Transperineal Versus Transrectal Magnetic Resonance Imaging–targeted and Systematic Prostate Biopsy to Prevent Infectious Complications: The PREVENT Randomized Trial

Abstract

Background and objective:

The transrectal biopsy approach is traditionally used to detect prostate cancer. An alternative transperineal approach is historically performed under general anesthesia, but recent advances enable transperineal biopsy to be performed under local anesthesia. We sought to compare infectious complications of transperineal biopsy without antibiotic prophylaxis versus transrectal biopsy with targeted prophylaxis.

Methods:

We assigned biopsy-naïve participants to undergo transperineal biopsy without antibiotic prophylaxis versus transrectal biopsy with targeted prophylaxis (rectal culture screening for fluoroquinolone-resistant bacteria and antibiotic targeting to culture and sensitivity results) through a multicenter, randomized trial. The primary outcome was postbiopsy infection captured by a prospective medical review and patient report on a 7-d survey. The secondary outcomes included cancer detection, noninfectious complications, and a numerical rating scale (0–10) for biopsy-related pain and discomfort during and 7 d after biopsy.

Key findings and limitations:

A total of 658 participants were randomized, with zero transperineal versus four (1.4%) transrectal biopsy infections (difference –1.4%; 95% confidence interval [CI] –3.2%, 0.3%; p = 0.059). The rates of other complications were very low and similar. Importantly, detection of clinically significant cancer was similar (53% transperineal vs 50% transrectal, adjusted difference 2.0%; 95% CI –6.0, 10). Participants in the transperineal arm experienced worse periprocedural pain (0.6 adjusted difference [0–10 scale], 95% CI 0.2, 0.9), but the effect was small and resolved by 7 d.

Conclusions and clinical implications:

Office-based transperineal biopsy is tolerable, does not compromise cancer detection, and did not result in infectious complications. Transrectal biopsy with targeted prophylaxis achieved similar infection rates, but requires rectal cultures and careful attention to antibiotic selection and administration. Consideration of these factors and antibiotic stewardship should guide clinical decision-making.

Patient summary:

In this multicenter randomized trial, we compare prostate biopsy infectious complications for the transperineal versus transrectal approach. The absence of infectious complications with transperineal biopsy without the use of preventative antibiotics is noteworthy, but not significantly different from transrectal biopsy with targeted antibiotic prophylaxis.

1. Introduction

In the USA and Europe, >2 million prostate biopsies are performed annually for the detection of prostate cancer [1]. Many men undergo more than one biopsy during their lifetime, particularly with increasing use of active surveillance [2], which requires serial biopsies to monitor for disease progression.

As recently as 2015, 99% of US prostate biopsies were taken by a transrectal approach [3], with the needle traversing the rectal mucosa into the prostate. This process consistently introduces bacteria into the prostate, urinary tract, and bloodstream [4]. Infectious complications occur in up to 7% of transrectal biopsy [5,6]. The risk of hospitalization has increased significantly from 0.6% in 1996 to 3.6% in 2005, in parallel with the increase in fluoroquinolone-resistant bacteria worldwide [7]. The United States Preventive Services Task Force considered biopsy-related adverse events among the harms of prostate-specific antigen (PSA) screening [8].

Antibiotic prophylaxis for transrectal prostate biopsy is used universally, although there is no consensus on best practice. Professional guidelines recommend fluoroquinolone prophylaxis [5]. An alternative is targeted prophylaxis, where a rectal swab culture is performed to evaluate antibiotic sensitivities to rectal flora, and the presence of fluoroquinolone resistance, leading to a switch to another antibiotic [9]. By contrast, for augmented prophylaxis, multiple antibiotics are given empirically to prevent infection [5]. Targeted prophylaxis is more time consuming due to multiple steps and is more costly, while augmented prophylaxis is widely used due to its relative ease. However, the optimal regimen for augmented prophylaxis is unknown due to significant variation in antimicrobial resistance patterns [10]. While fluoroquinolones are used for prophylaxis due to their high concentration in prostate tissue, overuse and misuse have exacerbated antimicrobial resistance and, accordingly, postbiopsy infections [5]. The European Commission banned empiric fluoroquinolone prophylaxis due to a negative benefit-risk balance; however, the regulations allow targeted fluoroquinolone prophylaxis due to better antibiotic stewardship [11].

The high number of biopsies and biopsy complications, along with the need for routine antibiotic prophylaxis, suggest that novel strategies might be of benefit. One obvious alternative is to use a transperineal approach, which avoids introducing rectal flora into the prostate (although there is an absence of consensus regarding antibiotic prophylaxis) [12]. Transperineal biopsy historically needed to be performed under general anesthesia, limiting the generalizability of its use. In recent years, new techniques have allowed transperineal biopsy to be performed under local anesthesia in the office setting [13], raising the possibility of routine implementation.

While in theory, the use of the transperineal approach should reduce the risk of infection, this has never been demonstrated in a randomized trial. Moreover, the transperineal biopsy needle traverses denser tissue over a greater distance, raising questions of cancer detection. The PReclude infection EVEnts with No prophylaxis Transperineal (PREVENT) study is a randomized controlled superiority trial comparing transperineal with transrectal biopsy, with infectious complications as the primary outcome, and cancer detection, noninfectious complications, and patient-reported outcomes of pain, discomfort, and anxiety as the secondary outcomes. We hypothesized the clean versus contaminated wound classification of the transperineal versus transrectal biopsy approach would result in fewer infectious complications.

2. Patients and methods

The trial protocol [14] was approved by the institutional review board at study sites and Biomedical Research Alliance of New York 18–02-365. The trial biostatistician (M.A.) analyzed the data. No commercial entity was involved in the trial. The trial was funded by the National Cancer Institute, which was not involved in protocol development, data analysis or interpretation, or manuscript preparation. Deviations were monitored by the New York Presbyterian Weill Cornell Medicine (NYP-WCM) Data Safety Monitoring Committee.

We assumed one infection (0.4%) in the transperineal arm and a 5% infection risk in the transrectal arm [5]. We estimated that randomization with 236 patients per arm would provide 80% power to reject the null hypothesis of no difference in infections, given a two-sided α of 0.05. The sample size was increased to 530 to allow for 10% withdrawal and loss to follow-up.

From March 2021 through May 2023, patients were recruited at ten centers and were eligible for enrollment if they had not undergone prior prostate biopsy, had an elevated PSA level and/or abnormal digital rectal examination, and had suspicious prostate magnetic resonance imaging (MRI) characteristics (Prostate Imaging Reporting and Data System, version 2.1 [15] scores 3–5). A small number of participants had biopsy without MRI (claustrophobia and metal prosthesis) and were included as our primary outcome was infection. The exclusion criteria included acute prostatitis in the last 6 mo or any current bacterial infection requiring antibiotic treatment. There have been no significant methodological (eligibility or outcomes) changes following trial commencement.

We used two-stage informed consent [16], although in some sites performing predominantly transperineal biopsy, traditional one-stage consent was used (see the Supplementary material for details). Research coordinators implemented the assignment sequence, which used randomly permuted blocks of unequal size stratified by the urologist and PSA (<4, 4–9.9, and ≥10 ng/ml), through central password-protected, web-based randomization, ensuring that allocation could not be predicted before or modified after randomized. Allocation was not blinded to researchers or participants.

A rectal culture to screen for fluoroquinolone-resistant organisms was obtained prior to transrectal biopsy, and targeted antibiotic prophylaxis was administered [10]. In the absence of resistance, fluoroquinolone prophylaxis was administered; an alternative based on culture sensitivities was used for participants allergic to fluoroquinolones. In the 15% of participants with rectal culture–fluoroquinolone resistance, an alternative antibiotic was given based on sensitivities. No antibiotic prophylaxis was given for transperineal biopsy. Investigators followed the trial protocol–specified biopsy procedural approach and used the transperineal local anesthetic approached described by Kubo et al [13]. A maximum of 20 ml of 1% lidocaine was used for both approaches, with 12 systematic biopsy cores and three targeted cores per MRI region of interest [14].

Immediately after biopsy, participants completed a numerical rating scale (0–10) assessment of pain, discomfort, and anxiety, with higher scores indicating greater symptom intensity [17].

In addition, we ascertained outcomes of interest through 7 d, because all infections, hospitalizations, and 99.9% postbiopsy adverse events requiring intervention (Common Terminology Criteria for Adverse Events v.5.0 of grade ≥2) occur within this time frame [18]. Adverse events were followed until resolution. Infectious complications included uncomplicated urinary tract infection (UTI), complicated UTI, and urosepsis [14].

The secondary outcomes included bleeding requiring intervention and urinary retention captured through a combination of a prospective medical review and the 7-d survey. The 7-d survey comprised the general symptom items from the validated biopsy TRUS-BxQ instrument (Supplementary material) [18]. Items were queried for fevers, chills, UTIs, acute urinary retention, new medications, additional postbiopsy healthcare encounters, and persistent pain and discomfort. If participants reported additional encounters, they were contacted to rule out adverse events treated at an outside facility. Healthcare providers were also queried regarding adverse events when reviewing biopsy pathology. We also compared the detection of clinically significant (grade group 2–5) and low-grade (grade group 1) cancer.

The primary analyses followed the intent-to-treat principles (Fig. 1). We also performed planned per-protocol analyses (Supplementary material) by type of biopsy received after excluding 26 participants with protocol deviations. Multivariable logistic regression with site as a covariate was used to analyze the association between the randomization arm and infection, urinary retention, bleeding requiring intervention, and detection of clinically significant and insignificant cancer. Adjusted absolute risk differences were calculated using regression least-squares adjusted mean difference. However, due to the small number of significant adverse events, we were unable to adjust for site, and instead used Fisher’s exact test and estimated the difference in proportions along with the Newcombe hybrid score 95% confidence intervals (CIs).

Fig. 1–

CONSORT diagram showing the flow of participants through various stages of a randomized trial. MRI = magnetic resonance imaging.

We tested for a difference in biopsy pain, discomfort, and anxiety at biopsy, and pain and discomfort were reported on the 7-d survey. Categorical responses were analyzed as described above, and continuous responses were analyzed using the analysis of covariance with randomization strata as covariates. All analyses were conducted using R version 4.2.2 with the tidyverse (v1.3.2), gtsummary (v1.6.3), and emmeans (v1.8.4.1) packages.

3. Results

A total of 658 participants underwent randomization, with 328 versus 330 assigned to transperineal versus transrectal biopsy (Fig. 1). Participant characteristics were well balanced at baseline (Table 1). Of the participants who underwent biopsy, 226 (79%) transperineal and 223 (80%) transrectal participants returned 7-d surveys with an 80% response rate.

Table 1–

Characteristics by randomization arm

Characteristic	Transperineal (N = 287), n (%)	Transrectal (N = 280), n (%)
Type of biopsy
Transperineal	272 (95)	19 (6.8)
Transrectal	15 (5.2)	261 (93)
Age	66 (61, 71)	66 (61, 70)
Race
Asian	12 (4.2)	16 (5.7)
Black or African American	32 (11)	44 (16)
Other	17 (5.9)	15 (5.4)
Unknown	37 (13)	37 (13)
White	189 (66)	168 (60)
Hispanic ethnicity	11 (4.7)	10 (4.3)
Unknown	51	49
BMI	27 (25, 30)	27 (24, 31)
Unknown	1	0
History of smoking	66 (23)	68 (24)
Unknown	1	1
Family history of prostate cancer	69 (24)	65 (23)
Unknown	2	3
Indication
Abnormal digital rectal exam	6 (2.1)	10 (3.6)
Elevated PSA	279 (98)	269 (96)
None of the above	1 (0.3)	1 (0.4)
Unknown	1	0
PSA	5.8 (4.4, 8.0)	5.8 (4.6, 8.3)
Prostate volume	41 (32, 57)	43 (32, 59)
Unknown	2	4
MRI	286 (100)	278 (99)
MRI PI-RADS score
1	7 (2.5)	6 (2.2)
2	22 (7.7)	27 (9.8)
3	67 (24)	52 (19)
4	119 (42)	123 (45)
5	69 (24)	68 (25)
No MRI performed	1 (0.3)	2 (0.6)
Unknown	2	2
Number of systematic cores	12 (12, 12)	12 (12, 12)
Number of targeted cores	3 (2, 5)	3 (2, 5)
Unknown	3	1

BMI = body mass index; MRI = magnetic resonance imaging; PI-RADS = Prostate Imaging Reporting and Data System; PSA = prostate-specific antigen.

Data are presented as n (%) and median (quartile 1, quartile 3).

There were no infections in the transperineal biopsy arm compared with four (1.4%) infections in the transrectal biopsy arm (difference −1.4%; Newcombe hybrid score 95% CI −0.3, 3.2; p = 0.059). The infections were febrile but did not require intensive care. Details regarding infection events are shown in the Supplementary material. Other complications were also very uncommon (Table 2). Notably prior literature suggests a higher risk of urinary retention with transperineal biopsy; however, only one (0.3%) participant experienced urinary retention in the transperineal biopsy arm. Clinically significant cancer was detected in 151 (53%) transperineal biopsy arm versus 141 (50%) transrectal biopsy arm (adjusted difference 2.0%; 95% CI −6.0%, 10%).

Table 2–

Adverse events by randomization arm where differences are calculated as the difference in the transrectal arm subtracted from the transperineal arm

Characteristic	Transperineal (N = 287), n (%)	Transrectal (N = 280), n (%)	Difference (%)	95% Confidence interval (%)	p value
Infection	0 (0)	4 (1.4)	−1.4	−3.6, 0.2	0.059
Urinary retention	1 (0.3)	3 (1.1)	−0.7	−2.8, 1.0
Bleeding requiring intervention	0 (0)	1 (0.4)	−0.4	−2.0, 1.0
Gleason grade group 2–5	151 (53)	141 (50)	2.0	−6.0, 10
Gleason grade group 1	49 (17)	62 (22)	−5.1	−12, 1.7

CI = confidence interval.

Values are presented as n (%); differences along with Newcombe hybrid score 95% confidence intervals and p values were calculated using Fisher’s exact test for the primary outcome of infection. For cancer detection outcomes, Gleason grade group differences adjusted for site along with 95% confidence intervals were calculated using the logistic regression least-squares adjusted mean difference (95% CI).

Patient-reported outcomes are shown in Table 3. Although the mean biopsy pain was worse with the transperineal approach, few participants experienced high levels of pain, and differences between groups did not persist at 7-d follow-up.

Table 3–

Patient-reported outcomes by randomization arm where differences are calculated as the difference in the transrectal arm subtracted from the transperineal arm adjusted for study site

Characteristic	N	Transperineal (N = 287)	Transrectal (N = 280)	Adjusted difference	95% Confidence interval
Biopsy pain	548	3.6 (2.3)	3.0 (2.1)	0.6	0.2, 0.9
Unknown		10	9
Biopsy pain ≥7 (severe)	548	33 (12%)	19 (7.0%)	5.0%	−0.1%, 10%
Unknown		10	9
Biopsy discomfort	554	4.2 (2.5)	3.8 (2.3)	0.4	0.0, 0.8
Unknown		9	4
Biopsy anxiety	565	3.9 (3.0)	4.2 (2.9)	−0.3	−0.8, 0.1
Unknown		1	1
7-d survey discomfort	448	2.1 (2.4)	1.7 (2.2)	0.3	–0.1, 0.7
Unknown		61	58
7-d survey pain present	449	22 (9.7%)	32 (14%)	−5.2%	−12%, 1.5%
Unknown		61	57
7-d survey pain score >3	444	15 (6.6%)	13 (6.0%)	0.8%	−4.6%, 6.3%
Unknown		61	62

ANCOVA = analysis of covariance.

For continuous variables, values are presented as mean (standard deviation); ANCOVA was used to generate the differences adjusted for site. For categorical variables, values are presented as n (%); differences were adjusted for site along with 95% confidence intervals calculated using the logistic regression least-squares adjusted mean difference.

4. Discussion

In the first multicenter, randomized trial comparing novel, office-based transperineal prostate biopsy with the traditional transrectal approach, we found zero infections with the transperineal approach, with negligible rates of other complications and similar detection of clinically significant disease. Although pain was worse in the transperineal approach, differences were moderate and resolved by 7 d. Indeed, a 1.3–1.6 difference is widely accepted to be the minimal clinically important difference [19,20], which is more than double the difference in pain scores from our trial. Our finding of greater procedural pain with transperineal biopsy is also consistent with a prior prospective, nonrandomized study [21], which showed a similar difference in pain that resolved the following day.

Our reported 1.4% infection rate in the transrectal arm is lower than that reported previously for cohorts undergoing augmented prophylaxis (5–7%) [5], although similar to the 1.9% infections from a prospective, nonrandomized, single-center study of targeted prophylaxis transrectal biopsy at an academic medical center [10]. Furthermore, no participants experienced sepsis, in contrast to a risk range of 0.3–3.1% from systematic reviews [5]. Moreover, Mian et al [22] recently completed a single-center randomized trial demonstrating 2.7% versus 2.6% infectious complications for transperineal biopsy without antibiotic prophylaxis versus transrectal biopsy with augmented prophylaxis. However, there are significant differences in defining infectious complications. Mian et al [22] examined 30-d composite definitions of fever (including undocumented), antibiotic prescriptions (for suspected infection), office visits, or phone calls. The definition by Mian et al [22] is notable given the very liberal criteria for infectious complications.

We used transrectal biopsy with targeted prophylaxis as the comparator because it combines optimal antibiotic stewardship, is more rigorous, and is considered best practice compared with augmented, empiric prophylaxis. However, the use of targeted prophylaxis is low. In a contemporary, national study, rectal swabs were obtained in only 1.8% of Medicare beneficiaries prior to prostate biopsy [6]. We demonstrated that targeted prophylaxis transrectal biopsy is an effective alternative for patients without access to transperineal biopsy, a key consideration given that fewer than half of US urology residents in 2022 were exposed to transperineal biopsy [23].

Our findings have significant implications for value-based care. The annual healthcare payments attributed to prostate biopsy infection hospitalizations is approximately $620 million [24], an estimate that excludes the treatment of infections that do not require hospitalization. Transperineal biopsy enables the exclusion of antibiotic prophylaxis and downstream costs associated with rectal swab collection and associated prescriptions, reduces antibiotic resistance, and eliminates antibiotic-related side effects. Moreover, the Food and Drug Administration recommended avoidance of fluoroquinolones due to rare but permanently disabling side effects [25].

An important finding was a very low rate of urinary retention for the transperineal approach. This is in contrast with prior studies demonstrating rates of ≥10% [5]. However, the majority of these reports included cohorts where transperineal biopsy was performed under general anesthesia, using saturation biopsy, in which >20 biopsy cores were sampled. The advent of MRI-targeted biopsy techniques contributes to the paradigm shift to the office setting, and we demonstrate a reassuringly low risk for urinary retention.

Critically, the detection of clinically significant prostate cancer was similar by approach, with the 95% CI excluding any clinically relevant decrease in detection rates for transperineal biopsy. The rate of clinically significant cancer that we report is slightly higher than that of a National Cancer Institute series conducted from 2007 to 2019 (52% vs 44%) [26]. This is best explained from secular trends associated with changes in PSA screening practices [27], but also serves as a benchmark for biopsy effectiveness. There was no central pathology review. However, all participating sites were major academic centers and randomization was stratified by site, making this unlikely to affect our results notably.

Our finding of greater procedural pain with transperineal biopsy is consistent with that of a prior nonrandomized study [21]. This study found no difference in postbiopsy pain the following day.

We did not compare transperineal biopsy with transrectal biopsy with augmented prophylaxis, the most common US approach to biopsy. Augmented prophylaxis contradicts the Centers for Disease Control and Prevention and World Health Organization recommendations due to poor antibiotic stewardship [28]. Antibiotics are a limited resource: the more these are used today, the less likely they will still be effective in the future. For instance, augmented prophylaxis increased fluoroquinolone-resistant bacteria on rectal swab cultures from 25% to 33% within 3 yr of implementation in the largest study of transrectal biopsy infection [29]. Moreover, a Michigan statewide quality improvement initiative using largely augmented prophylaxis demonstrated mixed results with no change and an increase in postbiopsy infections at some sites [30].

Targeted prophylaxis is also the most exacting prophylaxis approach as a comparator to demonstrate whether transperineal biopsy lowers infections significantly. Indeed, fluoroquinolone resistance was 23% in NYP-WCM, the site of the highest enrollment, antibiograms and demonstrated cross-resistance to sulfamethoxazole-trimethoprim, ceftriaxone, and gentamicin at 53%, 37%, and 18% (Supplementary material), respectively, highlighting the potential pitfalls of augmented prophylaxis due to broad and significant cross-resistance. These concerns are exacerbated by the significant US geographic variation in fluoroquinolone resistance, ranging from 9.4% to 60.9% [31], and our multicenter study design increases the generalizability of our findings. Although a prospective comparison of targeted versus augmented prophylaxis for transrectal biopsy is lacking, it is obsolete if others validate the absence of infections with transperineal biopsy, particularly with the European Commission ban on empiric fluoroquinolone prophylaxis [11].

Our findings must be interpreted in the context of the study design. The difference in infectious complications by the biopsy approach did not reach statistical significance. Several factors suggest that we underestimated the true difference in infection between the transrectal and transperineal approaches. First, we studied only first-time biopsy, and it has been reported that there is a higher risk of infection on repeat transrectal biopsy [32]. We are currently investigating this possibility in a second randomized trial that is actively accruing [33]. Second, the infection rate in the transrectal arm was very low compared with the 5–7% of infectious complications reported in guideline reviews [5]. This may be due to relative understudy of targeted transrectal biopsy prophylaxis, particularly in a multi-institutional fashion. Third, while the difference in infections did not reach statistical significance, the absence of transperineal infections without prophylaxis is noteworthy for several reasons: (1) antibiotic prophylaxis is the standard of care for transrectal prostate biopsy; (2) the absence of prescribing antibiotics for transperineal biopsy avoids the concerns of allergic reactions, side effects, and cross-reactions with other medications, and obviates the need to prescribe antibiotics and questions regarding the timing of antibiotic prophylaxis and side effects from patients, and clarifications from the pharmacy; and (3) the absence of prophylaxis avoids cancelling biopsies due to patient noncompliance and enables biopsies to be performed at the same visit as the initial evaluation, which is particularly beneficial for traveling patients. Finally, we excluded participants who were randomized but did not undergo biopsy from the intent-to-treat analysis because they are not at a risk for biopsy-related infection.

Our study also gives support to the use of two-stage consent methodology. Here, we demonstrate that two-stage consent can be implemented successfully with consequent excellent accrual. In particular (see the Supplementary material), accrual was about four-fold higher for urologists who used two-stage consent than for those who had adopted transperineal biopsy as the standard of care and thus needed to use a one-stage approach. This is noteworthy given that it is relatively common for even small trials to fail to accrue, particularly those involving randomization to surgical interventions [34].

5. Conclusions

We demonstrated that transperineal prostate biopsy without antibiotic prophylaxis was safe and infection free, and has similar detection of prostate cancer. While transperineal biopsy participants reported more pain and discomfort, these effects were small and resolved by 7 d.

Transrectal biopsy with targeted prophylaxis achieved similar infection rates but required rectal cultures and careful attention to antibiotic selection and administration. Consideration of these factors and antibiotic stewardship should guide clinical decision-making.