Intraocular surgery in a large diabetes patient population: risk factors and surgical results.

The prevalence of diabetes is on the increase in developed countries. Accordingly, the prevention and treatment of vision-threatening diabetic eye complications is assuming greater importance. The overall aim of this thesis is to analyse risk factors for intraocular surgery in a large diabetes population and to report surgical results. The specific objectives are to (1) estimate the incidence of diabetic vitrectomy and analyse risk factors (Study I), (2) report long-term results, prognostic factors and incidence of cataract surgery after diabetic vitrectomy (Study II), (3) report results and prognostic factors after cataract surgery in diabetes patients (Study III) and (4) analyse risk factors for diabetic papillopathy with emphasis on metabolic control variability (Study IV). All studies are based on a close-to-complete national surgery register and a large, closely followed diabetic retinopathy screening population. Study I (cohort study, 3980 type 1 diabetes patients) illustrates that diabetic vitrectomy is rarely required in a diabetes patient population with varying degrees of diabetic retinopathy. The risk of reaching diabetic vitrectomy increases fourfold with poor metabolic control, defined as glycosylated haemoglobin A1c > 75 mmol/mol (~9%), which points to good metabolic control as an important preventive measure. Study II (cohort study, 167 diabetes patients) shows that most diabetic vitrectomy patients stand to gain visual acuity ≥0.3 after surgery. Visual acuity is stable after 1 year, and the stability is maintained through 10 years of follow-up. The use of silicone oil for endotamponade is a consistent long-term predictor of low vision after surgery. The risk of requiring cataract surgery after diabetic vitrectomy is substantial, and the risk increases if silicone oil is used. Study III (cohort study, 285 diabetes patients) shows, on the other hand, that diabetes patients can expect a significant improvement in visual acuity after cataract surgery, regardless of the degree of diabetic retinopathy. Poor preoperative visual acuity, a high degree of diabetic retinopathy and advanced age are predictors of a poor visual acuity after surgery. The risk of diabetic macular oedema after surgery is 4%. Finally, Study IV (case-control study, 2066 type 1 diabetes patients) demonstrates that diabetic papillopathy shares characteristics with diabetic retinopathy. The risk of experiencing diabetic papillopathy increases markedly with a drastic, recent reduction in glycosylated haemoglobin A1c and a small optic disc. This lends support to the theory that diabetic eye complications may occur in anatomically predisposed patients in response to metabolic control variability. Overall, results after intraocular surgery in diabetes patients are favourable. Surgery, however, is associated with costs to society, patient discomfort and risk of complications. This thesis provides an analysis of risk factors for intraocular surgery and identifies prognostic factors for visual acuity after surgery, which can be used for preventive purposes, surgical decision-making and patient counselling.