CIGTS data: Large IOP reductions, initial trabeculectomy values validated

New Orleans-Researchers involved in the Collaborative Initial Glaucoma Treatment Study (CIGTS) have revealed several important findings on the relationship between large reductions in IOP and disease progression, the outcomes of surgical and medical treatment, and the quality of life of glaucoma patients.

Fifteen years after this large trial began, they are still mining the data to produce studies that will be applicable in both practice settings and clinical trials.

The key findings from CIGTS were that large IOP reductions (35% to 48%) were associated with little or no progression of glaucoma, that even early glaucoma is associated with a decrement in quality of life, and that initial trabeculectomy is an acceptable alternative and may be preferable in patients with advanced disease at diagnosis, said Richard P. Mills, MD, MPH, Glaucoma Consultants Northwest, Seattle.

Dr. Mills, vice chairman of the CIGTS trial, presented pearls from data collected through up to 9 years of follow-up during the glaucoma subspecialty day at the annual meeting of the American Academy of Ophthalmology.

CIGTS was a multicenter, randomized, controlled clinical trial designed to determine whether patients with an initial diagnosis of glaucoma are better treated by topical medication or by immediate filtration surgery. The study enrolled 607 patients aged 25 to 75 years with newly diagnosed primary, pigmentary, or pseudoexfoliative glaucoma who were randomly assigned to initial treatment with either trabeculectomy or eye drops. Recruitment began in fall 1993 and was completed in spring 1997. Follow-up visits occurred at 3 and 6 months after treatment initiation and every 6 months thereafter for at least 5 years.

CIGTS is unique among glaucoma studies in several respects, Dr. Mills said. First, patients were treated to achieve an individualized target pressure that was set depending on the initial pressure level and the severity of the glaucoma damage. Second, patients in both the medical and surgical arms achieved large IOP reductions averaging 35% in the medical arm and 48% in the surgical arm, and these reductions were sustained through 9 years of follow-up. These reductions are larger than those achieved in any other trial reported to date.

A comprehensive quality-of-life interview was obtained on every subject at 6-month intervals. The primary outcome of the study was visual field progression. At 5 years, no difference was seen between the medical and surgical arms, despite more than adequate power to detect a difference. Thus, it is acceptable to offer trabeculectomy as a first-line treatment option, especially when a patient wishes to avoid drops and laser, Dr. Mills said.

A subgroup analysis on the primary outcome variable showed that for patients with advanced visual field loss at baseline (mean defect -10 dB, on average), progression was less in the surgical arm than in the medical arm. This finding demonstrates that initial surgery is an attractive option in patients with advanced disease, according to Dr. Mills.

He added, though, that on average disease in neither the medical nor surgical groups progressed substantially over 5 years of observation. Although it always is dangerous to compare results across studies, the substantial IOP reductions achieved in CIGTS may have contributed to halting or dramatically slowing progression of damage, he explained.

In another subgroup analysis of visual field deterioration, it was found that IOP variability over the course of the study, rather than mean or maximum IOP, was the strongest predictor of progression.

This finding was also true in another important glaucoma trial, the Advanced Glaucoma Intervention Study, and suggests that it is important to achieve not only a low but also a steady pressure level during therapy, Dr. Mills emphasized.

The quality-of-life portion of the study consisted of a list of 43 possible symptoms that captured typical problems associated with medical and surgical treatment and their relative impact from mild to severe. Patients were asked about these symptoms during telephone interviews performed every 6 months. The outcomes showed that the surgery group reported more symptoms, especially local ones such as scratchiness and tearing, than the medical group. Based on this finding, Dr. Mills suggested that surgical informed consent forms always include information about local eye irritation.

Reports of symptoms and their negative impact on quality of life decreased over time in both groups and by 5 years were about equal in the surgical and medical arms.

Results for secondary outcomes showed that surgical patients had a greater risk of substantial visual acuity loss over all time points, Dr. Mills said. As expected, trabeculectomy patients developed cataracts more often, requiring cataract extraction in 20% by 5 years, versus 7% in the medically treated patients.

In early glaucoma, it has been learned that cataract removal restores vision and visual functioning nearly to baseline levels, and this finding was validated by results from CIGTS. The average visual acuity in both the medical and surgical groups was about equal at 4 years and after.

Continuing his summary of the outcomes of CIGTS, Dr. Mills said that visual field scores at baseline had a low degree of correlation with quality-of-life scores, but the correlation was highly significant.

This correlation indicates that even in very early glaucoma, a decrement in quality of life can on average be demonstrated and underscores the need for early glaucoma detection in populations, he said.

He also noted that the rather complicated visual field score invented for this study did not perform any better than mean defect, a standard parameter in the printout of tests for glaucoma diagnosis and management (Humphrey Field Analyzer, Carl Zeiss Meditec).

Accordingly, researchers and clinicians can safely use mean defect as a measure of glaucoma progression as long as other causes of mean defect decline, such as cataract, can be excluded.

The rate of surgical complications in CIGTS was low. Among 465 primary trabeculectomies performed without mitomycin-C in 300 surgical patients, no cases of severe sustained visual loss were reported. Older patients were more likely to experience choroidal detachment, whereas black patients were less likely than white patients to have hyphema but more likely to develop ptosis.

CIGTS has produced 13 peer-reviewed publications and 38 platform presentations and posters to date, Dr. Mills said. Further outcomes papers are planned and will appear over the next several years.