Glaskörperblutungen bei diabetischer Retinopathie: Verlauf ohne und mit Vitrektomie

Zusammenfassung

Bei diabetesbedingten massiven Glaskörperblutungen wurden die Spätresultate der Vitrektomie (80 Augen) mit denen nach konservativer Therapie (110 Augen) verglichen. Die Visusresultate nach Vitrektomie sind signifikant besser, Erblindungen nach Operationsmißerfolg aber neben anderen Komplikationen wie Rubeo-sis iridis, Sekundärglaukom, Phthisis bulbi beträchtlich häufiger als in der Vergleichsgruppe.

Summary

The natural course of 110 eyes with proliferative diabetic retinitis and massive vitreous haemorrhage was compared with the data of 80 eyes followed up after pars plana vitrectomy. The mean duration of follow-up was 27.3 and 17.6 months respectively (Table 1). Age, insulin dependence, arterial hypertension, and mortality rate were comparable in the two groups. The frequency of an insufficient control of hypertension was about 40% in both groups (Table 2). The mean duration of diabetes was significantly longer in the vitrectomy group (20 years) than in the control group (15.8 years). A spontaneous resolution of vitreous haemorrhages occurred in 43%, while there was no change in 28.5% and recurrent bleeding in 28.5%. In the vitrectomy group, the preoperative incidence of rubeosis iridis (10%), glaucoma (15%), and retinal detachment (37.5%) was significantly higher than in the control group. The postoperative incidence of rubeosis iridis and glaucoma increased by a factor of 3 (31% and 45% respectively) but remained low in the control group (5.5% and 12.7%). Intractable retinal detachment was the cause of blindness in 21 % of the eyes after vitrectomy. In the control group, phthisis (5%) and total blindness (5.5%) were less frequent than in the vitrectomy group (17.5% and 29% respectively). An improvement of visual acuity was highly significant only in the vitrectomy group (p< 0.0001 with Wilcoxon's matched pairs signed rank test). Good visual results, with acuities ranging from 20/100 to 10/ 10 were significantly more common after vitrectomy (41%) than without (12%). In the control group visual acuity remained below 10/200 in 66 eyes because of persistent vitreous opacities (74%), retinal detachment (47%), and central fibro-vascular proliferation (47%). These three parameters were significantly less frequent (6%, 16%, and 12% respectively) in 44 eyes with a better final vision (Table 4). A similar analysis in the vitrectomy group (Table 5) yielded the following results: a visual acuity below 10/200 was associated with peripheral and central retinal detachment in 52%, incomplete vitreous detachment in 75%, glaucoma in 58%, and rubeosis iridis in 53%. The corresponding rates for the eyes with visual acuities equal to or greater than 10/200 were 11%, 2%, 27%, 34% and 14%. According to this analysis, incomplete vitreous detachment with subsequent recurrent intravitreal bleeding and retinal traction detachment are the major causes of blindness of eyes with untreated diabetic vitreous haemorrhages. The same risk factors, as well as rubeosis iridis and glaucoma, will also hamper the functional outcome of vitreous surgery, although the overall visual benefit as compared to the natural history of these cases strongly justifies the application of vitrectomy. However, when vitrectomy is considered in a diabetic with vitreous haemorrhage one should be aware of the fact that following surgical failure the rate of severe complications such as rubeosis iridis, glaucoma, retinal detachment, phthisis, and total blindness with the consequence of possibly several secondary interventions is much higher than with any nonsurgical procedures.