Application value of intravoxel incoherent motion imaging in preoperative evaluating the perineural invasion of esophageal squamous cell carcinoma: a prospective study / 中华消化外科杂志

ABSTRACT

Objective:To investigate the application value of intravoxel incoherent motion (IVIM) imaging in preoperative evaluating perineural invasion (PNI) of esophageal squamous cell carcinoma (ESCC).Methods:The prospective study was conducted. The clinicopathological data of 63 ESCC patients who underwent IVIM imaging before surgery in the Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital from January 2016 to April 2019 were selected. Patients underwent routine magnetic resonance imaging scan and IVIM sequence examination. The apparent diffusion coefficient (ADC), true diffusion coefficient (D), pseudodiffusion coefficient (D*) and pseudodiffusion fraction (f) values were measured. Observation indicators (1) situations of enrolled patients; (2) IVIM examination of PNI of ESCC and its diagnostic efficiency. Measurement data with normal distribution were represented as Mean± SD, and comparison between groups was analyzed using the independent sample t test. Measurement data with skewed distribution were represented as M( Q1, Q3) or M(range), and comparison between groups was analyzed using the Wilcoxon rank sum test. Count data were described as absolute numbers, and comparison between groups was analyzed using the chi-square test. The interclass correlation coefficient (ICC) was used to evaluate cosistency of results. The receiver operating characrteristic curve was used to evaluate diagnostic efficiency. Results:(1) Situations of enrolled patients. A total of 63 patients were selected for eligibility. There were 41 males and 22 females, aged 60(range, 43?74)years. Of the 63 ESCC patients, 17 cases had PNI and 46 cases had no PNI. Cases with well differentiated tumor, moderate differentiated tumor or low differentiated tumor, cases in stage T1, T2, T3, T4 of infiltra-tion depth, cases with or without lymph node metastasis were 0, 6, 11, 0, 3, 13, 1, 15, 2 in patients with PNI, versus 6, 24, 16, 1, 22, 23, 0, 12, 34 in patients without PNI, showing significant differences between them ( Z=?2.32, ?2.49, χ2=19.58, P<0.05). (2) IVIM examination of PNI of ESCC and its diagnostic efficiency. The ADC, D, D*, f values were (2.21±0.33)×10 ?3 mm 2/s, (1.46±0.31)×10 ?3 mm 2/s, 32.50 (15.55,46.75)×10 ?3 mm 2/s, (0.34±0.09)×10 ?3 mm 2/s in patients with PNI, versus (2.45±0.45)×10 ?3 mm 2/s, (1.72±0.38)×10 ?3 mm 2/s, 34.85(23.43,51.20)×10 ?3 mm 2/s, (0.42±0.13)×10 ?3 mm 2/s in patients without PNI, showing significant differences in the ADC, D and f values between them ( t=?2.03, ?2.52, ?2.34, P<0.05) and no significant difference in the D* value between them ( Z=?0.84, P>0.05). The cosistency of ADC, D, D* and f values in the 63 ESCC patients measured by 2 radiologists was good, with the ICC as 0.96, 0.92, 0.89 and 0.95. The receiver operating characrteristic curve demons-trated the best cut-off value of ADC, D and f values were 2.50×10 ?3 mm 2/s, 1.68×10 ?3 mm 2/s and 0.41, with the areas under the curve as 0.67 (95% confidence interval as 0.54?0.79), 0.70 (95% confidence interval as 0.58?0.81) and 0.69 (95% confidence interval as 0.57?0.80), respectively. The combina-tion of D and f value for diagnosing PNI of ESCC had the area under the curve as 0.74 (95% confidence interval as 0.62?0.84), with the cut-off value as 0.20, sensitivity and specificity as 94.1% and 56.5%. Conclusions:IVIM parameters D and f values can evaluate and predict the PNI of ESCC preopera-tively. The combination of D and f values can further improve the diagnostic efficiency.