Sir,

The article by Gnanaraj et al1 contains inconsistencies, inaccurate statements, and misleading conclusions mandating clarification.

  • The first case reports associating retinoschisis and perimacular retinal folds with child abuse were published in 1986 and 1988, respectively.2, 3 However, Group 2 by Gnanaraj et al1 was extracted from a biased sampling of autopsies occurring between 1982 and 1989 tabulated by Gilliland et al.4 How can ocular findings be assessed when they had not yet been described in the literature?

  • According to Gnanaraj et al,1 none of the nine fatalities with putative crush head injuries in Group 2 had retinoschisis or folds, but peer-reviewed publication of these fundal findings has not occurred in any of the purported child abuse (shaking) fatalities from that 20-year-old data set.4

  • The authors misstate that crush injuries to the head are the 10th most common cause of injury in children under the age of 9 years. The cited reference lists crush injury as the 10th most common cause of injury in children under the age of 9 years (ICD-9 codes 925–929).5 This specifically excludes intracranial injury (ICD-9 codes 850.0–854.1).6

  • Gnanaraj et al1 claim that the perimacular retinal folds observed in the child from the evidence-based case report by Lantz et al7 were a bit atypical, more angulated at the apex; however, the cited reference does not discuss this subjective nuance nor was apical angulation of the retinal folds described in the case report or evident from the accompanying images.8

  • Gnanaraj et al1 state that retinal haemorrhages are well recognized although uncommon in accidental major head trauma citing a 1992 study.9 This article has been previously identified as exhibiting selection bias based on the relative minor head trauma sustained in the accident group.7 Similar systematic error is obvious in the selection of patients (Group 1) by Gnanaraj et al1 when compared to previously published studies characterizing the morbidity and mortality of head injuries associated with falling televisions.10, 11, 12 Remarkably, four of the nine children (44.4%) with accidental head injuries in Group 2 by Gnanaraj et al1 had retinal haemorrhages.

All too often, the human tendency is to embrace repetitious assertions that reinforce authoritative opinions, but trivialize or reject new findings that do not support entrenched beliefs.13 The perpetuated claim that retinoschisis and perimacular retinal folds in children are created by vitreoretinal traction during violent shaking resides in faith not in science.