- Email: [email protected]
Povidone-iodine for Ophthalmia Neonatorum Prophylaxis
Povidone-iodine for Ophthalmia Neonatorum Prophylaxis Sherwin J. Isenberg, M.D., Leonard Apt, M.D., Robert Yoshimori, Ph.D., Rosemary D. Leake, M.D., and Roberta Rich, R.N. PURPOSE: The agents currently used to prevent ophthalmia neonatorum are less than opti mal, with reports indicating evidence of bac terial resistance, ineffectiveness, and toxicity. Povidone-iodine ophthalmic solution, which has been shown to be effective in the preoperative preparation of the eye, generates no resistance, is an effective antimicrobial agent, and has low toxicity. We evaluated the effec tiveness and safety of povidone-iodine for ophthalmia neonatorum prophylaxis. METHODS: A bacterial culture was taken from the conjunctiva of each eye of 100 infants within 30 minutes of birth. A drop of 2.5% povidone-iodine solution was then placed on one eye, while the other eye received either one drop of silver nitrate 1% ophthalmic solu tion or 0.5% erythromycin ointment. Conjunctival bacterial cultures were again taken two to four hours after birth. At each culture and at 24 hours after birth, the eyes were examined for toxic changes. To measure the effective ness of the medications, the number of bacte rial colony-forming units and species from each culture was compared. RESULTS: All three agents significantly reduced the number of colony-forming units, but povidone-iodine caused the most significant decrease. The number of species was reduced significantly by povidone-iodine (P = .00051) and silver nitrate (P = .007), with povidoneiodine yielding the most significant decrease. Erythromycin did not significantly reduce the
Accepted for publication July 13, 1994. From the Jules Stein Eye Institute, Los Angeles, Cali fornia (Drs. Isenberg and Apt); Departments of O p h t h a l mology and Pediatrics, Harbor/UCLA Medical Center, UCLA School of Medicine, Torrance, California (Drs. Isenberg, Yoshimori, and Leake and Ms. Rich). This study was supported in part by grant RR00425 from the National Institutes of Health, Bethesda, Maryland, a n d by a grant from Allergan Corporation, Irvine, California. Reprint requests to Sherwin J. Isenberg, M.D., Depart ment of Ophthalmology, Harbor/UCLA Medical Cen ter, 1000 W. Carson St., Torrance, CA 90509.
number of species. Silver nitrate demonstrat ed more ocular toxicity at the 24-hour deter mination point than did either of the other two medications (P < .001). CONCLUSIONS: Povidone-iodine 2.5% ophthal mic solution is an effective antibacterial agent on the conjunctiva of newborns and causes less toxicity than silver nitrate. OPHTHALMIA NEONATORUM is generally denned
as conjunctivitis occurring within one month of birth. Untreated, this infection can cause blind ness, especially if the offending organism is Neisseria gonorrhoeae. Perhaps the greatest sin gle step toward preventing blindness in neonates was accomplished in 1881. By applying a 2% solution of silver nitrate to the eyes of neonates, Karl S. F. Credé reduced the inci dence of ophthalmia neonatorum from 10% to 0.3%. Subsequently, this prophylaxis has been mandated in many countries, including the United States. Erythromycin and tetracycline are also being used for this purpose. The choice of agent, however, has been controversial be cause of concerns of ineffectiveness and in duced ocular toxicity. For example, in a recent controlled clinical trial of 630 infants, the fre quency of conjunctivitis within 60 days after birth was studied. 1 Conjunctivitis occurred in 22% of newborns who did not receive any prophylaxis, but was reduced to 16% and 14% for those who received erythromycin and silver nitrate, respectively. The reduction induced by silver nitrate was statistically significant. A more ideal agent would have a broader spectrum of antibacterial activity than current ly used agents and would cause less resistance to microorganisms, induce fewer toxic reac tions, be easy to administer, be available at low cost, and be applicable worldwide. In a series of studies by our group and others, povidoneiodine was found to be an effective antibacterial agent when placed on the eye during the preoperative preparation. 2 " 4 Unlike the agents cur rently used for prophylaxis against ophthalmia neonatorum, povidone-iodine is also effective against microbes other than bacteria. Povidone-
©AMERICAN JOURNAL OF OPHTHALMOLOGY 118:701-706, DECEMBER, 1994
701
702
AMERICAN JOURNAL OF OPHTHALMOLOGY
iodine has been found to be potent against fungi.5 Thomas 6 showed that Chlamydia trachomatis was inactivated by povidone-iodine. The herpesvirus and the human immunodeficiency virus (HIV), among others, are sensitive to povidone-iodine applied topically. 7 Other im portant considerations regarding povidoneiodine are cost and availability. Both these factors favor the use of povidone-iodine to the degree that it is recommended for general use in developing countries to treat infectious con junctivitis. 8 Before the use of povidone-iodine is consid ered for prophylaxis against ophthalmia neonatorum, its toxicity and effectiveness should be assessed. In the current study of 100 neonates, we compared povidone-iodine, silver nitrate, and erythromycin with respect to effectiveness and toxicity.
Methods This protocol was reviewed and approved by the Human Subjects Committee of the Har bor-UCLA Medical Center, UCLA School of Medicine. Before the study of each neonate, a signed informed consent was obtained from the mother. In a pilot study, we first applied the 5% solution of povidone-iodine that had previous ly been found very effective in preoperative application to the conjunctiva of neonates. 2,3 We found, however, that in many of the infants an unsatisfactory hyperemia of the conjunctival vessels developed. On diluting the povidoneiodine to a 2.5% solution, we no longer saw the hyperemia. Because povidone-iodine has been found to have antimicrobial properties in a dilution as low as 0.25%, we continued to use the 2.5% solution in the current investigation. 6 We studied 100 infants born of vaginal deliv ery. Neonates with a maternal history of peri natal infection, including amnionitis, were ex cluded. Each infant had four bacterial cultures of the conjunctiva. An initial culture was taken within 30 minutes of birth from the inferotemporal conjunctiva of each eye. After this cul ture, one drop of 2.5% povidone-iodine solu tion was placed on the conjunctiva of one randomly selected eye. At the same time, on a random basis, either one drop of silver nitrate 1% solution or erythromycin 0.5% ophthalmic ointment was applied to the conjunctiva of the
December, 1994
other eye. In order to evaluate the effect of the medications, the culture was repeated two to four hours later from the inferonasal quadrant, or vice versa. With radioactively labeled technetium, Hardberger, Hanna, and Boyd9 found the half-life of ophthalmic ointments on the human eye to be 9.7 minutes for blinking sub jects and 32 minutes for nonblinking subjects. Thus, by two to four hours, there would not be a substantial amount of erythromycin ointment that might affect the results of the second cul ture. Each specimen was obtained with a sterile anaerobic transport swab. The swab was washed three times in 0.5 ml of Schaedler's broth and wrung out by pressing the swab along the sides of the tube. The swab was cultured in 10 ml of Schaedler's broth, whereas blood and chocolate agar were each inoculated with 0.1 ml of eluant and spread on the surface of the agar with a glass rod. The blood agar plates were incubated for seven days at 35 C in an anaerobic jar with a gas mixture of 80% nitrogen, 10% carbon dioxide, and 10% hydro gen. The chocolate agar plates were incubated in 5% to 10% carbon dioxide at 35 C. After incubation, the colonies were differentiated and enumerated by standard bactériologie pro cedures. All specimens were coded in the deliv ery room or nursery. The microbiologist did not know which eye was the source of the speci men, which prophylactic agent was used, or whether the specimen had been taken immedi ately after birth or two to four hours postnatally. To study toxic effects of the agents to the ocular surface, we devised a grading scale rang ing from 0 to 9, with 0 representing no abnor mal reaction and 9 a purulent hyperemic con junctiva with severe eyelid edema. The grading scale was as follows: for conjunctival hyper emia, a score of 0 represented no reaction; 1, mild; 2, moderate; and 3, severe. For eyelid edema, a score of 0 represented no reaction; 1, mild; 2, moderate; and 3, severe. For ocular discharge, 0 represented no reaction; 1, watery; 2, mild purulence; and 3, severe purulence. The appearance of the conjunctiva was graded at birth before applying any agent and then again 24 hours after birth. The Student f and Wilcoxon rank sum tests were used to compare the change induced in bacterial colonies and species and to evaluate the toxicity scores for each agent at each assess ment.
Povidone-iodine for Ophthalmia Neonatorum Prophylaxis
Vol. 118, No. 6
703
TABLE 2 EFFECT OF EACH AGENT ON THE NUMBER OF SPECIES CULTURED PER EYE (MEAN ± S.D.)
Results Of the 100 newborns studied, 51 were female. All 100 neonates received povidone-iodine in one eye. In the other eye, erythromycin 0.5% ophthalmic ointment was placed in 48 infants, and silver nitrate 1% solution in 52. The gestational ages were the same for the erythromycin group (39.8 ± 1 . 6 weeks) and the silver nitrate group (39.8 ± 1.9 weeks). Table 1 shows the reduction in colonyforming units engendered by each of the agents. Whereas all three agents significantly decreased the number of colony-forming units, povidone-iodine was the most effective; t val ues for povidone-iodine, silver nitrate, and erythromycin were 3.51, 2.76, and 2.03, respec tively. Erythromycin reduced the count by an amount that was barely significant (P = .046). The reduction in number of species cultured per eye is displayed in Table 2. Povidone-iodine reduced the number of species more than the other two medications. Erythromycin did not induce a significant reduction in the number of species per eye. By the Wilcoxon rank sum test, povidone-iodine caused a greater reduction in the number of colony-forming units (P < .05) and species (P < .002) than did erythromycin. Figure 1 demonstrates the percent reduction in the colony-forming units and number of species for each of the medications. Table 3 illustrates the toxicity caused to the anterior segment of the eye by each of the agents. By the rank sum test, silver nitrate had a significantly worse score 24 hours after birth than did the other two agents (P < .001). Figure 2 shows the frequency of significant toxic reac tions caused by each medication 24 hours after birth. At each level, silver nitrate was found to be the most frequent causative agent.
NO. OF SPECIES CULTURED
MEDICATION
AT BIRTH
TWO TO FOUR HOURS POSTNATALLY
P VALUE
Silver nitrate Erythromycin Povidone-iodine
0.96 ± 0.94 0.85 ± 0.87 1.1 ±1.1
0.58 ± 0.69 0.66 ± 0.83 0.60 ± 0.70
.021 .28* .00016
*Not significant.
Discussion Although our studies of povidone-iodine ap plied to the eye as part of preoperative prepara tion found it to be nontoxic and highly antibac terial, the possible use of povidone-iodine for ophthalmia neonatorum prophylaxis stimu lates other considerations. The conjunctival flo ra of the neonate is different from that of the adult. Isenberg and associates 10 investigated the bacterial flora within 15 minutes after birth and before any topical medications were applied. We found the largest group of bacteria to be microaerophilic, predominantly Lactobacillus species. The second largest group were true anaerobes, such as Bifidobacterium, Bacteroides, and Propionibacterium species. The smallest group were aerobic bacteria. The conjunctival bacteria of infants born by vaginal delivery reflect the bacterial flora of the vagina. In a subsequent investigation, we found the con junctival cultures of infants delivered by cesarean section within three hours of membrane
TABLE 1 EFFECT OF EACH AGENT ON THE NUMBER OF COLONY-FORMING UNITS PER EYE (MEAN ± S.D.) NO. OF COLONY-FORMING UNITS
MEDICATION
AT BIRTH
TWO TO FOUR HOURS POSTNATALLY
P VALUE
Silver nitrate Erythromycin Povidone-iodine
213 ± 517 146 ± 400 175 ± 425
14 ± 53 25 ± 107 22 ± 84
.007 .046 .00051
Povidone-iodine
Erythromycin
Silver-nitrate
Fig. 1 (Isenberg and associates). Percent reduction in the number per eye of colony-forming units and species induced by each of the medications.
704
December, 1994
AMERICAN JOURNAL OF OPHTHALMOLOGY
TABLE 3 TOXICITY SCORE* INDUCED BY EACH MEDICATION (MEAN ± S.D.)
25 20
MEDICATION
AT BIRTH«
24 HOURS POSTNATALLY
Silver nitrate Erythromycin Povidone-iodine
3.1 ± 1.3 3.1 ± 1.2 3.1 ± 1.2
3.4 ± 1.8* 3.0 ± 1.4 3.2 ± 1.6
"See text. 'Before any ocular medication. ♦Higher score than the other two 24-hour postnatal values (P < .001 by Wilcoxon rank sum test).
rupture to be sterile in 80% of cases and the rest to display only a few cutaneous bacteria. 11 Neonates delivered by cesarean section three hours after membrane rupture displayed a greater number of conjunctival bacteria that corre sponded more with the babies born by vaginal delivery. Aside from these bacteria, other or ganisms are of concern in the course of ophthal mia neonatorum, especially N. gonorrhoeas, C. trachomatis, herpes simplex virus type II, and possibly HIV. Benevento and associates 12 stud ied the effectiveness of 5.0%, 1.0% and 0.1% solutions of povidone-iodine in vitro against the first three aforementioned organisms. They found that all three concentrations sterilized inocula of herpes simplex virus type II and four different strains of N. gonorrhoeae. However, only the two stronger concentrations were ef fective against C. trachomatis. The choice of an agent best suited to prevent ophthalmia neonatorum has been controver sial. Silver nitrate, reduced to a 1% solution to decrease the frequency of toxic conjunctivitis, has remained popular. However, as a result of a perceived ineffectiveness of silver nitrate against Chlamydia and its tendency to cause toxic conjunctivitis, other medications, such as erythromycin 0.5% and tetracycline 1.0% oint ments, have been widely accepted. 13 Yet these antibiotics are not perfect. For example, Mooney and associates 14 reported a threefold increase in ophthalmia neonatorum when they switched from silver nitrate to erythromycin. Black-Payne, Bocchini, and Cedotal 15 calculated a minimum failure rate of erythromycin pro phylaxis against chlamydial ophthalmia neona torum to be between 7.0% and 19.5%. A Staphylococccus aureus conjunctivitis outbreak that was resistant to erythromycin occurred in a nursery of infants who were given the drug for
■z. LU
o
ΠLiJ
IS 10 5 D TOXICITY SCORE
Fig. 2 (Isenberg and associates). Percent of eyes demonstrating significant toxic reactions caused by each of the medications 24 hours after birth. prophylaxis. The infections ceased after the hospital switched to silver nitrate prophylaxis. 16 In a study of almost 50,000 births, Zanoni, Isenberg, and Apt17 reported ophthalmia neo natorum in general (P < .001) and chlamydial conjunctivitis specifically (P < .02) to be more frequent after erythromycin than silver nitrate prophylaxis. In the United States, gonococcal infections that are resistant to tetracycline are on the rise. 1819 The same trend has been ob served in other countries, such as the United Kingdom and The Netherlands. 20,21 Povidone-iodine has a number of potential advantages over the agents presently used to prevent ophthalmia neonatorum. In 1984, Apt and associates 2 studied the effect of applying a 5% solution of povidone-iodine to the con junctiva as part of the preoperative prepara tion. This agent reduced the colony count by 9 1 % andthespeciescountby 50%. In 1985, Isen berg and associates 3 described that, in combina tion with the outpatient preoperative use of a topical broad spectrum antibiotic preparation (gramicidin-neomycin-polymyxin B), povidoneiodine applied on the operating table preoperatively sterilized 83% of the conjunctivae. The broad antibacterial spectrum of povidoneiodine was statistically demonstrated in these studies. Subsequently, Speaker and Menikoff4 found the preoperative use of topical 5% povidone-iodine solution reduced the rate of postoperative bacterial endophthalmitis from 0.24% to 0.06%. In our previous studies involv ing povidone-iodine (as well as our ongoing use of it), we have seen no toxic reactions to the one-time application of povidone-iodine as part of the preoperative preparation. In the aforementioned study by Speaker and Meni koff,4 no toxic reactions resulted after more
Vol. 118, No. 6
Povidone-iodine for Ophthalmia Neonatorum Prophylaxis
t h a n 3,000 a p p l i c a t i o n s . It h a s a b r o a d e r spec t r u m that i n c l u d e s fungi, v i r u s e s , a n d all b a c teria. In contrast to silver nitrate a n d e r y t h r o m y c i n , p o v i d o n e - i o d i n e has not b e e n associated w i t h true bacterial resistance. 2 2 T h u s , t h e r e n e e d be little fear that resistant bacteria m i g h t affect the eye a n d possibly s p r e a d w i t h i n a n u r s e r y or h o m e situation, as occurred w i t h e r y t h r o m y cin. 15 The c u r r e n t study h a s s h o w n p o v i d o n e iodine to be highly effective a g a i n s t the bacteri al flora found on the eye at b i r t h . W h e r e a s all t h r e e a g e n t s r e d u c e d the n u m b e r of colonyforming u n i t s , p o v i d o n e - i o d i n e a c h i e v e d the best level of statistical significance a n d e r y t h r o mycin the worst. With r e g a r d to the r e d u c t i o n of the n u m b e r of species, p o v i d o n e - i o d i n e a g a i n achieved the best level of statistical signifi cance. Surprisingly, e r y t h r o m y c i n did not re duce the species count significantly. O u r data indicate that e r y t h r o m y c i n is n o t as effective in e l i m i n a t i n g bacterial flora of the n e o n a t e as the other two a g e n t s . As d e m o n s t r a t e d in this s t u d y , p o v i d o n e iodine is relatively nontoxic w h e n a p p l i e d to the sensitive eyes of n e o n a t e s . This lack of toxicity was a c c o m p l i s h e d by r e d u c i n g the con c e n t r a t i o n from the 5 % s o l u t i o n we have u s e d in the eyes of a d u l t s a n d c h i l d r e n to a 2 . 5 % solution. This milder c o n c e n t r a t i o n , n e v e r t h e less, was still effective against bacteria, as evi d e n c e d from our data. C o n v e r s e l y , silver n i t r a t e p r o d u c e d a toxic conjunctivitis m o r e frequently a n d of greater severity t h a n the o t h e r two a g e n t s . At the 2 4 - h o u r p o s t n a t a l d e t e r m i n a t i o n point, silver nitrate p r o d u c e d a toxicity score of 7 or more (generally i n d i c a t i n g m o d e r a t e to severe conjunctival h y p e r e m i a , eyelid e d e m a , and p u r u l e n t discharge) in 1 0 % of all cases. At the same e x a m i n a t i o n , the other two a g e n t s were found to p r o d u c e a similar r e s p o n s e only 1% to 2 % of the t i m e . Because silver nitrate o p h t h a l m i c s o l u t i o n a n d e r y t h r o m y c i n o i n t m e n t are t r a n s l u c e n t on the eye, it is s o m e t i m e s difficult to a p p r e c i a t e that these m e d i c a t i o n s have b e e n d e l i v e r e d . It is possible that a n u r s e may believe t h a t h e or she has a d m i n i s t e r e d the m e d i c a t i o n w h e n it actu ally may never have c o n t a c t e d the conjunctiva. However, p o v i d o n e - i o d i n e a p p e a r s dark b r o w n on the eye for a b o u t two m i n u t e s . This w o u l d e n s u r e its p r o p e r a p p l i c a t i o n a n d d e c r e a s e the possibility that an infant m i g h t go w i t h o u t prophylaxis.
705
The use of p o v i d o n e - i o d i n e offers a d v a n t a g e s in the a r e a s of availability a n d cost. Even in underdeveloped countries, povidone-iodine o p h t h a l m i c s o l u t i o n has b e e n r e c o m m e n d e d , a n d is b e i n g u s e d , as an a n t i - i n f e c t i o u s agent. 8 This study h a s s h o w n that a 2 . 5 % o p h t h a l m i c s o l u t i o n of p o v i d o n e - i o d i n e is an effective a n t i bacterial a n d a relatively n o n t o x i c a g e n t w h e n placed on the conjunctiva of a n e o n a t e . For this p u r p o s e , it h a s a d v a n t a g e s over the two most frequently u s e d p r o p h y l a c t i c m e d i c a t i o n s . This investigation, h o w e v e r , d o e s not a d d r e s s the effectiveness of p o v i d o n e - i o d i n e as a p r o p h y lactic agent specifically against hi. gonorrhoeae, C. trachomatis, h e r p e s simplex virus t y p e II, or HIV. To date, such effectiveness h a s only b e e n d e m o n s t r a t e d in vitro. To s t u d y this issue, a prospective clinical trial i n v e s t i g a t i n g the fre q u e n c y a n d type of o p h t h a l m i a n e o n a t o r u m arising after use of p o v i d o n e - i o d i n e a n d other m e d i c a t i o n s w o u l d be very informative. ACKNOWLEDGMENT
N a n c y Berman, P h . D . , D e p a r t m e n t of Pe diatrics, Harbor/UCLA Medical Center, UCLA School of M e d i c i n e , p r o v i d e d statistical analysis.
References 1. Bell TA, Grayston JT, Krohn MA, Kronmal RA, Eye Prophylaxis Study Group. Randomized trial of silver nitrate, erythromycin, and no eye prophylaxis for the prevention of conjunctivitis among newborns not at risk for gonococcal ophthalmitis. Pediatrics 1993;92:755-60. 2. Apt L, Isenberg S, Yoshimori R, Paez JH. Chemical preparation of the eye in ophthalmic sur gery, III: effect of povidone-iodine on the conjuncti va. Arch Ophthalmol 1984;102:728-9. 3. Isenberg SJ, Apt L, Yoshimori R, Khwarg S. Chemical preparation of the eye in ophthalmic surgery, IV: comparison of povidone-iodine on the conjunctiva with a prophylactic antibiotic. Arch Ophthalmol 1985;103:1340-2. 4. Speaker MG, Menikoff JA. Prophylaxis of endophthalmitis with topical povidone-iodine. Oph thalmology 1991;98:1769-75. 5. White JH, Stephens GM, Cinotti AA. The use of povidone-iodine for treatment of fungi in rabbit eyes. Ann Ophthalmol 1972;4:855-6. 6. Thomas BJ. The inactivation of Chlamydia tra chomatis by Betadine solution. In: Reber H, editor. The proceedings of the World Congress on Antisep sis. New York: HP Publishing Co, 1978:16-8.
706
AMERICAN JOURNAL OF OPHTHALMOLOGY
7. Durno AG, Kaplan JC, Schooley RT. Inactivation of human immune deficiency virus by povidone-iodine. In: Digenis GA, Agha BJ, Ansell J, Blecher L, editors. Proceedings of the Second Inter national Symposium on Povidone. Lexington: Uni versity of Kentucky Press, 1987:152-6. 8. Taylor J. Appropriate eye drugs for developing countries. Int Ophthalmol 1990;14:173-9. 9. Hardberger R, Hanna C, Boyd CM. Effects of drug vehicles on ocular contact time. Arch Ophthal mol 1975;93:42-5. 10. Isenberg SJ, Apt L, Yoshimori R, Alvarez SR. Bacterial flora of the conjunctiva at birth. J Pediatr Ophthalmol Strabismus 1986;23:284-6. 11. Isenberg SJ, Apt L, Yoshimori R, McCarty JW, Alvarez SR. Source of the conjunctival bacterial flora at birth and implications for ophthalmia neonatorum prophylaxis. Am J Ophthalmol 1988;106:458-62. 12. Benevento WJ, Murray P, Reed CA, Pepose JS. The sensitivity of Neisseria gonorrhoeae, Chlamydia trachomatis, and herpes simplex type II to disinfec tion with povidone-iodine. Am J Ophthalmol 1990; 109:329-33. 13. Dinsmoor MJ. Ophthalmia neonatorum. Contemp OB/GYN 1992;37:112-4. 14. Mooney BR, Green JA, Epstein BJ, Hashisaki PA Non-gonoccocal ophthalmitis associated with erythromycin ointment prophylaxis of gonococcal ophthalmia neonatorum. Infect Control 1984,5:13840. 15. Black-Payne C, Bocchini JA Jr, Cedotal C. Fail ure of erythromycin ointment for postnatal ocular
December, 1994
prophylaxis of chlamydial conjunctivitis. Pediatr In fect Dis J 1989;8:491-5. 16. Hedberg K, Ristinen TL, Soler JT, White KE, Hedberg CW, Osterholm MT, et al. Outbreak of erythromycin-resistant staphylococcal conjunctivitis in a newborn nursery. Pediatr Infect Dis J 1990; 9:268-73. 17. Zanoni D, Isenberg SJ, Apt L. A comparison of silver nitrate with erythromycin for prophylaxis against ophthalmia neonatorum. Clin Pediatr (Phila) 1992;31:295-8. 18. Schwarcz SK, Zenilman JM, Schnell D, Knapp JS, Hook EW, Thompson S, et al. National surveil lance of antimicrobial resistance in Neisseria gonorrhoeae. The Gonococcal Isolate Surveillance Project. JAMA 1990;264:1413-7. 19. Knapp JS, Zenilman JM, Biddle JW, Perkins GH, DeWitt WE, Thomas ML, et al. Frequency and distribution in the United States of strains of Neisse ria gonorrhoeae with plasmid-mediated, high-level resistance to tetracycline. J Infect Dis 1987;155:81922. 20. Ison CA, Terry P, Bindayna K, Gill MJ, Adams J, Woodford N. Tetracycline-resistant gonococci in UK [letter]. Lancet 1988;1:651-2. 21. van Klingeren B, Dessens-Kroon M, Verheuvel M. Increased tetracycline resistance in gonococci in The Netherlands [letter]. Lancet 1989,2:1278. 22. Houang ET, Gilmore OJA, Reid C, Shaw EJ. Absence of bacterial resistance to povidoneiodine. J Clin Pathol 1976;29:752-5.
Accepted for publication July 13, 1994. From the Jules Stein Eye Institute, Los Angeles, Cali fornia (Drs. Isenberg and Apt); Departments of O p h t h a l mology and Pediatrics, Harbor/UCLA Medical Center, UCLA School of Medicine, Torrance, California (Drs. Isenberg, Yoshimori, and Leake and Ms. Rich). This study was supported in part by grant RR00425 from the National Institutes of Health, Bethesda, Maryland, a n d by a grant from Allergan Corporation, Irvine, California. Reprint requests to Sherwin J. Isenberg, M.D., Depart ment of Ophthalmology, Harbor/UCLA Medical Cen ter, 1000 W. Carson St., Torrance, CA 90509.
number of species. Silver nitrate demonstrat ed more ocular toxicity at the 24-hour deter mination point than did either of the other two medications (P < .001). CONCLUSIONS: Povidone-iodine 2.5% ophthal mic solution is an effective antibacterial agent on the conjunctiva of newborns and causes less toxicity than silver nitrate. OPHTHALMIA NEONATORUM is generally denned
as conjunctivitis occurring within one month of birth. Untreated, this infection can cause blind ness, especially if the offending organism is Neisseria gonorrhoeae. Perhaps the greatest sin gle step toward preventing blindness in neonates was accomplished in 1881. By applying a 2% solution of silver nitrate to the eyes of neonates, Karl S. F. Credé reduced the inci dence of ophthalmia neonatorum from 10% to 0.3%. Subsequently, this prophylaxis has been mandated in many countries, including the United States. Erythromycin and tetracycline are also being used for this purpose. The choice of agent, however, has been controversial be cause of concerns of ineffectiveness and in duced ocular toxicity. For example, in a recent controlled clinical trial of 630 infants, the fre quency of conjunctivitis within 60 days after birth was studied. 1 Conjunctivitis occurred in 22% of newborns who did not receive any prophylaxis, but was reduced to 16% and 14% for those who received erythromycin and silver nitrate, respectively. The reduction induced by silver nitrate was statistically significant. A more ideal agent would have a broader spectrum of antibacterial activity than current ly used agents and would cause less resistance to microorganisms, induce fewer toxic reac tions, be easy to administer, be available at low cost, and be applicable worldwide. In a series of studies by our group and others, povidoneiodine was found to be an effective antibacterial agent when placed on the eye during the preoperative preparation. 2 " 4 Unlike the agents cur rently used for prophylaxis against ophthalmia neonatorum, povidone-iodine is also effective against microbes other than bacteria. Povidone-
©AMERICAN JOURNAL OF OPHTHALMOLOGY 118:701-706, DECEMBER, 1994
701
702
AMERICAN JOURNAL OF OPHTHALMOLOGY
iodine has been found to be potent against fungi.5 Thomas 6 showed that Chlamydia trachomatis was inactivated by povidone-iodine. The herpesvirus and the human immunodeficiency virus (HIV), among others, are sensitive to povidone-iodine applied topically. 7 Other im portant considerations regarding povidoneiodine are cost and availability. Both these factors favor the use of povidone-iodine to the degree that it is recommended for general use in developing countries to treat infectious con junctivitis. 8 Before the use of povidone-iodine is consid ered for prophylaxis against ophthalmia neonatorum, its toxicity and effectiveness should be assessed. In the current study of 100 neonates, we compared povidone-iodine, silver nitrate, and erythromycin with respect to effectiveness and toxicity.
Methods This protocol was reviewed and approved by the Human Subjects Committee of the Har bor-UCLA Medical Center, UCLA School of Medicine. Before the study of each neonate, a signed informed consent was obtained from the mother. In a pilot study, we first applied the 5% solution of povidone-iodine that had previous ly been found very effective in preoperative application to the conjunctiva of neonates. 2,3 We found, however, that in many of the infants an unsatisfactory hyperemia of the conjunctival vessels developed. On diluting the povidoneiodine to a 2.5% solution, we no longer saw the hyperemia. Because povidone-iodine has been found to have antimicrobial properties in a dilution as low as 0.25%, we continued to use the 2.5% solution in the current investigation. 6 We studied 100 infants born of vaginal deliv ery. Neonates with a maternal history of peri natal infection, including amnionitis, were ex cluded. Each infant had four bacterial cultures of the conjunctiva. An initial culture was taken within 30 minutes of birth from the inferotemporal conjunctiva of each eye. After this cul ture, one drop of 2.5% povidone-iodine solu tion was placed on the conjunctiva of one randomly selected eye. At the same time, on a random basis, either one drop of silver nitrate 1% solution or erythromycin 0.5% ophthalmic ointment was applied to the conjunctiva of the
December, 1994
other eye. In order to evaluate the effect of the medications, the culture was repeated two to four hours later from the inferonasal quadrant, or vice versa. With radioactively labeled technetium, Hardberger, Hanna, and Boyd9 found the half-life of ophthalmic ointments on the human eye to be 9.7 minutes for blinking sub jects and 32 minutes for nonblinking subjects. Thus, by two to four hours, there would not be a substantial amount of erythromycin ointment that might affect the results of the second cul ture. Each specimen was obtained with a sterile anaerobic transport swab. The swab was washed three times in 0.5 ml of Schaedler's broth and wrung out by pressing the swab along the sides of the tube. The swab was cultured in 10 ml of Schaedler's broth, whereas blood and chocolate agar were each inoculated with 0.1 ml of eluant and spread on the surface of the agar with a glass rod. The blood agar plates were incubated for seven days at 35 C in an anaerobic jar with a gas mixture of 80% nitrogen, 10% carbon dioxide, and 10% hydro gen. The chocolate agar plates were incubated in 5% to 10% carbon dioxide at 35 C. After incubation, the colonies were differentiated and enumerated by standard bactériologie pro cedures. All specimens were coded in the deliv ery room or nursery. The microbiologist did not know which eye was the source of the speci men, which prophylactic agent was used, or whether the specimen had been taken immedi ately after birth or two to four hours postnatally. To study toxic effects of the agents to the ocular surface, we devised a grading scale rang ing from 0 to 9, with 0 representing no abnor mal reaction and 9 a purulent hyperemic con junctiva with severe eyelid edema. The grading scale was as follows: for conjunctival hyper emia, a score of 0 represented no reaction; 1, mild; 2, moderate; and 3, severe. For eyelid edema, a score of 0 represented no reaction; 1, mild; 2, moderate; and 3, severe. For ocular discharge, 0 represented no reaction; 1, watery; 2, mild purulence; and 3, severe purulence. The appearance of the conjunctiva was graded at birth before applying any agent and then again 24 hours after birth. The Student f and Wilcoxon rank sum tests were used to compare the change induced in bacterial colonies and species and to evaluate the toxicity scores for each agent at each assess ment.
Povidone-iodine for Ophthalmia Neonatorum Prophylaxis
Vol. 118, No. 6
703
TABLE 2 EFFECT OF EACH AGENT ON THE NUMBER OF SPECIES CULTURED PER EYE (MEAN ± S.D.)
Results Of the 100 newborns studied, 51 were female. All 100 neonates received povidone-iodine in one eye. In the other eye, erythromycin 0.5% ophthalmic ointment was placed in 48 infants, and silver nitrate 1% solution in 52. The gestational ages were the same for the erythromycin group (39.8 ± 1 . 6 weeks) and the silver nitrate group (39.8 ± 1.9 weeks). Table 1 shows the reduction in colonyforming units engendered by each of the agents. Whereas all three agents significantly decreased the number of colony-forming units, povidone-iodine was the most effective; t val ues for povidone-iodine, silver nitrate, and erythromycin were 3.51, 2.76, and 2.03, respec tively. Erythromycin reduced the count by an amount that was barely significant (P = .046). The reduction in number of species cultured per eye is displayed in Table 2. Povidone-iodine reduced the number of species more than the other two medications. Erythromycin did not induce a significant reduction in the number of species per eye. By the Wilcoxon rank sum test, povidone-iodine caused a greater reduction in the number of colony-forming units (P < .05) and species (P < .002) than did erythromycin. Figure 1 demonstrates the percent reduction in the colony-forming units and number of species for each of the medications. Table 3 illustrates the toxicity caused to the anterior segment of the eye by each of the agents. By the rank sum test, silver nitrate had a significantly worse score 24 hours after birth than did the other two agents (P < .001). Figure 2 shows the frequency of significant toxic reac tions caused by each medication 24 hours after birth. At each level, silver nitrate was found to be the most frequent causative agent.
NO. OF SPECIES CULTURED
MEDICATION
AT BIRTH
TWO TO FOUR HOURS POSTNATALLY
P VALUE
Silver nitrate Erythromycin Povidone-iodine
0.96 ± 0.94 0.85 ± 0.87 1.1 ±1.1
0.58 ± 0.69 0.66 ± 0.83 0.60 ± 0.70
.021 .28* .00016
*Not significant.
Discussion Although our studies of povidone-iodine ap plied to the eye as part of preoperative prepara tion found it to be nontoxic and highly antibac terial, the possible use of povidone-iodine for ophthalmia neonatorum prophylaxis stimu lates other considerations. The conjunctival flo ra of the neonate is different from that of the adult. Isenberg and associates 10 investigated the bacterial flora within 15 minutes after birth and before any topical medications were applied. We found the largest group of bacteria to be microaerophilic, predominantly Lactobacillus species. The second largest group were true anaerobes, such as Bifidobacterium, Bacteroides, and Propionibacterium species. The smallest group were aerobic bacteria. The conjunctival bacteria of infants born by vaginal delivery reflect the bacterial flora of the vagina. In a subsequent investigation, we found the con junctival cultures of infants delivered by cesarean section within three hours of membrane
TABLE 1 EFFECT OF EACH AGENT ON THE NUMBER OF COLONY-FORMING UNITS PER EYE (MEAN ± S.D.) NO. OF COLONY-FORMING UNITS
MEDICATION
AT BIRTH
TWO TO FOUR HOURS POSTNATALLY
P VALUE
Silver nitrate Erythromycin Povidone-iodine
213 ± 517 146 ± 400 175 ± 425
14 ± 53 25 ± 107 22 ± 84
.007 .046 .00051
Povidone-iodine
Erythromycin
Silver-nitrate
Fig. 1 (Isenberg and associates). Percent reduction in the number per eye of colony-forming units and species induced by each of the medications.
704
December, 1994
AMERICAN JOURNAL OF OPHTHALMOLOGY
TABLE 3 TOXICITY SCORE* INDUCED BY EACH MEDICATION (MEAN ± S.D.)
25 20
MEDICATION
AT BIRTH«
24 HOURS POSTNATALLY
Silver nitrate Erythromycin Povidone-iodine
3.1 ± 1.3 3.1 ± 1.2 3.1 ± 1.2
3.4 ± 1.8* 3.0 ± 1.4 3.2 ± 1.6
"See text. 'Before any ocular medication. ♦Higher score than the other two 24-hour postnatal values (P < .001 by Wilcoxon rank sum test).
rupture to be sterile in 80% of cases and the rest to display only a few cutaneous bacteria. 11 Neonates delivered by cesarean section three hours after membrane rupture displayed a greater number of conjunctival bacteria that corre sponded more with the babies born by vaginal delivery. Aside from these bacteria, other or ganisms are of concern in the course of ophthal mia neonatorum, especially N. gonorrhoeas, C. trachomatis, herpes simplex virus type II, and possibly HIV. Benevento and associates 12 stud ied the effectiveness of 5.0%, 1.0% and 0.1% solutions of povidone-iodine in vitro against the first three aforementioned organisms. They found that all three concentrations sterilized inocula of herpes simplex virus type II and four different strains of N. gonorrhoeae. However, only the two stronger concentrations were ef fective against C. trachomatis. The choice of an agent best suited to prevent ophthalmia neonatorum has been controver sial. Silver nitrate, reduced to a 1% solution to decrease the frequency of toxic conjunctivitis, has remained popular. However, as a result of a perceived ineffectiveness of silver nitrate against Chlamydia and its tendency to cause toxic conjunctivitis, other medications, such as erythromycin 0.5% and tetracycline 1.0% oint ments, have been widely accepted. 13 Yet these antibiotics are not perfect. For example, Mooney and associates 14 reported a threefold increase in ophthalmia neonatorum when they switched from silver nitrate to erythromycin. Black-Payne, Bocchini, and Cedotal 15 calculated a minimum failure rate of erythromycin pro phylaxis against chlamydial ophthalmia neona torum to be between 7.0% and 19.5%. A Staphylococccus aureus conjunctivitis outbreak that was resistant to erythromycin occurred in a nursery of infants who were given the drug for
■z. LU
o
ΠLiJ
IS 10 5 D TOXICITY SCORE
Fig. 2 (Isenberg and associates). Percent of eyes demonstrating significant toxic reactions caused by each of the medications 24 hours after birth. prophylaxis. The infections ceased after the hospital switched to silver nitrate prophylaxis. 16 In a study of almost 50,000 births, Zanoni, Isenberg, and Apt17 reported ophthalmia neo natorum in general (P < .001) and chlamydial conjunctivitis specifically (P < .02) to be more frequent after erythromycin than silver nitrate prophylaxis. In the United States, gonococcal infections that are resistant to tetracycline are on the rise. 1819 The same trend has been ob served in other countries, such as the United Kingdom and The Netherlands. 20,21 Povidone-iodine has a number of potential advantages over the agents presently used to prevent ophthalmia neonatorum. In 1984, Apt and associates 2 studied the effect of applying a 5% solution of povidone-iodine to the con junctiva as part of the preoperative prepara tion. This agent reduced the colony count by 9 1 % andthespeciescountby 50%. In 1985, Isen berg and associates 3 described that, in combina tion with the outpatient preoperative use of a topical broad spectrum antibiotic preparation (gramicidin-neomycin-polymyxin B), povidoneiodine applied on the operating table preoperatively sterilized 83% of the conjunctivae. The broad antibacterial spectrum of povidoneiodine was statistically demonstrated in these studies. Subsequently, Speaker and Menikoff4 found the preoperative use of topical 5% povidone-iodine solution reduced the rate of postoperative bacterial endophthalmitis from 0.24% to 0.06%. In our previous studies involv ing povidone-iodine (as well as our ongoing use of it), we have seen no toxic reactions to the one-time application of povidone-iodine as part of the preoperative preparation. In the aforementioned study by Speaker and Meni koff,4 no toxic reactions resulted after more
Vol. 118, No. 6
Povidone-iodine for Ophthalmia Neonatorum Prophylaxis
t h a n 3,000 a p p l i c a t i o n s . It h a s a b r o a d e r spec t r u m that i n c l u d e s fungi, v i r u s e s , a n d all b a c teria. In contrast to silver nitrate a n d e r y t h r o m y c i n , p o v i d o n e - i o d i n e has not b e e n associated w i t h true bacterial resistance. 2 2 T h u s , t h e r e n e e d be little fear that resistant bacteria m i g h t affect the eye a n d possibly s p r e a d w i t h i n a n u r s e r y or h o m e situation, as occurred w i t h e r y t h r o m y cin. 15 The c u r r e n t study h a s s h o w n p o v i d o n e iodine to be highly effective a g a i n s t the bacteri al flora found on the eye at b i r t h . W h e r e a s all t h r e e a g e n t s r e d u c e d the n u m b e r of colonyforming u n i t s , p o v i d o n e - i o d i n e a c h i e v e d the best level of statistical significance a n d e r y t h r o mycin the worst. With r e g a r d to the r e d u c t i o n of the n u m b e r of species, p o v i d o n e - i o d i n e a g a i n achieved the best level of statistical signifi cance. Surprisingly, e r y t h r o m y c i n did not re duce the species count significantly. O u r data indicate that e r y t h r o m y c i n is n o t as effective in e l i m i n a t i n g bacterial flora of the n e o n a t e as the other two a g e n t s . As d e m o n s t r a t e d in this s t u d y , p o v i d o n e iodine is relatively nontoxic w h e n a p p l i e d to the sensitive eyes of n e o n a t e s . This lack of toxicity was a c c o m p l i s h e d by r e d u c i n g the con c e n t r a t i o n from the 5 % s o l u t i o n we have u s e d in the eyes of a d u l t s a n d c h i l d r e n to a 2 . 5 % solution. This milder c o n c e n t r a t i o n , n e v e r t h e less, was still effective against bacteria, as evi d e n c e d from our data. C o n v e r s e l y , silver n i t r a t e p r o d u c e d a toxic conjunctivitis m o r e frequently a n d of greater severity t h a n the o t h e r two a g e n t s . At the 2 4 - h o u r p o s t n a t a l d e t e r m i n a t i o n point, silver nitrate p r o d u c e d a toxicity score of 7 or more (generally i n d i c a t i n g m o d e r a t e to severe conjunctival h y p e r e m i a , eyelid e d e m a , and p u r u l e n t discharge) in 1 0 % of all cases. At the same e x a m i n a t i o n , the other two a g e n t s were found to p r o d u c e a similar r e s p o n s e only 1% to 2 % of the t i m e . Because silver nitrate o p h t h a l m i c s o l u t i o n a n d e r y t h r o m y c i n o i n t m e n t are t r a n s l u c e n t on the eye, it is s o m e t i m e s difficult to a p p r e c i a t e that these m e d i c a t i o n s have b e e n d e l i v e r e d . It is possible that a n u r s e may believe t h a t h e or she has a d m i n i s t e r e d the m e d i c a t i o n w h e n it actu ally may never have c o n t a c t e d the conjunctiva. However, p o v i d o n e - i o d i n e a p p e a r s dark b r o w n on the eye for a b o u t two m i n u t e s . This w o u l d e n s u r e its p r o p e r a p p l i c a t i o n a n d d e c r e a s e the possibility that an infant m i g h t go w i t h o u t prophylaxis.
705
The use of p o v i d o n e - i o d i n e offers a d v a n t a g e s in the a r e a s of availability a n d cost. Even in underdeveloped countries, povidone-iodine o p h t h a l m i c s o l u t i o n has b e e n r e c o m m e n d e d , a n d is b e i n g u s e d , as an a n t i - i n f e c t i o u s agent. 8 This study h a s s h o w n that a 2 . 5 % o p h t h a l m i c s o l u t i o n of p o v i d o n e - i o d i n e is an effective a n t i bacterial a n d a relatively n o n t o x i c a g e n t w h e n placed on the conjunctiva of a n e o n a t e . For this p u r p o s e , it h a s a d v a n t a g e s over the two most frequently u s e d p r o p h y l a c t i c m e d i c a t i o n s . This investigation, h o w e v e r , d o e s not a d d r e s s the effectiveness of p o v i d o n e - i o d i n e as a p r o p h y lactic agent specifically against hi. gonorrhoeae, C. trachomatis, h e r p e s simplex virus t y p e II, or HIV. To date, such effectiveness h a s only b e e n d e m o n s t r a t e d in vitro. To s t u d y this issue, a prospective clinical trial i n v e s t i g a t i n g the fre q u e n c y a n d type of o p h t h a l m i a n e o n a t o r u m arising after use of p o v i d o n e - i o d i n e a n d other m e d i c a t i o n s w o u l d be very informative. ACKNOWLEDGMENT
N a n c y Berman, P h . D . , D e p a r t m e n t of Pe diatrics, Harbor/UCLA Medical Center, UCLA School of M e d i c i n e , p r o v i d e d statistical analysis.
References 1. Bell TA, Grayston JT, Krohn MA, Kronmal RA, Eye Prophylaxis Study Group. Randomized trial of silver nitrate, erythromycin, and no eye prophylaxis for the prevention of conjunctivitis among newborns not at risk for gonococcal ophthalmitis. Pediatrics 1993;92:755-60. 2. Apt L, Isenberg S, Yoshimori R, Paez JH. Chemical preparation of the eye in ophthalmic sur gery, III: effect of povidone-iodine on the conjuncti va. Arch Ophthalmol 1984;102:728-9. 3. Isenberg SJ, Apt L, Yoshimori R, Khwarg S. Chemical preparation of the eye in ophthalmic surgery, IV: comparison of povidone-iodine on the conjunctiva with a prophylactic antibiotic. Arch Ophthalmol 1985;103:1340-2. 4. Speaker MG, Menikoff JA. Prophylaxis of endophthalmitis with topical povidone-iodine. Oph thalmology 1991;98:1769-75. 5. White JH, Stephens GM, Cinotti AA. The use of povidone-iodine for treatment of fungi in rabbit eyes. Ann Ophthalmol 1972;4:855-6. 6. Thomas BJ. The inactivation of Chlamydia tra chomatis by Betadine solution. In: Reber H, editor. The proceedings of the World Congress on Antisep sis. New York: HP Publishing Co, 1978:16-8.
706
AMERICAN JOURNAL OF OPHTHALMOLOGY
7. Durno AG, Kaplan JC, Schooley RT. Inactivation of human immune deficiency virus by povidone-iodine. In: Digenis GA, Agha BJ, Ansell J, Blecher L, editors. Proceedings of the Second Inter national Symposium on Povidone. Lexington: Uni versity of Kentucky Press, 1987:152-6. 8. Taylor J. Appropriate eye drugs for developing countries. Int Ophthalmol 1990;14:173-9. 9. Hardberger R, Hanna C, Boyd CM. Effects of drug vehicles on ocular contact time. Arch Ophthal mol 1975;93:42-5. 10. Isenberg SJ, Apt L, Yoshimori R, Alvarez SR. Bacterial flora of the conjunctiva at birth. J Pediatr Ophthalmol Strabismus 1986;23:284-6. 11. Isenberg SJ, Apt L, Yoshimori R, McCarty JW, Alvarez SR. Source of the conjunctival bacterial flora at birth and implications for ophthalmia neonatorum prophylaxis. Am J Ophthalmol 1988;106:458-62. 12. Benevento WJ, Murray P, Reed CA, Pepose JS. The sensitivity of Neisseria gonorrhoeae, Chlamydia trachomatis, and herpes simplex type II to disinfec tion with povidone-iodine. Am J Ophthalmol 1990; 109:329-33. 13. Dinsmoor MJ. Ophthalmia neonatorum. Contemp OB/GYN 1992;37:112-4. 14. Mooney BR, Green JA, Epstein BJ, Hashisaki PA Non-gonoccocal ophthalmitis associated with erythromycin ointment prophylaxis of gonococcal ophthalmia neonatorum. Infect Control 1984,5:13840. 15. Black-Payne C, Bocchini JA Jr, Cedotal C. Fail ure of erythromycin ointment for postnatal ocular
December, 1994
prophylaxis of chlamydial conjunctivitis. Pediatr In fect Dis J 1989;8:491-5. 16. Hedberg K, Ristinen TL, Soler JT, White KE, Hedberg CW, Osterholm MT, et al. Outbreak of erythromycin-resistant staphylococcal conjunctivitis in a newborn nursery. Pediatr Infect Dis J 1990; 9:268-73. 17. Zanoni D, Isenberg SJ, Apt L. A comparison of silver nitrate with erythromycin for prophylaxis against ophthalmia neonatorum. Clin Pediatr (Phila) 1992;31:295-8. 18. Schwarcz SK, Zenilman JM, Schnell D, Knapp JS, Hook EW, Thompson S, et al. National surveil lance of antimicrobial resistance in Neisseria gonorrhoeae. The Gonococcal Isolate Surveillance Project. JAMA 1990;264:1413-7. 19. Knapp JS, Zenilman JM, Biddle JW, Perkins GH, DeWitt WE, Thomas ML, et al. Frequency and distribution in the United States of strains of Neisse ria gonorrhoeae with plasmid-mediated, high-level resistance to tetracycline. J Infect Dis 1987;155:81922. 20. Ison CA, Terry P, Bindayna K, Gill MJ, Adams J, Woodford N. Tetracycline-resistant gonococci in UK [letter]. Lancet 1988;1:651-2. 21. van Klingeren B, Dessens-Kroon M, Verheuvel M. Increased tetracycline resistance in gonococci in The Netherlands [letter]. Lancet 1989,2:1278. 22. Houang ET, Gilmore OJA, Reid C, Shaw EJ. Absence of bacterial resistance to povidoneiodine. J Clin Pathol 1976;29:752-5.