- Email: [email protected]
The temporal development of retinopathy and nephropathy in type 1 diabetes mellitus during 15 years diabetes duration
Diabetes Research and Clinical Practice 45 (1999) 15 – 23 www.elsevier.com/locate/diabres
The temporal development of retinopathy and nephropathy in type 1 diabetes mellitus during 15 years diabetes duration M. Lo¨vestam-Adrian a, E. Agardh b, C.D. Agardh b,* b
a Department of Ophthalmology, Uni6ersity Hospital, S-221 85 Lund, Sweden Department of Endocrinology, Uni6ersity Hospital MAS, S- 205 02 Malmo¨, Sweden
Received 10 December 1998; received in revised form 15 March 1999; accepted 6 April 1999
Abstract In this prospective study, the 10- and 15-year incidence and course of retinopathy were examined in relation to medical risk indicators from 3.1 91.9 (mean9SD) years diabetes duration and onwards in 29 adult type 1 diabetic patients treated under routine care. A total of 28 patients were followed for 10 years and 20 patients for 15 years diabetes duration. After 10 years diabetes duration, 11 patients had developed any retinopathy (ten patients background retinopathy and one patient clinically-significant macular oedema). After 15 years diabetes duration, 16 patients had developed any retinopathy (12 patients developed background retinopathy and four patients developed potentially sight-threatening retinopathy, i.e. clinically significant macular oedema (n =2) or severe non-proliferative retinopathy (n =2)). None of the patients developed proliferative retinopathy. No differences were seen in mean HbA1c values between patients without any retinopathy and patients with background retinopathy, whereas patients who developed potentially sight-threatening retinopathy had higher mean HbA1c levels than patients without any retinopathy. Patients who developed potentially sight-threatening retinopathy had higher levels of mean HbA1c, both after 10 (9.79 1.6 vs 6.99 1.5%; P B0.05) and 15 years diabetes duration (9.39 1.2 vs 7.19 1.3%; PB0.05), compared to patients without any retinopathy. They also had higher levels of mean HbA1c than patients with background retinopathy after 15 years diabetes duration (9.3 9 1.2 vs 7.7 9 1.1%; PB 0.05). There were no differences in blood pressure levels between patients who developed retinopathy and those who did not. Only two patients developed clinical signs of nephropathy (urinary albumin 320 – 1590 mg/l) after 12 and 13 years diabetes duration, respectively. At those time points, both patients had already developed background retinopathy since 2 years. In conclusion, the present study shows that the incidence of retinopathy is associated with the duration of diabetes and that there is a strong association between the degree of metabolic control and development of potentially sight-threatening retinopathy. The study also indicates that the development of retinopathy does not seem to be associated with hypertension or clinical signs of nephropathy. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Retinopathy; Nephropathy; Metabolic control; Duration; Routine care; Diabetes mellitus
* Corresponding author. Tel.: + 46-40-331016; fax: + 46-40-337023. 0168-8227/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 9 9 ) 0 0 0 5 5 - 8
16
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
1. Introduction It is a well-known fact that type 1 diabetes mellitus is accompanied by micro- and macrovascular complications. Studies have shown an association between metabolic control and the incidence and progression of retinopathy and nephropathy [1–4]. In humans, intensified insulin treatment not only delays the onset of diabetic retinopathy but also retards the progression of both retinopathy and nephropathy [5,6], although a transient deterioration of retinopathy may occur during intensified glycaemic control [7,8]. However, there are few prospective studies of patients treated under routine care from near the onset of diabetes mellitus. In a recent study, in which patients were followed from the onset of diabetes, a possible reduction in the frequency of retinopathy was suggested in patients with strict metabolic control from the diagnosis of diabetes [9]. In the present study we were able to follow most of the patients for 15 years from near the onset of diabetes. The yearly mean levels of HbA1c that could be achieved under routine care were correlated to the incidence and course of retinopathy. We also studied a possible association between blood pressure levels, nephropathy and the development of retinopathy.
2. Materials and methods
2.1. Patients Adult patients (29) with type 1 diabetes mellitus participated in this prospective study which started in 1982. The patients have regularly attended the out-patient clinic at the Department of Internal Medicine, University Hospital, Lund, 4.3 92.3 (mean9 SD) times per year. At study entry, the mean duration of diabetes was 3.1 9 1.9 years. Of the patients, 26 received insulin treatment at the time of diagnosis and three patients received insulin treatment within 18 months of diagnosis. Of the patients, 28 were followed for 10 years and 20 patients for 15 years diabetes duration. During the observation period, two patients died, one from pneumonia after five years of
diabetes and the other from ketoacidosis after 13 years of diabetes. Among the seven patients who were followed for 10 years only, five patients moved from the area, one patient did not want to participate and one patient was recruited into the study 2 years later than the others. Age, age at onset, body mass index, and insulin dosage were registered. The degree of metabolic control was measured with mean glycosylated haemoglobin (HbA1c) values 3.2 9 2.0 times per year. Clinical signs of diabetic nephropathy were examined with albumin-creatinine clearance ratio, urinary albumin and serum creatinine concentrations. Normoalbuminuria was defined as a urinary albumin concentration B30 mg/l, and clinical nephropathy \ 299 mg/l. Glomerular filtration rate (GFR) was measured after 5, 10 and 15 years of diabetes duration. The blood pressure levels were measured 2.89 1.6 times per year in the supine position using a mercury sphygmomanometer. Diastolic blood pressure was taken at Korotkoff phase V.
2.2. Classification of retinopathy The patients were regularly examined at the Department of Ophthalmology, every other year in patients without retinopathy, and every 3–12 months, depending on the degree of retinopathy, in the others. The retinal examination was performed by fundus photography through a dilated pupil with three fields per eye (45°) nasal, temporal and central, including stereo photos of the macular region (Nikon NFC 50). The degree of retinopathy was based on fundus photo-grading of the eye that showed the most advanced retinopathy. In addition, fluorescein angiography was performed after five, 10 and 15 years diabetes duration. The different degrees of retinopathy were classified as no retinopathy, background retinopathy and potentially sight-threatening retinopathy. Potentially sight-threatening retinopathy was defined as clinically significant macular oedema or severe non-proliferative retinopathy, using ETDRS definitions [10] or proliferative retinopathy.
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
2.3. Analytical techniques HbA1c levels were analysed by ion-exchange chromatography using commercially available microcolumns (Bio-Rad, Richmond, CA) or by fast liquid chromatography (Kontron Instruments, Milan, Italy). Normal value for both methods is B 5.3%. Urine was collected in polystyrene tubes and stored at 4°C for less than 4 days before analysis. Urinary albumin concentration was measured with an electroimmunoassay using human albumin (Kabi Vitrum, Stockholm, Sweden) (detection limit 12.5 mg/l) or by turbidimetry with an automated analyser (Cobras Mira, Roche), antibodies (rabbit anti-human albumin) and technique as described by Dakopatts, Copenhagen (detection limit 5 mg/l). Urine- and serum-creatinine levels were analysed by a kinetic Jaffe reaction or by an enzymatic method (creatinine-hydrolase; EKTA Chem-analyser, Instrument Kodak). The second method results in about 8% higher values than the first one. This was adjusted for in order to create comparable values. Normal values were for women 48 – 100 mmol/l and for men 55 – 116 mmol/l. The urinary albumin creatinine clearance ratio was calculated as the ratio between the urine albumin clearance and the urine creatinine clearance; normal value B 0.01× 10 − 3. Plasma albumin was analysed by a bromcresolpurpur-method or by immuniturbidimetry. Glomerular filtration rate (GFR) was determined with a plasma clearance technique using the non-ionic X-ray contrast medium iohexol, as a marker. The method has a total imprecision of about 11% (CV), essentially all of which is accounted for by intra-individual biological variation.
2.4. Statistical methods The mean of 2– 4 observations per year for HbA1c, systolic blood pressure, diastolic blood pressure and urinary albumin was calculated. The statistical evaluation of the results was based on a repeated measure analyses-of-variance technique (ANOVA, SPSS). Values are given as mean9 SD. Urinary albumin concentration and urinary albumin creatinine clearance ratio are given as median
17
and range. Fisher’s exact test was used for categorical variables, the Student’s two-tailed t-test was used for equal standard deviations and Welch’s approximate t-test for unequal standard deviations. Correlations between variables were tested with Pearsons’ test.
3. Results
3.1. Patient characteristics At entry in 1982, the mean diabetes duration was 3.19 1.9 years. At this time point there were no differences regarding patient characteristics, except for HbA1c levels, which were higher among patients who later developed potentially sightthreatening retinopathy than those who did not (Table 1). None of the patients had anti-hypertensive medication and during the study, three patients received such treatment after 10 (n= 2) and 12 (n= 1) years diabetes duration. None of the five patients who moved, or the patient who did not want to participate any longer, developed any retinopathy during the first 10-year follow-up period. These patients did not differ at entry into the study from the other patients without retinopathy regarding age at onset (24.2911.3 years), HbA1c (10.49 1.5%), systolic blood pressure (1319 14 mmHg), diastolic blood pressure (73911 mmHg), S-Creatinine (859 11 mmol/l), insulin dosage (0.369 0.22 U kg-1) (mean9 SD) or urinary albumin levels 13 mg/l (5–13 mg/l; median, range).
3.2. Retinopathy grading At entry, no subject showed any signs of diabetic retinopathy. The incidence of any type of retinopathy after 10 years diabetes duration was 39% (eight males and three females) and after 15 years duration 80% (ten males and six females) (Table 1). The incidence of potentially sightthreatening retinopathy after 15 years diabetes duration was 20% (three males and one female). Three of the four patients with potentially sightthreatening retinopathy after 15 years diabetes duration had background retinopathy after 10
18
Sex (male/female) Age at onset of diabetes HbA1c (%) SBP (mmHg) DBP (mmHg) Albumin creatinine clearance ratio (×10−3) Urinary Albumin (mg/l) Serum creatinine (mmol/l) Insulin (U/kg) Body mass index (kg/m2) Antihypertensive Rx (n) a
No retinopathy at follow up
Background retinopathy at follow-up
Potentially sight-threatening retinopathy at follow-up
Entry (n =20)
Entry
Entry
12/8 28.0 95.2 10.3 9 2.8 130 99 80 97 B0.01 13 (5–13) 68 9 12 0.64 9 0.26 23.0 91.9
Follow-up (n = 4) 2/2 7.1 91.3 1339 15 75 99 B0.01 13 (5–27) 67 914 0.65 9 0.16 24.8 93.2 1
Follow-up (n=12)
Follow-up (n=4)
7/5 30.8 95.2 10.4 9 2.2 124 9 6 76 96 B0.01 13 (5–13) 74 9 12 0.52 90.24 22.2 9 2.0
7.7 91.1+ 12597 769 8 B0.01 12.5 (5–15) 719 12 0.63 9 0.15 24.49 2.4 1
3/1 28.3 9 6.4 11.992.0 1309 27 799 4 B0.01 13 (5-13) 66912 0.4190.08 22.992.8
9.3 91.2* 123 9 11 739 7 B0.01 (B0.01–0.04) (7–1590) 77 9 22 0.599 0.1 23.69 1.8 1
Values are given as mean(SD and for urinary albumin as median and range. SBP+systolic blood pressure; DBP+diastolic blood pressure. * PB0.05; patients without retinopathy vs. potentially sight-threatening retinopathy. + PB0.05; patients with background vs. potentially sight-threatening retinopathy.
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
Table 1 Patient characteristics at entry in relation to degree of retinopathy after 15 years diabetes durationa
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
years diabetes duration. The fourth patient developed background retinopathy after 12 years duration and potentially sight-threatening retinopathy after 15 years duration. The only patient who developed potentially sight-threatening retinopathy, i.e. clinically significant macular oedema, after 10 years duration, had an onset of diabetes in 1984, and has not yet been followed for 15 years. Among the other patients with potentially sightthreatening retinopathy, two patients developed clinically significant macular oedema and two patients severe non-proliferative retinopathy. None developed proliferative retinopathy. Patients with clinically significant macular oedema were treated with focal laser photocoagulation, and one of the patients with severe non-proliferative retinopathy received panretinal photocoagulation. In two of the five patients who moved after 10 years duration, fundus photos were available after 15 years duration and showed no signs of retinopathy. The fluorescein angiography confirmed the changes seen on fundus photos. The patients with clinically significant macular oedema showed central leakage in the mid-and late phase angiograms, and those with severe non-proliferative retinopathy showed numerous zones of 2 – 4 optic disc diameter areas of non-perfusion. In five patients, however, microaneurysms could be detected 2 – 3 years before they were visible on fundus photos.
3.3. Metabolic control Patients who developed potentially sight-threatening retinopathy had higher levels of mean HbA1c both after 10 (9.79 1.6 vs 6.991.5%; P B 0.05) and 15 years diabetes duration (9.39 1.2 vs 7.191.3%; P B 0.05) compared to patients without any retinopathy. They also had higher levels of mean HbA1c than patients with background retinopathy after the 15 year follow-up period (9.391.2 vs 7.791.1%; P B0.05). No differences were seen between patients with background retinopathy and patients without any retinopathy regarding HbA1c levels. On a group basis there was an improved metabolic control over time (Fig. 1). The improvement was significant between five and 10
19
years diabetes duration in patients with background retinopathy (9.39 1.3 vs.7.6 9 1.0%; PB 0.001) as well as in patients without any retinopathy (9.49 3.0 vs.6.9 9 1.5%; PB 0.05). Those patients who developed potentially sightthreatening retinopathy also had a tendency towards improved metabol control over time, however not significant.
3.4. Blood pressure and albuminuria There were no differences regarding systolic and diastolic blood pressure levels between patients who developed any degree of retinopathy and those who did not. Blood pressure levels were stable over the study period (Fig. 2). None of the patients received anti-hypertensive treatment at entry. After 15 years diabetes duration, three patients were treated with ACE-inhibitors. In two of these patients, the time for institution of anti-hypertensive treatment coincided with the deterioration of retinopathy. The mean level of systolic and diastolic blood pressure levels before institution of anti-hypertensive medication was 1529 10 and 9095 mmHg, respectively. Two patients developed clinical signs of nephropathy with urinary albumin levels of 320–1590 mg/l and an albumin creatinine-clearance ratio of 0.049 10 − 3 –0.089 10 − 3 after 12 and 13 years diabetes duration, respectively. At those time points, both patients had already developed background retinopathy since 2 years. One of the patients with diabetic nephropathy died from ketoacidosis 1 year after the development of nephropathy and still had background retinopathy at the time for death. In the other patient who developed diabetic nephropathy, retinopathy deteriorated to potentially sight-threatening retinopathy after another 2 years. Both patients had mean HbA1c levels \ 10% during their first 10 years of diabetes duration. The mean glomerular filtration rate (GFR) for the whole group was 1309 20 ml/min × 1.73 m2, 1219 21 ml/min × 1.73 m2 and 1079 22 ml/ min×1.73 m2 after 5, 10 and 15 years diabetes duration, respectively. There were no differences in GFR according to degree of retinopathy. The two patients who developed nephropathy did not
20
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
differ in GFR. The GFR values were correlated to the HbA1c levels (r =0.49, P B 0.01).
4. Discussion Long-term prospective follow-up studies are necessary in order to evaluate the effect of glycaemic control from the onset of diabetes mellitus on the development of retinopathy and nephropathy. The patients in this study all attended the out-patient clinic at the Department of Internal Medicine, University Hospital, Lund. This clinic takes care of all type 1 diabetic patients in the district and the patients in our study were treated under routine care. The sex-distribution in our patient group, 60% men and 40% women, indicates that diabetes mellitus is slightly more common among men. This is in agreement with figures for the whole country describing an incidence of 14.6/100 000 men and
8.8/100 000 women in the age group 25–29 years for type 1 diabetes mellitus [11]. However, the sex distribution according to the different levels of retinopathy, although based on small numbers, with 58% of patients with background retinopathy and 75% of patients with potentially sightthreatening retinopathy being men, shows a predominance for men in this study in agreement with another study [12], whereas the frequency of men and women was similar in others [13,14]. In the present study, the 10-year incidence of any retinopathy was 39% and the 15-year incidence was 80%. These incidence rates are in accordance with some studies [13,15] but lower than in others [16]. The incidence of potentially sightthreatening retinopathy after 15 years diabetes duration was 20% and none had developed proliferative retinopathy. Since we know, that two of the patients who moved did not have any retinopathy after 15 years diabetes duration and none of the other three patients who moved had
Fig. 1. HbA1c levels in relation to diabetes duration (mean9 SD). Circles () represent patients without retinopathy, triangles () patients with background retinopathy and filled squares ( ) patients with potentially sight-threatening retinopathy.*PB0.05 represents differences in HbA1c levels between 5 and 10 years diabetes duration. In addition there were differences in HbA1c levels between patients with potentially sight-threatening retinopathy and patients without any reinopathy after 10 and 15 years diabetes duration and between patients with sigh-threatening retinopathy and patients with background retinopathy after 15 years duration.
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
21
Fig. 2. Blood pressure levels in relation to diabetes duration (mean 9SD). Circles () represent patients without retinopathy, triangles () patients with background retinopathy and filled squares ( ) patients with potentially sight-threatening retinopathy.
any retinopathy after 10 years diabetes duration, the incidence figures after 15 years diabetes duration would have been even lower with these patients included. Although we found the development of potentially sight-threatening retinopathy but not background retinopathy to be associated with higher levels of HbA1c after 10 and 15 years diabetes duration, we found no statistically significant differences in HbA1c levels between patients who developed background retinopathy and patients who did not develop any retinopathy. However, there was a clear tendency towards higher levels in background retinopathy patients rendering it possible that this difference might well have reached statistical significance in a larger study. On a group basis, the degree of metabolic control was significantly improved in patients without any retinopathy and in patients who developed background retinopathy during the last 10 years, a phenomenon that has been previously described [17]. Among those who developed potentially sight-threatening retinopathy, however, HbA1c levels did not change significantly but showed a
tendency towards lower levels over time. In addition these patients seem to have been under-insulinised at entry compared to the others. The connection between hyperglycaemia and the development of retinopathy in type 1 diabetes is well documented [2,18,19]. However, the importance of good glycaemic control from the onset of disease for later development of retinopathy is under debate. The significance of excellent glycaemic control from the diagnosis has been emphasized [9], whereas a reduction in both the development and progression of retinopathy after intensified therapy has been demonstrated even later in the course of diabetes [4,6]. In a study from Israel, poor metabolic control during the last 3 years better predicted the risk for proliferative retinopathy than the mean of all HbA1 values [20]. In the report from the DCCT however, the degree of metabolic control, both before and during the study period, was found to influence the progression of retinopathy and pointed at the ‘total glycemic exposure’ to be the dominant risk factor for developing retinopathy. Anyhow, the benefits of improved metabolic control from early on of the disease [21] is of undisputable value.
22
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
Duration of diabetes is also known to be an important risk factor for the development of retinopathy [14,22,23]. However, not only the duration but also the age at onset of diabetes has been shown to play a role in the development of retinopathy, where age at onset of diabetes before or during puberty was found to be a significant risk factor for retinopathy [15,16,20], especially proliferative retinopathy [23]. In our study, the mean age at onset of diabetes was rather high, 26 years, with the youngest age at onset of 15 years. This could be one explanation for the lower incidence rates in our study and the fact that no patient developed proliferative retinopathy. Furthermore, the lack of other known risk factors such as proteinuria or hypertension [2,22,24,25] may have contributed to the lower incidence figures. We did not find any connection between retinopathy and hypertension, but in two of the four patients in whom the retinopathy deteriorated to potentially sight-threatening retinopathy, anti-hypertensive treatment was instituted during the follow-up period. Thus, it is possible that there could be a connection between hypertension and retinopathy in these patients in accordance with other studies [22,26]. Though the number of patients is small, it should be noted that the incidence of nephropathy was low and this is in agreement with another study, in which a declining incidence of nephropathy for type 1 diabetic patients has been shown over the last years [17]. In conclusion, the development of any type of retinopathy increased with the duration of diabetes and the development of severe but not background retinopathy was associated with worse metabolic control after 10 and 15 years diabetes duration. The metabolic control was improved during the last 10 years which shows that it is possible to achieve improved metabolic control over time, also under routine care.
Acknowledgements This study was supported by funds the Swedish Diabetes Federation, the Crafoord Foundation, the Crown Princess Margareta’s Committee for
the Blind, the Stig and Alme´rs Foundation, the Novo Nordisk Research Foundation and the Swedish Medical Research Council (Grant k9819x-12662-01A).
References [1] R. Klein, B.E.K. Klein, S.E. Moss, K.J. Cruickshanks, Relationship of hyperglycemia to the long-term incidence and progression of diabetic retinopathy, Arch. Intern. Med. 154 (1994) 2169 – 2178. [2] C-D. Agardh, E. Agardh, O. Torffvit, The association between retinopathy, nephropathy, cardiovascular disease and long-term metabolic control in type 1 diabetes mellitus: a 5 year follow-up study of 442 adult patients in routine care, Diabetes. Res. Clin. Pract. 35 (1997) 113 – 121. [3] C.E. Mogensen, Prediction of clinical diabetic nephropathy in IDDM patients. Alternatives to microalbuminuria?, Diabetes 39 (1990) 761 – 767. [4] The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N. Engl. J. Med. 329 (1993) 977 – 986. [5] P. Reichard, B-Y. Nilsson, U. Rosenqvist, The effect of long-term intensified insulin treatment on the development of microvascular complications of diabetes mellitus, N. Engl. J. Med. 329 (1993) 304 – 309. [6] The Diabetes Control and Complications Trial Research Group. Progression of retinopathy with intensive versus conventional treatment in the diabetes control and complications trial. Ophthalmology 102 (1995) 647 – 661. [7] K.F. Hanssen, H-J. Bangstad, O. Brinchmann-Hansen, K. Dahl-Jo¨rgensen, Blood glucose control and diabetic microvascular complications: long-term effects of nearnormoglycaemia, Diabet. Med. 9 (1992) 697 – 705. [8] The Diabetes Control and Complications Trial Research Group. The effect of intensive diabetes treatment on the progression of diabetic retinopathy in insulin-dependent diabetes mellitus. Arch. Ophthalmol. 113 (1995) 36 – 51. [9] R. Klein, M. Palta, C. Allen, G. Shen, D.P. Han, D.J. D’Alessio, Incidence of retinopathy and associated risk factors from time of diagnosis of insulin-dependent diabetes, Arch. Ophthalmol. 115 (1997) 351 – 356. [10] Early Treatment Diabetic Retinopathy Study Research Group. Photocoagulation for diabetic macular edema: ETDRS Report No. 1. Arch. Ophthalmol. 103 (1985) 1796 – 1806. [11] G. Blohme, L. Nystro¨m, H.J. Arnqvist, F. Lithner, B. Littorin, P.O. Olsson, B. Schersten, L. Wibell, J. O8 stman, Male predominance of Type 1 (insulin dependent) diabetes mellitus in young adults: results from a 5-year prospective nationwide study of the 15 – 34-year age group in Sweden, Diabetologia 35 (1992) 56 – 62.
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23 [12] C.E. Kullberg, H.J. Arnqvist, Good blood glucose control characterizes patients without retinopathy after long diabetes duration, Diabet. Med. 12 (1995) 314–320. [13] R. Klein, B.E.K. Klein, S.E. Moss, K.J. Cruickshanks, The Wisconsin Epidemiologic Study of Diabetic Retinopathy XV, Arch. Ophthalmol. 112 (1994) 1217– 1228. [14] A.S. Krolewski, J.H. Warram, L.I. Rand, A.R. Christlieb, E.J. Busick, C.R. Kahn, Risk of proliferative diabetic retinopathy in juvenile-onset type 1 diabetes: a 40-yr follow-up study, Diabet. Care 9 (1986) 443–452. [15] G. D’Annunzio, F. Malvezzi, L. Vitali, C. Barone, R. Giacchero, C. Klersy, S. Zanette, R. Lorini, A 3–19-year follow-up study on diabetic retinopathy an patients diagnosed in childhood and treated with conventional therapy, Diabet. Med. 14 (1997) 951–958. [16] D.E. Goldstein, K.J. Blinder, C.H. Ide, R.J Wison, H.M. Wiedmyer, R.R Little, J.D. England, M. Eddy, J.E. Hewett, S.K. Anderson, Glycemic control and development of retionpathy in youth-onset insulin-dependent diabetes mellitus, Ophthalmology 100 (1993) 1125–1132. [17] M. Bojestig, H.J. Arnqvist, G. Hermansson, B.E. Karlberg, J. Ludvigsson, Declining incidence of nephropathy in insulin-dependent diabetes mellitus, N. Engl. J. Med. 330 (1994) 15 – 18. [18] T. Danne, B. Weber, R. Hartman, I. Enders, W. Burger, G. Hovener, Long-term glycemic control has a nonlinear association to the frequency of background retinopathy in adolescents with diabetes, Diabet. Care 12 (1994) 1390– 1396. [19] E. Agardh, C-D. Agardh, O. Torffvit, A 5-year follow-up study on the incidence of retinopathy in type 1 diabetes
.
[20]
[21]
[22]
[23]
[24]
[25]
[26]
23
mellitus in relation to medical risk indicators, J. Intern. Med. 235 (1994) 353 – 358. O. Kalter-Leibovici, L. Leibovici, N. Loya, I. Kremer, R. Axer-Siegel, M. Karp, Z. Laron, The development and progression of diabetic retinopathy in type 1 diabetic patients: a cohort study, Diabet. Med. 14 (1997) 858 – 866. The Diabetes Control and Complications Trial Research Group. The relationship of glycemic exposure (HbA1c) to the risk of development and progression of retinopathy in the diabetes control and complications trial. Diabetes 44 (1995) 968 – 983. J. Stephenson, J.H. Fuller, Microvascular and acute complications in IDDM patients: the EURODIAB IDDM complications study, Diabetologia 37 (1994) 278 – 285. E. Agardh, O. Torffvit, C-D. Agardh, The prevalence of retinopathy and associated risk factors in type 1 (insulindependent) diabetes mellitus, J. Intern. Med. 226 (1989) 47 – 52. R. Klein, B.E.K. Klein, S.E. Moss, M.D. Davis, D.L. DeMets, The Wisconsin epidemiologic study of diabetic retinopathy II. Prevalence and risk of diabetic retinopathy when age at diagnosis is less than 30 years, Arch. Ophthalmol. 102 (1984) 520 – 526. A. Teuscher, H. Schnell, P.W.F. Wilson, Incidence of diabetic retinopathy and relationship to baseline plasma glucose and blood pressure, Diabet. Care. 11 (1988) 246 – 251. R. Klein, B.E.K. Klein, S.E. Moss, M.D. Davis, D.L. DeMets, Is blood pressure a predictor of the incidence or progression of diabetic retinopathy?, Arch. Intern. Med. 149 (1989) 2427 – 2432.
The temporal development of retinopathy and nephropathy in type 1 diabetes mellitus during 15 years diabetes duration M. Lo¨vestam-Adrian a, E. Agardh b, C.D. Agardh b,* b
a Department of Ophthalmology, Uni6ersity Hospital, S-221 85 Lund, Sweden Department of Endocrinology, Uni6ersity Hospital MAS, S- 205 02 Malmo¨, Sweden
Received 10 December 1998; received in revised form 15 March 1999; accepted 6 April 1999
Abstract In this prospective study, the 10- and 15-year incidence and course of retinopathy were examined in relation to medical risk indicators from 3.1 91.9 (mean9SD) years diabetes duration and onwards in 29 adult type 1 diabetic patients treated under routine care. A total of 28 patients were followed for 10 years and 20 patients for 15 years diabetes duration. After 10 years diabetes duration, 11 patients had developed any retinopathy (ten patients background retinopathy and one patient clinically-significant macular oedema). After 15 years diabetes duration, 16 patients had developed any retinopathy (12 patients developed background retinopathy and four patients developed potentially sight-threatening retinopathy, i.e. clinically significant macular oedema (n =2) or severe non-proliferative retinopathy (n =2)). None of the patients developed proliferative retinopathy. No differences were seen in mean HbA1c values between patients without any retinopathy and patients with background retinopathy, whereas patients who developed potentially sight-threatening retinopathy had higher mean HbA1c levels than patients without any retinopathy. Patients who developed potentially sight-threatening retinopathy had higher levels of mean HbA1c, both after 10 (9.79 1.6 vs 6.99 1.5%; P B0.05) and 15 years diabetes duration (9.39 1.2 vs 7.19 1.3%; PB0.05), compared to patients without any retinopathy. They also had higher levels of mean HbA1c than patients with background retinopathy after 15 years diabetes duration (9.3 9 1.2 vs 7.7 9 1.1%; PB 0.05). There were no differences in blood pressure levels between patients who developed retinopathy and those who did not. Only two patients developed clinical signs of nephropathy (urinary albumin 320 – 1590 mg/l) after 12 and 13 years diabetes duration, respectively. At those time points, both patients had already developed background retinopathy since 2 years. In conclusion, the present study shows that the incidence of retinopathy is associated with the duration of diabetes and that there is a strong association between the degree of metabolic control and development of potentially sight-threatening retinopathy. The study also indicates that the development of retinopathy does not seem to be associated with hypertension or clinical signs of nephropathy. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Retinopathy; Nephropathy; Metabolic control; Duration; Routine care; Diabetes mellitus
* Corresponding author. Tel.: + 46-40-331016; fax: + 46-40-337023. 0168-8227/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 9 9 ) 0 0 0 5 5 - 8
16
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
1. Introduction It is a well-known fact that type 1 diabetes mellitus is accompanied by micro- and macrovascular complications. Studies have shown an association between metabolic control and the incidence and progression of retinopathy and nephropathy [1–4]. In humans, intensified insulin treatment not only delays the onset of diabetic retinopathy but also retards the progression of both retinopathy and nephropathy [5,6], although a transient deterioration of retinopathy may occur during intensified glycaemic control [7,8]. However, there are few prospective studies of patients treated under routine care from near the onset of diabetes mellitus. In a recent study, in which patients were followed from the onset of diabetes, a possible reduction in the frequency of retinopathy was suggested in patients with strict metabolic control from the diagnosis of diabetes [9]. In the present study we were able to follow most of the patients for 15 years from near the onset of diabetes. The yearly mean levels of HbA1c that could be achieved under routine care were correlated to the incidence and course of retinopathy. We also studied a possible association between blood pressure levels, nephropathy and the development of retinopathy.
2. Materials and methods
2.1. Patients Adult patients (29) with type 1 diabetes mellitus participated in this prospective study which started in 1982. The patients have regularly attended the out-patient clinic at the Department of Internal Medicine, University Hospital, Lund, 4.3 92.3 (mean9 SD) times per year. At study entry, the mean duration of diabetes was 3.1 9 1.9 years. Of the patients, 26 received insulin treatment at the time of diagnosis and three patients received insulin treatment within 18 months of diagnosis. Of the patients, 28 were followed for 10 years and 20 patients for 15 years diabetes duration. During the observation period, two patients died, one from pneumonia after five years of
diabetes and the other from ketoacidosis after 13 years of diabetes. Among the seven patients who were followed for 10 years only, five patients moved from the area, one patient did not want to participate and one patient was recruited into the study 2 years later than the others. Age, age at onset, body mass index, and insulin dosage were registered. The degree of metabolic control was measured with mean glycosylated haemoglobin (HbA1c) values 3.2 9 2.0 times per year. Clinical signs of diabetic nephropathy were examined with albumin-creatinine clearance ratio, urinary albumin and serum creatinine concentrations. Normoalbuminuria was defined as a urinary albumin concentration B30 mg/l, and clinical nephropathy \ 299 mg/l. Glomerular filtration rate (GFR) was measured after 5, 10 and 15 years of diabetes duration. The blood pressure levels were measured 2.89 1.6 times per year in the supine position using a mercury sphygmomanometer. Diastolic blood pressure was taken at Korotkoff phase V.
2.2. Classification of retinopathy The patients were regularly examined at the Department of Ophthalmology, every other year in patients without retinopathy, and every 3–12 months, depending on the degree of retinopathy, in the others. The retinal examination was performed by fundus photography through a dilated pupil with three fields per eye (45°) nasal, temporal and central, including stereo photos of the macular region (Nikon NFC 50). The degree of retinopathy was based on fundus photo-grading of the eye that showed the most advanced retinopathy. In addition, fluorescein angiography was performed after five, 10 and 15 years diabetes duration. The different degrees of retinopathy were classified as no retinopathy, background retinopathy and potentially sight-threatening retinopathy. Potentially sight-threatening retinopathy was defined as clinically significant macular oedema or severe non-proliferative retinopathy, using ETDRS definitions [10] or proliferative retinopathy.
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
2.3. Analytical techniques HbA1c levels were analysed by ion-exchange chromatography using commercially available microcolumns (Bio-Rad, Richmond, CA) or by fast liquid chromatography (Kontron Instruments, Milan, Italy). Normal value for both methods is B 5.3%. Urine was collected in polystyrene tubes and stored at 4°C for less than 4 days before analysis. Urinary albumin concentration was measured with an electroimmunoassay using human albumin (Kabi Vitrum, Stockholm, Sweden) (detection limit 12.5 mg/l) or by turbidimetry with an automated analyser (Cobras Mira, Roche), antibodies (rabbit anti-human albumin) and technique as described by Dakopatts, Copenhagen (detection limit 5 mg/l). Urine- and serum-creatinine levels were analysed by a kinetic Jaffe reaction or by an enzymatic method (creatinine-hydrolase; EKTA Chem-analyser, Instrument Kodak). The second method results in about 8% higher values than the first one. This was adjusted for in order to create comparable values. Normal values were for women 48 – 100 mmol/l and for men 55 – 116 mmol/l. The urinary albumin creatinine clearance ratio was calculated as the ratio between the urine albumin clearance and the urine creatinine clearance; normal value B 0.01× 10 − 3. Plasma albumin was analysed by a bromcresolpurpur-method or by immuniturbidimetry. Glomerular filtration rate (GFR) was determined with a plasma clearance technique using the non-ionic X-ray contrast medium iohexol, as a marker. The method has a total imprecision of about 11% (CV), essentially all of which is accounted for by intra-individual biological variation.
2.4. Statistical methods The mean of 2– 4 observations per year for HbA1c, systolic blood pressure, diastolic blood pressure and urinary albumin was calculated. The statistical evaluation of the results was based on a repeated measure analyses-of-variance technique (ANOVA, SPSS). Values are given as mean9 SD. Urinary albumin concentration and urinary albumin creatinine clearance ratio are given as median
17
and range. Fisher’s exact test was used for categorical variables, the Student’s two-tailed t-test was used for equal standard deviations and Welch’s approximate t-test for unequal standard deviations. Correlations between variables were tested with Pearsons’ test.
3. Results
3.1. Patient characteristics At entry in 1982, the mean diabetes duration was 3.19 1.9 years. At this time point there were no differences regarding patient characteristics, except for HbA1c levels, which were higher among patients who later developed potentially sightthreatening retinopathy than those who did not (Table 1). None of the patients had anti-hypertensive medication and during the study, three patients received such treatment after 10 (n= 2) and 12 (n= 1) years diabetes duration. None of the five patients who moved, or the patient who did not want to participate any longer, developed any retinopathy during the first 10-year follow-up period. These patients did not differ at entry into the study from the other patients without retinopathy regarding age at onset (24.2911.3 years), HbA1c (10.49 1.5%), systolic blood pressure (1319 14 mmHg), diastolic blood pressure (73911 mmHg), S-Creatinine (859 11 mmol/l), insulin dosage (0.369 0.22 U kg-1) (mean9 SD) or urinary albumin levels 13 mg/l (5–13 mg/l; median, range).
3.2. Retinopathy grading At entry, no subject showed any signs of diabetic retinopathy. The incidence of any type of retinopathy after 10 years diabetes duration was 39% (eight males and three females) and after 15 years duration 80% (ten males and six females) (Table 1). The incidence of potentially sightthreatening retinopathy after 15 years diabetes duration was 20% (three males and one female). Three of the four patients with potentially sightthreatening retinopathy after 15 years diabetes duration had background retinopathy after 10
18
Sex (male/female) Age at onset of diabetes HbA1c (%) SBP (mmHg) DBP (mmHg) Albumin creatinine clearance ratio (×10−3) Urinary Albumin (mg/l) Serum creatinine (mmol/l) Insulin (U/kg) Body mass index (kg/m2) Antihypertensive Rx (n) a
No retinopathy at follow up
Background retinopathy at follow-up
Potentially sight-threatening retinopathy at follow-up
Entry (n =20)
Entry
Entry
12/8 28.0 95.2 10.3 9 2.8 130 99 80 97 B0.01 13 (5–13) 68 9 12 0.64 9 0.26 23.0 91.9
Follow-up (n = 4) 2/2 7.1 91.3 1339 15 75 99 B0.01 13 (5–27) 67 914 0.65 9 0.16 24.8 93.2 1
Follow-up (n=12)
Follow-up (n=4)
7/5 30.8 95.2 10.4 9 2.2 124 9 6 76 96 B0.01 13 (5–13) 74 9 12 0.52 90.24 22.2 9 2.0
7.7 91.1+ 12597 769 8 B0.01 12.5 (5–15) 719 12 0.63 9 0.15 24.49 2.4 1
3/1 28.3 9 6.4 11.992.0 1309 27 799 4 B0.01 13 (5-13) 66912 0.4190.08 22.992.8
9.3 91.2* 123 9 11 739 7 B0.01 (B0.01–0.04) (7–1590) 77 9 22 0.599 0.1 23.69 1.8 1
Values are given as mean(SD and for urinary albumin as median and range. SBP+systolic blood pressure; DBP+diastolic blood pressure. * PB0.05; patients without retinopathy vs. potentially sight-threatening retinopathy. + PB0.05; patients with background vs. potentially sight-threatening retinopathy.
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
Table 1 Patient characteristics at entry in relation to degree of retinopathy after 15 years diabetes durationa
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
years diabetes duration. The fourth patient developed background retinopathy after 12 years duration and potentially sight-threatening retinopathy after 15 years duration. The only patient who developed potentially sight-threatening retinopathy, i.e. clinically significant macular oedema, after 10 years duration, had an onset of diabetes in 1984, and has not yet been followed for 15 years. Among the other patients with potentially sightthreatening retinopathy, two patients developed clinically significant macular oedema and two patients severe non-proliferative retinopathy. None developed proliferative retinopathy. Patients with clinically significant macular oedema were treated with focal laser photocoagulation, and one of the patients with severe non-proliferative retinopathy received panretinal photocoagulation. In two of the five patients who moved after 10 years duration, fundus photos were available after 15 years duration and showed no signs of retinopathy. The fluorescein angiography confirmed the changes seen on fundus photos. The patients with clinically significant macular oedema showed central leakage in the mid-and late phase angiograms, and those with severe non-proliferative retinopathy showed numerous zones of 2 – 4 optic disc diameter areas of non-perfusion. In five patients, however, microaneurysms could be detected 2 – 3 years before they were visible on fundus photos.
3.3. Metabolic control Patients who developed potentially sight-threatening retinopathy had higher levels of mean HbA1c both after 10 (9.79 1.6 vs 6.991.5%; P B 0.05) and 15 years diabetes duration (9.39 1.2 vs 7.191.3%; P B 0.05) compared to patients without any retinopathy. They also had higher levels of mean HbA1c than patients with background retinopathy after the 15 year follow-up period (9.391.2 vs 7.791.1%; P B0.05). No differences were seen between patients with background retinopathy and patients without any retinopathy regarding HbA1c levels. On a group basis there was an improved metabolic control over time (Fig. 1). The improvement was significant between five and 10
19
years diabetes duration in patients with background retinopathy (9.39 1.3 vs.7.6 9 1.0%; PB 0.001) as well as in patients without any retinopathy (9.49 3.0 vs.6.9 9 1.5%; PB 0.05). Those patients who developed potentially sightthreatening retinopathy also had a tendency towards improved metabol control over time, however not significant.
3.4. Blood pressure and albuminuria There were no differences regarding systolic and diastolic blood pressure levels between patients who developed any degree of retinopathy and those who did not. Blood pressure levels were stable over the study period (Fig. 2). None of the patients received anti-hypertensive treatment at entry. After 15 years diabetes duration, three patients were treated with ACE-inhibitors. In two of these patients, the time for institution of anti-hypertensive treatment coincided with the deterioration of retinopathy. The mean level of systolic and diastolic blood pressure levels before institution of anti-hypertensive medication was 1529 10 and 9095 mmHg, respectively. Two patients developed clinical signs of nephropathy with urinary albumin levels of 320–1590 mg/l and an albumin creatinine-clearance ratio of 0.049 10 − 3 –0.089 10 − 3 after 12 and 13 years diabetes duration, respectively. At those time points, both patients had already developed background retinopathy since 2 years. One of the patients with diabetic nephropathy died from ketoacidosis 1 year after the development of nephropathy and still had background retinopathy at the time for death. In the other patient who developed diabetic nephropathy, retinopathy deteriorated to potentially sight-threatening retinopathy after another 2 years. Both patients had mean HbA1c levels \ 10% during their first 10 years of diabetes duration. The mean glomerular filtration rate (GFR) for the whole group was 1309 20 ml/min × 1.73 m2, 1219 21 ml/min × 1.73 m2 and 1079 22 ml/ min×1.73 m2 after 5, 10 and 15 years diabetes duration, respectively. There were no differences in GFR according to degree of retinopathy. The two patients who developed nephropathy did not
20
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
differ in GFR. The GFR values were correlated to the HbA1c levels (r =0.49, P B 0.01).
4. Discussion Long-term prospective follow-up studies are necessary in order to evaluate the effect of glycaemic control from the onset of diabetes mellitus on the development of retinopathy and nephropathy. The patients in this study all attended the out-patient clinic at the Department of Internal Medicine, University Hospital, Lund. This clinic takes care of all type 1 diabetic patients in the district and the patients in our study were treated under routine care. The sex-distribution in our patient group, 60% men and 40% women, indicates that diabetes mellitus is slightly more common among men. This is in agreement with figures for the whole country describing an incidence of 14.6/100 000 men and
8.8/100 000 women in the age group 25–29 years for type 1 diabetes mellitus [11]. However, the sex distribution according to the different levels of retinopathy, although based on small numbers, with 58% of patients with background retinopathy and 75% of patients with potentially sightthreatening retinopathy being men, shows a predominance for men in this study in agreement with another study [12], whereas the frequency of men and women was similar in others [13,14]. In the present study, the 10-year incidence of any retinopathy was 39% and the 15-year incidence was 80%. These incidence rates are in accordance with some studies [13,15] but lower than in others [16]. The incidence of potentially sightthreatening retinopathy after 15 years diabetes duration was 20% and none had developed proliferative retinopathy. Since we know, that two of the patients who moved did not have any retinopathy after 15 years diabetes duration and none of the other three patients who moved had
Fig. 1. HbA1c levels in relation to diabetes duration (mean9 SD). Circles () represent patients without retinopathy, triangles () patients with background retinopathy and filled squares ( ) patients with potentially sight-threatening retinopathy.*PB0.05 represents differences in HbA1c levels between 5 and 10 years diabetes duration. In addition there were differences in HbA1c levels between patients with potentially sight-threatening retinopathy and patients without any reinopathy after 10 and 15 years diabetes duration and between patients with sigh-threatening retinopathy and patients with background retinopathy after 15 years duration.
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
21
Fig. 2. Blood pressure levels in relation to diabetes duration (mean 9SD). Circles () represent patients without retinopathy, triangles () patients with background retinopathy and filled squares ( ) patients with potentially sight-threatening retinopathy.
any retinopathy after 10 years diabetes duration, the incidence figures after 15 years diabetes duration would have been even lower with these patients included. Although we found the development of potentially sight-threatening retinopathy but not background retinopathy to be associated with higher levels of HbA1c after 10 and 15 years diabetes duration, we found no statistically significant differences in HbA1c levels between patients who developed background retinopathy and patients who did not develop any retinopathy. However, there was a clear tendency towards higher levels in background retinopathy patients rendering it possible that this difference might well have reached statistical significance in a larger study. On a group basis, the degree of metabolic control was significantly improved in patients without any retinopathy and in patients who developed background retinopathy during the last 10 years, a phenomenon that has been previously described [17]. Among those who developed potentially sight-threatening retinopathy, however, HbA1c levels did not change significantly but showed a
tendency towards lower levels over time. In addition these patients seem to have been under-insulinised at entry compared to the others. The connection between hyperglycaemia and the development of retinopathy in type 1 diabetes is well documented [2,18,19]. However, the importance of good glycaemic control from the onset of disease for later development of retinopathy is under debate. The significance of excellent glycaemic control from the diagnosis has been emphasized [9], whereas a reduction in both the development and progression of retinopathy after intensified therapy has been demonstrated even later in the course of diabetes [4,6]. In a study from Israel, poor metabolic control during the last 3 years better predicted the risk for proliferative retinopathy than the mean of all HbA1 values [20]. In the report from the DCCT however, the degree of metabolic control, both before and during the study period, was found to influence the progression of retinopathy and pointed at the ‘total glycemic exposure’ to be the dominant risk factor for developing retinopathy. Anyhow, the benefits of improved metabolic control from early on of the disease [21] is of undisputable value.
22
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23
Duration of diabetes is also known to be an important risk factor for the development of retinopathy [14,22,23]. However, not only the duration but also the age at onset of diabetes has been shown to play a role in the development of retinopathy, where age at onset of diabetes before or during puberty was found to be a significant risk factor for retinopathy [15,16,20], especially proliferative retinopathy [23]. In our study, the mean age at onset of diabetes was rather high, 26 years, with the youngest age at onset of 15 years. This could be one explanation for the lower incidence rates in our study and the fact that no patient developed proliferative retinopathy. Furthermore, the lack of other known risk factors such as proteinuria or hypertension [2,22,24,25] may have contributed to the lower incidence figures. We did not find any connection between retinopathy and hypertension, but in two of the four patients in whom the retinopathy deteriorated to potentially sight-threatening retinopathy, anti-hypertensive treatment was instituted during the follow-up period. Thus, it is possible that there could be a connection between hypertension and retinopathy in these patients in accordance with other studies [22,26]. Though the number of patients is small, it should be noted that the incidence of nephropathy was low and this is in agreement with another study, in which a declining incidence of nephropathy for type 1 diabetic patients has been shown over the last years [17]. In conclusion, the development of any type of retinopathy increased with the duration of diabetes and the development of severe but not background retinopathy was associated with worse metabolic control after 10 and 15 years diabetes duration. The metabolic control was improved during the last 10 years which shows that it is possible to achieve improved metabolic control over time, also under routine care.
Acknowledgements This study was supported by funds the Swedish Diabetes Federation, the Crafoord Foundation, the Crown Princess Margareta’s Committee for
the Blind, the Stig and Alme´rs Foundation, the Novo Nordisk Research Foundation and the Swedish Medical Research Council (Grant k9819x-12662-01A).
References [1] R. Klein, B.E.K. Klein, S.E. Moss, K.J. Cruickshanks, Relationship of hyperglycemia to the long-term incidence and progression of diabetic retinopathy, Arch. Intern. Med. 154 (1994) 2169 – 2178. [2] C-D. Agardh, E. Agardh, O. Torffvit, The association between retinopathy, nephropathy, cardiovascular disease and long-term metabolic control in type 1 diabetes mellitus: a 5 year follow-up study of 442 adult patients in routine care, Diabetes. Res. Clin. Pract. 35 (1997) 113 – 121. [3] C.E. Mogensen, Prediction of clinical diabetic nephropathy in IDDM patients. Alternatives to microalbuminuria?, Diabetes 39 (1990) 761 – 767. [4] The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N. Engl. J. Med. 329 (1993) 977 – 986. [5] P. Reichard, B-Y. Nilsson, U. Rosenqvist, The effect of long-term intensified insulin treatment on the development of microvascular complications of diabetes mellitus, N. Engl. J. Med. 329 (1993) 304 – 309. [6] The Diabetes Control and Complications Trial Research Group. Progression of retinopathy with intensive versus conventional treatment in the diabetes control and complications trial. Ophthalmology 102 (1995) 647 – 661. [7] K.F. Hanssen, H-J. Bangstad, O. Brinchmann-Hansen, K. Dahl-Jo¨rgensen, Blood glucose control and diabetic microvascular complications: long-term effects of nearnormoglycaemia, Diabet. Med. 9 (1992) 697 – 705. [8] The Diabetes Control and Complications Trial Research Group. The effect of intensive diabetes treatment on the progression of diabetic retinopathy in insulin-dependent diabetes mellitus. Arch. Ophthalmol. 113 (1995) 36 – 51. [9] R. Klein, M. Palta, C. Allen, G. Shen, D.P. Han, D.J. D’Alessio, Incidence of retinopathy and associated risk factors from time of diagnosis of insulin-dependent diabetes, Arch. Ophthalmol. 115 (1997) 351 – 356. [10] Early Treatment Diabetic Retinopathy Study Research Group. Photocoagulation for diabetic macular edema: ETDRS Report No. 1. Arch. Ophthalmol. 103 (1985) 1796 – 1806. [11] G. Blohme, L. Nystro¨m, H.J. Arnqvist, F. Lithner, B. Littorin, P.O. Olsson, B. Schersten, L. Wibell, J. O8 stman, Male predominance of Type 1 (insulin dependent) diabetes mellitus in young adults: results from a 5-year prospective nationwide study of the 15 – 34-year age group in Sweden, Diabetologia 35 (1992) 56 – 62.
M. Lo¨6estam-Adrian et al. / Diabetes Research and Clinical Practice 45 (1999) 15–23 [12] C.E. Kullberg, H.J. Arnqvist, Good blood glucose control characterizes patients without retinopathy after long diabetes duration, Diabet. Med. 12 (1995) 314–320. [13] R. Klein, B.E.K. Klein, S.E. Moss, K.J. Cruickshanks, The Wisconsin Epidemiologic Study of Diabetic Retinopathy XV, Arch. Ophthalmol. 112 (1994) 1217– 1228. [14] A.S. Krolewski, J.H. Warram, L.I. Rand, A.R. Christlieb, E.J. Busick, C.R. Kahn, Risk of proliferative diabetic retinopathy in juvenile-onset type 1 diabetes: a 40-yr follow-up study, Diabet. Care 9 (1986) 443–452. [15] G. D’Annunzio, F. Malvezzi, L. Vitali, C. Barone, R. Giacchero, C. Klersy, S. Zanette, R. Lorini, A 3–19-year follow-up study on diabetic retinopathy an patients diagnosed in childhood and treated with conventional therapy, Diabet. Med. 14 (1997) 951–958. [16] D.E. Goldstein, K.J. Blinder, C.H. Ide, R.J Wison, H.M. Wiedmyer, R.R Little, J.D. England, M. Eddy, J.E. Hewett, S.K. Anderson, Glycemic control and development of retionpathy in youth-onset insulin-dependent diabetes mellitus, Ophthalmology 100 (1993) 1125–1132. [17] M. Bojestig, H.J. Arnqvist, G. Hermansson, B.E. Karlberg, J. Ludvigsson, Declining incidence of nephropathy in insulin-dependent diabetes mellitus, N. Engl. J. Med. 330 (1994) 15 – 18. [18] T. Danne, B. Weber, R. Hartman, I. Enders, W. Burger, G. Hovener, Long-term glycemic control has a nonlinear association to the frequency of background retinopathy in adolescents with diabetes, Diabet. Care 12 (1994) 1390– 1396. [19] E. Agardh, C-D. Agardh, O. Torffvit, A 5-year follow-up study on the incidence of retinopathy in type 1 diabetes
.
[20]
[21]
[22]
[23]
[24]
[25]
[26]
23
mellitus in relation to medical risk indicators, J. Intern. Med. 235 (1994) 353 – 358. O. Kalter-Leibovici, L. Leibovici, N. Loya, I. Kremer, R. Axer-Siegel, M. Karp, Z. Laron, The development and progression of diabetic retinopathy in type 1 diabetic patients: a cohort study, Diabet. Med. 14 (1997) 858 – 866. The Diabetes Control and Complications Trial Research Group. The relationship of glycemic exposure (HbA1c) to the risk of development and progression of retinopathy in the diabetes control and complications trial. Diabetes 44 (1995) 968 – 983. J. Stephenson, J.H. Fuller, Microvascular and acute complications in IDDM patients: the EURODIAB IDDM complications study, Diabetologia 37 (1994) 278 – 285. E. Agardh, O. Torffvit, C-D. Agardh, The prevalence of retinopathy and associated risk factors in type 1 (insulindependent) diabetes mellitus, J. Intern. Med. 226 (1989) 47 – 52. R. Klein, B.E.K. Klein, S.E. Moss, M.D. Davis, D.L. DeMets, The Wisconsin epidemiologic study of diabetic retinopathy II. Prevalence and risk of diabetic retinopathy when age at diagnosis is less than 30 years, Arch. Ophthalmol. 102 (1984) 520 – 526. A. Teuscher, H. Schnell, P.W.F. Wilson, Incidence of diabetic retinopathy and relationship to baseline plasma glucose and blood pressure, Diabet. Care. 11 (1988) 246 – 251. R. Klein, B.E.K. Klein, S.E. Moss, M.D. Davis, D.L. DeMets, Is blood pressure a predictor of the incidence or progression of diabetic retinopathy?, Arch. Intern. Med. 149 (1989) 2427 – 2432.