In reply
We agree with Drs Bryant, Cave, and Milne that full identification of potentially relevant studies is needed for performance of systematic review, and our search procedures were designed and implemented for this purpose. The letter by Bryant et al seems to assume that the 3 reports cited were not identified by our search; however, all 3 articles were identified. They were not included in the meta-analysis because they did not meet our prespecified inclusion criteria. As described, each study identified was analyzed thoroughly in the review process to assess whether it met our rigorous inclusion criteria. The article by Cowell et al was not included because the sample incorporated children with small birth size and because baseline data for each study group were not provided. The study of Ackland et al did not provide outcome data for growth parameters required in our "Materials and Methods" section. The Barton et al study was excluded because it did not have the specified GH levels required by our inclusion criteria. In addition, studies that do not provide or make data available in a format that can be used in meta-analysis cannot be included.