Role of Ultrasound Elastography in Patient Selection for Prostatic Artery Embolization

Clinical question
Is baseline ultrasound elastography predictive of prostate artery embolization efficacy?

Take away point
Baseline prostatic elastic modulus and shear wave velocity measurements each predict efficacy of prostate artery embolization at 1 year.

Reference
de Assis A.M., Moreira A.M., Carnevale F.C., Marcelino A.S.Z., Antunes A.A., Srougi M., and Cerri G.G. Role of Ultrasound Elastography in Patient Selection for Prostatic Artery Embolization. J Vasc Interv Radiol. 2021 Oct; 32 (10): 1410-1416.

Click here for abstract

Study design
Prospective, observational, descriptive, single-center cohort study.

Funding Source
No reported funding

Setting
Academic setting. University of Sao Paulo Medical School, Sao Paulo, Brazil.

Figure

Figure Scatter plots showing a moderate positive correlation between baseline SWV and 12-month IPSS (R = 0.68, P = .002) (a) and EM and 12-month IPSS (R = 0.62, P = .007) (b). EM = elastic modulus, IPSS = International Prostate Symptom Score (IPSS < 8, optimal; IPSS ≥ 8, suboptimal), SWV = shear wave velocity.

Summary

Ultrasound elastography (US-E) has recently been described as a novel tool to evaluate prostate artery embolization (PAE) outcomes with anatomical and functional assessments in patients with benign prostatic hyperplasia (BPH). US-E demonstrated that PAE not only treats bladder outlet obstruction (BOO) by reducing prostate volume, but also by improving the α-adrenergic mediated effect on prostatic muscular tonus, which addresses the dynamic component of BOO. While recent US-E research gave interesting insights on PAE mechanism of action, research on the prognostic role of US-E is lacking.

The authors performed a prospective, observational, descriptive, single-center cohort study involving 20 consecutive patients. Inclusion criteria were lower urinary tract symptoms LUTS) attributed to BPH for at least 6 months refractory to standard medical treatment (α-1 adrenergic receptor antagonist with or without 5-α reductase inhibitor), international prostate symptom score (IPSS) ≥ 8, and prostate volume ≥ 40 cm3 and ≤ 200 cm3. Exclusion criteria were biopsy-confirmed cancer, active urinary infection, presence of urinary retention requiring use of Foley urinary catheter, serum creatinine ≥ 2 mg/dL, and previous pelvic surgery or radiotherapy. From February 2018 to October 2019, 20 patients met inclusion criteria but 2 did not return for follow up and 1 refused to complete evaluation therefore the remaining 17 entered statistical analysis.

Primary endpoints were prostatic elastic modulus (EM) and shear wave velocity (SWV) assessed with US-E up to 30 days before and at 23–37 days after PAE. Patients on α-1 adrenergic antagonist were maintained until after the second US-E evaluation, after which the medication was permanently withdrawn. Secondary endpoints included IPSS questionnaire, IPSS-Quality of Life (QoL) item, and peak urinary flow rate (Qmax) assessed using uroflowmetry, among others. All ultrasound examinations were performed transrectally with an 11C3 endocavity probe (Toshiba Medical Systems Corporation, Otawara, Japan) on a Toshiba Aplio i800 (Toshiba Medical Systems Corporation) utilizing 2-dimensional sound wave elastography (SWE) via acoustic structure quantification and dynamic energy generation via acoustic radiation force impulse that generated multifocal stress and enabled EM and SWV measurement by building colored elastographic maps. This meant the exams did not depend on tissue compression by the operator during examinations. EM and SWV values were obtained from maps including the transitional zone of both lobes of the prostate simultaneously in at least 2 axial slices at the middle third of the prostate. All examinations were performed by the same radiologist (A.S.Z.M.).

Pearson product-moment correlation tests were used to assess association of IPSS at 1 year with baseline EM and SWV. Receiver operating characteristic (ROC) curves were used to determine predictive cutoff values of baseline EM and SWV. PAE efficacy at 1 year was considered optimal at IPSS < 8, QoL < 2, or Qmax ≥ 12 mL/s, respectively. Area under ROC curves were obtained at 95% confidence interval and statistical significance was defined as bicaudal type I error (p) < 0.05.

16 of the 17 patients (94.1%) underwent bilateral PAE. Catheterization was technically unsuccessful in the remaining 1 patient (5.9%) due to atherosclerosis. EM was moderately positively correlated with IPSS at 1 year (R = 0.62) and SWV was slightly more positively correlated with IPSS at 1 year (R =0.68), with both correlations being statistically significant (p = 0.007 and p = 0.002, respectively). SWV was also slightly positively correlated with QoL at 1 year but without statistical significance (p = 0.09, R = 0.41). Baseline EM ≥ 50.14 kPa predicted suboptimal IPSS with 84.6% specificity, suboptimal QoL with 68.8% specificity, and suboptimal Qmax with 80.0% specificity; and baseline SWV ≥ 5.9 m/s predicted suboptimal IPSS with 100% specificity, suboptimal QoL with 81.3% specificity, and suboptimal Qmax with 90.0% specificity. Results consistently demonstrated inferior PAE outcomes for patients with very low initial prostate elasticity, with potential to predict patient suitability for PAE using the possible prognostic cutoff values of EM < 50.14 kPa and/or SWV < 5.9 m/s.

Commentary

This study (1) further described PAE’s effect on prostate elasticity as measured via EM and SWV, and (2) explored the prognostic value of US-E in categorizing patients as good candidates for PAE by assessing IPSS, QoL, and Qmax at 1 year. A notable advantage to this study was the use of ultrasound that utilized dynamic energy generation via acoustic radiation force impulse, which meant ultrasound exams were not operator dependent. 

Possible limitations include the fact that all ultrasounds were read by the same radiologist, who is one of the authors and no indication of blinding was described, and the sample size of 17. 

This research further supports the hypothesis that PAE’s efficacy is based, at least partially, on prostate elasticity, which is a dynamic component to BOO. The results suggest a potential prognostic role for US-E in patient stratification for PAE. Additional future study, perhaps multi-center and/or with blinding, can be helpful in continuing to explore these initial promising results.

Additional Reading:

Video presentation on the history and development of shear wave elastography: https://www.spiedigitallibrary.org/conference-proceedings-of-spie/11242/112420B/Shear-wave-elastography--from-ultrasound-to-optics/10.1117/12.2540509.full?SSO=1

Review on acoustic radiation force impulse imaging: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3337770/

Post Author
Zayn Mohamed, JD
MS-4 Doctor of Medicine Candidate,
University of South Florida Morsani College of Medicine

Edited and formatted by @NingchengLi

Interventional Radiology Resident

Dotter Institute, Oregon Health and Science University