In Reply.

plex mechanisms. Women who have severe ophthalmologic findings during preeclampsia do not have associated nondiabetic traction retinal detachments or new onset retinal breaks.2 Another internal inconsistency is the finding of an increased incidence of retinal tears but not of rhegmatogenous retinal detachments. The two should track together. The authors’ division of retinal diseases to be studied into “retinal detachment, retinopathy (persistent inflammation and vascular remodeling of the retina), and other retinal disorders” does not correlate with clinically used terms. Additional concerns about the robustness of these associations include: 1) methodologic issues seen in retrospective claims studies, for example, coding errors and misclassification bias; 2) duration of follow-up could be overestimated because of a common end date rather than an actual end date, which would increase follow-up for women without preeclampsia; 3) no information on how women with multiple births (60% of total group) were treated; 4) definition of metabolic disease as occurring at any time, which could overestimate prevalence; and 5) lack of information on women treated outside of the hospital since many retinal breaks are treated in the ambulatory setting and women without preeclampsia or metabolic disease may not be hospitalized. In summary, we recommend following the current Preferred Practice Patterns3,4 until future studies provide a preponderance of evidence that preeclampsia is a risk factor for traction retinal detachment, retinal breaks, and diabetic retinopathy. We recommend that affected patients should be educated about the symptoms of retinal breaks, tears, or traction, and, if these occur, patients should be evaluated promptly.