SEARCH

SEARCH BY CITATION

Keywords:

  • prostate;
  • biopsy;
  • infection;
  • E. coli;
  • drug resistance;
  • complications

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Study Type – Prognosis (case series)

Level of Evidence 4

What's known on the subject? and What does the study add?

Septicaemia is the most frequent cause of hospitalization after transtrectal prostate biopsy; fatalities have been reported and the incidence is on the rise. This study shows that men with a history of recent international travel or antibiotic use have up to four times increased risk of septicaemia and hospitalization. When they do occur, infections are usually due to multi-resistant E coli and additional care, e.g. delay before biopsy, different antibiotic prophylaxis or transperineal biopsy, should be considered in these cases.

OBJECTIVE

  • • 
    To study the infection rate after prostate biopsy in those who have travelled overseas or used antibiotics in the 4 weeks before biopsy.

PATIENTS AND METHODS

  • • 
    A total of 316 men with a mean (range) age of 61 (45–85) years were studied. All had undergone transrectal ultrasonography (TRUS)-guided prostate biopsy after standard antibiotic prophylaxis.
  • • 
    Before their biopsy the patients were risk stratified and a history of recent international travel or antibiotic use was recorded.
  • • 
    Those who suffered sufficiently severe infection/sepsis so as to require hospitalization were identified at the end of the study period.
  • • 
    The characteristics of these patients and the types of infections were explored and the relative risk (RR) of infection after recent travel or antibiotic use was calculated.

RESULTS

  • • 
    Of the 316 men, 16 were hospitalized with infection.
  • • 
    The group with (n= 16) and without (n= 300) infection were equivalent in age, prostate-specific antigen level, disease status and number of biopsy cores taken.
  • • 
    Either recent travel or antibiotic use were independent risk factors for infection [travel: 8/16 vs 76/300; P= 0.04; RR 2.7 and antibiotic use: 4/16 vs 20/300; P= 0.025; RR 4]. There was no significant pattern in the countries visited or the type of antibiotic used.
  • • 
    Culture results were positive in 10/16 men, and all cultures grew multiresistant Escherichia coli. The strains were uniformly resistant to ciprofloxacin and amoxycillin, and variably resistant to gentamicin and co-amoxiclav, but nearly all were sensitive to meropenem.
  • • 
    All patients made a full recovery after antibiotic and supportive treatment.

CONCLUSIONS

  • • 
    Either recent international travel or antibiotic use are independent risk factors for severe infection after TRUS-guided prostate biopsy.
  • • 
    When infection does occur it should be treated aggressively as the causative agent is usually a multiresistant E. coli.

Abbreviation
RR

relative risk

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Infection and sepsis after TRUS-guided prostate biopsy is a risk, even after pre-procedural antibiotic prophylaxis, and as prostate biopsy is such a commonly performed procedure, many men will suffer this complication. The causative organisms are usually the colonic commensals and infection can be unusually severe, amounting to full blown septicaemia requiring hospital care in some men. The frequency is estimated to be 1–2% [1,2], but there are concerns that the infection rate is increasing and more recent series cite an infection rate of up to 5% [3]. Rising resistance to the 5-fluoroquinolone agents, which are the most commonly used prophylactic antimicrobials [1,2], has been blamed. As a group, the 5-fluoroquinolone agents have high bio-availability within prostate tissue. Until recently, these agents have also had a good range of coverage across the colonic flora, e.g. Escherichia coli, but fluoroquinolone resistance is spreading [3] and threatens to undermine the risk–benefit profile of transrectal prostate biopsy. These concerns have been widely publicized [4].

Resistant bacteria are also more prevalent in some countries. Antibiotic overuse or misuse has been blamed, but a wider dissemination of resistant organisms resulting from globalization and international travel may also be a factor [5]. The spread of multiresistant E. coli is of particular concern, and those who have recently been exposed to antibiotic therapy, or travellers returning from areas with endemic antibiotic resistance, may be at increased risk of infection after prostate biopsy. The primary aim of this prospective study was to see whether a history of recent overseas travel or antibiotic use increases the risk of infection after prostate biopsy, an area that has not been studied previously. The secondary aims were to explore the infection types and antibiotic susceptibility in this sub-group. The study focused on infections of sufficient severity to require inpatient hospital care.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The setting of the study was an outpatient unit in a major, international city (London, UK) and specializing in the management of prostate diseases. All patients who underwent TRUS-guided prostate biopsy between June 2008 and June 2010 were recruited. Patients had been referred with suspected prostate cancer, either because of an increased age-corrected PSA level and/or an abnormal DRE. All patients gave their consent for TRUS-guided prostate biopsy.

The start date of the study coincided with the introduction of a policy of pre-biopsy risk stratification. A history of allergy, recent gastroenteritis or UTI was noted. Those on anticoagulants or antiplatelet agents (e.g. warfarin, clopidogrel or aspirin) underwent appropriate pharmacological manipulation to normalize their bleeding status before the biopsy. A mid-stream urine sample was sent for culture to identify those with concurrent UTI or colonization. A history of recent overseas travel (defined as within the preceding 4 weeks), including destinations, was recorded. Antibiotic use in the preceding 4 weeks was also identified, and the indication and the agents used were recorded.

All patients received the same antibiotic prophylaxis, regardless of a history of travel or recent antibiotic use: 5 days of ciprofloxacin (500 mg twice daily) starting 24 h before biopsy. Immediately before the biopsy all patients were given 120–160 mg gentamicin i.v. and 1 g metronidozole suppository per rectum. Between 6 and 12 prostate biopsies were undertaken under TRUS guidance. An 18-G disposable needle was used. Patients were discharged after a short period of observation (typically 30 min) and once having urinated. A high oral fluid intake was advised and written information was provided regarding complications; in particular, patients were advised to call the centre on a 24-h emergency number if they noted any signs of sepsis (defined as a high fever with rigors). Those with clinical suspicion of septicaemia were immediately admitted to hospital. On admission, urine and blood samples were taken for culture and i.v. antibiotics commenced. The antibiotic regime was subsequently adjusted according to the culture results.

All patients were seen for follow-up 1–2 weeks after the biopsy and a full clinical outcome record of all cases was available, supplemented with retrieved information from the hospital records. Relevant data for this study were extracted from the main database and entered into a separate spreadsheet. Once the characteristics of all patients were recorded, the incidence of infection was calculated and the relative risk (RR) in travellers and after recent antibiotic use was calculated. A commercially available statistics program (GraphPad InStat, version 3.05 for Windows 95/NT, GraphPad Software, San Diego, USA) was used and a P value of <0.05 was considered to indicate statistical significance.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Over the study period 316 men underwent prostate biopsy. Sufficient clinical data were available for every patient and there were no exclusions. In all, 16 patients were hospitalized with infection (nine in the first year of study and seven in the second year; P≥ 0.05) and the onset of fever and rigors was at a median (range) of 2 (1–4) days after biopsy. The characteristics of the patients with and without infection are shown in Table 1. The two groups are comparable. All mid-stream urine cultures taken before the biopsy were sterile, except one that showed mixed flora that were interpreted as contamination.

Table 1.  Characteritics of 316 men who underwent TRUS-guided prostate biopsy
 No infectionInfection
  • *

    In those patients with malignancy.

Patients, n30016
Mean (range) age, years63 (45–83)67 (53–85)
Median (range) PSA, µmol/L8.4 (2.8–172)7.3 (3.2–17)
Median (range) no. of biopsy cores12 (6–12)12 (12)
Benign/malignant126/1746/10
Median (range) Gleason score*6 (6–10)6 (6–9)
Median (range) no. of cores positive*3 (1–12)4 (1–7)
Travel <4 weeks before biopsy, n76 (46 Europe/North America; 17 Asia; 11 Africa; 2 South America)8 (4 Africa; 3 Europe/North America; 1 Asia)
Antibiotic use <4 weeks before biopsy20 (10 unknown; 3 amoxycillin; 3 ciprofloxacin; 2 trimethoprim; 1 penicillin)4 (3 ciprofloxacin; 1 unknown)

Eighty four patients had travelled overseas in the previous 4 weeks, and there was a significantly higher proportion of travellers among those who were hospitalized than among those who were not (8/16 vs 76/300; P= 0.04, Fisher's exact test). The RR of septicaemia after recent travel was 2.7 (Table 2). The countries visited are given in Table 1. More of those who suffered infection had recently visited Africa, Asia or South America but this difference was not significant compared with other countries (5/8 vs 30/76; P= 0.26, Fisher's exact test). Patients with a history of recent antibiotic use also had an increased incidence of septicaemia (P= 0.025; RR 4; Table 2). A wide variety of antimicrobials were used (Table 1) and there was no specific association with individual agents used. We could not show that combined travel and antibiotic use had a higher RR in this cohort.

Table 2.  The increased RR of septicaemia after prostate biopsy after recent travel or antibiotic use
 Travel <4 weeks before biopsyAntibiotic use <4 weeks before biopsyAntibiotics and travel <4 weeks before biopsy
  • *

    16/316 (5%) were hospitalized with post-prostate biopsy septicaemia.

  • †Fisher's exact test.

  • ‡Chi-squared test.

No infection76/30020/3006/300
Infection*8/164/162/16
P0.040.0250.32
RR (95% CI)2.7 (1.0–7.1)4 (1.4–11) 

All 16 patients made a full recovery from septicaemia. Positive cultures, either urine or blood, were obtained from 10/16 cases. All 10 had multiresistant E. coli infection: all cultures were resistant to ciprofloxacin and amoxycillin, with variable resistance to gentamicin and co-amoxiclav, but 9/10 were sensitive to meropenem (meropenem sensitivity was not tested for in the 10th patient). These details are given in Table 3. Those with sterile cultures were empirically treated with meropenem for 7 days. One patient (patient number 8, Table 3) suffered a clinical relapse 10 days later. On the first admission urine cultures grew multiresistant E. coli, sensitive to and treated with meropenem for 7 days. On the second admission, blood cultures grew Enterococcus faecalis, sensitive to teicoplanin and gentamicin. This patient made a full recovery after further treatment with gentamicin.

Table 3.  Bacterial culture results and antibiotic sensitivity in 10 patients hospitalized with septicaemia after transrectal prostate biopsy
Patient numberOrganismBloodUrineResistanceSensitive
  1. Cipro, ciprofloxacin; Amoxy, amoxycillin; Tri, trimethoprim; Co-A, co-amoxiclav; Gent, gentamicin, Taz, tazocin; Ceph, cephradine; Nitro, nitrofurantoin; Ami, amikacin; Mer, meropenem; Cef, cefuroxime. *Patient suffered a clinical relapse (details in text).

 1E. coli+Cipro, Amoxy, TriGent, Ceph, Nitro, Ami, Co-A, Mer, Taz
 2E. coli+Cipro, Tri, AmoxyGent, Co-A, Cef, Nitro
 3E. coli+Cipro, Co-A, Amoxy TriMer, Gent, Nitro, Co-A
 4E. coli++Cipro, Amoxy, GentNitro, Ami, Cef, Co-A, Mer, Taz
 5E. coli+Cipro, Amoxy, TriMer, Ami, Cipro, Taz, Cef, Nitro
 6E. coli+Cipro, Amoxy, Gent, Ceph, Tri, Co-AAmi, Cef, Mer, Taz
 7E. coli+Cipro, Amoxy, Gent, TriMer,Ceph,Nitro,Ami, Taz
 8*E. coli+Cipro, Amoxy, Co-A, Tri, Gent, TazAmi, Cef, Mer, Nitro
 9E. coli+Cipro, Amoxy, Tri, Ceph, TazMer, Gent, Ami
10E. coli+Cipro, Amoxy, TriGent, Mer, Taz

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The present study shows that recent overseas travel or antibiotic use are independent risk factors for severe infection after prostate biopsy, with a 2.7 and 4 times greater risk, respectively. The present study is the first to show an increased complication rate in this sub-group of patients. We have been able to study this because of our particular patient population, which includes a high percentage of frequently travelling patients, but, because of the growth in global travel, our findings should be of relevance to all prostate clinics, especially those based in major or urban centres serving an international population. The emerging medical challenges of globalization and medical tourism, have been discussed recently [5] and increased risk after prostate biopsy in travellers, and those who have recently used antibiotics, should be seen as part of this developing medical scenario. Our results should be interpreted in the context of these global changes, but their importance is underscored by a suspected general increase in infection after prostate biopsy. In a study of >75 000 men who underwent prostate biopsy between 1996 and 2005, the 30-day hospitalization rate increased from 1 to 4.1% [6]. In another study, infections increased from 1.7% to 4.8% between the years 2004 and 2006, with fluoroquinolone-resistant organisms accounting for most of the increase [3]. These infections are also often severe, can be fatal [7] and the 30-day mortality rate has been estimated to be 0.09% [6].

Antibiotic resistance is increasing worldwide, but is particularly prevalent in Asia [8] and Africa [9]. In one document, 23–37% of clinical isolates of enterobacteria (including E. coli) had reduced antibiotic susceptibility in India and China [8]. Between 12 and 18% of E. coli infections were 5-fluoroquinolone resistant in one major African city [9] and this is recognized as a part of a general trend in that continent. Parallel developments are also being seen in Europe. In one recent study, fluoroquinolone-resistant E. coli strains were seen in 3–29% of infected patients in various European nations [10]. Similar trends have been seen in clinical isolates in the USA [11] and nosocomial infections in the UK [12]. These changes are complicated by multiresistance and coresistance. Resistance to ciprofloxacin often co-exists with resistance to other fluoroquinolones, as well as some of the other prophylactic agents used before prostate biopsy, e.g. co-amoxiclav and amoxicillin.

The normal colonic flora is part of a dynamic environment, continually harmonizing with the local conditions. Hospitalization and recent antibiotic use are recognized causes of change in the intestinal commensals. In a longitudinal study of outpatients the prevalence of resistant organisms doubled during antibiotic therapy from 18–33% to 38–67% [13]. Travel can also adversely affect the bowel milieu. In one study of travellers, the frequency of antibiotic resistant E. coli commensals increased from 7.8% to 49% after travel [14], with antibiotic use during travel being an additional risk factor. Colonization was greater in those who had visited Asia, and 18% still had resistant strains 6 months after returning. In a similar study of healthy volunteers, travellers were found to be carriers of resistant strains: 24% of healthy travellers returned with resistant E. coli on stool culture, with gastroenteritis during the holiday period being an additional risk factor favouring colonization [15]. The bowel flora also changes very quickly, within 10 days of travel in the study by Orskov et al. [16]. The link between travel and changed bowel commensals has been shown by other studies as well [17], but although these previous studies have proven that the background intestinal flora can rapidly transform, our study is the first to show a link between travel and an increase in a specific procedure-related morbidity.

In recognition of the increasing infection rate after prostate biopsy, different antibiotic prophylaxis agents have been explored. Amikacin prophylaxis has recently been shown to improve the infection rate after prostate biopsy [18]. The superior coverage of amikacin was also noted in the study by Feliciano et al. [3], and has been recommended for initial treatment of severe infections after prostate biopsy [19]. An alternative approach would be wider use of transperineal prostate biopsy [20], which has a much lower serious infection rate as the rectum is not transgressed [21]; however, this procedure is not as convenient as it requires day-case facilities and generally also needs to be performed under general anaesthesia. Urinary retention is also more frequent [22].. Wider use of transperineal prostate biopsy may not be cost- or resource-efficient, but with the rise in infection and antibiotic resistance, there should be an interest in identifying high-risk groups that may be selected for transperineal biopsy or offered alternative antimicrobial agents and more intensively followed after biopsy. As we have shown, risk stratification before biopsy can help identify a high-risk group. Delaying prostate biopsy after antibiotic use or recent travel is a simple option, but it is not known for how long the biopsy should be deferred. One study showed that intestinal flora can revert to its native state as early as 2 weeks after antibiotic use [8], but travellers may harbour resistant organisms even 6 months after their return [13]. Further data on these aspects would be useful.

Some limitations of our study should be discussed. The study was centred on severe infections as these are of greatest concern after prostate biopsy and, although we are confident that we have captured all such cases, because of the study design we may have underestimated the total number of infections after prostate biopsy. Some may only have had mild prostatitis or epididymitis managed with a prolonged course of oral antibiotics. Data about antibiotic use and gastroenteritis during travel is dependent on patient recall and we could not control for the possibility of recall bias. Similar limitations apply to recollections about the antibiotic agents used. Our results may also not be universally applicable, as different countries and cities will have varying patterns in their travelling population, antibiotic usage and microbial resistance. Further data from other countries would be helpful.

In conclusion, we have shown that either recent travel or antibiotic use are independent risk factors for an increased risk of severe infection after TRUS-guided prostate biopsy. When cultures were positive, infections were all attributable to multiresistant E. coli. We now more closely counsel those we feel are at increased risk or advise transperineal prostate biopsy instead; and when an infection does occur we urgently arrange for hospital care. Patients are started on meropenem until antibiotic susceptibilities are known. Our data show that amikacin or tazocin would be suitable alternative choices for blind treatment of post-prostate biopsy septicaemia.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The authors are grateful to all the staff in the Prostate Centre, who helped with data collection. Prokar Dasgupta acknowledges financial support from the Department of Health via the National Institute for Health Research (NIHR) comprehensive Biomedical Research Centre award to Guy's & St Thomas' NHS Foundation Trust in partnership with King's College London and King's College Hospital NHS Foundation Trust. He also acknowledges the support of the MRC Centre for Transplantation.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES