Eye (1991) 5, 70-74
Quantitative Alterations of the Commensal Eye Bacteria in Contact Lens Wear
D. F. P. LARKIN' and J. P. LEEMING2 Bristol
Summary A study was performed on the commensal external eye flora in 34 long-term contact lens wearers and a matched control group, in order to identify the effectof lens wear. Samples were taken from the lid margin and conjunctiva by semi-quantitative, and tear filmby quantitative methods. Cultured bacteria were identified with particular attention to coagulase-negative staphylococci. No qualitative alteration in the com mensal bacteria was found, although lens wearers were found to have a significantly higher number of species at all sites than controls. Significant quantitative changes were identified on the lid margin, with particularly high counts in some lens wearers, and tear film. Increased numbers of bacteria obtained from the conjunctiva were not statistically significant. Quantitative changes in the tear film are thought to be secon dary to changes at the lid margin, for which no explanation is apparent.
Several bacteriological investigations of evidence that whilst conjunctival bacterial normal human eyes have demonstrated that counts are suppressed in the first few months the flora of the conjunctiva is qualitatively of lens wear, they may rise to levels higher similar to that of the lid margins, although than control subjects after six months wear. 4.8 viable microbial counts may be considerably There are no quantitative data for conjunc lower on the conjunctiva. 1.2 The bacteria iso tival microbial carriage in long-term lens lated (coagulase positive and negative staphy wearers. Acanthamoeba is a recently recog lococci, coryneforms, micrococci, and less nised cause of devastating keratitis in lens commonly a-haemolytic streptococci and wearerslO,11 and we were interested in the various Gram-negative bacteria) are typical possibility that one factor contributing to the of areas of skin with a high humidity.3 It is increased susceptibility of long-term contact likely that microorganisms found on the lid lens wearers to this condition might be high margin and conjunctiva represent transient bacterial populations on the external eye, residents rather than autochthonous especially because the appetite of Acantha populations.' moeba for bacteria is the basis for their in vitro The possibility that contact lens wear dis isolation and culture. turbs the ecology of the conjunctiva has been We therefore instigated a survey of bact investigated previously. These studies report erial carriage on the conjunctiva, lid margin similar types of bacteria in lens wearers and and tear film of lens wearers and a matched normal subjects;4-7 however an increase in control group. The 'quantitative' swab pro Gram-negative organisms in lens wearers has cedure introduced by Cagle and Abshirel2 was been found in two studies.8,9 There is some used for lid margin and conjunctiva silmple
From lUniversity Department of Ophthalmology, Bristol Eye Hospital, and 'Department of Microbiology, Bristol Royal Infirmary, Bristol, United Kingdom.
Correspondence to: Mr D. F. P. Larkin, Moorfields Eye Hospital, City Road, London ECIV 2PD. QUANTITATIVE ALTERATIONS OF THE COMMENSAL EYE BACTERIA IN CONTACT LENS WEAR 71 collection: this is in fact a semi-quantitative with a sterile calibrated glass sampling pipette method. Five ilL samples were collected for (Corning Scientific Products, Corning, New quantitative analysis of tears. The term 'quan York). Care was taken to avoid corneal con titative' is used below to describe numerical tact and reflextear secretion by instructing the analysis of bacteria at each site, although patient to gaze superiorly at time of sampling. strictly correct only for tear filmexamination. The tear fluid was then expressed from the pipette to a sterile tube containing 0.5 ml pep Patients and Methods tone water, using a micro-pi petting aid Patients: Thirty-four asymptomatic contact (Micropipex capillary pipetter, Richardsons lens wearers were recruited. None had a of Leicester, England). history of eye disease and all had been using Laboratory method: A 0.9 ml volume of Cal daily-wear lenses for correction of refractive gon Ringer solution (Lab M, Bury, England) error only, for at least 12 months (range 12 was added to each tube containing an alginate months-24 years, mean 8. 4 years). This group swab. Calcium alginate was dispersed by mix comprised 18 soft lens wearers and 16 who ing the tube contents on a vortex mixer for used either hard or rigid gas-permeable two minutes. From each tube containing lenses. Patients complied with instructions on swabs or tear fluid, 0. 4 ml was inoculated on lens hygiene and all had disinfected their to a plate of heated blood agar (blood agar lenses on the day of sample collection. Soft base (Lab M) supplemented with 10% defi lens wearers used hydrogen peroxide, chlo brinated horse blood) and spread with a glass rhexidine; thiomersal or halazone for disin hockey stick. Plates were incubated in 7% CO for 48 hours. fection; hard or gas-permeable lens wearers 2 used chlorhexidine, thiomersal, benzalko Colonies were counted and grouped nium chloride or chlorbutol. The group con according to colony morphology, Gram reac tion, catalase and oxidase production. Sta sisted of 12 males and 22 females, age range phylococci and micrococci were further 21-51 (mean 30) years. A group of 34 control, subdivided on the basis of furazolidone and non-lens wearing subjects were also studied. novobiocin sensitivity; coagulase, phospha These were also asymptomatic and had no tase and urease production; production of history of external eye infection. The control acid from mannitol, trehalose, mannose, group comprised 11 males and 23 females, age maltose, sucrose and xylose, and pattern of range 19-44 (mean 27) years. growth in tubes of semi-solid thioglycolate Slit-lamp examination of the external eye medium. 13 Gram-negative bacilli were identi was undertaken to exclude clinical abnormal fiedusing commercial microtest systems (API ity prior to sample collection. Contact lenses 20E and 20NE; API Laboratory Products, were not removed at the time of sample col Basingstoke, UK). Counts were expressed as lection and had been worn four to eight hours the number of colony forming units (cfu) of a on that day. given bacterial group at each site. Sample collection: Six samples were taken Statistical analysis: For the purpose of statis from each patient: one from each lower lid tical analysis counts from right and left eyes margin, inferior tarsal conjunctiva and infer were treated as independent variables (n = 68 ior forniceal tear film. All samples were taken in each group); counts from eyes of the same and processed by the same person. individual often differed markedly. The fol A calcium alginate swab (Medical Wire and lowing methods were applied: (1) comparison Equipment Co, Corsham, England) moist of the number of different species isolated ened in sterile normal saline, was used to col from each patient was carried out using Stu lect lid margin samples. The swab was rubbed dent's t test, (2) comparison of the colony from the lateral to medial end of the everted counts at equivalent sampling sites was lid margin, and the swab then placed in a ster carried out using the Mann-Whitney U test, ile tube containing 0.1 ml peptone water. The (3) correlation between bacterial counts swab procedure was repeated for sampling of obtained from different sites was evaluated by the inferior tarsal conjunctiva. Five III of tear the Spearman Rank Correlation method. The fluid was collected from the inferior fornix significance level was taken as 5% (p<0.05). 72 D. F. P. LARKIN AND J. P. LEEMING
Table I. Comparison of the prevalence of bacteria on spp' are those which yielded a test profile the lid margin, conjunctiva and tear filmof lens wearers inconsistent with recognised speciesY and controls. Figures indicate the number of subjects in Quantitative assessment: Histograms record each group from whom bacteria indicated were isolated the number of cfu isolated in each group from Lens wearers Controls lid margins [Fig. 1], conjunctiva [Fig. 2] and
Species (n = 34) (n = 34) tear film [Fig. 3]. Lens wearers yielded higher mean bacterial counts than did control sub Coagulase-negative jects at each site sampled. These differences staphylococci 33 30 were statistically significant for counts from Staph. aureus 7 2 (U Corynebacterium spp. 11 6 the lid margin = 1823, P = 0. 033) and tear Micrococcus spp. 9 14 film (U = 1918, P = 0.04), but not from con Acinetobacter spp. 3 1 junctiva (U = 1964, P = 0.105). The elevation Streptococcus spp. 2 1 of bacterial counts from lids of lens wearers Neisseria 2 1 spp. was particularly notable; eight swabs yielded Moraxella spp. 1 2 Pseudomonas spp. 1 2 counts up to 14 times higher than the maxi Bacillus sp. 1 mum counts obtained from lids of control sub Serratia marcescens 1 jects (maximum counts 2,500 cfu/swab and Serratia liquifaciens 1 175 cfu/swab respectively). Corresponding figures at the other sites studied were less impressive; four conjunctival swabs from lens Results wearers had counts up to 2.2 times higher than Species isolated: Organisms were isolated all control swabs and two tear samples from from at least one of the sampling sites in 33 subjects in each group. Most patients in each lens wearers had counts up to 3.6 times higher group had similar flora at each of the six sites than all tear samples from control subjects. sampled, and there was no qualitative differ There was no association between carriage of ence between the florain the lens wearing and particularly dense bacterial populations and control groups. The proportion of Gram length of lens wear, details of lens care or any negative organisms was similar in both other patient-specific information obtained. groups. Lens wearers tended to have a larger No quantitative difference in the commensal number of species at all sites (mean 3.9 per popUlation was observed between soft and patient) than controls (mean 2.7 per patient) hardlrigid gas-permeable lens wearers [Figs 1, [Student's ttest, p Table II. Sub-classification of coagulase-negative staphylococci isolated G> � 20 " z Species Lens wearers Controls Staph. epidermidis 30 26. 10 Staph. capitis 10 6 Staph. hominis 6 1 Staph. simulans 4 4 0-10 11-100 101-1000 >1000 Staph. warneri 4 1 Staph. haemolyticus 2 Lid Margin Counts (c.f.u./swab) Staph. saprophyticus 1 Staph. spp. 4 7 Fig. 1 Numbers of cfu isolated from the lower lid margin in control and lens wearing groups. QUANTITATIVE ALTERATIONS OF THE COMMENSAL EYE BACTERIA IN CONTACT LENS WEAR 73 60 for these observations is that increases in bac • Hard IG.P. Lens Wearers terial numbers on conjunctiva and in tears are Soft 50 o Lens Wearers due to increases in the numbers of organisms Controls III washed from the lids into the eye, and are 40 therefore secondary to increases in lid popula t ,iii tions. The mechanism by which lens wear '0 30 might effect an increase in microbial popula J tions on the eyelids is not readily apparent. � The data presented are not likely to be z 20 explained by carry-over from lens disinfectant solutions or microbial growth around the 10 lenses. If this were the case, the quantitative increases observed should be accompanied by. >64 qualitative shift towards Gram-negative organisms, which constitute the majority of Conjucllval Counts (c.l.u./ swab) organisms isolated from lens storage cases 14-16 Fig. 2 Numbers of cfu isolated from inferior tarsal and various in-use lens care solutions. The conjunctiva in control and lens wearing groups. observation that lens wearers did not yield sig nificantlymore Gram-negative bacteria in this tions between counts from the three sites study may well reflect the efficiency of host studied. Greatest correlation was evident in antimicrobial defences at the ocular surface the lens wearing group: lid margin-conjunc rather than fastidious attention to hygiene by tiva rs = 0.47 (p 11 Key words: commensal eye flora, contact lens wear. Moore MB, McCulley JP, Newton C, et al: Acantha moeba keratitis: A growing problem in soft and hard contact lens wearers. Ophthalmology 1987, References 94: 1654-61. 12 1 Locatcher-Khorazo D and Seegal BC: Microbiology Cagle GD and Abshire RL: Quantitative ocular bac of the eye. St Louis. CV Mosby, 1972: 13 -23. teriology: a method for the enumeration and 2 Allansmith MR, Ostler B, Butterworth M: Concom identification of bacteria from the skin-lash mar itance of bacteria in various areas of the eye. Arch gin and conjunctiva. Invest Ophthalmol Vis Sci Ophthalmol1969, 82: 37-42. 1981,20: 751-7. 3 Noble WC: Microbiology of the human skin. 2nd ed. 13 Schleifer KH (Ed): Section 12. Gram-positive cocci. London. Lloyd-Luke, 1981. In: Sneath PHA, Mair NS, Sharpe ME, Holt JG 4 Rauschl RT and Rogers JJ: The.effect of hydrophilic eds. Bergey's Manual of Systematic Bacteriology, contact lens wear on the bacterial flora of the Vol 2, Baltimore: Williams and Wilkins 1986: human conjunctiva. Int Contact Lens Clin1978, 5: 999-1103. 56-62. 14 Mayo MS, Schlitzer RL, Ward MA, Wilson LA, 5 McBride ME: Evaluation of the microbial flora of Ahearn DG: Association of Pseudomonas and the eye during wear of soft contact lenses. Appl Serratia corneal ulcers with use of contaminated Environ Microbiol1979, 37: 233-6. solutions. J Clin Microbiol1987, 25: 1398-400. 6 Smolin G, Okumoto M, Nozik RA: The microbial 15 Donzis PB, Mondino BJ, Weissman BA, Bruckner flora in extended-wear soft contact lens wearers. DA: Microbial contamination of contact lens care Am J Ophthalmol1979, 88: 543-7. systems. Am J Ophthalmol1987, 104: 325-33. 7 Callender MG, Tse LSY, Charles AM, Lutz D: Bac 16 Larkin DFP, Kilvington S, Easty DL: Contamina terial flora of the eye and contact lens cases during tion of contact lens storage cases by Acantha hydrogel lens wear. Am J Optom Physiol Optics moeba and bacteria. Br J Ophthalmol 1990, 74: 1986,63: 177-80. 133-5. 7 8 Hovding G: The conjunctival and contact lens bacte 1 Mayo MS, Cook WL, Schlitzer RL, Ward MA, Wil rial floraduring lens wear. Acta Ophthalmol1981, son LA, Ahearn DG: Antibiograms, serotypes 59: 387-401. and plasmid profiles of Pseudomonas aeruginosa 9 Morgan JF: Complications associated with contact associated with corneal ulcers and contact lens lens solutions. Ophthalmology1979,86: 1107-19. wear. J Clin Microbiol1986, 24: 37 2-6 . 10 Centers for Disease Control: Acanthamoeba kera 18 Larkin DFP and Easty DL. External eye bacteria as titis associated with contact lenses-United a nutrient source for Acanthamoeba. Graefe's States. MMWR 1986,35: 405-8. Arch Clin Exp Ophthalmol1990, 228: [in press].