Urethral catheter-related urinary tract infection (CRUTI) is one of the most frequent infections in patients admitted to the Intensive Care Unit (ICU), after ventilator associated pneumonia (VAP) and primary or vascular catheter-related bacteremia.1,2 In most cases the diagnosis is based on the isolation of a significant number of bacteria or fungi per milliliter of urine in patients with clinical evidence of systemic inflammatory response. The presence of signs or symptoms specific of CRUTI is less common.

The etiology of CRUTI in critical patients admitted to the ICU presents some differences with respect to the urinary infections seen in the rest of the patients admitted to hospital, due to the presence of certain pathogens both in the patient and in the surroundings, selected as a result of frequent antibiotic use.3 The increase in resistances on the part of these pathogens can condition changes in the empirical treatment of these infections. Nevertheless, the information referred to this infectious problem in critical patients is limited.

For years, the Infectious Diseases Working Group of the Spanish Society of Intensive and Critical Care Medicine and Coronary Units (Grupo de Trabajo de Enfermedades Infecciosas de la Sociedad Española de Medicina Intensiva, Crítica y Unidades Coronarias, GTEI-SEMICYUC) has promoted the vigilance of infections acquired in the ICU and related to the use of medical devices. The National Study for the Vigilance of Nosocomial Infections in the ICU (ENVIN-ICU) has registered infections of this kind in Spanish ICUs since 1994.4 The results referred to the evolution of the CRUTI rates, the underlying etiologies, and the multiple resistance markers of the most common pathogens are described in the present study.

During the 6 years analyzed, a total of 78,863 patients were included, of which 2329 (2.95%) suffered one or more CRUTI episodes during admission to the ICU. The characteristics of the patients according to the presence or absence of CRUTI are shown in Table 2. The presence of such infection was frequent in women, in the more seriously ill patients upon admission, and in the subjects classified as corresponding to clinical or surgical cases. Likewise, CRUTI predominated among patients with invasive devices, replacement therapies (parenteral nutrition, extrarenal filtration) and in those with risk factors related to their disease (neutropenia, immune suppression and immune depression–the latter referring to acquired depression and HIV infection).

Table 3 includes the number of CRUTI episodes detected in each of the years of the study, along with the number of patients in which these infections were identified, the number of patients with urethral catheterization, and the total patients-day with urethral catheterization. Based on these variables, calculation was made of the four ways of expressing the frequency of these infections. The ratio between the number of days corresponding to the patients admitted to the ICU with urethral catheterization and the total number of days corresponding to the patients admitted to the ICU remained constant over time, in the order of 0.85 (95%CI: 0.83–0.86), while the different CRUTI incidence indicators decreased significantly (p<0.001) (Table 3).

Table 4 shows the etiology of all the CRUTI episodes. Gramnegative bacilli were seen to predominate during the analyzed time period, representing between 55 and 60% of all isolates, and with a particularly important presence of E. coli. This was followed by fungi (19–25% of all isolates), all pertaining to the genus Candida, where C. albicans was identified in one-half of the episodes. Among the grampositive cocci, enterococci were seen to predominate (particularly E. faecalis).

The systemic response associated to CRUTI is reported in Table 5. In one-half of the cases there was no systemic inflammatory response, and in less than 15% of the cases the response was classified as serious.

The multiresistance markers of the main pathogens causing CRUTI showed important differences (Table 6). While in the case of E. coli we observed a progressive increase in strains resistant to ciprofloxacin, no such increase was observed in relation to cefotaxime. In the case of P. aeruginosa, resistance to ciprofloxacin and imipenem exceeded 30%, while resistance to ampicillin and vancomycin was practically inexistent in E. faecalis. Likewise, no C. albicans strains were found to be resistant to fluconazole.

The patients with CRUTI had a longer stay in the ICU (25.59 versus 4.39 days in those without CRUTI) and greater global mortality in the ICU (22.4% versus 8.6%)–though the available data do not allow us to establish a relationship between these variables and the presence of CRUTI.

The main contribution of this study has been the description of the CRUTI rates, etiologies and multiresistance markers in critical patients admitted to the ICU in the last 6 years (2005–2010). The evolutive data reflect a significant decrease in CRUTI, independent of the rate used to express the infections. In the last year of the study, the rate expressed as incidence density (ID) was 4.18 episodes per 1000 days of urethral catheterization. This is considered to be within the limits published by the National Healthcare Safety Network (NHSN) in December 2009 with the cumulative data corresponding to the period between 2006 and 2008 in American ICUs.13 The ID of CRUTI varied between 7.4 episodes per 1000 days of urethral catheterization in the ICU for burn or neurological patients and 3.1 episodes per 1000 days of urethral catheterization in clinical-surgical ICUs with more than 15 beds–the national average calculated from the data being 4.08 episodes per 1000 days of urethral catheterization. The references to CRUTI in the literature range from 1.4 episodes per 1000 days of urethral catheterization in 12 ICUs in India14 to 5.1 episodes per 1000 days of urethral catheterization in four Peruvian ICUs,15 while the most recently published data report 15.7 episodes per 1000 days of urethral catheterization in four ICUs in Alexandria (Egypt).16

The CRUTI rates in Spanish ICUs have decreased significantly in recent years, without the application of any specific national project or initiative for the prevention of these infections. During this period of time there have been specific campaigns aiming to promote the washing of hands among healthcare workers in most Spanish hospitals.17 Furthermore, a national campaign has been started for preventing catheter-related bacteremias in the ICU (the bacteremia zero project), based on the introduction of a package of measures (one of which consists of washing of the hands before and after manipulating the catheter),18 and the concept of integral patient safety has been divulgated through courses and practical cases. Globally, these activities designed to promote a safe environment in the ICU may have contributed to improve the CRUTI rates.The etiology of CRUTI has remained without change during these years, being headed by gramnegative bacilli (especially E. coli), followed by other microorganisms such as E. faecalis and C. albicans.19 Other pathogens that may be endemic to ICUs and remain in the environment are much less frequent, including methicillin-resistant S. aureus (MRSA), P. aeruginosa and Acinetobacter baumannii (A. baumannii), which are the main causes of other nosocomial infections such as ventilator associated pneumonia (VAP).20 This underscores the importance of the intestinal flora in the etiology of these infections and the need to maximize all measures that limit or complicate the contamination of urethral catheters.21–23

The treatment of CRUTI is conditioned by the evolution of the resistances of the most frequent pathogens. Until recently, the quinolones and semisynthetic penicillins were the treatment of choice in the presence of gramnegative bacilli in urine staining. At present, in those cases where CRUTI is accompanied by signs of severe sepsis or septic shock, empirical treatment must consider the presence of extended spectrum betalactamase (ESBL)-producing E. coli (resistant to cephalosporins) or of a non-fermenting gramnegative bacillus (NFGNB), where the combination of a carbapenem and an aminoglycoside active against NFGNB could be the treatment of choice. This combination cannot be maintained for the full treatment period, however, and must be adjusted as soon as possible to the sensitivity profile of the causal pathogen. In cases characterized by the presence of yeasts, the treatment of choice is fluconazole. In contrast, in those situations in which CRUTI is not associated to a systemic inflammatory response, the possibility of not prescribing antimicrobial treatment may be considered.

The mortality rate and duration of stay in the ICU in patients with CRUTI are greater than in patients without this type of infection, though from the information obtained in our study it cannot be deduced that CRUTI increases either mortality or the duration of admission. More complex studies, based on case-control methodology and with patient populations more homogeneous than in our series would be needed in order to draw such conclusions. The published data referred to mortality are contradictory. Recently, a metaanalysis has been published that suggests a statistically significant impact upon mortality and upon the duration of admission to the ICU and hospital stay–though the analysis fails to specify whether use was made of those studies that perform adjustments for the included variables. A probability of in-ICU mortality of 1.94 (95%CI: 1.61–2.34) was identified.24,25

Like all multicenter studies involving voluntary participation, our work is affected by patient selection bias that is offset by the broad participation of ICUs and the volume of patients accumulated. On the other hand, application of the definition of CRUTI in the critical patient (often sedated and/or subjected to analgesia) is difficult, particularly in reference to differentiation with respect to asymptomatic bacteruria; the true rates therefore may have been overestimated. This circumstance is aggravated by the increasing tendency to conduct active multiresistant flora vigilance studies that include urine samples from asymptomatic patients or individuals with other infectious foci.

It can be concluded that CRUTI shows a significant decrease in incidence in Spain, without the adoption of any specific interventions designed to secure such a decrease. The distribution of causal pathogens remains stable, with a predominance of gramnegative bacilli (particularly E. coli). Empirical treatment of the more serious cases must consider coverage of multiresistant pathogens that have become increasingly prevalent–particularly E. coli and P. aeruginosa–while in patients without a systemic inflammatory response the possibility of not prescribing antimicrobial treatment may be considered.

The authors declare that they have no conflicts of interest.