardle and Hull, 24 acute pelvic pain, dysmenorrhea, primary infertility and secondary infertility had no statistically significant influence on the diagnosis of endometriosis.
To date, there is no consensus on the relationship between the extent of endometriosis and the intensity of pelvic pain. 25t has been shown that there is a correlation between certain histopathological findings (a well-differentiated pattern or a diagnosis of stromal disease) and the intensity of pelvic pain. 26In the present study, 98.84% of all patients (regardless of endometriosis stage) reported pelvic pain.Pelvic pain was found to correlate significantly with endometriosis stage (P = 0.03) (Table 3).
In other studies, it was reported that the severity of dysmenorrhea presented no significant association with the stage or location of endometriosis. 20,27Our results are in accordance with those of such studies, in that no positive correlation was found between the degree of endometriosis and the intensity of dysmenorrhea. 27We observed dysmenorrhea in 37.39% of our patients with confirmed endometriosis and in 26.71% of our patients without endometriosis, although the difference was not statistically significant.
Topalski Fistes et al. 27 carried out a comparative study with a control group of 200 fertile women.They found that the frequency of endometriosis was 32% among infertile women and 5% among fertile women, which was a statistically significant difference (P = 0.001).In the present study, the frequencies of primary or secondary infertility were comparable, regardless of the severity of the disease.
When we compared the laterality of ovarian involvement in the 345 women evaluated, we found similar frequencies (left ovary versus right ovary: 25.79% versus 22.31%; P > 0.05).Several studies evaluating endometriotic ovarian cysts have shown a predisposition towards left-sided lesions. 8,28,29However, this was not confirmed in our study.
In addition, we observed a greater incidence of the disease in the peritoneum (79.3%), regardless of the stage of endometriosis, whereas the incidence of peritoneal lesions described in the literature ranges from 17.5% to 31%. 19,20,22n the present study, the number of biopsies testing positive for endometriosis was directly proportional to the severity of the endometriosis.This shows that, whether laparoscopy or histopathology is used, it is more difficult to make a definitive diagnosis when the lesions are minimal or mild.
In a study assessing macroscopic findings of anatomical abnormalities and confirmation of endometriosis, it was found that 85.7% of the patients presented pelvic anatomical abnormalities consistent with endometriotic lesions and that 31.1% of them were identified through histopathology as endometriosis. 20In our study, 468 patients presenting pelvic pain and anatomical abnormalities typical of endometriosis were evaluated, and the diagnosis of endometriosis was confirmed in 337 (72%).
Comparison between these studies reveals that, despite the validity of laparoscopy for diagnosing endometriosis, its use without histopathological confirmation gives rise to discrepancies in relation to the macroscopic findings. 10There is a need for an informal consensus regarding study design, and good surgical practice should be supported by detailed documentation in order to systematize the diagnosis. 10he findings from the present study allow us to conclude that endometriosis demonstrated a significant positive correlation with chronic pelvic pain, although not with dysmenorrhea or infertility.A greater frequency of peritoneal endometriosis was observed, in comparison with the involvement of other sites, such as the rectovaginal septum or ovaries.

CONCLUSIONS
The results obtained suggest that laparoscopy alone is of limited efficacy.Therefore, it needs to be combined with histopathological examination in order to achieve diagnostic confirmation of endometriosis.Among the cases of ovarian endometriosis, there was no difference in laterality.

Table 1 .
Laparoscopic endometriosis staging, by patient age and clinical manifestation *Age expressed as mean ± standard deviation; all other values expressed as number and percentage.Figure 1.Flowchart of the distribution of the patients selected for this study.

Table 3 .
Staging of endometriosis in the literature