Limits...
Impact of changes in metabolic control on progression to photocoagulation for clinically significant macular oedema: a 20 year study of type 1 diabetes.

Sander B, Larsen M, Andersen EW, Lund-Andersen H - Diabetologia (2013)

Bottom Line: We therefore examined the effect of changes in glycaemia and arterial blood pressure on the incidence of clinically significant macular oedema in a population of diabetic patients.A recent decrease of ≥ 0.5 percentage points or an increase in HbA1c of >0.5 percentage points per 6 months was associated with HRs of 3.04 and 1.28, respectively, compared with lesser changes in HbA1c.The effects of metabolic and haemodynamic stability on diabetic retinopathy should be examined in prospective studies.

View Article: PubMed Central - PubMed

Affiliation: Department of Ophthalmology, Glostrup Hospital, University of Copenhagen, Ndr. Ringvej 57, 2600, Glostrup, Denmark, bisan@regionh.dk.

ABSTRACT

Aims/hypothesis: Although increasing hyperglycaemia, arterial hypertension and longer duration of diabetes raise the risk of progression of diabetic retinopathy, short-term benefits in terms of improved metabolic control and lowered blood pressure have not been demonstrated. We therefore examined the effect of changes in glycaemia and arterial blood pressure on the incidence of clinically significant macular oedema in a population of diabetic patients.

Methods: We performed a retrospective review of all patients with type 1 diabetes who attended the retinopathy screening clinic at the Steno Diabetes Center from 1988 to 2008, using the endpoint referral to first photocoagulation treatment for clinically significant diabetic macular oedema. The analysis included 1,878 patients (median observation, 8 years). Changes were defined as the inter-visit change; in the case of an event the last event-free interval before referral, where the median screening interval was 6 months.

Results: Risk of progression to photocoagulation for macular oedema increased with duration of diabetes (p < 0.001), current HbA1c (p < 0.0001) and with the magnitude of changes in HbA1c (p = 0.0002) and systolic blood pressure (p < 0.0001) in a multiple regression model. A recent decrease of ≥ 0.5 percentage points or an increase in HbA1c of >0.5 percentage points per 6 months was associated with HRs of 3.04 and 1.28, respectively, compared with lesser changes in HbA1c.

Conclusions/interpretation: In this study, large recent changes in metabolic control and systolic blood pressure, irrespective of direction, were independent risk factors for progression to photocoagulation for diabetic macular oedema. The effects of metabolic and haemodynamic stability on diabetic retinopathy should be examined in prospective studies.

Show MeSH

Related in: MedlinePlus

(a–d) The effects of metabolic changes were entered into the models as continuous variables, as described in the Methods. The curves are calculated with a linear spline, i.e. a piecewise linear model, the HRs (loge scale, solid line) and the 95% CIs (broken line) are shown from a joint model similar to Table 3. (a–c) For both HbA1c and blood pressure, the minimal risk for photocoagulation was found for zero changes, with increasing risks both for decreases and increases in the variables. The curves also suggest a peak in the hazard ratio for HbA1c changes of 0.5 percentage points (corresponding to a change of 5.5 mmol/mol) (a) and 10 mmHg for systolic (b) and 5 mmHg (c) for diastolic blood pressure, with a plateau for larger changes. Due to the piecewise linear model, the exact position of the peaks and the distinct discontinuation points should be interpreted with caution. The HR for diabetes duration (d) reaches a plateau after approximately 40 years
© Copyright Policy - OpenAccess
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3824341&req=5

Fig1: (a–d) The effects of metabolic changes were entered into the models as continuous variables, as described in the Methods. The curves are calculated with a linear spline, i.e. a piecewise linear model, the HRs (loge scale, solid line) and the 95% CIs (broken line) are shown from a joint model similar to Table 3. (a–c) For both HbA1c and blood pressure, the minimal risk for photocoagulation was found for zero changes, with increasing risks both for decreases and increases in the variables. The curves also suggest a peak in the hazard ratio for HbA1c changes of 0.5 percentage points (corresponding to a change of 5.5 mmol/mol) (a) and 10 mmHg for systolic (b) and 5 mmHg (c) for diastolic blood pressure, with a plateau for larger changes. Due to the piecewise linear model, the exact position of the peaks and the distinct discontinuation points should be interpreted with caution. The HR for diabetes duration (d) reaches a plateau after approximately 40 years

Mentions: Using a discrete time proportional hazards model, current values for HbA1c, systolic blood pressure and diastolic blood pressure remained significant risk factors for progression to photocoagulation for CSME (HbA1cp < 0.0001, systolic blood pressure p = 0.0003, diastolic blood pressure p = 0.0038, duration p < 0.0001; Table 3, left column) when analysed separately and adjusted for current age, calendar period and duration (univariate model). With the same model, changes for both HbA1c, and systolic and diastolic blood pressure were all significant (p < 0.0001). In a multiple regression model containing all three covariates (joint model), only current HbA1c remained significant overall (p < 0.0001; Table 3, right column). For the effect of changes, both the change in HbA1c and systolic blood pressure were significant (HbA1c change p = 0.0002; systolic blood pressure p < 0.0001). Duration has both a linear and squared effect as the effect levels out after approximately 40 years duration (Fig. 1).Table 3


Impact of changes in metabolic control on progression to photocoagulation for clinically significant macular oedema: a 20 year study of type 1 diabetes.

Sander B, Larsen M, Andersen EW, Lund-Andersen H - Diabetologia (2013)

(a–d) The effects of metabolic changes were entered into the models as continuous variables, as described in the Methods. The curves are calculated with a linear spline, i.e. a piecewise linear model, the HRs (loge scale, solid line) and the 95% CIs (broken line) are shown from a joint model similar to Table 3. (a–c) For both HbA1c and blood pressure, the minimal risk for photocoagulation was found for zero changes, with increasing risks both for decreases and increases in the variables. The curves also suggest a peak in the hazard ratio for HbA1c changes of 0.5 percentage points (corresponding to a change of 5.5 mmol/mol) (a) and 10 mmHg for systolic (b) and 5 mmHg (c) for diastolic blood pressure, with a plateau for larger changes. Due to the piecewise linear model, the exact position of the peaks and the distinct discontinuation points should be interpreted with caution. The HR for diabetes duration (d) reaches a plateau after approximately 40 years
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3824341&req=5

Fig1: (a–d) The effects of metabolic changes were entered into the models as continuous variables, as described in the Methods. The curves are calculated with a linear spline, i.e. a piecewise linear model, the HRs (loge scale, solid line) and the 95% CIs (broken line) are shown from a joint model similar to Table 3. (a–c) For both HbA1c and blood pressure, the minimal risk for photocoagulation was found for zero changes, with increasing risks both for decreases and increases in the variables. The curves also suggest a peak in the hazard ratio for HbA1c changes of 0.5 percentage points (corresponding to a change of 5.5 mmol/mol) (a) and 10 mmHg for systolic (b) and 5 mmHg (c) for diastolic blood pressure, with a plateau for larger changes. Due to the piecewise linear model, the exact position of the peaks and the distinct discontinuation points should be interpreted with caution. The HR for diabetes duration (d) reaches a plateau after approximately 40 years
Mentions: Using a discrete time proportional hazards model, current values for HbA1c, systolic blood pressure and diastolic blood pressure remained significant risk factors for progression to photocoagulation for CSME (HbA1cp < 0.0001, systolic blood pressure p = 0.0003, diastolic blood pressure p = 0.0038, duration p < 0.0001; Table 3, left column) when analysed separately and adjusted for current age, calendar period and duration (univariate model). With the same model, changes for both HbA1c, and systolic and diastolic blood pressure were all significant (p < 0.0001). In a multiple regression model containing all three covariates (joint model), only current HbA1c remained significant overall (p < 0.0001; Table 3, right column). For the effect of changes, both the change in HbA1c and systolic blood pressure were significant (HbA1c change p = 0.0002; systolic blood pressure p < 0.0001). Duration has both a linear and squared effect as the effect levels out after approximately 40 years duration (Fig. 1).Table 3

Bottom Line: We therefore examined the effect of changes in glycaemia and arterial blood pressure on the incidence of clinically significant macular oedema in a population of diabetic patients.A recent decrease of ≥ 0.5 percentage points or an increase in HbA1c of >0.5 percentage points per 6 months was associated with HRs of 3.04 and 1.28, respectively, compared with lesser changes in HbA1c.The effects of metabolic and haemodynamic stability on diabetic retinopathy should be examined in prospective studies.

View Article: PubMed Central - PubMed

Affiliation: Department of Ophthalmology, Glostrup Hospital, University of Copenhagen, Ndr. Ringvej 57, 2600, Glostrup, Denmark, bisan@regionh.dk.

ABSTRACT

Aims/hypothesis: Although increasing hyperglycaemia, arterial hypertension and longer duration of diabetes raise the risk of progression of diabetic retinopathy, short-term benefits in terms of improved metabolic control and lowered blood pressure have not been demonstrated. We therefore examined the effect of changes in glycaemia and arterial blood pressure on the incidence of clinically significant macular oedema in a population of diabetic patients.

Methods: We performed a retrospective review of all patients with type 1 diabetes who attended the retinopathy screening clinic at the Steno Diabetes Center from 1988 to 2008, using the endpoint referral to first photocoagulation treatment for clinically significant diabetic macular oedema. The analysis included 1,878 patients (median observation, 8 years). Changes were defined as the inter-visit change; in the case of an event the last event-free interval before referral, where the median screening interval was 6 months.

Results: Risk of progression to photocoagulation for macular oedema increased with duration of diabetes (p < 0.001), current HbA1c (p < 0.0001) and with the magnitude of changes in HbA1c (p = 0.0002) and systolic blood pressure (p < 0.0001) in a multiple regression model. A recent decrease of ≥ 0.5 percentage points or an increase in HbA1c of >0.5 percentage points per 6 months was associated with HRs of 3.04 and 1.28, respectively, compared with lesser changes in HbA1c.

Conclusions/interpretation: In this study, large recent changes in metabolic control and systolic blood pressure, irrespective of direction, were independent risk factors for progression to photocoagulation for diabetic macular oedema. The effects of metabolic and haemodynamic stability on diabetic retinopathy should be examined in prospective studies.

Show MeSH
Related in: MedlinePlus