Fast 3-T MR-guided transrectal prostate biopsy using an in-room tablet device for needle guide alignment: a feasibility study

For this institutional review board (IRB)-approved feasibility study, 20 males scheduled for MR-guided prostate biopsy with a single CSR with a PI-RADS v2 [14] score ≥ 4 on diagnostic multiparametric prostate MRI (mpMRI) were prospectively enrolled between June and October 2016. Indications for diagnostic prostate mpMRI and subsequent biopsy were either elevated PSA (> 4.0 ng/ml) and clinical suspicion of primary PCa (n = 12) or follow-up in patients under active surveillance for minimal low-risk PCa (Gleason score ≤ 6) diagnosed on previous TRUS-biopsy (n = 8). None of the patients had received previous prostate treatment. Informed consent was obtained from all patients. A summary of patient and lesion characteristics is given in Table 1.

Prior to IRB application for this study, an MR safety assessment was conducted to determine safe operation conditions for the tablet device (iPad 2, Apple, Cupertino, CA, USA) in a 3-T MRI suite. The magnetically induced displacement force on the tablet device was determined according to American Society for Testing of Materials (ASTM) standard test method F2052-06 [15]. The tablet device was attached to a non-magnetic fixture using a nylon string such that it hung free in space. Using a protractor with 1° graduated markings, the deflection angle due to magnetic attraction of the tablet device was recorded at multiple locations in the scanner room and varying distances from the bore entry. Additional ASTM safety tests [16] were conducted to establish whether the tablet device or wireless network connection affected MR image quality.

For the proposed method, the tablet device was installed in the MR room. The tablet device was connected to a stand-alone computer outside the scanner room via a remote desktop application (VNC Viewer, RealVNC, Cambridge, UK) on a secured wireless network connection (Fig. 1a). To prevent an accidental approach to the scanner bore during operation, the device was incorporated into a Perspex holder mounted to a non-magnetic fixture secured to the MR scanner table (Fig. 1b).

All patients received antibiotic prophylaxis (oral ciprofloxacin, 2dd 500 mg) for 3 days, starting on the day before biopsy. All biopsy procedures were performed on a 3-T clinical MR system (MAGNETOM Skyra, Siemens, Erlangen, Germany) by one prostate interventionalist (C.O., 4-year experience). Patients were positioned in a prone position on the MR table. A commercially available transrectal biopsy device (DynaTrim, Invivo, Gainesville, FL, US) was used. Axial T2-weighted turbo spin echo (T2-TSE) and diffusion-weighted imaging (DWI) were acquired to reproduce pre-biopsy MRI findings. An additional extended axial T2-TSE data set incorporating the entire needle guide was acquired for planning purposes. On the stand-alone computer, two points were identified in the extended T2-TSE set using interventional planning software [Interactive Front End (IFE), Siemens Healthcare GmbH, Erlangen, Germany]: (1) the biopsy target, defined as the center of the lesion, and (2) the needle guide pivoting point, which had been previously determined at 9.2 cm from the proximal edge of the contrast-filled compartment of the needle guide in in-house experiments. The software then calculates the planned trajectory between these two points, representing the desired needle guide position to target the CSR (Supplemental Figure 1). Two orthogonal MR scan planes were aligned to this trajectory in axial and sagittal orientation and returned to the scanner console, providing the slice positions for an interactive real-time balanced steady-state free precession (BEAT) MR fluoroscopy sequence (temporal resolution ~3 imagings/s). The interventionalist then entered the MR room and manipulated the needle guide into both scan planes under MR fluoroscopy feedback visualized on the in-room tablet device while the patient was inside the scanner bore (Supplemental Video 1). Upon completion, short balanced steady-state free precession (bSSFP) scans were performed in axial and sagittal orientation to confirm correct alignment with the CSR. In case of correct alignment, subsequent biopsy was performed using an 18-gauge fully automatic MR-compatible biopsy gun (Invivo, Gainesville, FL, US). In case of incorrect alignment, additional routine manual adjustment steps were performed for correction. Axial and sagittal bSSFP confirmation scans were obtained to map each biopsy location. Detailed imaging parameters are summarized in Table 2. An overview of the biopsy procedure is shown in Fig. 2.

Per- and post-procedural complications were recorded with rates and description of adverse events. Complications were scored according to the Clavien grading system [17].

Technical feasibility and single-step targeting success of needle guide alignment were assessed. Single-step targeting success was defined as obtainment of a representative biopsy core directly after a single alignment step under MR fluoroscopy feedback on the in-room tablet device without requiring further needle guide repositioning. Representativity of each biopsy core was indicated on a five-point scale (i.e., 1 = not representative to 5 = representative) by a prostate radiologist reviewing each biopsy confirmation scan, where a score of ≥ 4 was considered a representative sample.

Biopsy procedure times were also recorded. The time to first biopsy was defined as the time from acquisition of the DWI sequence until the confirmation scan of the first biopsy of the CSR (Fig. 2). The total procedure time was defined as the time between localizer acquisition and the last scan of the biopsy procedure. Finally, the MR fluoroscopy time was defined as the duration the physician required to align the needle guide under MR fluoroscopy feedback. The setup time of the required hardware was not recorded as multiple subsequent biopsy procedures only required a single installation of ± 5-10 min.

After the biopsy procedures, core samples underwent histopathological workup with hematoxylin-eosin staining and were subsequently evaluated for the presence of PCa or benign findings. When applicable, a Gleason score was assigned.

In the 3-T suite, it was found that the specified unfixed tablet device could be safely operated at ≥ 50 cm from the bore entry without significant magnetic attraction according to ASTM tolerance levels. No noticeable interferences or artifacts affecting MR image quality were observed because of the presence of the tablet device or wireless network connection.

A total of 20 CSRs was successfully sampled in 20 patients (1 per patient). A median of two biopsy cores (range: 2-3) was taken per CSR resulting in 46 biopsy samples. There were no per-procedural or post-biopsy complications.

Needle guide alignment using the in-room tablet device was technically feasible in all patients (Fig. 3). Single-step targeting was successful in all but one patient (19/20 lesions; 95%). In the latter patient, an additional routine repositioning step was needed because the lesion was located anteriorly outside of the maximum range of the biopsy device. After forward adjustment of the needle guide, biopsy was performed successfully. The median radiological score of biopsy representativity was 5 out of 5 (range: 4-5).

With use of the in-room tablet device, mean time to first biopsy was 5.8 ± 1.0 min and mean total procedure time was 23.7 ± 4.1 min. There was a downward trend in total procedure time with increasing procedure number, with the last ten procedures being performed in less than 25 min (Fig. 4). The mean MR fluoroscopy time was 1.1 ± 0.3 min.

Histopathology revealed prostate cancer in 18/20 (90%) patients. Gleason scores were 3+3 (3), 3+4 (10), 3+5 (1), 4+3 (2), 4+4 (1) and 4+5 (1). In total, 42 of 46 (91%) biopsy samples contained cancer. In two patients (10%), biopsy samples contained prostatitis.