Patients
This study included 86 patients (46 men,
40 women,
mean age 61,7±12,9 years) with a new diagnosis of rectal carcinoma established on site at the Kazakh State Institution of Oncology and Radiology,
between January 2015 and June 2018.
The study was approved by the institutional review board.
The inclusion criterion was a new diagnosis of path-proved rectal carcinoma below the sacral promontorium with no gender or age predilection.
MRI technique
MRI was performed using a 3.0-T system pre-operatively (GE Discovery MR750w,
USA),
utilizing a pelvic phased – array surface coil (Gem body coil 8ch.).
T2 – weighted turbo-spin echo (TSE) sequences in sagittal,
paracoronal,
and para-axial planes were used.
Slice thickness was 3 mm without gap,
field of view 26 cm,
matrix 320x320.
Image analysis
The MRI data were assessed by two radiologists with 5 and 15 years of experience in oncology imaging respectively.
MRI reporting criteria for T staging of rectal carcinoma were listed as standard and described elsewhere [5,6].
Briefly,
the tumor confined to only to submucosa is considered as T1,
with involvement of muscular layer but sparing perirectal fat (PF) as T2,
with PR invasion as T3,
and extension to other pelvic organs or distant metastases as T4.
Images were assessed for infiltration of the rectal wall layers,
perirectal fat and pelvic structures,
and for the evidence of extramural vascular invasion.
The mesorectal fascia was regarded as intact at a distance from the tumor margin or involved mesorectal lymph nodes> 2 mm; a possible invasion of 1-2 mm,
and is considered as an involved in the process at a distance ⩽1 mm [10,11].
Extramural vascular invasion was assessed as the part of the standard MR protocol [12-16].
EMVI was divided as the involvement of large vessels (more than 3 mm) and small vessels (less than 3 mm in diameter).
Histopathology evaluation
All patients underwent surgery; 71 out of total 86 patients underwent preoperative neoadjuvant therapy consisting of radiotherapy and\or chemotherapy.
In all 86 patients,
histological examination and pathological staging was performed on the surgical specimens through conventional pathology examination.
Pathological staging was carried out according to TNM criteria [3,4].
The pathologist with 20 years’ experience examined the specimen blinded to the preoperative radiological staging.
Statistical analysis
Correlative analysis of MRI and histopathological findings was performed.
The sensitivity,
specificity,
and accuracy were calculated [17,18].
The MRI data were assessed by two radiologists with 5 and 15 years of experience in oncology imaging respectively.
Interobserver agreement was assessed using Cohen’s Kappa coefficient.
Agreement was defined as substantial (0,78) for T staging,
and for extramural vascular invasion assessment (0,63).