Diagnostic performance of magnetic resonance imaging (MRI) in patients with operable cervical cancer: a retrospective analysis
- Author:
Rohini KULKARNI
1
;
Jay THAKKAR
;
Meenakshi THAKUR
;
Kedar DEODHAR
;
Pabashi PODDAR
;
Amita MAHESHWARI
Author Information
- Publication Type:Young Doctor Session
- From:Journal of Gynecologic Oncology 2022;33(S1):S14-
- CountryRepublic of Korea
- Language:English
-
Abstract:
Objective:Magnetic resonance imaging (MRI) has become an integral part of pre-operative work up of patients with early cervical cancer. It has superior soft tissue delineation and therefore useful in evaluation of parametrial, cervical stromal, and regional lymph node involvement. In 2018, FIGO also incorporated radiology into staging of carcinoma cervix, with upstaging to stage III based on radiological involvement of lymph nodes. However, there is scant data available from low-and middle-income countries’ regarding the diagnostic performance of MRI in patients with operable cervical cancer. Therefore, this retrospective study was planned.
Methods:A retrospective review of electronic medical records of all patients who underwent radical surgery for carcinoma cervix between 2013–2018 at our institute and had a pre-operative MRI. Diagnostic performance of MRI with respect to assessment of tumor size, cervical stromal invasion, parametrial, vaginal and lymph nodal involvement was evaluated using sensitivity, specificity, positive and negative predictive values (PPV, NPV), positive and negative likelihood ratios (PLR, NLR) and false negative and false positive rates (FNR, FPR). Final histopathological report was considered as the gold standard.
Results:Total 79 patients who underwent surgery for early cervical cancer and had pre-operative MRI, were included in the study. For parametrial involvement MRI showed high sensitivity (100%) with moderate specificity (71.8%) but high FPR (28.2%) and low PPV (4.7%). For cervical stromal involvement MRI revealed moderate sensitivity (65.6%) with high specificity (100%) but high FNR (34.4%) and low NPV (15.4%). Tumor size assessment on MRI had 85.5% concordance with tumor size on gross pathological examination. For pelvic lymph node and vaginal involvements, MRI showed poor sensitivity; 14.3%, 25%, respectively.
Conclusion:In patients with operable cervical cancer MRI pelvis cannot replace a well conducted clinical examination to assess loco-regional disease spread. Clinical examination still plays a major role in triaging patients for radical surgery or radical radiotherapy. We should be cognizant of the effects of stage migration of cervical cancer based on radiology with resultant possibility of change of management.
