The Adnexal Mass : Benign or Malignant ? Evaluation of Risk of Malignancy Index

Background: Screening progmmme for ovarian malignancy is an international public demand. Ojective: To validate a risk of malignancy index (RMI) incorporating serum CA125, abdominal ultrasound findings and menopausal status for preoperative diagnosis of ovarian malignancy among patient with adnexal mass. Methods: This prospective observational study was conducted on 57 patients having adnexal mass admitted consecutively for surgical exploration in BSMMU from January 2000 to March 2001. S. CAl25 assay & abdominal ultrasonography was done within 10 days preoperatively.Ultrasoundscore(0, 1&3 )using5 sonographicfeatures&menopausalscore(1 &3)were calculated. RMI was detected by the formula "RMI = Ultrasound score E Menopausal score I S. CA125 level" . Definitive diagnosis was based on histopathological examination. Sensitivity, specificity, positive and negative predictive values of S.CA 125 , ultrasound score & menopausal score individually and in the form of RMI to diagnose Ovarian malignancy preoperatively were detected . For statistical analysis Chi square & student's 't' test were used. Results: Sensitivity , specificity , positive and negative predictive values of RMI at the cut-offvalue of 200 wereTSYo ,88yo ,58% and 95% respectively. Conclusion: RMI was more accurate than any individual criterion in diagnosing Ovarian malignancy . This scoring system can be introduced easily into clinical practice to facilitate selection ofpatient for appropiate surgery at proper centre by skilled surgeon .


Introduction:
The management of a woman having adnexal mass is a problem frequently faced by the Gynecologists.It may represerrt a wide and varied differential diagnosis including both benign and malignant gynecological conditions and also some non-gynecological conditions.lIn a clinically diagnosed adnexal mass traditional strategy for establishing final diagnosis has been to perform art exploratory laparotomy. 2   Ovarian malignancy is an important adnexal pathology.It shows the highest mortality among the gynecological malignarLcy and has become the greatest challenge in Address for Correspondence: Dr. Shamsun Nahar Associate Professor, Dept. of OB-GYN, Central Medical College, Paduar Bazar, Bishwa Road Comilla, Bangladesh.Mobile: 0 I 7 1 65 51,941, E-mail: drsnl 0 1 @gmail.comIBSMMU J 201 3 ; 6 (2) : ]27-1 34J gynecological oncology throughout the globe including Bangladesh.
In recent years, there has been developed considerable international public demand for ovarian malignancy screening.Multiple studies have concluded the in-effectiveness of the screening programme due to the absence of pre-cancerous condition and low prevalence of ovarian malignancy among the population .3   The prognosis of ovarian malignancy may be greatly influenced by appropriate first line surgery.The initial laparotomy is importarrt for accurate staging and cytore- ductive surgery.aStaging of neoplastic disease has assumed great importance in selection of the best form of therapy for the patient.sThe aim of cytoreductive surgery is to achieve optimal residue and the amount of residual r27 disease after primary surgery is the most important prognostic factor .6'7Moreover, chemotherapy and radio- therapy are likely to be effective in optimal residual disease.
However, in-spite of known benefit of meticulous comprehensive surgical staging and cytoreductive surgery many women do not receive approprrate first line surgery.This is because of difficulty in making an acsxate pre-operative diagnosis of ovarian malignancy as well as skilled and lengthy surgery of cytoreduction .In practice, the diagnosis of ovarian malign aficy is often difficult to make pre-operatively and inappropriate & inadequate surgical exploration by junior or inexperienced surgeon at improper centre is not uncofirmon Therefore, sensitive and specific methods for diagnosis of ovarian malignancy are required which may influence the selection of the institution and also the seniority and expertise of surgeon.This would provid e a rational basis for refercal before diagnostic laparotomy.8,eUltrasound reflects the gross morphological characteris- tics of tumour , not the histology.Evidence of solidity , mixed echogenicity , multiloculation , thick/irregular wall and septations, papillary projections, bilaterality, larger size , ascites & evidence of metastasis are suggestive of malignancy."'t'To improve ultrasonic diagnostic accutacy in ovarian mass , many scoring systems have been developed for objective evaluation of changes in many morphological parameters .12 The tumour marker, CAl25 mostly elevates in epithelial ovarian malignancy.Its normal value is up to 35 U/ml .The diagnostic accuracy of CA I25 assay is sufficiently high to be of clinical value to discriminate benign from malignant ovarian conditions.t3'r4s.CAl25 also rises in other pelvic malignancies and various benign conditions.

It may or may not rise in non-epithelial ovarian malignacy.ls
The risk of malignancy index (RMI) is a simple but novel scoring system introduced by Jacobs & his associates in 1990 uttLizing 3 diagnostic criteriaultrasound findings , serum CA 125 level and menopausal status of the patient .These researchers formulated as, RMI -U x M x S CAl25 (u: ultrasound score , Mmenopausal score and t28 S. CAl25absolute value of serum CAl25).They suggested that to distinguish benign and malignant adnexal mass the RMI is superior to 3 findings in isolation and a clinician would expect to classify correctly approximately 90% of patient with benign or malignant adnexal mass using the RMI .RMI utilizes currently available non-invasive tests , which can be applicable in clinical practice and would provide rational basis for specialist referral of patient with malignancy prior to surgery .tuThis study was conducted to validate the role of RMI to discriminate malignant from benign ovarian conditions in our population .Proper permission to conduct the study was taken from the chairmen of the concerned Departments after approval of the protocol .Verbal consent from each patient was obtained after informing about the nature and aim of the study .
Data was collected in data collection instrument which was a pre-designed data sheet of mixed form and questionnaire type. 2 methods were followed for data collection by interview and by some investigations Measurement of outcome variables were serum CA 125 , ultrasound score , menopausal score and RMI .
In interview method, the socio-demographic characteris- tics and clinical informations were recorded by interview.
In investigative method,2 ml of peripheral venous blood sample was obtained with full aseptic precaution by disposable syringe from each patient within 10 days preoperatively.Blood was drawn for the single time from each patient and was poured into a clean glass test fube.
Then the sample was allowed to clot at room temperafure and the serum was separated.S.CA 125 was determined by the technician using commercial kit (IMX, Abbott labora- Ultrasound scan of whole abdomen was performed by the Sonologist of Radiology and Imaging Department using 3.5 MHz transabdominal transducer in full bladder technique within 10 days preoperatively Ultrasonic features noted were solidity , multilocularity ,bllaterality , ascites and evidence of intra-abdominal metastasis Ultrasound score (U) was calculated for each patient as followsscore of 0, if none of the feafures was present, score of 1 if one of the feafures was present and score of 3,if > 2 ofthe features were present .
Post menopausal stafus was defined as )1 year of amenor- rhoea or ) 50 years of d$e, if previously had undergone hysterectomy.Pre-menopausal stafus was considered as the state outside the aforementioned criteria.Menopausal score (M) was scored as 1 if pre-menopausal and 3 if postmenopausal.
RMI score was calculated using the formula proposed by Jacobs et al. (1990) : RMI -U: X M X serum CAI25 where, U & M were ultrasound and menopausal score as defined above .
Immediately after surgery the resected specimen was sent to the Department of Pathology preserving in formalin solution.Histopathological examination was performed by the Pathologist .Definitive diagnosis was based on histopathological finding.In one case of illegal ectopic pregnancy, the surgical diagnosis was assumed to be correct as histopathological examination was not performed on patient's request.
Data was analyzed using SPSS program of computer with the help of statistician.SD was taken as a measure of variation and frequency of data was expressed as mean t SD.Student's 't' test was used to test the differences in distribution of age , pffit$, serum CAl25, ultrasound score and RMI among the patients with benign and malignant adnexal mass.Chi-square test was applied in testing differences in distribution of socio-economic and menopausal stafus in the same groups .Level of significance was expressed as p value and values showing < .05 was considered as significant to assess the performance of serum CAl25, ultrasound score , menopausal score and RMI score to discriminate malignant from benign adnexal mass.The sensitivity, specifi city, positive predictive values(PPv) and negative predictive values (NPV) were calculated at different cut-off values.

Results:
Though 57 patients were initially enrolled for the study, 3 were dropped out due to refusal of surgery by patient herself in 1 and decision of giving primary chemotherapy in the rest 2. Later on another 5 were excluded from the sfudy where 4 were proved to be non-adnexal malignancy (uterine leiomyosarcoma -2, frbrosarcoma -l, ffidmalignant fibrous histocytoma 1) and 1 to be uterine leiomyoma on histopathological examination .Thus, 49 patients comprised the study of population .Out of 49 patients ,9 (18%) had ovarian malignancy , where majority(3) were serous cystadeno-carcinoma.On the other hand , 40 (82%) benign adnexal mass were detected.Majority (9) were functional cyst followed by serous cystadenoma (7) , ovarian endometrioma (6) , dermoid cyst (5) and the rest were other benign pelvic conditions.
Data regarding the patient characteristics and relevant informations have been presented in the following differ-    Ovarian malignancy is a major cancer killer in gynaeco- logical oncology throughout the globe including Bangladesh.Absence of pre-cancerous condition, silent nafure or non-specific symptoms at earlier stages and low preva- lence are responsible factors for the ineffectiveness of screening programme .Detection at earlier stages can at least reduce the cancer mortality by giving opportunity of primary cytoreductive surgery with accurate staging.But the diagnosis of this adnexal malignancy is difficult due to existence of wide varieties of adnexal pathologies includirrg both the gynaecological and non-gynaecological.A reliable diagnostic method is very much demanding to reduce the problem.Because only then appropriate t32 surgery by experienced gynaecologist at institution with adequate facilities locally or after referral may be possi- ble.
RMI is reliable diagnostic scoring system to discriminate malignant from benign adnexal mass.To test the validity of RMI this study has been performed.Therefore , the sensitivity , specificity and predictive values of serum CAI25, ultrasound score , menopausal stafus and RMI to discriminate malignant from benign adnexal mass were determined and compared.
In present study, optimal cut-off value of serum CAl25 was >T20 U/ml in contrast to other studies, where the cut-off values ranges from 30-70 U/ml.Both sensitivity and specificity of ultrasound score in present study were in agreement with the previous studies.
In comparison to Jacobs et al.e sensitivity was higher (78% vs 7I%) and specificity was lower (50% vs 83 oh) but the variations were acceptable.In comparison to Davies et a1.8 specificity was higher (80% vs 7 4Yo), but sensitivity was slightly lower (78% vs \7Yo)in our population , that was also acceptable.Moreover, like the previous researchers, it was found that along with malignancy, positive ultrasound finding was also associated with benign mass.
Performance of menopausal stafus was found to be less

Conclusion:
RMI is a simple , non-invasive and reliable diagnostic method , which can be useful in clinical practice .Ultra- sonic features used in RMI can be easily identified by sonologist with trans-abdominal transducers and now a days ultrasound is widely available and less expensive in Bangladesh.
Routine use of RMI at institution with proper facilities in cases of ovarian pathology could provide early and reliable prediction of ovananmalignancy with consequent appropriate surgical approach.This includes preoperative consultation with general surgeon , necessary preoperative preparation , planning of extendable midline incision and adequate surgery.Moreover , detection of ovarian malignancy at centre with lack of adequate facilities of appropriate surgical approach may provide a basis for referral to proper instifution , to experienced gynaecolo- gist or to the gynaecologist wtih special interest in oncol- ogy Such a study on larger population may further improve diagnostic precision of RMI and that is certainly demanding .
Methods: This was a prospective observational study.The study was conducted during the period of January 2000 to March 2001 in the Department of obstetrics and Gynecology of BSMMU, Dhaka, with the full co-operation of Dep art- ment of ClinicalPathology , Department of Pathology & Department of Radiology & Imaging of BSMMU, Dhaka.The girls and women admitted consecutively in the Department of Obstetrics and Gynecology for the surgical exploration of adnexal mass and having informed willirrgness in participation into the study (n-57) were enrolled for the study.Patient with histo-pathological report show- ing non-adnexal mass was excluded .
tories , USA) by Micropartical Enzyme Immuno Assay ( MEIA) technique at the Department of clinical pathology, BSMMU, Dhaka.

Table - I
Distribution of Serum CA 125 levels in Patient with Adnexal Mass fu -a9) ? Evaluation of of Malienancv Index '

Table - II
Distribution of Age, Parity, Socio -economic status among Patients with Adnexal Mass ( n-49)
sensitive in present study (22% vs 79o/o and 59% in the study of Jacobs et al.e and Davies et a1.8 respectively) of same results at both cut-off values were possibly due to contribution of RMI score of 200 and 25A by the same group of patient as sample size was small.thatscore of RMI.On the other hand , positive predic- tive value of 5 8o/o means that 58% of adnexal mass will be ovarian malignancy at that score.Negative predictive value of 95% means that 95% of adnexal mass will be benign atthat score.In this study, RMI sensitivity (78%) was higher than the sensitivity (71%) found by Tingulstad et a1.2, but it was lower than that ofDavieset al.s and Jacobs et aL.e which was 87% and 85% respectively.The specificity was almost similar to Davies et al.8,but slightly lower than other two studies.The negative predictive value was higher (95%) but positive predictive value was lower (58%) than those of other reseachers.This study also showed that RMI could discriminate 83% of malig- nant from benign mass.But this discrimination was 9IoA, 88% and 84% in the studies of Jacobs et &l.e , Davies et &1.8, and Tingulstad et al.2 respectively.The overall diagnostic accuracy of RMI was in agreement with the previous studies, but slight variation may be due to smaller sample size.
finding at