63 (1985) 530-534 ACTA OPHTHALMOLOGICA

Proliferative diabetic retinopathy: at risk patients identified by early detection of microalbuminuria

J. Vigstrupl and C. E. MogensenZ

Department of Ophthalmology1 (Head: N. Ehlers) and Second University Clinic of Internal Medicine2 (Head: C. E. Mogensen),

Arhus Kommunehospital, University of Aarhus, Arhus. Denmark

Abstract. The aim of the study was to clarify if these diabetics, with the greatest risk for development of proliferative retinopathy, could be identified by early measurement of urinary albumin excretion (UAE), using sensitive methods. In addition, the possible role of blood pressure was assessed. All male diabetics originally exa- mined in 1969-1976 were reexamined by measuring U A E and blood pressure together with an ordinary ophthalmoscopic examination. Patients fulfilling the fol- lowing criteria were incorporated in the study: age at diagnosis of diabetes < 20 years; initial duration of diabetes 7-20 years; follow-up period > 7 years, and no clinical proteinuria. 44 patients were identified, all but one accepted reexamination. 29 patients initially had U A E < 15 Fg/min, one of these developed proliferative retinopathy. 14 patients initially had UAE 2 15 pg/min, 9 of these developed poliferative retinopathy. Blood pres- sure was elevated both initially and at follow-up in diabetics showing progression to proliferative retin- opathy. It is concluded that even a slightly elevated UAE ‘micro-albuminuria’, is a strong predictor with respect to development of proliferative retinopathy.

Key words: urinary albumin excretion - microalbuminu- ria - proliferative diabetic retinopathy -blood pressure.

Although diabetes is responsible for about 10% of all new cases of blindness in the Western world (Feldman et al. 1982; Kahn & Hiller 1974), and about 15% of all patients with insulin dependent diabetes develop proliferative retinopathy (Nielsen 1984), about 20% of diabetics remain free of retinopathy. Furthermore, in many patients retinal lesions do not impair quality of life (Chazan et al. 1970).

530

It would be of considerable clinical interest if, at an early stage of the disease, it could be determined whether a patient was a risk for developing proli- ferative retinopathy or not. It has been shown that elevated urinary albumin excretion (UAE) is a strong predictor for clinical nephropathy in insulin dependent diabetics (Parving et al. 1982; Mathie- sen et al. 1984, Mogensen & Christensen 1984; Viberti et al. 1982). The aim of this study was to clarify if proliferative retinopathy in insulin depen- dent diabetics is predictable by UAE.

Since it has been demonstrated that antihyper- tensive treatment considerably retards progression of nephropathy (Mogensen 1982), it seems reason- able to examine the possible role of associated hypertension in patients developing proliferative diabetic retinopathy.

Patients and Method

Data on all young male patients with insulin de- pendent diabetes studied during the years 1969- 1976 were evaluated in 1983, and patients fulfilling the following criteria were reexamined: 1) age at diagnosis of diabetes 5 20 years; 2) diabetes dura- tion within 7 to 19 years at the time of the first examination; 3) the follow-up renal function test and the ophthalmoscopic examination should take place at least 7 years after the initial examination; 4) no clinical proteinuria present initially (< 0.5 g/24 h at repeated examinations). Forty-four pa-

tients were identified, and all but one accepted reexamination. At the follow-up examination, in addition to ophthalmoscopic examination, urinary albumin excretion at baseline was measured, to- gether with several other kidney parameters. The results of these examinations have been recently described elsewhere (Mogensen & Christensen 1984).

The initial studies of the natural history of dia- betic renal involvement were carried out, for prac- tical reasons, primarily in male subjects, and there- fore this study include only male diabetics.

The 43 patients underwent fundoscopy through a dilated pupil. Patients were subdivided into 5 groups: 0) retina without signs of diabetic retino-

Initial duration

of diabetes

Follow-~p UAE Fg/min period

years years initial follow-up

pathy; I) the number of microaneurisms in each eye 5 3; 11) > 3 microaneurisms or haemorrhages but no exudates; 111) exudates with or without haemorrhages; and IV) proliferative retinopathy, defined as proliferations of new vessels or glial tissue formation.

At the time of the reexamination, the ophthal- mologist was unaware of the results of UAE mea- surements. The results of the initial ophthalmo- scopic examinations were obtained from the pa- tient's clinical records, and a subdivision as above was done. Urinary albumin excretion was mea- sured at rest in all patients, albumin being mea- sured by radioimmunoassay (Miles et al. 1970). Progression within the microalbuminuria range

Retinopathy

group initial follow-up

Age at diagnosis

years

No progression 12.4 f 5.3 (n = 27)

Progression 11.3 f 4.3 to microalbuminuria

(n = 4)

Progression to diabetic nephropathy or UAE > 150 Fg/min

Initial UAE 14.1 f 3.5 < 70 pg/min (n = 9)

Initial UAE 10.3 f 3.1 2 70 pg/min (n = 3)

Not re-examined 1 1 (n = 1)

12.7 f 3.6 10.5 f 2.6 7.3 f 4 . 7 5.9 f 3.3 0 I I1 I11 I V

12.8 f 3.5 10.3 f 3.2 5.9 f 2.9 41.1 f 17.4 0 I I1 111 I V

0 I I 1

10.8 f 2.1 9.6 f 2.7 26.6 f 8.2 2373 f 2488 I11 I V

16.3 f 0.8 12.0 f 3.4 79 f 15 3254 f 1279 0 I I1 I11 I V

10 13 12.6

23 12 2 5 2 7 0 3 0 0

2 0 1 1 1 0 0 2 0 1

3 1 1 0 2 0 3 2 0 6

0 0 0 0 1 0 1 0 1 3

Mean f SD is shown.

34' 53 1

Table2. Urinary albumin and development of proliferative

retinopathy. I I I

Initial

Initial urinary Non-P Proliferation albuminexcretion I Total I I

FO~IOW-UP

~~~~~ ~

< 15 pg/min 29 28 1* (3.4%) 2 15 pglmin 14 5 9(69.3%)

*Developed microalbuminuria at follow-up.

was defined as UAE increased by more than a factor 3 at the follow-up study, while development of diabetic nephropathy was defined as presence of clinical proteinuria or UAE > 150 pg/min at re- examination.

Blood pressure was measured auscultatorily af- ter 20 min at rest in a supine position.

For the 43 re-examined patients, age at diag- noses of diabetes averaged 12.3 years f 5 SD (range 1-20). Mean initial diabetes duration was 12.6 years k 3 SD (range 7- 19). Mean follow-up time was 10.4 years f 3 SD (range 7-14). In the subgroups almost identical data were obtained. For the one patient not accepting re-examination, similar data was noted. Clinical, renal, blood pressure as well as fundoscopic data of the subgroups are shown in Tables 1 and 3.

Results

In Table 1 the patients were subdivided according to the results of follow-up data of UAE with respect to progression. Those exhibiting progression to clinical nephropathy were subdivided according to initial UAE, also. In those subgroups (shown in Table 1) the patients were devided into the earlier mentioned groups 0-IV according to the ophthal- moscopic pictures.

Twenty-seven patients showed no progression in UAE, and none of these developed proliferative diabetic retinopathy, and as many as 12 had com- pletely normal fundus. Sixteen patients showed progression in UAE, of these 10 developed proli- ferative retinopathy. Of the 16 patients, 4 progres- sed from completely normal value of UAE to the microalbuminuria level (UAE > 15 pg/min, but no clinical proteinuria), and 1 of these 4 developed proliferative changes. 12 patients progressed to diabetic nephropathy or to UAE > 150 pg/min. Of these, 9 developed proliferative retinopathy.

Initially, 29 patients had UAE < 15 pg/min, and out of these, only 1 developed proliferative retin- opathy, i.e. 3.4% (the patient had developed micro- albuminuria at follow-up). 14 patients had UAE 1 15 pg/min, 9 of these developed proliferative changes, i.e. 69.3% as seen from Table 2. These results are further illustrated in Fig. 1, showing follow-up fundoscopy as a function of the initial

Progression in UAE but no proliferative retinopathy

132 f 12/86 f 7* 131 f 10/86 f 9*

(n = 6)

Progression in UAE and proliferative retinopathy

138 f 17/91 f 8** 156 f 17/105 f 12**

(n = 10)

*No significant increase for the whole group comparet to non-progressors. **Significantly increased for the whole group compared to non-progressors

(P < 0.0 1).

532

Follow-up fundoscopy . .

IV

111

11

1

0

.. .

. . ..

0 .

. .. ..

...... .

...... moo ...... ..

I

0 I 11 111 IV Initial fundoscopy.

initial UAE < 15 pglmin, . initial UAE 1 1 5 pg/min.

0 initial UAE 2 15 pg/min,but no progression in nephropathy.

Fig. I. Follow-up fundoscopy as a function of the initial retinal

examination.

Renal score for future

retinopathy*

examination. Patient are subdivided according to initial UAE (2 and < 15 pg/min). Of the 2 patients initially fallen into group I1,and at follow-up 0, 1 initially had mocroaneurisms and a small haemorr- hage, and the second had more than 3 microaneu- risms in one eye.

In diabetics developing proliferative retinopa- thy, both systolic and diastolic blood pressure ini- tially and at follow-up, were significantly elevated (P < 0.01) compared to non-progressors, while no significant increase could be shown when diabetics showing progression to microalbuminuria but without proliferative retinopathy were compared to non-progressors (Table 3). Thus it can be seen that patients with high UAE (2 15 pg/min) show a clear tendency toward development of prolifera- tive retinopathy, especially if renal changes pro- gresses and high blood pressure prevails. Non-

proliferation Total Proliferation

Discussion

Under the age of 30 years proliferative retinopathy starts to appear 10 to 15 years after the diagnosis of diabetes and develops ultimatively in about 25 % of cases with insulin dependent diabetes. In this study,

the mean duration of diabetes at the time of the initial fundoscopy and UAE measurement was 12.6 years, and only one showed proliferations. On the other hand, 10 out of 43 patients had proliferative changes at follow-up, i.e. 23%. Great effort has been made to prevent the development of proli- ferative retinopathy. Good diabetes control pro- bably reduces the frequency and delays the onset of retinopathy (Pirarts 1978), but still, it has been estimated that about 350 new cases of proliferative diabetic retinopathy occur per year among about 20000 insulin dependent diabetics in Denmark (Nielsen 1984). It is an open question how often patients should be controlled for development of retinopathy. The following measures have been recommended (Gregersen & Deckert 1984): a young person with no retinopathy and a short duration of diabetes should have an ophthalmo- scopic examination every 3.-5. year. With longer duration of diabetes (i.e. more than 9-10 years) or with presence of retinopathy, ophthalmoscopic examination each year or more often should be performed, depending on the nature of the diabe- tic changes. By following these guidelines, there should be a good possibility of detecting and per- haps preventing or at least postponing a large number of new cases of blindness caused by proli- ferative retinopathy, but it implies a rather exten- sive ophthalmoscopic service (Gregersen & Deckert 1984).

This study documents, however, that even a slightly raised UAE, far below the level of clinical proteinuria, is a strong predictor with respect to development of proliferative diabetic retinopathy, as well as nephropathy. It is also shown that diabe- tic patients with a normal or nearly normal UAE

Table 4. Renal factors predicting proliferative retinopathy.

3 7 2 5 2 5 1 4 1 6 6 0 0 25 25 0

*Each point counts one score: Initial UAE 2 15 pg/min; Progression in UAE; Initial blood pressure L 140/90.

533

have a very small risk of developing proliferative changes over the next 10 years. It therefore seems reasonable to use UAE to select those patients needing a closer control and those where control could be less frequent. Such a programme will depend on the availability of measurements of urinary albumin in the diabetic and/or the oph- thalmoscopic clinic.

It has been established that antihypertensive treatment can retard the decline in renal function in patients with clinical overt diabetic nephropathy (Mogensen 1982), but there is still no conclusive evidence that hypertension influences the develop- ment and severity of diabetic retinopathy (Chahal & Kohner 1983). This study does only partly con- tribute to the solution of this problem. However, as shown in Table 4, the diabetic has a high risk for developing proliferative retinopathy if initial UAE 115 pg/min, initial blood pressure 2140/90 and UAE is showing progression. At least this repre- sents a fact with diabetics.

insulin dependent young male

References

Chahal P S & Kohner E M (1985): The relationship between diabetic retinopathy and diabetic nephropa- thy. Diabetic Nephropath 6: 4-6.

Chazan B I, Balodimos M C, Ryan J R & Marble A (1970): Twenty-five to forty-five years of diabetes with and without vascular complications. Diabetologia 6:

Feldman J N, Hirsch S R, Beyer M M, James W A, L'Esperance F A & Friedman E A (1982): Prevalence of diabetic nephropathy at time of treatment for diabe- tic retinopathy. In: Friedman E A & L'Esperance F A (eds). Diabetic Renal-Retinal Syndrome, pp 9-22. Grune & Stratton, New York.

Gregersen E & Deckert T (1984): Forebyggelse og be- handling af retinopathia diabetica. Ugeskr Laeg 146:

565-569.

531-532.

Kahn H A & Hiller R (1974): Blindness caused by diabetic retinopathy. Am J Ophthalmol78: 58-67.

Mathiesen E R, Oxenball B, Johansen K, Svendsen P Aa & Deckert T (1984): Incipient nephropathy in Type 1 (insulin-dependent) diabetes. Diabetologia 26: 406- 4 10.

Miles D W, Mogensen C E & Gundersen H J G (1970): Radioimmunoassay for urinary albumin using a single antibody. Scan J Clin Lab Invest 26: 5- 11.

Mogensen C E (1982): Long-term antihypertensive treat- ment inhibiting- progression of diabetic nephropathy. Br Med 1285: 685-688.

Mogensen C E, Christensen C K & Vittinghus E (1983): The stages in diabetic renal disease. With emphasis on the stage of incipient nephropathy. Diabetes 32. Suppl

Mogensen C E & Christensen C K (1984): Predicting diabetic nephropathy in insulin dependent patients. N Engl J Med 3 11 : 89-93.

Nielsen N V (1984): Diabetisk 0jensygdom. Ugeskr Laeg

Parving H H, Oxenball B, Svendsen P Aa, Christiansen J S & Andersen A R ( 1982): Early detection of patients at risk of developing diabetic nephropathy. A longitu- dinal study of urinary albumin excretion. Acta Endo- crinol (Copenh) 100: 550-555.

Pirart J (1978): Diabetes mellitus and its degenerative complications: A prospective study of 4400 patients observed between 1947 and 1973. Diabetes Care 1: 252-263.

Viberti G C, Jarrett R J. MAhmud U, Hill R D, Argyro- poulos A & Kain H (1982): Microalbuminuria as a predictor of clinical nephropathy in insulin dependent diabetes mellitus. Lancet 1: 1430- 1432.

2: 64-68.

146: 533-538.

Received on March 14th, 1985.

Author's address:

J. Vigstrup, Department of Ophthalmology. Aalborg Sygehus Syd, DK-9000 Aalborg, Denmark.

534