Sepsis and ‘superbugs’: should we favour the transperineal over the transrectal approach for prostate biopsy?




  • To determine the rate of hospital re-admission for sepsis after transperineal (TP) biopsy using both local data and worldwide literature, as there is growing interest in TP biopsy as an alternative to transrectal ultrasonography (TRUS)-guided biopsy for patients undergoing repeat prostate biopsy.

Patients and Methods

  • Pooled prospective databases on TP biopsy from multiple centres in Melbourne were queried for rates of re-admission for infection.
  • A literature review of PubMed and Embase was also conducted using the search terms: ‘prostate biopsy, fever, infection, sepsis, septicaemia and complications’.


  • In all, 245 TP biopsies were performed (111 at Alfred Health, 92 at Epworth Healthcare, 38 at Peter MacCallum Cancer Centre, and four at other institutions).
  • The rate of hospital re-admission for infection was zero.
  • The literature review showed that the rate of sepsis after TRUS biopsy appears to be rising with increasing rates of multi-resistant bacteria found in rectal flora, and is as high as 5%.
  • However, the rate of sepsis from published series of TP biopsy approached zero.


  • Both local and international data suggest a negligible rate of sepsis with TP biopsy.
  • This compares to a concerning rise in the rate of sepsis after TRUS biopsy due to the increasing prevalence of multi-resistant bacteria in rectal flora.
  • Although TRUS biopsy is convenient, cheap and quick to perform, we think that TP biopsy should now be offered as an option, not only to patients undergoing repeat prostate biopsy, but to all patients in whom a prostate biopsy is indicated.

carbapenem-resistant enterobacteriaceae


extended-spectrum β-lactamase


intensive care unit


radical prostatectomy




Victorian Transperineal Biopsy Collaboration


Prostate biopsy provides the most important information for clinical decision-making in men suspected of prostate cancer. TRUS-guided biopsy is considered the ‘gold standard’ for obtaining a histological diagnosis of prostate cancer (European Association of Urology Guidelines on Prostate Cancer 2013), and is the most commonly used technique. TRUS biopsy has a short learning curve, can be performed quickly using a simple rectal probe, and is often undertaken in the urologist's office under local anaesthesia. Due to these advantages, millions of TRUS biopsies [1] have been performed around the world since the introduction of systematic sextant biopsy in 1989 [2].

In recent years, there has been a noticeable rise in the rate of TRUS-biopsy sepsis along with the increasing prevalence of multi-resistant organisms [1, 3]. Extended-spectrum β-lactamase (ESBL) and quinolone-resistant bacteria are now commonly found in rectal flora, and their prevalence continues to rise [4]. As a result, broad-spectrum carbapenems, currently one of the last lines of antibiotic defence against such bacteria, are being increasingly used both in treating TRUS-biopsy sepsis and in TRUS-biopsy prophylaxis. Furthermore, there are now reports of carbapenem-resistant Enterobacteriaceae (CRE) [5, 6].

Transperineal (TP) biopsy, using a TRUS probe and brachytherapy template grid, is performed via the perineal skin, avoiding trocar passage through the rectal mucosa. It is therefore thought to confer a comparatively low risk of sepsis and may obviate the use of broad-spectrum antibiotics altogether.

We sought to determine the rate of sepsis after TP biopsy by measuring the rate of hospital re-admission for TP biopsy patients across multiple institutions in Melbourne, Australia. We also performed a systematic review of the current published data to determine the overall rate of TP biopsy sepsis reported to date.

Patients and Methods

The Victorian Transperineal Biopsy Collaboration (VTBC) was formed and ethics approval was obtained from each institution's Institutional Review Board. TP biopsy databases from VTBC centres in Melbourne (Alfred Health, Epworth Healthcare and Peter MacCallum Cancer Centre) were kept prospectively, except for Alfred Health patients between September 2009 and 2011, and included data points for infective complications. Data was obtained from patients’ charts in the public setting and the treating urologists’ patient files in private. The databases were pooled and queried for rates of re-admission for infection from September 2009 to April 2013. The patient characteristics analysed included patient age, PSA level, indication for TP biopsy, α-blocker use, prophylactic antibiotics and number of cores taken. Other results analysed were rate of cancer detection and urinary retention.

All TP biopsies were taken under general anaesthesia in the extended dorsal lithotomy position using a biplanar TRUS probe mounted on a stabiliser and stepper with a brachytherapy template grid. All patients attended for review consultation either in the public outpatient clinic of the hospital at which the TP biopsy was performed, or by the urologist who performed the TP biopsy in the private setting.

A systematic literature review of PubMed and Embase was also conducted using the search terms ‘transperineal, prostate biopsy, fever, infection, sepsis, septicaemia and complications’ from 2003 to the present.

Articles were excluded for the following reasons:

  1. Re-admission or sepsis rate not reported.
  2. Same patient cohorts published in different series.
  3. Patients underwent simultaneous TRUS biopsy where sepsis occurred (unable to determine cause).
  4. Non-English language articles.


From the VTBC databases, 245 TP biopsies were undertaken on 244 patients (111 at Alfred Health, 92 at Epworth Healthcare, 38 at Peter MacCallum Cancer Centre and four at other institutions). Patient characteristics, according to age, PSA level, indication for TP biopsy, number of cores taken, usage of α-blockers and antibiotics are summarised in Table 1.

Table 1. VTBC TP biopsies – patient and procedure characteristics
Age, years:<6061–70>71
Number of patients (%)94 (38.5)131 (53.7)19 (7.8)
PSA level, ng/mL:<44.1–10.0>10.1
Number of patients (%)33 (14)132 (54)80 (32)
Indication for TP biopsy:Rising PSA level with previous negative TRUS biopsyASOther (e.g. high sepsis risk, TRUS contraindicated, IMPACT study, patient/physician preference)
Number of patients (%)106 (44)96 (39)42 (17)
Number of patients (%)100 (41)144 (59) 
Number of cores taken<1415–20>21
Number of TP biopsies (%)5 (2)125 (51)115 (47)
Prophylactic antibiotics, n (%)Yes, 244 (100)  
Antibiotic used, n (%):   
cephalosporin alone14 (6)
cephalosporin and gentamicin38 (16)
cephalosporin, quinolone and gentamicin110 (45)
quinolone alone60 (25)
not specified22 (8)

In our series of 245 TP biopsies, no patients were re-admitted for infective complications. In all, 10 patients (4%) developed acute urinary retention and three (1%) patients had clot retention.

In all, 96 patients were undergoing active surveillance (AS). Of these, 30% were found to have disease upgrading, with 70% of patients undergoing radical prostatectomy (RP) and 30% of patients undergoing radiotherapy.

Prostate cancer was detected in 39% of the remaining 148 patients. Of this cohort, 59% of patients went on to undergo RP, 23% underwent radiotherapy and 16% of patients commenced AS.

A systematic review of the literature found 16 mutually exclusive series of TP biopsies that reported on infective complications (Table 2) [7-22]. In a total of 6609 patients, only five were re-admitted to hospital for sepsis, for an overall rate of just 0.076%.

Table 2. Systematic literature review of re-admission for sepsis after TP biopsy
StudyJournal/year of publicationNumber of patientsPatient admissions for sepsis, n (%)Antibiotic prophylaxisNumber of cores
  1. N/A, not available.
Pepe et al. [7]Urology/201330000Levofloxacin12–24
Vyas et al. [8]BJU Int/20136340Amikacin24–38
Symons et al. [9]BJU Int/20134091Norfloxacin/gentamicin22
Kuru et al. [10]J Urol/20133470N/AN/A
Ekweune et al. [11]BJU Int/20132701Gentamicin/metronidazole28
Dimmen et al. [12]BJU Int/2012691NilN/A
Pal et al. [13]BJU Int/2012400Coamoxiclav/gentamicin36
Suzuki et al. [14]Int J Urol/20095390Levofloxacin14
Kubo et al. [15]Int J Urol/2009450Levofloxacin14
Merrick et al. [16]BJU Int/20081290Not specified24
Hara et al. [17]Urology/20081260Levofloxacin12
Li et al. [18]Urology/20073030N/A24
Yamamoto et al. [19]Int J Clin Oncol/20053001Sublactam/cephalexin/ levofloxacin12
Pinkstaff et al. [20]Urology/20052100Fluoroquinolone21
Miller et al. [21]ANZ J Surg/2005811Nil6
Emiliozzi et al. [22]Urology/20031070Quinolone6
Total 66095 (0.076)  


TRUS Biopsy

In TRUS biopsy, the trocar traverses the rectal mucosa, passing from ‘dirty’ to ‘clean’, contravening the basic surgical principle of avoiding contamination of a sterile environment. As a result, faecal flora may gain access not only to the highly vascular prostate gland, but thereon to the bloodstream, causing life-threatening septicaemia. From the outset of the use of TRUS biopsy, this risk of sepsis was recognised [23]. Prophylactic antibiotics were therefore recommended and are routinely given for TRUS biopsy today, although regimens used vary widely [24].

However, despite prophylactic antibiotics, evidence suggests that the rate of TRUS-biopsy sepsis is rising. Nam et al. [3] reported a population study of 75 190 Canadian men undergoing TRUS biopsy from 1996 to 2005. During this period, the rate of re-admission for infection rose dramatically from 0.6% to 3.6%.

In an international prospective study of 521 TRUS biopsy patients in 2010 and 2011, 3.1% required hospitalisation [1]. Over the same 2 years in Christchurch, New Zealand, 1421 men underwent TRUS biopsy; 2.8% were re-admitted for infection, 10% of whom required admission to an intensive care unit (ICU). In a 2012 report, as many as 5% of 316 UK men undergoing TRUS biopsy using standard antibiotic prophylaxis were re-admitted with infection. They noted independent risk factors of recent travel or antibiotic use and all positive cultures grew fluoroquinolone-resistant E. coli [25].

Multi-resistant Bacteria

ESBL and quinolone-resistant bacteria are now commonly found in rectal flora, and their prevalence continues to rise [26]. In a large Spanish study, 28 307 E. coli blood isolates were examined from 2001 to 2009. The rate of ciprofloxacin-resistance increased from 17.6% to 32.7%. Over the same period, the community use of levofloxacin increased 307.2%. They concluded that community use of quinolones was a major factor in the rapid increase in resistance [4].

In the face of increasing resistance, carbapenems are increasingly being used not only to treat TRUS-biopsy sepsis, but also as routine prophylaxis. These antibiotics are nearing the last line of defence against such bacteria. Furthermore, there are now reports of CRE found across the world, in patients in a paediatric hospital in UK [6] and an ICU in Australia [5].

Routine use of fluoroquinolones for prophylaxis in TRUS biopsy is therefore becoming less effective [27], and may be accelerating the development of antibiotic resistance. Based on current trends, its ongoing use in TRUS-biopsy prophylaxis would appear to be unsustainable. Recognising the serious nature of this problem, alternative approaches have been sought. The use of pre-biopsy rectal screening swabs and instillation of disinfectant enemas have met with limited success as these techniques retain the fundamental problem of transrectal trocar passage [28, 29].

TP Biopsy

TP biopsy has typically been reserved for patients undergoing repeat biopsy, either for AS of low-risk cancer, or for a rising PSA level despite a negative TRUS biopsy [12].

TP biopsy has several drawbacks that have prevented more widespread use. It usually requires a general anaesthetic, although nerve block techniques and local anaesthesia have been reported [7, 18, 21, 30]. TP biopsy also takes longer to take than a TRUS biopsy, and typically involves taking more cores. It is therefore more costly and a significant drain on resources, including reporting of the pathology. Furthermore, for template grid biopsy, specialised equipment is required, including a brachytherapy grid, stepper, stabiliser and bi-plane transducer.

A systematic review and meta-analysis of all randomised and case-control trials comparing TRUS to TP biopsy was reported by Shen et al. [31] in 2012. They found equivalence in cancer detection rate when comparing sextant, extended core and saturation biopsies for each technique. Studies have even reported superiority of TP over TRUS biopsy for detection of anterior tumours [12, 32, 33].

The reported rate of urinary retention in TP biopsy is similar to TRUS biopsy [31]. The largest series of TP biopsy reported a rate of 6.7% of 3000 patients, with only 56% taking an α-blocker [7]. However, the Guy's Hospital series of 634 men reported a retention rate of 1.7% [8].

TP biopsy is not associated with a greater risk of erectile dysfunction [16] and its effect on dissection in RP appears to be anecdotal only [34].

Perhaps the most significant advantage of TP biopsy is its avoidance of transrectal trocar passage. Evidence of this advantage was first described in a small but elegant study in 1982. Thompson et al. [23] found lower rates of bacteraemia and plasma endotoxin levels when using the TP approach, and that any bacteraemia was predominantly with skin commensals only.

The present finding of zero sepsis echoes that of published TP biopsy series, showing negligible rates of sepsis, which are 40 to 70 times lower than those currently reported for TRUS biopsy. The importance of this lies not only in the dramatic risk reduction of significant and potentially life-threatening morbidity to patients. From the public health and antimicrobial stewardship perspective, TP biopsy also enables the safe avoidance of the use of carbapenems which, in turn, may slow the development of further resistance. Indeed, even fluoroquinolones are unlikely to be required for TP biopsy. This subject is currently under investigation by the VTBC.

One limitation of the present study was data collection by chart review. Although the vast majority of the database was populated prospectively, it is possible that episodes of re-admission for infective complications were missed. However, this is felt to be unlikely as all patients undergoing the procedure were reviewed either in the clinic of the institution where the biopsy was performed, or in the office of the urologist who performed the procedure.

Another weakness of the present study was the lack of standardisation of TP biopsy techniques across our institutions. Varying templates, core numbers and prophylactic antibiotics were used. However, despite these variations the sepsis rate of zero was consistent. Standardisation of technique and alignment of our database with the minimal dataset proposed by the Ginsburg Study Group [35] is underway.

In conclusion, in today's environment of rising rates of TRUS-biopsy sepsis and antibiotic resistance, we think that the risk–benefit ratio has now shifted sufficiently to warrant offering TP biopsy as an option to all men in whom a prostate biopsy is indicated.

This paradigm shift is likely to have a significant impact on health resources. Whilst the procedure of TP biopsy alone is clearly more costly than TRUS biopsy, the savings from its lack of infective complications must also be considered. This will include savings on hospital re-admission, carbapenem use, stays in ICU, prolonged antibiotic therapy in the community, and loss of productivity. This subject is also under investigation by our collaborative group.

Conflict of Interest

None declared.