Bacteria on Urine Microscopy Is Not Associated with Systemic Infection in Patients with Obstructing Urolithiasis

Abstract

Introduction:

Determining whether bacterial presence in urine microscopy represents infection is important as ureteral stent placement is indicated in patients with obstructing urolithiasis and infection. We aim to investigate whether the presence of bacteria on urine microscopy is associated with other markers of infection in patients with obstructing urolithiasis presenting to the emergency room.

Methods:

We performed a cross-sectional study of 199 patients with obstructing urolithiasis and divided patients into two groups according to the presence of bacteria on urine microscopy. The primary outcome was serum white blood cell count and secondary outcomes were objective fever, subjective fever, tachycardia, pyuria, and final urine culture. Univariate and multivariate analysis were used to assess whether the presence of bacteria on microscopy was associated with other markers of infection.

Results:

The study included 72 patients in the bacteriuria group and 127 without bacteriuria. On univariate analysis, the presence of bacteria was not associated with leukocytosis, objective fever, or subjective fever, but it was associated with gender (p < 0.001), pyuria (p < 0.001), positive nitrites (p = 0.001), positive leukocyte esterase (p < 0.001), and squamous epithelial cells (p = 0.002). In a multilinear regression model including the presence of squamous cells, age, and sex, the presence of bacteriuria was not related to serum white blood cell count (coefficient −0.47; 95% confidence interval [CI] −1.1, 0.2; p = 0.17), heart rate (coefficient 0.85; 95% CI −2.5, 4.2; p = 0.62), presence of subjective or objective fever (odds ratio [OR] 1.5; 95% CI 0.8, 3.1; p = 0.18), or the presence of squamous epithelial cells (coefficient −4.4; 95% CI −10, 1.2; p = 0.12). However, the presence of bacteriuria was related to only the degree of pyuria (coefficient 16.4; 95% CI 9.6, 23.3; p < 0.001).

Conclusions:

Bacteria on urine microscopy is not associated with other markers of systemic infection and may largely represent a contaminant. Renal colic may be a risk factor for providing a contaminated urine specimen.

Introduction

Obstructing urolithiasis are a common cause of emergency room (ER) visits with over 600,000 visits to the ER occurring annually in the United States.^1,2 During the evaluation of these patients, it must be decided whether immediate kidney drainage with ureteral stent or nephrostomy tube placement is necessary. Concomitant urinary tract infection is considered an indication for emergent kidney drainage in patients with obstructing urinary calculi, although the presence or absence of a urinary tract infection in these patients is not always clear.^3,4

Urine analysis with microscopy is usually obtained in patients presenting to the ER with obstructing urinary calculi to investigate for infection. Specifically, the presence of leukocyte esterase and/or nitrite are commonly assessed and considered indicators of possible infection when positive. The presence or absence of bacteria is also assessed on urine microscopy, however, the clinical significance of bacteria on urine microscopy in these patients is unclear.^5,6 A recent study of ER patients found that neither bacteria nor pyuria alone on urinalysis are diagnostic of urinary tract infection, especially in asymptomatic patients.⁷ Further understanding of whether or not the presence of bacteria on urinalysis indicates infection in patients with obstructing calculi is important for clinical decision making. To this end, we aim to investigate the association between the presence of bacteria on urine microscopy with other markers of infection in ER patients with obstructing urinary calculi. We hypothesize that the presence of bacteria on urinary microscopy is not associated with other markers of infection in ER patients with obstructing urinary calculi.

Materials and Methods

We conducted a retrospective cross-sectional study of patients presenting to the ER at our institution with obstructing urolithiasis between October 1, 2014 and October 20, 2016. Patients were identified by International Classification of Diseases (ICD)-9 and ICD-10 codes (591, 592.0, 592.1, 788.0, N23, N13.2, N13.3, N20.0, N20.1, Q61.5, R10.9, R32), and their records were reviewed for study eligibility. Inclusion criteria were patients greater than 18 years of age presenting to the ER with obstructing urolithiasis diagnosed on CT with a laboratory work-up completed during evaluation, including complete blood count, urinalysis, and urine culture. Patients were instructed to provide a clean-catch mid-stream urine sample in accordance with emergency department protocol.⁸ Exclusion criteria were patients with urinary drains (i.e., nephrostomy tubes, urinary stents, indwelling urethral catheters, etc.), urinary diversions, ER visits within the preceding 90 days, active pregnancy, and urine contamination as defined as the presence of moderate or many squamous epithelial cells on urine analysis. Patients with none, few, or occasional squamous cells on urinalysis were included. The medical records of patients meeting study inclusion and exclusion criteria were reviewed for the presence of objective fever defined by a temperature >38°C (both at initial presentation and throughout the visit), subjective fever (on review of systems), leukocytosis >10,800 × 10⁶/L, tachycardia >100 bpm, positive urine culture from the initial urine studies defined as greater than 100,000 CFU/mL, length of hospital stay, and intervention. Cultures with greater than or equal to three organisms were considered contaminated and were excluded from the study.

For the analysis, patients were divided into two groups according to the presence of bacteria on urine microscopy: (1) bacteriuria group (few, moderate, or many bacteria on microscopy) and (2) no bacteriuria group (no bacteria on microscopy). Student's t-test was used to compare continuous variables among the groups, and Chi-square or Fisher's Exact tests were used to compare categorical variables among the groups. Multiple regression analysis was used to assess the relationship between bacteria and other markers of infection while controlling for confounding factors, including the presence of squamous epithelial cells, gender, and pyuria. The smallest difference between the normal white blood cell count at our institution (upper limit of 10,800/mm³) and the minimum white blood cell count in patients with leukocytosis from sepsis (12,000/mm³)⁹ was 1200/mm³. We fixed alpha at 0.05 and calculated that we needed 34 patients in each group (bacteriuria vs no bacteriuria) to provide 90% power to detect a difference of 1200/mm³ white blood cells between the groups. Additional patients were included to allow for multivariable analysis.

Results

A total of 261 charts were identified by ICD codes and reviewed. Sixty-two patients did not meet inclusion/exclusion criteria (e.g., active pregnancy, urinary drain, recent admission, absence of urolithiasis on CT scan, and chronic indwelling catheters). The final analysis therefore included 199 patients who met study criteria. Of these patients, 72 (36%) had bacteria on urinalysis (bacteriuria group) and 127 (64%) patients did not (no bacteriuria group). The mean age and body mass index were 49.3 (standard deviation [SD] 15.7) and 29.3 (SD 5.6), respectively, and there were no differences between the two groups (Table 1). More women had bacteriuria compared to men (p < 0.001). Mean stone size measured on CT was 5.9 mm (SD 3.3). Most patients were Caucasian/white (93.5%) with no significant difference between the bacteriuria and no bacteriuria group (p = 0.61). Stone laterality included 82 (41.2%) right-sided calculi, 114 (57.3%) left-sided calculi, and 3 (1.5%) bilateral calculi, and there was no difference between the groups (p = 0.71). There was also no difference in the degree of hydronephrosis (none, mild, moderate, severe) among the two groups (p = 0.24).

Table 1.

Demographics of Study Population

Demographics	No bacteriuria (n = 127)	Yes bacteriuria (n = 72)	Total	p
Age mean (SD), years	49.2 (15.3)	49.4 (16.3)	49.3 (15.7)	0.9
BMI (SD), kg/m²	29.6 (5.8)	28.7 (5.2)	29.3 (5.6)	0.36
Diabetes, n (%)	25 (19.7)	10 (13.9)	35 (17.6)	0.30
Sex, n (%)
Male	81 (63.8)	25 (34.7)	106 (53.3)	<0.001
Female	46 (36.2)	93 (46.7)	93 (46.7)
Mean stone size (SD), mm	5.7 (3.2)	6.2 (3.5)	5.9 (3.3)	0.31

BMI = body mass index; SD = standard deviation.

On univariate analysis, bacteriuria was not found to correlate with subjective fever (p = 0.49), objective fever (p = 0.29), heart rate (p = 0.65), positive final urine culture (p = 0.1), or leukocytosis (p = 1.0) (Table 2). The two groups showed statistically significant differences in pyuria on microscopy (bacteriuria vs no bacteriuria; p < 0.001), positive nitrites (bacteriuria vs no bacteriuria; p = 0.001), positive leukocyte esterase (bacteriuria vs no bacteriuria; p < 0.001), and squamous epithelial cells (bacteriuria vs no bacteriuria; p = 0.01). There was no difference in management according to group; 62 (47.7%) patients from the no bacteriuria group and 29 (40.3%) patients from the bacteriuria group were discharged on medical expulsive therapy (MET) (p = 0.21) (Table 2). Of the 91 total patients discharged from the emergency department on MET, 5 patients (5.4%) returned to the emergency department within 30 days, 2 (3.2%) from the no bacteriuria group, and 3 (10.3%) from the bacteriuria group (Table 2). The degree of bacteriuria (none, few, moderate, many) was correlated with the degrees of squamous epithelial cells (none, few, occasional) (coefficient 0.16; p = 0.02), leukocyte esterase (none, trace, small, moderate, large) (coefficient 0.42; p < 0.001), and pyuria on microscopy (coefficient 0.29; p < 0.001). However, the degree of bacteriuria was not correlated with length of stay (0.0; p = 0.96), tachycardia (0.22; p = 22), leukocytosis (−0.11; p = 0.14), or size of stone (0.84; p = 0.25).

Table 2.

Selected Clinical Signs and Urinalysis Data Comparisons Between Groups

Parameters	No bacteriuria	Yes bacteriuria	Total	p
Serum WBC mean (SD)	10.6 (3.2)	10.2 (3.5)	10.4 (3.3)	0.39
WBC >10,800, n (%)^a	56 (43.1)	31 (43.1)	87 (43.1)	1.0
Heart rate mean (SD)	86.4 (17.0)	88.2 (16.0)	87.1 (16.6)	0.48
Tachycardia >100 bpm, n (%)	22 (16.9)	14 (19.4)	36 (17.8)	0.65
Squamous epithelial cells present, n (%)	63 (48.5)	48 (66.7)	11 (55.0)	0.01
Nitrite positive, n (%)	5 (3.9)	13 (18.1)	18 (9.0)	<0.001
Leukocyte esterase present, n (%)	41 (31.5)	45 (62.5)	86 (42.6)	<0.001
Urine RBC mean (SD)	50.7 (68.8)	64.9 (74.1)	55.9 (65.8)	0.18
Urine WBC mean (SD)	7.8 (19.4)	32.6 (53.2)	16.8 (37.4)	<0.001
Objective fever >38°C, n (%)	1 (0.8)	2 (2.7)	3 (1.5)	0.29
Subjective fever, n (%)	6 (4.6)	5 (7.0)	11 (5.5)	0.49
Fever during hospital course >38°C, n (%)	2 (1.5)	3 (4.2)	5 (2.5)	0.35
Any fever, n (%)	6 (4.6)	6 (8.3)	12 (5.9)	0.28
Hospital length days, n (SD)	0.56 (0.62)	0.85 (1.4)	0.6 (1.0)	0.46
Positive final urine culture, n (%)	13 (10.0)	13 (18.1)	26 (12.9)	0.10
Discharged on MET, n (%)	62 (47.7)	29 (40.3)	91 (45.1)	0.21
Returned to ED after MET trial, n (%)	2 (3.2)	3 (10.3)	5 (0.05)	0.16

10,800 is considered the upper limit of normal at our institution.

ED = emergency department; MET = medical expulsive therapy; RBC = red blood cell; WBC = white blood cell.

In multilinear regression models including the presence of squamous cells, age, and sex as fixed covariates, the presence of bacteriuria was not related to serum white blood cell count (coefficient −0.47; 95% confidence interval [CI] −1.1, 0.2; p = 0.17), heart rate (coefficient 0.85; 95% CI −2.5, 4.2; p = 0.62), presence of subjective or objective fever (odds ratio [OR] 1.5; 95% CI 0.8, 3.1; p = 0.18), or the presence of squamous epithelial cells (OR −4.4; 95% CI −10, 1.2; p = 0.12). However, the presence of bacteriuria was related to the degree of pyuria (coefficient 16.4; 95% CI 9.6, 23.3; p < 0.001).

Discussion

In this retrospective cross-sectional study of patients presenting to the ER with obstructing urolithiasis, the presence of bacteriuria on urine microscopy was not associated with signs of systemic infection. However, the presence of bacteria was associated with pyuria, even after accounting for confounding factors such as specimen contamination (degree of squamous epithelial cells).

While the presence of bacteriuria in patients with obstructing urolithiasis may represent isolated lower urinary tract infection, more likely, bacteria on urine microscopy may be an additional marker of specimen contamination. Mohr and colleagues has shown that squamous cells on urinalysis predict a poor predictive value of the additional findings on urinalysis, and therefore samples with high squamous cell content were excluded.¹⁰ However, we did not find a greater proportion of positive urine cultures in patients with bacteria on microscopy. On univariate analysis, bacteriuria was more common in women who had nitrite, squamous epithelial cells, and pyuria. Therefore, this would also suggest that bacteriuria is a marker of contamination rather than infection.

In this study, gender was associated with a statistically significant difference between the two groups, with women representing more than half of the bacteriuria group, and men representing more than half of the no bacteriuria group. From previous studies it has shown that contamination rates in women can be as high as 30%, and it is well known that women have a higher urine specimen contamination rate when compared to men.¹¹ Of note, urine contamination in the ER is common, and methods to decrease contamination by patient instruction and visual aid have not proven to be effective.^12,13 Additionally, patients with renal colic who may be in severe pain may have a difficult time providing a clean urine sample.

The strength of this study is the exclusion of patients with moderate and many squamous epithelial cells on urinalysis, as these patients may have a high rate of urine specimen contamination.¹⁰ By excluding those patients at high risk of a contaminated sample, this study sought to assess a patient population that had already selected against a known marker of contamination. Taking into account this exclusion criterion, the results of this study appear to indicate that bacteria on urinalysis, with or without squamous epithelial cells, could independently be a marker of a contaminated sample rather than infection. Additionally, it should be considered that bacteria on urinalysis may be a marker of urinary microbiome. Although it is currently suggested that the microbiome of other body systems, such as the gastrointestinal tract, is critically involved in health maintenance and development of disease, research into the microbiome of the urinary tract is currently underway.^14,15 Future testing may evolve to look at the urinary microbiota of patients presenting to the emergency department with obstructing stones, but it is unclear at this time how the urinary microbiome relates to symptomatology and how the analysis of the microbiome may or may not be used as a marker of infection.

Although this study suggests the isolated presence of bacteriuria does not appear to be a marker of infection, there are other signs to suggest infection in patients with obstructing stones, such as serum white blood cell count and clinical appearance. Clinicians should be reminded to rely on the overall clinical picture when assessing a patient with obstructing urolithiasis. This is especially important when making the decision regarding whether MET is appropriate for a patient. However, the presence or absence of bacteriuria may be considered an important data point for clinical decision making, and thus it is important to decipher whether bacteriuria is a marker of infection.

Of the five patients discharged on MET that returned to the emergency department within 30 days, two patients were from the no bacteriuria group, and three were from the bacteriuria group. The three returning patients from the bacteriuria group presented with symptoms such as lethargy and hypotension, potentially concerning for severe complications such as sepsis. However, these patients had significant comorbidities (one patient had stage IV pancreatic cancer, one patient had severe dementia, and one patient had recurrent multidrug-resistant urinary tract infections). While these findings may suggest that bacteriuria incurs a higher risk of serious complications, it is difficult to make this distinction based on the small number of patients that returned to the emergency department with concomitant comorbidities. Further studies specifically investigating the clinical course following discharge may help elucidate any differences between the two groups presented in this study. Overall, only a small proportion of total patients discharged on MET returned to the hospital with continued complaints.

This study is limited in that patients were not catheterized to minimize possible contamination of urine samples. However, as most patients who present to the ER for obstructing urolithiasis are not catheterized to obtain a urine sample, this study is generalizable to the majority of patients. Definitions for systemic infection may also be controversial as the recent publication by Sepsis-3 describes screening for sepsis should be inclusive of respiratory rate, altered mental status, and hypotension with systolic blood pressure <100 mm Hg.¹⁶

Future research investigating whether patients who are discharged with positive bacteriuria return to the ER with related urinary complaints can be investigated. Additionally, a retrospective and prospective study that looks at patients with bacteriuria as the sole abnormality on urinalysis would help to properly assess the clinical significance of this one component of urinalysis.

Conclusion

While urinalysis is commonly obtained in patients with renal colic presenting to the emergency department, the clinical significance of bacteriuria is unclear. We found that bacteria on urine microscopy was not associated with markers of systemic infection and may largely represent a contaminant. It is possible that renal colic may be a risk factor for providing a contaminated urine specimen.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Graham

, Luber

, and Wolfson

. Urolithiasis in the emergency department. Emerg Med Clin North Am, 2011; 29:519–538.

Fwu

C-W

, Eggers

, Kimmel

, et al. Emergency department visits, use of imaging, and drugs for urolithiasis have increased in the United States. Kidney Int, 2013; 83:479–486.

Geerlings

. Clinical presentations and epidemiology of urinary tract infections. Microbiol Spectr, 2016; 4:1–11.

Pearle

et al. Medical management of kidney stones: AUA guideline. J Urol, 2014; 192:316–324.

Lohr

, Portilla

, Geuder

, et al. Making a presumptive diagnosis of urinary tract infection by using a urinalysis performed in an on-site laboratory. J Pediatr, 1993; 122:22–25.

Van Nostrand

, Junkins

, Bartholdi

. Poor predictive ability of urinalysis and microscopic examination to detect urinary tract infection. Am J Clin Pathol, 2000; 113:709–713.

Schulz

, Hoffman

, Pothof

, et al. Top ten myths regarding the diagnosis and treatment of urinary tract infections. J Emerg Med, 2016; 51:1–6.

Maher

, Brown

AEC

, Gatewood

. The effect of written posted instructions on collection of clean-catch urine specimens in the emergency department. J Emerg Med, 2016; 52:639–644.

Bone

, Balk

, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest, 1992; 101:1644.

10.

Mohr

, Harland

, Crabb

, et al. Urinary squamous epithelial cells do not accurately predict urine culture contamination, but may predict urinalysis performance in predicting bacteriuria. Acad Emerg Med, 2016; 23:323–330.

11.

Frazee

, et al. Abnormal urinalysis results are common, regardless of specimen collection technique, in women without urinary tract infections. J Emerg Med, 2015; 48:706–711.

12.

Cai

, Koves

, Johansen

TEB

. Asymptomatic bacteriuria, to screen or not to screen-and when to treat?. Curr Opin Urol, 2017; 27:107–111.

13.

Singer

, Deutschman

, Seymour

, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA, 2016; 315:801–810.

14.

Whiteside

, Razvi

, et al. The microbiome of the urinary tract—A role beyond infection. Nat Rev Urol, 2015; 12.2:81–90.

15.

Kartsens

, Asquith

, Davin

, et al. Does the urinary microbiome play a role in urgency urinary incontinence and its severity?. Front Cell Infect Microbiol, 2016; 6:78.

16.

Assimos

, Krambeck

, Miller

, et al. Surgical management of stones: American Urological Association (AUA) Endourological Society Guideline. Journal of Urology, 2016; 196:1153–1160.