Purpose: To study clinicians' approach to distinguishing urinary tract infection (UTI) with sepsis from inconsequential bacteriuria with fever from other sources in the intensive care units (ICUs).
Materials and Methods: The microbiology results (November 1, 2004-March 31, 2005) were retrospectively screened. All adult ICU patients with positive urine culture (≥105 colony-forming unit per milliliter) were identified, and their medical records were reviewed. The following information was recorded: demographics, comorbidity, vital signs, urinary catheter placement, and antibiotic treatment. The incidence of diarrhea was estimated based on the number of stool samples submitted for culture and Clostridium difficile tests.
Results: We encountered 90 evaluable cases. Their age was 62.9 ± 17.6 years; 80 (89%) had indwelling catheters, 66 (73.3%) had leukocytosis (>113 white blood cell counts per microliter), 42 (46.7%) were febrile (≥38.3°C) or septic, and 5 (5.6%) had urinary symptoms. Other possible causes for fever/sepsis were present in 28 (70.0%) febrile/septic patients. Clinicians opted to initiate antibiotics in 43 (91.5%) of 47 patients with fever/sepsis or urinary symptoms (27 of 30 with other causes, 11 of 12 patients without other causes, and 5 of 5 with urinary symptoms without fever) and 25 (58.1%) of 43 patients without symptoms or fever/sepsis. The majority (86.0%) of asymptomatic patients had indwelling catheters. Antibiotic treatment was associated with higher incidence of diarrhea (relative risk, 2.8; 95% confidence interval, 1.03-7.74; P = 0.04).
Conclusions: Clinicians often treat UTI in the ICU in the absence of symptoms and in the presence of infections in other sites. This approach is inappropriate in asymptomatic patients and questionable in patients with other conditions. Urinary tract infection treatment guidelines for ICU patients is urgently needed.
Department of Medicine, St John Hospital and Medical Center, Detroit, MI.
No external financial support was received.
None of the authors have any conflict of interest for disclosure.
This study was approved by our institutional review board and the Human Investigation Committee; patient consent was not required because of the study's observational nature.
Address correspondence and reprint requests to Riad Khatib, MD, Medical Education, St John Hospital and Medical Center, 19251 Mack Ave, Suite 340, Grosse Pointe Woods, MI 48236. E-mail: firstname.lastname@example.org.
Patients in the intensive care units (ICUs) often have fever because of infectious and noninfectious causes.1-4 They are also at high risk of infection. Distinguishing urinary tract infection (UTI) with sepsis from inconsequential bacteriuria with fever from other sources poses a challenge. Therefore, it is tempting to initiate antibiotic therapy when urine culture (UC) is positive. However, many of these patients have indwelling urinary catheters that predispose them to urinary tract colonization or infection.5-8 Additionally, bacteriuria is common in patients with multiple comorbid conditions,9,10 and attempting to treat this condition usually results in treatment side effects and superinfection with antibiotic-resistant organisms. Guidelines for managing bacteriuria are well established.11 However, application of these guidelines to ICU patients has not been studied. We examined clinicians' approach to UTI in the ICU.
Urine culture results from adult patients in the ICU between November 1, 2004, and March 31, 2005, were retrospectively screened. All patients with positive UC (≥105 colony-forming unit per milliliter) were identified, and their records were reviewed. Demographics, comorbidity, reasons for ICU admission, indwelling urinary catheter placement, vital signs, white blood cell (WBC) counts, urine analysis, and antibiotic treatment were recorded. Definitions are the following: fever was defined as temperature of 38.3°C or higher; leukocytosis was defined as WBC count of greater than 113 WBC/μL; and sepsis/systemic inflammatory response syndrome was defined as 2 or more of the following conditions: temperature of 38.3°C or higher or 36°C or lower, respiratory rate of greater than 20/min or Paco2 of less than 32 torr, heart rate of greater than 90 beats/min, and leukocyte count of greater than 113 or less than 43 WBC/μL.
The frequency of diarrhea and Clostridium difficile colitis was estimated based on the number of stool examination for C. difficile toxins and the test results.
We encountered 119 positive UC during the study period; 29 were excluded: death/discharge before UC results (n = 25) and duplicate UC/same day (n = 4). The remaining 90 positive UC were detected among 87 patients. One patient had 2 separate admissions, and 3 patients had 2 positive UC during the same admissions separated by 2 to 9 days. Each episode was counted as a case.
Mean age of the study group was 62.9 ± 17.6 years. Their characteristics are shown in Table 1. The most common reason for ICU admission was a cardiac cause (26.7%), a central nervous system event (21.1%), and respiratory failure (11.1%). Many patients had leukocytosis (73.3%) and fever/sepsis (46.7%); overall, 20% were considered septic. Urine analysis was obtained in 83 patients. Pyuria was noted in 62.7% of tested patients. Most patients had indwelling urinary catheter for a median of 12 days (range, 1-120 days).
The type of microorganisms recovered from the urine are shown in Table 2. Antibiotics were given in 27 episodes of positive UC in patients with fever plus other possible causes for fever, 11 episodes among febrile patients without other causes, and 5 episodes in patients with urinary symptoms without fever. The other possible causes for fever included pneumonia (n = 5), intracerebral hemorrhage or other cerebrovascular events (n = 5), device-associated infection (n = 5), soft tissue infections (n = 4), meningitis (n = 3), peritonitis (n = 2), C. difficile colitis (n = 2), and acute deep venous thrombosis (n = 1).
Among asymptomatic patients without fever, physicians opted to treat for the possible UTI in 25 (58.1%) of 43 instances, many (52%) were without pyuria (Figure 1). Leukocytosis was noted in 13 of 25 treated episodes and 9 of 18 untreated episodes. Treatment was initiated in 22 (59.5%) of 37 instances in patients with indwelling catheters and 3 (50.0%) of 6 instances in patients without urinary catheters.
Antibiotic treatment was associated with a trend toward higher incidence of diarrhea (37/68 vs. 7/22; P = 0.06) and C. difficile colitis (12/68 vs. 1/22; P = 0.1). No significant differences in the incidence of diarrhea were noted between different treatment groups.
Patients in the ICUs are at high risk of infections, with UTI as the leading cause.1,2 Most published studies address the epidemiology, risk factors, antibiotic choices, and risk reduction.5-8,12-14 Urinary tract infection in ICU patients has been implicated in increased morbidity and length of stay in the ICU.14 However, the benefit of treatment of asymptomatic UTI in the ICU has not been well studied.10 Treatment guidelines addressing asymptomatic bacteriuria were recently updated.11 Their applicability to critically ill patients has not been evaluated. These treatment guidelines do not address patients with sepsis due to infection at other sites or with fever secondary to a noninfectious etiology. Our findings demonstrate that clinicians often treat UTI in the ICU in the absence of symptoms and in the presence of infections at other sources. This approach is inappropriate in asymptomatic patients and questionable in the presence of other conditions. Common sense suggests that treatment without modifying the risk factors in ICU patients will likely lead to untoward side effects, infection with highly resistant pathogens, and C. difficile-associated diarrhea without documented benefit.
The limitations of our study are the retrospective nature with its inherent bias and the small sample size. Our intent was to point out that treatment of UTI in the ICU is often initiated without evidence-based medicine to support this approach. This practice probably adds to antibiotic expenditure, contributes to selection of resistance, and predisposes to C. difficile colitis and other complications. We hope that our study will increase the awareness of the prevalence of UTI in the ICU and the questionable impact on patient outcome and encourage the development of appropriate guidelines. Hence, until UTI practice guidelines address patients in the ICU, we recommend treatment of patients with fever or sepsis that could not be explained by other condition, in pregnancy, recent transplant recipients, and after urinary instrumentations.10,11 Treatment of other patients is probably not warranted. Furthermore, removal of unnecessary catheters should be encouraged.
1. Kennedy V, O'Heron S, Jaloway J, et al. Nosocomial infections in intensive care patients. Crit Care Nurs Clin North Am
2. Laupland KB, Zygun DA, Davies HD, et al. Incidence and risk factors for acquiring nosocomial urinary tract infection in the critically ill. J Crit Care
3. Commichau C, Scarmeas N, Mayer SA. Risk factors for fever in the neurologic intensive care unit. Neurology
4. O'grady NP, Barie PS, Bartlett JG, et al. Practice guidelines for evaluating new fever in critically ill adult patients. Task force of the Society of Critical Care Medicine and the Infectious Diseases Society of America. Clin Infect Dis
5. Bagshaw SM, Laupland KB. Epidemiology of intensive care unit-acquired urinary tract infections. Curr Opin Infect Dis
6. Mnatzaganian G, Galai N, Sprung CL, et al. Increased risk of bloodstream and urinary infections in intensive care unit (ICU) patients compared with patients fitting ICU admission criteria treated in regular wards. J Hosp Infect
7. Laupland KB, Bagshaw SM, Gregson DB, et al. Intensive care unit-acquired urinary tract infections in a regional critical care system. Crit Care
8. Huang WC, Wann SR, Lin SL, et al. Catheter-associated urinary tract infections in intensive care units can be reduced by prompting physicians to remove unnecessary catheters. Infect Control Hosp Epidemiol
9. Nicolle LE. Asymptomatic bacteriuria: when to screen and when to treat. Infect Dis Clin North Am
10. Nicolle LE, Zhanel GG, Harding GKM. Microbiological outcomes in women with diabetes and untreated asymptomatic bacteriuria. World J Urol
11. Nicolle LE, Bradley S, Colgan R, et al. Infectious Diseases Society of America guidelines for the diagnosis and treatment of asymptomatic bacteriuria in adults. Clin Infect Dis
12. Liu H, Mulholland SG. Appropriate antibiotic treatment of genitourinary infections in hospitalized patients. Am J Med
13. Avecillas JF, Mazzone P, Arroliga AC. A rational approach to the evaluation and treatment of the infected patient in the intensive care unit. Clin Chest Med
14. Leone M, Garnier F, Avidan M, et al. Catheter-associated urinary tract infections in intensive care units. Microbes Infect