Rationale: The Baerveldt glaucoma drainage device (GDD) successfully reduces intraocular pressure but also involves a risk of corneal endothelial deterioration. Supposedly, the tip of a GDD tube with a thinner diameter, such as the Paul implant, will remain at a larger distance from the cornea and, thereby, cause less damage.

Objective: To determine whether the Paul tube induces less damage to the corneal endothelium than the Baerveldt GDD.

Study design: Randomized clinical trial. Study population: Phakic patients scheduled for surgical GDD implantation. Intervention: Either a Baerveldt or a Paul GDD implant. Main study parameters/endpoints: Endothelial cell density and tube position at 24 months.

Nature and extent of the burden and risks associated with participation, benefit and group relatedness: The Paul GDD may have a less harmful effect on corneal endothelium. Otherwise, both GDDs will probably have a similar risk/benefit profile. The risks of study-related assessments are negligible, burden is low, extra time is about 5 x 1.5 h (total 7.5 h) in two years.

Inclusion Criteria:

• Age > 18 years.
• Informed consent.
• Caucasian ethnicity (to facilitate comparison of results with those of earlier work). Ethnicity will be based on self-reported origin of subject and/or parents.
• Primary open-angle glaucoma, pseudoexfoliative glaucoma, or pigmentary glaucoma.

Exclusion Criteria:

• History of intraocular surgery (e.g. vitrectomy, cataract surgery, cyclodestructive procedures etc).
• History of ocular comorbidity (e.g. active uveitis, proliferative diabetic retinopathy).
• Pseudophakia.
• Functionally monocular patients.
• Need for glaucoma surgery combined with other ocular procedures (i.e. cataract surgery, keratoplasty, or retinal surgery) or an anticipated need for additional ocular surgery.
• Narrow anterior chamber angle.
• Best corrected visual acuity less than 0.1.
• Severe blepharitis.