Biofilm-Based Infections in Long-term Care Facilities

Gianfranco Donelli; Claudia Vuotto


Future Microbiol. 2014;9(2):175-188. 

In This Article

Catheter-Associated Urinary Tract Infections

Urinary tract infections (UTIs) are known to contribute to the morbidity of patients recovered in LTCFs even if asymptomatic bacteriuria is very frequent, as confirmed by several prevalence studies performed in different countries, reporting that 15–30% of urine samples from men and 25–50% from women were culture positive for at least one uropathogen.[16] The chronic comorbid conditions requiring patient admittance to the LTCF are widely recognized as the main causative factors of bacteriuria. In fact, cerebrovascular accidents as well as degenerative neurological diseases, including Alzheimer's and Parkinson's disease, are associated with an impaired emptying of the bladder (neurogenic bladder) and related ureteric reflux, with both these phenomena contributing to bacteriuria.[17] Approximately 80% of UTIs in LTCFs are due to the short- or long-term insertion of a urinary catheter.[18] In hospitals for neuromotor rehabilitation, a long-term bladder catheterization is required for most patients.[19,20] The main causes of CAUTIs are the following:

  • Transfer of microorganisms from the urethra to the bladder following catheter insertion. However, when the urinary tract is not affected by specific diseases, bacteria are promptly cleared by the defense mechanism of bladder mucosa, the most efficacious being represented by polymorphonuclear leukocytes;

  • Bladder contamination as a consequence of the entry of bacteria through the thin fluid layer present at the urethral mucosa–catheter interface. This contamination route is favored by catheter movements and involves intestinal bacteria colonizing the patient perineum that are able to ascend the urethra after catheter insertion;

  • Bacterial migration through the catheter lumen as a consequence of contamination of the drainage bag system. Approximately 30% of CAUTIs are estimated to be associated with this intraluminal contamination route, while the first two extraluminal contamination routes account for 70% of CAUTIs. The involved microorganisms are usually acquired from external sources, mainly via the hands of clinical staff, with this source reaching up to 15% of infections in groups of hospitalized patients.[21]

Patients with a urinary catheter in place for over 30 days are considered to have a long-term, or chronic, catheter.[22] Of course, the longer urinary catheters remain in place, the greater the tendency of bacteria to colonize their surfaces and grow as biofilm (Figure 2A). It has been repeatedly reported that two to five different microorganisms are usually detectable in biofilm-based CAUTIs, the most common ones being Gram-negative Enterobacteriaceae, including Escherichia coli, Klebsiella pneumoniae and Citrobacter freundii, as well as urease-producing bacteria, such as Proteus mirabilis, Morganella morganii and Providencia stuartii (Figure 2B). Other Gram-negative species frequently occurring in CAUTIs are Pseudomonas aeruginosa and Acinetobacter baumannii. Gram-positive microorganisms, such as coagulase-negative staphylococci, enterococci and group B streptococci, are also quite often isolated. Yeast may be sometimes isolated from patients treated with repeated courses of antibiotics, Candida albicans being the most common, followed by Candida glabrata and Candida tropicalis.[23,24]

Figure 2.

Field emission scanning electron microscopy micrographs of a polymicrobial biofilm grown in the lumen of a urinary catheter removed from a patient recovered at the research hospital for neuromotor rehabilitation Fondazione Santa Lucia in Rome. The species identified by culture methods were Klebsiella pneumonia, Enterococcus faecalis and Acinetobacter baumanni. (A) 2000×; (B) 20,000×.

Concerning the process of biofilm formation, the insertion of a urinary catheter in a patient is soon followed by adhesion to the device surfaces of host proteins, forming a 'conditioning layer' able to promote microbial adherence and biofilm formation.[25,26] Given the significant role of biofilm in the pathogenesis, treatment and prevention of CAUTIs,[27,28] appropriate attention has been recently focused on the clinical implications of microbial sessile growth. In fact, the establishment of bacterial biofilm communities on urinary catheters is believed to inhibit the effectiveness of antibiotic treatment, bypass host defense mechanisms and facilitate bacterial communication, leading to the expression of virulence determinants.[29] Specimens of urinary catheters removed from patients after several days of permanence and observed by scanning electron microscopy (SEM) frequently show the multispecies nature of these biofilms, sometimes adding useful information on the species involved not always identified by culture-dependent techniques. If a catheter remains in situ for a sufficient number of days to allow biofilm formation, just the culture of a urine sample collected from the catheterized patient could be used to detect the bacterial composition of the biofilm developed in the catheter lumen, avoiding the need to analyze the catheter itself. However, it must be emphasized that this approach could lead to false negative results.

Among the bacterial species frequently isolated from CAUTIs,[30] the most investigated for their role in biofilm formation are E. coli, K. pneumoniae, P. mirabilis and, more recently, the emerging A. baumannii. In recent years, these species have recorded an increasing spread at hospital level and important changes in their antimicrobial resistance patterns,[31] exacerbated by the well-known reduced susceptibility of biofilms to antibiotics.[32]

E. coli is a Gram-negative, facultative anaerobic, rod-shaped bacterium, causing more than 30% of nosocomially acquired UTIs and is the most commonly isolated pathogen from urines of patients affected by CAUTIs.[33–35]

Similarly to the uropathogenic E. coli strains, strains isolated from CAUTIs express specific O antigen and capsule (K) types, and exhibit a large number of virulence factors, including toxins and adhesins, the best characterized ones being P and type 1 fimbriae, both involved in adherence to catheters.[36–38] Several factors have been reported to be associated with E. coli biofilm development, including curli, flagella, type 1 and F9 fimbriae,[39,40] and antigen 43.[41,42] The production and functional role of type 3 fimbriae were found to influence biofilm growth of E. coli.[43]

Few studies have so far compared different E. coli phylogenetic groups involved in UTIs with the presence of virulence factors, biofilm-forming ability and antimicrobial resistance.[42,44] In particular, Ejrnæs and coworkers have tried to find a correlation between biofilm production and relapse or persistence of uropathogenic E. coli strains isolated from community-acquired lower UTIs in women.[42] We have reported, in collaboration with Portuguese colleagues, on the lack of a positive correlation between biofilm production ability of different E. coli clonal groups belonging to phylogenetic group D and their high transmissibility among different settings. Briefly, biofilm production seems not to be directly related to the epidemiological success of these strains mostly isolated from urine.[45]

The Gram-negative, facultatively anaerobic, rod-shaped P. mirabilis is the second most frequent etiological agent of CAUTIs after E. coli. It exhibits a significant number of virulence factors, including fimbriae, flagellar motility, capsule antigen, lipopolysaccharide (LPS), struvite and apatite crystal formation, urease and haemolisin production, adhesiveness, and swarming ability, the latter enabling Proteus to colonize and survive.[46–51] The well-known phenomena of encrustation and occlusion of catheters are related to the establishment of catheter colonization by urease-producing bacteria, able to generate ammonia from urea and elevate the urine pH. As the pH value increases, long rectangular crystals made of magnesium ammonium phosphate (struvite) and hydroxylated calcium phosphate microcrystals (apatite) precipitate, giving rise to the development of a microcrystalline layer, which is then rapidly coated by a dense polymicrobial biofilm consisting of P. mirabilis, E. coli and P. aeruginosa.[27,52]

Biofilm development causes the slowing of urine flow within the catheter,[53–55] all types of Foley catheters being affected by this problem. Currently, there are no effective procedures available for its control. In fact, experiments in a laboratory model developed by Stickler and Morgan have shown that even antimicrobial catheters, such as those coated with silver or impregnated with nitrofurazone, are rapidly encrusted and blocked by biofilm-forming, urease-producing bacteria, particularly P. mirabilis.[53] Observations on encrusted silver catheters removed from a patient have confirmed that the crystalline layer forms also in vivo.[53] Apart from the pivotal role of urinary pH in the adhesion of P. mirabilis to catheter surfaces, surface irregularities of the Foley catheter eyeholes promote colonization and crystallization, and lead to device blockage, as revealed by SEM investigations.[56]

The Gram-negative, rod-shaped, encapsulated and nonmotile K. pneumoniae is today considered the causative agent of approximately 10–20% of CAUTIs.[35,57]

In recent years, special attention has been focused on the increased incidence of extended-spectrum β-lactamases and carbapenemase-producing K. pneumonia, not only in general hospital but also in rehabilitation centers.[58]

The pathogenicity of K. pneumoniae has been attributed to different virulence factors, including membrane transporters, antiphagocytic capsule, LPS and siderophores.[59–62] On the other hand, its persistence seems to be associated with the ability to adhere to specific epithelial tissues or abiotic surfaces, and this process, responsible for the first step of biofilm formation,[63] has been reported to be mediated by type 1 and 3 fimbriae.[64,65] Furthermore, both the capsule and type 1 and 3 fimbriae, together with the contribution of a type 2 quorum-sensing regulatory system, have been shown to be involved in the ability of K. pneumoniae to form biofilm.[66–70] Moreover, signature-tagged mutagenesis (STM) showed that, beyond the extracellular components (capsule and fimbriae), the genes ORF-4 (wza homologous, outer membrane protein for the transport of polysaccharides), wbbM (synthesis of D-galactan I), wzm (an ABC-2 membrane family involved in the LPS transport pathway) and pgaA (encoding poly-β-1,6-N-acetyl-D-glucosamine) are also involved in K. pneumoniae biofilm formation.[71,72]

Until now, only one comparative study is available on the effect of different classes of antibiotics on different phases of biofilm development in K. pneumoniae. Antibiotics compared were piperacillin, amikacin sulfate and ciprofloxacin hydrochloride, belonging to β-lactam, aminoglycoside and fluoroquinolone classes, respectively. Of the antibiotics tested, amikacin appeared to be the most effective in the eradication of K. pneumoniae young biofilm, the higher production of EPS making the mature biofilm less permeable to antibiotics.[73]

A. baumannii is known to survive in clinical settings and very effectively colonize the surfaces of urinary catheters. The long persistence in medical environments[74] and the rising antibiotic resistance reported for these species[75] could be related to their ability to form biofilm.[76–78] Multidrug- and pan-drug-resistant A. baumannii have been increasingly isolated from compromised patients with an underlying illness.

Some features of this microorganism have been associated with biofilm formation, such as a pilus assembly system,[79] a homolog to the staphylococcal Bap involved in intercellular adhesion within the mature biofilm,[80] the outer membrane protein OmpA needed for the formation of robust biofilms on polystyrene,[81] a poly-β-(1-6)-N-acetylglucosamine[82] and an autoinducer synthase, which is involved in quorum-sensing and required for normal biofilm formation.[83]

An increasing number of A. baumannii clinical isolates have been reported worldwide as multidrug-resistant microorganisms; that is, resistant to three or more of the following antibiotic classes: β-lactams, including penicillins, cephalosporins and monobactams; carbapenems; fluoroquinolones; and aminoglycosides. On the other hand, pan-drug-resistant A. baumannii strains, defined as resistant to all available antibiotic options, are often isolated at hospital level.[84–86] Vidal and coworkers firstly evaluated the efficacy of antibiotic molecules on A. baumannii biofilm, reporting a bactericidal effect of sulbactam and imipenem on young biofilms.[87] On the contrary, on mature biofilms, sulbactam exhibited a greatly reduced activity, while imipenem substantially maintained its bactericidal effect. A more recent multicenter cohort study reported that biofilm-forming A. baumannii isolates (56/92 strains; 63%) were less frequently resistant to imipenem or ciprofloxacin than nonbiofilm-forming isolates were (25 vs 47%; p = 0.04 and 66 vs 94%; p = 0.004, respectively).[78] Data on the susceptibility of biofilm-growing A. baumannii to colistin report that the full bactericidal effect was obtained when its concentration reached 8 μg/ml, a value approximately 50-times higher than the MIC (0.125 μg/ml). This polymyxin was more effective on mature than young biofilm.[88]