AUTHOR’S QUERY SHEET

Author(s): Romero-Aroca Pedro et al.Article title: Article no: NOPE 498661

Dear Author,

Please check these proofs carefully. It is the responsibility of the corresponding author to check against the original manuscript and approve or amend these proofs. A second proof is not normally provided. Informa Healthcare cannot be held responsible for uncorrected errors, even if introduced during the composition process. The journal reserves the right to charge for excessive author alterations, or for changes requested after the proofing stage has concluded.

The following queries have arisen during the editing of your manuscript and are marked in the margins of the proofs. Unless advised otherwise, submit all corrections using the CATS online correction form. Once you have added all your corrections, please ensure you press the “Submit All Corrections” button.

AQ1. Define AAO - American Academy of Ophthalmology?

AQ2. Define - used later again.

AQ3. Thaere are several places that CT/HDL appears but is never defined. I believe it may be typo for TC/HDL which is defined. Please either correct or define.

AQ4. A declaration of interest statement reporting no conflict of interest has been inserted. Please confirm the statement is accurate.

AQ5. Please check and approve the running head or provide an alternative.

AQ6. Currently your figures will appear in black and white for the print edition. If you would like to pay for figures 1 and 2 to print in color, please indicate so that we may request a cost estimate.

Any figure submitted as a color original will appear in color in the journal’s online edition free of charge. Color reproduction in the print edition is available if the authors, or their funding body, bear the associated costs. The charge for the first color page is USD $1000, second and third pages are charged at USD $500 each. For four or more pages please contact us for a quote. All orders or inquiries must be made on or before the proof due date.

1. Article details

◻ I wish to pay for color in print. The manuscript (CATS) ID of my article is: _________________ (six digit number)

Article title: ________________________________________________________________________________________

Author name: ______________________________________________________________________________________

Please detail the figure/part numbers for color printing: _________________________________________________

Number of color pages: ____________ Total color costs: ________________(see table above)

2. Payment optionsAn invoice or credit card payment will be raised when the issue containing your paper is sent to press.

EITHER pay by credit card

OR invoice to: ___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________Signed: ________________________________________________________ Date: ______________________________Print name: ________________________________________________________________________________________

NOTESIf requested, and if possible, at proof stage we can group images from various pages onto one page to reduce color costs. Please give guidance in your proof corrections if this is the case. Color costs are by PAGE not per image, and size has no relevance.

RightsLink author reprints are manufactured separately from the printed journal and are printed in black and white by default. Author reprints ordered prior to your color in print order being processed will be printed in black and white. If you wish to order color reprints of your article, please wait until your color in print order is processed then access the RightsLink reprint ordering link where you will be able to select an option for color.

If you do not wish to pay for color printing, please ensure that all figures/images and legends are suitably amended to be read, and understood, in black and white. It is the author’s responsibility to provide satisfactory alternatives at proof stage. Please access the instructions for authors for

Number of color pages 1 2 3 4+

Total price $1000 $1500 $2000 Contact us for a quotation

Credit Card Please charge my AMEX/VISA/masterCard(delete as appropriate)

Card No / / / Exp.date /

CVV number (last 3 or 4 digits on reverse of card):

Name on credit card:

Address where registered

Signature: Date:

COLOR IN PRINT ORDER

OPHTHALMIC EPIDEMIOLOGY color policy

Scan and email the completed form to the journal’s production editor

further guidance on figure quality at: http://informahealthcare.com/ope

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556

57585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112

1

IntroductIon

Diabetes mellitus (DM) is defined as a group of metabolic diseases whose common feature is an elevated blood glucose level (hyperglycaemia) and is

Ophthalmic Epidemiology, 00(00), 000–000, 2010Copyright © 2010 Informa UK Ltd.ISSN: 0928-6586 print/ 1744-5086 onlineDOI: 10.3109/09286586.2010.498661

ORIGINAL ARTICLE

Prevalence and relationship between diabetic retinopathy and nephropathy, and its risk factors in the north-east of Spain, a

population-based study

Romero-Aroca Pedro1, Sagarra-Alamo Ramon2, Baget-Bernaldiz Marc1, Fernández-Ballart Juan3, and Méndez-Marin Isabel4

1Ophthalmic Service, Hospital Universitario Sant Joan, Institut de Investigacio Sanitaria Pere Virgili, Universidad Rovira I Virgili, Reus, Spain

2Primary Care Health Area Reus-Priorat, Reus, Spain3Department of Basic Sciences, University Rovira i Virgili, Tarragona, Spain

4Hospital Universitario Sant Joan, Institut de Investigacio Sanitaria Pere Virgili, Universidad Rovira I Virgili, Reus, Spain

AbStrAct

Purpose: To determine the prevalence of microangiopathy, and its risk factors in a population-based study of diabetes mellitus patients in the north-eastern area of Spain.Methods: A population-based transversal study of 8,187 type 2 (83.37% of the diagnosed patients) and 488 type 1 (85.76% of the diagnosed patients) underwent a detailed medical history that included: diagnoses of diabetic retinopathy, macular edema, microalbuminuria or overt nephropathy. A study of its risk factors, and the relationship between diabetic retinopathy and renal lesion was performed.Results: In type 1 diabetes patients we observed a prevalence of 36.47% of diabetic retinopathy and 5.73% with macular edema; in type 2 diabetes patients the prevalence of diabetic retinopathy was 26.11% and 6.44% with macular edema. Microalbuminuria prevalence was 25.61% in type 1 and 17.78% in type 2 patients, and overt nephropathy prevalence was 8.60% in type 1 and 6.74% in type 2 diabetic patients. The risk factors for diabetic retinopathy were: diabetes duration, high glycosy-lated level, and arterial hypertension, and insulin treatment in type 2. The Total-cholesterol/High Density-cholesterol (TC/HDL) ratio and triglycerides were significant for diabetic macular edema (DME). Microalbuminuria and overt nephropathy were well correlated to diabetes duration, arterial hypertension and glycosylated haemoglobin.Conclusions: Prevalence and risk factors for microangiopathy are similar to other studies, and the important finding is that the TC/HDL ratio was significant for DME. Microalbuminuria is a risk factor for diabetic retinopathy in type 1 diabetes mellitus patients but not for type 2. Overt nephropathy is well correlated with diabetic retinopathy.

KEYWordS: Diabetic retinopathy; Diabetic macular edema; Microalbuminuria; Overt nephropathy; Prevalence; Epidemiological study; TC/HDL ratio

Received 10 November 2009; Revised 23 March 2010; Accepted 14 April 2010

Correspondence: Pedro Romero, C/Ample 55 1° 43202 Reus ( Tarragona). E-mail: promero@grupsagessa.com

10 November 2009

23 March 2010

14 April 2010

© 2010 Informa UK Ltd

2010

Ophthalmic Epidemiology

0928-65861744-5086

10.3109/09286586.2010.498661

00

000000

00

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

2 Romero-Aroca Pedro et al.

Ophthalmic Epidemiology

a major health problem worldwide. In 2010 more than 200 million patients will have diabetes, and the global population is predicted to increase by 62% between 1995 and 2025.1

In 2000, the World Health Organization Expert Committee on Diabetes2 redefined diabetes using these criteria: > 7.0 mmol/l (126 mg/dl) or 2 hour post 75g glucose load plasma glucose > 11.1 mmol/l (200mg/dl). The disease is classified into several cat-egories, which were published in 20002: Type 1 DM (formerly known as insulin-dependent diabetes mel-litus or juvenile-onset diabetes mellitus) was defined as a disorder caused by autoimmune destruction of pancreatic β-cells, rendering the pancreas unable to synthesize and secrete insulin; Type 2 DM accounts for almost all of the remaining cases of diabetes as the other forms are rare. It is a condition that pre-dominantly affects middle-aged and older people but prevalence is increasing among children and young adults in countries with high levels of obesity.

The major microvascular complications, ie, retinopa-thy and nephropathy, are the most important causes of blindness and end-stage renal disease in Europe. The detection of retinopathy is easy using periodical fun-dus retinography, but diagnosis of the early stages of nephropathy need microalbuminuria to be determined in urinary analysis. After development of clinical grade proteinuria (> 80%) patients go on to develop a decreased glomerular filtration rate and, given enough time, end-stage renal disease.3

Several factors appear to influence susceptibil-ity to the microvascular complications of DM, but our knowledge of the role and importance of these genetic and environmental factors is still incomplete. The most powerful risk factor for microvascular complications is the duration of diabetes, but the frequency of both retinopathy and nephropathy is related to the level of plasma glucose at the time of examination, as various multicentric studies have demonstrated.4,5

The aim of the present study is to determine the prevalence of diabetic retinopathy, diabetic macular edema, microalbuminuria, overt nephropathy and its risk factors, in a population-based study of DM patients censused in our Health Care Area (HCA).

MAtErIAl And MEthodS

Setting

The Hospital St. Joan is the only public ophthal-mology centre in Reus (Spain), with a dependent population of 218,740 inhabitants, and all diabe-tes patients, referred by general practitioners and

endocrinologists, are examined once a year at that center. There are two nonmydriatic fundus cameras for diabetes screening.

design

The study was a population-based transversal study on patients previously diagnosed with diabetes mellitus type 1 or 2 by family physicians or endocrinologist and censused in its Health Care Centers. The study was carried out between 1 January 2008 and 31 December 2008. All patients were screened during this period by the two nonmydriatic fundus cameras. After screening, the patients with diabetic retinopathy were evaluated by retina specialists of the hospital.

Power of the study

Using the database, our epidemiologist estimated that in order to detect a 95% increase in risk at an accuracy interval of 3% we needed a sample of 90 patients in the type 1 DM group, and we needed 1,152 in the type 2 DM group.

Prevalence of diabetes in our area and sample of the study

Type 1 DM patients: since 1987 a register has been kept of all new cases in Catalonia (Spain) and there has been a steady increase of 11.4 new cases per 100,000 inhabitants (13.2 men and 9.6 women).6,7 At Hospital St Joan, 569 type 1 DM patients were registered in 2008, and 488 patients (85.76%) were recruited for the study.

Type 2 DM patients: the prevalence of type 2 DM (diagnosed and undiagnosed) determined according to National Health studies was 10.30% of patients > 30 years old, a total of 12,144 patients with DM in our area.7 According to the records of Primary Care Units, a total of 9,820 patients were registered as type 2 DM patients in our area (representing 80.86% of all dia-betes patients). A sample of 8,187 type 2 DM patients (83.37% of those registered) were recruited for the study.

Ethical adherence

The local scientific investigative ethical committee also gave us their agreement, according to the Helsinki Dec-laration. All patients of the second group signed an informed consent to participate in the study.

AQ5

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

Prevalence of Microangiopathy in Spain 3

© 2010 Informa UK Ltd.

Methods

Diabetic retinopathy was evaluated by non mydriatic retinal photographs through dilated pupils, two retino-graphs were taken of each eye at 45° fields centered firstly at the temporal to the macula and secondly from the nasal to the papilla.8 The total number of images taken of each patient, in order to obtain the best quality retinographies was 2.92 ± 0.67.

Diabetic retinopathy (DR) is diagnosed when at least four or more microaneurysms are present in the fundus photograph, with or without hard or soft exudates, in the absence of other known causes of the changes (ie, a branch retinal vein occlusion).

According to a modified version of the AAO classification,9 seven grades of severity were estab-lished: non-DR (NDR), mild nonproliferative DR (NPDR), moderate NPDR, severe NPDR, prolifera-tive diabetic retinopathy (PDR) with or without high risk characteristics, and patients treated previously with laser (LTP). The eye with the most severe dia-betic retinopathy was used in order to classify the patient. With our non-stereoscopic digital images, the presence of retinal thickening with no hard exu-dates makes it difficult to detect severity. After the detection of macular exudates or microaneurysms by nonmydriatic fundus camera, the patients with any suspicion of macular edema were examined by ste-reoscopic viewing of the macula with a slit lamp and a Goldman fundus contact lens. DR was considered present if we found any clinically significant macu-lar edema (CSME) according to ETDRS criteria,10 as follows:

retinal thickening at or within 500 • μ m of the centre of the maculahard exudates at or within 500 • μ m of the centre of the macula, if associated with thickening of adja-cent retina (but not hard exudates remaining after retinal thickening had disappeared)a zone or zones of retinal thickening, one disc area •or larger in size, any part of which is within 1 disc diameter of the centre of the macula

In all patients with diabetic macular edema, a fluo-rescein angiography centered on the macular region was obtained to determine any leakage. An optical coherence tomography (OCT) was also performed.Inclusion criteria:

Patients with type 1 DM diagnosed and censused •at the Hospital, and its Health Care Centers.Patients with type 2 DM censused from the list of •diabetic patients diagnosed and controlled by the family physicians in the Health Care Centers.

Exclusion criteria

Patients with other specific types (ie, diseases of •the exocrine pancreas, endocrinopathies, genetic defects in ß-cell function, genetic defects in insulin action).Patients included in the group IV. Gestational • diabetes mellitus (GDM)

definition of variables

Visual acuity in each eye was measured on the Snellen chart and recorded as a decimal value, with the best refraction for distance, and the data were applied to the logMAR (logarithm of the minimum angle of resolu-tion) ETDRS chart. All data handling on visual acuity is expressed in logMAR format. The legal blind subject is defined as corrected visual acuity of less than or equal to 0.1 in the better eye; reduced visual acuity is defined as ≤ 0.4 and >0.1 in the better eye.

The epidemiological risk factors included in the study were:

gender•age in type 1 DM patients (≤ 35 years old versus > •35 years old), in type 2 DM (≤ 65 years old versus >65 years old)duration of DM, classified in the statistical study into •two groups: ≤15 years and >15 years duration.arterial hypertension, which is indicated by a •systolic/diastolic measurement according to the report of the sixth joint National Committee on the Prevention, Detection, Evaluation and Treatment of High Blood Pressure11: systolic and diastolic blood pressures were the average of the 2 measurements, hypertension was defined as (i) a mean systolic blood pressure of 160 mmHg and/or a mean diastolic blood pressure 95 mmHg or a history of antihypertensive medication at the time of examination in individuals 25 years of age or (ii) a mean systolic blood pressure of 140 mmHg and/or a mean diastolic blood pres-sure of 90 mmHg, and/or a history of antihyper-tensive medication at the time of examination in younger persons.levels of glycosylated haemoglobin (HbA• 1c) defined according to the American Diabetes Association12 recommendation, was measured every 3 months, and a mean of all values was applied to the study. The control of glycemia was considered according to the European Diabetes Policy Group 1999, creating two groups of patients, over or under 7.0%.13

presence of microalbuminuria, defined as increased •albumin excretion (30–300 mg of albumin/24 h or

AQ1

AQ2

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

4 Romero-Aroca Pedro et al.

Ophthalmic Epidemiology

20–200 μg/min of creatinine) in two out of three tests repeated at intervals of 3–6 months as well as exclusion of conditions that invalidate the test.3

After the diagnosis of microalbuminuria was docu-mented, there was repeated testing over a period of 3–4 months, thus:

presence of overt diabetic nephropathy, defined by •clinical albuminuria or overt nephropathy by the American Diabetes Association,14 corresponding to protein excretion >300 mg/24 h (>200 μg/min or >300 μg/mg of albumin / creatinine ratio). At this stage we included the last three stages of diabetic nephropathy (early overt diabetic nephropathy, advanced diabetic nephropathy and end-stage renal disease) as defined by Mogensen.15

levels of the fractions of the cholesterol (HDL-•cholesterol and LDL-cholesterol) and triglyc-erides. In the statistical analysis, we classified patients into normal or higher values: HDL- cholesterol normal value ≥1.04 mmol/L in men and ≥1.16 mmol/L in women. LDL-cholesterol normal value < 4.16 mmol/L. Triglycerides normal value < 2.26 mmol/L, according to the Adult Treat-ment Panel III guidelines recently issued by the National Cholesterol Education Program.16 Total- cholesterol/HDL-cholesterol (TC/HDL) ratio, normal value < 4.1 in women and <4.0 in men, equivalent to quintile 2 of Rifai17 defined levels, with a cardiovascular relative risk < 1.2.

Statistical methods

All statistical analysis was carried out using the SPSS software package version 17.0 (Statistical Package for the Social Sciences; SPSS Inc., Chicago, IL). Results were expressed as mean ± standard error, a P-value of <0.05 was considered to indicate statistical significance. Differences between those included in analysis were examined using the two sample Student’s T-tests or Kruskal-Wallis, for continuous or quantitative vari-ables, such as visual acuity or current age. For the qualitative or categorical variables such as gender, systolic hypertension, diastolic hypertension, diabetes duration, levels of HbA1c, LDL-Cholesterol, HDL-cho-lesterol, CT/HDL ratio and triglycerides, we used the Chi-square test in the univariate phase of study, with determination of the Odds Ratio for each variable.

In the multivariate phase of analysis, the relation-ship of diabetic retinopathy and macular edema (as independent variables) to the risk factors, was examined by multiple logistical regression analysis. Variables included in the multivariate analysis were

selected in stepwise fashion from the following list: age, gender, HbA1c, arterial hypertension, insulin treatment (only for type 2 DM), LDL-cholesterol, HDL-cholesterol, triglycerides, CT/HDL ratio. When we introduced CT/HDL ratio we did not introduce LDL-cholesterol, nor HDL-cholesterol. The inclusion of microalbuminuria excluded overt nephropathy in the regression mode, and if we included overt nephropathy we excluded microalbuminuria in the multivariate analysis.

In the multivariate phase, analyzing by multiple logistical regression the relationship of microalbu-minuria and overt nephropathy to the risk factors, we included the same variables analyzed in the multivari-ate analysis selected in stepwise fashion, and the rela-tionship with diabetic retinopathy was included in two forms of risk factor: the first as diabetic retinopathy (present / absent), the second as severity of diabetic retinopathy (proliferative / non proliferative), The introduction of the first variable excluded the second, and the other way around. The results were presented as significance (p), relative risk and 95% CI for each variable.

The variables analyzed for each independent variable included: age (in type 1 patients cut off at ≤ 20 years old, in type 2 cut off at ≤ 65 years ), gender, arterial hypertension, systolic blood tension, diastolic blood tension, diabetes duration (cut off at 15 years), levels of HbA1c (Normal value ≤ 7%), LDL-Cholesterol (≤ 4.16 mmol/L ), HDL-cholesterol (normal value ≥1.04 mmol/L in men and ≥1.16 mmol/L in women), Total-cholesterol/HDL-cholesterol (CT/HDL) ratio (normal value < 4.1 in women and <4.1 in men) and triglycerides (normal value ≤ 2.26 mmol/L).

rESultS

demographic variables of the patients

The differences with respect to age and gender were not significant (Table 1). Diabetes duration was higher in type 1 DM (20.42 ± 7.57 years) than type 2 DM (12.42 ± 6.30 years). As could be expected, the preva-lence of arterial hypertension is higher in patients with type 2 DM than in type 1 patients (29.1% and 50.99%, respectively). There were no significant differences in mean HbA1c between the two diabetes types (type 1: 7.32SD1.42, and type 2: 7.34SD1.23).

Visual acuity

The number of patients considered blind (Visual Acu-ity [VA] ≤ 0.1 in the Snellen chart) was inferior in type 2 DM (4.90%) to type 1 DM (8.00%).

AQ3

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

Prevalence of Microangiopathy in Spain 5

© 2010 Informa UK Ltd.

Prevalence of diabetic retinopathy and macular edema

The prevalence of diabetic retinopathy in patients with type 1 DM was 36.47% (178 patients) and in patients with type 2 DM the prevalence was 26.11% (2138 patients). Table 2 shows the prevalence of dia-betic retinopathy. The proliferative form of retinopa-thy is present in 1.02% of type 1 DM and in 0.56% of type 2 DM patients. The number of the patients in the LTP (treated previously with laser) group with a ratio of 7.17% of type 1 DM and 6.92% of type 2 DM.

According age-mean in type 1 DM the prevalence of DR under or equal 35 years old was 35.89 and over 35 years old was 36.92%; in type 2 DM the prevalence of DR ≤65 years old was 26.10% and over 65 years old 26.12%.

Stratification by age and type of DM demonstrated differences between both groups, thus for Type 1 DM prevalence of DR by age-stratification were: 20.53% for the group < 25 years old, 28.15% in the group between 25 to 35 years old, 37.99% in the group 35 to 45 years old and 39.77% in the group >45 years old. In Type 2 DM the prevalence of DR was similar in all age groups:

TABLE 1 Demographic data. Prevalence of visual acuity, diabetic retinopathy, macular edema and diabetic renal diseaseType of diabetes mellitus Type 1 diabetes (488 patients) Type 2 diabetes (8187 patients)Men / women (%) 272/216 (55.74% / 44.26%) 4520 / 3667 (55.21% / 44.79%)Age (mean SD) 34.87 SD 10.46 years old

(7–59 years)64.60 SD 10.78 (34–87 years)

Diabetes duration (mean SD) 20.42 SD 7.57 years (1–35 years)

12.42 SD 6.30 years (1–37 years)

Arterial hypertension 146 (29.91%) 4175 (50.99%)Systolic mean tension 124.42 SD 17.5 mm Hg 142.87 SD 18.17 mm HgDiastolic mean tension 78.14 SD 10.88 mm Hg 81.37 SD 11.32 mm HgDiabetes treatment

Diet•Oral•Insulin•

488 (100%)

1391 (17.00%) 4175 (50.99%) 2621 (32.01%)

Glycosilated haemoglobin (HbA1c ) 7.32 % SD 1.42% 7.34% SD 1.23%Visual acuity

≤ 0.1•0.2–0.4•> 0.4•

39 (8.00%) 54 (11.06%) 395 (80.94%)

401 (4.90%)

1138 (13.90%) 6648 (81.20%)

SD = standard deviation.* Percentage respect to the total of 488 type 1 and 8187 type 2 diabetes patients.** Patients treated previously by focal laser for macular oedema.

TABLE 2 Prevalence of diabetic retinopathy, diabetic macular oedema, microalbuminuria and overt nephropathyType of diabetes mellitus Type 1 diabetes(488 patients) Type 2 diabetes(8187 patients)Prevalence of diabetic retinopathy 178 patients (36.47%) 2138 patients (26.11%)Diabetic retinopathy level

Nonproliferative mild• 87 (17.82%) 985 (12.03%)Nonproliferative moderate• 37 (7.59%) 417 (5.09%)Nonproliferative severe• 14 (2.87%) 123 (1.50%)Proliferative• 5 (1.02%) 46 (0.56%)Patients treated previously with laser• 35 (7.17%) 567 (6.92%)

Prevalence of diabetic macular oedemaPatients treated previously with focal laser*

28 patients (5.73%) 26 patients (5.32%)

528 patients (6.44%) 477 patients (5.86%)

Prevalence of microalbuminuria 125 patients (25.61%) 1456 patients (17.78%)Prevalence of overt nephropathy 42 patients (8.60%) 552 patients (6.74%)Without microalbuminuria or retinopathy 282 (57.78%) 5043 (61.59%)Patients with microalbuminuria only 28 (4.09%) 1006 (12.29%)Patients with retinopathy only 132 (27.05%) 1526 (18.64%)Patients with retinopathy and microalbuminuria 46 (9.42%) 612 (7.47%)Without overt nephropathy or retinopathy 306 (62.70%) 5785 (70.66%)Patients with overt nephropathy only 4 (0.81%) 264 ( 3.22%)Patients with retinopathy only 134 (27.46%) 1633 (19.94%)Patients with retinopathy and overt nephropathy 44 (9.01%) 505 (6.16%)

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

6 Romero-Aroca Pedro et al.

Ophthalmic Epidemiology

in the group of age < 45 years old the prevalence was 24.33%, in the group 45 to 55 years old the prevalence was 29.34%, in the group 55 to 65 years old was 25. 37% and in the group > 65 years old the prevalence of DR was 26.01%.

Figure 1 shows the changes in prevalence of diabetic retinopathy according to the duration of diabetes. In type 1 DM prevalence increases from 0% in the group <5 years duration to 70.19% >15 years diabetes dura-tion. In type 2 DM the prevalence of diabetic retin-opathy increases from 9.19% in the group < 5 years to 65.80% in the group >15 years duration. This shows that after 15 years the prevalence of diabetic retinopathy is similar in both DM types.

The prevalence of diabetic macular edema for patients with type 1 DM was 5.73% (28 patients). For type 2 patients prevalence was 6.44% (528 patients). Age and gender adjusted data did not alter the prevalence of diabetic macular edema. Figure 1 shows an increase in diabetic macular edema according to the duration of diabetes in both DM types, with a low prevalence in the <5 years group (0% in type 1 DM and 1% in type 2 DM) and a higher prevalence in the group >15 years, with a similar prevalence for both DM types (14.98% in type1 and 15.92% in type 2). The mean of HbA1c in patients with diabetic macular edema was 8.47SD1.75 (type 1 DM) and 9.11SD1.47 (type 2 DM). In patients without macular edema, the mean of HbA1c was 7.38SD1.37 (type 1 DM) and 7.98SD1.67 (type 2 DM). The mean of triglycerides in patients with diabetic macular edema was 1.579 ± 0.375 mmol/l and in patients without macular edema the mean of triglycerides was 0.947 ± 0.293 mmol/l.

Prevalence of nephropathy

Prevalence of microalbuminuriaThere were many differences between the two types of DM. For type 1, prevalence was 25.61% (125 patients), slightly higher than microalbuminuria in type 2 with a prevalence of 17.78% (1,456 patients). Also, the preva-lence of overt nephropathy is slightly higher in type 1 with 8.60% (42 patients) against 6.74% (552 patients) in type 2. Age and gender adjusted data did not alter the prevalence of microalbuminuria or overt nephropathy.

In the relationship between microalbuminuria and duration of diabetes, type 1 patients showed a peak of 12.99% in prevalence at 11–15 years, and type 2 patients showed a peak in prevalence at 11–15 years of 7.60 (Figure 2). After a decrease at 16-20 years, an increase in both DM types was observed after 20 years, more importantly in type 1 patients (7.88% and 3.46% respectively).

The number of patients with overt diabetic neph-ropathy also decreased. There was a peak in prevalence of diabetic nephropathy between 11–15 years duration of diabetes in both DM types. In type 1 DM there was a peak of 4.30%, and 2.42% in type 2 DM patients. As in the prevalence of microalbuminuria, an increase in overt nephropathy was observed in the group after 20 years (Figure 2).

Statistical study of diabetic retinopathy

The risk factors for both DM types were: >15 years duration of DM, high levels of arterial hypertension (diastolic blood pressure was significant, but not

0%

10%

20%

30%

40%

50%

60%

70%

80%

Type 1 DR Type 2 DR

Type 1 DME Type 2 DME

Type 1 DR 0% 3,52% 28,09% 70,19%

Type 2 DR 9,19% 13,59% 48,74% 65,80%

Type 1 DME 0% 2,00% 13,22% 14,98%

Type 2 DME 1% 6% 8,98% 15,92%

0--5 6--10 11--15 >15

DR = diabetic retinopathy DME = diabetic macular edema

FIGURE 1 Prevalence of diabetic retinopathy and macular edema in type 1 and type 2 diabetes patients related to the duration of diabetes mellitus.

0%

2%

4%

6%

8%

10%

12%

14%

Type 1 MA Type 2 MAType 1 ON Type 2 ON

Type 1 MA 1% 3,21% 12,99% 3,22% 7,88%

Type 2 MA 0,78% 2,76% 7,60% 2,42% 3,46%

Type 1 ON 0% 0,00% 4,30% 1,07% 3,22%

Type 2 ON 0% 1% 2,42% 0,86% 1,38%

0--5 6--10 11--15 16--20 >20

MA = microalbuminuria ON = overt nephropathy

FIGURE 2 Prevalence of microalbuminuria and diabetic nephropathy in type 1 and type 2 diabetes patients related to the duration of diabetes mellitus

AQ6 AQ6

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

Prevalence of Microangiopathy in Spain 7

© 2010 Informa UK Ltd.

systolic), and overt nephropathy. Microalbuminuria was significant for type 1 DM but not for type 2. Lastly, insulin treatment was significant for type 2 DM patients. The stratification according to age showed differences of DR prevalence in the group of patients with type 1DM, but included in the statistical analysis diabetes duration, it was shown that the time of evolu-tion of DM, and not age was the most important risk factor in the development of DR, the reason was that the patients with the less age has the group with lower DM duration, and the groups with older age have the higher DM duration (Table 3).

Statistical study of diabetic macular edema

The risk factors for both DM types were: >15 years duration of DM, high levels of HbA1c, presence of arterial hypertension (diastolic blood pressure was significant, but not systolic), and overt nephropathy. The study of lipids demonstrates that LDL-cholesterol, TC/LDL ratio, and triglycerides were significant for both DM types. Lastly, microalbuminuria was signifi-cant only for type 1 DM. Age and gender adjusted data were not significant (Table 4).

Statistical study of diabetic microalbuminuria

The risk factors for both DM types were: >15 years duration of DM, high levels of HbA1c, presence of arte-rial hypertension (diastolic blood pressure was signifi-cant, but not systolic), and the severity of proliferative diabetic retinopathy. Insulin treatment was significant only for type 2 DM patients. Age and gender adjusted data were not significant (Table 5).

Statistical study of diabetic overt nephropathy

The risk factors for both DM types were: >15 years duration of DM, high levels of HbA1c, presence of arterial hypertension (diastolic blood pressure was significant, but not systolic), high triglycerides levels, the presence and severity of proliferative diabetic retinopathy. Insulin treatment was significant only for type 2 DM. Age and gender adjusted data were not significant (Table 6).

relationship between microalbuminuria and diabetic retinopathy

Table 2 shows that the relationship between microalbu-minuria and diabetic retinopathy is different between

types of diabetes. In type 2 DM, there were more patients only with microalbuminuria in type 1. Despite the risk factors for microalbuminuria being the same for both diabetes types, Table 5 shows that the sever-ity of diabetic retinopathy, rather than diabetic retin-opathy itself, is a risk factor for microalbuminuria, but at a low level of significance P = 0.030 OR 3.136, 95% IC0.098–4.829 for type 1 DM and P = 0.030 OR 2.455, 95% IC 0.762–6.759 for type 2 DM. If we consider the presence of microalbuminuria as a risk factor for dia-betic retinopathy and macular edema, we observe that the presence of microalbuminuria is a risk factor only in type 1 DM patients (Tables 3 and 4).

relationship between overt nephropathy and diabetic retinopathy

Table 2 shows the relationship between overt neph-ropathy and diabetic retinopathy, and the distribution of the four groups of patients was similar in both dia-betes types. Table 6 shows the risk factors for overt nephropathy, which were the same for both types: >15 years duration of DM, high levels of HbA1c, presence of arterial hypertension (diastolic blood pressure was significant, but not systolic), high triglyceride level, the presence and severity of proliferative diabetic retinopathy. Insulin treatment was significant only for type 2 DM.

If we consider the presence of overt nephropathy as a risk factor for diabetic retinopathy and macular edema, we observe that it is a risk factor for both types, despite the risk being higher in type 1 DM than in type 2 DM patients (Tables 3 and 4).

dIScuSSIon

In Spain there is a Public National Health System where family physicians (FP) act as gatekeepers. Diabetes patients have their FP to refer them for diabetes care, though the common specialists for referral for retin-opathy screening are public based ophthalmologists. There is a register of diabetes patients in Health Care Centers (HCC). In some regions of Spain (Andalusia, Catalonia and Valencia) a program for diabetic retin-opathy has been initiated that uses non-mydriatic fundus camera. This particular system of diabetes care has allowed us to evaluate a large number of patients in the present study, and to determine the true preva-lence of diabetic retinopathy and renal lesion more accurately than has been possible in other studies. In fact, in our previously published studies, determina-tion of prevalence was based on a random sample of patients, but we think the present study, in which we

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

8 Romero-Aroca Pedro et al.

Ophthalmic Epidemiology

TAB

LE

3 S

tati

stic

al s

tud

y of

ris

k fa

ctor

s fo

r d

iabe

tic

reti

nopa

thy

Ris

k fa

ctor

Type

1 d

iabe

tes

Type

2 d

iabe

tes

Chi

squ

ared

Log

isti

cal r

egre

ssio

nC

hi s

quar

edL

ogis

tica

l reg

ress

ion

OR

95

%C

IO

R

95%

CI

pO

R

95%

CI

OR

95

%C

I

pG

end

er (m

ale)

0.98

8, 0

.393

–1.2

420.

174,

0.3

14–2

.105

, 0.5

440.

991,

0.3

84–1

.870

0.28

6, 0

.242

–2.8

82, 0

.238

Age

*0.

997,

0.7

76–1

.231

0.99

2, 0

.607

–2.1

01, 0

.404

1.05

2, 0

.921

–1.2

010.

883,

0.6

75–2

.705

, 0.4

42A

rter

ial h

yper

tens

ion(

pres

ent)

2.78

3, 1

.906

–4.0

633.

081,

1.9

94–8

.946

, 0.0

013.

646,

1.0

20–5

.692

2.99

2, 1

.201

–7.3

22, 0

.021

Syst

olic

blo

od p

ress

urep

er 1

0 m

m H

g1.

033,

0.0

79–2

.560

Not

incl

uded

in th

e fin

al m

ulti

vari

ate

mod

el1.

159,

1.0

20–1

.317

Not

incl

uded

in th

e fi

nal m

ulti

vari

ate

mod

elD

iast

olic

blo

od p

ress

urep

er 1

0 m

m H

g3.

706,

1.9

79–1

7.56

0N

ot in

clud

ed in

the

final

mul

tiva

riat

e m

odel

3.37

9, 1

.003

–8.3

17N

ot in

clud

ed in

the

fina

l mul

tiva

riat

e m

odel

Dia

bete

s d

urat

ion>

15 y

ears

7.63

0, 4

.490

–12.

751

6.62

5, 4

.828

–10.

386,

<0.

001

7.77

6, 3

.002

–11.

839

6.77

2, 1

.081

–29.

546,

<0.

001

Lev

el o

f HbA

1cpe

r 1%

4.78

7, 2

.446

–9.3

716.

324,

2.5

74–9

.332

, <0.

001

6.37

0, 3

.533

–11.

487

6.29

9, 1

.082

–27.

802,

<0.

001

HD

L-c

hole

ster

ol1.

229,

0.8

59–1

.759

0.96

8, 0

.537

–1.8

87, 0

.466

1.02

4, 0

.284

–3.3

700.

733,

0.1

02–5

.529

, 0.1

89L

DL

-cho

lest

erol

1.89

0, 0

.788

–3.5

220.

988,

0.8

91–5

.954

, 0.0

921.

177,

0.8

72–4

.897

1.02

7, 0

.980

–5.5

99, 0

.071

TC

/H

DL

rati

o2.

203,

1.0

88–5

.177

1.73

4, 0

.581

–3.6

85, 0

,060

1.1.

923,

0.6

63–3

.976

1.38

7, 1

.233

–3.3

69, 0

.062

Trig

lyce

rid

es1.

110,

0.9

77–3

.877

1.48

3, 0

.778

–4.6

67, 0

.112

1.12

7, 0

.866

–1.4

670.

997,

0.8

70–5

.002

, 0.1

02In

sulin

trea

tmen

tN

AN

A5.

779,

2.4

02–1

0.32

76.

168,

1.7

85–9

.659

, <0.

001

Mic

roal

bum

inur

ia(p

rese

nt)

2.21

8, 1

.543

–4.7

782.

467,

0.6

83–5

.718

, 0.0

121.

981,

1.0

27–2

.877

1.11

,7 0

.519

–2.7

78, 0

.060

Ove

rt n

ephr

opat

hy(p

rese

nt)

2.71

0, 1

.328

–4.4

97,

3.65

5, 0

.289

–5.7

76, 0

.011

2.63

7, 0

.501

–4.8

992.

869,

0.7

22–7

.401

, 0.0

22H

DL

= H

igh

den

sity

cho

lest

erol

LD

L =

Low

den

sity

cho

lest

erol

TC

/H

DL

= T

otal

cho

lest

erol

/hi

gh d

ensi

ty.

Age

*= g

roup

s of

age

are

des

crib

ed in

Met

hod

s. O

R=

Od

ds

rati

o, N

A =

Not

ava

ilabl

e, b

ecau

se a

ll pa

tien

ts w

ith

type

1 d

iabe

tes

mel

litus

are

insu

lin tr

eate

d. N

ot in

clud

ed in

the

final

mul

tiva

riat

e m

odel

= in

dic

ate

that

var

iabl

e w

as n

ot s

igni

fican

t and

thus

not

incl

uded

in th

e fin

al m

ulti

vari

ate

mod

el.

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

Prevalence of Microangiopathy in Spain 9

© 2010 Informa UK Ltd.

TAB

LE

4 S

tati

stic

al s

tud

y of

ris

k fa

ctor

s fo

r d

iabe

tic

mac

ular

ed

ema

Ris

k fa

ctor

Type

1 d

iabe

tes

Type

2 d

iabe

tes

Chi

squ

ared

Log

isti

cal r

egre

ssio

nC

hi s

quar

edL

ogis

tica

l reg

ress

ion

OR

95

%C

IO

R

95%

CI

pO

R

95%

CI

OR

95

%C

I p

Gen

der

(mal

e)0.

998,

0.6

62–1

.772

1.39

2 0.

523–

2.40

3, 0

.107

0.84

3, 0

.286

–1.0

240.

839,

0.3

43–1

.966

, 0.5

31A

ge*

0.87

8, 0

.576

–1.6

411.

002

0.91

9–2.

196,

0.3

270.

778,

0.5

51–1

.434

1.09

6, 0

.778

–1.3

34, 0

.563

Art

eria

l hyp

erte

nsio

n(pr

esen

t)5.

571,

1.6

33–1

9.00

93.

992,

1.6

77–6

.113

, 0.0

303.

778,

2.7

81–5

.631

3.66

5, 2

.334

–6.9

76, 0

.020

Syst

olic

blo

od p

ress

urep

er 1

0 m

m H

g1.

775,

1.8

42–6

.772

Not

incl

uded

in th

e fin

al m

ulti

vari

ate

m

odel

1.00

2, 0

.840

–4.0

22N

ot in

clud

ed in

the

fina

l mul

tiva

riat

e m

odel

Dia

stol

ic b

lood

pre

ssur

eper

10

mm

Hg

8.03

3, 3

.899

–23.

771

Not

incl

uded

in th

e fin

al m

ulti

vari

ate

m

odel

6.07

5, 2

.077

–8.7

72N

ot in

clud

ed in

the

fina

l mul

tiva

riat

e m

odel

Dia

bete

s d

urat

ion>

15 y

ears

10.9

33, 4

.365

–27.

383

6.99

8, 5

.442

–21.

236,

<0.

001

9.88

7,4.

542–

20.0

036.

107,

2.6

02–9

.047

<0.

001

Lev

el o

f HbA

1cpe

r 1%

8.55

4, 5

.992

–17.

292

6.02

7, 4

.921

–26.

773,

<0.

001

8.77

9, 3

.078

–21.

337

5.65

5, 3

.441

–8.8

02, 0

.001

HD

L- c

hole

ster

ol0.

908,

0.3

20–6

.888

1.00

2, 0

.402

–2.7

77, 0

.245

0.77

8, 0

.221

–1.9

770.

701,

0.5

01–1

.702

, 0.5

01L

DL

- cho

lest

erol

3.44

7, 1

.214

–17.

002

1.60

7, 0

.990

–4.1

78, 0

.072

2.78

3, 1

.427

–7.4

951.

348,

1.5

55–5

.877

, 0.0

81T

C/

HD

L ra

tio

8.29

0, 4

.004

–19.

555

4.26

6, 1

.102

–10.

335,

0.0

206.

539,

3.8

87–1

7.24

03.

983,

0.8

57–7

.599

, 0.0

20Tr

igly

ceri

des

4.48

7, 1

.938

–10.

390

1.93

2, 0

.910

–6.1

32, 0

.041

3.55

7, 1

.526

–8.4

951.

882,

0.6

03–4

.469

, 0.0

50In

sulin

trea

tmen

tN

AN

A8.

332,

2.6

02–1

7.32

76.

174,

3.0

12–7

.997

, 0.0

01M

icro

albu

min

uria

(pre

sent

)5.

069,

2.2

56–1

1.38

82.

592,

0.8

30–7

.897

, 0.0

303.

230,

1.0

43–1

0.43

01.

890,

0.4

37–8

.176

, 0.3

94O

vert

nep

hrop

athy

(pre

sent

)6.

110,

1.7

88–1

1.10

16.

631,

2.0

19–1

9.77

3, 0

.010

6.33

7, 2

.001

–10.

899

3.93

5, 1

.174

–6.4

01, 0

.010

HD

L =

Hig

h d

ensi

ty c

hole

ster

ol L

DL

= L

ow d

ensi

ty c

hole

ster

ol T

C/

HD

L =

Tot

al c

hole

ster

ol/

high

den

sity

.A

ge*=

gro

ups

of a

ge a

re d

escr

ibed

in M

etho

ds.

CI =

Con

fiden

ce in

terv

al O

R=

Od

ds

rati

o, N

A =

Not

ava

ilabl

e, b

ecau

se a

ll pa

tien

ts w

ith

type

1 d

iabe

tes

mel

litus

are

insu

lin

trea

ted

. Not

incl

uded

in th

e fin

al m

ulti

vari

ate

mod

el =

ind

icat

e th

at v

aria

ble

was

not

sig

nific

ant a

nd th

us n

ot in

clud

ed in

the

fina

l mul

tiva

riat

e m

odel

.

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

10 Romero-Aroca Pedro et al.

Ophthalmic Epidemiology

TAB

LE

5 S

tati

stic

al s

tud

y of

ris

k fa

ctor

s fo

r m

icro

albu

min

uria

Ris

k fa

ctor

Type

1 d

iabe

tes

Type

2 d

iabe

tes

Chi

squ

ared

Log

isti

cal r

egre

ssio

nC

hi s

quar

edL

ogis

tica

l reg

ress

ion

OR

95

% C

IO

R

95%

CI

pO

R

95%

CI

OR

95

%C

I

p

Gen

der

(mal

e)1.

090,

0.6

10–1

.884

0.27

4, 0

.393

–1.2

42, 0

.323

0.98

6, 0

.523

–1.8

590.

933,

0.4

65–1

.874

, 0.8

46A

ge*

0.99

7, 0

.338

–1.7

730.

677,

0.7

76–1

.231

, 0.5

441.

728,

0.8

65–3

.453

0.87

7, 0

.354

–1.2

03, 0

.746

Art

eria

l hyp

erte

nsio

npre

sent

5.11

1, 2

.738

–9.5

413.

655,

1.4

79–7

.560

, 0.0

014.

972,

0.9

26–8

.344

3.29

9, 0

.736

–5.8

28, 0

.001

Syst

olic

blo

od p

ress

ureP

er 1

0 m

m H

g1.

834,

1.0

02–4

.339

Not

incl

uded

in th

e fin

al m

ulti

vari

ate

mod

el1.

022,

0.3

42–1

.899

Not

incl

uded

in th

e fi

nal m

ulti

vari

ate

mod

elD

iast

olic

blo

od p

ress

ureP

er 1

0 m

m H

g6.

131,

1.3

63–9

.789

Not

incl

uded

in th

e fin

al m

ulti

vari

ate

mod

el5.

881,

1.2

06–9

.863

Not

incl

uded

in th

e fi

nal m

ulti

vari

ate

mod

elD

iabe

tes

dur

atio

n2.

735,

0.7

42–4

.401

2382

8, 1

.438

–6.3

98, 0

.020

2.08

8, 0

.867

–3.4

412.

544,

0.6

57–5

.105

, 0.0

20L

evel

of H

bA1c

per

1%8.

318,

3.2

27–1

2.36

36.

394,

2.53

3–9.

052,

<0.

001

3.58

7, 1

.355

–4.9

415.

233,

2.1

03–7

.445

, 0.0

01H

DL

-cho

lest

erol

1.13

3, 0

.472

–2.7

230.

436,

0.7

46–

3.91

0, 0

.179

1.00

8, 0

.518

–1.9

571.

046,

0.5

86–2

.049

, 0.5

67L

DL

-cho

lest

erol

2.19

5, 0

.998

–4.1

641.

022,

0.7

88–3

.522

, 0.3

502.

323,

0.9

04–4

.969

1.18

8, 0

.736

–3.2

77, 0

.168

TC

/H

DL

rati

o2.

884,

1.6

45–6

.992

1.10

2, 1

.233

–3.0

02, 0

.207

2.67

7, 1

.102

–5.1

041.

185,

0.2

63–2

.662

, 0.1

03Tr

igly

ceri

des

1.90

1, 0

.621

–3.0

021.

102,

0.1

77–3

.877

, 0.2

601.

466,

0.5

59–3

.843

1.09

5, 0

.382

–3.2

44, 0

.866

Insu

lin tr

eatm

ent

NA

NA

2.52

4, 1

.202

–5.3

005.

394,

2.5

33–9

.052

, 0.0

01D

iabe

tic

reti

nopa

thy

1.50

8, 1

.028

–2.8

510.

655,

0.3

77–2

.384

, 0.2

671.

508,

1.0

28–2

.851

0.95

5, 0

.264

–2.7

88, 0

.436

Seve

rity

of D

iabe

tic

reti

nopa

thy

2.67

7, 1

.227

–5.0

243.

136,

0.0

98–4

.829

, 0.0

302.

377,

1.2

27–4

.024

2.45

5, 0

.762

–6.7

59, 0

.030

HD

L =

Hig

h d

ensi

ty c

hole

ster

ol L

DL

= L

ow d

ensi

ty c

hole

ster

ol T

C/

HD

L - T

otal

cho

lest

erol

/hi

gh d

ensi

ty.

Age

*= g

roup

s of

age

are

des

crib

ed in

Met

hod

s. C

I = co

nfid

ence

inte

rval

OR

= O

dd

s ra

tio,

NA

=N

ot a

vaila

ble,

bec

ause

all

pati

ents

wit

h ty

pe 1

dia

bete

s m

ellit

us a

re in

sulin

tr

eate

d. N

ot in

clud

ed in

the

final

mul

tiva

riat

e m

odel

= in

dic

ate

that

var

iabl

e w

as n

ot s

igni

fican

t and

thus

not

incl

uded

in th

e fi

nal m

ulti

vari

ate

mod

el.

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

Prevalence of Microangiopathy in Spain 11

© 2010 Informa UK Ltd.

TAB

LE

6 S

tati

stic

al s

tud

y of

ris

k fa

ctor

s fo

r ov

ert n

ephr

opat

hy

Ris

k fa

ctor

Type

1 d

iabe

tes

Type

2 d

iabe

tes

Chi

squ

ared

Log

isti

cal r

egre

ssio

nC

hi s

quar

edL

ogis

tica

l reg

ress

ion

OR

95

% C

IO

R

95%

CI

pO

R

95%

CI

OR

95

%C

I

p

Gen

der

(mal

e)0.

659,

0.2

76–1

.573

0.91

2,0.

329–

2.52

4, 0

.859

0.63

0, 0

.138

–2.8

680.

462,

0.0

84–2

.542

, 0.3

75A

ge*

1.03

5, 0

.431

–2.0

570.

687,

0.14

6–2.

229,

0.6

340.

980,

0.6

40–1

.722

0.78

6, 0

.366

–1.9

04, 0

.476

Art

eria

l hyp

erte

nsio

npre

sent

7.28

6, 2

.865

–18.

530

4.17

0,1.

221–

8.01

0, 0

.001

6.11

2, 1

.989

–10.

344

5.74

4, 2

.021

–6.4

63, 0

.001

Syst

olic

blo

od p

ress

urep

er 1

0 m

m H

g2.

103,

1.8

55–4

.302

Not

incl

uded

in th

e fin

al m

ulti

vari

ate

mod

el1.

566,

1.0

22–5

.632

Not

incl

uded

in th

e fi

nal m

ulti

vari

ate

mod

elD

iast

olic

blo

od p

ress

urep

er 1

0 m

m H

g8.

344,

2.8

58–1

6.37

4N

ot in

clud

ed in

the

final

mul

tiva

riat

e m

odel

8.03

4, 1

.452

–13.

473

Not

incl

uded

in th

e fi

nal m

ulti

vari

ate

mod

elD

iabe

tes

dur

atio

n2.

750,

1.1

50–6

.576

2.97

3,0.

797–

5.96

1, 0

.020

2.65

3, 1

.053

–3.4

063.

392,

0.8

67–3

.027

, 0.0

20L

evel

of H

bA1c

per

1%7.

445,

2.9

46–1

6.33

53.

829,

0.83

7–6.

229,

0.0

016.

778,

2.7

45–9

.67

5.94

8, 2

.988

–8.1

83, 0

.001

HD

L-c

hole

ster

ol0.

650,

0.1

44–2

.928

0.49

3,0.

235–

1.46

3, 0

.159

1.33

3, 0

.254

–6.7

020.

969,

0.1

46–3

.435

, 0.5

47L

DL

-cho

lest

erol

2.77

8, 1

.308

–7.9

211.

173,

0.79

7–3.

953,

0.3

251.

925,

0.9

67–2

.310

1.10

2, 0

.344

–2.1

04, 0

.435

TC

/H

DL

rati

o3.

473,

1.5

87–8

.672

1.29

0,0.

302–

3.12

9, 0

.267

1.

253,

0.4

71–3

.003

, 0.3

82Tr

igly

ceri

des

2.25

4, 1

.278

–5.7

862.

891,

1.02

2–6.

109,

0.0

502.

651,

0.9

89–7

.607

2.34

1, 0

.094

–5.7

72, 0

.050

Insu

lin tr

eatm

ent

NA

NA

1.08

4, 0

.988

–1.6

892.

702,

0.8

73–5

.643

, 0.0

40D

iabe

tic

reti

nopa

thy

5.25

6, 2

.081

–13.

276

3.17

6,1.

114–

7.48

2, 0

.020

4.76

2, 2

.331

–8.0

233.

273,

1.0

25–4

.332

, 0.0

20D

iabe

tic

reti

nopa

thy

seve

rity

6.77

2, 2

.871

–16.

055

4.88

9,1.

279–

6.97

6, 0

.001

6.10

2, 2

.027

–9.1

025.

882,

2.9

87–7

.657

, 0.0

01H

DL

= H

igh

den

sity

cho

lest

erol

LD

L =

Low

den

sity

cho

lest

erol

TC

/H

DL

= T

otal

cho

lest

erol

/hi

gh d

ensi

ty.

Age

*= g

roup

s of

age

are

des

crib

ed in

Met

hod

s. C

I= c

onfid

ence

inte

rval

OR

= O

dd

s ra

tio,

NA

= N

ot a

vaila

ble,

bec

ause

all

pati

ents

wit

h ty

pe 1

dia

bete

s m

ellit

us a

re in

sulin

tr

eate

d. N

ot in

clud

ed in

the

final

mul

tiva

riat

e m

odel

= in

dic

ate

that

var

iabl

e w

as n

ot s

igni

fican

t and

thus

not

incl

uded

in th

e fi

nal m

ulti

vari

ate

mod

el.

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

12 Romero-Aroca Pedro et al.

Ophthalmic Epidemiology

have studied 8,187 type 2 and 488 type 1 patients, is much more highly representative of the diabetes popu-lation. These numbers represent 83.37% of diagnosed (registered) type 2 DM and 85.76% type 1 DM patients in our Health Care Area (HCA).

In the present study, there is an observable decrease in the prevalence of diabetic retinopathy in patients with type 2 DM (from a previously published study) from 39.41% in 1993 to 26.11% in the present study.18 In Spain, there has also been a decrease in the prevalence of diabetic retinopathy. The most recent studies pub-lished have reported a decrease, with values of preva-lence between 20.41%–34.50%. These differences may be due to the methodology used, patient selection, or using population-based studies against clinical-setting studies.19–21

Regarding the published studies around the world, there are also differences in the ratio of diabetic retin-opathy. If the studies have a clinical setting, the preva-lence is higher than in population-based studies. Thus in the UKPDS, Kohner made a clinical-based study5 and encountered a prevalence of 39% in men and 35% in women in a large population of 2,964 patients in the UK. However, in the same country a population-based study by Davis22 on 3,027 patients, found a prevalence of 21%. In addition, Lundman23 in Sweden, following a population-based study of 4,127 patients identified in a health district, determined a 27% prevalence of diabetic retinopathy. We can conclude therefore that if the studies are population-based the prevalence of diabetic retinopathy decreases against the results from clinical studies.

On the other hand, in type 1 DM patients the preva-lence of diabetic retinopathy is higher, with a value of 36.47%. This is similar data to the studies in the UK, with prevalence between 33.6% and 36.7% and in the EURODIAB study (performed from 31 centers in 16 European countries), in which the overall mean prevalence for all the participating centers was found to be 35.9%.24 Taking into account that the mean dura-tion of diabetes in the present study is higher in type 1 DM (20.42 SD 7.57 years [1–35 years]) than in type 2 DM (12.42 SD 6.30 years [1–37 years]) patients, this may be a confounding risk factor that may explain the differences.

For the most severe forms of diabetic retinopa-thy (nonproliferative severe and proliferative) we observed a low prevalence in both types of patients, only 1.07% in type 1 DM and 0.56% in Type 2 DM with proliferative diabetic retinopathy present, against the number of patients treated previously with laser, with values of 7.17% in type1 DM and 6.92% in type 2 DM patients. These findings may be due to a better con-trol of the complications of ophthalmic diabetes by ophthalmologists in our area.5

The statistical study shows that the more important diabetic retinopathy risk factors are: diabetes dura-tion, high HbA1c levels, arterial hypertension (more related to an increase in diastolic than systolic blood pressure). For type 2 DM, treatment with insulin is also a risk factor, findings that are the same as previously published studies.5,25

In the present study, the prevalence of diabetic macular edema (the leading cause of decrease in visual acuity) gives a similar value as other studies. Figure 1 shows that the duration of DM was the most important at p<0.001. Macular edema increases from 0% to 14.98% in type 1 DM and from 1% to 15.92% in type 2 patients DM. a finding similar to that obtained in the Wisconsin Epidemiological Study of Diabetic Retinopathy.26 Other risk factors in the present study were higher glycosy-lated hemoglobin and overt nephropathy (proteinuria) in both DM types, and insulin treatment for type 1 patients. We can speculate that bad diabetic metabolic control of the patients, measured by HbA1c and the use of insulin in type 2 patients, were major causes in the appearance of diabetic macular edema.

The relationship between lipids and diabetic macu-lar edema has been widely studied in the literature with different results. In a cross sectional study in the WESDR,27 higher total cholesterol was positively associated with the presence of hard exudates in type 1 DM patients, but in a prospective analysis there was no association of lipids with macular edema after controlling for HbA1c and other risk factors.28 In a prospective analysis of the ETDRS data,29 the develop-ment of hard exudates were 50% faster among subjects with elevated baseline levels of total cholesterol and triglycerides and 36% faster among participants with higher baseline levels of LDL-Cholesterol compared with participants with normal lipid levels at baseline. In the present study, although we observed that the adjustment for HbA1c considerably attenuated many of the relationships between lipids and diabetic retin-opathy end points, several associations persisted with high triglyceride levels but not with LDL-cholesterol. However, if we introduce the TC/HDL ratio, we observe that this factor is significant in both DM types, so we could suppose that LDL-cholesterol levels are significant with macular oedema. The TC/HDL ratio,17,30 known as the atherogenic or Castelli index is an important indicator of vascular risk, the predictor of which is greater than the isolate param-eters (Total-cholesterol, LDL-Cholesterol or HDL-cholesterol). Determining it is important when there is no reliable calculation of LDL-cholesterol, as when triglyceridemia exceeds 300 mg/dl (3.36 mmol/L). Individuals with a high TC/HDL ratio have greater cardiovascular risk due to an imbalance between the cholesterol carried by atherogenic and protective

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

Prevalence of Microangiopathy in Spain 13

© 2010 Informa UK Ltd.

lipoproteins. This may be due to an increase in the atherogenic component contained in the numerator (LDL- cholesterol + VLDL-cholesterol), a decrease in the anti-atherosclerotic trait of the denominator, or both. Its association with diabetic macular edema in the present study is significant and its Odds Ratio is greater than LDL-cholesterol and triglycerides in univariate analysis, a relationship that remained after controlling for other potential confounders, such as HbA1c levels, arterial hypertension, duration of diabe-tes or overt nephropathy. This last risk factor is impor-tant because overt nephropathy produces an increase in triglyceride levels that may reduce the importance of LDL levels as a risk factor. Also, the inclusion of the TC/HDL ratio in the study allows us to eliminate the confounding action of high triglyceride levels and to verify the relationship of LDL-cholesterol with macular edema. In the literature, we found only one reference in which high levels of TC/HDL and other lipoproteins has been associated with maculopathy in a study by Mohan,31 but it is a case series with a limited number of patients. This relationship found in the present study should be corroborated by more lon-gitudinal studies in order to clarify any relationship between macular edema and TC/HDL ratio. This ratio is a marker of the risk factor for atherogenesis and dia-betic macroangiopathy that includes cardiovascular diabetic diseases and stroke, and may be the clue to the association between micro and macro angiopathies in diabetes patients.

A relationship between lipid levels and macular edema appears to be biologically plausible. High lipid levels are known to cause endothelial dysfunction30 via a local inflammatory response, with consequent release of cytokines and growth factors, activation of oxygen sensitive biological changes in vessel walls, increases in LDL oxidation, and quenching of nitric oxide. In turn, endothelial dysfunction in the diabetic vasculature results in blood retinal barrier breakdown in animal models of DR,32–34 though data is lacking in humans.

The prevalence of nephropathy, represented by microalbuminuria and overt nephropathy in the pres-ent study, with values of 25.61% for microalbuminu-ria and 8.60% for overt nephropathy in type 1 DM, is similar to other published studies in our country.6,7 In type 1 DM patients, published studies have shown that approximately 25 to 45% develop clinically evi-dent renal disease during their lifetime, developing 10 to 15 years after onset. Figure 2 shows a peak of renal lesions at around 11–15 years of diabetes duration (a finding which agrees with published studies), fol-lowed by a decrease in prevalence and a later increase in patients after 20 years.

As in other studies, the risk factors for nephropa-thy (microalbuminuria and overt nephropathy) in the

present study were: above 15 years duration of DM, high levels of HbA1c, presence of arterial hypertension (diastolic blood pressure was significant, but not sys-tolic), high triglycerides levels, the presence and sever-ity of proliferative diabetic retinopathy, and insulin treatment for type 2 DM. But in the analysis of diabetic overt nephropathy we should bear in mind that the increases in blood pressure or lipid levels may be a consequence of renal failure and not the cause.35–40

The relationship between diabetic retinopathy and nephropathy in the present study showed that for dia-betic retinopathy microalbuminuria is a good predic-tor for type 1 DM but not for type 2. This relationship is similar in the case of diabetic macular edema. On the contrary, the relationship with overt nephropathy is significant in both types of diabetes mellitus, with diabetic retinopathy and macular edema. For the study of renal lesions (microalbuminuria and overt nephropathy), the proliferative diabetic retinopathy form is correlates well with both forms of renal lesion. Despite the relationship between diabetic retinopathy and nephropathy being more related in type 1 DM patients, Table 2 shows that the prevalence of patients only with nephropathy (microalbuminuria or overt nephropathy) is higher in type 2 DM patients, a finding that may be due to the presence of other renal injuries than to diabetes in this group of patients.

The more important limitation of the present study is that the design is transversal not prospective, so dif-ferent risk factors may be altered by the later follow up of patients throughout the life of the study.

In conclusion, the present study is highly represen-tative of our diabetes population, and the obtained prevalence values of diabetic retinopathy and neprhopathy are certainly representative. Despite a decrease in prevalence of severe forms of diabetic retinopathy in our country, the retinal diabetic lesion persists as an important Health Care problem, and diabetic patients need to be strictly monitored in order to lower the number of patients who develop blindness or loss of vision. An important finding analyzed in present study is the TC/HDL ratio, which has been poorly studied in the literature, and has allowed us to determine the importance of LDL-cholesterol as a risk factor for diabetic macular edema, despite the presence of high levels of Triglyc-erides. This is frequent in diabetes patients so more studies should be made to determine this positive association. In the study of the relationship between diabetic nephropathy and retinopathy we found that microalbuminuria is a risk factor for type 1 DM but not type 2 DM patients, and correlates well with the severity of diabetic retinopathy. Also, overt neph-ropathy is correlates well with diabetic retinopathy and macular edema.

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

14 Romero-Aroca Pedro et al.

Ophthalmic Epidemiology

AcKnoWlEdgMEntS

To all the Family Physicians in our area who helped us to implement the new screening system using the non-mydriatic fundus camera and to our camera technicians for their work and interest in the new method.The local scientific investigative ethical committee also gave us their agreement, according Helsinki declara-tion. All patients of second group signed an informed consent to be given into the study

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

rEfErEncES

1. King H, Aubert RE, Herman WH. Global burden of diabetes, 1995–2025: prevalence, numerical estimates, and projections. Diabetes Care. 1998;21(9):1414–1431.

2. The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus (2000). Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 23:S4–S19.

3. Geiss L, Engelgau M, Fraizer E, Tierney E. Diabetes sur-veillance, 1997. Centers for Disease Control and Prevention. U.S. Department of Health and Human Services Atlanta GA. 1997.

4. Diabetes Control and Complications trial/Epidemiology of Diabetes Interventions and Complications Research Group. Prolonged effect of intensive therapy on the risk of retinopa-thy complications in patients with type 1 diabetes mellitus. Arch Ophthalmology. 2008;126(12):1707–1715.

5. Kohner EM, Aldington SJ, Stratton IM, Manley SE, Hol-man RR, Turner RC. United Kingdom Prospective Diabetes Study, 30: diabetic retinopathy at diagnosis of non-insulin-dependent diabetes mellitus and associated risk factors. Arch Ophthalmol. 1998;116(3):297–303.

6. Consell Assesor sobre Diabetes a Catalunya. Registre de diabetes tipus 1 de Catalunya, 1995–2002. Butlleti Epide-miològic de Catalunya. Vol XXIII, num 4. Barcelona. Depar-tament de Sanitat i Seguretat Social, 2002.

7. Castell C, Tresserras R, Lloveras G, Goday A, Serra J, Salleras L. Prevalence of diabetes in Catalonia, an OGTT-based population study. Diab Res Clin Prac. 1999;43(1): 33–40.

8. Aldington SJ, Kohner EM, Meuer S, Klein R, Sjolie AK. Methodology for retinal photography and assessment of diabetic retinopathy, The EURODIAB IDDM complications study. Diabetologia. 1995;38(4):437–444.

9. Wilkinson CP, Ferris FL 3rd, Klein RE, Lee PP, Agardh CD, Davis M, Dills D, Kampik A, Pararajasegaram R, Ver-daguer JT; Global Diabetic Retinopathy Project Group. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthal-mology. 2003;110(9):1677–1682.

10. ETDRS. Detection of diabetic macular edema study n°5. Ophthalmology. 1989;96:746–751.

11. The Sixth Report of the Joint National Committee on Pre-vention, Detection, Evaluation and treatment of High Blood Pressure. Arch Intern Med. 1997;157(21):2413–2442.

12. American Diabetes Association. Clinical practice recommen-dations. Diabetes Care. 2002;25, suppl 1.

13. European Diabetes Policy Group. A Desktop guide to type I (insulin-dependent) diabetes mellitus 1998–199. Guidelines for diabetes care. Diabetic Med. 1999; 16(3):253–266.

14. American Diabetes Association. Report of the Expert Com-mittee on the diagnosis and classification of diabetes mel-litus. Diabetes Care. 1997;20:1183–1201.

15. Mogensen CE, Christensen CK, Vittinghus E. The stages in diabetic renal disease, with emphasis on the stage of incipient diabetic nephropathy. Diabetes. 1983;32 Suppl 2:64–78.

16. Expert Panel on Detection, Evaluation, And Treatment of high Blood Cholesterol in Adults (Adult Treatment Panel III). Executive summary of the third report of the National Cholesterol Education Program (NCEP). JAMA. 2001;285:2486–2497.

17. Rifai N, Ridker PM. Proposed cardiovascular risk assess-ment algorithm using high-sensitivity C-reactive protein and lipid screening. Clin Chem. 2001;47(1):28–30.

18. Romero-Aroca P, del Castillo-Déjardin D. Estudio de preva-lencia de retinopatía diabética en la población del Baix Camp (Tarragona). Arch Soc Esp Oftalmol. 1996;71:261–268.

19. López IM, Díez A, Velilla S, Rueda A, Alvarez A, Pastor CJ. Prevalence of diabetic retinopathy and eye care in a rural area of Spain. Ophthalmic Epidemiol. 2002;9(3):205–214.

20. Santos-Bueso E, Fernández-Vigo J, Fernández-Pérez C, Macarro-Merino A, Fernández-Perianes J. Prevalencia de retinopatía diabética en la Comunidad Autónoma de Extremadura. 1997–2001 (Proyecto Extremadura para prevención de la ceguera). Arch Soc Esp Oftalmol. 2005;80(3):187–194.

21. Teruel-Macias C, Fernández-Real JM, Ricart W, Valent-Ferrer R, Vallés-Prats M.Prevalencia de la retinopatía diabética en la población de diabéticos diagnosticados en las comarcas de Girona. Estudio de los factores asociados. Arch Soc Esp Oftalmol. 2005;80(3):85–91

22. Davis TM, Stratton IM, Fox CJ, Holman RR, Turner RC. UK Prospective Diabetes Study 22. Effect of age at diagnosis on diabetic tissue damage during the first 6 years of NIDDM. Diabetes Care. 1997;20 (9):1435–1441.

23. Lundman B, Engstrom L. Diabetes and it’s complications in a Swedish county. Diabetes Res Clin Pract. 1998;39 (2): 157–164.

24. Toeller M, Buyken AE, Heitkamp G, Berg G, Scherbaum WA.Prevalence of chronic complications, metabolic con-trol and nutritional intake in type I diabetes: comparison between different European regions. EURODIAB Complica-tions Study group. Horm Metab Res. 1999;31:680–685.

25. Klein R, Klein BE, Moss SE, Cruickshanks KJ. The Wis-consin Epidemiologic Study of Diabetic Retinopathy: XVII. The 14-year incidence and progression of diabetic retinopathy and associated risk factors in type 1 diabetes. Ophthalmology. 1998;105(10):1801–1815.

26. Klein R, Moss SE, Klein BE, Davis MD, DeMets DL. The Wisconsin epidemiologic study of diabetic retinopathy. XI. The incidence of macular edema. Ophthalmology. 1989;96(10):1501–1510.

27. Klein BE, Moss SE, Klein R, Surawicz TS: The Wisconsin Epidemiologic Study of Diabetic Retinopathy. XIII. Relation-ship of serum cholesterol to retinopathy and hard exudate. Ophthalmology. 1991;98:1261–1265.

28. Klein BE, Klein R, Moss SE. Is serum cholesterol associated with progression of diabetic retinopathy or macular edema in persons with younger onset diabetes of long duration? Am J Ophthalmol. 1999;128(5):652– 654.

AQ4

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

Prevalence of Microangiopathy in Spain 15

© 2010 Informa UK Ltd.

29. Chew EY, Klein ML, Ferris FL 3rd, Remaley NA, Mur-phy RP, Chantry K, Hoogwerf BJ, Miller D. Association of elevated serum lipid levels with retinal hard exudate in diabetic retinopathy: Early Treatment Diabetic Retin-opathy Study (ETDRS) report 22. Arch Ophthalmol. 1996;114(9):1079 –1084.

30. Millán J, Pintó X, Muñoz A, et al. Lipoprotein ratios: Physi-ological significance and clinical usefulness in cardiovascu-lar prevention. Vasc Health Risk Manag. 2009;5:757–65.

31. Mohan R, Mohan V, Susheela L, Ramachandran A, Viswa-nathan M. Increased LDL cholesterol in non-insulin-de-pendent diabetics with maculopathy. Acta Diabetol Lat. 1984;21(1):85–89.

32. Landmesser U, Hornig B, Drexler H. Endothelial dysfunc-tion in hypercholesterolemia: mechanisms, pathophysi-ological importance, and therapeutic interventions. Semin Thromb Hemost. 2000;26(5):529 –537.

33. Joussen AM, Murata T, Tsujikawa A, Kirchhof B, Bursell SE, Adamis AP. Leukocyte mediated endothelial cell injury and death in the diabetic retina. Am J Pathol. 2001;158(1):147–152.

34. Joussen AM, Poulaki V, Mitsiades N, et al. Nonsteroidal anti-inflammatory drugs prevent early diabetic retinopathy via TNF-alpha suppression. FASEB J. 2002;16(3):438–440.

35. Brancati FL, Cusumano AM. Epidemiology and prevention of diabetic nephropathy. Current Opinion in Nephrolology and Hypertension. 1995;4(3):223–229.

36. Jawa A, Kcomt J, Fonseca VA. Diabetic nephropathy and retinopathy. Med Clin N Am. 2004;88(4):1001–1036.

37. Hovind P, Tarnow L, Rossing K, Rossing P, Eising S, Larsen N, Binder C, Parving HH. Decreasing incidence of severe diabetic microangiopathy in type 1 diabetes. Diabetes Care. 2003;26(4):1258–1264.

38. Rossi MC, Nicolucci A, Pellegrini F, et al. Identifying patients with type 2 diabetes at high risk of microalbu-minuria: results of the DEMAND (Developing Education on Microalbuminuria for Awareness of Renal and cardio-vascular risk in Diabetes) Study. Nephrol Dial Transplant. 2008;23(4):1278–1284.

39. Skrivarhaug T, Bangstad HJ, Stene LC, Sandvik L, Hanssen KF, Joner G. Low risk of overt nephropathy after 24 yr of childhood-onset type 1 diabetes mellitus (T1DM) in Norway. Pediatr Diabetes. 2006;7(5):239–246.

40. Pambianco G, Costacou T, Ellis D, Becker DJ, Klein R, Orchard TJ. The 30-year natural history of type 1 diabetes complications: the Pittsburgh Epidemiology of Diabetes Com-plications Study experience. Diabetes. 2006;55(5):1463–1469.