The definition of a positive urine culture has been the subject of controversy for decades. Using a cutoff of 100,000 colony-forming units per milliliter (CFU/mL) to define urinary tract infection (UTI) in adult patients was based largely on a small case-control study reported by Kass1 in 1956 in the which compared urine culture results from women with clinically diagnosed pyelonephritis and asymptomatic controls; most women with pyelonephritis had colony counts greater than 100,000 CFU/mL and most asymptomatic women had colony counts less than 10,000 CFU/mL.
Almost 30 years later, in a cross-sectional study of young children who underwent bladder catheterization to rule out UTI, Hoberman and colleagues compared the characteristics of children who had growth from 10,000 to 49,000 CFU/mL and 50,000 to 99,000 CFU/mL.2 Among the 35 samples with growth between 10,000 and 99,000 CFU/mL, mixed growth and/or Gram-positive cocci were observed more frequently among children with colony counts of 10,000 to 49,000 CFU/mL compared to children with colony counts of 50,000 to 99,000 CFU/mL.
Since then, the limit of 50,000 has been the accepted cutoff for interpreting culture results from specimens collected by catheterization in children younger than 2 years of age.3 However, because a standard of culture-independent reference in neither of these 2 studies, they can only be considered as clarification4 and cannot provide more than approximations of a limit that could be useful in clinical practice.
In fact, there have been reports of problems with the currently accepted pediatric limit of 50,000 CFU/mL. A notable example comes from a study by Swerkersson et al 5 in which a considerable proportion of young children with radiologically confirmed pyelonephritis had colony counts below the currently accepted limit of 50,000 CFU/mL.
To investigate the trade-offs between sensitivity and specificity at various cut-off points, a cross-sectional study is needed in which both a urine culture (the index test) and a culture-independent reference standard are performed on unselected samples from subjects in whom it is clinically sensible to suspect a UTI.
Recent advances in 16S sequencing, which uses the exact sequence of the highly conserved 16S ribosomal RNA (rRNA) gene to identify bacteria present in samples, now provides us with a sensitive and relatively unbiased reference standard for the identification of organisms in the urine.
In a previous study, they found high concordance between conventional urine culture 16S rRNA gene amplicon sequencing (hereinafter referred to as 16S sequencing) in a cohort of young children (not overlapping with the current cohort) who are being evaluated for UTI. .6
In this cohort of young, febrile children undergoing bladder catheterization to rule out a urinary tract infection, using 16S sequencing as the reference standard, the authors calculated the accuracy of conventional culture at different cut-off points to identify the one that provides the optimal balance between sensitivity and specificity.
| Methods |
Between June 2019 and May 2020, they enrolled consecutive children who presented to the emergency department at the Children’s Hospital of Pittsburgh who had urine remaining after completing all clinical tests. The study was approved by the University of Pittsburgh Institutional Review Board.
Children were included if they were between 1 month and 2 years and 11 months of age, had fever (documented temperature ≥38°C in the emergency department or by parent report) within 24 hours of presentation, and had a blood test taken. urine sample via catheter to rule out a UTI.
They excluded newborns because this study is part of a larger study examining biomarkers of pyelonephritis. The diagnosis of pyelonephritis is achieved by performing a renal scan, which is difficult to perform in neonates.
Children were excluded if they had received systemic antibiotics or corticosteroids within 3 days before enrollment, had other concurrent systemic bacterial infections, were immunocompromised, had a neurogenic bladder, or had significant genitourinary anomalies (e.g., spina bifida, dysplastic kidneys). , grade IV or IV vesicoureteral reflux).
Conventional urine culture was performed in the hospital clinical microbiological laboratory using standard microbiological methods and the number of colony forming units (CFU)/mL was reported as <10,000, 10,000 to 49,000, 50,000 to 99,000, and ≥ 100,000.
> Processing of urine samples for 16S sequencing
An aliquot of residual urine was used for 16S sequencing. The aliquot generally occurred within 1 hour of collection; however, if delays were anticipated, samples were kept refrigerated.
The aliquot was frozen in a cryovial at −80°C without preservatives. Before shipping, they added 70 µLs of urine conditioning buffer (Zymo, D3061-1-8) per 1 mL of frozen urine sample. Samples were shipped overnight in a cold package to Pangea Laboratory, Tustin, CA, USA, for 16S sequencing analysis using the PrecisionBIOME NGS Microbial Test.
DNA was extracted from the urinary specimen using the ZymoBIOMICS DNA Miniprep Kit according to the manufacturer’s instructions (Zymo Research Corporation, Irvine, CA). Extracted DNA was prepared for microbiome analysis following the PrecisionBIOME workflow, which included library preparation using the Quick-16S NGS Prep Kit (regions V1-3, Zymo Research Corporation, Irvine, CA), sequencing of barcoded amplicons with the MiSeq sequencing platform (Illumina, San Diego, CA) and bioinformatics analysis using the PrecisionBIOME bioinformatics pipette capable of producing species-level resolution of bacterial and fungal sequences (data on fungi present and cytokines are will report separately).
Negative controls (transport medium alone and unused swabs) were included. To control for contamination, they also included cells and mock DNA communities as positive controls. Possible sequencing errors and chimeric sequences were removed with the DADA2 pipette.
> 16S sequencing data processing
They used Uclust to perform taxonomic classifications using a custom database from PrecisionBIOME. They calculated phylotypes as percentage proportions based on the total number of sequences in each sample.
The species-level resolution of this sequencing approach was previously confirmed by shotgun sequencing.7 We excluded samples with <1000 sequences per sample.
> Target condition being diagnosed
In a previous study carried out between 2011 and 2017,6 the authors found a high concordance rate between conventional urine cultures and 16S sequencing in the identification of bacteria in urine. This established, the authors felt it was now appropriate in the context of this exploratory study to substitute 16S sequencing in place of urine culture in the diagnosis of UTI (the target condition).
In clinical practice, in addition to the presence of bacteriuria in the culture, elevation of inflammatory markers is required to diagnose UTI.3
Consequently, to diagnose UTI in the context of this proof-of-concept study, in addition to requiring that ≥80% of the sequences belonged to a single taxon (i.e., relative abundance of any taxon ≥80%), they also require the elevation of urinary markers of inflammation.
In constructing their reference standard, the authors chose a cutoff of 80% (for the primary analysis) because they felt that, as physicians, they would be forced to diagnose a UTI in a febrile child with urinary markers of inflammation whose urine was at this level. of bacteriuria.
The rest of the children were classified as “no UTI.” As a sensitivity analysis, we also examined results using 50% and 90% relative abundance limits.
> Urinary markers of inflammation
Each child underwent a microscopic urine analysis in which white blood cells (WBCs) were observed (per cubic millimeter or per high-power field) and a dipstick test in which the leukocyte esterase test was reported. semiquantitative form (none, stroke, 1+, 2+, 3+).
Because the sensitivity of both WBC and leukocyte esterase tests is generally low,8 they also measured, using methods used in previous studies,9 neutrophil gelatinase-associated lipocalin (NGAL) in an aliquot of residual urine.
NGAL is an inflammatory marker released by neutrophils or intercalated cells in the kidney in response to UTI.9,10 Although the use of urinary NGAL for the diagnosis of UTI is a relatively recent development, given the strong evidence supporting its use for diagnose UTI,10 the authors decided, a priori, to categorize a child as having evidence of inflammation if any of the following were present: ≥10 WBC/mm3, ≥5 WBC per high power field (Hpf), ≥trace leukocyte esterase or NGAL level greater than 39.9 ng/mL.10 They performed a sensitivity analysis examining the results if they had not considered NGAL as a marker of inflammation.
> Index test in evaluation
The authors defined conventional culture results as positive if the urine culture showed growth of at least 1 organism with a count of at least 10,000 CFU/mL and if at least 1 of the urinary markers of inflammation was elevated. Other cut-off points evaluated were 50,000 CFU/mL and 100,000 CFU/mL.
> Statistical analysis
For the primary analysis, they calculated the sensitivity and specificity of urine culture for detecting UTI (using a 90% relative abundance cutoff in a child with urinary markers of inflammation), along with corresponding 95% Wald confidence intervals. They then repeated the analyzes using different definitions of UTI (i.e., using 50% and 80% relative abundance limits instead of 90%). They summarized and analyzed the data using SAS version 9.4 (SAS Institute Inc).
| Results |
In total, 341 children were included in the study. The majority of children enrolled were female (74%), white (67%), and the majority (64%) had a documented temperature of 39°C or higher. The mean age of the children at diagnosis was 12.5 months and the mean temperature at presentation was 39.3°C. In children with urinary markers of inflammation but without at least 1 organism with a culture count of 10,000 CFU/mL or more, the median relative abundance of the predominant identifiable organisms was 15% (interquartile range: 7%–32 %).
Using a relative abundance cutoff of 80%, 46 children in this sample had a UTI. Of these, 41 (89%) had E. coli as the predominant pathogen. When a cutoff of ≥10,000 is used to define a positive urine culture, among 46 children with UTI, 45 were correctly identified by conventional urine culture (sensitivity 98%, confidence interval [CI]: 93% to 100%).
The missing child was a 4-month-old infant with dipstick leukocyte esterase 3+, a WBC count of 73 per Hpf, an elevated NGAL of 319, in whom >99% sequences were identified by species-assigned 16S sequencing. of Klebsiella . On culture, this child had <10,000 CFU/ml of gram-negative organisms and <10,000 CFU of gram-positive cocci; no further organisms were identified due to low colony counts. Of the 295 children without UTI, 291 were correctly identified as such by conventional culture (99% specificity, CI: 97% to 100%).
Using a cutoff of ≥50,000 CFU/ml decreased urine culture sensitivity to 80% (95% CI: 68%–93%). Changing the cutoff to 50,000 had a negligible effect on specificity (i.e., specificity remained at 99%, CI: 98%-100%). The 8 extra children with UTIs who would have been missed if a limit of 50,000 had been used; all children with missed UTIs had organisms currently considered uropathogenic, and, by definition, all were symptomatic and had elevated levels of urinary markers of inflammation.
Using a cutoff of 100,000 CFU/mL would have reduced sensitivity to 70% (95% CI: 55%-84%). Changes in the definition of the reference standard had little effect on precision estimates.
Sensitivity and specificity estimates would have been similar if they did not include NGAL as a marker of inflammation.
| Discussion |
Previous studies have attempted to understand urine culture results by comparing colony counts under extreme conditions (i.e., pyelonephritis versus asymptomatic) 1 or by comparing characteristics of subjects at certain high and low limits of colony counts.2
Such study designs, although necessary during the early exploratory phases of diagnostic testing research, cannot provide a true assessment of a test’s accuracy in clinical practice, nor can they be used to identify cut-points that optimize sensitivity and the specificity. 4
Here, using 16S sequencing as the gold standard method for UTI diagnosis, the authors were able to evaluate the diagnostic accuracy of urine culture at various colony count limits.
They found that, in cases of febrile children younger than 3 years undergoing bladder catheterization to rule out UTI, using a colony count of 10,000 CFU/ml would have resulted in fewer cases of missed UTI than using a colony count of 50,000 CFU. /mL (misses 20% of cases) or 100,000 CFU/ml (misses 30% of cases).
According to the study design, all children with missed UTI had elevated urinary markers of inflammation and were febrile. Consequently, the data suggest that a cutoff of 10,000 CFU/mL best differentiates young children with and without a true UTI.
There are many reasons why 16S sequencing, especially when combined with urinary markers of inflammation, is a suitable reference standard for the diagnosis of UTI.
First, unlike urine culture, which is optimized to detect E. coli , 11 16S sequencing analysis provides a largely unbiased assessment of the bacteria present in urine. In fact, a growing body of evidence12,13 suggests that the results of 16S sequencing analysis align well with the results of expanded quantitative urine culture, a more sensitive variation of urine culture that uses additional culture media, larger volumes of urine to inoculate plates culture, longer incubation times, and a variety of atmospheric conditions.14,15
However, although the data on the higher sensitivity of 16S sequencing compared to urine culture seems indisputable, its higher sensitivity could in theory come at the cost of lower specificity, especially due to the required amplification step. Thus, one could imagine a scenario in which many clinically non-relevant organisms could be identified by 16S sequencing, leading to a large number of inappropriate UTI diagnoses.
In this study, however, the use of 16S sequencing (along with inflammatory markers) revealed only 1 child in whom UTI was missed. This suggests that 16S sequencing analysis (at a relative abundance threshold of 80%) combined with inflammatory markers of inflammation, was a suitable reference standard to use in this study.
When they began this study, the authors were concerned about using a relative measure of abundance as a standard reference. However, the results (in this study and the previous one) show that, for UTI diagnosis, the relative abundance of 16S sequencing and absolute culture results were strikingly similar.
The authors hypothesized that for a pathogen such as E. coli to account for 80% or more of all sample reads in 16S sequencing, its gene copy numbers would have to exceed the combined counts16 of all other bacteria normally present. as part of the urobiome by four. Thus, the relative abundance threshold of 80% will only be reached when the absolute abundance of a uropathogen is quite high.
The results support the continued use of urine culture (at least a limit of 10,000 CFU/mL) in the study population. However, due to continued technological advances in the accuracy, speed, and cost of 16S sequencing, it may soon be available in clinical centers. If this occurs, identification of organisms could be achieved in a matter of hours instead of days. Therefore, it is important that comparison studies be conducted to establish the pros and cons of using this technology in the clinical setting.
Several limitations of this study need to be considered. They used a definition of UTI that required the presence of symptoms, elevated urinary markers of inflammation, and a high relative abundance of organisms in 16S sequencing analysis. Using a less strict definition (for example, not requiring urinary markers of inflammation) could have led to a higher rate of undetected UTIs, but they would have been less certain that all of these UTIs represented bona fide UTIs.
The choice of 80% for the relative abundance threshold required to define UTI, although based on data from a previous study by the authors, was, however, to some extent, arbitrary. They recognize that less abundant organisms may also be capable of causing important diseases.
In short, as demonstrated by the sensitivity analysis, the cut-off point chosen for the primary analysis had little influence on the conclusions. This is probably because, in most cases, the samples were dominated by sequences from a single known uropathogen or had very low abundance of a variety of organisms that, in previous studies, had been detected in urine samples of asymptomatic people.14
The findings only apply directly to children who undergo bladder catheterization to rule out UTI; A higher limit for urine culture may be necessary if there is more contamination, for example related to the collection method. The authors did not use preservatives before freezing the samples; However, the samples were never left at room temperature, and the results do not suggest bias due to the approach they used.
The clinical laboratory did not report accurate colony counts for conventional urine cultures; This could have been useful to locate the threshold count less than 50,000 CFU/mL that best optimized the sensitivity and specificity of the urine culture. The majority of UTIs in this sample were caused by E. coli .
Larger studies are needed to further examine the accuracy of colony count thresholds for less common uropathogens.
Finally, 16S sequencing, although less biased than urine culture, may be subject to certain biases, especially when the biomass in the samples is relatively low17; however, these biases are unlikely to change the observed pattern from one in which many organisms are present at relatively low abundances to that in which the sample is dominated by 1 organism.
Strengths of the study included the consecutive enrollment of symptomatic children with suspected UTI, the use of a reference standard based on previous data, the performance of the index test and the reference standard in all included children, the use of clinical definitions a priori, the use of positive and negative controls during extraction and sequencing, and the use of a validated sequencing approach.
Inevitably, lowering the threshold for UTI diagnosis from 50,000 CFU/mL to 10,000 CFU/mL will increase the number of children falsely labeled as having UTI. However, this is appropriate because (1) the number of children falsely labeled as having a UTI will be less than the number of children with missed UTIs that will be discovered, and (2) the negative consequences of missing a febrile UTI generally outweigh those of prescribe an additional course of antimicrobial agents.
In conclusion, using 16S sequencing as the reference standard, they found preliminary empirical evidence supporting the use of a cutoff of 10,000 CFU/mL for the diagnosis of UTI in young children with fever.
