Interventions for preventing ophthalmia neonatorum
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
The objectives of this review are:
(1) to determine if any type of ophthalmic prophylaxis reduces the incidence of conjunctivitis in neonates.
(2) to determine which ophthalmic prophylactic medication is most effective at reducing the incidence of conjunctivitis in neonates.
Background
Ophthalmia neonatorum, also called neonatal conjunctivitis, is an inflammatory disorder of the ocular surface of neonates caused primarily by bacteria and less often by viruses and chemical agents (Albert 1994). It can lead to permanent damage of the eye and blindness.
The major pathogens responsible for ophthalmia neonatorum are Chlamydia trachomatis, Staphylococcus aureus, Neisseria gonorrhoeae, Streptococcus and Haemophilus (Deschenes 1990). The most serious type of ophthalmia neonatorum is gonococcal conjunctivitis which is caused by Neisseria gonorrhoeae (Isenberg 1994). This can lead to perforation of the intact corneal epithelium and loss of the eye (Canadian 1992). In some settings, the most common type of ophthalmia neonatorum is chlamydial conjunctivitis which is caused by Chlamydia trachomatis (Isenberg 1994).
Ophthalmia neonatorum is mainly contracted from the mother's infected birth canal during delivery, but can also be contracted in‐utero by ascending infections and postnatally by contaminated secretions from family members (Deschenes 1990; Isenberg 1994).
The World Health Organization estimates that in 1995 there were 62.2 million new cases of gonorrhoea and 89.1 million new cases of chlamydia in adults aged 15‐49 years of age globally (Gerbase 1998). Most of these new cases occurred in the developing world. Approximately 28 per cent of the infants born to women with Neisseria gonorrhoeae will develop gonococcal conjunctivitis and 18 to 50 per cent of infants born to women with Chlamydia trachomatis will develop chlamydial conjunctivitis (Canadian 1992). Hence, the worldwide potential for ocular morbidity and blindness is large (Isenberg 1995).
Prophylaxis has significantly reduced the risk of the newborn developing ophthalmia neonatorum. However, the relative efficacy of various prophylactic agents has not been resolved. Karl S.F. Crede introduced silver nitrate prophylaxis in 1881, which reduced the incidence of gonococcal conjunctivitis from 10 to 0.3 per cent (Isenberg 1994). However, the emergence of Chlamydia trachomatis as the leading cause of ophthalmia neonatorum and the ineffectiveness of silver nitrate against Chlamydia trachomatis has driven a re‐examination of the prophylactic agent (Canadian 1992; Isenberg 1994).
Erythromycin and tetracycline gained acceptance as prophylactic agents in the 1980's because of their allegedly superior activity against Chlamydia trachomatis and because they lacked some side effects of silver nitrate, such as chemical conjunctivitis (Isenberg 1994). However, it remains unresolved whether these antibiotic agents are, in fact, any more effective than silver nitrate in preventing chlamydial conjunctivitis. Further, the emergence of beta‐lactamase‐producing Neisseria gonorrhoeae has reduced the prophylactic effectiveness of erythromycin and tetracycline (Ison 1998). Recently, promising results have been found for povidone‐iodine as a prophylactic agent (Isenberg 1995; Isenberg 1994b). It has many advantages over silver nitrate, erythromycin and tetracycline, including a broader antibacterial spectrum and lack of bacterial resistance.
Currently, it remains unresolved which drug regimen to administer for ophthalmia neonatorum prophylaxis. This is partly due to the lack of studies, partly due to a lack of good quality, conclusive studies, and partly due to the lack of appropriate reviews of the existing studies (Canadian 1992).
Objectives
The objectives of this review are:
(1) to determine if any type of ophthalmic prophylaxis reduces the incidence of conjunctivitis in neonates.
(2) to determine which ophthalmic prophylactic medication is most effective at reducing the incidence of conjunctivitis in neonates.
Methods
Criteria for considering studies for this review
Types of studies
We will consider randomised and quasi‐randomised trials.
Types of participants
Participants will be all newborn infants receiving some prophylactic medication or placebo for ophthalmia neonatorum after birth.
Types of interventions
We will consider any topical, systemic or combination medication which has been compared to placebo and/or other topical, systemic or combination medication.
Types of outcome measures
The following outcomes will be included:
(1) Proportion of infants developing infectious conjunctivitis by type of organism, for example gonococcal conjunctivitis, chlamydial conjunctivitis
(2) Number of colony‐forming units per eye
(3) Number of different species of bacteria cultured per eye
(4) Time to onset of ophthalmia neonatorum
(5) Cost‐effectiveness
(6) Impairment of maternal‐infant bonding
Maternal and infant behaviours during care and during breast feeding have been assessed in some studies by observation and interviews. During observation, behaviour has been quantified by tally four to five days postpartum after administration of prophylaxis. Structured interviews have been carried out four to five days postpartum and six to eight weeks postpartum.
(7) Visual alertness
Visual alertness has been assessed by the ability of the infant to open their eyes for alertness and scanning after administration of eye prophylaxis. This has been scored according to the scale developed by Korner and Thoman (Wahlberg 1982):
(0) does not open eyes
(1) opens eyes for short moments several times
(2) opens eyes over a sustained period
(3) scans drowsily or non alertly
(4) alert with bright and shiny eyes
(5) scans with bright, alert eyes.
(8) Adverse outcomes:
(a) proportion of infants developing noninfectious conjunctivitis, for example chemical conjunctivitis
(b) incidence of nasolacrimal duct obstruction
(c) pain reaction. Assessed by measuring the length of the cry reaction (Wahlberg 1982).
(d) eye irritation including swelling of eyelids, redness of conjunctiva, secretion
(e) other severe adverse outcomes
(f) other minor adverse outcomes
(9) Compliance
Compliance includes aspects such as ease of administration of the prophylactic medication, ease of preparation of the prophylactic medication, stability of the medication, and whether the prophylactic medication marks the eye in some way (for example, colour) to determine if the medication has been administered.
Search methods for identification of studies
Electronic searches
We will identify studies by searching the Cochrane Controlled Trials Register ‐ CENTRAL (which includes the Cochrane Eyes and Vision Group specialised register) and MEDLINE.
See: Appendices for details of search strategies for the electronic databases.
Searching other resources
We will use the Science Citation Index to locate studies which have cited the identified trials. We will check the reference lists of identified trial reports and existing review articles to identify additional trials. We will contact the authors of identified trials, pharmaceutical companies and experts in the area to locate further trials. Hand‐searching efforts will be undertaken to identify additional trials.
Data collection and analysis
Selection of trials
Three reviewers will assess the titles and abstracts resulting from the electronic searches. The full copy of all relevant or potentially relevant trials will be obtained and assessed according to the 'Criteria for considering studies for this review'. Only trials meeting these criteria will be assessed for methodological quality. The reviewers will not be masked to any trial details during the assessment. Disagreements about whether a trial should be included will be resolved by discussion and consensus. In cases where additional information is needed before a decision can be made whether to include a trial, we will attempt to obtain this information from the study investigator.
Assessment of methodological quality
Trial quality will be assessed according to the methods set out in Section 6 of the Cochrane Reviewer's Handbook and the Cochrane Eyes and Vision Group Methodology Guide for Reviewers. This assessment will be made by at least two reviewers. We will use the following components to determine methodological quality:
(1) Allocation concealment: This is determined by the method of allocation used in the randomised trial and graded as adequate, unclear or inadequate. In this way, any selection bias can be ascertained.
(2) Performance bias: Trials will be graded based on whether the recipients of care were unaware of their assigned treatment and whether the persons providing care were unaware of the assigned therapy.
(3) Detection bias: The trials will be graded according to the awareness of the persons responsible for outcome assessment to the assigned therapy.
(4) Attrition bias: This will be assessed in the trials by comparing the rates of follow‐up in the treatment groups and by determining whether the analysis was 'intention to treat'.
A global rating will be made of the trial as 'low risk of bias', 'moderate risk of bias' or 'high risk of bias' based on the responses to the above four components. This rating will be applied according to the guidelines in the Cochrane Eyes and Vision Group Methodology Guide for Reviewers.
The independent appraisals will be compared for differences and any discrepancies will be resolved by discussion. The consensus agreement will be recorded using a separate printed form. Reliability will be examined throughout the data collection process to avoid 'coder drift'.
Data collection
Two reviewers independently will extract data onto forms developed by the Cochrane Eyes and Vision Group using Section 7 of the Cochrane Reviewer's Handbook, the Cochrane Eyes and Vision Group Methodology and Statistical Guides for Reviewers. Authors of trials will be contacted to try to obtain missing data.
Data analysis
Data analysis will be conducted using the following as guides: Section 8 of the Cochrane Reviewer's Handbook and the Cochrane Eyes and Vision Group Statistical Guide for Reviewers.
For dichotomous data, results will be expressed as odds ratio estimates or risk ratio estimates (95% confidence intervals). Also, the risk difference or the number needed to treat will be obtained (95% confidence intervals). Results will be summarised across studies using the odds ratio or relative risk and/or the risk difference.
For continuous data, the mean and standard deviation will be obtained; if the data are skewed, the median and inter quartile range will be obtained. Standard errors will be converted to standard deviations. If trial results are only reported as mean differences, investigators will be contacted to obtain the mean and standard deviation. Results will be summarised across studies using weighted mean differences (95% confidence intervals).
The effect of differences in the definitions and measurement of outcomes, such as diagnosis of conjunctivitis and infections and differences in length of follow‐up, will be examined using methods for addressing clinical heterogeneity: sub‐group analysis and meta‐regression. If any other significant heterogeneity is detected between studies, sub‐group analysis will be carried out and pooling of results will be avoided.
Sensitivity analysis will be conducted with the following adjustments:
(1) Excluding studies of lower methodological quality
(2) Excluding unpublished studies
(3) Changing the inclusion criteria:
(a) excluding quasi‐randomised trials
(b) excluding trials conducted in developed country settings
(c) excluding trials conducted in developing country settings
(4) Excluding studies which have assumed that eyes within an individual are independent
(5) Excluding studies where follow‐up has not been continued for a critical period of one month
We will perform cost‐effectiveness analysis where data permit.
