A Novel Antibiofilm Technologyfor Contact Lens Solutions

Bruce F. Farber, MD, Hsi-Chia Hsieh, PhD, Eric D. Donnenfeld, MD,Henry D. Perry, MD, Arthur Epstein, OD, Arlene Wolff, BS

Purpose: Nonsteroidal anti-inflammatory drugs, including sodium salicylate, inhibitextracellular bacterial biofilm production . The authors studied the effect of the additionof sodium salicylate on bacterial adherence and biofilm formation on contact lenses andcases and commonly used medical polymers,

Methods: The study was done in vitro with bacterial adherence and biofilm measuredon lenses and cases that were exposed to saline contaminated with Staphylococcusepidermidis and Pseudomonas aeruginosa with and without 1 and 3 mm sodium salic­ylate. Bacterial adherence to contact lenses was quantitated by a vortex assay and byscanning electron microscopy. Biofilm formation on contact lens cases and other poly­mers was measured by an optical density assay and a radiolabeling assay .

Results: Inhibition of biofilm formation was demonstrated on plastic contact lenscases in a dose-related manner with 1 and 3 mm sodium salicylate. A dose-relateddecrease in bacterial adherence also was noted. Assays with contact lenses also dem­onstrated less adherence in the presence of sodium salicylate. Electron micrographs ofthe contact lens showed less biofilm, most noticeable with 3 mm salicylate. Other studiesdemonstrated decreased adherence of S. epidermidis to polyethylene and polystyrene.Sodium salicylate also decreased biofilm on plastic tissue culture wells, but sorbic acidparadoxically increased deposition.

Conclusion: The authors found that the addition of low-dose sodium salicylate tosaline decreased the adherence of S. epidermidis and P. aeruginosa to contact lensesand lens cases. Biofilm production also was decreased on the lens cases and on medicalpolymers used to make plastic cases. These studies suggest that sodium salicylatedeserves additional study to determine its use in contact lens solutions.Ophthalmology 1995;102:831-836

Infectious keratitis remains a serious complication ofcontact lens use. Poggio et all estimated an annual inci­dence of infectious keratitis of 20.9 per 10,000 patientswearing cosmetic extended-wear contact lenses and 4.1per 10,000 patients wearing daily wear cosmetic soft con­tact lenses. A conservative estimate based on approxi­mately 20 million Americans wearing only daily wear softcontact lenses would create an annual incidence of 8200

Originally received: lune 10, 1994.Revision accepted: November 23, 1994.

From the Departments of Medicine and Ophth almology, North ShoreUniversity Hospital-Cornell University Medical College, Manhasset.

Presented in part at the ARVO Annual Meeting, Sarasota , May 1994.Dr. Farber holds patent rights to th is techn ology.

Reprint requests to Bruce F. Farber , MD, North Shore Uni versity Hos­pital, 300 Community Dr, Manhasset , NY ) 1030.

cases of infectious keratitis. The majority of contact lens­related corneal ulcers are due to Pseudomonas aerugi­nosa.r" Many factors have been postulated to be impor­tant influences on the risk of infection. These factors in­clude the lens material, wearing schedule, disinfectiontechniques, adherence of bacteria to the lens, and con­tamination of the lens care system and solutions. l"

Contact lens deposits or spoilage creates an undesirableinterface between the contact lens and the conjunctivaand corneal surface. Contact lens deposit symptomatologyincludes decreased visual acuity , foreign body sensation,mucous discharge, tearing , redness, photophobia, and ab­normal lens fit.8.

9 Severe deposits may result in decreasedoxygen permeability of the contact lens with the formationof superficial and deep cornea) neovascularization, in­trastromallipid, and intracorneal hemorrhage. 10-14 Proteindeposits play an important role in the development of

831

Ophthalmology Volume 102, Number 5, May 1995

contact lens-related allergy, including giant papillary con­junctivitis. Contact lens deposits harbor infectious organ­isms and increase the risk of infectious keratitis." Thesedeposits include tear film proteins, sebaceous secretions,exogenous materials, mineral deposits, and bacterial bio­film."

Recent data have demonstrated that the attachment ofmicroorganisms to medical devices is mediated by biofilm,or capsular polysaccharides which form a protective ma­trix around a device. Biofilm produces a scaffold whichfacilitates bacterial adhesion and colonization.P In ad­dition, biofilm protects and insulates bacteria from anti­infectives and disinfectant systems. Biofilm formation hasbeen demonstrated on both contaminated lenses and inlens cases, even after appropriate sterilization.v" Thesebacteria likely serve as an inoculum for contact lens-re­lated bacterial keratitis. Several authors have documentedculturing organisms from the contact lens and contactlens case that were identical to the patients' infectiouskeratitis.t-'

Nonsteroidal anti-inflammatory drugs, including so­dium salicylate, inhibit extracellular bacterial biofilmproduction.P'?" These compounds inhibit bacterial at­tachment to a number of medical polymers, includingsilastic and polyurethane catheters. We investigated theaddition oflow-dose sodium salicylate to contact lens so­lutions. We attempted to determine whether the additionofsodium salicylate to contact lens solutions can decreasebiofilm formation on contact lenses and their cases andbacterial attachment to contact lenses. We chose to studyP. aeruginosa because it is the most common cause ofcontact lens-related infectious keratitis and Staphylococcusepidermidis because it is the organism most commonlycultured from contact lenses.i"

Methods

Strain

S epidermidis strain 91 was used in these experiments.This is a slime-positive strain that has been described pre­viously.!" A clinical isolate ofP. aeruginosa also was used.

Lens Cases

Studies on biofilm formation were performed in plasticcontact lens cases (Allergan, Irvine, CA). An overnightculture of the test organism (108 colony-forming units[CFU]/ml) in trypticase soy broth (TSB) was added tosaline with and without sodium salicylate in a 1:1 dilution.The bacterial suspension was incubated for 5 hours at 37°C, and the suspension was removed. In the experimentswith P. aeruginosa, nonadherent film floating on the cul­ture was removed with a cotton swab. The remaining bio­film was washed gently two times with 5 to 10 ml salineto remove nonadherent bacteria and debris. The adherentfilm was removed mechanically with a pipette and dis­solved in 3 ml saline. The optical density was read on aBausch and Lomb spectronic 601 at 686 nm (Rochester,

832

NY). In addition, bacterial counts were done on the sam­ples by serial dilution and plating onto Mueller-Hintonagar plates, which were incubated at 37° C for 18 hours.

Low-density Polyethylene

Low-density polyethylene was used as a primary referencematerial. It had been prepared under controlled conditionsand has been used as a standard in experimental studies.'The material was obtained from Abiomed, Inc (Danvers,MA), and has been described in detail. The sheets werecut into 10 X 26-mm rectangular sections. A bacterialsuspension similar to that described previously was used.The polyethylene sheets were placed in the suspensionswith and without sodium salicylate for 24 hours at 37°C. They were removed, and the white biofilm was me­chanically removed with a no. 1014 brush and placed in5 ml saline. The optical density then was read as previouslydescribed.

Polystyrene

Bacterial suspensions were prepared as previously notedand incubated with and without sodium salicylate at 3rC for 24 hours in 50 ml clear polystyrene tubes (Corning,NY). The supernatant was removed, and the white filmon the walls of the tube was inspected. The biofilm thenwas dissolved in 5 ml saline by mechanically removing itwith a plastic transfer pipette. The preparation was readat an optical density of 686 JLm. In addition, similar ex­periments were performed, but 14C-sodium acetate (4 JLCi)(1_14C) was added to all tubes before incubation. Thesupernatant was removed from the tubes, and 2 ml Aqua­sol was added. The biofilm was removed mechanicallyfrom the sides of the tube, resuspended in Aquasol andwas counted in scintillation vials.

Contact Lenses

An overnight TSB culture was standardized to one-halfoptical density at 686 JLm and diluted 1:100 in broth. Thebacterial suspension then was added in a 1:1 dilution tosaline with and without sodium salicylate. Acuvue (John­son and Johnson, Raritan, NJ) contact lenses were in­cubated in the contaminated saline for 20 hours at 3rc. The lenses were removed and placed in 5 ml salinewith gentle washing. They then were removed and placedin 5 ml fresh saline, vortexed at number 7 for 1 minutefive times. The optical density of the saline was measuredat 686 JLm, and colony counts were performed by serialdilution. Lenses also were incubated with P. aeruginosa(107 CFU/ml) for 2 hours in saline alone or supplementedwith sodium salicylate. The lenses were removed andwashed in saline three times. They were incubated with3H-leucine (2 JLCi/ml) for 30 minutes and then washedand counted with Aquasol in scintillation vials. The resultsare expressed in counts per minute.

The effect of the addition of sodium salicylate to H 20 2

during the disinfection of contact lenses was studied.Lenses were placed in lens cases with H20 2 supplemented

Farber et al . Antibiofilm Technology for Contact Lens

Figure 1. Electron micrographs of contact lenses that were incubated in saline and 3 mm sodium salicylate in the presence of Staphylococcus epidermidis.The amount of biofilm seen on the 3-mm lens was less than that seen in the control lens (original magnification, X2000).

with 1 and 3 mmolj(mm) salicylic acid. The HzOz wasneutralized with catalase. The lens were rinsed in salineand incubated in P. aeruginosa for 2 hours. They thenwere washed in saline three times, and incubated in ra­diolabeled leucine (2 l1Cijml) for 30 minutes at 20° C.They were washed three times, suspended in Aquasol,and counted in scintillation vials.

Electron Microscopy

Lens were incubated in saline or saline with 3 mm sodiumsalicylate (Fig 1). They then were contaminated with anS. epidermidis bacterial suspension as previously describedfor 4 hours. The soft contact lenses were fixed in 2% buf­fered glutaraldehyde. A rapid dehydration was performedwith graded alcohols. They then were processed by criticalpoint drying. A palladium gold carbonation was used forplating by sputter coating, and the lenses were viewedwith a scanning Jeol electron microscope (Mexford, MA).

Keuls comparison test was used to detect differences be­tween the salicylate and sorbic acid group.

Results

Inhibition ofS. epidermidis biofilm was demonstrated onthe sides of plastic contact lens cases with 1 and 3 mmsodium salicylate as seen in Figure 2. There was 39% in­hibition with 1 mm and 95% inhibition with 3 mm. Thedifference was significant with P < 0.0001. A dose-relateddecrease in corresponding bacterial counts also was ob­served. The control lens cases contained 1.2 X 107 CFUjml in the dissolved biofilm. There were 9.6 X 106 CFUjml in 1 mm and 2.4 X 106 CFU/ml in 3 mm sodiumsalicylate. Results of a similar assay performed with P.aeruginosa are demonstrated in Figure 3. A dose-relateddecrease in adherent biofilm was noted. A corresponding

<,r-,

-,-,

"~

Studies Using Salicylate and Sorbic Acid

Contact lenses were incubated overnight in 24 well platescontaining S. epidermidis in TSB diluted 1:1 with sterilesaline with or without sodium salicylate (3 mm) or sorbicacid (0.015%). They then were rinsed gently with salineand transferred to a I5-ml centrifuge tube and vortexedwith 4 ml NaCl vigorously fivetimes for 1 minute. Colony­forming units then were done. In addition, overnight cul­tures were placed in the wells without lens. The mediawere removed, and the remaining biofilm was scrapedout of the wells and suspended in 1 ml saline. Opticaldensities were read at 686 11m.

0.12

0.1

0.08

Optical 0.06

density0.04

0.02

0o Imm 3mm

Statistical Analysis

Simple linear regression analysis was used to detect dose­related differences between the control, the 1-, and 3-mmgroups. One way analysis of variance with a Newman-

Concentrationsalicylate (mm).

Figure 2. Biofilm formation measured in optical density versus concen­tration of salicylate (millimolar) with Staphylococcus epidermidis.

833

Ophthalmology Volume 102, Number 5, May 1995

95,53563,23546,784

Counts/min(%)

(100)(89)(47)

0.6180.5540.296

Optical Density"(um)

Table 2. Inhibition of Biofilm to Polystyrene UsingRadiolabeled Staphylococcus epidermidis

• Mean of three assays.

Controls1mm3mm

"" '-.0" " .....

!lA<,

........~.oa

.0"

01

Il

o

007

o

o.

Optical 0

Density 0

o

Table 1. Inhibition of Biofilm on Polyethylene

Figure 3. Biofilm formation measured in optical density versus concen­tration of salicylate (millimolar) with Pseudomonas aeruginosa.

Concentrationsalicylate (mm)

decrease in the number of CFU/ml bacteria also was ob­served.

The results of the studies using S. epidermidis withstandardized polyethylene are presented in Table I. Thestudy was performed in triplicate. There was minimal in­hibition noted at I mm and 21% at 3 mm (P < 0.04).Studies using polystyrene are presented in Table 2. Therewas a dose-related linear inhibition ofbiofilm as measuredby both optical density (P < 0.00 I) and uptake of radio­labeled sodium acetate. Table 3 gives the results of thestudies using contact lenses and S. epidermidis. In thesestudies, the number ofbacteria adhering to contact lensesincubated in contaminated saline was quantitativelymeasured in two ways. Both the optical density and num­ber of colony-forming units per milliliter bacteria fell ina dose-related manner with increasing concentrations ofsodium salicylate. Although there was a dose-related de­crease in adherence, it did not achieve statistical signifi­cance.

Table 4 lists the results ofan assay in which adherenceto lenses was studied with P. aeruginosa. Both assaysdemonstrate a dose-related inhibition of growth (colony­forming units per milliliter); however, it was not statisti-

Despite adequate disinfection, corneal infection continuesto be a problem associated with soft contact lens use. Thepathophysiology of these infections is complex and prob­ably multifactorial. It seems logical that contaminationof lens care cases and solutions is a major risk factor ofinfection. In addition, recent data from studies of othermedical devices and lenses suggest that adherence ofbac­teria is significantly dependent on biofilm formation. Mi­croorganisms produce a protective polysaccharide filmthat is resistant to disinfectants and antimicrobial agents. 17

Our previous work has demonstrated that the salicy­lates and other nonsteroidal anti-inflammatory agents in­terfere with bacterial attachment to a number of medicalpolymers. 17

, 19,20 In addition, we have demonstrated that

cally significant. The radiolabeling method produced asimilar degree of inhibition. In Table 5, adherence of P.aeruginosa to lenses was studied after exposure in con­taminated neutralized H202 . A dose-related decrease inadherence again was noted (P < 0.003).

Optical densities ofthe biofilm scraped out of the plasticwells were 0.788 nm for controls, 0.194 Jl-m for salicylate,and 1.435 Jl-m for sorbic acid (mean trials, 7). There wasa highly significant difference among the groups (P <0.00 I). Of note was the paradoxical increase in biofilmfound with sorbic acid. Colony counts done on lenses thatwere incubated demonstrated 6 X 105 CFU/rnl in thecontrol saline, 1.3 X 104 CFU/ml with salicylate, and 2.5X 105 CFU/ml with sorbic acid. These differences werenot significant.

Discussion

3mmlmmo

Controls

1mm

3mm

Optical Density(~m)

0.1320,1670,1830.1660.1530.1460.1120.1350.136

Mean

0.160

0.155

0,127

(%) Table 3. Adherence of Staphylococcus epidermidisto Contact Lenses

(100)Optical Density (%) CFU/ml

Controls 0.044 (100) 5.2 X 107

(96) 1mm 0.026 (59) 4.4 X 107

3mm 0.008 (18) 2.8 X 107

6mm 0.004 (9) 1.2 X 107

(79)CFU = colony-forming units.

834

Farber et al . Antibiofilm Technology for Contact Lens

Table 4. Adherence of Pseudomonas aeruginosa toContact Lenses

the salicylates decrease the amount of S. epidermidis ex­tracellular slime. 16 Although experience with biofilm hasbeen most extensive with P. aeruginosa, the observed ef­fect has been demonstrated with a variety of other organ­isms.

In this study, we demonstrated that the addition oflow-dose sodium salicylate to saline decreases biofilm for­mation on contact lens cases. In addition, there was re­duced bacterial adherence to lenses exposed to saline con­taminated with P. aeruginosa and S. epidermidis com­pared with control saline. The concentrations of sodiumsalicylate used were 0.013% to 0.039%.The concentrationof other nonsteroidal anti-inflammatory drugs used incommercially available topical ophthalmic drops is 0.1%to 1.0% or approximately 2.5 to 25 times greater than thehighest concentration ofsodium salicylatewe used. Hence,we are using concentrations that would be otherwise con­sidered subtherapeutic. It is unclear whether other non­steroidal anti-inflammatory drugs have this property.

Sorbic acid currently is used as a preservative in contactlens solutions at a concentration of 0.1%, which is higherthan the concentrations that we used. Our studies suggestthat it paradoxically may increase biofilm deposition onlenses, opposite of the effect with sodium salicylate.

Previous research has shown that P. aeruginosa pref­erably adheres to focal deposits on contact lenses." Thestrong association of Pseudomonas to contact lens-relatedbacterial keratitis may be explained by the production oflarge amount of extracellular slime. This may allow Pseu­domonas to adhere to the surface of the contact lens. Itis unclear whether the addition of a nonsteroidal anti­inflammatory drug to the lens solutions would decreasethe rate of corneal infection. However, by decreasing bac­terial adhesion to the contact lens, we postulate a de­creased risk of inoculation into the cornea. This hopefullywould decrease the incidence of bacterial keratitis. It isalso possible that decreasing biofilm might decrease theincidence oflens intolerance due to allergic problems. Wesuggest that further studies on biofilm formation andnonsteroidal anti-inflammatory drugs are indicated, aswell as toxicity studies of the chronic effect of low dosenonsteroidal anti-inflammatories on the ocular surface.

Vortex Method

Controls1 mm3mm

Radiolabel Method

ControlSalicylate (1 mm)

CFU = colony-forming units.

Triall

CFU/ml

4.0 X 106

1.4 X 1068.5 X 105

Counts/min

209,434119,046

Trial 2

CFU/ml

1.8 X 106

1.2 X 1068.4 X lOS

Table 5. Adherence of Pseudomonas aeruginosa toContact Lenses after HzOz Disinfection

Counts/min Final pH

HzOz 490,524 7.0HzOz + 1 mm 398, 149 6.8HzOz + 3 mm 215,831 6.4

References

I . Poggio EC, Glynn RJ, Schein 00, et al. The incidenceof ulcerative keratitis among users of daily-wear and ex­tended-wear soft contact lenses. N Engl J Med 1989;321:779-83.

2. Alfonso E, Mandelbaum S, Fox MJ, Forster RK. Ulcerativekeratitis associated with contact lens wear. Am J Ophthalmol1986;10I :429-33.

3. Donnenfeld ED, Cohen EJ, Arentsen JJ , et al. Changingtrends in contact lens associated corneal ulcers: an overviewof 116 cases. CLAO J 1986;12:145-9.

4. Wilhelmus KR. Review ofclinical experience with microbialkeratitis associated with contact lenses. CLAO J 1987;13:211-14.

5. Duran JA, Refojo MF, Gipson IK, Kenyon KR. Pseudo­monas attachment to new hydrogel contact lenses. ArchOphthalmoI1987;105:106-9.

6. Donzis PB, Mondino BJ, Weissman BA, Bruckner DA. Mi­crobial contamination of contact lens care systems. Am JOphthalmol 1987; I04:325-33.

7. Slusher MM, Myrvik QN , Lewis JC, Gristina AG. Extended­wear lenses, biofilm , and bacterial adhesion. Arch Ophthal­mol 1987;105:110-15.

8. Nesburn AB. Symposium: contact lenses. Complicationsassociated with therapeutic soft contact lenses. Ophthal­mology 1979;86:1130-7.

9. Tripathi RC, Tripathi BJ, Millard CB. Physiochemicalchanges in contact lenses and their interactions with thecornea and tears: a review and personal observations. CLAOJ 1988; 14:23-32.

10. Donnenfeld ED, Ingraham H, Perry HD, et al. Contact lens­related deep stromal intracorneal hemorrhage. Ophthal­mology 1991;98:1793-6.

II. Rozenman Y, Donnenfeld ED, Cohen EJ, et al. Contactlens-related deep stromal neovascularization. Am JOphthalmol 1989;107:27-32.

12. Bloomfield SE, Jakobiec FA, Theodore FH. Contact lensinduced keratopathy: a severe complication extending thespectrum of keratoconjunctivitis in contact lens wearers.Ophthalmology 1984;91:290-4.

13. Richard NR, Anderson JA, Tasevska ZG , Binder PS. Eval­uation of tear protein deposits on contact lenses from pa­tients with and without giant papillary conjunctivitis. CLAOJ 1992;18:143-7.

14. Allansmith MR, Korb DR , Greiner JV, et al. Giant papillaryconjunctivitis in contact lens wearers . Am J Ophthalmol1977;83:697-708.

15. Tojo M, Yamashita N, Goldmann DA, Pier GB. Isolationand characterization of a capsular polysaccharide adhesionfrom Staphylococcus epidermidis. J Infect Dis 1988;157:713-22.

835

Ophthalmology Volume 102, Nur.lber 5, May 1995

16. Teichberg S, Farber BF, Wolff AG, Roberts B. Salicylic aciddecreases extracellular biofilm production by Staphylococcusepidermidis: electron microscopic analysis [letter]. J InfectDis 1993;167:1501-3.

17. Farber BF, Wolff AG. The use of nonsteroidal antiinflam­matory drugs to prevent adherence of Staphylococcus epider­midis to medical polymers. J Infect Dis 1992;166:861-5.

18. Wang I, Anderson JM, Marchant RE. Staphylococcus epi-

836

dermidis adhesion to hydrophobic biomedical polymer ismediated by platelets. J Infect Dis 1993;167:329-36.

19. Farber BF, Wolff AG. The use of salicylic acid to preventthe adherence of Escherichia coli to silastic catheters. J Urol1993;149:667-70.

20. Farber BF, Wolff AG. Salicylic acid prevents the adherenceof bacteria and yeast to silastic catheters. J Biomed MaterRes 1993;27:599-602.