2.1

Study design, population, and definitions

In this retrospective observational before-and-after study, we reviewed the medical records of patients who underwent TRUS-PBx between January 2014 and December 2023 at the Komaki City Hospital (558 beds until April 2018 and 520 beds since May 2019), a tertiary care center in Aichi Prefecture, Japan.

AP administration during TRUS-PBx was divided into 3 periods. The route of administration and dosage of the antimicrobials were as follows: cefazolin (CEZ) (IV), 1 g; CMZ (IV), 1 g; and LVFX (oral), 500 mg. From January 2014 to July 2017, the AP consisted of a single dose (SD) of CEZ before TRUS-PBx and multiple doses (MD, 5 days) of LVFX after TRUS-PBx. From August 2017 to April 2019, CEZ and LVFX SDs were administered as AP before TRUS-PBx. Starting from May 2019, SDs of CMZ and LVFX were administered as AP before TRUS-PBx. The exclusion criteria were as follows: age < 18 years, untraceable cases, use of alternative antibiotics, use of antibiotics for infections other than those related to TRUS-PBx within 14 days after TRUS-PBx, use of antibiotics within 14 days before TRUS-PBx, and data loss.

All TRUS-PBx procedures included overnight hospitalization for observing complications. Patients without any complications were discharged the following morning. All biopsies were performed after rectal cleansing using a single glycerin enema and intrarectal disinfection using povidone-iodine.

The following data were collected from electronic medical records: age, serum prostate-specific antigen, diagnosis of prostate cancer, prostate size, obesity (body mass index [BMI] ≥ 25), history of antibiotic use within 6 months, hospitalization within last 3 months, history of prostatitis, and history of prostate biopsy. The following data were collected as risk factors: the presence of an indwelling urethral catheter, dysuria (residual urinary volume ≥ 100 ml), prostate volume ≥ 75 ml, poorly controlled diabetes mellitus (HbA1c ≥ 8.0%), and corticosteroid use [ ].

The primary outcome was the incidence of febrile UTI, defined as a diagnosed UTI with a fever of 37.8°C or higher within 14 days after TRUS-PBx. Secondary outcomes included incidences of urosepsis, bacteremia, abscess, and readmission. Urosepsis was diagnosed if the symptoms met two or more of the following criteria: (1) a temperature > 38°C or < 36°C, (2) heart rate > 90/min, (3) respiratory rate > 20/min or PaCO ₂ < 32 mmHg, and (4) a peripheral blood leukocyte count > 12,000/mm ³ or < 4,000/mm ³ [ ].

2.2

Comparison of costs

The costs associated with TRUS-PBx and subsequent ICs were based on a cost-effectiveness model [ ]. In this model, we estimated healthcare-related costs based on the total cost of AP for TRUS-PBx, the cost per febrile UTI, and the incidence of febrile UTI after TRUS-PBx. The cost of AP was calculated based on the prevailing drug prices in Japan as of December 2023. The cost of each IV antibiotic agent was calculated after adding the cost of the two-chamber formulation saline. The cost of each drug (in Japanese Yen, JPY) was as follows: LVFX, JPY 77.4; CMZ, JPY 441.0; CEZ, JPY 291.0; and two-chamber saline formulation, JPY 204.0. The cost per febrile UTI indicates the average cost of additional hospitalization and outpatient medical expenses incurred owing to ICs per patient. Healthcare-related costs per patient were estimated using the following formula: cost of total prophylactic antibiotics (including the two-chamber formulation of saline) + incidence of febrile UTI × average cost per febrile UTI. A conversion rate of 1 US dollar (USD) to 144 JPY was used as of December 2023.

2.3

Statistical analyses

Categorical variables were expressed as totals and percentages and analyzed using Fisher’s exact test, whereas continuous variables were expressed as means and interquartile ranges (IQR) and analyzed using the Mann–Whitney U test. We performed a stratified analysis to evaluate the relationship between risk factors and the incidence of febrile UTI by comparing patients with and without risk factors using Fisher’s exact test.

For the primary analysis, we used multivariable logistic regression to assess whether using CMZ+LVFX as a prophylactic reduced the incidence of febrile UTI. Adjusted variables were first selected based on their clinical relevance and then further refined using statistical methods [ ]. From a clinical perspective, the following adjusted variables were considered: age ≥ 73, number of biopsy cores > 12, obesity (BMI ≥ 25), history of antibiotic use within 6 months, hospitalization within last 3 months, history of prostatitis, history of prostate biopsy, indwelling urethral catheter, prostate volume ≥ 75 ml, dysuria (residual urinary volume ≥ 100 ml), diabetes mellitus (HbA1c ≥ 8.0%), and corticosteroid use. Using statistical methods, we selected the main model as the one with the lowest Akaike Information Criterion, with CMZ+LVFX designated as the prophylactic and included in the model as the primary explanatory variable.

A significance level of 5% was used for all analyses. All statistical analyses were performed using EZR, which is based on R (version 4.3.1; R Foundation for Statistical Computing, Vienna, Austria) [ ].

2.4

Ethics statement

This study adhered to the Japanese ethical guidelines for epidemiological studies, and all study protocols were approved by the Institutional Review Board of Komaki City Hospital (Approval No. 231048). The requirement for written informed consent was waived owing to the retrospective nature of the study.