Impact of Antimicrobial Stewardship Interventions on Antimicrobial Utilization in Asymptomatic Bacteriuria

Abstract

Background: Antimicrobial therapy for asymptomatic bacteriuria (ASB) is often unnecessary and is a common reason for inappropriate antimicrobial use in hospitalized patients. Unnecessary ASB treatment leads to collateral damage such as resistance, and Clostridium difficile infections. This study evaluated the impact of interdisciplinary antimicrobial stewardship interventions on antimicrobial utilization in ASB. Methods: This was a quasi-experimental institutional review board (IRB)-approved study evaluating the impact of antimicrobial stewardship on antibiotic utilization for ASB in a pilot medical-surgical unit. The control phase was from August-October 2017 and the postintervention phase was from December-March 2018. In the control phase, electronic medical records of patients with positive urine cultures were retrospectively reviewed. Patients were classified as either having ASB or urinary tract infection (UTI) based on the absence or presence of UTI symptoms documented in the medical record. The intervention phase consisted of educational in-services to providers, nurses, and pharmacists. Clinical pharmacists for the pilot unit utilized an electronic real-time surveillance system to identify patients with positive urine cultures. With nurses’ collaboration, clinical pharmacists classified these patients as either having UTI or ASB. Stewardship interventions were made in real-time to discontinue antibiotics in patients with ASB. Results: There were 65 and 77 patients with bacteriuria in the pre- and postintervention phases. Among these, ASB was present in 29 (45%) and 27 (35%) patients, respectively. After excluding those receiving antibiotics for concurrent nonurinary indications, the combination of education with pharmacist and nursing interventions decreased unnecessary ASB treatment from 18 (62%) to 6 (22%) patients (relative risk: 0.36, 95% confidence interval: 0.16-0.72, P = .003). Conclusion: The findings of this study highlight the importance of interdisciplinary interventions in reducing unnecessary antimicrobial therapy for the treatment of ASB. With increasing antimicrobial resistance, healthcare institutions should evaluate the role of these interdisciplinary interventions to reduce unnecessary treatment for ASB.

Introduction

Antimicrobial resistance is considered as one of the biggest threats to global health by the World Health Organization. ¹ Within the United States, it is estimated annually at least 2 million people acquire a serious bacterial infection that is resistant to antimicrobials. ² Due to antimicrobial resistant infections, approximately 23 000 people die each year. ² Mechanisms must be put into place to help treat and protect patients who acquire these infections, and reduce the rise in antimicrobial resistance. The Choosing Wisely campaign highlights the importance of avoiding overuse of antimicrobial therapy in the hospital setting and encourages a stewardship initiative as a way to reduce the emergence and spread of antimicrobial resistance. ³

One area of concern is the use of antimicrobial therapy in patients with asymptomatic bacteriuria (ASB). Asymptomatic bacteriuria is characterized by the isolation of at least one bacterial species in the urine without signs or symptoms suggestive of a urinary tract infection (UTI). ⁴ Majority of the patients do not need treatment for ASB, exceptions include pregnant patients and patients undergoing urologic procedures. ⁴ Infectious Diseases Society of America (IDSA) guidelines recommend not screening or treating ASB in the following populations due to high prevalence of ASB: women with diabetes (10.8%-16%), elderly persons living in the community (3.6%-19%), elderly persons in a long-term care facility (15%-50%), persons with long-term indwelling catheter (100%). ⁴ Treatment neither decreases the frequency of symptomatic infection nor prevents further episodes of ASB. Despite recommended guidelines, the use of antimicrobial therapy in patients who are not symptomatic is highly prevalent. This leads to increased drug resistance, increased healthcare costs, and collateral damage such as Clostridium difficile infections and drug toxicities.^4,5 In addition, unnecessary use of antimicrobial therapy can be costly with estimates of more than $1.1 billion spent annually according to the Centers for Disease Control and Prevention. ⁶

Effective antimicrobial stewardship among members of the healthcare team can help reduce unnecessary utilization of antimicrobial therapies. The objective of this study was to evaluate a multifaceted interdisciplinary stewardship approach to reduce the number of patients treated for ASB with antimicrobials.

Methods

We conducted an institutional review board (IRB)-approved, quasi-experimental study at a pilot medical-surgical unit in a 340-bed community hospital. The control phase was from August to October 2017 and the postintervention phase was from December 2017 to March 2018. In the control phase, an electronic real-time surveillance system, Vigilanz (Minneapolis, Minnesota), was used to identify patients ≥18 years old who had a positive urine culture with ≥10³ colony-forming unit (CFU) bacteria. Electronic medical records of these patients were retrospectively reviewed by the clinical pharmacist. Patients were classified as either having ASB or UTI based on the absence or presence of UTI symptoms documented in the medical record. Patients were excluded from data collection if they were pregnant or had a recent or pending urologic procedure.

Educational in-services to physicians, nurses, and pharmacists were performed in November 2017. The educational sessions consisted of case studies highlighting the importance of assessing patients for UTI symptoms and were performed by nursing educators and pharmacists. In the postintervention phase, the clinical pharmacist of the pilot unit would identify patients ≥18 years old who have a positive urine culture with ≥10³ CFU bacteria daily using Vigilanz in real-time. Patients were excluded from data collection if they were pregnant or had a recent or pending urologic procedure. Clinical collaboration was established between nurses and clinical pharmacist of the pilot unit when patients from Vigilanz were identified in real-time. Urinary tract infection symptoms were determined based on nursing interview of the patient and documentation in the electronic medical record. Subsequently, the clinical pharmacist classified the patient as either having ASB or UTI. The clinical pharmacist would then make stewardship interventions via verbal communications with the provider to discontinue unnecessary antimicrobial therapy for ASB.

Symptomatic UTI was defined as the presence of UTI symptoms with a positive urine culture of at least 10³ CFU bacteria. Urinary tract infection symptoms included fever, acute hematuria, rigors, flank pain, pelvic discomfort, urgency, frequency, dysuria, suprapubic pain, spasticity, or delirium (meeting 2 of the following criteria: disorganized thinking, acute onset and fluctuating course, altered level of consciousness, and inattention). Patients with bacteriuria 10³ CFU without symptoms were defined to have ASB. Asymptomatic bacteriuria was further classified as treated with antibiotics, not treated or antibiotics discontinued after pharmacist intervention, and on antibiotics for another indication. Treatment of ASB was defined by medical record documentation of a clinician order for a specific antibiotic treatment for bacteriuria. The primary outcome was inappropriate treatment of ASB. Inappropriate treatment was defined as patients who received antibiotics without urinary symptoms. Electronic medical records were reviewed to obtain demographic characteristics, urine culture, urinalysis results, presence of urinary catheters, UTI symptoms, misleading symptoms, and if on antibiotics for another indication.

Statistical Analysis

Data analysis was performed using GraphPad Prism 7 software (GraphPad Software, Inc, La Jolla, California). Categorical data were summarized utilizing frequency counts and percentages and compared using the Fisher exact or chi-square test. For continuous data, means and standard deviations (normally distributed) or medians and interquartile ranges (nonnormally distributed) were calculated. Continuous variables were compared with the Mann-Whitney U test. In all cases, a 2-sided P value of less than .05 was considered statistically significant.

Results

In this quasi-experimental study, we included 65 patients in the control group and 77 patients in the postintervention group with bacteriuria in the final analysis. Two patients in the postintervention group were excluded because these were preoperative urine specimens. Baseline characteristics obtained were similar between the groups except for a significant difference observed with racial distribution and median age which was 71 years in the preintervention group and 80 years in the postintervention group (P = .02). Most of the samples were obtained in the emergency room 52% vs 70% (P = .09) and via clean catches 38% vs 40% (P = .37) in the control vs postintervention group, respectively.

The most commonly isolated pathogens from the urine culture in the control and postintervention groups were Escherichia coli (n = 31 vs 39), Enterococcus spp (n = 10 vs 10), Klebsiella spp (n = 6 vs 11), and Pseudomonas aeruginosa (n = 9 vs 4). Urine cultures with isolate counts of ≥ 10⁶ CFU/mL were present in 76 vs 78 (P = .23) samples in the control group and the postintervention group, respectively. The median urine white blood cell count was 71.5 WBC/HPF (high power field) in the control group and 68 WBC/HPF in the postintervention group (P = .62). Other baseline characteristics are summarized in Table 1.

Table 1.

Baseline Characteristics of Patients in the Control Group and Post Intervention Group.

Variable	Control (n = 65)	Postintervention (n = 77)	P value
Median age, years	71 (60-84)	80 (69-87.5)	.02
Sex, n (%)
Male	20 (30.8)	26 (33.8)	.72
Female	45 (69.2)	51 (66.2)
Race, n (%)
Asian	3 (4.6)	10 (13)	.02
Black	25 (38.5)	14 (18.2)
White	33 (50.8)	50 (64.9)
Other	4 (6.1)	3 (3.9)
Median serum, WBC k/µL	9.7 (7.7-14.9)	10 (6.85-13.8)	.62
Median urine, WBC/HPF	71.5 (9.75-200)	68 (14-200)	.57
Median temperature, °F	98.7 (98.1-99.6)	98.9 (98.3-100.2)	.20
Urine culture location, n (%)
ER	52 (80)	70 (90.9)	.09
Floor	13 (20)	7 (9.1)
Organism count, n
<10 000 CFU/mL	3	7	.46
≥10 000 to < 50 000 CFU/mL	2	3
≥100 000 CFU/mL	76	78
Catheter type, n (%)
None	38 (58.5)	49 (63.6)	.37
Suprapubic	4 (6.2)	2 (2.6)
Transurethral (Foley)	17 (26.2)	17 (22.1)
Intermittent catheterization	4 (6.1)	8 (10.4)
External (condom) catheter	0 (0.0)	1 (1.3)
Nephrostomy	2 (3.0)	0 (0.0)

Note. WBC = white blood cells; HPF = high power field; ER = emergency room; CFU = colony-forming units.

In the control and postintervention groups, ASB was present in 29 (45%) and 27 (35%) of the patients based on predefined criteria. After excluding patients receiving antibiotics for concurrent nonurinary infections and patients who did not receive antibiotics, the effect of education alone decreased the number of patients initiated on treatment for ASB from 18 (62%) to 12 (44%) (relative risk [RR]: 0.72, 95% confidence interval [CI]: 0.42-1.17, P = .28) in the postintervention group. The combined effect of real-time pharmacist and nursing interventions in addition to education as described in the methods, further significantly decreased the treatment of ASB from 18 (62%) to 6 (22%) (RR: 0.36, 95% CI: 0.16-0.72, P = .003). A subgroup analysis of patients with clinically relevant pyuria (WBC ≥ 10 WBC/HPF) was conducted to determine whether pyuria in the presence of bacteriuria was a strong indicator of a UTI diagnosis. There were 47 (72%) patients in the control group and 65 (84%) patients in the postintervention group with pyuria. Of these, ASB was present in 34% vs 32% of the patients in both groups, respectively (P = 1.00). The most frequently documented UTI symptoms in both groups were fever, hematuria, rigors, flank pain, dysuria, and delirium. Of these, there were significantly more febrile patients, 14 (22%) vs 29 (38%) (P = .04), in the postintervention group.

Discussion

Antimicrobial resistance is a global health concern and the unwarranted use of antimicrobial therapy adds to this growing issue. ¹ Antimicrobial stewardship remains one of the most important strategies for reducing the emergence and spread of antimicrobial resistance. A potential antimicrobial stewardship strategy is to reduce unnecessary antimicrobial therapy for ASB. ⁷ Reducing unnecessary treatment for ASB is part of the American Board of Internal Medicine “Choosing Wisely” campaign and has been proposed as an area of focus for hospital antibiotic stewardship programs. ³

In a retrospective observational study published by Grein and colleagues, among 300 patients with bacteriuria, ASB was present in 71%. After excluding those given antibiotics for nonurinary indications, antibiotics were given to 38% (62/164) with ASB. ⁵ Similarly, in a Veterans Affairs hospital, Linares and colleagues observed that 26% (30/117) of patients with ASB received antimicrobials. ⁸ In the preintervention phase of our study, ASB was prevalent in 45% of the study population which is similar to previous studies. After excluding patients who were on antibiotics for another infection and patients that were not treated, 62% of patients with ASB received antimicrobials in the preintervention phase.

Despite IDSA guideline recommendations that pyuria accompanying ASB is not an indication for antimicrobial treatment, ⁴ in our study we observed that abnormal urinalysis findings was associated with ASB treatment. Treatment for bacteriuria should not solely be based on laboratory criteria of abnormal urinalysis findings; it should rather be coupled with patient symptoms. Despite provider education, our rate of patients treated with ASB did not significantly improve. However, with active stewardship interventions, the percentage of patients treated for ASB decreased significantly. These findings are consistent with IDSA guidelines on implementing an antimicrobial stewardship program that recommend prospective audit and feedback to optimize antimicrobial utilization, and to use educational activities to complement other stewardship activities. ⁹

Our study looked at clinical pharmacists and nursing interventions coupled with education to reduce unnecessary antimicrobial therapy for ASB. Several studies have been published evaluating various antimicrobial stewardship strategies to reduce unnecessary antimicrobial therapy for ASB. Linares and colleagues significantly decreased the mean duration of antimicrobial therapy for ASB from 6.3 to 2.2 days (P < .001) in their intervention group. Their intervention consisted of placing an electronic educational memorandum in patients’ charts to remind providers of appropriate management of ASB. ⁸ A multifaceted intervention by Trautner and colleagues resulted in a significant decrease in the overall rate of urine culture ordering (incidence rate ratio [IRR]: 0.57; 95% CI: 0.53-0.61) and overtreatment of ASB (IRR: 0.35; 95% CI: 0.22-0.55) during the intervention period. ⁶ This multifaceted intervention included a kicking catheter-associated UTI diagnostic algorithm, provider education, audit and feedback and surveys. ⁶ Kelley and colleagues conducted educational initiatives provided to physicians and pharmacists to minimize inappropriate treatment of ASB. Post education, unnecessary empirical antibiotics for ASB decreased significantly from 62 to 26% (P < .001). ¹⁰ In our study, although educational intervention alone did not significantly reduce unnecessary empirical antibiotics for ASB (62%-44%, P = .28), real-time clinical pharmacists and nursing interventions significantly reduced unnecessary antibiotic therapy for ASB (62%-22%, P = .003). Despite communication with the provider and lack of urinary symptoms, 6 patients (22%) were continued on antibiotics for ASB.

Our study was strengthened by including a noninterventional control group. In addition, patient characteristics in the control group and the intervention group were well matched except for race and median age. Our study has several limitations. First, the quasi-experimental, nonrandom design could result in selection bias which could confound our findings. However, this was minimized because we included all consecutive patients that met our inclusion criteria in the study time frame. Second, because there was a retrospective chart review of the patients in the control group, this study relied on appropriate documentation in the patient’s electronic medical record to determine patient symptoms.

In conclusion, real-time pharmacist and nursing interventions in combination with education led to a significant decrease in inappropriate antimicrobial therapy for ASB. Although this study was successful, our findings are preliminary and hypothesis generating for future larger prospective studies. Given the high prevalence of treatment among patients with ASB, health care institutions should evaluate the role of interdisciplinary antimicrobial stewardship interventions to reduce unnecessary antimicrobial utilization for ASB.