Methodological quality of the review: Low confidence

Authors: Bbumelo LL,Bai L

Region: Spain, Canada, Korea, Iran, Germany, China, Greece, France and India.

Sector: Glaucoma

Sub-sector: Risk factor

Equity focus: None specified

Review type: Other review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background:

The level of the intraocular pressure (IOP) associated with optic nerve changes is not the same in every eye and some individuals may tolerate for the long periods a pressure that would rapidly blind another. The key to preventing irreversible blindness caused by glaucoma lies in early diagnosis, which can only be possible through mass screening of the population, because in this case everyone is a suspect until proven otherwise by the screening tests. Several studies, including most notably the Ocular Hypertension Treatment Study, have determined a significant relationship between central corneal thickness (CCT) and IOP, and the need to adjust IOP applanation tonometry readings by a factor to consider the effect of CCT in order to understand the “true” IOP.

Objectives:

To evaluate the relationship between central corneal thickness and intraocular pressure in healthy and glaucomatous eyes of adults.

Main findings:

The authors included 13 studies, of which 12 were observational and one was a case control study. Seven studies were conducted in Europe, four were Asian studies and two were in North America. In the meta-analysis for IOP, the authors reported that heterogeneity was present (Q=51.61, I²=76.8%, P<0.001). The authors reported that compared to control subjects, patients had significantly increased IOP levels (SMD: 0. 50, 95% CI: 0.30 ~ 0.70, Z = 4. 88, P <0.001).

The authors found that, through the meta-regression analysis, none of the covariates had a statistically significant relationship to the observed IOP difference between patients and controls. The authors found no publication bias within the review. The sensitivity analysis showed that two studies had greater influence on findings.  After omitting these two studies, the combined effect size of meta-analysis became 0.437 (0.253, 0. 620) and 0. 447 (0. 256, 0. 638) respectively.

Meta-analysis for CCT showed mild heterogeneity (Q=18.06, I²= 33.5, P = 0.114). Compared to control subjects, patients had significantly decreased CCT (SMD: -0.14, 95% CI: -0.23 ~ -0.05, Z =3. 14, P =0.002). Meta-regression revealed that age mean (P = 0.025) was found to have a statistically significant relation to the observed CCT difference between the two groups. The authors found no publication bias. Based on the sensitivity analysis, two studies had greater influence on overall findings of the meta-analysis. After omitting these two studies, the combined effect size of meta-analysis became -0.172 (-0.279, -0.067) and -0.110 (-0.203, -0.017) respectively.

Methodology: 

The authors conducted a search on Medline and Science Direct databases for studies investigating the relationship between CCT and IOP. Search key words included: central corneal thickness, intraocular pressure, glaucoma, ocular hypertension, exfoliative glaucoma, applanation tonometry, pachymetry, primary open-angle glaucoma, Goldmann applanation tonometry. In addition, a manual search of “The Year Book of Ophthalmology” Journals 2004 to 2006, Issues in the Southern Medical University Library, English language section, was done.

Authors applied the following exclusion criteria: non-English media studies; studies done before 2005; case series and case reviews; studies involving treatment protocols or surgical techniques; studies comparing glaucomatous eyes with other conditions such as diabetes, hypertension or cardiovascular disease as secondary variables;  studies with children as study subjects; studies with animals as study subjects.

Inclusion criteria applied was as follows: case control studies; observational studies;  studies involving primary open angle glaucoma, ocular hypertension, pseudo exfoliation glaucoma and normal tension glaucoma combined and results reported as glaucoma with means of central corneal thickness and intraocular pressure as outcomes; intraocular pressure measured only by Goldmann applanation tonometry, an average of at least two measurements; and patients with no corneal dystrophy, edema, scar or prior history of corneal surgery.

A single investigator sifted studies for inclusion in the review and extracted data of included studies. The authors note that since the principal summary measure was difference in means, the effect size was standard mean difference. The meta-analysis was done in two parts: firstly, meta-analysis of the difference in means of IOP of the patients and controls; and secondly, meta-analysis of the difference in means of the CCT of the patients and that of controls.

For both meta-analyses, a test of heterogeneity was done with weighting of individual studies and a sum of weights for all the 13 studies included by way of a forest plot. Meta regression was carried out to explore the sources of heterogeneity. Evaluation of publication bias was done using a funnel plot, in addition the Eggers and Beggs tests were employed by the authors. Sensitivity analysis was also done to assess the agreement of the summary results.

Applicability/external validity:

The authors noted in the review that it has been established that glaucomatous eyes tend to have thinner CCT and higher IOP compared to normal eyes. However, based on the meta-regression for CCT studies, revealed that age had an influence on the findings of the review.

Geographic focus

The authors included studies from Asia, Europe and North America. However, they did not discuss the applicability of findings to low- and middle- income stings.

Summary of quality assessment:

Overall, low confidence was attributed in the conclusions about the effects of this review as important limitations were identified. One major limitation includes the search strategy, which was not comprehensive enough to ensure that all relevant studies were included in the review. In addition, the authors did not use rigorous methods to screen studies and extract data of included studies, impacting on the validity of overall findings.