Target reader: Radiologists and ENT(ear, nose and throat) doctors.

Purpose: To compare pure molecular diffusion (D), perfusion-related diffusion (D*), perfusion fraction (f) based on intravoxel incoherent motion (IVIM) theory and apparent diffusion coefficient (ADC) in patients with nasopharyngeal carcinoma (NPC).

Methods: Patients. The institutional research ethics committee approved this study. Written informed consent was obtained from all patients and the caretakers on behalf of minors. Sixty-five consecutive patients (48 men, 17 women; mean age, 51 years; age range, 16–69 years) with suspected NPC. MRI protocol. Both IVIM-DWI MRI and conventional DWI MRI were performed using a GE 3T MRI scanner. The IVIM DW imaging sequence was performed prior to the injection of Gd-DTPA (Bayer Healthcare, Berlin, Germany). Thirteen b values (0, 10, 20, 30, 50, 80, 100,150, 200, 300, 400, 600 and 800 s/mm2) were applied with a single-shot diffusion-weighted spin-echo echo-planar sequence. The lookup table of gradient directions was modified to allow multiple b value measurements in one series. Parallel imaging was used with an acceleration factor of 2. A local shim box covering the nasopharynx region was applied to minimize susceptibility artefacts. In total, 14 axial slices covering the nasopharynx were obtained with a 24-cm field of view, 4 mm slice thickness, 1 mm slice gap, 3,000 ms TR, 58 ms TE, 128×128 matrix and NEX=2. Statistics. A nonparametric Mann–Whitney test was used to compare IVIM parameters between primary NPC and enlarged adenoid cases. The D value (with independent significance between two groups) and ADC were assessed using a ROC curve to estimate the diagnostic tolerance. All statistical analyses were performed using SPSS 13.0 for Windows (SPSS, Chicago) and MedCalc (MedCalc Software, Acacialaan 22, B-8400 Ostend, Belgium). P < 0.05 was considered significant.

Results: IVIM DWI was successfully conducted at the primary site in 60 out of 65 patients (45 men and 15 women; mean age, 52 years; age range, 16–69 years), including 37 NPC cases and 23 enlarged adenoid cases confirmed by subsequent nasopharyngeal biopsy. IVIM DWI was failed in the remaining five patients because of susceptibility artefacts around the skull base and paranasal sinuses (three cases of NPC) or motion artefacts due to swallowing (two cases of enlarged adenoids). The mean tumour volumes (±SD) for patients with NPC and enlarged adenoids were 22.50 ± 6.06 and 8.52 ± 2.78 cm3, respectively, according to the summation-of-areas technique. As shown in Figure 1, IVIM DW images of NPC were performed with 13 b values (range: 0–800 s/mm2). The signal fitting curve proved that IVIM fitting is a bi-exponential model. We successfully implemented bi-exponential IVIM model to calculate both diffusion and perfusion parameters. Our results showed that D (P = 0.001) and f (P < 0.0001) were significantly lower in patients with primary NPC than in patients with enlarged adenoids, whereas D* was significantly higher (P < 0.0001) in the NPC group. However, the difference in the ADC observed between the two groups did not reach significance (P > 0.05). Box plots comparing D, D* and f between patients with NPC and patients with enlarged adenoids are shown in Figure 2. The ROC analysis of ADC and D indicated that when both sensitivity and specificity were adjusted to produce the highest accuracy, the optimal D and ADC thresholds for distinguishing primary NPC from enlarged adenoids were 0.75 × 10−3 mm2/s and 0.936 × 10−3 mm2/s respectively. The AUC for D (0.849) was significantly larger than ADC (0.566) (P < 0.05).

Conclusion: Our study demonstrates the feasibility of using IVIM MRI to investigate primary NPC. D was significantly decreased and D* was remarked increased in primary NPC, increased D* reflected increased blood vessel generation and parenchymal perfusion in primary NPC.