To determine the association between diabetes mellitus (DM) and primary open-angle glaucoma (POAG).
This is a systematic review and meta-analysis of case-control and cohort studies. The literature search included two databases (PubMed and Embase) and the reference lists of the retrieved studies. Separate meta-analyses for case-control studies and cohort studies were conducted using random-effects models, with results reported as adjusted odds ratios (ORs) and relative risks (RRs), respectively.
Thirteen studies—seven case-control studies and six population-based cohort studies—were included in this meta-analysis. The pooled RR of the association between DM and POAG based on the risk estimates of the six cohort studies was 1.40 (95% CI, 1.25–1.57). The pooled OR of the association between DM and POAG based on the risk estimates of the seven case-control studies was 1.49 (95% CI, 1.17–1.88). There was considerable heterogeneity among the case-control studies that reported an association between DM mellitus and POAG (P<0.001) and no significant heterogeneity among the cohort studies (P = 0.377). After omitting the case-control study that contributed significantly to the heterogeneity, the pooled OR for the association between DM and POAG was 1.35 (95% CI, 1.06–1.74).
Citation: Zhou M, Wang W, Huang W, Zhang X (2014) Diabetes Mellitus as a Risk Factor for Open-Angle Glaucoma: A Systematic Review and Meta-Analysis. PLoS ONE 9(8): e102972. https://doi.org/10.1371/journal.pone.0102972
Editor: Demetrios Vavvas, Massachusetts Eye & Ear Infirmary, Harvard Medical School, United States of America
Received: February 9, 2014; Accepted: June 24, 2014; Published: August 19, 2014
Copyright: © 2014 Zhou et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the National Natural Science Foundation of China (81170849, 81371008). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Glaucoma is a significant cause of irreversible blindness worldwide . Primary open angle glaucoma (POAG) is the most common type . Older age , a family history of POAG , myopia , central corneal thickness , and ocular hypertension ,  are relatively consistent risk factors for POAG.
Diabetes mellitus (DM) is a serious and increasingly prevalent health problem worldwide due to lifestyle changes and an aging population. DM is associated with severe acute and chronic complications, which negatively influence both the quality of life and the survival of affected individuals . The prevalence of diabetes among all age groups worldwide was estimated to be 2.8% in 2000 and 4.4% in 2030 . Some studies also found that DM is another possible risk factor for POAG , , , . However, the relationship between DM and POAG is controversial , .
An earlier meta-analysis in 2004 indicated that DM is associated with an increased risk of developing POAG . However, that meta-analysis of seven cross-sectional and five case-control studies of DM and the risk of POAG did not include any cohort studies. The major drawback of cross-sectional studies is that they cannot establish clear temporal relationship between exposure and outcome. Prospective cohort studies would be a good way to assess the relationship between exposure to diabetes and development of POAG. Since then, much larger-scale epidemiological evidence, especially cohort studies, of an association between DM and POAG has been reported , , , , , , . However, these epidemiological studies of the relationship between DM and POAG were somewhat contradictory and inconclusive, and two studies reported discordant results , .
To provide a more accurate estimate of the association between DM and POAG, we conducted a meta-analysis of all case-control and cohort studies involving DM and POAG.
We conducted a systematic review and meta-analysis to examine the association of DM with POAG based on the Meta-analysis of Observational Studies in Epidemiology guidelines . We conducted a computerized search of the PubMed and Embase up to Feb 08, 2014. The following terms, adapted for each database, were used for the searches: diabetes mellitus, DM, impaired glucose tolerance, hyperglycemia, insulin resistance, insulin secrete dysfunction, glaucoma, intraocular hypertension, intraocular pressure, ocular hypertension. To achieve maximum sensitivity, limits or filters were not placed on the searches. No language restrictions were imposed. A manual search was performed by checking the reference lists of the original reports.
Inclusion and exclusion criteria
The full-length articles were required to meet the following inclusion and exclusion criteria for the meta-analysis. The inclusion criteria were: (1) they had a case-control or a cohort design, (2) the exposure of interest was DM, (3) the outcome of interest was POAG, and (4) the odds ratios (ORs) or the relative risk (RR) estimates with their 95% confidence intervals (CIs) (or data to calculate them: raw data, P value, and/or variance estimates) were reported.
We excluded the following: (1) studies involving secondary glaucoma or angle-closure glaucoma, (2) studies without a detailed description of the POAG assessment, (3) crude data that could not calculate the adjusted ORs or the adjusted RRs. When multiple publications from the same study population were available, we checked for duplicate analyses and included only the most recent publication. We excluded one study that reported only crude data because the ORs and their 95% CIs could not be calculated.
Data extraction and quality assessment
Two authors (Z.M.W. and W.W.) independently extracted the following data from each publication: publication data (author, year of publication, and country of the population studied); study design (cohort study or case-control study); methods of DM ascertainment (self-report, medical records, and blood glucose measurement); definition of glaucoma; type of DM; participant's age; study population; number of cases and controls (for case-control studies); number of exposed and comparison group (for cohort studies); number of cases (for cohort studies); number of DM patients (for case-control studies); follow-up period (for cohort studies); summary estimates and corresponding 95% CI, and confounding factors adjusted for.
Two reviewers independently assessed the quality of each study using the Newcastle-Ottawa Scale (NOS) . The NOS consists of three parameters of quality: selection, comparability, and exposure (case-control studies) or outcome (cohort studies). The NOS assigns a maximum of four points for selection, two points for comparability, and three points for exposure/outcome. Therefore, nine points reflect the highest quality. Any discrepancies were addressed by a joint re-evaluation of the original article with a third reviewer.
The data from the cohort studies and the case-control studies were analyzed separately. The RR was used as a common measure of the association between DM and the risk of POAG in the cohort studies. The incidence rate ratio (IRR) and the hazard ratio (HR) were considered as RRs, and the pooled adjusted RRs with the corresponding 95% CIs were calculated. For the case-control studies, the pooled adjusted ORs with the corresponding 95% CIs were calculated. The maximally adjusted RRs or ORs were used to assess the association between DM and POAG. Considering the differences in the characteristics of the study groups and the variation in the sample sizes, we assumed that heterogeneity was present even when no statistical significance was identified. Thus, we combined the data using a random-effects model. Statistical heterogeneity between the studies was evaluated using Cochran's Q test and the I2 statistic. For the Q statistic, P<0.05 was considered to indicate statistically significant heterogeneity. For the case-control studies, a sensitivity analysis was also conducted, in which one study at a time was removed and the rest were analyzed to estimate whether a single study could have markedly affected the results. For the cohort studies, stratified analysis was performed according to the type of DM, type of effect measure, and geographic area. To detect publication biases, we calculated Begg and Egger measures. P<0.05 was considered statistically significant in the test for overall effect. The analysis was conducted using the Stata software package (Version 11.0; Stata Corp., College Station, TX).
Identification and selection of reports for the systematic review and meta-analysis
We identified 4616 articles from the database search. After the removal of 654 duplicate publications, there were 3962 studies. In total, 78 articles were retrieved for full-text review. Finally, we identified seven case-control studies , , , , , ,  and six cohort studies , , , , ,  that presented results on DM and the risk of POAG (Figure 1). One cohort study was excluded because it only provided an unadjusted but not adjusted RR .
POAG = primary open-angle glaucoma; DM = diabetes mellitus.
Characteristics of the case-control studies
The main characteristics of the case-control studies are presented in Table 1. The studies were published between 1987 and 2009. Three studies originated from the United States, one from Korea, one from the Congo, and two from Europe (France and Denmark). In total, 11,472 cases and 75,631 controls were included in this meta-analysis. Four studies reported a positive association between DM and the incidence of POAG. The definition of POAG varied across the studies. Four studies included an increased IOP in their case definition of POAG , , , , and one study included a history of glaucoma treatment . Six studies included disc cupping abnormalities (a measure of optic nerve damage) and the visual field test , , , , , . DM was ascertained by self-reporting , , , , medical records , , and blood glucose level .
Characteristics of the cohort studies
The main characteristics of the cohort studies are listed in Table 2. The studies were published between 2000 and 2011. Three studies originated from the United States, one from the Netherlands, one from the Barbados, and one from the United Kingdom. A total of 46,360 cases of POAG in a cohort of 3,393,011 individuals were included in this meta-analysis. Three studies reported a positive association between DM and POAG. The definition of POAG varied across the studies. Two studies included an increased intraocular pressure (IOP) in their case definition of POAG , , and all the studies included a visual field test and disc cupping abnormalities , , , , , . DM was ascertained by self-reporting , , , medical records , and the patient's blood glucose level , .
Quality assessment results
With regard to the case-control studies, all the studies were of high quality (NOS score >6), with an average NOS score of 7.43. All the cohort studies were also of high quality (NOS score >6), with an average NOS score of 7.83. The most common bias was ascertainment of exposure, with four studies assessing DM by self-reporting or medical records.
Pooled estimates of the association between DM and POAG analysis with the case-control studies
The pooled OR for the seven case-control studies was 1.49 (95% CI, 1.17–1.88) under the random-effects model. Figure 2 shows the multivariate ORs for each study separately and for the seven case-control studies combined. There was statistically significant heterogeneity among the seven case-control studies (I2 = 82.1%; P<0.001). The sensitivity analysis showed that Welinder's study  substantially influenced the pooled OR (Table 3). After excluding this study, the pooled OR was 1.35 (95% CI, 1.06–1.74), with no evidence of heterogeneity (I2 = 37.1%; P = 0.159).
OR = odds ratio; CI = confidence interval; DM = diabetes mellitus; POAG = primary open-angle glaucoma.
Analysis with the cohort studies
The pooled RR for the five cohort studies was 1.40 (95% CI, 1.25–1.57) under the random-effects model, and the heterogeneity was statistically insignificant (I2 = 6.2%; P = 0.377). Figure 3 shows the multivariate RRs for each study separately and for the six cohort studies combined. Subgroup analyses according to type of DM, type of effect measure, and geographic area were performed to examine the impacts of these factors on the association (Table 4). No evidence of heterogeneity was observed within any subgroup. A significant positive association between DM and POAG was observed in all subgroups.
RR = risk ratio; CI = confidence interval; DM = diabetes mellitus; POAG = primary open-angle glaucoma.
The assessment of publication bias (Figure 4) using Begg's test (P = 0.548) and Egger's test (P = 0.939) doesn't show evidence of publication bias in the case-control studies. Publication bias also was not statistically detected by Begg's test (P = 0.452) and Egger's test (P = 0.923) in the cohort studies (Figure 5).
Begg's regression asymmetry test (P = 0.548). OR = odds ratio; DM = diabetes mellitus; POAG = primary open-angle glaucoma.
Begg's regression asymmetry test (P = 0.452). RR = risk ratio; DM = diabetes mellitus; POAG = primary open-angle glaucoma.
Although several risk factors for the development of POAG have been evaluated, this is a field of ongoing investigation . DM is a possible risk factor for POAG. The current literature does not provide a definitive link between DM and POAG. Hence, the purpose of this systematic review and meta-analysis was to summarize all the available relevant evidence with reference to the relationship between DM and POAG. The findings from this meta-analysis support the evidence for a positive relation between DM and an increased risk of POAG. The results, based on 13 epidemiological studies, are consistent with those of a previous meta-analysis . However, unlike previous studies, we included several newly published case-control and cohort studies, and we excluded cross-sectional studies in this updated meta-analysis. This allowed for a greater number of subjects and, hence, a more detailed and accurate risk estimation than in prior meta-analyses.
The results from the case-control and cohort studies were quite similar. The findings from this meta-analysis showed that compared with non-diabetic individuals, individuals with DM have an approximately 1.4-fold increased risk of developing POAG in cohort studies. The results from the case-control studies showed that they have an about 49% increased odds of developing POAG compared with individuals without DM. Moreover, the results from the cohort studies subgroup analyses were quite similar, with a significant association found between DM and POAG in all the subgroups. The results were not substantially affected by the DM type. In the current meta-analysis, no study was excluded based on sample size, and both Begg's test and Egger's test doesn't show evidence of publication bias. Thus, the results of this meta-analysis are robust.
Heterogeneity is often a concern in a meta-analysis. Substantial heterogeneity was observed in case-control studies, which was expected given the between-study variation, such as inconsistent data collecting methods, different ethnic populations, and different sample size. In the leave-one-out sensitivity procedure, we found that removing Welinder's study  from the case-control studies altered the results. This study had the largest sample size and the greatest differences in the strength of the association. These factors may have been the main sources of the heterogeneity. After this study was excluded, the remainder showed a positive association between DM and POAG, with no evidence of heterogeneity. However, the high quality of Welinder's study showed the positive relation between DM and POAG, which is consist with the pooled outcome. Thus, despite exist the heterogeneity in case-control studies, the results of this meta-analysis are still robust. Little evidence of heterogeneity was observed in the subgroups of the cohort studies.
There seems to be a direct relationship between DM and POAG. Several hypotheses on biological links between DM and POAG have been proposed. First, there is a growing body of evidence that the presence of long-standing hyperglycemia, along with lipid anomalies, may increase the risk of neuronal injury from stress . In particular, laboratory data have provided robust evidence for such an association , . Second, studies showed that diabetic eyes have a reduced capacity to auto-regulate blood flow and that they exhibit decreased retinal blood flow . As a result, they show relative hypoxia and overexpression of hypoxia-inducible factor-1 (HIF-1α , . Importantly, levels of HIF-1α increased in ganglion cells, in the retina, and in the optic nerve head of human glaucomatous eyes in response to elevated IOP . These might be another important association between DM and POAG. Third, the observed association between DM and POAG may be explained by the remodeling of the connective tissue of the optic nerve head. The remodeling might reduce compliance at the trabecular meshwork and the lamina cribrosa, resulting in increased IOP and greater mechanical stress on the optic nerve head, respectively , . Research has demonstrated that diabetes can exacerbate connective tissue remodeling and amplify these biomechanical changes . More importantly, the Barbados Eye Study had found that diabetes was a risk factor for increased IOP in follow-up . Genetic factors and diabetes-related autonomic dysfunction are likely to play a role of this association . Further research is needed to better understand these phenomena.
The strengths of the present study are as follows: First, our meta-analysis of 13 studies involving a relatively large number of cases and participants enhanced the power to detect a significant association and provided more reliable estimates. Second, most of the included studies evaluated multiple potential confounding factors, some of which were considered to be risk factors for POAG, such as race, hypertension, family history of glaucoma, smoking, and body mass index. Third, the large majority of the studies included were of high quality.
As with any meta-analysis of observational studies, there are several potential limitations with regard to the results. First, significant heterogeneity existed in the case-control studies. We found that one study was the main source of this heterogeneity, but when we exclude this study, the remainder studies still showed the similar association between DM and POAG. Second, in the cohort studies, the effect size of one study was the IRR; it was the HR in another study but not the RR effect size. However, the positive association between DM and POAG was consistently observed in RRs and non-RRs subgroup. Third, DM was self-reported in some studies, and this may have introduced a recall bias. Fourth, in general, publication bias is a major problem in published studies and in meta-analyses of published studies. Publication bias may be an issue because studies that report statistically significant results are more likely to be published than studies that report non-significant results, and this could have distorted the findings of our meta-analyses . However, the results obtained from the funnel plot analysis and the formal statistical tests did not provide evidence for such bias. Finally, no cohort studies addresses detection bias and it is another limitation in this meta-analysis.
Our findings have important public health implications. Glaucoma is an important cause of irreversible blindness worldwide. Controversy continues regarding the effects of DM on the risk of POAG. The findings from our study aimed at addressing this issue and resolving the inconsistency are both important and timely. In summary, the results of this meta-analysis point to a significant association between DM and the risk of POAG. Further studies are needed to elucidate the exact underlying mechanisms linking DM with POAG.
Conceived and designed the experiments: MWZ XLZ. Performed the experiments: MWZ WW. Analyzed the data: WW WBH. Contributed reagents/materials/analysis tools: WW MWZ. Wrote the paper: MWZ.
- 1. Quigley HA (1996) Number of people with glaucoma worldwide. Br J Ophthalmol 80: 389–393.
- 2. Peters D, Bengtsson B, Heijl A (2013) Lifetime Risk of Blindness in Open-Angle Glaucoma. Am J Ophthalmol
- 3. Leske MC, Connell AM, Schachat AP, Hyman L (1994) The Barbados Eye Study. Prevalence of open angle glaucoma. Arch Ophthalmol 112: 821–829.
- 4. Sommer A, Tielsch JM (1996) Risk factors for open-angle glaucoma: the Barbados Eye Study. Arch Ophthalmol 114: 235.
- 5. Topouzis F, Wilson MR, Harris A, Founti P, Yu F, et al. (2011) Risk factors for primary open-angle glaucoma and pseudoexfoliative glaucoma in the Thessaloniki eye study. Am J Ophthalmol 152: 219–228.
- 6. Francis BA, Varma R, Chopra V, Lai MY, Shtir C, et al. (2008) Intraocular pressure, central corneal thickness, and prevalence of open-angle glaucoma: the Los Angeles Latino Eye Study. Am J Ophthalmol 146: 741–746.
- 7. Heijl A, Leske MC, Bengtsson B, Hyman L, Bengtsson B, et al. (2002) Reduction of intraocular pressure and glaucoma progression: results from the Early Manifest Glaucoma Trial. Arch Ophthalmol 120: 1268–1279.
- 8. Kass MA, Heuer DK, Higginbotham EJ, Johnson CA, Keltner JL, et al. (2002) The Ocular Hypertension Treatment Study: a randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary open-angle glaucoma. Arch Ophthalmol 120: 701–830, 701-713, 829-830.
- 9. Vigneri P, Frasca F, Sciacca L, Pandini G, Vigneri R (2009) Diabetes and cancer. Endocr Relat Cancer 16: 1103–1123.
- 10. Wild S, Roglic G, Green A, Sicree R, King H (2004) Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 27: 1047–1053.
- 11. Dielemans I, de Jong PT, Stolk R, Vingerling JR, Grobbee DE, et al. (1996) Primary open-angle glaucoma, intraocular pressure, and diabetes mellitus in the general elderly population. The Rotterdam Study. Ophthalmology 103: 1271–1275.
- 12. Chopra V, Varma R, Francis BA, Wu J, Torres M, et al. (2008) Type 2 diabetes mellitus and the risk of open-angle glaucoma the Los Angeles Latino Eye Study. Ophthalmology 115: 227–232.
- 13. Pasquale LR, Kang JH, Manson JE, Willett WC, Rosner BA, et al. (2006) Prospective study of type 2 diabetes mellitus and risk of primary open-angle glaucoma in women. Ophthalmology 113: 1081–1086.
- 14. Klein BE, Klein R, Jensen SC (1994) Open-angle glaucoma and older-onset diabetes. The Beaver Dam Eye Study. Ophthalmology 101: 1173–1177.
- 15. de Voogd S, Ikram MK, Wolfs RC, Jansonius NM, Witteman JC, et al. (2006) Is diabetes mellitus a risk factor for open-angle glaucoma? The Rotterdam Study. Ophthalmology 113: 1827–1831.
- 16. Tielsch JM, Katz J, Quigley HA, Javitt JC, Sommer A (1995) Diabetes, intraocular pressure, and primary open-angle glaucoma in the Baltimore Eye Survey. Ophthalmology 102: 48–53.
- 17. Bonovas S, Peponis V, Filioussi K (2004) Diabetes mellitus as a risk factor for primary open-angle glaucoma: a meta-analysis. Diabet Med 21: 609–614.
- 18. Wise LA, Rosenberg L, Radin RG, Mattox C, Yang EB, et al. (2011) A prospective study of diabetes, lifestyle factors, and glaucoma among African-American women. Ann Epidemiol 21: 430–439.
- 19. Newman-Casey PA, Talwar N, Nan B, Musch DC, Stein JD (2011) The relationship between components of metabolic syndrome and open-angle glaucoma. Ophthalmology 118: 1318–1326.
- 20. Ellis JD, Evans JM, Ruta DA, Baines PS, Leese G, et al. (2000) Glaucoma incidence in an unselected cohort of diabetic patients: is diabetes mellitus a risk factor for glaucoma? DARTS/MEMO collaboration. Diabetes Audit and Research in Tayside Study. Medicines Monitoring Unit. Br J Ophthalmol 84: 1218–1224.
- 21. Welinder LG, Riis AH, Knudsen LL, Thomsen RW (2009) Diabetes, glycemic control and risk of medical glaucoma treatment: A population-based case-control study. Clin Epidemiol 1: 125–131.
- 22. Motsko SP, Jones JK (2008) Is there an association between hypothyroidism and open-angle glaucoma in an elderly population? An epidemiologic study. Ophthalmology 115: 1581–1584.
- 23. Leske MC, Wu SY, Hennis A, Honkanen R, Nemesure B (2008) Risk factors for incident open-angle glaucoma: the Barbados Eye Studies. Ophthalmology 115: 85–93.
- 24. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, et al. (2000) Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 283: 2008–2012.
Wells GA, Shea B, D OC The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses.: Ottawa, Canada: Dept of Epidemiology and Community Medicine, University of Ottawa.
- 26. Wilson MR, Hertzmark E, Walker AM, Childs-Shaw K, Epstein DL (1987) A case-control study of risk factors in open angle glaucoma. Arch Ophthalmol 105: 1066–1071.
- 27. Uhm KB, Shin DH (1992) Glaucoma risk factors in primary open-angle glaucoma patients compared to ocular hypertensives and control subjects. Korean J Ophthalmol 6: 91–99.
- 28. Kaimbo DK, Buntinx F, Missotten L (2001) Risk factors for open-angle glaucoma: a case-control study. J Clin Epidemiol 54: 166–171.
- 29. Charliat G, Jolly D, Blanchard F (1994) Genetic risk factor in primary open-angle glaucoma: a case-control study. Ophthalmic Epidemiol 1: 131–138.
- 30. Katz J, Sommer A (1988) Risk factors for primary open angle glaucoma. Am J Prev Med 4: 110–114.
- 31. Jiang X, Varma R, Wu S, Torres M, Azen SP, et al. (2012) Baseline risk factors that predict the development of open-angle glaucoma in a population: the Los Angeles Latino Eye Study. Ophthalmology 119: 2245–2253.
- 32. Rivera JL, Bell NP, Feldman RM (2008) Risk factors for primary open angle glaucoma progression: what we know and what we need to know. Curr Opin Ophthalmol 19: 102–106.
- 33. Kong GY, Van Bergen NJ, Trounce IA, Crowston JG (2009) Mitochondrial dysfunction and glaucoma. J Glaucoma 18: 93–100.
- 34. Toda N, Nakanishi-Toda M (2007) Nitric oxide: ocular blood flow, glaucoma, and diabetic retinopathy. Prog Retin Eye Res 26: 205–238.
- 35. Neufeld AH, Hernandez MR, Gonzalez M (1997) Nitric oxide synthase in the human glaucomatous optic nerve head. Arch Ophthalmol 115: 497–503.
- 36. Clermont AC, Bursell SE (2007) Retinal blood flow in diabetes. Microcirculation 14: 49–61.
- 37. Arjamaa O, Nikinmaa M (2006) Oxygen-dependent diseases in the retina: role of hypoxia-inducible factors. Exp Eye Res 83: 473–483.
- 38. Danylkova NO, Pomeranz HD, Alcala SR, McLoon LK (2006) Histological and morphometric evaluation of transient retinal and optic nerve ischemia in rat. Brain Res 1096: 20–29.
- 39. Tezel G, Wax MB (2004) Hypoxia-inducible factor 1alpha in the glaucomatous retina and optic nerve head. Arch Ophthalmol 122: 1348–1356.
- 40. Roberts MD, Grau V, Grimm J, Reynaud J, Bellezza AJ, et al. (2009) Remodeling of the connective tissue microarchitecture of the lamina cribrosa in early experimental glaucoma. Invest Ophthalmol Vis Sci 50: 681–690.
- 41. Johnson EC, Morrison JC, Farrell S, Deppmeier L, Moore CG, et al. (1996) The effect of chronically elevated intraocular pressure on the rat optic nerve head extracellular matrix. Exp Eye Res 62: 663–674.
- 42. Francis-Sedlak ME, Uriel S, Larson JC, Greisler HP, Venerus DC, et al. (2009) Characterization of type I collagen gels modified by glycation. Biomaterials 30: 1851–1856.
- 43. Hennis A, Wu SY, Nemesure B, Leske MC (2003) Hypertension, diabetes, and longitudinal changes in intraocular pressure. Ophthalmology 110: 908–914.
- 44. Mapstone R, Clark CV (1985) Prevalence of diabetes in glaucoma. Br Med J (Clin Res Ed) 291: 93–95.
- 45. Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR (1991) Publication bias in clinical research. Lancet 337: 867–872.