|
|
| |||||||||||||||||||||||||||||||||||||||||
Background Neonatal conjunctivitis (ophthalmia neonatorum) continues to cause blindness, because the agents used prophylactically to prevent this condition are not completely effective and are not widely available in many parts of the world. Povidone–iodine ophthalmic solution is an effective antibacterial agent with broad antibacterial and antiviral activity to which no bacteria are known to be resistant, and it is far less expensive and less toxic than the agents currently used to prevent neonatal conjunctivitis.
Methods We conducted a masked, prospective trial involving 3117 infants born over a period of 30 months in a hospital in Kenya. Shortly after birth each infant received a 2.5 percent solution of povidone–iodine (n = 1076), a 1 percent solution of silver nitrate (n = 929), or 0.5 percent erythromycin ointment (n = 1112) in both eyes. Randomization was achieved by rotating the three medications after each was used for a week.
Results Of the neonates treated with povidone–iodine, 13.1 percent had infectious conjunctivitis, as compared with 17.5 percent of those treated with silver nitrate (P<0.001) and 15.2 percent of those treated with erythromycin (P = 0.01). Povidone–iodine was more effective against Chlamydia trachomatis than was silver nitrate (P<0.001) or erythromycin (P = 0.008). There were 104 cases of noninfectious conjunctivitis (9.7 percent) in the povidone–iodine group, as compared with 129 in the silver nitrate group (13.9 percent, P<0.001) and 148 in the erythromycin group (13.3 percent, P = 0.004). Many cases of noninfectious conjunctivitis were probably due to a toxic reaction to the treatment itself. The incidence of Neisseria gonorrhoeae and Staphylococcus aureus infections was similar in the three groups.
Conclusions A 2.5 percent ophthalmic solution of povidone–iodine as prophylaxis against ophthalmia neonatorum is more effective than treatment with silver nitrate or erythromycin, and it is less toxic and costs less.
Ophthalmia neonatorum was the main cause of childhood blindness in 19th-century Europe, but treatment with an instillation of silver nitrate solution on the conjunctiva at birth reduced the incidence by a factor of 20 to 30.3 This innovation may be the single greatest advance toward preventing blindness. Many centers have switched from silver nitrate to erythromycin or tetracycline preparations because of their allegedly superior activity against Chlamydia trachomatis and their lower toxicity.4 Today, some form of prophylaxis is legally required in the United States and in many other countries.
Povidone–iodine has many potential advantages over the two currently used drugs, including a broader antibacterial spectrum. In a concentration as low as 0.1 percent, povidone–iodine is effective against Neisseria gonorrhoeae; in a concentration as low as 1 percent it is effective against C. trachomatis5; and in a concentration of 0.5 percent or lower its antiviral spectrum includes the human immunodeficiency virus and herpes simplex virus.5,6 Povidone–iodine turns the surface of the eye brown for a few minutes, a characteristic that can serve as an indicator that it has been properly applied. The possibility of misapplication is greater with the other two agents because they are colorless. With povidone–iodine, unlike antibiotics, bacterial resistance has not been encountered. Finally, it is cheaper than the other agents (Table 1). In many developing countries, where ophthalmia neonatorum is most common, no prophylaxis is used, mainly because of the expense and lack of availability.
|
Methods
Patients
The protocol was reviewed and approved by the Human Subjects Committee of the Harbor–UCLA Medical Center and the Ministry of Health of Kenya. All babies born at the Presbyterian Church Hospital in Kikuyu, Kenya, from March 1991 through August 1993 were candidates for the study. Infants were excluded if they had any obvious ocular malformations, the mother had received antibiotics during the last month of pregnancy, or the mother was unable to bring the infant back to the hospital in the event that conjunctivitis developed. Verbal consent was obtained from each child's mother.9
All infants in the study received one of three ophthalmic preparations in both eyes within 20 minutes of birth, after the eyes and face were cleaned: a drop of 2.5 percent povidone–iodine solution, a 1-cm strip of 0.5 percent erythromycin ophthalmic ointment, or a drop of 1 percent silver nitrate ophthalmic solution. The three medications were rotated after each was used for a week in the maternity unit. Thus, each infant was assigned to receive a drug according to the week of birth. This allocation system was easier to use in a busy maternity unit in a developing country, resulting in less confusion and possible misallocation, than a strict randomization of infants. Each mother was shown pictures of inflamed eyes and instructed to return to the hospital with her infant if the child's eye began to have a discharge or became red within a month of birth. The instructions were the same regardless of the allocation group.
Specimen Collection
Infants returning with conjunctivitis were taken to the clinical laboratory of the Nairobi Hospital for microbiologic analysis of the inflamed eyes. Because the cultures were plated at the laboratory, no transport mediums were necessary. Conjunctival specimens obtained for detection of bacteria were inoculated immediately onto plates containing blood and Thayer–Martin medium, the inocula were streaked across the plates to promote isolation of the bacteria, and the plates were incubated in an atmosphere of 5 to 10 percent carbon dioxide at 35°C for 24 to 48 hours. Cultured organisms were identified by standard procedures. Two smears were made, stained with Gram's and Giemsa stains, and observed for C. trachomatis inclusions and bacteria. For direct fluorescent-antibody assay for C. trachomatis, specimens were collected, fixed, stained with fluorescein-conjugated monoclonal antibody (Syva Microtrak, Palo Alto, Calif.), and observed for elementary bodies. The laboratory technicians had no knowledge of which prophylactic ocular medication was given to any particular infant.
Statistical Analysis
Logistic regression was used to determine the effect of the medications on the risk of infection. The infants with conjunctivitis were grouped according to the results of the bacteriologic culture and direct fluorescent-antibody assay as having infection with C. trachomatis, all bacteria under study, coagulase-negative staphylococci, Staphylococcus aureus, or N. gonorrhoeae. The infants were considered to have noninfectious conjunctivitis if no organisms were cultured. Each group was compared with the group of infants with no inflammation. The model included sex and the presence or absence of prenatal care, maternal vaginal infection, and birth in an unhygienic environment (indicated by the presence of meconium, among other factors) as covariates.
Results
The characteristics of the 3117 newborns according to treatment group are presented in Table 2. The risk factors were distributed evenly across the groups. Table 3 shows the distribution and percentages of infants with infectious and noninfectious conjunctivitis. The most frequently identified organism was C. trachomatis (found in 50.5 percent of all infected infants), and S. aureus was the second most common (found in 39.7 percent of all infected infants). N. gonorrhoeae was cultured in 5 percent of all infected infants. Table 4 shows the distribution of the risk factors according to the development of conjunctivitis.
|
|
|
|
Fewer infections with coagulase-negative staphylococci were seen in the povidone–iodine group than in the erythromycin group (P = 0.01), but the rate was virtually identical to that in the group given silver nitrate. With respect to infections caused by S. aureus or N. gonorrhoeae, no difference in effectiveness was found among the three agents. The povidone–iodine group had fewer cases of noninfectious conjunctivitis than either the silver nitrate group (P<0.001) or the erythromycin group (P = 0.004).
Babies whose mothers had infections had more bacterial infections overall (P = 0.009), as shown in Table 5, and more gonococcal infections in particular (P<0.001). Female infants had marginally more S. aureus infections than male infants (P = 0.045). No significant difference in the frequency or type of infection was seen whether the infant was delivered vaginally or by cesarean section. Infants born in an unhygienic environment had a greater incidence of infections with coagulase-negative staphylococci than those born in a hygienic environment (P = 0.002).
Discussion
We undertook this test of the effectiveness of povidone–iodine as prophylaxis against ophthalmia neonatorum for two reasons. First, there is an increasing recognition of the inadequacy of the commonly used agents. When used prophylactically, topical erythromycin has a failure rate of 7 to 19.5 percent.10 Ophthalmia neonatorum has been reported more often after erythromycin prophylaxis than after silver nitrate treatment11 and more often after silver nitrate treatment than after tetracycline treatment (P<0.05).12 Prophylaxis with erythromycin has resulted in outbreaks of erythromycin-resistant staphylococcal conjunctivitis in neonates.13 Since the most dreaded pathogen of ophthalmia neonatorum is N. gonorrhoeae, the many reports of tetracycline resistance from such countries as the United Kingdom,14 the Netherlands,15 and the United States16 are alarming. Tetracycline is no longer recommended as first-line therapy for gonococcal infections.17 Ophthalmia neonatorum has occurred after the use of silver nitrate.18,19 In addition, silver nitrate causes a serious toxic conjunctivitis, which may impede infant–mother bonding,20 far more frequently than erythromycin or povidone–iodine.7
The second reason to study povidone–iodine as prophylaxis against ophthalmia neonatorum is that its antibacterial spectrum exceeds that of any of the other agents. True bacterial resistance to povidone–iodine, unlike the other agents, has not been demonstrated.21 In a controlled clinical trial of povidone–iodine given before ocular surgery, we showed that the application of a 5 percent solution reduced the number of bacterial colony-forming units by 91 percent and the number of species by 50 percent.22 Subsequently, Speaker and Menikoff, using this ophthalmic solution preoperatively in 3489 adults undergoing cataract surgery, found that the incidence of postoperative endophthalmitis was significantly reduced.23 In none of the thousands of patients treated by us and mentioned in the report of Speaker and Menikoff did this solution cause a discernible toxic or allergic reaction.
In a pilot study, we determined that a 2.5 percent solution of povidone–iodine was not irritating to the sensitive eyes of neonates, whereas the 5.0 percent solution that we had used in previous studies occasionally produced some conjunctival hyperemia. We then studied the antibacterial effect of a 2.5 percent solution of povidone–iodine on the eyes of 100 newborns in California.7 Although both povidone–iodine and silver nitrate significantly reduced the number of both bacterial colony-forming units and species, povidone–iodine caused a greater decrease.7 As compared with erythromycin, povidone–iodine caused a greater reduction in the number of colony-forming units (P<0.05) and in the number of bacterial species (P<0.002). Twenty-four hours after birth, there were more serious toxic reactions to silver nitrate than to povidone–iodine (P<0.001). Encouraged by these results, we decided to use the 2.5 percent concentration in the Kenyan trial.
Povidone–iodine is also active against viruses, including herpes simplex. Herpetic keratoconjunctivitis can be an insidious infection. Fifty to 70 percent of neonates with herpes simplex are born to mothers with no history of genital herpes infection.24 Although only 2 percent of cases of ophthalmia neonatorum are caused by herpes,25 17 percent of newborns with herpes simplex virus will have ocular involvement.26 Benevento and associates showed povidone–iodine to be effective in a concentration as low as 0.1 percent against a challenge of 10 million plaque-forming units of herpes simplex virus type II.5
A final but important factor favoring the use of povidone–iodine is cost. Many hospitals in developing countries cannot afford to purchase the currently used medications, but povidone–iodine ophthalmic solution can be manufactured from powder available to pharmacies worldwide. The price of povidone–iodine, at least in Kenya, ranges from 1.4 percent to 30 percent of the cost of the other agents (Table 1). Producing the solution locally minimizes cost and maximizes the possibility that the newborns will receive proper prophylaxis.
Despite the use of prophylactic agents, the incidence of ophthalmia neonatorum in the present study was 13.1 percent, 15.2 percent, and 17.6 percent after prophylaxis with povidone–iodine, erythromycin, and silver nitrate, respectively. Although high for the developed world, where the incidence generally is less than 2 percent, these rates are not high for developing countries. After silver nitrate prophylaxis, the incidence of ophthalmia neonatorum was 10.5 percent in Gabon27 and 19.4 percent in Cameroon.28 Our high rate of positive cultures may also result from our methods. In many previous studies specimens were transported to the laboratory for analysis, whereas in our study all specimens and cultures were collected and processed at the laboratory.
From the 23 percent incidence of ophthalmia neonatorum reported in a Kenyan population similar to the one we studied when prophylaxis was not used,8 it can be estimated that the incidence of ophthalmia neonatorum is reduced 44 percent by prophylaxis with povidone–iodine, 34 percent by prophylaxis with erythromycin, and 24 percent by prophylaxis with silver nitrate. A recent American study found that, as compared with no treatment, erythromycin decreased the incidence of ophthalmia neonatorum by 31 percent and silver nitrate decreased it by 39 percent,29 whereas a Taiwanese study found that silver nitrate reduced the incidence by 31 percent (P<0.05) and erythromycin did not cause a significant reduction.30
The most feared cause of ophthalmia neonatorum is N. gonorrhoeae, because it can quickly cause blindness.31 Laga and colleagues reported that the incidence of gonococcal and chlamydial infections in Kenya was 2.8 percent and 7.3 percent, respectively, when no prophylaxis was used.8 In our study, all three agents were equally effective against gonococcus; the frequency of gonococcal infections in the three groups ranged from 0.43 percent to 0.99 percent. The second most feared microorganism, C. trachomatis, may decrease vision as a result of corneal and conjunctival scarring, vascularization,32 and the formation of pseudomembranes.33 Significantly fewer cases of C. trachomatis infection occurred in infants receiving povidone–iodine than in those receiving either silver nitrate (P<0.001) or erythromycin (P = 0.008). This benefit is especially important because the incidence of chlamydial infections is escalating.34 Erythromycin and silver nitrate were not superior to povidone–iodine against any other bacteria encountered in the study.
By convention, ophthalmia neonatorum is defined as conjunctivitis arising within one month after birth. Thus, an infection could have originated at home within a few weeks after the infant was discharged from the hospital and still be considered ophthalmia neonatorum. This may explain why there was no difference among the three drugs in the proportion of infections produced by S. aureus (about 6 percent). Krohn and colleagues found that some apparent cases of ophthalmia neonatorum were actually acquired from the infants' nasopharyngeal passages or from their care givers after birth.35
Of the 3117 infants we studied, 12.2 percent had signs of ocular inflammation, but no growth on cultures and a negative chlamydial antibody test. Toxic conjunctivitis, a reaction to the prophylactic medication, is generally seen within 24 hours of the instillation of the drug and resolves without treatment. We could not distinguish between cases of true infection with no bacterial growth and cases of toxic conjunctivitis. The povidone–iodine group had fewer infants with no evidence of any organism than did the silver nitrate group (P<0.001) or the erythromycin group (P = 0.004). Since in infants the ophthalmic solution of povidone–iodine is less toxic than silver nitrate (P<0.001),7 we believe that many of the cases of noninfectious conjunctivitis represent toxic reactions.
Consistent with the report of Laga and associates,8 the presence of maternal infection predisposed a baby to the development of conjunctivitis in general (P = 0.009) and gonococcal conjunctivitis in particular (P<0.001). A lack of prenatal care was also associated with an increased incidence of conjunctivitis (P<0.001). Thus, health professionals should be especially cautious to administer prophylaxis properly to such babies and to those born in unhygienic environments, because of the greater incidence of infections with coagulase-negative staphylococci (P = 0.002) in these situations.
In conclusion, prophylaxis with a 2.5 percent ophthalmic solution of povidone–iodine resulted in fewer cases of ophthalmia neonatorum overall and, notably, fewer cases of chlamydial conjunctivitis than did prophylactic treatment with either erythromycin or silver nitrate. In contrast to other prophylactic agents, which are colorless, povidone–iodine turns the surface of the eye brown, which indicates that the medication has been properly administered. Povidone–iodine is less expensive than the other drugs and can be prepared locally. For these reasons, povidone–iodine deserves serious consideration as prophylaxis against ophthalmia neonatorum, especially in developing countries.
Supported by the Karl Kirchgessner Foundation.
We are indebted to Sisters Silpa Odero and Margaret Wandia Kimonye as well as the staff members of the eye and maternity units of the Presbyterian Church Hospital (Kikuyu, Kenya) for data collection; to Roberta Rich, R.N., for data entry and processing; and to Dr. Nancy Berman for statistical analysis.
Source Information
From the Department of Ophthalmology, Jules Stein Eye Institute, Harbor–UCLA Medical Center, UCLA School of Medicine, Los Angeles and Torrance, Calif. (S.J.I., L.A.); and the Eye Department, Presbyterian Church Hospital, Kikuyu, Kenya (M.W.).
Address reprint requests to Dr. Isenberg at the Jules Stein Eye Institute, UCLA School of Medicine, 100 Stein Plaza, Los Angeles, CA 90024-7001.
References
| |||||||||||||||||||||||||||||||||||||||||
Related Letters:
Povidone-Iodine to Prevent Ophthalmia Neonatorum
Rotta A. T., Petri A. R., Pietsch M., Barton L. L., Isenberg S. J., Apt L., Foster A.
Extract |
Full Text
N Engl J Med 1995;
333:126-127, Jul 13, 1995.
Correspondence
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
