Introduction. Extramural vascular invasion (EMVI) in rectal cancer is an predictor of poor prognosis of the patients affected by rectal cancer. EMVI is identifiable with MRI but its detection is more difficult in patients who underwent preoperative chemoradiotherapy (pCRT) for locally advanced rectal cancer (LARC). The aim of this study was to determine the accuracy of T2-weighted, diffusion weighted imaging (DWI) and contrast enhanced (CE) MRI sequences in the detection of EMVI in LARC patients after pCRT. Materials and methods. We retrospectively enrolled 103 patients (M=67, median age 66, range 44-83; F=36, median age 61, range 43-84) treated with pCRT for LARC that underwent restating contrast-enhanced pelvic MRI and surgery (total mesorectal excision or abdominoperineal resection). T2-weighted, DWI (b 50, 500 and 1000) and CE sequences were evaluated jointly by two radiologist experts in abdominal imaging that scored the probability of EMVI presence on each type of sequence using a grading score ranging from 0 (no evidence of EMVI) to 4 (strong radiological evidence of EMVI), blinded to clinical and histopathological data. Results from 0 to 2 were ranked as EMVI negative and from 3 to 4 as EMVI positive. ROC curves were drawn for each technique, using histopathological results as reference standard. Results. All the three techniques showed an area under the ROC curve (AUC) higher than that under the identity line (T2-weighted 0.610, 95% CI:0.509-0.704; DWI 0.729, 95% CI:0.633-0.812; CE 0.624, 95% CI:0.523-0.718), and DWI sequence showed a significantly higher AUC than that of both T2-weighted sequences (p=0.0494) and CE sequences (p=0.0315). Conclusions. DWI performed better than T2-wighted and CE sequences for the identification of EMVI in restaging of LARC patients, hence these sequences should be used as the mainstay to perform a correct evaluation of EMVI in this clinical setting.
Introduzione. L’invasione vascolare extramurale (EMVI) nel cancro del retto è un fattore prognostico negativo in pazienti affetti da carcinoma rettale. EMVI è identificabile con la risonanza magnetica ma la sua rilevazione è più difficile in pazienti che hanno subito radiochemioterapia preoperatoria (pCRT) per cancro del retto localmente avanzato (LARC). L’obbiettivo dello studio è di determinare l’accuratezza di determinare l’accuratezza di sequenze T2, diffusione (DWI) e con mezzo di contrasto (CE) nell’individuazione di EMVI in pazienti con LARC dopo pCRT. Materiali e metodi. Abbiamo selezionato retrospettivamente 103 pazienti (M=67, età mediana 66, range 44-83; F=36, età mediana 61, range 43-84) trattati con pCRT per LARC che hanno avuto una ristadiazione con risonanza magnetica con mezzo di contrasto e successivamente un’operazione chirurgica (escissione totale mesorettale o resezione addominoperineale). Sequenze T2, DWI (b 50, 500 e 1000) e CE sono state valutate congiuntamente da due radiologi esperti di imaging addominale, i quali hanno valutato la probabilità della presenza di EMVI per ogni sequenza usando un grading score da 0 (nessuna evidenza di EMVI) a 4 (forte evidenza radiologica di EMVI), senza conoscere i relativi dati clinici e istopatologici. Risultati da 0 a 2 sono stati classificati come EMVI negativi e da 3 a 4 come EMVI positivi. Sono state tracciate curve ROC per ogni tecnica, usando risultati istopatologici come standard di riferimento. Risultati. Tutte e tre le tecniche in esame hanno mostrato una area sotto la curva ROC (AUC) maggiore che sotto la linea di riferimento (T2 0.610, 95% CI:0.509-0.704; DWI 0.729, 95% CI:0.633-0.812; CE 0.624, 95% CI:0.523-0.718), e le sequenze DWI hanno mostrato una AUC significativamente maggiore sia di quelle T2 (p=0.0494) che CE(p=0.0315). Conclusioni. Le sequenze DWI hanno dato migliori risultati di quelle T2 e CE nell’identificazione di EMVI nel restaging di pazienti con LARC, quindi queste sequenze dovrebbero essere utilizzate per ottenere una corretta valutazione di EMVI in questo contesto clinico.
La risonanza magnetica nella valutazione dell’infiltrazione venosa extramurale (EMVI) nel cancro del retto post radio-chemioterapia: confronto tra sequenze T2, DWI e T1 con mezzo di contrasto.
DEL NEGRO, MARCO
2021/2022
Abstract
Introduction. Extramural vascular invasion (EMVI) in rectal cancer is an predictor of poor prognosis of the patients affected by rectal cancer. EMVI is identifiable with MRI but its detection is more difficult in patients who underwent preoperative chemoradiotherapy (pCRT) for locally advanced rectal cancer (LARC). The aim of this study was to determine the accuracy of T2-weighted, diffusion weighted imaging (DWI) and contrast enhanced (CE) MRI sequences in the detection of EMVI in LARC patients after pCRT. Materials and methods. We retrospectively enrolled 103 patients (M=67, median age 66, range 44-83; F=36, median age 61, range 43-84) treated with pCRT for LARC that underwent restating contrast-enhanced pelvic MRI and surgery (total mesorectal excision or abdominoperineal resection). T2-weighted, DWI (b 50, 500 and 1000) and CE sequences were evaluated jointly by two radiologist experts in abdominal imaging that scored the probability of EMVI presence on each type of sequence using a grading score ranging from 0 (no evidence of EMVI) to 4 (strong radiological evidence of EMVI), blinded to clinical and histopathological data. Results from 0 to 2 were ranked as EMVI negative and from 3 to 4 as EMVI positive. ROC curves were drawn for each technique, using histopathological results as reference standard. Results. All the three techniques showed an area under the ROC curve (AUC) higher than that under the identity line (T2-weighted 0.610, 95% CI:0.509-0.704; DWI 0.729, 95% CI:0.633-0.812; CE 0.624, 95% CI:0.523-0.718), and DWI sequence showed a significantly higher AUC than that of both T2-weighted sequences (p=0.0494) and CE sequences (p=0.0315). Conclusions. DWI performed better than T2-wighted and CE sequences for the identification of EMVI in restaging of LARC patients, hence these sequences should be used as the mainstay to perform a correct evaluation of EMVI in this clinical setting.File | Dimensione | Formato | |
---|---|---|---|
Tesi EMVI (1).pdf
accesso aperto
Dimensione
2.54 MB
Formato
Adobe PDF
|
2.54 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/30533