The combined effects of plasma and hydrogel coating on adhesion of Staphylococcus epidermidis and Staphylococcus aureus to polyurethane catheters

MICROBIOLOGY LETTERS

EL-SEWER

FEMS Microbiology Letters 144 (1996) 241-247

The combined effects of plasma and hydrogel coating on adhesion of Staphylococcus epidermidis and Staphylococcus aureus to polyurethane catheters Stephanie

F. John ‘, Max R. Derrick b, Alan E. Jacob ‘, Pauline S. Handley aj*

a School of Biological Sciences, 1.800 Stopford Building, Manchester University. Oxford Road, Manchester, Ml3 9PT, UK b Ohmeda, Faraday Road, Dorcan. Swindon SN3 SJH, UK

Received 20 June 1996; revised 26 August 1996; accepted 27 August 1996

Abstract The adhesion of three Staphylococcus epidermidis and three S. aureus clinical isolates, to uncoated and hydrogel-coated polyurethane catheters was tested, following pretreatment of catheters with human plasma. Plasma significantly decreased the adhesion of S. epidermidis strains to uncoated polyurethane catheters, but had no significant effect on the adhesion to hydrogel-coated catheters. The influence of plasma on adhesion of S. aureus strains to catheters was strain dependent. Plasma significantly increased the adhesion of one strain (SA6) to uncoated catheters. For two other strains (SA3 and SA14) plasma produced no clear effect on their adhesion to uncoated catheters; adhesion values for each strain showed either a small but significant increase or a replicate-dependent increase or decrease. However, plasma significantly increased the adhesion of all S. aureus strains to hydrogel-coated polyurethane catheters. Overall, with the exception of one batch culture of S. epidermidis strain SE3 tested, attachment to plasma-treated hydrogel-coated catheters was statistically significantly lower, by up to 85%, than

attachment

to plasma-treated

uncoated

catheters

for both

S. epidermidis and S. aureus.

Keywords: Staphylococcus epidermidis; Staphylococcus aureus; Plasma; Catheter; Hydrogel coating; Adhesion

1. Introduction

bacterial

colonisation

Central venous catheter (CVC) related sepsis is a significant cause of morbidity and mortality in hospitals and occurs in up to 15% of patients with a CVC [l]. The most frequent colonisers of CVCs are Staphylococcus epidermidis and S. aureus [2]. Attempts are being made to modify the polymers used in the manufacture of CVCs, in order to reduce * Corresponding author. Tel.: +44 (0161) 275 5265; Fax: +44 (0161) 275 5656; E-mail: [email protected] 0378-1097 /96/ $12.00 Copyright PIISO378-1097(96)00352-7

0 1996 Federation

of European

and

so increase

the longevity

[3]. One such modification is the coating of polyurethane catheters with a hydrophilic hydrogel (polyvinylpyrollidine). The hydrogel adsorbs water on contact with the blood and produces an extremely smooth surface [4]. As well as improved haemocompatibility [4], the hydrogel significantly reduces bacterial adhesion in in vitro assays [S-7]. On insertion into a deep vein, a catheter becomes coated with a variety of blood proteins [8]. It has been suggested that individual proteins or their comof the device

Microbiological

in the patient

Societies. Published

by Elsevier Science B.V.

cific activity of 49 Ci mmol~~‘. Each radiolabelled, separately inoculated, broth culture represented one batch of cells.

bined effects may influence bacterial adhesion to catheters [8]. This study investigates whether the reduction in adhesion of S. epidermidis and S. uureus, due to the hydrogel, is maintained when the catheters are coated with plasma.

2.2. Prepurution

Hydrogel-coated and uncoated polyurethane catheters were cut into 1 cm lengths and were either incubated in human plasma (donated by the Manchester blood bank) or buffer for 1 h at 37°C. The buffer used throughout this study consisted of Tris (0.05 M), containing NaCl (0.145 M), at pH 7.5. Plasma-treated catheter pieces were washed 3 times in buffer, to remove excess protein. ‘Catheter assemblies’ were constructed from either uncoated polyurethane or hydrogel-coated polyurethane catheters. These consisted of 8 pieces of catheter held firmly together at the ends by 2 plastic holders, with projections fitting exactly into the lumen of each catheter piece. For each assembly of 8 uncoated or 8 coated catheters, 4 pieces were pretreated with plasma and 4 pretreated with buffer only. In this way, adhesion to plasma-treated catheters could be compared

2. Materials and methods 2.1. Bacterial

struins

Three strains of S. epidermidis were used: NCTC 11047, F337, a clinical isolate from a CVC. donated by Dr. R. Bayston (Nottingham City Hospital) and SE3, a clinical isolate from a CVC, donated by M. Howard (Special Care Baby Unit, Manchester Royal Infirmary). Three clinical isolates of S. aureu~ were also used: SA14, SA6 and SA3. These were obtained from CVC tips tested at the Withington Hospital. Manchester. For the adhesion assay, strains were grown in tryptone soya broth (TSB) for 20 h at 37°C. in the presence of 8 pCi ml ’ tritiated methylthymidine (Amersham, Bucks, UK), with spe-

Table I Effect of plasma on adhesion Strain

of S. epidermidis and S. uuwus to hydrogel-coated

Adhesion Bottle

to hydrogel-coated

catheters

I

r$’ cutheters

catheters CPlC x 100%

(‘%I)

Bottle

Bottle 2

C

CP

C

CP

S. vpidermidis NCTC 11047 A NCTC 11047 B SE3 A SE3 B F337 A F337 B

0. I? * 0.04’ 0.10~0.03 0.3 I + 0.07 0.22 + 0.05 0.04 + 0.01 0.23 k 0.06

0. I I i 0.04 0.09 + 0.02 0. I7 + 0.07 0. I8 2 0.05 0.04 + 0.01 0.21 k 0.08

0.13 -t 0.04 0.10i0.01 0.17~0.01 0.14 f 0.02 0.04 f 0.01 0.19 k 0.05

0.13~O.OY 0.14+0.06 0.10f0.04 0.12+0.06 0.06 + 0.03 0.22 + 0.06

s. DUleUS SA3 A SA3 B SA14 A SA14 B SA6 A SA6 B

0.25 + 0.04 0.21 f 0.07 0.05 * 0.02 0.08 + 0.02 0.12*0.04 0.1010.01

1.12f0.37 0.66f0.17 0.27+0.11 0.42+0.13 0.85 + 0.05 0.61 kO.07

0. I5 k 0.03 0.09 + 0.02 0.07 f 0.02 0.09 + 0.06 0.05 i 0.00 0. I I i 0.03

0.94 i 0.21 0.93 i 0.26 0.25 ?I 0.09 0.56t0.09 0.38 + 0.04 0.48 f 0.03

C, hydrogel-coated catheters; catheters (CP) expressed as a two separate adhesion assays; “Standard deviation. hP values, calculated from the

I

Bottle 2

~~~ X9 (N/S)” 89 (N/S) 55 (N/S) 80 (N/S) 93 (N/S) 89 (N/S)

448 3 I8 519 501 723 585

(0.05) (0.05) (0.05) (0.05) (0.01) (0.01)

101 136 61 83 149 I I8

(N/S) (N/S) (N/S) (N/S) (N/S) (N/S)

614 978 370 613 794 443

(0.01) (0.01) (0.05) (0.01) (0.01) (0.01)

CP, plasma-treated, hydrogel-coated catheters; CPK x IOO’X~,adhesion to plasma-treated, hydrogel-coated percentage of adhesion to hydrogel coated catheters (C); A and B. two separate batches of cells tested in bottles I and 2. two separate universal bottles containing 4 XC and 4 X CP pieces of catheter. comparison

of CP and C values, using analysis of variance;

N/S, not significantly

different.

S.F. John et al. IFEMS

in the same bottle pieces.

with the control,

Microbiology

buffer-treated

2.3. Radiolabelling adhesion assay The radiolabelling adhesion assay method has been described previously [5]. Radiolabelled cells of S. epidermidis or S. aureus were washed three times in buffer and resuspended to an ODddo of 0.393. Resuspended cells (4 ml) were pipetted into 4 acidOwashed, glass universal bottles. Three 50-f.tl samples of cell suspension were removed for scintillation counting. For each strain, a catheter assembly was placed into each of the 4 bottles so that two bottles contained uncoated, polyurethane catheters (buffer and plasma-treated) and two bottles contained hydrogel-coated, polyurethane catheters (buffer and plasma-treated). Bottles were incubated for 2 h at 37°C with agitation. Catheter assemblies were removed from the glass bottles, washed three times in distilled water and allowed to dry. Each catheter piece was placed into a separate vial and the radioactivity (dpm) determined for 3 min in a scintillation

Table 2 Effect of plasma on adhesion Strain

of S. epidermidis and S. aureus to uncoated

Adhesion Bottle

to uncoated

catheters

I

U

Letters 144 (1996)

241-247

243

counter (2000 CA Tri Carb Packard, United Technologies, Berks, UK). The percentage adhesion for each strain per catheter piece was calculated by dividing the radioactivity from each catheter piece by the initial radioactivity per ml of culture. The mean of these values (n = 4) was calculated. Two batch cultures were tested for each strain. All data was statistically tested using analysis of variance, and significance was measured to P < 0.05 and P < 0.01 ievels.

3. Results

Three S. epidermidis and three S. aureus strains were tested for their adhesion to the standard lengths of polyurethane catheters. Half of the catheters had a hydrogel coating and half of the catheter lengths of each type had a pretreatment with plasma. Two batch cultures (A and B) of each strain were tested and two bottles (see Section 2) were set up for each batch. The effect of plasma pretreatment on the ability of the staphylococci to adhere to uncoated and

catheters

(%)

UP/U x

100%

Bottle 1

Bottle 2 UP

U

UP

0.18kO.03 0.21 * 0.04 0.13kO.02 0.77 ?r 0.12 0.2 1 f 0.03 0.28 + 0.03

0.91 fO.10 0.53~0.17 0.46 + 0.03 0.98 +O.lO 0.66f0.15 1.23f0.12

0.24 + 0.04 0.36? 0.13 0.17f0.03 0.68 ? 0.16 0.37 k 0.02 0.61 5 0.23

21 47 34 82 43 4s

6.13 k 1.80 3.20 f 0.40 1.08 + 0.05 0.78 f 0.08 1.28f0.16 0.98+0.17

8.33 f 2.40 1.59 2 0.37 0.74f0.10 0.84f0.12 0.35 + 0.11 0.61 f. 0.22

7.74k0.36 2.68 kO.35 0.96kO.11 0.82kO.14 1.36kO.18 1.14+0.09

89 80 140 98 321 465

Bottle 2

S. epidermidis NCTC 11047 A NCTC 11047 B SE3 A SE3 B F337 A F337 B

0.86 0.44 0.38 0.95 0.49 0.63

t 0.06” + 0.06 + 0.04 I!I0.06 5~0.08 + 0.16

(O.Ol)h (0.01) (0.01) (N/S) (0.01) (0.05)

26 68 36 69 56 49

(0.01) (N/S) (0.01) (0.05) (0.05) (0.05)

93 168 130 98 391 186

(N/S) (0.05) (N/S) (N/S) (0.01) (0.05)

S. aureus SA3 A SA3 B SA14 A SA14 B SA6 A SA6 B

6.92 L!Z 0.91 4.01 + 0.30 0.77 + 0.12 0.80 t 0.07 0.40? 0.08 0.21 t 0.03

(N/S) (0.05) (0.05) (N/S) (0.01) (0.01)

U, polyurethane (uncoated) catheters; UP, plasma-treated, uncoated catheters; UP/U X lOO%, adhesion to plasma-treated, uncoated catheters (UP) expressed as a percentage of adhesion to uncoated catheters (U); A and B, two separate batches of cells tested in two separate adhesion assays; bottles 1 and 2, two separate universal bottles containing 4 X U and 4 X UP pieces of catheter. “Standard deviation. ‘p values. calculated from the comparison of UP and U values, using analysis of variance; N/S, not significantly different.

S.b John rf ul. I FEMS Microbiology Letters 144 i I YY6j 241-247

244

coated catheters was analysed statistically. This analysis was performed on the mean values for adhesion for each batch culture (n = 8). The data for the six strains are shown in Tables I-3. The results for the adhesion for S. epidermidis NCTC 11047 batches A and B and for S. aUreUS SA6 batches A and B are illustrated in Figs. 14, respectively. In confirmation of an earlier report in which a full statistical analysis was presented [5], all six strains showed a significant reduction in adhesion to the hydrogel-coated catheters compared to that with the uncoated catheters (compare Table 1, columns C with Table 2, columns U, and Figs. 1 and 2, histograms C with U, and Figs. 3 and 4, histograms C with U). In the test of the effect of plasma pretreatment on S. epidermidis adhesion to hydrogel-coated catheters. no statistically significant difference in adhesion was observed between plasma-treated and untreated catheters for any of the batches tested (compare Table 1, columns CP with columns C, summarised in the last column, and Figs. 1 and 2, histograms CP with C). In sharp contrast, the adhesion of S. aureus strains was increased by approx. 3-9-fold by the plasma pretreatment (compare Table 1, columns CP with C, summarised in the last column, and Figs. 3 and 4, histograms CP with C). The effect of the plasma pretreatment of uncoated catheters on S. epidermidis adhesion was consistently

Table 3 Adhesion to plasma-treated, hydrogel-coated percentage of adhesion to plasma-treated,

catheters (CP) as a uncoated catheters

(UP) Strain

CPKJP x

1OO’%

_ S. epidermidi.~ NCTC 11047 SE3 F337

Batch A

Batch B

57 (0.05), 93 (N/S) 17 (0.01)

40 (0.01) 21 (0.01) 49 (0.05)

s. uureuS SA3 SAl4 SA6

15 (0.01) 25 (0.01) 46 (0.01)

27 (0.01) 61 (0.01) 52 (0.01)

Batches A and B, two separate batches of cells tested in two separate adhesion assays. Values calculated from data from two bottles. ‘P values, calculated from the comparison of CP and UP values. using analysis of variance: N/S, not significantly different.

1.6

1.4

1.2

G

1.0

g .z 0)

0.8

e a

0.6

0.4

0.2

0

II a 1

T

c

CP

1

r

c CP

U’UP

U JP

hydrogel

hydrogel

uncoated

uncoated

Bottle

Bottle

Bottle

Bottle

1

2

Catheter

1

2

type

Fig. I. Influence of plasma (P) on adhesion of S. epidermidis NCTC 11047 batch A to hydrogel-coated (C) and plain polyurethane (U) catheters. Each bar represents the mean of 4 catheter pieces tested for adhesion in two separate bottles. The error bar represents the standard deviation.

to reduce it (compare Table 2, columns UP with U, summarised in the last column, and Figs. 1 and 2, histograms UP with U). For S. aureus, adhesion was either unaffected or increased by the plasma pretreatment of uncoated catheters (compare Table 2, columns UP with columns U, summarised in the last column, and Figs. 3 and 4, histograms UP with U). Statistically significant variation in adhesion values between batches and bottles was seen for the S. epidermidis and S. aureus strains and is shown in the representative histograms for S. epidermidis NCTC 11047 (Figs. 1 and 2) and S. aweus SA6 (Figs. 3 and 4) and in Tables 1 and 2. However, this variation was taken into account by the experimental design, where adhesion to catheters with and without plasma was compared in the same bottle and analysis of variance was used to statistically test the results. For all strains of S. epidermidis and S. aureus tested, with the exception of one batch of S. epider-

S.F. John et al. IFEMS

Microbiology Letters 144 (1996) 241-247

245

midis SE3, there was a significant

reduction in adhesion to the plasma-treated, hydrogel-coated catheters, compared with the plasma-treated, uncoated catheters (Table 3). Reduction in adhesion of S. epidermidis strains ranged from 43% to 83% and 39% to 85% for the S. aureus strains.

4. Discussion The adhesion of S. epidermidis to uncoated polyurethane catheters was significantly reduced by plasma coating, an effect which has previously been demonstrated for a range of polymers, including fluorinated polyethylene-propylene (FEP) [9], polyethylene [lo], silicone [11,12], pellethane and polyethylene [13]. An identical effect was observed when serum was used to coat catheters made of silicone, polyurethane and hydrogel-coated polyur-

1.6 -

.g t

0.8

G (

6 a

0.6

0.4

0.2

0 C’CP

C’CP

U’UP

U’UP

hydragel hydrogel

uncoated

uncoated

Bottle

Bottle

Bottle

1

Bottle

2

1

2

type

Fig. 3. Influence of plasma (P) on adhesion of S. aureus SA6 batch A to hydrogel-coated (C) and plain polyurethane (U) catheters. Each bar represents the mean of 4 catheter pieces tested for adhesion in two separate bottles. The error bar represents the standard deviation.

1.2 -

l.O-

0.8

Catheter

1.4 -

E

E .E 9)

0.6 I

Ii u

U’UP

UP

hydrogel

hydrogel

uncoated

uncoated

Bottle

Bottle

Bottle

Bottle 2

1

2

Catheter

1

type

Fig. 2. Influence of plasma (P) on adhesion of S. epidermidis NCTC 11047 batch B to hydrogel-coated (C) and plain polyurethane (U) catheters. Each bar represents the mean of 4 catheter pieces tested for adhesion in two separate bottles. The error bar represents the standard deviation.

ethane [ 141, Teflon@ [15] and polymethylmethacrylate (PMMA) [16]. Therefore, for all polymers tested to date, plasma and serum coating consistently reduced adhesion of S. epidermidis strains. This reduction could be due to a decrease in the biomaterial surface hydrophobicity after plasma coating [9], which Pascual et al. [15] have attributed to the presence of the hydrophilic protein albumin. In addition, electrostatic repulsion between adsorbed proteins and bacterial cells could contribute to the reduction in adhesion [17]. In contrast to S. epidermidis, the effects of plasma on the adhesion of S. aureus to catheters are not consistent as different studies have shown that adhesion may be increased [18] or decreased [11,16] by plasma. This study has shown either no effect or an increase in adhesion of S. aureus strains to plasmatreated catheters and that the increase to hydrogelcoated, polyurethane catheters is statistically significant.

S.F. John et ~1, IFEMS

246

Microhiolo,qv

1.6

1.4

1.2

0 hydrogel

hydrogel

uncoated

uncoated

Bottle

Bottle

2

Bottle

Bottle

Catheter

type

1

1

2

Fig. 4. Influence of plasma (P) on adhesion of S MWII.S SAh batch B to hydrogel-coated (C) and plain polyurethane (U) catheters. Each bar represents the mean of 4 catheter pieces tested for adhesion in two separate bottles. The error bar represents the standard deviation.

S. aureus strains carry receptors for a number of blood proteins, such as fibrinogen or fibrin [I 81 and fibronectin [16]. Therefore, it is likely that S. aureus adhesion to plasma-treated catheters is mediated by these specific receptors, producing the increase in adhesion clearly exhibited on plasma-treated, hydrogel-coated surfaces in this study. It is not clear why blood proteins would reduce S. aureus adhesion [11,16] but it could be a strain specific phenomenon. We have previously demonstrated that the hydrogel itself causes a significant decrease in S. epidermidis and S. aureus adhesion compared to the uncoated polyurethane [5]. Negatively charged, hydrophilic surfaces discourage bacterial adhesion [ 191 and so the hydrogel may prevent adhesion by electrostatic repulsion. The statistically significant variation noted between bottles and between different batches of both S. epidermidis and S. aureus has already been documented [5] although the reasons for this variation are

Let/em

144 IIYY6I 241-247

not understood. The experimental design and statistical analysis, previously used to take this variation into account was adapted for this study to evaluate adhesion to plasma and non-plasma-treated catheter pieces in the same bottle [5]. The variation in adhesion between culture batches was not due to cell clumping. Washed and resuspended culture batches were all checked for clumping by visual assessment with a light microscope. Cells were generally found in singles or doubles, with only occasional triplets and quadruplets. When the effect of plasma pretreatment on the adhesion of both staphylococcal species to uncoated and hydrogel-coated catheters was compared, there was an overall decrease in adhesion to the hydrogelcoated catheters. S. epidermidis strains showed variation in the statistical significance of this decrease. However, the difference observed for S. aureus, a pathogen with specific adhesion mechanisms, was highly significant. This study shows that, with one exception, the hydrogel coating on the polyurethane catheters remains effective in reducing bacterial adhesion of both S. epidermidis and S. uureus, by up to 85% even when the catheters are coated with plasma. The exception was that for one batch of S. epidermidis SE3 cells tested no significant reduction was observed. However, a second batch of these cells did show significant reduction (Table 3). This overall reduction is in spite of the conspicuous difference in the ability of the two staphylococcal species to adhere to hydrogel-coated catheters in the presence of plasma; plasma causes an increase in S. aureus adhesion but no significant effect on S. epidermidis adhesion. We do not yet know whether these results have a clinical significance. Clinical trials are not yet available to prove unequivocally whether such a reduction in adhesion brought about by coating catheters with hydrogel would allow a longer period of catheterization before the possible onset of infection.

Acknowledgments

This work was supported by a BBSRC CASE studentship with industrial support from Ohmeda (Swindon, UK).

S.R John et al. IFEMS

Microbiology

References [l] Elliott, T.S.J. (1993) Line associated bacteraemias. CDR Review, Communicable Disease Report. 3, no. 7, R91lR97. [2] Elliott, T.S.J. (1988) Intravascular-device infections. J. Med. Microbial. 27, 161-167. [3] Bridgett, M.J., Davies, M.C. and Denyer, S.P. (1992) Control of staphylococcal adhesion to polystyrene surfaces by polymer surface modification with surfactants. Biomaterials 13, 41 l416. [4] Jansen, B., Jansen, S., Peters, G and Pulverer, G. (1992) Invitro efficacy of a central venous catheter (‘Hydrocath’) loaded with teicoplanin to prevent bacterial colonisation. J. Hosp. Infect. 22, 933107. [5] John, SF., Hillier, V.F., Derrick, M.R. and Handley, P.S. (1995) Adhesion of staphylococci to polyurethane and hydrogel-coated polyurethane catheters assayed by an improved radiolabelling technique. J. Med. Microbial. 43, 1333140. [6] Tebbs, S.E. and Elliott, T.S.J. (1994) Modification of central venous catheter polymers to prevent in vitro microbial colonisation. Eur. J. Clin. Microbial. Infect. Dis. 13, 111-117. [7] Denyer, S.P., Hanlon, G.W., Davies, M.C. and German, S.P. (1993) Antimicrobial and other methods for controlling microbial adhesion and infection. In: Microbial Biofilms: Formation and Control (Denyer, S.P., Gorman, S.P. and Sussman, M., Eds.), pp. 147-165, Blackwell, Oxford. [8] Goldmann, D.A. and Pier, G.B. (1993) Pathogenesis of infections related to intravascular catheterisation. Clin. Microbial. Rev. 6, 176192. [9] Hogt, A.H., Dankert, J. and Feijen, J. (1985) Adhesion of Staphylococcus epidermidis and Staphylococcus saprophyticus to a hydrophobic biomaterial. J. Gen. Microbial. 131, 24852491. [lo] Wang, I., Andersen, J.M. and Marchant, R.E. (1993) Staphy-

Letters 144 (1996)

241-247

247

lococcus epidermidis adhesion to hydrophobic biomedical polymer is mediated by platelets. J. Infect. Dis. 167, 329-336. B., Thamdrup [ill Espersen, F., Wilkinson, B.J., Gahm-Hansen, Rosdahl, V. and Clemmensen, I. (1990) Attachment of staphy lococci to silicone catheters in vitro. APMIS 98, 471478. v21 Muller, E., Takeda, S., Goldmann, D.A. and Pier, G.B. (1991) Blood proteins do not promote adherence of coagulase-negative staphylococci to biomaterials. Infect. Immun. 59, 33233326. Stollenwerk, M., [I31 Paulsson, M., Kober, M., Freij-Larsson, Wesslen, B. and Ljungh, A. (1993) Adhesion of staphylococci to chemically modified and native polymers, and the influence of preadsorbed fibronectin, vitronectin and fibrinogen. Biomaterials 14, 8455853. v41 Kristinsson, K.G. (1989) Adherence of staphylococci to intravascular catheters. J. Med. Microbial. 28, 2499257. u51 Pascual, A., Fleer, A., Westerdaal, N.A.C. and Verhoef, J. (1986) Modulation of adherence of coagulase-negative-staphylococci to teflon catheters in vitro. Eur. J. Clin. Microbial. 5, 518-522. R., V61 Herrmann, M., Vaudaux, P.E., Pittet, D., Auckenthaler, Lew, P.D., Schumacher-Perdreau, F., Peters, G. and Waldvogel, F.A. (1988) Fibronectin, fibrinogen and laminin act as mediators of adherence of clinical staphylococcal isolates to foreign material. J. Infect. Dis. 158, 6933701. u71 Reynolds, E.C. and Wong A. (1983) Effect of adsorbed protein on hydroxyapetite zeta potential and Streptococcus mutans adherence. Infect. Immun. 39, 128551290. [181 Cheung, A.L. and Fischetti, V.A. (1990) The role of fibrinogen in staphylococcal adherence to catheters in vitro. J. Infect, Dis. 161, 117771186. u91 Fletcher, M. and Loeb, G.I. (1979) Influence of substratum characteristics on the attachment of a marine pseudomonad to solid surfaces. Appl. Environ. Microbial. 37. 67-72.