Novel contrast mixture achieves contrast resolution of human bladder wall suitable for T1 mapping: applications in interstitial cystitis and beyond

Abstract

Purpose: Instillation of novel contrast mixture (NCM) was recently shown to improve the contrast resolution of rat bladder wall with high contrast-to-noise ratio (CNR). Here, the clinical safety and the feasibility of NCM-enhanced MRI to achieve artifact-free visualization of human bladder wall suitable for quantitative measurement of the magnetic resonance (MR) longitudinal relaxation time (T1) was assessed.

Methods: Six female subjects [two controls and two with Hunner-type interstitial cystitis IC and two with non-Hunner-type IC] consented for MRI at 3 T before and after instillation of NCM [4 mM gadobutrol and 5 mM ferumoxytol in 50 mL of sterile water for injection]. Single breath-hold fast MR acquisition in large readout bandwidth for 5-mm-thick single slice with variable flip angle was applied to minimize the motion and chemical shift artifacts in measurements of bladder wall thickness (BWT), CNR and T1 from 20 pixels.

Results: NCM instillation in subjects did not evoke pain or discomfort. Fourfold increase in bladder wall CNR (*p < 0.02) and pixel size of 0.35 mm with minimal influence of artifacts allowed accurate determination of bladder wall thinning ~ 0.46 mm from 50 mL NCM (*p < 0.05). Pre-contrast bladder wall T1 of 1544 ± 34.2 ms was shortened to 860.09 ± 13.95 ms in Hunner-type IC (*p < 0.0001) relative to only 1257.42 ± 20.59 and 1258.16 ± 6.16 ms in non-Hunner-type IC and controls, respectively.

Conclusion: Findings demonstrate the safety and feasibility of NCM-enhanced MRI to achieve artifact-free differential contrast and spatial resolution of human bladder wall, which is suitable for measuring BWT and pixel-wise measurement of T1 in post-contrast setting.

Keywords: Ferumoxytol; Gadolinium; Human bladder; MRI; T1 mapping.

Figures

Fig. 1

NCM-enhanced MRI relies on the differences in the particle size and contrast mechanisms of two FDA-approved agents for increasing the CNR of human bladder wall. a Schematic illustration depicting the diffusion of smaller-sized gadolinium-based contrast agent GBCA (Gd) molecules into the ECV of the bladder wall leads to shortened T1 (positive contrast), while large molecular size of ferumoxytol (Fe) allows hypointensity to dominate within the lumen due to T2 relaxation (signal loss) in T1w images. b T1w image acquired at 7 T (TR/TE 500/7.3 ms) of ex vivo rat bladder phantom instilled with 0.3 mL NCM. GBCA diffusion into the bladder wall from lumen increases the differential contrast of the hyperintense bladder wall with inner (asterisk) and outer border (open circle). Retention of ferumoxytol in lumen quenches the GBCA-mediated enhancement leading to the darkened lumen

Fig. 2

Effect of FA on T1w signal intensity at 3 T—the signal intensity (arbitrary units) of the GBCA phantom doubled in T1w 2D-FLASH image (b) acquired at FA of 15° (614), compared to the image acquired at FA of 6° (272) shown in (a). Since signal intensity of GBCA phantom doubled from FA of 6°–15°, T1w 2D-FLASH image acquired at FAs of ≤ 15° was used for measuring CNR, BWT and T1 of human subjects. Lack of difference in the signal intensity of the NCM and Fe phantoms demonstrates that contrast enhancement by GBCA in NCM phantom is quenched by ferumoxytol (Fe) and the observed quenching of GBCA is independent of the FA. All four phantoms were imaged together at different FA with 2D-FLASH sequence (TR/TE 5/2 ms) at 3-T scanner

Fig. 3

BWT is underestimated in T2w MRI. Representative pre-contrast T1w 2D-FLASH (TR/TE 5.5/2 ms) and T2w images (TR/ TE 6.94/3.06 ms) of the controls (a, b) and Hunner-type IC/BPS subjects (c, d). A slight underestimation of BWT by T2w is suggested by BWT measurements of 3.39 ± 0.74 versus 3.01 ± 0.82 mm in T1w and T2w images, respectively (n = 6). Thickened bladder wall noted in the Hunner-type IC/BPS patient shown here (c, d) corroborates the increased circumferential thickness noted by other methods

Fig. 4

NCM-enhanced MRI improves the human bladder wall CNR. Representative pre- and post-contrast images of female control subjects, non-Hunner-type IC and Hunner-type IC patients acquired in orthogonal orientation at 3-T scanner are shown. NCM instillation raised the CNR of human bladder wall in post-contrast images (57.84 ± 32.01 vs. 12.34 ± 9.63; *p < 0.02, paired Student’s t test; n = 6) (b, d, f) by fourfold over the pre-contrast images (a, c, e) acquired with the same parameters. Fast image acquisition over single breath hold of ~ 17 s at the FA of 14° with slice thickness of 5 mm, matrix size 256 × 256, FOV of 18 × 18 cm2 and 10 averaging achieved in-plane resolution of 0.35 mm for CNR measurement in 20 pixels. Increased differential contrast between lumen and bladder wall, smaller pixel size and large bandwidth 574 Hz per pixel allowed accurate determination of the significant bladder wall thinning from 3.39 ± 0.74 to 2.93 ± 0.8 mm (paired Student’s t test, *p < 0.05, n = 6) in post-contrast images acquired after 50 mL instillation. (BWT is shown by the inset legend.) Trace amount of blood oozing from lesions in Hunner-type IC/BPS patient is shown by gray granules (indicated by red dotted line) against the bright background of lumen near the catheter shown by letter C in post-contrast images. The catheter is shown by the letter C in post-contrast images, and the dark areas within the lumen are caused by air bubble introduced from the instillation procedure. Titanium hip implant casts a dark shadow in pelvic region of Hunner-type IC patient, but the effect on bladder wall imaging was minimal

Fig. 5

T1 measurement of human bladder wall (a–f) corresponds to respective images shown in Fig. 4. Calculated T1 values over the entire field of view are illustrated in a color-coded map, where dark blue-colored pixel assigns T1 of 0 ms and dark red-colored pixel assigns T1 of 2500 ms. a, c, e The pixels representing bladder wall (within the red dotted line regions in all figures) in pre-contrast images of six subjects are in the range of green to yellow, which together generate a mean value of 1544 ± 34.2 ms. Values of at least 20 pixels were chosen for T1 measurement in each subject. b, d, f The color-coded T1 map for post-contrast bladder wall depicts a color gradient indicating a greater shortening of T1 from GBCA diffusion in pixels representing the inner layer (urothelium) (dark blue color) > pixels representing the middle-layer lamina propria (blue to green) > pixel representing the outer detrusor layer (green to yellow). Larger blue- to green-colored band of pixels in the bladder wall of Hunner-type IC patients indicates deeper bladder wall diffusion of GBCA compared to controls and non-Hunner-type IC/BPS patients. Catheter in the center is shown in (c), and the hip implant of Hunner-type IC patient is indicated

Fig. 6

The plot shows the pre-contrast and post-contrast values of the bladder wall T1 in milliseconds (ms). Pre-contrast values of six subjects were not different from each other with a mean value of 1544 ± 34.2 ms, which was shortened by the diffusion of GBCA into the ECV of bladder wall. Fourfold greater T1-shortening was noted in Hunner-type IC patients with value of 860.09 ± 13.95 ms compared to controls and non-Hunner-type IC subjects with values of 1258.16 ± 6.16 and 1257.42 ± 20.59 ms, respectively. Reduction from pre-contrast T1 values (open circles) was ~ 44% for Hunner-type IC patients compared to only ~ 18% for controls and non-Hunner-type IC subjects. Pre-contrast (open circles) and post-contrast (filled square) mean values of individual subjects are shown for comparison and only the statistical difference between post-contrast bladder wall T1 values for individual subjects was assessed (*p < 0.0001, two-way ANOVA followed by Tukey’s test)