Refractive status of type ii diabetes mellitus patients at Kenyatta National Hospital
Abstract
A hospital based cross sectional study of 96 type II diabetes mellitus patients attending
the diabetic medical clinic at Kenyatta National Hospital.
Aim: To determine the prevalence and pattern of refractive errors among African type II
diabetes mellitus patients and establish the relationship between baseline refractive status
and degree of diabetic retinopathy and indicators of glycaemic control.
Method: The study was carried out in the month of November, 2005. The statistically
predetermined sample size was 94 patients. The first 10 of the patients seen on each day
of the diabetic medical clinic were included in the study. These patients were randomly
booked at the diabetic medical clinic and had no prior knowledge of the study, hence no
bias in case selection. The actual level of metabolic control was evaluated from
measurement of HBAI C and FBS. The patients had full ocular examination including
OR and SLE. Two eyes from 2 patients were excluded due to dense cataracts. After these
eyes were excluded, data from both eyes were reported (190 eyes).
Results: The total number of subjects examined was 96. There were 58 females and 38
males. The mean age was 52 (range 28-76) years and the median was 53 years. The
prevalence of myopia was 39.5% (75/190 eyes) and that of hypermetropia was 19.0%
(36/190 eyes). To estimate the short term fluctuation in refraction caused by current level
of metabolic control, the power of patients' own distant spectacles for 31(32.3%) patients
and the measured refraction at presentation were correlated, statistically significant
correlations were found (rho=0.945, p-value =0.001). Each patient was requested to come
back for HBAIC results after 14 days, but only 84 (87.5%) patients came and these were
reexamined to check for variations in refractive status.
There was a statistically significant correlation between refractive status at first
presentation and day 14 (rho=0.978, p=0.001). This suggests that our prevalence estimates
were unlikely to have been influenced by acute metabolic dysregulations. Of the 96
patients, 22.6% had DR and no patient was blind. Of the eyes with DR, 20.0% (15175
eyes) were myopic, 19.4% (7/36 eyes) were hypermetropic and 26.6% (21179 eyes) were
emmetropic. There was no statistically significant correlation at first presentation,
between refractive status and diabetic retinopathy (p=0.358), HBA 1C (rho=0.130, pvalue=
0.249 among myopes) and FBS (rho=-0.089, p-value=0.438 among myopes and
rho=0.158, p-value=0.350 among hyperopes). There was a statistically significant
correlation between baseline refractive status and duration of OM (rho=0.260, p=0.001)
and hypermetropic refractive status and HBAIC (rho=0.401, p-value=0.014) at first
presentation.
Conclusions:
• The patients had poor glycaemic control i.e. 47.9% had HBAIC > 7.3% while
47.9% had FBS > 10.1 mrnol/l.
• Refractive errors were seen in 58.5% of the patients, myopia was the commonest
refractive error (39.5%) while 19.0% were hypermetropic.
• There was no statistically significant relationship between baseline refractive
status and indicators of glycaemic control except for hypermetropic refractive
status and HBAIC (rho=0.401, p-value=0.014).
• The number of OM patients having eye examination for the first time was less than
in previous studies.
Recommendations:
• A study looking at the relationship between refractive status and DR should be
conducted on patients with DR.
• According to the results of this study, it is not mandatory to ask for HBAIC or
FBS results before issuing spectacle prescription to adult patients with type II
diabetes mellitus. However, there is need to emphasize the need for good
glycaemic control to minimize the other ocular complications. A similar study
should be done on young people with type I diabetes mellitus.
Citation
Masters of Medicine (Ophthalmology), University of Nairobi, 2006Publisher
Universityl of Nairobi