Which indwelling urethral catheters resist encrustation by Proteus mirabilis biofilms?

Transcription

1 British Journal of Urology (1997), 80, Which indwelling urethral catheters resist encrustation by Proteus mirabilis biofilms? N.S. MORRIS, D.J. STICKLER and C. WINTERS School of Pure and Applied Biology, University of Wales Cardi, Wales UK Objective To test the resistance of currently available Results The mean times to blockage ranged from 21 h types of indwelling urethral catheters to blockage for the Bard hydrogel/silver-coated latex catheter to by encrustation with mineralized Proteus mirabilis 56 h for the Eschmann Folatex S all-silicone catheter. biofilms. The calcium and magnesium salts were mainly Materials and methods Encrustation was studied in a deposited on the 10 cm below the eye-holes of the simple laboratory model of the catheterized bladder. catheters, complete blockage generally occurring in Artificial urine was supplied to the bladder chamber the 2 cm immediately below the eye-hole. at 0.5 ml/min. The bladder urine was inoculated with Conclusion None of the 18 types of catheter tested, a clinical strain of P. mirabilis that had been isolated including those coated with hydrogel or silver, were from an encrusted catheter. The models were operated capable of resisting encrustation by P. mirabilis biofilm. until the catheters blocked and atomic absorption Keywords Catheter encrustation, bacterial biofilms, spectrometry was used to assess the amounts of Proteus mirabilis calcium and magnesium deposited on the catheters. Scanning electron microscopy was also used to locate and assess the degree of encrustation. Introduction Long-term indwelling urethral catheters are commonly subject to encrustation [1 3]. Crystalline deposits can cover the balloon and obstruct the eye-hole and lumen of the catheter. They can be extremely hard and can cause trauma to the bladder mucosa and to the urethra on withdrawing the catheter. Blockage of the catheter lumen can lead to retention of urine and painful distension of the bladder. As bacteriuria is inevitably associated with the encrustation and blockage, retention can facilitate ascending infection of the urinary tract, culminating in episodes of pyelonephritis, septicaemia and shock [4]. Unnoticed catheter blockage can thus be very dangerous, particularly for patients being cared for in the community where professional care is not immediately available. Several surveys have reported that around half of patients undergoing this form of long-term bladder management will experience recurrent encrustation and blockage of catheters [5 7]. The catheter encrustations resemble infection-induced stones in their crystalline structure, being composed of a mixture of struvite (magnesium ammonium phosphate hexahydrate) and the partly crystalline form of calcium Accepted for publication 4 February 1997 phosphate, hydroxyapatite [8,9]. The early work of Grith et al. [10] showed the central role of ureaseproducing bacteria in the generation of the bladder and kidney stones. The enzymatic hydrolysis of the urea produces an alkaline environment in which calcium and magnesium phosphates crystallize out of solution. Examination of these infection-induced stones by scanning electron microscopy confirmed the presence of microbial cells throughout their structure [11]. Similarly, scanning electron microscopy by Cox et al. [12] showed the presence of many bacilli associated with catheter encrustations and substantiated the view that they are generated by the processes responsible for the formation of infection-induced stones. Bacteriological analysis subsequently confirmed that the urease producer Proteus mirabilis was the predominant organism in encrusted catheter biofilms [13]. Recent work [14 17] on the mechanism of encrustation suggests that the steps in the process are: (i) the infection of the urinary tract by P. mirabilis, or other urease-producing species, (ii) the formation of an organic conditioning film on the catheter surface, (iii) the adherence of the urease-producing bacteria to the catheter, (iv) the development of a bacterial biofilm community within a matrix of bacterial exopolysaccharide, (v) the elevation of the ph of the urine and biofilm matrix by British Journal of Urology

2 ENCRUSTATION OF P. MIRABILIS OF INDWELLING CATHETERS 59 the action of urease on urea, (vi) the attraction of di-sodium oxalate 0.02 g/l, potassium dihydrogen phosphate calcium and magnesium ions into the gel of the matrix, 2.8 g/l, potassium chloride 1.6 g/l, ammonium and (vii) the alkali-induced, gel-stabilized crystallization chloride 1.0 g/l, urea 25 g/l, gelatine 5.0 g/l. The ph of calcium and magnesium ammonium phosphates. It of the medium was adjusted to 6.1 and then sterilized also seems that throughout the encrustation process, by membrane filtration. Tryptone Soya Broth (Oxoid, crystals formed in the alkaline urine of the bladder can UK) was prepared separately, autoclaved and added to attach to the surfaces of the catheter and the biofilm. the sterile basal medium to a final concentration of Attempts have been made to make urethral catheters 1.0 g/l. less attractive to bacteria by applying coatings of hydrophilic hydrogels or antibacterial agents such as silver. While some manufacturers claim that they have developed The bladder model catheters that are highly resistant to encrustation, The model of the catheterized bladder has been described there is little information available on the relative performance previously [18]. In essence, it consists of a glass fermen- of the various types of catheters that are tation flask maintained at 37 C by a water jacket. The currently available. Stickler et al. [18] devised a simple model was sterilized by autoclaving and then a 14 F physical model of the catheterized bladder to study the catheter was inserted into the flask through a section of encrustation process in the laboratory [17]. We have silicone tubing (a urethra ) attached to a glass outlet at now used the model to compare the ability of a range of the base of the flask. The catheter balloon inflated in the commercially available catheters to resist encrustation usual way, securing the catheter in position and sealing by mineralized P. mirabilis biofilms. the outlet from the bladder. Sterile urine was then supplied to the bladder at 0.5 ml/min such that a residual volume of 30 ml collected in the bladder below the level Materials and methods of the catheter eyelet and then flowed through the Growth media catheter and drainage tube to a collecting bag. Eighteen dierent types of 14 F catheters were tested (Fig. 1). The artificial urine used in the experimental work was based on that devised by Grith et al. [10] and contained calcium chloride 0.49 g/l, magnesium chloride hexahydrate Experimental protocol 0.65 g/l, sodium chloride 4.6 g/l, di-sodium Sets of models were assembled and the bladders primed sulphate 2.3 g/l, tri-sodium citrate dihydrate 0.65 g/l, with urine. The test organism, P. mirabilis NSM6 had 18 Bard Hydrogel/Silver Coated Latex 17 Bard Hydrogel Coated Latex 16 Warne Silicone Treated Latex 15 Warne Teflon Coated Latex 14 Simpla Silicone Coated Latex 13 Rusch Ultrasil Silicone Coated Latex 12 Bard Teflon Coated Latex 11 Eschmann Folatex Silicone Treated Latex 10 Eschmann Silicone Treated Latex Tiemann Tip 9 Bard Silicone Coated Latex 8 Rusch Silkolatex SL Cath (Silver impregnated) 7 Rusch Simplastic 6 Rusch Silikon Simpla All Silicone 4 Eschmann Silicone Treated Latex Stewart Tip 3 Rusch Simplastic Whistle Tip 2 Bard All Silicone 1 Eschmann Folatex S All Silicone Time (h) Fig. 1. The time taken for 18 dierent types of catheter to block in the bladder model infected with Proteus mirabilis (mean of four replicates). Catheter

3 60 N.S. MORRIS, D.J. STICKLER and C. WINTERS been isolated from a patient s encrusted catheter 3 indwelling catheters in elderly patients who were blockers months before the start of this study. For experimental and concluded that the encrustation and blockage purposes, cells were sub-cultured monthly from a stock was significantly less with silicone catheters than with suspension in 5% glycerol stored at 80 C. The bladders Teflon-coated or latex catheters. A problem in comparing were inoculated with 10 ml of a 4 h urine culture of the performance of catheters in this way is ensuring that the test strain. After one hour to allow the organisms to in each case the catheters are exposed to a standard establish themselves in the model, urine was supplied. challenge. Although these patients all had significant Catheters were examined in sets of 4 6 in each experiment bacteriuria, it was not confirmed that throughout the with the Bard all-silicone catheter included in each trial period, the patients were all infected with P. mira- set as a reference. The models were operated until the bilis. An advantage of the bladder model is that during catheters blocked, recording the time to blockage in each the test period it is possible to ensure that each catheter case. The experiments were each replicated four times. is exposed to a defined population of P. mirabilis in urine In three cases the catheters were removed from the (108 cells/ml) throughout the test period. A chemically models and the extent of encrustation was assessed on defined artificial urine was used as it allowed a compari- serial sections cut from the tip to the base. The first son of the relative performance of a range of catheters section (1 cm) included the eye-hole and the second under standard experimental conditions. However, we section (1 cm) included the portion of the catheter under recognize that factors present in human urine may also the balloon; the subsequent sections were all 2 cm long. influence the encrustation process. Each section was placed into 4% nitric acid (10 ml) in In the laboratory model, 14 F catheters were used, as a universal container. The encrustations were then in our experience these are used most commonly in disrupted by ultrasonication for 5 min (Transsonic Water patients undergoing long-term care in the community. Bath, Camlab, UK). The calcium and magnesium content The system provides a severe but controlled test of the of the resulting solutions were then determined by ability of the catheters to resist encrustation and is atomic absorption spectroscopy. equivalent to placing a new catheter into the bladder of The encrustation was assessed visually on the fourth a patient who already has significant bacteriuria with set of each catheter using low-vacuum scanning electron P. mirabilis. The results presented in Fig. 1 show that all microscopy. Cross-sections were made of each catheter types of catheter currently available are susceptible to at 1, 4, 6, 12, 22 and 32 cm from the tip, mounted on encrustation by mineralized P. mirabilis biofilms. Overall, carbon discs on aluminium stubs and viewed directly the all-silicone catheters took longer to block than the using the low-vacuum setting of a JEOL 5200 LV SEM Teflon, silicone or hydrogel-coated catheters. It was at 20 kv. The low-vacuum facility allows the direct particularly interesting that some of the catheters that examination of specimens that have not been fixed, have been designed to resist bacterial colonization and stained or treated in any way. encrustation blocked most rapidly. Hydrogel- and hydrogel/silver-coated catheters all blocked within h; these coatings obviously failed to prevent encrustation. Results Previous studies on hydrogel coatings suggested that All catheter types tested encrusted rapidly; the mean they should perform at least as well as all-silicone time taken for each type of catheter to block, calculated catheters. For example, Cox et al. [19], in an in vitro from four series of experiments, is presented in Fig. 1. study, examined encrustation on 6 cm sections of cath- The mean times to blockage ranged from just 21 h for eters (that had also been cut longitudinally). The sections the Bard hydrogel/silver coated-latex to 56 h for the were exposed to artificial urine in which crystallization Eschmann Folatex S all silicone catheter. An examination of struvite and apatite was induced by adding the enzyme of the catheter design revealed considerable dierences urease. In a similar test system, Liedberg et al. [20] found in the dimensions of the central channels and the eye- that significantly (P<0.05) less encrustation took place holes (Fig. 2). on silver-coated latex than on latex or hydrogel-coated The electron micrographs (Fig. 3) confirmed that the latex. These test systems examined the deposition of blockage of the catheters occurred in the first few struvite and apatite crystals onto catheter surfaces, but centimetres below the eye-hole. Amorphous apatite and did not examine the ability of catheters to resist blockage large crystals of struvite were clearly visible (Fig. 3). brought about by colonization of the luminal surfaces by mineralizing bacterial (P. mirabilis) biofilm. Bull et al. [21] reported a clinical study which Discussion compared the performance of the Bard hydrogel-coated Kunin et al. [5] reported a clinical study of the eect of catheter with a silicone-coated catheter in a group of 69 catheter materials on the encrustation of long-term patients. The catheters were left in situ for 16 weeks and

4 ENCRUSTATION OF P. MIRABILIS OF INDWELLING CATHETERS 61 Fig. 2. Cross-sections of catheters taken just below the eye-holes, illustrating the dierences in dimensions and designs of the central channels. The eye-holes of each type of catheter are also shown. only changed when necessary for a medical reason. The have larger diameters than those of the coated catheters hydrogel catheters remained in place for significantly (Fig. 2). For example, the Bard all-silicone catheter has longer than the silicone-coated ones. The patients were an internal diameter of 2.8 mm compared with 1.8 mm reported to prefer the hydrogel catheters and leakage of for the hydrogel and hydrogel/silver-coated latex cath- urine was less of a problem than with the silicone eters. Regression analysis gave a significant linear correlation catheters. However, interestingly, while no indication between internal diameter and time to blockage was given of the incidence of P. mirabilis infections in (correlation coecient, r=0.72, P<0.01). The internal the patients over the 16 weeks, 11 hydrogel catheters diameter is thus not surprisingly a major factor in had to be changed because of encrustation, compared determining time to blockage. The eye holes of some of with nine of the silicone-coated catheters. the catheters are also very small and vulnerable to While it is clear from Fig. 1 that the all-silicone blockage by clumps of crystals formed in the bladder. catheters took longer to block than the other types it is Silver is a potent antibacterial agent [22] but failed to not possible to conclude that this was due to the catheter block colonization by a mineralizing bacterial biofilm. In surface as the central channels of all-silicone catheters a previous report on the formation of encrustations on

5 62 N.S. MORRIS, D.J. STICKLER and C. WINTERS are needed to develop catheters which resist encrustation and blockage. The ideal catheter should have surfaces which resist bacterial colonization and crystal deposition. It should also have a wide lumen and large eye-holes. In the meantime, the only sure way to avoid the problem of catheter encrustation and blockage is to avoid the use of the catheter. Acknowledgements a This study was supported by a grant from the Welsh Scheme for the Development of Health and Social Research. b Fig. 3. Low-vacuum scanning electron micrographs of sections of a blocked Bard hydrogel-coated latex catheter. a, A section 3 cm from the tip. b, A section 12 cm from the tip. catheter surfaces in the bladder model, we showed that crystals formed in the bladder urine can attach directly to the catheter surface. Biofilm then developed by bac- terial colonization of the crystals [17]. It also appears that pioneering cells killed as they attempt to colonize 10 Grith DP, Musher DM, Itin C. Urease: the preliminary an antibacterial surface provide receptors for subsequent cause of infection-induced urinary stones. Invest Urol 1976; bacterial colonization, shielded from the antibacterial 13: agent on the catheter. Therefore, the results from these 11 Takeuchi H, Takayama H, Konishi T, Tomoyoshi T. References 1 Bruce AW, Sira SS, Clark AF, Awad SA The problem of catheter encrustation. Can Med Assoc J 1974; 111: Hukins DWL, Hickey DS, Kennedy AP Catheter encrustation by struvite. Br J Urol 1983; 55: Getlie KA, Mulhall AB The encrustation of indwelling catheters. Br J Urol 1991; 67: Kunin CM. Detection, Prevention and Management of Urinary Tract Infections. 4th edn. Philadelphia: Lea & Febiger, Kunin CM, Chin QF, Chambers S. Formation of encrustations on indwelling urinary catheters in the elderly: a comparison of dierent types of catheter materials in blockers and non-blockers. J Urol 1987; 138: Getlie KA. The characteristics and management of patients with recurrent blockage of long-term urinary catheters. J Adv Nursing 1994; 20: Kohler-Ockmore J, Feneley RCL. Long-term catheterisation of the bladder: prevalence and morbidity. Br J Urol 1996; 77: Hedelin H, Eddeland A, Larsson, L, Pettersson S, Ohman S. The composition of catheter encrustations, including the eects of allopurinol treatment. Br J Urol 1984; 56: Cox AJ, Hukins DWL. Morphology of mineral deposits on encrusted urinary catheters investigated by scanning electron microscopy. J Urol 1989; 142: in vitro tests suggest that the simple strategy of incorporinfection Scanning electron microscopy detects bacteria within ating an antibacterial agent into the catheter material stones. J Urol 1984; 132: 67 9 is unlikely to prevent the bacteria adhering to and 12 Cox AJ, Hukins DWL, Sutton TM. Infection of catheterised colonizing available surfaces [23]. patients: bacterial colonisation of encrusted Foley catheters The chemical analysis and the electron microscopy of shown by scanning electron microscopy. Urol Res 1989; 17: the catheter sections (Fig. 3) showed that encrustation 13 Stickler D, Ganderton L, King J, Winters C. Proteus mirabilis formed mainly on the 10 cm length of catheter below biofilms and the encrustation of urethral catheters. Urol the eye-hole. Similar observations were made by Kunin Res 1993; 21: et al. [24] on encrusted catheters removed from patients. 14 Ohkawa M, Sugata T, Sawaki M, Nakashima T, Fuse H, The results presented in Fig. 1 show that none of the Hisazumi H. Bacterial and crystal adherence to the surfaces catheters currently available are able to resist coloniz- of indwelling urethral catheters. J Urol 1990; 143: ation by mineralized P. mirabilis biofilm. New approaches 15 Clapham I, Mclean RJC, Nickel JC, Downey J, Costerton

6 ENCRUSTATION OF P. MIRABILIS OF INDWELLING CATHETERS 63 JW. The influence of bacteria on struvite crystal habit and randomised, parallel group study of the Bard Biocath its importance in urinary stone formation. J Crystal Growth catheter and a silicone elastomer coated catheter. Br J Urol 1990; 104: ; 68: Dumanski AJ, Hedelin H, Edin-Lilijegren A, Beachemin D, 22 Russell AD. Antimicrobial activity and action of silver. Prog McLean RJC. Unique ability of the Proteus mirabilis capsule Med Chem 1994; 31: to enhance mineral growth in infectious urinary calculi. 23 Costerton JW, Lewandowski Z, Caldwell DE, Korber DR, Infect Immun 1994; 62: Lappin-Scott HM. Microbial Biofilms. Ann Rev Microbiol 17 Winters C, Stickler DJ, Howe NS, Williams TJ, Wilkinson 1995; 49: N, Buckley CJ. Some observations on the structure of 24 Kunin CM, Chin QF, Chambers S. Indwelling urinary encrusting biofilms of Proteus mirabilis on urethral catheters. catheters in the elderly. Am J Med 1987; 82: Cells Materials 1995; 5: Stickler DJ, Howe NS, Winters C. Bacterial biofilm growth on ciprofloxacin treated urethral catheters. Cells Materials 1994; 4: Authors 19 Cox AJ, Millington RS, Hukins DWL, Sutton TM. Resistance N.S. Morris, BSc, Postgraduate Research Student. of catheters coated with a modified hydrogel to encrustation D.J. Stickler, BSc, MA, DPhil, Senior Lecturer in Medical during an in vitro test. Urol Res 1989; 17: Microbiology. 20 Liedberg H, Ekman P, Lundeberg T. Urease-induced C. Winters, CBiol, Electron Microscopist. precipitation on latex and silver coated latex in vitro. Scand Correspondence: Dr D.J. Stickler, School of Pure and Applied J Urol Nephrol Suppl 1991; 138: Biology, University of Wales Cardi, PO Box 915, Cardi CF1 21 Bull E, Chilton CP, Gould CAL, Sutton TM. Single-blind, 3TL, UK.