Quantitative ocular bacteriology: a method for the enumeration and identification of bacteria from the skin-lash margin and conjunctiva

Transcription

1 Quantitative ocular bacteriology: a method for the enumeration and identification of bacteria from the skin-lash margin and conjunctiva Gerald D. Cagle and Robert L. Abshire Type 1 Calgiswabs were used to collect bacterial specimens from the skin-lash margins and conjunctivae of normal volunteers. The swabs were held in a nonnutritive balanced salt solution during transportation to the laboratory. Alginate swabs were dissolved in a two-step procedure in TC Eagle's medium and 2.5% sodium hexametaphosphate solution. Aliquots of suspended bacteria from the dissolved swab were plated on blood agar. After incubation, bacterial colonies present were enumerated and identified. Cultures from both the skin-lash margin and conjunctiva show significant variation in the number and types of microorganisms isolated. Quantities ofbaceteria isolated from the lid margin were usually greater than from the conjunctiva. Staphylococcus epidermidis was the most prevalent bacterium isolated. The adsorption of bacteria onto alginate swabs from bacterial suspensions and the recovery of a red pigment-producing organism from the conjunctivae of rabbit eyes previously inoculated show that the method is sensitive and highly reproducible. Key words: bacteria, bacterial enumeration, bacterial identification conjunctiva, lid margin I n order to identify the etiologic agents involved in bacterial infections of the eye, microorganisms present must be isolated. Isolation must be followed by identification of the microbe(s) by laboratory personnel and interpreted by the clinician as (1) a member of the normal resident ocular microflora, (2) transient contaminating microbiota, (3) an extraneous ocular microorganism introduced during specimen collection or processing, or (4) the responsible ocular pathogen. Therefore a knowledge of both the qualitative and quanti- From New Product Research-Medical and Product Support Microbiology, Alcon Laboratories, Inc., Fort Worth, Texas. Submitted for publication March 20, Reprint requests: Gerald D. Cagle, Ph.D., New Product Research-Medical, Alcon Laboratories, Inc., 201 S. Freeway, Box 1959, Fort Worth, Texas tative aspects of ocular microbial infections is essential for diagnosis and treatment. Numerous studies have reported the occurrence and relative incidence of various bacteria from cultures of normal and infected eyes. 1 " 9 However, few methods exist to enumerate and identify ocular bacteria. In this study we describe a reproducible quantitative bacteriological procedure for the enumeration and identification of microorganisms obtained from the lid margin and conjunctiva of human volunteers. The results of in vitro and in vivo bacterial test methods used to validate the procedures employed are also reported. Materials and methods Swabs. Calcium alginate swabs (Calgiswabs type 1; Inolex Corp., Glenwood, 111.) were used to collect specimens from the lid margin and con /81/ $ Assoc. for Res. in Vis. and Ophthal., Inc. 751

2 752 Cagle and Abshire Invest. Ophthahnol. Vis. Sci. June 1981 Table I. Physical characteristics and in vitro bacterial recovery from calcium alginate and cotton swabs* Calcium alginate Cotton t value p value Physical property Mass (gm): Range n = Surface area (mm 2 ): Range n = Absorption (gm) Range n = Bacterial recovery S. aureus ATCC6538P: n = 63 S. pneumoniae strain BS-2: n = x x 10" 6.8 x 10" 1.0 x "Swab material only, shaft not considered in mass or surface area determinations. junctiva. Cotton swabs were used in comparative studies. Both swab types were presterilized. Some physical characteristics of 20 each of both swab types were determined, including mass (weight), surface area, and absorption. The mass of the swab was determined by cutting the swab from the shaft onto tared glassine paper and weighing each swab separately on a Mettler Model H single pan balance. Surface area was calculated for each swab type by the formula for solids bounded by curved surfaces: 4TTI- 2. Absorption was determined by immersing tared swabs into water at 15 sec intervals and repeating measurements of swab weight. Specimen collection. Bacterial cultures of the conjunctival sac and lid margins were obtained with Calgiswabs moistened with several drops of balanced salt solution (BSS; Alcon Laboratories, Inc., Fort Worth, Texas). BSS is an isotonic, physiological salt solution containing no antimicrobial preservative. Conjunctival cultures were obtained without anesthetic by inserting the alginate swab in the lower conjunctival fornix at the nasal margin and rotating it along the fornix to the temporal margin. A second alginate swab, moistened in the same manner, was rotated along the lash margin of the lower eyelid in a similar fashion. The alginate swabs were placed in separate sterile 13 by 100 mm test tubes containing 0.3 ml of BSS (transport buffer) and five 3 mm glass beads. The shaft of the swab was cut with sterile scissors (or the scissor blades were surface-disinfected with a 70% isopropyl alcohol swab) below the area handled in obtaining the specimen. The procedure used to dissolve the alginate swab and suspend the adhering bacteria is described below. 1. Each tube was vortexed (Vortex-Genie mixer; Scientific Products, Evanston, 111.) gently (setting of 4 on scale of 10) for 2 min. 2. A 1.5 ml aliquot of tissue culture medium (TC Eagle-Earl BSS; Difco Laboratories, Detroit, Mich.) was added, and the tube was allowed to stand at ambient temperature for 15 min. The action of the Eagle's medium loosens the strands of the alginate swab. 3. The tube was then vortexed vigorously for 1 min. 4. A 1.3 ml volume of 2.5% sodium hexametaphosphate (Na 3 PO 4 ) 6 solution, ph 7.2, was added to the tube, and the mixture was vortexed vigorously for an additional minute. The hexametaphosphate solution dissolves the loosened alginate strands. In vitro determination of sampling reproducibility. Suspensions of Staphylococcus aureus ATCC 6538P or Streptococcus pneumoniae (strain BS-2, Alcon Laboratories Ophthalmic Microbiology Culture Collection) were grown on slants of trypticase soy agar (TSA) (BBL Microbiology

3 Volume 20 Number 6 Quantitative ocular bacteriology Time(hours) Fig. 1. Graphic representation of the log number of survivors of S. pyogenes, S. aureus, A. calcoaceticus and M. lacunata incubated in a nonnutritive salt solution at 25 C for 8 hr. Table II. Average recovery of the biological indicator from rabbit eyes Swab Test organism Time A (hr) Calcium alginate Cotton t value p value S. marcescens TNTC B TNTC <1 ND C p < 0.05 A Time after instillation of test organism. "Too numerous to count, >300 CFU. c Not done. Systems, Cockeysville, Md.) supplemented with 5% defibrinated sheep blood (blood agar). The cells were removed from the agar surface and resuspeneded in sterile 0.1M phosphate buffer (solution 3, 21 Code of Federal Regulations [CFR] ). The suspensions of bacteria were adjusted to a 50% transmittance (T) at 520 nm with a Bausch & Lomb Spectronic 20 spectrophotometer. The suspensions contained approximately 10 8 organisms/ml as determined by viable colonyforming units. Calcium alginate or cotton swabs were immersed in the bacterial suspensions for 15 sec. The number of viable bacteria was determined by processing both cotton and Calgiswabs as described above. Aliquots of the bacteria suspended from the dissolved swab were plated on blood agar and incubated at 37 C in an atmosphere enriched with 4% CO 2. Colonies were counted after 24 hr. In vivo determination of sampling sensitivity. Ten normal rabbit eyes were cultured, and the swabs were processed by the Eagles-hexametaphosphate dissolution procedure. Suspended bacteria were plated on TSA and incubated at room temperature in order to confirm the absence of microorganisms capable of producing a red pigment (prodigiosin). A prodigiosin-producing strain of Serratia marcescens (ATCC 14756) was grown in trypticase soy broth (TSB) (BBL) for 18 hr and diluted to a 50% T in 0.1M phosphate buffer as described above. The suspension was diluted 1: 100 in the buffer, and 0.05 ml of the suspension

4 754 Cagle and Abshire Invest. Ophthalmol. Vis. Set. June 1981 Table III. Quantitative bacterial results of normal volunteers cultured at two ophthalmic centers A Patient No. Eye Boston S. epidermidis Alpha-hemolytic streptococci S. aureus P. aeurginosa Houston S. epidermidis Corynehacterium sp. Micrococci Neisseria sp. B Gram-negative rods OD OS 51/186 0/1 84/258 6/0 3120/ /0 3/3 c 0/60 D 0/6 111/150 22/225 14/20 3/0 2100/195 0/66 33/0 93/0 0/24 12/183 6/6 3/3 3/0 K 18/2 3/0 3/ /1326 0/9 120/140 A Results presented as number of conjunctival bacterial isolates/skin-lash margin bacteria. Cultures obtained with Calgiswabs dissolved in Eagle's medium and hexametaphosphate solution. B NotN. gonorrhoeae or N. meningitidis. c Pseudomonas sp., not P. aeruginosa. D Pasteurella sp. t: Alcaligenes. f Flavobacterium sp. (approximately 50,000 bacteria) was placed in the cul-de-sac of 10 rabbit eyes. At intervals of 10 and 30 min and 1, 2, and 4 hr after instillation of S. marcescens into one cul-de-sac of the rabbit eyes, the lower conjunctiva of each eye was cultured in a random manner with either a cotton swab or a Calgiswab. Swabs collected at each sampling interval were processed by the dissolution procedure described above. Aliquots of the bacteria suspended in the swab dissolute were placed on TSA and incubated at room temperature to enhance red pigment production by S. marcescens. Colonies with red pigment were counted after 48 hr. Human bacteriology. Bacterial cultures from the lid margin and conjunctiva were obtained from 10 individuals at each of two geographic locations (Boston, Mass., and Houston, Texas). Specimens were collected with Calgiswabs and dissolved by the two-step Eagle's-hexametaphosphate procedure. Investigators were requested to process specimens as soon as possible, but in any case within 4 hr of collection. Interim storage of cultures was at ambient temperature. Aliquots of the dissolved swabs were placed on blood agar and chocolate agar and incubated at 37 C for 24 hr in an atmosphere enriched with 4% CO 2. In order to study the viability of bacteria in the transport tube containing BSS and to validate a 4 hr holding period of the collected ocular samples, we suspended four bacterial species in 9 ml vol of BSS. Initial viable colony counts were performed on each of the four microorganisms maintained at approximately 25 C. At intervals of 1, 2, 4, and 8 hr viable cell counts were repeated. For these studies S. aureus (ATCC 6538P) and Acinetobacter calcoaceticus (ATCC 19606) were grown on TSB, Streptococcus pyogenes (Alcon Culture Collection No. 201) was cultured on blood agar, and Moraxella lacunata (CDC strain D-7066) was grown on heart infusion agar (Difco) supplemented with 2.5% bovine serum. All cultures were incubated for 24 hr at 37 C. Bacteria from initial viable cell counts and those performed at each subsequent time interval were plated and incubated on the same media used to culture the organisms. Bacterial counts at each time interval were performed in

5 Volume 20 Number 6 Quantitative ocular bacteriology 755 Patient No /25 0/ /20 0/20 6/0 1900/384 3/0 0/2 2/22 18/56 46/51 6/0 39/ /60 900/120 63/0 120/123 42/660 27/0 1440/ /48 0/18 730/15 0/60 168/ / /3 840/ /0 0/2400 3/0 K triplicate, and the viability of each organism was determined in duplicate experiments. Cultures were incubated for 24 hr at 37 C. Tests of statistical differences between pairs of data collected with alginate vs. cotton swabs were performed with paired t tests. Probability values based on the number of pairs (n) in the test were obtained from tables. Results The comparative properties of alginate and cotton swabs used in this study are shown in Table I. Calgiswabs and cotton swabs used were selected from four different lots during the course of the investigation. Cotton swabs are considerably larger than Calgiswabs (type 1) and have, on the average, a swab mass (shaft excluded) eight times, a surface area of more than four times, and an absorption capacity of 19 times that of alginate swabs. However, the relative number of S. aureus and S. pneumoniae retrieved from Calgiswabs, processed by the two-step dissolution procedure, was greater than that retrieved from cotton swabs by a factor of approximately 10 (Table I). Statistical tests showed that a significantly greater number ot ii. aureus and S. pneumoniae () were recovered with alginate swabs. The recovery of an indicator organism from the conjunctiva of rabbits was performed to determine whether in vivo bacterial recovery with alginate swabs would parallel results obtained from in vitro studies. After placing approximately 10 4 viable S. marcescens into the lower fornix of 10 rabbit eyes, we cultured the conjunctiva and inspected inoculated TSA plates for red colonies. Table II shows that at 10 and 30 min colony counts from each swab were about equal but were increasingly different at 1, 2, and 4 hr. These data indicate that when large numbers of microorganisms were present, both swab types were satisfactory for bacterial recovery from the eye. However, when the number of bacteria present was significantly lower (e.g., after antibiotic therapy), the alginate swab/dissolution procedure was the more sensitive recovery method. In vivo results obtained with alginate swabs (Table II) at 2 and 4 hr correlated well with in vivo data (Table I); S.

6 756 Cagle and Abshire Invest. Ophthalmol. Vis. Set. June 1981 marcescens was recovered by a factor greater than 10 vs. bacterial recovery with cotton swabs. Paired t tests of these data at each sampling showed that at 30 min, significantly greater numbers of S. marcescens were recovered from the conjunctivae of rabbit eyes with cotton swabs (p < 0.05). However, at each succeeding sampling time, significantly more bacteria were cultured from alginate conjunctival swabs (). The viability of four bacteria incubated in transport buffer (BSS) at 25 C is presented in Fig. 1. The log number of survivors was unchanged for each of the microorganisms over the 8 hr incubation period. Variations in the actual viable cell counts may be attributed to technical errors involved in repetitive serial dilutions and counting of colony-forming units. The utility of the alginate swab dissolution procedure was tested in swabs taken from the eyes of 10 volunteers without apparent infection at two ophthalmic centers. The results of these studies are shown in Table III. For each participant the number of conjunctival isolates is shown over the number of skin-lash margin bacteria. The number of bacteria recovered from normal human lid margin and conjunctiva ranged from 0 to 1.2 X 10 4 /ml. Staphylococcus epidermidis was the most common bacterium present and was isolated from 30 to 40 (75.0%) skin-lash margin cultures and from 27 of 40 (76.5%) conjunctival cultures. Corynebacteria and micrococci were the next most common microbes and were isolated from 35% of the bacterial cultures. Significant numbers ofpseudomonas aeruginosa and alpha-hemolytic streptococci were isolated in Boston. Fewer numbers of S. aureus and four gram-negative rods were also cultured, including species of Alcaligenes, Pasteurella, Flavobacterium, and Pseudomonas (notp. aeruginosa). ANeisseria species was also isolated from one individual. Discussion The methods of specimen collection and processing are extremely important to the isolation of ophthalmic microbial isolates. Ordinary cotton swabs contain unsaturated fatty acids known to be toxic to certain fastidious bacteria, and the importance of employing moistened swabs has been documented. 9 In addition, when conventional semisolid transport media are employed, significant numbers of viable bacteria may remain in the holding medium. Comparative studies of the physical characteristics and of in vitro and in vivo sample reproducibility of alginate and cotton swabs demonstrate the efficiency and reproducibility of this quantitative method. In these studies the mass, surface area, and uptake of fluid by Calgiswabs proved uniform compared to cotton swabs. Bacterial recovery in vitro from suspensions (Table I) and from rabbit conjunctiva inoculated 2 hr previously with S. marcescens (Table II) proved to be more efficient with alginate swabs by a factor of 10 or more. The viability of four bacterial species, including S. aureus, A. calcoaceticus, S. pyogenes, and M. lacunata, in the transport buffer (BSS) (Fig. 1) supports the use of a nonnutritive balanced salt solution as an interim holding menstruum. Prior to the development of this quantitative culture technique, only qualitative and semiquantitative evaluation of ocular microflora was possible with relative ease and precision. Hadley et al. 10 reported reproducible results using a cellulose microsponge. The procedure we employed is completed rapidly, avoiding the patient inconvenience caused by the presence of a microsponge in the cul-de-sac. The use of the alginate swab also allowed us to obtain separate cultures of the lid margin and conjunctiva. In addition, since a nonabsorptive material is used to collect the specimen and because no fluid is added to the conjunctival sac to enhance sponge wetting, this quantitative bacteriological procedure can be less laborious and completed in a more rapid and reproducible manner. Mechanical movement of the alginate swab over most of the conjunctival surface also releases bacteria attached to surface cells or cellular convolutions. It is important to follow closely the procedures outlined for dissolution of the swab.

7 Volume 20 Number 6 Quantitative ocular bacteriology 757 The "loosening fluid" we use is TC Eagle's medium, a solution capable of supporting bacteria growth. However, incubation in the loosening fluid is restricted to 15 min. This period of time is generally not sufficient for bacterial reproduction and therefore does not alter relative numbers of bacteria isolated from the culture sites. The spectrum of bacterial isolates from the 20 individuals cultured is similar to that in studies of normal eyes reported previously. 9 Individuals harboring bacteria commonly termed pathogenic, including P. aeruginosa and species of Pseudomonas and Alcaligenes, alpha-hemolytic streptococci (not group D), and S. aureus showed no apparent signs of infection. However, these organisms may represent normal ocular bacterial microbiota or ocular bacterial transients. In this study we have presented the results of a rapid and reproducible quantitative ocular bacteriological technique. Data collected in normal human volunteers indicate that significant variation occurs in the number and species of aerobic ocular bacteria flora. Anaerobes were not cultured in these studies. Subsequent studies will report on the reestablishment of ocular microflora after topical antibiotic therapy and evaluate these methods in the collection of bacterial specimens from patients with external ocular infections. We acknowledge the technical contribution of Ms. Nettie Robinson in development of the alginate swab dissolution procedure, and R. H. Stewart, M.D., and H. M. Liebowitz, M.D., in collecting ocular microbial specimens. REFERENCES 1. Axenfeld T: The Bacteriology of the Eye, McNab A, translator. London, 1908, Bailliere, Tindall & Casell Ltd. 2. Lucic H: Bacteriology of the normal conjunctival sac. Am J Ophthalmol 10:829, Khorazo D and Thompson R: Bacterial flora of normal conjunctiva. Am J Ophthalmol 28:306, Rodin RH: Bacteriologic study of human conjunctival flora. Am J Ophthalmol 28:306, Cason L and Winkler CH Jr: Bacteriology of the eye. I. Normal flora. Arch Ophthalmol 51:196, Smith CH: Bacteriology of the healthy conjunctiva. Br J Ophthalmol 38:719, Soudakoff PS: Bacteriologic examination of the conjunctiva. Survey of 3,000 patients. Am J Ophthalmol 38:374, Allansmith MR, Ostler MB, and Butterworth M: Concomitance of bacteria in various areas of the eye. Arch Ophthalmol 82:37, Locatcher-Khorazo D and Seegal BS: Microbiology of the Eye. St. Louis, 1972, The C. V. Mosby Co. chap Hadley WK, Aronson SB, and Goodner EK: Quantitative conjunctival bacteriology. Arch Ophthalmol 90:386, 1973.