Aqueous humor was collected in the operating room from 63 eyes of 63 human subjects (age range, 19 to 83 years; mean±standard deviation, 49.88±17.58) undergoing cataract or glaucoma surgery.

Patients were classified into four groups: those with POAG (group P; 14 patients), those with NVG complicated with diabetes (group N; 14 patients), those with SOAG complicated with uveitis (group S; 15 patients), and those with only cataract (group C, control group; 20 patients).

In group P, the average preoperative intraocular pressure (IOP) was 21.8±5.7 mmHg. The patients had established diagnosis of advanced POAG that correlated with advanced visual field loss. The POAG patients received 2.51±0.76 anti-glaucoma medications, and none of the patients had previous ocular surgery of any kind. In group N, the average preoperative IOP was 33.7±6.2 mmHg. All patients had proliferative diabetic retinopathy with rubeosis iridis. The patients were treated with 2.23±1.18 anti-glaucoma medications and panretinal photocoagulation (PRP). The anterior segment neovascularization was reduced in nine patients, but peripheral anterior synecia formation progressed. Five patients had peripheral anterior synecial closure at the time of PRP. In group S, the average preoperative IOP was 27.3±4.7 mmHg. The patients had uveitis with anterior, intermediate, posterior, and panuveitis of different origins. In group C, patients underwent elective cataract extraction with intraocular lens implantation. All patients had a normal ocular examination. Patients were excluded if they were taking topical medication or had an ocular condition that was being treated with topical or systemic medications.

All surgery was performed under local anesthesia, consist of 2% lidocaine hydrochloride, administered by retrobulbar and nadbath injection.

All samplings were performed by the same surgeon. At the time of cataract or glaucoma surgery, a limbal paracentesis was made with a 20 G MVR knife (BD Ophthalmic systems, Bidford on Avon, Warks, B50 4JH, UK). Aqueous humor (100~200 µL) was aspirated into a tuberculin microsyringe prior to any incisional surgical procedures. All samples were rapidly frozen and stored in a -70℃ freezer until analysis.

The concentration of TGF-β₂ in aqueous humor samples was quantified by a double antibody 'sandwich-indirect' enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, MN, U.S.A.), with a sensitivity of 2 pg/ml. Intra- and inter-assay coefficients of variation for ELISA were within 7%. The concentration of TGF-β₂ in the aqueous humor samples was determined by comparison with a standard curve of activity of standard TGF-β₂.

Briefly, 100 µL standard TGF-β₂ or aqueous humor sample was added to a 96-well ELISA plates coated with murine monoclonal anti-TGF-β₂-antibodies. After 2 hours of incubation at room temperature, the wells were washed three times with 400 µL wash buffer. Each well was coated with 200 µL of polyclonal anti-TGF-β₂ antibodies conjugated to horseradish peroxidase. The mixture was incubated and washed, as previously described. 200 µL substrate solution was added to each well, and the plate was incubated for 20 minutes at room temperature to allow the reaction to proceed. 50 µL stop solution was then added to each well. Absorbance at 450 nm was measured with an immunoreader system (BIO-TEK, Winooski, VT, USA).

For the determination of total TGF-β₂, all of the aqueous humor samples were treated with 10 µL 1N HCl, and left for 10 minutes at room temperature to allow activation. Then, 10 µL 1.2 N NaOH/0.5 M HEPES was added for neutralization, and diluted with 340 µL Calibrator Diluent. Determination of total TGF-β₂ was performed immediately.

Statistical analysis was performed, using the Kruskall-Wallis test program. A multiple comparison was conducted, using the Mann-Whitney test. Significant differences were defined by P values below 0.05.