ABSTRACT
OBJECTIVE: The purpose of this study is to investigate the diagnostic accuracy of 3.0Tesla (3T) diffusion-weighted magnetic resonance imaging (MRI) in addition to conventional MRI for the detection of lymphadenopathy in patients with early stage cervical cancer compared to histopathological evaluation of the systematically removed pelvic lymph nodes as reference standard. METHODS: 68 fédération internationale de gynécologie obstétrique (FIGO) stage Ia2 to IIb cervical cancer patients were included. Sensitivity and specificity rates for two experienced observers were computed for the detection of lymphatic metastasis. Reproducibility of conventional MRI was tested by kappa statistics. The variables included in the analysis were: size of the long axis, short axis, ratio short to long axis and apparent diffusion coefficient (ADC). RESULTS: Nine patients had 15 positive pelvic nodes at histopathological examination. The sensitivity and specificity of lymphatic metastasis detection by predefined conventional MRI characteristics was 33% (95% Confidence Interval (CI) 3-64) and 83% (95% CI 74-93) on patient level, and 33% (95% CI 7-60) and 97% (95% CI 95-99) on regional level respectively for observer 1. For observer 2 the sensitivity was 33% (95% CI 3-64) and the specificity 93% (95% CI 87-100) on patient level, and 25% (95% CI 1-50) and 98% (95% CI 97-100) on regional level, respectively. The kappa-value for reproducibility of metastasis detection on regional level was 0.50. The short axis diameter showed the highest diagnostic accuracy (area under the curve (AUC)=0.81 95% CI 0.70-0.91); ADC did not improve diagnostic accuracy (AUC=0.83 95% CI 0.73-0.93). CONCLUSIONS: Diffusion-weighted MRI did not result in additional diagnostic value compared to conventional MRI.
Subject(s)
Adenocarcinoma/secondary , Carcinoma, Squamous Cell/secondary , Diffusion Magnetic Resonance Imaging/methods , Lymphatic Metastasis/diagnosis , Neoplasm Staging/methods , Uterine Cervical Neoplasms/pathology , Adenocarcinoma/surgery , Adenocarcinoma, Clear Cell/secondary , Adenocarcinoma, Clear Cell/surgery , Adult , Aged , Carcinoma, Adenosquamous/secondary , Carcinoma, Adenosquamous/surgery , Carcinoma, Squamous Cell/surgery , Confidence Intervals , Female , Humans , Lymph Node Excision , Middle Aged , Observer Variation , Pelvis , Prospective Studies , ROC Curve , Reproducibility of Results , Sensitivity and Specificity , Uterine Cervical Neoplasms/surgery , Young AdultABSTRACT
OBJECTIVE: Pelvic lymphadenectomy is considered the gold standard to diagnose and possibly treat lymphatic metastases in gynaecological cancer patients. The aim of this study is to evaluate whether all presurgical MRI detected lymph nodes were removed during the systematic pelvic lymph node dissection (PLND) in cervical cancer patients. METHODS: 21 consecutive cervical cancer patients who were scheduled to undergo a PLND were evaluated by a MRI scan prior to surgery and 6 weeks afterwards. All patients had tumour metastasis negative lymph nodes at histopathological examination. RESULTS: On average, 10 pelvic lymph nodes (range 5-17) per patient were detected by presurgical MRI. Postsurgical MRI scans showed that on average 1 (range 0-3) pelvic node per patient was not removed by surgery. In total, 14% of the presurgical MR detected nodes were not removed by surgery (31/225). Approximately half of all lymph nodes that remained after surgery were located in the obturator region. In spite of the remaining nodes, surgery and histopathological examination did detect more nodes than MRI: on average 21 lymph nodes per patient (range 9-59) were removed. Another 2 lymph nodes (range 0-6 per patient) were judged to be newly developed after surgery. CONCLUSION: Although surgery was able to remove many more lymph nodes than those detected by presurgical MRI, 14% of presurgical MRI detected lymph nodes were not removed by PLND. The value of MRI prior to surgery for the detection of pathological lymph nodes is a subject of further research.
