Methods

After obtaining approval from the Institutional Review Board of the Nagata Eye Clinic, the medical records of 1887 eyes that had undergone pars plana vitrectomy at the Nagata Eye Clinic between January 9, 2004 and March 29, 2011 were reviewed. From these, 981 eyes were excluded because of the presence of possible contributing factors for the IOP elevation, for example, preexisting glaucoma (suspect), ocular hypertension, chronic systemic or topical corticosteroid use, previous intravitreal or periocular triamcinolone injections, history of anterior segment inflammation, ocular trauma, rubeosis iridis, peripheral anterior synechia, proliferative diabetic retinopathy, proliferative vitreoretinopathy, history of more than 2 vitrectomy procedures, and use of silicone oil. Additionally, 109 eyes were excluded because the follow-up period was ≤6 months. To evaluate the incidence of ocular hypertension in the pseudophakic conditions, we excluded 2 phakic eyes that had pars plana vitrectomy alone. In addition, we also excluded 28 eyes that had pars plana vitrectomy in both eyes because we wanted to compare the IOP to that of the fellow eye that had not undergone pars plana vitrectomy. In the end, 767 eyes of 767 patients were studied, which represents 88% of the 876 consecutive cases. An informed consent for the surgical procedures was obtained from all patients before the surgery. The research procedures conformed to the tenets of the Declaration of Helsinki.

All patients underwent conventional 3-port pars plana vitrectomy, and all surgeries were performed by a single experienced surgeon (H.K.). One hundred and seventy-six eyes underwent pars plana vitrectomy alone because these eyes had undergone cataract surgery before the pars plana vitrectomy. Five hundred and ninety-one eyes underwent combined phacoemulsification and aspiration through a corneal or sclerocorneal incision followed by an implantation of an intraocular lens. The vitrectomy was performed after the implantation of the intraocular lens. In all eyes, the vitreous body was made visible by an intravitreal injection of triamcinolone acetonide during the pars plana vitrectomy. Postoperatively, patients were treated with a topical corticosteroid 3 times a day for 1 month.

All IOPs were measured with a Goldmann applanation tonometer (Hagg-Streit, Koeniz, Switzerland). Late-onset ocular hypertension was defined as an IOP >21 mm Hg that was detected more than 2 months postoperatively on at least 2 postoperative visits. In addition, the increase in the IOP must have been 4 mm Hg higher than the preoperative IOP. The medical records of all vitrectomy patients with a follow-up period >6 months were examined. In all eyes, documented gonioscopic findings confirmed that the anterior chamber angle was open.

Results

The mean age at the time of vitrectomy was 63 ± 11 years. The mean duration of the follow-up period was 47.8 ± 25.3 months. There were 383 men and 384 women. The diagnosis of the cases requiring vitrectomy were: 308 eyes with rhegmatogenous retinal detachment, 202 eyes with epiretinal membrane, 169 eyes with macular hole, 44 eyes with vitreous hemorrhage (attributable to retinal arteriolar macroaneurysm in 19 eyes, age-related macular degeneration in 8 eyes, posterior vitreous detachment in 8 eyes, retinal tear in 4 eyes, retinal vein occlusion in 4 eyes, and retinal hemangioma in 1 eye), 16 eyes with subretinal hemorrhage (attributable to retinal arteriolar macroaneurysm in 8 eyes, age-related macular degeneration in 7 eyes, and suprachoroidal hemorrhage in 1 eye), 15 eyes with vitreomacular traction syndrome, 12 eyes with vitreous opacity (attributable to asteroid hyalosis in 6 eyes, vitreous floaters in 4 eyes, and organized vitreous hemorrhage in 2 eyes), and 1 eye with retinoschisis. The gauge of the instruments used for the pars plana vitrectomy was 20 gauge in 240 eyes, 23 gauge in 51 eyes, and 25 gauge in 476 eyes.

Of the 767 eyes of 767 patients, 32 eyes of 32 patients (4.2%) developed late-onset ocular hypertension. There were 21 men and 11 women ( P = .07). There were no significant differences in the incidence of ocular hypertension among the different diseases for which pars plana vitrectomy was performed ( P = .49, Table 1 ) and for the gauge of the instruments used for the pars plana vitrectomy ( P = .26, Table 2 ). In addition, there were no significant differences in the incidence of late-onset ocular hypertension between the 478 eyes that required gas tamponade (including 308 eyes with rhegmatogenous retinal detachment, 169 eyes with macular hole, and 1 eye with vitreomacular traction syndrome), and the 289 eyes that required no gas tamponade (including 202 eyes with epiretinal membrane, 44 eyes with vitreous hemorrhage, 16 eyes with subretinal hemorrhage, 14 eyes with vitreomacular traction syndrome, 12 eyes with vitreous opacity, and 1 eye with retinoschisis; P = .06).

Of the 32 eyes with ocular hypertension, 18 eyes required glaucoma medications and 3 eyes required glaucoma surgery owing to persistent IOP elevation with medication. The IOP was successfully controlled in all of these eyes. The remaining 11 eyes were observed without treatment. In most patients, late-onset ocular hypertension did not cause major changes in visual function. In 4 eyes, open-angle glaucoma was newly diagnosed after the vitrectomy, but only 2 eyes had a progression of a glaucomatous visual defect. The characteristics of the eyes that developed late-onset ocular hypertension are shown in Table 3 .

The highest IOPs of all eyes that developed late-onset ocular hypertension following pars plana vitrectomy are shown in Figure 1 . The 3 eyes with an asterisk within the oval had severely increased IOP that required glaucoma surgery ( Figure 1 ). The follow-up durations after vitrectomy of all eyes are shown in Figure 2 . The proportion of eyes followed for more than 60 months was 34.3%, which is indicated by an asterisk in Figure 2 . Taken together, these data indicate that the actual rate of late-onset ocular hypertension might have been higher than our recorded results if longer follow-up durations had been studied.

Highest intraocular pressures of all pseudophakic eyes that developed late-onset ocular hypertension following uncomplicated pars plana vitrectomy, and the time of development of the highest intraocular pressure (ie, an IOP >21 mm Hg), as a function of the time of the IOP elevation measured. IOP = intraocular pressure. Oval indicated by an asterisk (*) surrounds 3 eyes that required glaucoma surgery owing to persistent IOP elevation. Oval indicated by an arrow surrounds 7 of 32 eyes (21.9%) that developed late-onset ocular hypertension after 60 months postoperatively.

Histogram showing the number of pseudophakic eyes that developed late-onset ocular hypertension following uncomplicated pars plana vitrectomy as a function of the duration of the follow-up period after vitrectomy of all eyes. * = Of all eyes studied, 263 eyes (34.3%) were followed for more than 60 months after vitrectomy. The mean follow-up duration in all eyes after vitrectomy was 47.8 ± 25.3 months.

Among the eyes that developed late-onset ocular hypertension following pars plana vitrectomy, 31 eyes of 32 patients developed ocular hypertension unilaterally in the vitrectomized eye. The mean preoperative IOP in the vitrectomized eyes was 16.4 ± 2.1 mm Hg and the mean postoperative IOP, which was the average of 2 consecutive IOP measurements before antiglaucoma treatment, was 27.5 ± 6.8 mm Hg. In addition, the mean preoperative IOP in nonvitrectomized fellow eyes was 16.4 ± 2.0 mm Hg and the mean postoperative IOP, which was the average of the last 2 visits, was 16.7 ± 2.3 mm Hg. The mean postoperative IOP in the vitrectomized eyes was significantly higher than that in nonvitrectomized eyes ( P < .0001).

For all eyes, 176 eyes underwent pars plana vitrectomy alone and 591 eyes underwent pars plana vitrectomy combined cataract surgery. We found no significant difference in the incidence of ocular hypertension between the lens status regarding the presence of the crystalline lens at the time of pars plana vitrectomy ( P = .15).