FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1562024881> ?p ?o ?g. }

• W1562024881 endingPage "448" @default.
• W1562024881 startingPage "443" @default.
• W1562024881 abstract "Purpose: This study set out to map the associations between retinal lesions, visual acuity (VA) and the presence of clinically significant macular oedema (CSMO) in diabetes subjects. Methods: This cross-sectional study comprised 656 type 1 and 328 type 2 diabetes subjects undergoing retinopathy screening in the County of North Jutland, Denmark. Numbers of specific retinal lesions were quantified from retinal photographic recordings. Associations between CSMO, number of specific retinal lesions and VA were established. The percentages of eyes with CSMO ascribed to retinal lesions were calculated. Results: The presence of CSMO, number of specific retinal lesions and VA were all significantly associated. The parameter with the highest statistical association with CSMO measured by Spearman’s correlation coefficient was hard exudates (type 1: 0.524; type 2: 0.715), followed by microaneurysms (type 1: 0.298; type 2: 0.508), retinal haemorrhages (type 1: 0.227; type 2: 0.595), cottonwool spots (type 1: 0.207; type 2: 0.259) and VA (type 1: − 0.137; type 2: − 0,175). Conclusions: All retinal lesions are significantly associated with CSMO and together can predict for up to 42.3% (in type 1 diabetes) and 64.3% (in type 2 diabetes) of CSMO cases. Diabetic maculopathy is a leading cause of reduced visual acuity (VA) and blindness (National Society to Prevent Blindness 1980; Klein & Klein 1995). In diabetes subjects, maculopathy is unambiguously defined as a focal or diffuse thickening of the central macular region and is designated ‘clinically significant macular oedema’ (CSMO) (Early Treatment Diabetic Retinopathy Study [ETDRS] Group 1987). To date, the presence of CSMO has been measured according to a simplified retinopathy scale (Wilkinson et al. 2003) and the highest prevalence rate found in severe simplex retinopathy (Wisconsin Epidemiologic Study of Diabetic Retinopathy 1984, 1989; Sjølie 1985; Agardh et al. 1989a, 1989b;Knudsen et al. 2006). The use of such a retinopathy scale is convenient and relatively simple, but it may exclude detailed information about the associations between CSMO and the numbers of various retinal lesions. The present study therefore explores the relative influence of individual retinal lesions and the combined influence of all retinal lesions on the prevalence of CSMO. In the period between 1 April 2000 and 30 April 2004, 656 subjects with type 1 diabetes and 328 subjects with type 2 diabetes underwent diabetic retinopathy (DR) screening in the County of North Jutland, Denmark. The type 1 diabetes subjects came almost exclusively from central Aalborg (an urban area in the county) and represented 70–75% of all adult type 1 diabetes subjects in this region. The type 2 diabetes subjects were enrolled from the entire county, mainly as a result of the malregulation of the registration of diabetes subjects. These individuals accounted for < 5% of registered type 2 diabetes subjects in the county. The majority of participants were White. The method of DR screening has been described previously (Knudsen et al. 2006) and is briefly summarized below. Standardized VA was measured using a decimal progression scale. The retina was photographically recorded using a digital camera (Zeiss DSC 420, resolution 1524*1012; Carl Zeiss GmbH, Oberkochen, Germany). One photograph was centred on the macular region; the other included the optic disc and the nasal part of the retina (Fig. 1). Representative photographic recordings of (A) the macular region and surrounding retina, and (B) the nasal part of the retina. The digitized retinal recordings were electronically transferred to the Department of Ophthalmology for additional evaluation. The retinal recordings were examined on a high-resolution screen (Nokia 446 PRO; Nokia Corp., Espoo, Finland) and quantitatively assessed for several retinal lesions by Knudsen & Skriver (2006). In cases of doubt or when detectable pathology was present in the macular region, subjects were called for a clinical examination (22% of all subjects) to decide the presence of CSMO. A high number of patients were recalled for an additional clinical examination in order to exclude CSMO. The quality of the photographic recording was assessed by the photographer and the evaluating ophthalmologists, who concluded that fewer than 10 subjects had non-gradable retinopathy. If a subject failed to appear, he or she was summoned again, resulting in 100% participation. The type of diabetes was defined according to usual clinical practice. Fasting plasma C-peptide or GAD (glutamic acid decarboxylase) antibodies were not measured routinely in our diabetes outpatient clinic, and we did not routinely carry out genetic testing for maturity diabetes of the young (MODY). Subjects with type 1 diabetes were usually < 30 years of age, of normal body weight and had a history of ketoacidosis. Subjects with type 2 diabetes were usually > 30 years of age at diagnosis, of normal weight or overweight at diagnosis and without a history of ketoacidosis. The majority of type 2 diabetes subjects had been treated with antidiabetes tablets for ≥ 1 year from the debut of diabetes. A few patients with a type 2 phenotype had been treated with insulin from debut because of high primary plasma glucose levels. The presence of CSMO was established by a clinical, 3-dimensional, stereoscopic evaluation of the macular region in accordance with ETDRS criteria (ETDRS 1987). Initially, we used a slit-lamp together with a 78-D lens. In cases of doubt, we used the Ocular–Mainster high-magnification lens (Ocular Instruments, Inc., Bellevue, WA, USA) designed for detecting and treating macular problems. The classification was carried out by two ophthalmology consultants who had been previously trained and assessed in the classification of macular oedema (Knudsen & Skriver 2006). We defined the macular region as the region within the two temporal vascular arcades with a temporal limitation two disc diameters from the foveola. Numbers of retinal lesions were quantitatively assessed as follows: microaneurysms (0–50); retinal haemorrhages (0–20); hard exudates (0–50), and cottonwool spots (0–20). We also quantitatively assessed the prevalence of white vessels, fibrous tissue, venous beading, venous loop, double-contoured vessels, intraretinal microvascular abnormalities, neovascularization of the disc and elsewhere, preretinal bleedings and vitreous haemorrhages. However, because of the low frequencies of the latter types, they were subsequently excluded from the statistical analysis. The relative occurrences of the various lesions in the macular region were explored in a sub-study which examined photographic recordings in 100 randomly selected right eyes. Numbers of lesions inside and outside the macular region were assessed. Data were collated from several sources. Typing errors and mismatch of cases were controlled for by visual inspections of scatterplots etc. and by the validation of selected cases. Multiple logistic regressions were used to calculate prevalences, odds ratios and confidence intervals adjusted for various parameters. The relationship between the prevalence of CSMO and a specific retinal parameter was established by logistic regression assuming a log-linear relationship between the odds of CSMO and the retinal parameter. Associations between variables were assessed by non-parametric correlations (Spearman’s rho). To avoid possible bias in the statistical analysis, we only included CSMO in the right eye. Statistical analysis was carried out using spss 12.0.2 for Windows (SPSS, Inc., Chicago, IL, USA) and R 2.1.1 (R Development Core Team 2005). The prevalence of CSMO increased gradually with the number of registered retinal lesions (Fig. 2). The prevalence and curve slope (number of lesions versus logit-transformed prevalence) were statistically equivalent for type 1 and type 2 diabetes subjects with respect to hard exudates and cottonwool spots. For microaneurysms and retinal haemorrhages, the prevalence of CSMO was found to be increased in type 2 diabetes subjects. The presence of each type of retinal lesion significantly increased the prevalence of CSMO, even when corrected for age and duration of diabetes. Similarly, the inverse association between VA and CSMO was found to be significant, both for crude data and when corrected for age and duration of diabetes. When corrected for the presence of other retinal lesions, this association consistently prevailed for hard exudates in both types of diabetes and for microaneurysms and VA in type 1 diabetes. Associations between the number of retinal lesions and the prevalence of clinically significant macular oedema in type 1 and type 2 diabetes for (A) hard exudates, (B) microaneurysms, (C) cottonwool spots and (D) retinal haemorrhages. The prevalence of CSMO was found to be significantly and inversely associated with VA; the slope of this curve was significantly steeper for type 2 than for type 1 diabetes subjects (Fig. 3). The various retinal lesions were all significantly and inversely associated with VA (Table 1). Associations between visual acuity and the prevalence of clinically significant macular oedema in type 1 and type 2 diabetes. The associations between CSMO, hard exudates, microaneurysms and retinal haemorrhages were relatively high. The associations with VA and cottonwool spots were lower. The strongest association between retinal lesions and the presence of CSMO measured by Spearman’s rho was observed for hard exudates (type 1: 0.524; type 2: 0.715); this was followed by associations between CSMO and retinal haemorrhages (type 1: 0.227; type 2: 0.595), microaneurysms (type 1: 0.298; type 2: 0.508), cottonwool spots (type 1: 0.207; type 2: 0.259) and VA (type 1: − 0.137; type 2: − 0.175) (Table 1). Predicting CSMO status from retinal parameters gives a sensitivity of 42.3% and specificity of 98.7% for type 1 diabetes subjects and a sensitivity of 64.3% and specificity of 98.3% for type 2 diabetes subjects, using a 50/50 cut point. Among all type 1 diabetes subjects, 45.7% revealed no diabetic retinal manifestations and 54.3% had some type of diabetic retinal lesion. The corresponding values for type 2 diabetes subjects were 60.7% and 39.3%, respectively. Numbers of the various retinal lesions were significantly increased in CSMO subjects compared with the non-oedema group (Table 1). Visual acuity was significantly reduced in the oedema group (Table 2). The influence of the various manifestations on the prevalence of CSMO was calculated for type 1 and type 2 diabetes subjects (Table 3). In a sub-study, the numbers of retinal lesions within (in) and outside (out) the macular region were assessed. Results were as follows: microaneurysms (in: 4.86, out: 5.88; non-significant [NS], p = 0.339); retinal haemorrhages (in: 1.63, out: 1.88; NS, p = 0.538); cottonwool spots (in: 0.0, out: 0.2; NP, p = 0.07), and hard exudates (in: 3.95, out: 0.86; significant, p = 0.026). In this study the presence of CSMO was found to be significantly associated with all retinal lesions examined. It was most strongly associated with hard exudates, followed by microaneurysms, retinal haemorrhages, cottonwool spots and VA. In combination, the various retinal lesions accounted for 42.3% and 64.3% of all cases of CSMO in type 1 and type 2 diabetes subjects, respectively. In general, the statistical association was found to be higher in type 2 than in type 1 diabetes. This study explores the associations between the presence of CSMO, the total number of specific lesions in the photographically visualized retina and VA. Although the macular region comprises only a minor proportion of the entire retinal surface, the total number of all retinal lesions was found to be significantly associated with the presence of CSMO, suggesting that macular changes are closely associated with changes in the entire retina. This is supported by the sub-study findings, which reported similar numbers of retinal lesions within and outside the macular region, excluding hard exudates, which dominate in the macular region. Generally, the association between the various retinal lesions, VA and the prevalence of CSMO was found to be higher among type 2 diabetes subjects. This may suggest that CSMO is a more dominating feature in these subjects, as is evident in daily clinical practice. Consequently, CSMO can presumably be considered to represent part of a generalized affection of the entire diabetic retina. However, some focal variation can be seen as the result of anatomical and physiological differences in the various parts of the retina, particularly in type 2 diabetes. The study demonstrated that all the retinal lesions examined and VA were mutually and significantly associated (Table 1). Thus CSMO cannot be unambiguously characterized by one specific retinal lesion type. The composition of the various lesions seems to give a more meaningful impression of macular status, as is evident in daily clinical practice. The various retinal lesions influence the prevalence of CSMO differently. The greatest influence on the prevalence of CSMO is exerted by a single cottonwool spot, followed by retinal haemorrhages, hard exudates and microaneurysms. However, the statistical association between CSMO and the various retinal lesions was found to be strongest for hard exudates, followed by microaneurysms, retinal haemorrhages and cottonwool spots. This apparent inconsistency is caused by differences in the frequencies of the various lesions. As cottonwool spots are relatively uncommon, their influence on CSMO is less than that of hard exudates, microaneurysms or retinal haemorrhages. Several large-scale studies (Wisconsin Epidemiologic Study of Diabetic Retinopathy 1984, 1989; Sjølie 1985) have explored the prevalence of CSMO and its association with the level of retinopathy. The present study confirms these findings but adds a more detailed description of this association because it includes a quantitative assessment of the various lesion types. Microaneurysms represent focal capillary dilation and represent the earliest clinical signs of DR (Friedenwald 1950). They are known to leak water and various substances from plasma, but must be present in certain numbers before they result in detectable retinal thickening. An intermediate association between CSMO and the presence of retinal microaneurysms, as seen in this study, therefore seems understandable. Hard exudates represent lipoprotein leakage across the microaneurismal wall (Ryan 2006), indicating relatively severe vascular damage and suggesting a pronounced leakage of water and various substances from plasma. A relatively high statistical association between CSMO and the presence of hard exudates, as seen in this study, is therefore to be expected. Cottonwool spots represent inner capillary closure and therefore a relatively severe affection of the retinal perfusion system (Ashton & Henkind 1965). They might therefore be expected to be strongly associated with the presence of CSMO. In this study, one cottonwool spot increased the prevalence of CSMO more than any other lesion type, but as this manifestation occurs relatively rarely, the statistical significance of this association was comparatively weak. Although VA testing was performed under standardized conditions, it is subject to various limitations and can only document relatively pronounced changes in macular function. The relatively low and inverse statistical association between VA and prevalence of CSMO therefore seems understandable. It highlights the fact that the various retinal lesions give a more detailed impression of macular status than does VA. As all the retinal lesions can only account for up to 42.3% and 64.3% of all cases of CSMO in type 1 and type 2 diabetes, respectively, identifying additional causes of this clinical manifestation would seem relevant. In a previous study we explored the influence of metabolic control and blood pressure on the presence of CSMO and found a very low statistical association (Knudsen et al. 2006), which was in accordance with findings by Williams et al. (2004). One possible model of interpretation refers to the vascular system of the eyes (i.e. the retinal and choroidal vasculature). Affection of the retinal capillaries includes the various retinal lesions described in this study, which account for 42.3% and 64.3% of CSMO in type 1 and type 2 diabetes subjects, respectively. The choroid (and the pigment epithelium) is usually assumed to be involved in the control of water in the retina, which is affected in patients with CSMO. A dysfunction of the outer part of the blood–retina barrier may therefore possibly be involved in the development of CSMO and may be of interest in future studies to improve our understanding of CSMO. Other parameters, such as size of lesion, the depth at which it is located and its distance from the foveola may also be of importance (Hove et al. 2006; Massin et al. 2006; Knudsen 2007; Neubauer et al. 2007). Likewise, the influence of factors relating to vascular perfusion, such as nocturnal hypotension (Hayreh 2008) and slow inflammatory processes (Knudsen 2004), may also be worth investigating." @default.
• W1562024881 created "2016-06-24" @default.
• W1562024881 creator A5025303610 @default.
• W1562024881 creator A5044728245 @default.
• W1562024881 creator A5053640365 @default.
• W1562024881 creator A5090218585 @default.
• W1562024881 date "2009-08-14" @default.
• W1562024881 modified "2023-09-26" @default.
• W1562024881 title "The North Jutland County Diabetic Retinopathy Study (NCDRS)" @default.
• W1562024881 cites W1535393253 @default.
• W1562024881 cites W1966206262 @default.
• W1562024881 cites W1985290610 @default.
• W1562024881 cites W1988293663 @default.
• W1562024881 cites W1995095177 @default.
• W1562024881 cites W2006567136 @default.
• W1562024881 cites W2017936203 @default.
• W1562024881 cites W2033858945 @default.
• W1562024881 cites W2035646659 @default.
• W1562024881 cites W2045338681 @default.
• W1562024881 cites W2079677326 @default.
• W1562024881 cites W2144818564 @default.
• W1562024881 cites W2171006864 @default.
• W1562024881 cites W2207135888 @default.
• W1562024881 cites W4236519347 @default.
• W1562024881 doi "https://doi.org/10.1111/j.1755-3768.2009.01555.x" @default.
• W1562024881 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/19681762" @default.
• W1562024881 hasPublicationYear "2009" @default.
• W1562024881 type Work @default.
• W1562024881 sameAs 1562024881 @default.
• W1562024881 citedByCount "3" @default.
• W1562024881 countsByYear W15620248812012 @default.
• W1562024881 countsByYear W15620248812013 @default.
• W1562024881 crossrefType "journal-article" @default.
• W1562024881 hasAuthorship W1562024881A5025303610 @default.
• W1562024881 hasAuthorship W1562024881A5044728245 @default.
• W1562024881 hasAuthorship W1562024881A5053640365 @default.
• W1562024881 hasAuthorship W1562024881A5090218585 @default.
• W1562024881 hasBestOaLocation W15620248811 @default.
• W1562024881 hasConcept C118487528 @default.
• W1562024881 hasConcept C119767625 @default.
• W1562024881 hasConcept C134018914 @default.
• W1562024881 hasConcept C2779829184 @default.
• W1562024881 hasConcept C555293320 @default.
• W1562024881 hasConcept C71924100 @default.
• W1562024881 hasConceptScore W1562024881C118487528 @default.
• W1562024881 hasConceptScore W1562024881C119767625 @default.
• W1562024881 hasConceptScore W1562024881C134018914 @default.
• W1562024881 hasConceptScore W1562024881C2779829184 @default.
• W1562024881 hasConceptScore W1562024881C555293320 @default.
• W1562024881 hasConceptScore W1562024881C71924100 @default.
• W1562024881 hasIssue "4" @default.
• W1562024881 hasLocation W15620248811 @default.
• W1562024881 hasLocation W15620248812 @default.
• W1562024881 hasOpenAccess W1562024881 @default.
• W1562024881 hasPrimaryLocation W15620248811 @default.
• W1562024881 hasRelatedWork W1563850031 @default.
• W1562024881 hasRelatedWork W2032287785 @default.
• W1562024881 hasRelatedWork W2094113608 @default.
• W1562024881 hasRelatedWork W2233866314 @default.
• W1562024881 hasRelatedWork W2383009242 @default.
• W1562024881 hasRelatedWork W2401570230 @default.
• W1562024881 hasRelatedWork W2415759662 @default.
• W1562024881 hasRelatedWork W2594752101 @default.
• W1562024881 hasRelatedWork W3036934084 @default.
• W1562024881 hasRelatedWork W4212878237 @default.
• W1562024881 hasVolume "88" @default.
• W1562024881 isParatext "false" @default.
• W1562024881 isRetracted "false" @default.
• W1562024881 magId "1562024881" @default.
• W1562024881 workType "article" @default.