Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Diagnosis of bacteriuria in men: specimen collection and culture interpretation.

Author(s): Lipsky BA, Ireton RC, Fihn SD, Hackett R, Berger RE

Publication: J Infect Dis, 1987, Vol. 155, Page 847-54

PubMed ID: 3559288 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study investigated if clean-catch midstream urine more accurately reflects bladder bacterial loads found in aspiration or catheter specimens than uncleansed first void urine and investigated the effects of antibiotic use and the presence of symptoms on bacteriuria. Urine was collected from 66 patients (76 collection sets) who could void at will and were undergoing a urologic procedure or had acute dysuria, irritative genitourinary symptoms, bacteriuria. Patients were excluded if they had a bladder neoplasm, coagulation disorder, were morbidly obese, had prior lower abdominal surgery, or radiation therapy, but the study included 10 patients who were on antibiotics. Ten  mL urine specimens were collected first by suprapubic aspiration and then by voiding (no cleaning). Following the collection of 10 mL in the first void, the urethral meatus was cleaned with povidone-iodine and the next approximately 100 mL was collected. Subsequently, a final 10 mL collection was made by catheterization. Specimens were delivered to the laboratory, cultures were inoculated within 30 min. Cultures were examined after 18-24 h and contamination was defined as <10² cfu/mL in the bladder specimen and three or more organisms present in the other specimens.

   Summary of Findings:

   Thirty-six of the seventy-six specimen sets grew a single predominant organism in the bladder specimen.  When bacteria was present in the bladder specimen, it was also found at greater than 10³ cfu/ml in the clean-catch specimens in 34 of 35 cases (97.1%) and in the uncleansed first void in 33 of 34 cases (94.1%) with the discrepant case displaying low levels of growth in the bladder specimen and no growth in the clean-catch specimen.  When the bladder specimen was found to be sterile (<10² cfu/ml), all 40 matched clean-catch specimens were also sterile but only 30 of 38 matched uncleansed first void specimens were sterile with five displaying contamination (3 or more organisms found) and three with growth of a single organism. Overall, the agreement between bladder and clean-catch specimens had a weighted κ=0.924 (P<0.0001) and the agreement between bladder and uncleansed first void specimens had a weighted κ=0.906 (P<0.0001). When growth in the voided specimen was defined as >10³cfu/mL, there was not a significant difference for bacteriuria between the clean-catch and uncleansed first void urine in sensitivity (97% for both), specificity (100% versus 91-92%), positive predictive value (100% versus 80-92%), and or negative predictive values (8% versus 97%). Of the 62 cases where both an aspiration and catheter specimen were obtained, 56 had identical results with three cases differing in colony count by 1 log, one case differing in colony count by 2 log, and two cases having sterile aspiration specimens and 10² cfu/mL growth of a single organism in the catheter specimen.

   Among the specimens collected from the 10 patients on antibiotics, five contained bacteria in the bladder specimen and the clean-catch midstream urine and five displayed no growth in either the bladder or clean-catch specimen. Of the seven subjects who provided specimens more than once, five had the same organism at the second sampling time, one had mixed growth at the second sampling, and the final subject showed no growth at the subsequent sampling (after antibiotic use).  Finally, pyuria and bacteriuira were more common in patients with irritative symptoms than those without (37% versus 30.6%, P<0.01).

   Biospecimens

   • Bodily Fluid - Urine

   Preservative Types

   • None (Fresh)

   Diagnoses:

   • Other diagnoses
   • Not specified

   Platform:

   Analyte	Technology Platform
   Cell count/volume	Light microscopy
   Cell count/volume	Microbiological assay

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Preaquisition	Other drugs	Antibiotics No antibiotics
   Biospecimen Acquisition	Method of fluid acquisition	Different urine collection procedures compared Needle Catheterized urine Voided urine (spot collection) Collection site preparation methods compared
   Preaquisition	Diagnosis/ patient condition	Acute dysuria Irritative genitourinary symptoms Bacteriuria Undergoing a urologic procedure

   View more

2. Study Purpose

   This study compared leukocyte counts in clean-catch midstream, uncleansed first void, and bladder urine and investigated the effects of bacteriuria. Urine was collected from 66 patients (76 collection sets) who could void at will and had acute dysuria, irritative genitourinary symptoms, bacteriuria, or were undergoing a urologic procedure. Patients were excluded if they had a bladder neoplasm, coagulation disorder, were morbidly obese, had prior lower abdominal surgery, or radiation therapy, but the study included 10 patients who were on antibiotics. Ten mL urine specimens were collected first by suprapubic aspiration and then by voiding (no cleaning). Following the collection of 10 mL in the first void the urethral meatus was cleaned and with povidone-iodine and the next approximately 100 mL was collected. Subsequently, a final 10 mL collection was made by catheterization. Specimens were delivered to the laboratory and cultures were inoculated within 30 min. Cultures were examined after 18-24 h. Leukocyte counts were determined microscopically. 

   Summary of Findings:

   Average leukocyte counts were highly variable but were at least 10-fold higher in culture positive specimens.  The leukocyte count was approximately 4-fold higher in the uncleaned first void specimen than the clean-catch specimen but only 30% higher in the clean-catch specimen than the bladder specimen. The presence of more than 10 leukocytes/mL in clean-catch urine specimens had a positive predictive value of 77% and a negative predictive value of 69% for bacteriuria. 

   Biospecimens

   • Bodily Fluid - Urine

   Preservative Types

   • None (Fresh)

   Diagnoses:

   • Other diagnoses
   • Not specified

   Platform:

   Analyte	Technology Platform
   Cell count/volume	Microbiological assay
   Cell count/volume	Light microscopy

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Preaquisition	Diagnosis/ patient condition	Bacteriuria No bacterial growth
   Biospecimen Acquisition	Method of fluid acquisition	Catheterized urine Different urine collection procedures compared Needle Voided urine (spot collection) Collection site preparation methods compared

   View more