Effectiveness of Preanalytic Practices on Contamination and Diagnostic Accuracy of Urine Cultures: a Laboratory Medicine Best Practices Systematic Review and Meta-analysis.

BACKGROUND: Urinary tract infection (UTI) in the United States is the most common bacterial infection, and urine cultures often make up the largest portion of workload for a hospital-based microbiology laboratory. Appropriately managing the factors affecting the preanalytic phase of urine culture contributes significantly to the generation of meaningful culture results that ultimately affect patient diagnosis and management. Urine culture contamination can be reduced with proper techniques for urine collection, preservation, storage, and transport, the major factors affecting the preanalytic phase of urine culture. OBJECTIVES: The purposes of this review were to identify and evaluate preanalytic practices associated with urine specimens and to assess their impact on the accuracy of urine culture microbiology. Specific practices included collection methods for men, women, and children; preservation of urine samples in boric acid solutions; and the effect of refrigeration on stored urine. Practice efficacy and effectiveness were measured by two parameters: reduction of urine culture contamination and increased accuracy of patient diagnosis. The CDC Laboratory Medicine Best Practices (LMBP) initiative's systematic review method for assessment of quality improvement (QI) practices was employed. Results were then translated into evidence-based practice guidelines. SEARCH STRATEGY: A search of three electronic bibliographic databases (PubMed, SCOPUS, and CINAHL), as well as hand searching of bibliographies from relevant information sources, for English-language articles published between 1965 and 2014 was conducted. SELECTION CRITERIA: The search contained the following medical subject headings and key text words: urinary tract infections, UTI, urine/analysis, urine/microbiology, urinalysis, specimen handling, preservation, biological, preservation, boric acid, boric acid/borate, refrigeration, storage, time factors, transportation, transport time, time delay, time factor, timing, urine specimen collection, catheters, indwelling, urinary reservoirs, continent, urinary catheterization, intermittent urethral catheterization, clean voided, midstream, Foley, suprapubic, bacteriological techniques, and microbiological techniques. MAIN RESULTS: Both boric acid and refrigeration adequately preserved urine specimens prior to their processing for up to 24 h. Urine held at room temperature for more than 4 h showed overgrowth of both clinically significant and contaminating microorganisms. The overall strength of this body of evidence, however, was rated as low. For urine specimens collected from women, there was no difference in rates of contamination for midstream urine specimens collected with or without cleansing. The overall strength of this evidence was rated as high. The levels of diagnostic accuracy of midstream urine collection with or without cleansing were similar, although the overall strength of this evidence was rated as low. For urine specimens collected from men, there was a reduction in contamination in favor of midstream clean-catch over first-void specimen collection. The strength of this evidence was rated as high. Only one study compared midstream collection with cleansing to midstream collection without cleansing. Results showed no difference in contamination between the two methods of collection. However, imprecision was due largely to the small event size. The diagnostic accuracy of midstream urine collection from men compared to straight catheterization or suprapubic aspiration was high. However, the overall strength of this body of evidence was rated as low. For urine specimens collected from children and infants, the evidence comparing contamination rates for midstream urine collection with cleansing, midstream collection without cleansing, sterile urine bag collection, and diaper collection pointed to larger reductions in the odds of contamination in favor of midstream collection with cleansing over the other methods of collection. This body of evidence was rated as high. The accuracy of diagnosis of urinary tract infection from midstream clean-catch urine specimens, sterile urine bag specimens, or diaper specimens compared to straight catheterization or suprapubic aspiration was varied. AUTHORS' CONCLUSIONS: No recommendation for or against is made for delayed processing of urine stored at room temperature, refrigerated, or preserved in boric acid. This does not preclude the use of refrigeration or chemical preservatives in clinical practice. It does indicate, however, that more systematic studies evaluating the utility of these measures are needed. If noninvasive collection is being considered for women, midstream collection with cleansing is recommended, but no recommendation for or against is made for midstream collection without cleansing. If noninvasive collection is being considered for men, midstream collection with cleansing is recommended and collection of first-void urine is not recommended. No recommendation for or against is made for collection of midstream urine without cleansing. If noninvasive collection is being considered for children, midstream collection with cleansing is recommended and collection in sterile urine bags, from diapers, or midstream without cleansing is not recommended. Whether midstream collection with cleansing can be routinely used in place of catheterization or suprapubic aspiration is unclear. The data suggest that midstream collection with cleansing is accurate for the diagnosis of urinary tract infections in infants and children and has higher average accuracy than sterile urine bag collection (data for diaper collection were lacking); however, the overall strength of evidence was low, as multivariate modeling could not be performed, and thus no recommendation for or against can be made.

Impact of AmpC overexpression on outcomes of patients with Pseudomonas aeruginosa bacteremia.

AmpC overexpression (AmpC++) is a significant mechanism of beta-lactam resistance in Pseudomonas aeruginosa, but its impact on clinical outcomes is not well established. To examine the influence of AmpC++ on clinical outcomes of patients with P. aeruginosa bacteremia, we screened all bloodstream P. aeruginosa isolates obtained from 2003 to 2006 for AmpC++. Demographics and outcomes were retrospectively compared between patients with P. aeruginosa bacteremia caused by AmpC++ and pan-susceptible strains (wild-type controls). Of the 263 isolates screened, 63 (24.0%) were nonsusceptible to ceftazidime. Clinical data of 42 AmpC++ isolates from 21 patients were compared with 33 control patients. The 2 groups were similar in sex and race. Patients in the AmpC++ group was more likely to receive inappropriate empiric antibiotics (odds ratio [OR] = 67.5; 95% confidence interval [CI], 6.3-720.0) and experience microbiologic persistence (OR = 12.2; 95% CI, 1.7-87.7). In institutions with a high prevalence of AmpC++, empiric therapy with agents with activity against AmpC++ strains may be warranted.

Increased bacterial adherence and biomass in Pseudomonas aeruginosa bacteria exposed to clarithromycin.

Long-term low-dose macrolides alter response in patients with chronic sessile Pseudomonas aeruginosa colonization. We examined the effect of clarithromycin on 1) adherence of P. aeruginosa cells and 2) biofilm formation. A suspended-coupon continuous-flow biofilm reactor model was used. Adherent P. aeruginosa bacteria were established for 24 h, immediately followed by a 24-h continuous-flow operation (CFO) phase with serial sampling. In addition, the effect of clarithromycin on adherent biomass was assessed quantitatively using a colorimetric assay. Isolates preexposed to clarithromycin were more adherent to the suspended coupons than nonexposed isolates (P=0.021). After 2 h of CFO, a 1.30+/-0.86 log colony-forming unit (CFU)/cm(2) decrease was observed in controls compared with a 0.08+/-0.55 log CFU/cm(2) decrease in isolates exposed to clarithromycin. Furthermore, a concentration-dependent increase in biofilm biomass was observed with the addition of clarithromycin in a standard mucoid P. aeruginosa strain (1-64 microg/mL, P<0.001) and 44 clinical P. aeruginosa strains (2 or 32 microg/mL, P<0.001). Clarithromycin increased bacterial adherence to the suspended coupons, and increased biomass was observed in isolates treated with clarithromycin.

Outcomes of bacteremia due to Pseudomonas aeruginosa with reduced susceptibility to piperacillin-tazobactam: implications on the appropriateness of the resistance breakpoint.

BACKGROUND: Bacteremia due to Pseudomonas aeruginosa is associated with grave clinical outcomes. Recent studies have emphasized the importance of appropriate empirical therapy, but controversy arises when piperacillin-tazobactam is used against isolates with reduced susceptibility. METHODS: We performed a retrospective cohort study of pseudomonal bacteremia from 2002 to 2006. Patients were identified by the microbiology laboratory database, and pertinent clinical data (demographic characteristics, baseline Acute Physiology and Chronic Health Evaluation [APACHE] II scores, source of bacteremia, and therapy) were retrieved from the electronic medical records. All patients received appropriate empirical therapy within 24 h of positive culture results. Patients receiving piperacillin-tazobactam were compared with those receiving other agents (control subjects). The primary outcome was 30-day mortality from the first day of bacteremia. RESULTS: A total of 34 bacteremia episodes were identified involving isolates with reduced susceptibility to piperacillin-tazobactam (minimum inhibitory concentration, 32 or 64 mg/L, reported as susceptible); piperacillin-tazobactam was empirically given in 7 episodes. There was no significant difference in baseline characteristics between the 2 groups. Thirty-day mortality was found to be 85.7% in the piperacillin-tazobactam group and 22.2% in the control group (P = .004). Time to hospital mortality was also found to be shorter in the piperacillin-tazobactam group (P < .001). In the multivariate analysis, 30-day mortality was found to be associated with empirical piperacillin-tazobactam therapy (odds ratio, 220.5; 95% confidence interval, 3.8-12707.4; P = .009), after adjustment for differences in age and APACHE II score. CONCLUSIONS: In P. aeruginosa bacteremia due to isolates with reduced piperacillin-tazobactam susceptibility, empirical piperacillin-tazobactam therapy was associated with increased mortality. Additional studies are warranted to examine the appropriateness of the current Clinical Laboratory Standards Institute resistance breakpoint of piperacillin-tazobactam.

Prevalence, mechanisms, and risk factors of carbapenem resistance in bloodstream isolates of Pseudomonas aeruginosa.

We examined the prevalence of various carbapenem resistance mechanisms in Pseudomonas aeruginosa bloodstream isolates from a university-affiliated hospital. Isolates obtained in 2003 and 2004 were screened for meropenem/imipenem resistance, and clonality was assessed by repetitive-element-based polymerase chain reaction. The presence of carbapenemase and AmpC overexpression was ascertained by spectrophotometric assays. Outer membrane protein profiles were examined by sodium dodecyl sulfate polyacrylamide gel electrophoresis, and efflux pump overexpression was confirmed by Western blotting. We examined 129 nonrepeat isolates; 21 isolates (from 13 distinct clones) were resistant to meropenem or imipenem (prevalence rate = 16.3%). Nineteen (90.5%) carbapenem-resistant isolates had reduced OprD expression, and 6 (28.6%) isolates had overexpression of MexB. Increased length of hospital stay was identified as a significant risk factor for bacteremia due to carbapenem-resistant P. aeruginosa. Understanding the prevalence and mechanism of carbapenem resistance in P. aeruginosa may guide empiric therapy for nosocomial infections in our hospital.

An integrated pharmacoeconomic approach to antimicrobial formulary decision-making.

PURPOSE: The utility of a novel interdisciplinary approach to antimicrobial formulary decision-making was studied. METHODS: Pseudomonas aeruginosa minimum inhibitory concentration (MIC) distribution data for cefepime and ceftazidime were retrieved from nonrepeat isolates obtained from November 2002 to October 2003. Unbound drug exposures were simulated for 5000 patients using the Monte Carlo method. Weighted target attainment rates (TARs) were calculated for cefepime and ceftazidime 1 g every 8 hours and 1 g every 12 hours (infused over 0.5, 2, and 4 hours), using three representative pharmacodynamic targets (percentage of time above the MIC of 67%, 100%, and 400%). RESULTS: MIC data for 1230 nonrepeat P. aeruginosa were analyzed. The MIC at which 90% of the P. aeruginosa isolates were inhibited was 16 and 32 mg/L for cefepime and ceftazidime, respectively. Drug acquisition cost was the highest with cefepime 1 g given every 8 hours (37.56 dollars/day), followed by cefepime 1 g every 12 hours (25.04 dollars/day) and ceftazidime 1 g every 8 hours (22.26 dollars/day). When infused over 0.5 hour, the highest TAR was achieved with cefepime 1 g every 8 hours (82%), followed by ceftazidime 1 g every 8 hours (77%) and cefepime 1 g every 12 hours (66%); ceftazidime 1 g every 8 hours was 70% more cost-effective than cefepime 1 g every 8 hours. Cefepime 1 g every 12 hours, infused over 4 hours, increased the TAR to 89% and was similar in cost-effectiveness to ceftazidime 1 g every 8 hours infused over 0.5 hour. CONCLUSION: An integrated pharmacoeconomic approach to antimicrobial formulary decision-making addressed local resistance patterns, population pharmacokinetics, pharmacodynamics, dosing regimens, and drug acquisition costs. This method appeared to be more realistic and objective than the conventional approach of considering only drug acquisition costs, especially for agents in a similar structural or functional class.

Tolerance of vancomycin for surgical prophylaxis in patients undergoing cardiac surgery and incidence of vancomycin-resistant enterococcus colonization.

BACKGROUND: In 2001, vancomycin replaced cefuroxime for antibiotic prophylaxis in patients undergoing cardiac surgery at our institution due to high rates of surgical site infections caused by methicillin-resistant Staphylococcus spp. However, few data supported the use of vancomycin for surgical prophylaxis. OBJECTIVE: To determine the tolerance of vancomycin for antibiotic prophylaxis and incidence of vancomycin-resistant Enterococcus (VRE) in cardiac surgery patients. METHODS: In 2 separate studies, we assessed the adverse effects in patients given perioperative vancomycin (study 1) and the incidence of VRE in patients given perioperative vancomycin (study 2). Study 1 was a prospective cohort study of patients undergoing coronary artery bypass graft (CABG) or valve replacement surgery given vancomycin (1 dose preoperatively/2 doses postoperatively) for antibiotic prophylaxis between October 2003 and December 2004. Patients were assessed for tolerance to the antibiotic regimen. In study 2, cardiac surgery patients receiving perioperative vancomycin were screened for VRE before therapy and at day 7 of hospitalization. VRE was detected using standard microbiologic procedures. RESULTS: In study 1, 1161 patients (CABG = 75%; valve = 19%; both = 6%) were evaluated. All patients but one (99.9%) were prescribed preoperative vancomycin. Therapy was changed for 34 (2.9%) patients, of which 20 changes were due to physician preference for another antibiotic. The only toxicity that required a change in the vancomycin regimen was red man's syndrome, which was experienced by 9 (0.8%) patients. Four patients did not receive a second postoperative dose due to prior renal insufficiency. Patients were most commonly switched to cefuroxime (n = 26), linezolid (n = 2), cefepime (n = 2), gatifloxacin, cefazolin, levofloxacin, or ceftriaxone (n = 1, each). In study 2, 100 patients were screened for the emergence of VRE colonization. No patient was VRE positive at baseline and 4 (4%) were positive at day 7. CONCLUSIONS: Surgical antibiotic prophylaxis with vancomycin was reasonably well tolerated in CABG and valve replacement surgery, with a 4% incidence of VRE colonization.