Diagnostic utility of diffusion‑weighted magnetic resonance imaging in two common renal tumors

Introduction

Renal cell carcinoma is the most common primary malignant tumor of the kidney, accounting for 85–90% of malignant kidney tumors (1) and 3% of all malignant tumors, with an incidence that is increasing yearly (2). Angiomyolipoma is the most common benign tumor of the kidney, comprising 7–9% of kidney tumors. The accurate evaluation of tumor behavior is critical in selecting an appropriate therapy and determining the prognosis. At present, the most commonly employed imaging modalities for renal tumors include computed tomography (CT), magnetic resonance imaging (MRI) and contrast-enhanced imaging. To differentiate between benign and malignant tumors or tumor stages, tissue density is measured on CT, the tissue signal intensity if measured on plain MRI and the pattern of contrast enhancement is examined on enhanced images. Renal angiomyolipoma usually contains significant fat and therefore, shows typical characteristics of fat on CT or MRI. The differential diagnosis of renal carcinoma is usually not difficult, but lesions with little or no fat and small tumors are difficult to discern on conventional CT, plain MRI and contrast-enhanced imaging (3–5). In addition, certain patients are allergic to the contrast material used during enhanced CT (6,7). Contrast-enhanced MRI may produce fewer allergic reactions, but it can also accelerate the progression of nephrogenic systemic fibrosis in patients with renal insufficiency (8–10).

With recent developments in imaging technology and the increased use of software for high-field MRI (1.5T and 3.0T), diffusion-weighted MRI (DWI) has gained wide use clinically. This diagnostic technique offers the advantage of good safety and does not require a contrast agent. DWI is the only non-invasive functional imaging able to assess the water diffusion status in vivo, and as a result, it is widely used in the examination of the central nervous system, particularly for the early diagnosis of cerebral infarction (11) and the staging of astrocytoma. The apparent diffusion coefficient (ADC) is an important index that is measured during DWI. In recent years, follow-up DWI has been increasingly employed in imaging the abdominal organs, breast, pancreas, liver and kidney (12–16). DWI provides functional information through three parameters: The dispersion diagram, the ADC diagram and the ADC value, which has gained significant attention due to its role in tumor diagnosis (17).

In the present study, a retrospective analysis was performed on 54 cases of renal carcinoma and 31 cases of angiomyolipoma assessed by DWI in order to investigate the utility of DWI in the differential diagnosis of common benign and malignant tumors in the kidney.

Materials and methods

General data

The complete medical records of 85 renal tumor cases presenting to the Linyi People's Hospital (Linyi, Shangdong, China) between March 2006 and December 2010 were collected. All patients underwent plain MRI and DWI pre-operatively, and the diagnosis was confirmed surgically by the Department of Urology. Among the 85 patients, 59 were male and 26 were female; the mean patient age was 53.8 years (range, 23–81 years). All 85 cases presented with unilateral renal tumors, 48 on the right and 37 on the left. The upper pole of the kidney was affected in 30 cases, the lower pole in 37 cases and the renal hilus in 18 cases. Clinical signs were absent in 38 cases, and in the remainder, the following physical findings were observed: Gross hematuria (n=14); lower back pain (n=13); lower back pain with hematuria (n=15); and a waist mass (n=5). Of the 54 cases of renal cell carcinoma, 8 patients underwent tumor enucleation, 37 patients underwent a nephrectomy and 9 patients underwent a radical nephrectomy. Of the 31 cases of renal angiomyolipoma, 4 patients underwent a nephrectomy and 27 patients underwent tumor enucleation. Tumors were diagnosed on post-operative histopathological staging as follows: Clear cell renal cell carcinoma (n=46); papillary carcinoma (n=7); chromophobe cell carcinoma (n=1); and renal angiomyolipoma (n=31). This study was conducted in accordance with the declaration of Helsinki and with approval from the Ethics Committee of Linyi People's Hospital. Written informed consent was obtained from all participants.

MRI examination

All 85 patients underwent plain MRI and DWI (1.5T; Twin Speed Infinity with Excite II; GE Healthcare Life Sciences, Pittsburgh, PA, USA). The patient was placed in the supine position and the phased array surface coil was positioned at the abdomen, with the Torso PA coils at the abdominal wall and side. The abdominal wall was secured in front of the coil using a bandage, and patients were trained to breathe deeply prior to holding their breath in order to reduce motion artifacts during scanning. Axial MRI was performed in a fast recovery fast spin echo (SE) accelerated sequence [axial T1-weighted imaging (WI), T2WI and fat suppression] and a coronal fast imaging employing steady state acquisition sequence. A single-shot SE echo planar imaging was obtained during axial DWI scanning. The two scans were performed with the same body position, and layer thickness on axial T2WI, and imaging parameters were as follows: b=0 or 800 sec/mm2; stimulated repetition time/echo time of 4,000/56 msec one time; received bandwidth of 125 kHz; thickness/distance of 5/0 mm; field of view of 36×34 cm; and a matrix of 128×128 cm. Sensitive gradient pulses were applied in the X-, Y- and Z-axes. Patients were required to breath-hold during scanning.

Image transmission

The DWI images were transferred to an AW4.0 workstation and processed using FuncTool 2.0 software (both GE Healthcare Life Sciences).

Image processing

During processing, the threshold was defined by removing fat, bone and gas regions from the image, and the b-value was entered, resulting in the DWI and ADC maps.

The region of interest (ROI) measured 60–100 mm2. Using the conventional axial T2WI and DWI images as a reference, the ROI was set at the same size in three successive levels of the ADC diagram, and the ADC was measured. The mean value was designated as the final value. In lesions with a homogeneous signal, the ROI avoided the margins as much as possible in order to reduce the partial volume effect. In lesions with an inhomogeneous signal, the ROI was positioned in the solid region of the lesion, and cystic change, necrosis, hemorrhage and calcification areas were avoided when possible.

The ADC values were measured and recorded, and the ADC diagram was stored. The ADC was calculated according to the following formula: ADC = (S0 / S1) / (b1 − b0), where b0 was 0 sec/mm2, b1 was 800 sec/mm2, S0 was the DWI signal intensity at a b-value of 0 sec/mm2 and S1 was the DWI signal strength with a b-value of 800 sec/mm2.

Statistical analysis

The normal distribution of ADC values was evaluated in the renal cell carcinoma group and the renal angiomyolipoma group using SPSS14.0 statistical analysis software (SPSS, Inc., Chicago, IL, USA). Subsequent to confirming that the data were normally distributed, the mean ADC values were compared between the two groups using an analysis of variance. P<0.05 was considered to indicate a statistically significant difference.

Results

MRI, DWI, and ADC in the renal angiomyolipoma group

On T1WI, the 31 angiomyolipoma cases showed a high partial fat signal intensity and a low signal intensity in the soft tissue. When the fat signal was suppressed, it showed a low signal intensity on T2 images, while the soft tissues showed a slightly high signal intensity. The corresponding DWI image showed a lower fat signal intensity and slightly higher signal intensity for soft tissue; the fat signal intensity on the pseudo-color ADC map was high, the soft tissue was slightly low and the signal intensity was inhomogeneous (Fig. 1A).

Figure 1. Map of MRI scan, DWI and ADC of (A) renal vascular leiomyoma group and renal cell carcinoma and (B) renal cell carcinoma groups. (a) T1WI and (b) T2WI images, and (c) T2WI and (d) ADC maps. MRI, magnetic resonance imaging; DWI, diffusion-weighted MRI; ADC, apparent diffusion coefficient; WI, weighted imaging.

MRI, DWI and ADC in the renal cell carcinoma group

Among the 54 renal cell carcinoma cases, 18 cases showed a homogeneous low signal intensity on T1WI images, a homogeneous high signal intensity on T2WI and a homogeneous high signal intensity on DWI. The corresponding pseudo-color ADC map showed a slightly low signal intensity and a uniform signal strength. In 36 cases containing necrosis, cystic change or hemorrhage, the T1WI showed an inhomogeneous low signal intensity, and the T2WI and DWI showed an inhomogeneous high signal intensity. The corresponding pseudo-color ADC map showed a heterogeneous slightly low signal intensity (Fig. 1B).

ADC comparison

As detailed in Table I, the ADC for the renal cell carcinoma group was significantly lower than that in the renal angiomyolipoma group (P<0.05).

Table I. Comparison of the ADC in the renal cell carcinoma and renal angiomyolipoma groups when the b-value was set at 800 sec/mm2.

Table I.

Comparison of the ADC in the renal cell carcinoma and renal angiomyolipoma groups when the b-value was set at 800 sec/mm2.

Group	n	ADC, x10−3 mm2/sec	F	P-value
Renal angiomyolipoma	31	1.8271±0.3486	25.718	0.000
Renal carcinoma	54	1.4181±0.1643

[i] ADC, apparent diffusion coefficient.

Discussion

The present study demonstrated that the ADC in renal cell carcinoma is lower than that in renal angiomyolipoma, which is consistent with the results of studies by Doğanay et al (18), Mytsyk et al (19) and Zhang et al (20), which found that the ADC was lower in malignant kidney tumors than in benign tumors. Doğanay et al reported a mean ADC of 2.21±0.63×10−3 mm2/sec for renal cell carcinoma and 2.55±0.49×10−3 mm2/sec for benign tumors. Mytsyk et al reported a similar trend of 2.11±0.25×10−3 mm2/sec in renal cell carcinoma and 2.36±0.32×10−3 mm2/sec in angiomyolipoma. Zhang et al reported a mean ADC of 1.264±0.271×10−3 mm2/sec for renal cell carcinoma and 1.717±0.431×10−3 mm2/sec for angiomyolipoma. The present analysis showed a mean ADC of 1.4181±0.1643×10−3 mm2/sec for renal cell carcinoma and 1.8271±0.3486×10−3 mm2/sec for renal angiomyolipoma.

The diagnosis of tumors on DWI is based on organizational, structural, cell density and karyoplasmic ratio changes within lesions, as well as changes to the large molecular distribution in the intra- and extracellular spaces; these variations alter the Brownian motion of water molecules, generating an abnormality in the DWI signal (21). The cell concentration increases in renal cell carcinoma, which decreases the extracellular clearance and slows the extracellular water molecular motion (22). In addition, the cell nuclei are relatively larger and the cytoplasm lessened, which further reduces water molecular motion within cells. These mechanisms explain the lower ADC value in renal cell carcinoma compared with normal tissues. The ADC is an index that measures the magnitude of random molecular motion within a given volume and is calculated as follows: ADC = ln(low S / high S) / (high b − low b), where low S and high S represent the signal intensity measured by DWI corresponding to a low b and high b, respectively, and b is a diffusion gradient factor. By calculating the ADC at each voxel and arranging the results in a grayscale image, an ADC diagram can be generated, providing a map of the ADCs of the entire lesion. In the ADC map, high signal areas represent high dispersion; the ADC value is high, while the signal on the corresponding DWI map is low. By contrast, a low signal zone on the ADC map represents a low dispersion region, indicating a low ADC value and a high signal on the corresponding DWI map. In addition, the ADC map eliminates the effect of T2 transmission and does not depend on the strength of the magnetic field and gradient; thus, the ADC value reflects the magnitude of molecular water movement in tissues, and can be measured and compared graphically (23).

The present study has two notable limitations. Firstly, the majority of the renal angiomyolipomas contained significant amounts of fat, and only a few lesions contained little or no fat. Furthermore, other common benign renal tumors, such as renal oncocytoma, were not included in this small study.

This study indicates that DWI may be valuable in the differential diagnosis of benign and malignant renal tumors in the kidney; it can differentiate between benign and malignant tumors, and could prove particularly useful for small renal vascular leiomyoma, tumors with little to no fat and renal cell carcinoma. DWI should be considered an important supplementary tool in the diagnosis and differentiation of renal tumors.

References