In Reply We are happy that our systematic review1 raised important discussion on intra-arterial chemotherapy (IAC) for retinoblastoma. The development of IAC is clearly one of the most important recent advances in the management of children with retinoblastoma.
Abramson and colleagues asked why we rejected some articles based on judgment of scientific value, while including 2 patients from a social media survey.2 We excluded articles based on specified inclusion criteria rather than judgment. We included only 1 article from the Abramson group because we could not discern whether patients were duplicated across multiple publications. The patients from the social media survey were not included in our analyses or tables, but mentioned in our discussion. This is akin to postmarketing surveillance, allowable for crucially important outcomes such as death. There is no reason for us to believe that these deaths were untrue. However, they are not yet reported in any primary publication, supporting our statement that “retrospective studies are invariably subject to some selection and detection bias, risking underdetection.”1