Objective
The purpose of this study was to assess the prevalence of endometriosis in women with premenstrual spotting and to determine the predictive value of this symptom in the diagnosis of endometriosis.
Study Design
We conducted a retrospective cohort study of 80 consecutive women who presented to the infertility clinic for evaluation and who subsequently underwent laparoscopic assessment for infertility with or without pelvic pain. Our main outcome measure was the presence or absence of histologically confirmed endometriosis in women with and without premenstrual spotting.
Results
Endometriosis was significantly more prevalent in subfertile women who reported premenstrual spotting for ≥2 days relative to women without this symptom (89% [34/38 women] vs 26% [11/42 women]; P < .0001). Multinomial logistic regression analysis demonstrated the presence of premenstrual spotting for ≥2 days to be associated significantly with the presence of endometriosis (odds ratio, 16; 95% confidence interval, 3.9–65.4; P < .01) and red vesicular lesion type (odds ratio, 52.6; 95% confidence interval, 8.6–323.1; P < .001).
Conclusion
In this cohort of women with infertility, premenstrual spotting of ≥2 days was associated strongly with histologically confirmed endometriosis and a better predictor than dysmenorrhea or dyspareunia of finding endometriosis at laparoscopy. Premenstrual spotting of at least 2 days was also associated strongly with both higher stage disease and the red vesicular peritoneal endometriosis phenotype.
Endometriosis is a debilitating gynecologic condition classically defined as the presence of endometrial glands and stroma in ectopic locations. Affecting 6-10% of reproductive-aged women, endometriosis is associated with pain and infertility. Currently, this disorder can be reliably diagnosed only by visual inspection of the abdomen and pelvis with histologic confirmation of biopsied lesions. Although laparoscopy is a relatively safe minimally invasive approach, the procedure poses surgical risk to patients. In addition, laparoscopy is expensive in terms of both procedural cost and convalescence-associated absence from work. Despite the association of endometriosis with well-characterized pain symptoms, nearly one-half of women with chronic pelvic pain are found to have no identifiable disease at laparoscopy. These considerations highlight the importance of research toward minimizing negative laparoscopies with accurate preoperative identification of patients with endometriosis.
The requirement for invasive surgery for the diagnosis of peritoneal implants contributes to an average latency of 6.7 years from onset of symptoms to definitive diagnosis. Delayed diagnosis and treatment may have significant consequences, as endometriosis is more advanced in women whose surgical evaluation is delayed, suggesting progression of disease over time. Consequently, the discovery of a nonsurgical biomarker for the diagnosis of endometriosis is considered a main priority and an area of active research.
To date, an accurate, noninvasive diagnostic test for endometriosis is unavailable, and decisions to perform laparoscopy are based on clinical judgment with the use of medical history, pelvic examination, and ultrasound scanning. In women with an endometrioma, ultrasound scanning is very accurate in the preoperative prediction of endometriosis but rather limited when only peritoneal disease is present. A previous study reported only 38% of cases of nonovarian endometriosis were predicted accurately by the combination of symptoms, pelvic examination, and ultrasound scanning. Of note, this study did not include irregular bleeding of any type in the preoperative assessment.
Several groups have reported an association between shorter menstrual cycle length and endometriosis. However, these studies do not comment on whether premenstrual spotting abbreviated the interval. Compared with women with luteal phase defect, a higher prevalence of endometriosis was observed in women with premenstrual spotting of ≥3 days. Herein, we sought to determine whether the symptom of premenstrual spotting has predictive value in the diagnosis of endometriosis in women with infertility. If correlated, inclusion of this clinical symptom in the menstrual history may assist in the identification of the most appropriate surgical candidates.
Materials and Methods
This study was approved by the Institutional Review Board of the Madigan Healthcare System. The records of all consecutive women who underwent laparoscopy for infertility with or without pelvic pain from March 2009 to March 2011 at a single tertiary care center were reviewed. All women were reproductive age with regular menses in terms of cycle length and at least unilateral tubal patency at hysterosalpingogram or chromopertubation. The latter stipulation was maintained in view of abundant evidence that supported retrograde menstruation in the pathogenesis of endometriosis.
All laparoscopic procedures were performed by 1 of 2 surgeons (G.E.C., R.O.B.) who are experienced in the diagnosis and treatment of endometriosis. At laparoscopy, the surgeon surveyed the entire pelvis and upper abdomen. Biopsies were performed on suspected lesions, as per standard clinical practice, and read by a pathologist who is experienced in the histologic appearance of endometriosis. Any remaining endometriotic lesions were removed by either surgical resection or thermal cautery ablation. All patients had at least 1 biopsy specimen sent for histologic confirmation. The extent of endometriosis was staged according to the revised American Fertility Society (rAFS) classification system. Additionally, peritoneal implants were classified as either red vesicular or powder burn phenotype. Representative images of these lesional phenotypes are provided in the Figure .
The medical records of women who met the study inclusion criteria were abstracted for historic findings, ultrasound scan results, pelvic examination, infertility history, and indication for laparoscopy; the results were correlated with surgical findings. For purposes of assessing the relative accuracy of symptoms in the prediction of endometriosis, the medical records were reviewed specifically for the presence and duration of premenstrual spotting, dysmenorrhea, and dyspareunia.
Patient history at the intake visit was collected both by written questionnaire and by physician interview. Before the initial infertility consultation, patients completed a standard 3-page infertility questionnaire that was developed by the Reproductive Endocrinology and Infertility Division. The questionnaire was reviewed by the physician with the patient at the time the history is taken. Questions specifically related to endometriosis symptoms and menstrual irregularities were included in the general infertility questionnaire. Patients were asked specifically about dysmenorrhea (“Do you have severe cramping or pelvic pain with your periods”) and dyspareunia (“Do you have pain with intercourse”). Women who selected “Yes” on the questionnaire with corroborative documentation by the physician who obtained the verbal history at intake were considered to be affected with these symptoms. The intake survey specifically queried whether patients experienced “spotting before the onset of full menstrual flow.” Likewise, physicians specifically asked and documented response to the question, “Do you experience spotting before the onset of full menstrual flow?” For women who acknowledged premenstrual spotting, the duration of spotting was recorded. We defined premenstrual spotting as bleeding on the order of spotting before the onset of full menstrual flow. To eliminate confounding by normal variants in menstrual onset and other conditions that may result in brief premenstrual spotting, the symptom was considered significant only if the reported duration was at least 2 days.
The demographic parameters of age, gravidity, parity, and body mass index were compared between groups with the use of the Student t test. The chi-square statistic was used to calculate the significance of the association between endometriosis presence/absence and premenstrual spotting presence/absence. The sensitivity, specificity, positive and negative predictive values, and accuracy (percentage correct) were calculated for the symptoms of dysmenorrhea, dyspareunia, and premenstrual spotting. Relationships between each variable (premenstrual spotting, dysmenorrhea, dyspareunia, dyschezia, age, parity, and body mass index) and each outcome (presence/absence of endometriosis, rAFS stage, lesional phenotype) were indicated by the phi-coefficient; nomial/multinomial logistic regression was used to determine their odds ratio, both unadjusted and adjusted, in contributing to the outcome. Unweighted kappa analysis was performed to evaluate the variability between self-reported premenstrual spotting and histologic findings. Statistical analysis was performed using SPSS software (version 18; SPSS Inc, Chicago, IL). Probability values of < .05 were considered statistically significant.
Results
Of the 80 consecutive patients who met inclusion criteria, 38 women reported premenstrual spotting of ≥2 days, and 42 women denied premenstrual spotting on both intake questionnaire and during menstrual history. All patients initially were seen for infertility, which was the primary indication for surgery in all but 3 patients for whom chronic pelvic pain was the primary indication. However, these 3 patients also were seen for infertility evaluation and were maintained in the analysis. Women who reported ≥2 days of premenstrual spotting were older than those women without this history. Otherwise, there were no significant differences between the 2 groups with respect to gravidity, parity, or body mass index ( Table 1 ).
| Characteristics | Premenstrual spotting (n = 38) | No premenstrual spotting (n = 42) | P value a |
|---|---|---|---|
| Age, y b | 30.7 ± 0.8 (22–40) | 28.2 ± 0.8 (20–40) | .03 |
| Gravidity, n b | 0.6 ± 0.1 (0–3) | 0.8 ± 0.2 (0–8) | NS |
| Parity, n b | 0.2 ± 0.1 (0–1) | 0.3 ± 0.1 (0–3) | NS |
| Body mass index, kg/m 2 b | 24.6 ± 0.6 (19.5–33.0) | 25.7 ± 0.7 (19.0–35.6) | NS |
| Preoperative diagnosis, n | |||
| Unexplained | 30 | 21 | |
| Tubal factor | 2 | 17 | |
| Uterine septum | 4 | 1 | |
| Chronic pain | 2 | 3 |
a Student t test with significance, P < .05
In the group of women without premenstrual spotting, 26% of the women (11/42 women) were diagnosed with endometriosis at laparoscopy. Of note, all cases were staged as minimal (rAFS stage I) biopsy-proven disease ( Table 2 ). On the other hand, 89% of the women (34/38 women) with premenstrual spotting were found to have biopsy-proven endometriosis at laparoscopy (89% vs 26%; P < .0001). Furthermore, in 85% of these cases (29/34 women), advanced stage disease (defined as greater than rAFS stage I) was documented at surgery. In women with premenstrual spotting affected with endometriosis, 85% (23/27 women) had lesions of red vesicular type. In contrast, only 27% of affected women (3/11 women) without premenstrual spotting evidenced red vesicular lesions ( P < .001).
| Finding | Premenstrual spotting, n (%) | No premenstrual spotting, n (%) | P value a |
|---|---|---|---|
| No endometriosis | 4 (11) | 31 (74) | |
| Endometriosis | 34 (89) | 11 (26) | < .0001 |
| Stage | |||
| I | 5 (15) | 11 (100) | |
| II | 11 (32) | 0 | |
| III | 13 (38) | 0 | |
| IV | 5 (15) | 0 | |
| Lesion phenotype | < .001 | ||
| Red vesicular | 27 (79) | 3 (27) | |
| Powder burn | 7 (21) | 8 (73) |
Compared with presurgical symptoms of dysmenorrhea and dyspareunia, premenstrual spotting demonstrated the highest positive predictive value and a negative predictive value very near that of dysmenorrhea ( Table 3 ). Of the 3 symptoms, premenstrual spotting was the most accurate in correctly identifying women with and without endometriosis: 81% compared with 76% accuracy for dysmenorrhea and 58% accuracy for dyspareunia. Unweighted kappa analysis also demonstrated accuracy between self-reported spotting and histologic findings (κ = 0.63; 95% confidence interval [CI], 0.46–0.80). Only 6% of the biopsy specimens (3/48 women) were negative for endometriosis, which demonstrated high accuracy for biopsy determination at laparoscopy.
| Symptom | Sensitivity | Specificity | Positive predictive value | Negative predictive value | Percent correct (accuracy) a |
|---|---|---|---|---|---|
| Dysmenorrhea | 0.87 | 0.63 | 0.75 | 0.79 | 0.76 |
| Dyspareunia | 0.38 | 0.83 | 0.74 | 0.51 | 0.58 |
| Premenstrual spotting | 0.76 | 0.90 | 0.96 | 0.74 | 0.81 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
