Utility of Urine Cultures in Predicting Blood Culture Susceptibilities in Patients with Bacteremic Urinary Tract Infection

In a review of 428 patients with bacteremic urinary tract infections, urine culture susceptibility results accurately predicted blood culture results when the same organism was isolated from both cultures. Early targeted therapy using urine culture results can potentially reduce broad-spectrum antibiotic exposure and its associated adverse effects and the length of hospitalization.

ABSTRACT

In a review of 428 patients with bacteremic urinary tract infections, urine culture susceptibility results accurately predicted blood culture results when the same organism was isolated from both cultures. Early targeted therapy using urine culture results can potentially reduce broad-spectrum antibiotic exposure and its associated adverse effects and the length of hospitalization.

INTRODUCTION

Urine and blood cultures are frequently collected from patients undergoing evaluation for infection, but due to differences in culturing methods and inoculum concentration, urine culture susceptibility results are often available before blood culture results are finalized. We studied how well urine culture susceptibility results predict blood culture results in patients diagnosed with bacteremic urinary tract infection (UTI).

We conducted a retrospective review of adult patients admitted to Sunnybrook Health Sciences Centre (SHSC) from 1 May 2010 to 31 July 2015 with a diagnosis of bacteremic UTI, with urine and blood cultures (drawn within the same 24-hour period) growing the same organism. Gram-positive organisms and polymicrobial bacteremia were excluded. Urine culture susceptibility results (susceptible and nonsusceptible) were compared with the corresponding blood culture results for the following antibiotics: ampicillin, amoxicillin-clavulanate, cefazolin, ceftriaxone, ceftazidime, ciprofloxacin, ertapenem, gentamicin, meropenem, piperacillin-tazobactam, tobramycin, and trimethoprim-sulfamethoxazole. Antibiotic susceptibilities reported as intermediate or resistant were categorized together as nonsusceptible. SHSC is a 627-bed academic health sciences center in Toronto, Canada. The microbiology laboratory uses smudge plate preparation and serum separator tubes (1) with matrix-assisted laser desorption ionization–time of flight mass spectrometry (Bruker Biotyper; Bruker Daltonics, Inc., MA, USA) to expedite the identification of Gram-negative blood culture isolates and Vitek-2 antimicrobial cards (bioMérieux, Marcy l'Etoile, France) in conjunction with CLSI performance standards for susceptibility testing.

We considered the urine culture as a diagnostic test to predict the susceptibility of the blood culture isolate as the reference standard and computed test characteristics, including sensitivity, specificity, positive predictive value, negative predictive value, and positive and negative likelihood ratios. The potential combinations of matched urine-blood susceptibility results were populated into a two-by-two contingency table, and test characteristics were calculated using the combined totals in each quadrant. Test characteristics were calculated for the overall population and for subgroups based on the antibiotic, organism, presence of a urinary catheter, and location of specimen collection. When calculating test characteristics, each individual organism-antibiotic combination served as the unit of analysis. To assess the time frame for urine and blood susceptibility reporting, the time from laboratory specimen receipt to susceptibility result from 15 consecutive unmatched urine and blood cultures positive for Escherichia coli in January 2018 were recorded. Study approval from the institutional research ethics board was obtained.

A total of 428 patients with bacteremic UTI and matching urine and blood isolates were identified. The most commonly isolated organisms were Escherichia coli (69.4%) and Klebsiella (13.6%) and Enterobacter species (4.7%). The most commonly prescribed empirical therapies were ceftriaxone (49.0%), piperacillin-tazobactam (32.9%), and ciprofloxacin (17.8%).

Among the 428 urine-blood culture pairs, there were 4,807 individual antimicrobial susceptibility test results, of which 4,734 (98.5%) were concordant between urine and blood. Fully concordant susceptibility results, as defined by concordance for all antibiotics tested for that organism, were observed in 381 of 428 (89.0%) urine-blood culture pairs; 33 (7.7%) pairs differed by one antibiotic, 6 (1.4%) by two antibiotics, and 8 (1.9%) by three or more antibiotics. Of the 47 discordant pairs, there were 30 (63.8%) pairs where the discordant urine susceptibility result was more resistant than the blood isolate (urine nonsusceptible isolate with a susceptible blood isolate). Discordant pairs were not associated with any specific antibiotic type, the presence of a urinary catheter, or the location of urine specimen collection. Among the 141 patients treated empirically with piperacillin-tazobactam, 117 (83.0%) had a urine culture result demonstrating susceptibility to a narrower-spectrum antibiotic. Similarly, 19/29 (65.5%) patients treated empirically with a carbapenem had a urine culture demonstrating susceptibility to a narrower-spectrum antibiotic.

The urine culture served as an accurate diagnostic test to predict blood culture susceptibility, with a sensitivity of 0.99 (95% confidence interval [CI], 0.98 to 0.99), specificity of 0.98 (95% CI, 0.97 to 0.99), positive predictive value of 0.99 (95% CI, 0.99 to 1.00), negative predictive value of 0.95 (95% CI, 0.93 to 0.96), positive likelihood ratio of 44.58 (95% CI, 29.21 to 68.05), and negative likelihood ratio of 0.01 (95% CI, 0.01 to 0.02). These test characteristics remained robust across the analyzed subgroups (Table 1). Although Proteus spp. and meropenem-specific urine culture results appeared to have a lower calculated negative predictive value, there were very few resistant organisms within these subgroups, resulting in wide confidence intervals.

TABLE 1.

Test characteristics for urine culture predicting blood culture susceptibility for 428 paired isolates^a

Characteristic (no. of pairs)	SN (95% CI)	SP (95% CI)	PPV (95% CI)	NPV (95% CI)	+LR (95% CI)b	−LR (95% CI)
Overall	0.99 (0.98–0.99)	0.98 (0.97–0.99)	0.99 (0.99–1.00)	0.95 (0.93–0.96)	44.58 (29.21–68.05)	0.01 (0.01–0.02)
Organism
Enterobacter species (20)	0.98 (0.95–1.00)	1.00 (0.94–1.00)	1.00 (0.98–1.00)	0.95 (0.87–0.99)		0.02 (0.01–0.06)
Escherichia coli (297)	0.98 (0.98–0.99)	0.97 (0.96–0.98)	0.99 (0.99–1.00)	0.93 (0.91-0.95)	36.86 (22.73–59.77)	0.02 (0.01–0.02)
Klebsiella species (58)	0.99 (0.98–1.00)	1.00 (0.97–1.00)	1.00 (0.99–1.00)	0.97 (0.93–0.99)		0.01 (0.00–0.02)
Proteus species (19)	0.99 (0.96–1.00)	0.80 (0.44–0.97)	0.99 (0.97–1.00)	0.73 (0.39–0.94)	4.93 (1.43–17.03)	0.02 (0.01–0.06)
Pseudomonas aeruginosa (15)	1.00 (0.95–1.00)	0.99 (0.94–1.00)	0.99 (0.93–1.00)	1.00 (0.96–1.00)	87.00 (12.39–610.68)	0
Other (19)c	1.00 (0.98–1.00)	0.96 (0.87–1.00)	0.99 (0.96–1.00)	1.00 (0.93–1.00)	27.00 (6.93–105.20)	0
Antibioticd
Ampicillin (426)	0.98 (0.94–1.00)	0.98 (0.96–0.99)	0.97 (0.93–0.99)	0.99 (0.97–1.00)	53.53 (22.45–127.61)	0.02 (0.01–0.06)
Amoxicillin-clavulanate (247)	0.98 (0.94–0.99)	0.96 (0.89–0.99)	0.98 (0.95–1.00)	0.95 (0.87–0.99)	24.74 (8.16–75.02)	0.02 (0.01–0.06)
Cefazolin (423)	0.97 (0.94–0.98)	0.99 (0.95–1.00)	1.00 (0.98–1.00)	0.91 (0.85–0.96)	103.61 (14.73–728.92)	0.03 (0.02–0.06)
Ceftriaxone (417)	0.99 (0.98–1.00)	1.00 (0.94–1.00)	1.00 (0.99–1.00)	0.96 (0.88–0.99)		0.01 (0.00–0.03)
Ceftazidime (419)	0.99 (0.98–1.00)	1.00 (0.93–1.00)	1.00 (0.99–1.00)	0.95 (0.85–0.99)		0.01 (0.00–0.03)
Ciprofloxacin (428)	0.98 (0.96–1.00)	0.99 (0.95–1.00)	1.00 (0.98–1.00)	0.95 (0.89–0.98)	98.48 (14.01–692.28)	0.02 (0.01–0.04)
Ertapenem (416)	1.00 (0.99–1.00)	1.00 (0.75–1.00)	1.00 (0.99–1.00)	0.93 (0.66–1.00)		0.00 (0.00–0.02)
Gentamicin (428)	0.99 (0.98–1.00)	0.95 (0.83–0.99)	0.99 (0.98–1.00)	0.93 (0.81–0.99)	20.34 (5.26–78.60)	0.01 (0.00–0.03)
Meropenem (418)	1.00 (0.98–1.00)	1.00 (0.02–1.00)	1.00 (0.99–1.00)	0.33 (0.01–0.91)		0.00 (0.00–0.02)
Piperacillin-tazobactam (341)	0.97 (0.94–0.99)	0.98 (0.89–1.00)	1.00 (0.98–1.00)	0.84 (0.73–0.93)	48.45 (6.96–337.30)	0.03 (0.02–0.06)
Tobramycin (419)	0.99 (0.97–1.00)	0.96 (0.85–0.99)	0.99 (0.98–1.00)	0.91 (0.80–0.98)	22.26 (5.74–86.28)	0.01 (0.00–0.03)
Trimethoprim-sulfamethoxazole (425)	0.98 (0.96–0.99)	0.95 (0.90–0.98)	0.98 (0.96–0.99)	0.96 (0.91–0.99)	20.49 (9.38–44.73)	0.02 (0.01–0.04)
Presence of urinary catheter
Yes (90)	0.99 (0.98–0.99)	0.97 (0.94–0.99)	0.99 (0.98–1.00)	0.97 (0.94–0.98)	33.10 (16.73–65.51)	0.01 (0.01–0.02)
No (338)	0.99 (0.98–0.99)	0.98 (0.97–0.99)	1.00 (0.99–1.00)	0.94 (0.92–0.96)	51.66 (30.15–88.50)	0.01 (0.01–0.02)
Location of specimen collection
Emergency department (314)	0.99 (0.98–0.99)	0.98 (0.97–0.99)	1.00 (0.99–1.00)	0.94 (0.92–0.96)	51.75 (29.55–90.63)	0.01 (0.01–0.02)
Ward (88)	0.98 (0.97–0.99)	0.97 (0.94–0.99)	0.99 (0.98–1.00)	0.94 (0.90–0.97)	35.18 (15.98–77.44)	0.02 (0.01–0.03)
Critical care unit (26)	0.99 (0.97–1.00)	0.97 (0.92–0.99)	0.98 (0.95–1.00)	0.99 (0.95–0.100)	34.81 (11.41–106.20)	0.01 (0.00–0.04)

^aSN, sensitivity; SP, specificity; PPV, positive predictive value; NPV, negative predictive value; +LR, positive likelihood ratio; −LR, negative likelihood ratio; CI, confidence interval.

^bPositive likelihood ratios could not be calculated in subgroups where there were no urine culture-susceptible and blood culture-nonsusceptible pairs.

^cOrganisms (n) within this category include Citrobacter species (6), Morganella species (2), Providencia stuartii (2), Raoultella planticola (1), and Serratia marcescens (8).

^dDue to changes in culture reporting over the 5-year period, certain antibiotic susceptibility results were unavailable. Most notably, amoxicillin-clavulanate and piperacillin-tazobactam susceptibilities were not reported for 181 and 67 matching pairs, respectively.

Based on 15 consecutive E. coli urine and blood culture isolates from January 2018, the times from specimen receipt to susceptibility results for urine and blood cultures were 46.0 and 61.7 hours, respectively, resulting in a time differential of 15.7 hours (95% CI, 8.7 to 22.7; P < 0.001).

In patients diagnosed with bacteremic UTI with urine and blood cultures yielding the same pathogen, the urine culture susceptibility results accurately predict blood culture susceptibility results. Therefore, urine culture results can be used to choose a targeted therapy when the same organism is isolated in blood cultures and susceptibility results are still pending.

A significant time difference from specimen receipt to susceptibility report was noted for urine and blood cultures, highlighting the potential for earlier targeted antimicrobial therapy. Earlier targeted therapy can potentially reduce broad-spectrum antibiotic exposure and its associated adverse effects and the length of hospitalization in patients who have responded to treatment. A longer time difference may occur in laboratory settings with limited operating hours, lower specimen volumes, and those without access to techniques (such as smudge plate preparation) aimed at shortening the time to susceptibility results from positive blood cultures.

This study was limited to patients with a presumed urinary source of infection. The applicability of these results to patients with nonurinary sources of Gram-negative bacteremia and concomitant, incidental bacteriuria is unknown. Although our results appear robust across clinically relevant subgroups, further research is needed in other jurisdictions to confirm their generalizability. Finally, with new technologies aimed at further shortening the time from blood culture positivity to the antimicrobial susceptibility result, the time difference between urine and blood culture susceptibility results may be reduced in the future, thereby attenuating the time advantage offered by the study findings.

In summary, urine culture susceptibility results accurately predict blood culture results in patients with bacteremic UTI. Therefore, it is reasonable for clinicians to streamline antibiotic therapy based on urine culture susceptibility results when the same pathogen is growing in blood cultures and susceptibility results are pending.