To describe the incidence of hypotony after 25-gauge vitrectomy and to identify preoperative and intraoperative factors that influence the occurrence of hypotony.
Retrospective, nonrandomized, interventional case series.
We reviewed 122 consecutive cases of 25-gauge vitrectomy for all surgical indications. The primary outcome measure was intraocular pressure (IOP) at postoperative day 1, measured with Goldmann tonometry. Secondary outcome measures were clinical signs of hypotony and other complications.
Hypotony, defined as an IOP of 5 mm Hg or less, was found in 13.1% of cases on postoperative day 1. Clinical signs of hypotony were encountered in 7 eyes (5.7%). The risk of hypotony was significantly lower in cases with air or gas tamponade (3.3%) than in cases without tamponade (22.4%). Hypotony was encountered more often in reoperations (29.9%) than in primary operations (9.2%; statistically significant difference). In cases in which intravitreal triamcinolone was used, the risk of hypotony was significantly higher (35.3%) than in cases without triamcinolone (10.3%). Phakic eyes had significantly less chance of hypotony (6.7%) than pseudophakic eyes (15.6%) and eyes undergoing combined phacoemulsification and vitrectomy (25.0%). At postoperative day 7, all cases of hypotony recovered spontaneously. None of our cases developed endophthalmitis.
Our results show that a transient hypotony occurs commonly after 25-gauge vitrectomy. Hypotony was significantly influenced by tamponade, reoperation, intraoperative lens status, and use of intravitreal triamcinolone. Although all cases of hypotony recovered spontaneously without permanent damage, the high frequency of hypotony does impose potential risks. Increased vigilance with focus on perioperative antisepsis and low tolerance of sclerotomy leakage are important for the prevention of endophthalmitis. Strategies aimed at lowering the risk of hypotony are needed to improve the safety of this promising technique.
Since its introduction, 25-gauge vitrectomy is gaining popularity as a vitrectomy platform. Advantages of this sutureless, trocar-based technique are diminished conjunctival scarring, improved patient comfort, decreased postoperative inflammation, and reduced astigmatic changes.
Postoperative hypotony is considered inherent to nonsuture vitrectomy techniques. It is undesirable because it might lead to maculopathy, choroidal detachment, and choroidal hemorrhage. Even more seriously, however, hypotony implies inadequate closure of sclerotomies, increasing the risk for postoperative endophthalmitis. Two large retrospective studies reported a higher incidence of endophthalmitis for 25-gauge vitrectomy compared to 20-gauge vitrectomy, although more recent studies reported no difference in risk.
The aim of our present study is to search for predisposing factors for the occurrence of hypotony, with the ultimate goal to identify strategies to avoid this potentially dangerous complication.
We reviewed the clinical charts of all consecutive patients who underwent 25-gauge vitrectomy between August 1, 2008, and November 1, 2008. We excluded eyes that had preoperative choroidal hemorrhage or detachment because of potential interference with sclerotomy closure. All operations were performed at the Academic Medical Center, University of Amsterdam. Preoperative factors included age, sex, preoperative intraocular pressure (IOP), previous cataract extraction or vitrectomy, surgical indication, and lens status. Intraoperative factors included operator experience (consultants or fellows), operating time, tamponading agent, wound leakage (defined as subconjunctival leakage of intraocular fluid, air, or gas requiring suturing of the sclerotomy), and intraoperative complications.
All patients underwent 3-port pars plana vitrectomy under general or local anesthesia, depending on patient preference. All operations were performed with the Alcon Accurus, using the Alcon 25-gauge system (Accurus 600 DS; Alcon Laboratories, Fort Worth, Texas, USA) and the BIOM wide-angle viewing system (Binocular Indirect Ophthalmo Microscope; Oculus, Inc, Wetzlar, Germany). Infusion pressure was set at 45 mm Hg during vitrectomy and posterior vitreous detachment induction, cutting rate varied between 800 and 1000 cuts/minute, and aspiration pressure was set at 600 mm Hg for core vitrectomy and lowered to about 200 mm Hg in peripheral areas. After thorough disinfection of the periocular skin with povidone-iodine 5% and instillation of povidone-iodine into the inferior fornix, the conjunctiva and the Tenon capsule were displaced away from the intended sclerotomy site using forceps or a cotton swab. All incisions were created in an oblique (beveled, angled) fashion 3.5 to 4 mm posterior to the limbus. The eye was penetrated with the trocar as tangentially as possible parallel to the limbus with the bevel up or down according to surgeon preferences. Once past the trocar sleeve, the angle was set perpendicular to the ocular surface and the cannula was further inserted into the eye. The trocars were inserted with a solid shaft-type trocar-cannula system with a beveled tip. This type of trocar-cannula system creates a V-shaped scleral surface opening.
The infusion cannula was placed in the inferotemporal quadrant, while the 2 other cannulas were placed in the superotemporal and superonasal areas. In cases with attached posterior vitreous, we always induced a posterior vitreous detachment. In all indications, a core vitrectomy was performed, followed by removal of vitreous in the periphery. In pseudophakic cases, the vitrectomy was carried up to the vitreous base. In phakic cases, we intentionally remove less vitreous, leaving a vitreous skirt, to minimize the risk of lens touch. We did not shave the vitreous base for any indication.
If no tamponade was necessary, the vitreous cavity was left with balanced saline solution (BSS; Alcon Laboratories Inc), or had a partial air/fluid exchange aiming at a 20% to 30% fill at the end of the procedure. Internal tamponade was instituted in all cases of rhegmatogenous retinal detachment, in all macular hole cases, and in cases where retinal breaks were found. Either air, SF 6 20%, or SF 6 30% were used for tamponade. The choice of tamponading agent and mixture was based on surgeon preference. Silicone oil was used in 3 cases of rhegmatogenous retinal detachment with proliferative vitreoretinopathy. In 1 case, 0.3 mL of pure C 3 F 8 was used to tamponade a superior hole. At the end of the procedure, the cannulas were removed by slowly pulling them out, following the angled entry path. The sclerotomy sites were then gently massaged with a cotton swab and carefully inspected for fluid or gas leakage. If a sclerotomy site was found leaking, that sclerotomy was sutured with a single stitch of 9–0 vicryl. The patients received a standard nonpressure ocular bandage. Starting from the first postoperative day, patients were put on TobraDex 4 times a day, tapering over 2 weeks.
The primary outcome of the study was postoperative IOP on day 1.
IOP was always measured by Goldmann applanation tonometry in all patients, in all instances. For statistical analysis of effects of predisposing factors, we converted IOP data to a dichotomous dataset. We defined “hypotony” as an IOP of 5 mm Hg or less and “normal pressure” as an IOP of more than 5 mm Hg. Patients with hypotony at the first postoperative control were rechecked after 3 days. If the IOP still did not exceed 5 mm Hg, a third visit was planned at 7 days postoperatively. With each visit, the incidence of hypotony-related complications was evaluated and recorded. We devised a severity scale for the clinical signs of hypotony: grade 1 = corneal Descemet folds; grade 2 = choroidal folds; grade 3 = collapsed eye. No additional therapy was instituted in cases of hypotony.
Statistical analysis was performed using SPSS software for Windows version 16.0 (SPSS Inc, Chicago, Illinois, USA) for Pearson χ 2 and Mann-Whitney analyses.
One hundred twenty-two eyes from 121 patients met the study inclusion criteria. There were 67 right eyes and 55 left eyes. There were 57 female and 64 male patients. The operation was performed under local anesthesia in 79.5% (97/122) of cases. Median age at the time of operation was 65 years (range 27–86 years). Forty-five cases were pseudophakic, 60 were phakic, in 16 cases vitrectomy was combined with phacoemulsification and implantation of an intraocular lens, and 1 case was aphakic. In combined procedure, the corneal wound was always sutured.
Indication for surgery was floaters in 9 eyes, macular pucker in 46 eyes, macular hole in 18 eyes, retinal detachment in 37 eyes, and other indications in 12 eyes (vitreous hemorrhage, retained lens fragments, intraocular lens luxation). Ninety-eight cases were primary vitrectomies and in 24 cases the eye had undergone 1 or more previous vitrectomies.
One sclerotomy was sutured in 4 cases and 2 sclerotomies were sutured in 4 cases because of visible leakage of sclerotomies at the end of surgery, after removal of the trocars. These 8 eyes were included in our further analyses. Of these 8 cases, none became hypotonous. Median IOP at postoperative day 1 was 12 mm Hg (range 0–55 mm Hg). In 17 cases, the IOP was higher than 21 mm Hg, and 8 eyes had an IOP above 30 mm Hg.
Table 1 shows some preoperative and intraoperative characteristics in relation to the occurrence of hypotony on the first postoperative day. Overall, hypotony occurred in 13.1% of cases, varying from 0% (0/9) in floater surgery to 19.6% (9/46) in pucker surgery. The differences in hypotony risk between the different indications, however, did not reach statistical significance.
|Hypotony, n (%)|
|Total eyes (n = 122)||16 (13.1%)|
|Indication||P = .381 a|
|Floater (n = 9)||0 (0.0%)|
|Pucker (n = 46)||9 (19.6%)|
|Macular hole (n = 18)||1 (5.6%)|
|Retinal detachment (n = 37)||5 (13.5%)|
|Else (n = 12)||1 (8.3%)|
|Eye||P = .513 a|
|OD (n = 67)||10 (14.9%)|
|OS (n = 55)||6 (10.9%)|
|Surgeon||P = .327 a|
|Consultant (n = 108)||11 (10.2%)|
|Fellow (n = 14)||3 (21.4%)|
|Air/gas tamponade||P = .002 a|
|No tamponade (n = 58)||13 (22.4%)|
|Tamponade (n = 60)||2 (3.3%)|
|Lens status||P = .012 a|
|Combined procedure (n = 16)||4 (25.0%)|
|Phakic (n = 60)||4 (6.7%)|
|Pseudophakic (n = 45)||7 (15.6%)|
|Aphakic (n = 1)||1 (100%)|
|Reoperation||P = .009 a|
|Primary operation (n = 98)||9 (9.2%)|
|Reoperation (n = 24)||7 (29.2%)|
The side of operation (OD or OS) or surgeon experience were not found to be significant predisposing factors ( Table 1 ). Cases with air or gas tamponade had a significantly lower risk of hypotony compared to cases with no tamponade ( Table 1 ). In the 60 tamponade cases, we used air in 17 cases, SF 6 20% in 30 cases, and SF6 30% in 13 cases. Of the 2 cases with hypotony in this group, 1 occurred with air and 1 with SF 6 20%. There was no statistical difference in hypotony risk for the different agents and mixtures ( P = .840, χ 2 test). In the cases where no tamponade was used, 14 cases were left with BSS and 44 cases had a partial air fill. Hypotony occurred in 14.3% (2/14) of the BSS cases and in 25.0% (11/44) of the partial-air-fill cases. However, this difference was not statistically significant ( P = .402; χ 2 test).
Lens status was significantly predictive for the occurrence of hypotony. Eyes that were phakic and did not undergo combined lens extraction had a lower risk of developing postoperative hypotony. Eyes that were pseudophakic and eyes that underwent a combined procedure were both found to have a higher risk for hypotony. We also found a statistically significant difference in risk for hypotony between primary operations and reoperations ( Table 1 ).
The median duration of the operations was 33 minutes in patients with hypotony on postoperative day 1 (range 12–110 minutes) and 36 minutes in the patients with normal IOP (range 10–93 minutes). This difference was small and not statistically significant ( P = .722, Mann-Whitney U test). Patient age was not found to be a predisposing factor for the occurrence of hypotony. The median age was 65 years (range 46–85 years) in the normal IOP cases and 64 years (range 27–86 years) in the hypotony cases ( P = .242, Mann-Whitney U test). Furthermore, postoperative hypotony was not found to be related to preoperative IOP ( P = .159; Mann-Whitney U test).
Clinical hypotony at the first postoperative day was found in 7 cases (5.7% [7/122]). See Table 2 for characteristics of these cases. In all 7 eyes, IOP was below 5 mm Hg on the first postoperative day. A Seidel test was performed in all these cases, in all instances, but no apparent leakage was identified. Clinical hypotony only occurred after macular pucker procedures (n = 6) and 1 retinal detachment procedure. There were corneal Descemet folds in 6 eyes and choroidal folds in 1. We did not encounter collapsed eyes. The IOP spontaneously increased to a value over 5 mm Hg in 6 of 7 eyes on day 3. The IOP of the remaining case spontaneously increased after 7 days. All signs of clinical hypotony disappeared spontaneously. There was no anatomic or functional permanent damage in any of these cases.
|Case No.||Indication||Triam||Reop||Postop Day 1||Postop Day 3||Postop Day 7||VA pre||VA 3mo|
|IOP||Severity a||IOP||Severity a||IOP||Severity a|