From May 1, 2005, through March 31, 2007, 92 iris specimens from 92 unique patients were evaluated by the 3 reviewing pathologists. The specimens consisted of 22 latanoprost-treated darkened irides, 35 latanoprost-treated irides without darkening, and 35 non–latanoprost-treated irides. The 22 latanoprost-treated darkened irides included 11 cases obtained from Arranz-Marquez et al4
and 11 other cases from the LPC. Evaluation of the composite pathology grading forms indicated that the latanoprost-treated groups with or without darkening and the control group were comparable in terms of quality, location, and orientation of the iridectomy specimens. All specimens were obtained from the peripheral iris.
No melanomas were reported on the pathology grading form in either of the latanoprost-treated groups or in the control group. The iris specimens in all of the groups were evaluated for melanocytic atypia (). The various features suggestive of melanocytic atypia included intranuclear inclusions, nuclear invaginations, prominent nucleoli, and mitotic figures. No mitotic figures were noted in any of the iris specimens. There were no statistically significant differences in the number of intranuclear inclusions among the 3 groups. There was a statistically significant decrease in the number of nuclear invaginations and prominent nucleoli in latanoprost-treated darkened irides compared with the other groups (P=.004 and P=.005, respectively).
The 22 latanoprost-treated darkened irides were further examined in terms of the following 2 subgroups: 11 cases obtained from the series studied by Arranz-Marquez et al4
and 11 other cases from the LPC. There were more intranuclear inclusions in the cases obtained from Arranz-Marquez et al than in those from the LPC (mean, 2.2 vs 1.1; P
=.01). However, there were fewer nuclear invaginations (mean, 1.7 vs 3.4; P
=.01) and prominent nuclei (mean, 1.1 vs 2.7; P
=.003) in the cases obtained from Arranz-Marquez et al compared with those from the LPC.
The average (SD) thickness of the anterior border layer was greater in the latanoprost-treated darkened irides (2.2 [0.9] cells) than in the latanoprost-treated irides without darkening (1.8 [0.7] cells) and the control group (1.7 [0.7] cells) (P=.03). There was also a statistically significant increase in pigmentation of the anterior border layer in the latanoprost-treated darkened irides (21 of 22 irides [95%] with moderate or heavy pigmentation) compared with the latanoprost-treated irides without darkening (20 of 35 irides [57%] with moderate or heavy pigmentation) and the controls (21 of 34 irides [62%] with moderate or heavy pigmentation) (P=.02; ). Similarly, the latanoprost-treated darkened irides had increased pigmentation of the stroma (P<.001), stromal fibroblasts (P<.001), melanocytes (P=.005), vascular endothelium (P=.02), and adventitia (P<.001) relative to the latanoprost-treated irides without darkening and the controls ( and ). In the areas of increased pigmentation, the melanocytes were differentiated from pigmented fibroblasts by the plump, fusiform cytoplasm in melanocytes, compared with fibroblasts, which have wispy cytoplasm and narrow, spindle-shaped nuclei.
Figure 1 Hematoxylin-eosin–stained sections of iridectomy specimens showing increased thickness of the anterior border layer in latanoprost-treated darkened irides (A) compared with latanoprost-treated irides without darkening (B) and control irides (C) (more ...)
Constituent Cells Graded with Heavy (Maximum) Pigmentation According to Location and Treatment
Figure 2 Hematoxylin-eosin–stained sections of iridectomy specimens showing increased pigmentation of the stroma, stromal fibroblasts, melanocytes, vascular endothelium, and adventitia in latanoprost-treated darkened irides (A) compared with latanoprost-treated (more ...)
A subset analysis was performed for the extent of pigmentation in the various layers of iris in the latanoprost-treated darkened irides (11 cases obtained from Arranz-Marquezet al4
and 11 cases from the LPC). Arranz-Marquez et al also noted a thickening of the anterior border layer, but this was given as a measurement in millimeters rather than in terms of the number of cells. However, in the LPC cases, there was no statistically significant difference in pigmentation of the anterior border layer (11 of 11 [100%] vs 10 of 11 [91%] with heavy or moderate pigmentation; P
=.45), stroma (9[82%] vs 7[64%]; P
=.51), stromal fibroblasts (11 [100%] vs 8 [73%]; P
=.39), melanocytes (11 [100%] vs 9 [82%]; P
=.22), vascular endothelium (0 vs 0; P
>.99), and adventitia (1 [9%] vs 2 [18%]; P
A detailed evaluation of all the iris specimens also found no evidence of stromal inflammation or abnormal appearance of stromal blood vessels in any irides. The posterior iris pigment epithelium was also unremarkable in all specimens. There was no evidence of rubeosis, pseudoexfoliation material, synechiae, descemetization, or endothelialization in any of the iris specimens. Two specimens, one from the latanoprost-treated darkened irides group and the other from the control group, showed attached Descemet membrane. Three specimens, including 1 latanoprost-treated iris without darkening and 2 controls, had hemorrhage on the anterior border layer and/or the posterior iris pigment epithelium.