PURPOSE: Glaucoma is a frequent and often severe problem in patients needing keratoprosthesis (KPro) surgery. Standard glaucoma shunts in these cases often develop a very dense capsule that obstructs flow and causes intraocular pressure elevation. Therefore, we developed shunts that deliver aqueous to distant epithelialized cavities where an obstructing capsule is less likely to form. In this retroactive study of such shunts, the risk of postoperative infection (especially endophthalmitis) has been assessed. METHODS: Ahmed shunts were connected with a tube to the lacrimal sac or ethmoid sinuses. In subsequent designs the polypropylene plate was eliminated, the valve was enclosed and a distal tube added (New World Medical Inc, Rancho Cucamonga, CA). The latter shunts were connected to the maxillary sinus or lower lid fornix. Between 2001 and 2005 the devices were implanted in 34 patients with very severe ocular disease. (31 had KPro implanted). Thus, the distal tube was led to the lacrimal sac (2 cases), to the ethmoid sinuses (6), to the maxillary sinus (16 cases), and to the lower lid fornix (10 cases). The patients were followed for cumulatively 145 shunt years, with a mean follow-up of 4 years and 3 months. They were instructed to administer low-dose topical antibiotics indefinitely. RESULTS: Only 1 case of acute bacterial endophthalmitis occurred, thus an incidence of 0.7% per shunt year. A slow-growing Mycobacterium was cultured in a disintegrating eye and 1 bacterial maxillary sinusitis occurred, both in autoimmune diseases. Three valves became exposed and had to be removed (1 replaced). Four eyes developed hypotony, probably from valve failure. CONCLUSIONS: The shunt arrangement, which connects the anterior chamber to potentially microbe-populated cavities or to the lower lid fornix and its flora, might be suspected to allow rapid retrograde invasion of infective agents, resulting in endophthalmitis. In this series of KPro patients, however, the incidence of severe infection was very low, in fact comparable to that after standard trabeculectomy.
PURPOSE:Glaucoma is a frequent and often severe problem in patients needing keratoprosthesis (KPro) surgery. Standard glaucoma shunts in these cases often develop a very dense capsule that obstructs flow and causes intraocular pressure elevation. Therefore, we developed shunts that deliver aqueous to distant epithelialized cavities where an obstructing capsule is less likely to form. In this retroactive study of such shunts, the risk of postoperative infection (especially endophthalmitis) has been assessed. METHODS: Ahmed shunts were connected with a tube to the lacrimal sac or ethmoid sinuses. In subsequent designs the polypropylene plate was eliminated, the valve was enclosed and a distal tube added (New World Medical Inc, Rancho Cucamonga, CA). The latter shunts were connected to the maxillary sinus or lower lid fornix. Between 2001 and 2005 the devices were implanted in 34 patients with very severe ocular disease. (31 had KPro implanted). Thus, the distal tube was led to the lacrimal sac (2 cases), to the ethmoid sinuses (6), to the maxillary sinus (16 cases), and to the lower lid fornix (10 cases). The patients were followed for cumulatively 145 shunt years, with a mean follow-up of 4 years and 3 months. They were instructed to administer low-dose topical antibiotics indefinitely. RESULTS: Only 1 case of acute bacterial endophthalmitis occurred, thus an incidence of 0.7% per shunt year. A slow-growing Mycobacterium was cultured in a disintegrating eye and 1 bacterial maxillary sinusitis occurred, both in autoimmune diseases. Three valves became exposed and had to be removed (1 replaced). Four eyes developed hypotony, probably from valve failure. CONCLUSIONS: The shunt arrangement, which connects the anterior chamber to potentially microbe-populated cavities or to the lower lid fornix and its flora, might be suspected to allow rapid retrograde invasion of infective agents, resulting in endophthalmitis. In this series of KPro patients, however, the incidence of severe infection was very low, in fact comparable to that after standard trabeculectomy.
Authors: Eleftherios I Paschalis; James Chodosh; Ralph A Sperling; Borja Salvador-Culla; Claes Dohlman Journal: PLoS One Date: 2013-06-28 Impact factor: 3.240