A Trial of China Laser and Surgery Study Glaucoma in Rural China

Glaucoma is one of the three leading causes of blindness in China. Although this condition can be controlled with medication to lower intraocular pressure , such treatment must continue over a patient's lifetime. This is not practical in resource-poor settings due to cost and inconvenience for patients. Trabeculectomy is the standard treatment offered in rural China but is rarely used because of the following issues:

1. need for hospitalization, which increases patient expense and inconvenience
2. high refusal rates by patients for all ocular surgery
3. high rates of infection due to poor quality instruments and material
4. need for multiple follow-up visits post-surgery (our earlier work shows a return rate of only 30% 1 month post-surgery among rural patients in China).

A viable alternative for patients in rural China is selective laser trabeculoplasty , based on the following advantages:

1. simplicity (done as a 5-minute out-patient procedure requiring only topical anaesthesia)
2. effectiveness (18-32% reduction in intraocular pressure, which is comparable to medication)
3. safety (complications are rare and do not include serious infectious as with trabeculotemy)
4. convenience (minimal follow-up, with one visit within 6 weeks post-surgery)
5. cost-effectiveness (compared to other glaucoma therapies in many settings)
6. acceptability to patients (due to low-cost out-patient care with easy follow-up).

Study Design (Quantitative) This phase of the proposed China Laser and Surgery Study research project is a multi-centre open-label pragmatic randomised controlled trial.

Expected timeline:

• Enrolment: 12 months
• Follow-up: 12 months
• Data analysis: 4 months

Patient Population and Setting: Patients presenting with glaucoma at rural hospitals (n=4) in Guangdong province, China, all of whom are currently capable of performing trabeculectomy surgery, and have been provided an selective laser trabeculoplasty machine and training in its use.

Sample Size: One hundred ninety patients (95 in each group) will be needed to obtain a power of 0.90, with a two-sided alpha of 0.05. Patients are the unit of randomization. Expected IOP difference between groups of 3.0 mm Hg + 4.0 mmg HG, and a correlation of 0.30 between baseline and final intraocular pressure, allowing for loss to follow up of 30%.

Baseline Data:

aseline data (to be collected prior to determination of eligibility and randomization but after obtaining informed consent) include the following:

• Presence or absence of relevant pre-existing health conditions (i.e., hypertension, diabetes, cardiovascular disease)
• Ophthamologic test results
• Best corrected visual acuity and presenting visual acuity in both eyes
• Intraocular pressure in both eyes
• Cup to disk ratio in both eyes
• presence or absence of visually significant cataract in both eyes; if cataract present in study eye, whether cataract surgery has already been performed
• degrees of visible Trabecular Meshwork in the study eye on gonioscopy without "dynamic gonioscopy"
• score on glaucoma knowledge test using a simple, previously validated questionnaire
• Demographic data
• date of birth or self-reported age in years
• sex
• Contact information
• Address
• mobile phone number where can be reached (own or neighbours, etc.)
• Socioeconomic status
• completed education in years
• wealth, as measured by ownership of items on a pre-determined list of 13 common household devices in China
• job title/description
• Level of difficulty traveling to hospital with eye clinic
• distance traveled
• time spent traveling
• direct or out-of-pocket costs for travel, meals, etc.

Main analyses:

• Following requirements for intention-to-treat analyses, we will measure mean intraocular pressure at 12 months comparing the two groups and adjusting for baseline intraocular pressure, including all subjects who accepted and refused treatment (primary analysis, unadjusted). We will also measure intraocular pressure as above, adjusting for cataract surgery, age, gender, prior use of medication, degrees of open angle in study eye at baseline, and severity of glaucoma (Cup to disk ratio) in study eye at baseline (primary analysis, adjusted). Both of the above analyses will be repeated using a per protocol approach (including only patients who accepted their assigned therapy, or secondary analysis).
• We will test for differences in baseline characteristics by study group with t-tests (for continuous variables) and Chi-square tests (for categorical variables). We will use generalized linear models with Poisson regression (for variables with binary outcomes) and linear regression (for variables with continuous outcomes) to estimate the relative risk for the intervention arm. Multiple imputation in Stata 13.1 (StataCorp, College Station, TX, USA) will be used to address missing data. All statistical analysis will be performed using Stata 13.1. A two-tailed P-value of <0.05 will be considered statistically significant.