Evaluation of the risk malignancy index diagnostic value in patients with adnexal masses

Background/Aim. Ovarian cancer is the leading cause of death from gynecologic malignancies. Risk of malignancy index (RMI) is recommended in assessment of patients with adnexal masses. The aim of this study was to verify the effectiveness of the RMI in the discrimination between benign lesions and malignant adnexal masses in clinical practice. Methods. Ultrasounds were performed for all the patients and menopausal status, CA125 level and calculated RMI were defined. All the patients were divided into 3 groups depending on RMI (< 25, 25–200, > 200). After operations all adnexal masses were analyzed histopathologically (HP) and then sensitivity, specificity and predictive value of RMI were calculated. Results. Out of a total of 81 patients involved benign tumor had 51 (62.96%) and malignant 30 (37.04%) of the patients. The average value of CA125 in the group of patients with benign adnexal masses was 68.3 U/mL and in the group of patients with malignant adnexal masses it was 581.95 U/mL. In the group of patients with benign adnexal masses the average RMI was 284.9 and in the group of patients with malignant adnexal masses RMI was 469.2. All the results showed a positive correlation between both HP categories and RMI categories. The more malignant HP result produced higher RMI and the cut off value was RMI = 200. Sensitivity of RMI was 83.33%, specificity was 94.12%, positive predictive value was 89.29% and negative predictive value was 90.57%. Conclusion. Our study showed that RMI is very reliable in differentiation benign from malignant adnexal masses.


Introduction
Ovarian cancer is a leading cause of death among gynecological malignant tumors.Treatment efficiency in patients with ovarian cancers could be increased by standardization of preoperative evaluation.Jacobs et al. 1 have introduced risk of malignancy index (RMI) and that was the first diagnostic model which has combined demographic, sonographic and biochemical parameters for investigating patients with adnexal masses.RMI was modified by Tingulstad et al. 2,3 for the first time in 1996 (RMI2) and for the second time in 1999 (RMI3).These three versions of RMI were assessed in many prospective and retrospective clinical studies.Yamamoto et al. 4 made even RMI4, but its validity is due to be confirmed in future studies.The RMI value of 200 has been proven to be the best for distinction of benign from malignant adnexal masses, with the high level of sensitivity (51%-90%) and specificity (51%-97%) 5,6 .
The aim of the study was to evaluate the risk of malignancy index efficiency in differentiation benign from malignant adnexal tumors in clinical practice.

Methods
The study involved all patients with adnexal tumors who were hospitalized at the Institute of Gynecology and Obstetrics, Clinical Center of Serbia during the first six months of 2010.Ultrasonographic examination of pelvic organs was performed, menopausal status and level of cancer antigen 125 (CA125) were assessed and finally RMI was calculated for all the patients.RMI was calculated using the formula: RMI = U × M × CA125.In the formula U represents the ultrasound index.Multilocular and bilateral tumors, the presence of solid parts in tumor, metastasis and ascites are marked with one point each.The sum of these points, are scored so that in the formula U 0 = 0 points, U 1 = 1 points, U 2-5 = 3 points.In the formula M represents menopausal status (1 for premenopausal and 3 for postmenopausal women).Values of CA125 are taken in all patients prior to surgery.The patients were divided into three

Results
The study involved 81 patients, out of which 51 (62.96%) were still in the reproductive period, while 30 (37.04%) were in the menopause.Out of all analyzed patients benign tumor had 51 (62.96%) and malignancy 30 (37.04%) women.In the group of premenopausal women benign tumor was registered in 38 (74.51%) and malignancy in 13 (25.5%)patients.In the group of postmenopausal women benign tumor was registered in 13 (43.33%)and malignant in 17 (56.67%)patients.From all benign tumors ovarian cyst was the most frequent, while from the malignant tumors adenocarcinoma was found to be the most usual (Figure 1).The HP diagnoses of benign adnexal masses are shown in Table 1while Table 2 shows HP diagnoses of malignancy.

Fig. 1 -Adnexal tumors' histological findings of all the investigated patients 1-benign ovarian cyst; 2 -adenocarcinoma; 3 -endometriosis; 4 -other
The average age of the investigated patients with benign tumors was 38.3 years, whereas, the average age of the investigated patients with malignant tumors was 51.45 years, which was significantly different (p = 0.001, p < 0.05).The average value of CA125 in the group of patients with benign adnexal tumors was 68.3 U/mL, while in the group of those with malignant tumors it was 581.95 U/mL.However, there were no significant differences in CA125 values concerning categories of HP findings (benign, malignant) (p = 0.342, p > 0.05).These data are shown in the Table 3.
Multilocular changes with solid tumor components were registered in 5 (6.71%) of the patients who were all premenopausal.Only multilocular changes were registered in 24 (29.63%) of the patients, 14 pre-and 10 postmenopausal.Tumor had only solid parts in 18 (22.22%)patients, 8 pre-and 10 postmenopausal.Metastases were registered in 11 (13.58%) of the patients, out of which 6 were postmenopausal and ascites was registered in 13 (16.05%) of the patients, out of which only 4 were premenopausal.Tumors were present on both ovaries in 13 (16.05%)women, 8 pre-and 5 postmenopausal.
Using the given formula, RMI was calculated for each patient.In the group of patients with benign adnexal masses average RMI was 284.9 (min = 0, max = 2,889), while in the group with malignant changes it was 469.2 (min = 0, max = 122,607).Relation between RMI, HP findings and menopausal status is shown in the  Data shown in Table 4 indicate a positive correlation between both categories of HP findings (benign, malignant) and categories of RMI in all the examined women.This correlation shows that with an increase of RMI values, HP findings are more frequently malignant and that the significant cut off value of RMI is 200 (Ro xy = 0.428; df = 78; p = 0.000; p < 0.01).
There were 3 patients with benign tumors and RMI higher than 200.Those were false positive cases.On the other hand, 5 patients with malignant tumors had RMI less than 200.Those were false negative cases.In the group of premenopausal women false positive results were 2 and false negative 1, while in the group of postmenopausal women there was 1 false positive result and 4 false negative results.The most frequently misdiagnosed were borderline tumors (3 out of 8) or low differentiated tumors (1 out of 4) as well as endometriotic cysts (2 out of 13) and benign mucinous cystadenomas (2 out of 8).Therefore, RMI sensitivity is 83.33%, specificity 94.12%, while the positive predictive value is 89.29% and the negative predictive value is 90.57%.

Discussion
The investigated patients from our population more frequently had benign adnexal masses, which were more frequent in premenopausal women.On the other hand, patients who had malignant tumors were significantly older than those with benign masses and therefore, in postmenopausal group malignant tumors were the most usual.
Our study shows that the risk for malignancy of adnexal tumor based on RMI statistically significantly corresponds with postoperatively obtained HP findings.Among the patients we investigated, there were a lot of cases of benign tumors and high CA125 values.It is well known that the levels of CA125 in premenopausal women can vary regarding the phase of menstrual cycle and that the peak values are during menstrual bleeding 7 .Also, entities such as endometriosis and pelvic inflammatory disease can be the reason for high blood levels of CA125 8 .In our study, out of all benign tumors 25.49% were endometriotic cysts and that lead to the increase of CA125 levels in the group of women with benign adnexal masses.
On the other hand, borderline tumors were sorted in the group of malignant tumors and they were present in 26.66% of the patients with malignant tumors.That might have influenced a bit lower the average values of CA125 in the group of patients with malignant adnexal masses.Nevertheless, according to the data from the literature, values of CA125 are still of high importance in evaluation of adnexal masses, and therefore must not be disregarded [9][10][11][12][13] .Concerning the above mentioned, RMI has shown to be a significant factor in routine clinical practice and therefore we evaluated its efficiency very precisely.
Based on the RMI analysis that we have done, it can be concluded that all aspects of RMI efficiency are satisfactory (high sensitivity, specificity, positive predictive value and negative predictive value), which can indicate significance of RMI in discrimination between benign and malignant adnexal tumors.The results of our study are in correlation with the data from the literature 14,15 .
However, when we compare the investigated parameters, it can be seen that higher values are achieved for specificity and negative predictive value than for sensitivity and positive predictive value.Good specificity means that test will always disregard negative findings, i.e. in our case that benign tumors will always have low RMI values.That is also confirmed by high negative predictive values.Tests with better specificity are more useful in early diagnostic phases and in population screening and RMI calculation and evaluation is used precisely in those cases.If we compare pre-and postmenopausal women, RMI is more sensitive and specific for the premenopausal group.On the other hand, in both groups as well as in all the patients examined together, a higher specificity was observed.On the contrary, a positive predictive value had maximal values in the group of postmenopausal women, which indicates that high RMI of the patients from this group will usually (92.86% cases) really be a sign of malignant adnexal tumor.A bit lower sensitivity indicates that final diagnostics can not be based only on RMI, as there is still a chance that this method does not diagnose the presence of malignant tumor.

Conclusion
Our study confirms that RMI is very reliable for differentiation between benign and malignant adnexal masses.

R E F E R E N C E S
1. Jacobs I, Oram D, Fairbanks J, Turner J, Frost C, Grudzinskas JG.
A risk of malignancy index incorporating CA 125, ultrasound and menopausal status for the accurate preoperative diagnosis of ovarian cancer.Br J Obstet Gynaecol 1990; 97(10): 922-9.
groups according to the RMI values (low risk < 25, intermediate risk 25-200, high risk > 200).After surgery, histopathological (HP) findings of excised tumors were analyzed in order to determine the final diagnosis and the stage the disease.Finally, based on the standard formulas, sensitivity [(true positive / true positive + false negative) × 100], specificity [(true negative / true negative + false positive) × 100], positive [(true positive / true positive + false positive) × 100] and negative [(true negative / true negative + false negative) × 100] predictive values of RMI, as a test which could indicate tumor malignancy, were calculated.For statistical analysis of the achieved data standard methods of data description (mean values, variability measures such as confidence interval and standard deviation) as well as the tests for examining the significance of correlations and differences between achieved data (chi-square test, Fisher T-test and Spearman Rank correlation) were used.For statistical data analysis SPSS 15 computer program was used.