To explore the long-term efficacy and safety of laser peripheral iridectomy for primary angle closure glaucoma (PACG). It was a retrospective case series study. Data were collected from those patients who received laser peripheral iridectomy (LPI) for acute or chronic PACG from April 1992 through October 2002 at the Peking Union Medical College Hospital. Only patients who were followed for at least 5 years were included in this study. The control of intraocular pressure (IOP), visual acuity and managements after LPI were analyzed. All of the studied eyes were re-classified into three categories according to the status of anterior chamber angle, optic nerve head and visual field before LPI: primary angle closure suspect (PACS), primary angle closure (PAC) and primary angle closure glaucoma (PACG). Satisfactory control of IOP was defined as the IOP was less than 21 mm Hg (1 mm Hg = 0.133 kPa) without any medications after LPI. No satisfactory control of IOP was defined as the IOP was greater than 21 mm Hg after LPI, yet could be controlled below 21 mm Hg by anti-glaucoma medications. A failure in IOP control was defined as an acute attack of angle closure developed or filtering surgery was required to control IOP after LPI. Chi-square analysis was used for comparison of IOP control in different groups. One hundred and thirty one patients (251 eyes) with PACG were eligible for this study. The mean follow-up period was (9.2 +/- 3.7) years. Of the 251 eyes, 18 eyes (7.2%) were identified as PACS, 98 eyes (39.0%) PAC, 129 eyes (51.4%) PACG, and 6 eyes (2.4%) could not be classified owing to the lack of the information on the optic nerve head and visual field before LPI. The rates of satisfactory control of IOP were 27.1% in all eyes, and 88.9% (16/18), 38.8% (38/98) and 10.9% (14/129) in PACS, PAC and PACG eyes respectively. The rates of no satisfactory control of IOP were 59.8% in all eyes, and 5.6% (1/18), 48.0% (47/98) and 75.2% (97/129) in PACS, PAC and PACG eyes respectively. The rates of failure in IOP control were 13.1% in all eyes, and 5.6% (1/18), 13.3% (13/98) and 14.0% (18/129) in PACS, PAC, PACG eyes respectively. The difference in IOP control between PACS, PAC and PACG eyes was statistically significant (chi(2) = 59.08, P = 0.000). Only 8 eyes had an acute attack of angle closure after LPI. No long-term complications after LPI were observed in all eyes. The IOP control after LPI in PACG eyes is not so good as expected. However, most of PACG eyes after LPI are free of acute attack of angle closure. PACG eyes should be given close and regular follow-up in a long-term to monitor the IOP control and the progression of PACG after LPI.