Vitamin E prophylaxis to reduce retinopathy of prematurity: A reappraisal of published trials

• Published in: The Journal of pediatrics 1997-12, Vol.131 (6), p.844-850
• Main Authors: Raju, Tonse N.K., Langenberg, Patricia, Bhutani, Vinod, Quinn, Graham E.
• Format: Article
• Language: English
• Subjects: Biological and medical sciences; Confidence Intervals; Diseases of the eye; Humans; Infant, Newborn; Infant, Very Low Birth Weight; Medical sciences; Odds Ratio; Radiotherapy. Instrumental treatment. Physiotherapy. Reeducation. Rehabilitation, orthophony, crenotherapy. Diet therapy and various other treatments (general aspects); Randomized Controlled Trials as Topic; Research Design; Retinopathy of Prematurity - prevention & control; Sample Size; Treatment Outcome; Vitamin E - administration & dosage; Vitamin E - blood
• ISSN: 0022-3476; 1097-6833
• DOI: 10.1016/S0022-3476(97)70031-3

Objective: We conducted a meta-analysis of the published randomized clinical trials of vitamin E prophylaxis designed to reduce retinopathy of prematurity (ROP) to determine whether increased serum concentrations of vitamin E reduced the incidence of severe, threshold (Stage 3+) ROP in the very low birth weight (VLBW) infant subset. Study Design: Among the six trials considered eligible for analyses, the incidence for all stages of ROP and for Stage 3+ ROP was computed for all randomly assigned infants (intention-to-treat analysis) and for those infants completing the trials. Odds ratios (ORs) and pooled estimates for event rate reductions (and the respective 95% confidence intervals [CIs]) were calculated for these outcome end points. Results: There were 704 VLBW infants in the vitamin E prophylaxis groups and 714 in control groups; 536 (76.1%) infants in the vitamin E and 551 (77.2%) in the control groups completed the trials. In all trials the mean serum vitamin E concentrations were within or above the physiologic range in the vitamin-treated groups and at or below the physiologic ranges in the control groups. The overall incidence of any stage ROP was similar between the groups: 39.8% in the vitamin E group and 43.5% in the control group. The overall incidence for Stage 3+ ROP was lower in the vitamin E–treated groups than in the control group (2.4% in vitamin E vs 5.3% in control). The pooled OR for developing Stage 3+ ROP with vitamin E prophylaxis was 0.44 (95% CI, 0.21, 0.81, p < 0.02). The reciprocal of the pooled estimate of mean risk reduction (2.8%, 95% CI: 0.55%, 5.1%) for Stage 3+ ROP revealed that on average, vitamin E prophylaxis given to 35 VLBW infants would prevent one case of threshold, Stage 3+ ROP. Conclusions: Considering that there was a 52% reduction in the incidence of Stage 3+ ROP, we suggest that the role of vitamin E in reducing severe ROP be re-evaluated. We could not assess the adverse effect rates from vitamin E therapy in the trials analyzed; therefore, we recommend a well-controlled and focused trial in which the issues of benefit, adverse effects, and cost can be assessed with vitamin E prophylaxis in extremely low birth weight (