PURPOSE: To evaluate which parameters may affect endothelial cell loss after Descemet membrane endothelial keratoplasty (DMEK) by comparing eyes in the low vs high quartile of endothelial cell loss... Show morePURPOSE: To evaluate which parameters may affect endothelial cell loss after Descemet membrane endothelial keratoplasty (DMEK) by comparing eyes in the low vs high quartile of endothelial cell loss over a follow-up period of 4 years.DESIGN: Retrospective cohort study.METHODS: Donor endothelial cell density (ECD) decline was evaluated for 351 eyes of 275 patients up to 4 years after DMEK for Fuchs endothelial corneal dystrophy (FECD). Eyes with a postoperative endothelial cell loss in the lower quartile at all available follow-up moments were assigned to Group 1 (n = 51) and those in the upper quartile to Group 2 (n = 42). Multinomial regression was used to assess which covariates were related to greater ECD decline.RESULTS: Mean endothelial cell loss as compared to preoperative donor ECD for the entire study group was 33 (+/- 16)%, 36 (+/- 17)%, and 52 (+/- 18)% at 1, 6, and 48 months postoperatively. Endothelial cell loss of Group 1 was 12 (+/- 7)%, 13 (6)%, and 26 (+/- 8)% at, respectively, 1, 6, and 48 months postoperatively, and 59 (+/- 10)%, 64 (+/- 9)%, and 75 (+/- 5)% in Group 2. Partial graft detachment, donor death cause cardiovascular/stroke (vs cancer), postoperative complications other than graft detachment, and severity of preoperative FECD (all P<.01) showed the strongest relation with greater ECD decline.CONCLUSIONS: DMEK eyes with a completely attached graft and operated in an early stage of FECD may show the lowest endothelial cell loss postoperatively. ((C) 2019 Elsevier Inc. All rights reserved.) Show less
Feature Network Models (FNM) are graphical structures that represent proximity data in a discrete space with the use of features. A statistical inference theory is introduced, based on the... Show moreFeature Network Models (FNM) are graphical structures that represent proximity data in a discrete space with the use of features. A statistical inference theory is introduced, based on the additivity properties of networks and the linear regression framework. Considering features as predictor variables leads in a natural way to a univariate multiple regression problem with positivity restrictions on the parameters, which represent edge lengths in the network representation. Theoretical standard errors and confidence intervals are obtained for the parameters and their performance is evaluated by Monte Carlo simulation. When the feature structure is not known in advance, a strategy is proposed to select an adequate subset of features that takes into account a good compromise between model fit and model complexity using Gray codes and the positive lasso. The same statistical inference theory also holds for additive trees that are special cases of FNM. Standard errors and confidence intervals, model tests and prediction error are obtained for the estimates of the branch lengths of additive trees. The dissertation concludes by demonstrating that there exists a universal network representation of city-block models based on key elements of the network representation consisting of betweenness, metric segmental additivity and internal nodes. Show less
