SUMMARY Diabetes mellitus has considerable impact on the eye, accounting for one of commonest causes of blindness in the world. This study therefore aims to determine the pattern of anterior segment manifestations of diabetes mellitus and its impact on visual status of the diabetic patients at the Endocrinology Unit of the University of Port Harcourt Teaching Hospital, Port Harcourt. Also, it is intended to evaluate the knowledge of patients about the eye complications of diabetes and suggest ways for early detection and management of diabetic eye complications in order to reduce the blindness rate due to disease in the Teaching Hospital. A hospital-based cross-sectional study of diabetic patients attending the Endocrinology clinic of University of Port Harcourt Teaching Hospital was carried out between 25th March and 28th June, 2011. Consecutive adult diabetic patients, 19 years and older on each clinic day was recruited into the study. They were randomly booked on each clinic day and had no prior knowledge of the study, which ensured unbiased selection of the patients. A total of 225 patients consisting of 110 (48.9%) males and 115 (51.1%) females were examined within a three month period. Informed consent was obtained before examination which began with administration of interviewer-administered semi-structured questionnaires at the Endocrinology clinic. All patients were then taken to the Eye Clinic for visual acuity measurement using the Snellen’s chart (alphabet and tumbling E); anterior segment examination using torch light and slit lamp. The anterior segment angle was evaluated with Goldmann 3-mirror goniolens. Noncycloplegic manual refraction was carried out on all patients and intraocular pressure measured with Perkin’s hand-held applanation tonometer. Dilated fundoscopy with direct ophthalmoscope was done on patients with clear ocular media. Patients with vertical cup disc ratio (VCDR) > 0.5 had full threshold central visual field analysis. The data obtained was recorded in data sheets and analysed using the Statistical Package for Social Sciences (SPSS) version 16 (2007). Simple statistics such as means, frequencies and ratios were determined. Comparisons of associated variables were made using p–values, chi-square and Fisher exact test. A p-value of < 0.05 was regarded as statistically significant (95% confidence interval). Mean value was calculated for intraocular pressure (IOP); frequencies were calculated for age, sex, educational status, duration of diabetes, types of treatment and glycaemic control. Pearson’s correlation was calculated for age and awareness of eye complications of diabetes; and educational status and the eye complications of diabetes mellitus. Positive values indicate direct relationship while negative values imply an inverse relationship. Of the 225 patients examined, 13 (5.8%) were bilaterally blind while 10 patients (4.4%) had unilateral blindness. Blindness was commoner in the males (n=15) than females (n=8), with a ratio of 1.9:1; but this difference was not statistically significant (P>0.005). Twenty patients (8.9%) had unilateral visual impairment while 26 (11.5%) had bilateral visual impairment. Age significantly affected the prevalence of blindness (P=0.001) but not the prevalence of visual impairment (P=0.293). The duration of diabetes did not significantly affect the prevalence of blindness and visual impairment in this study (P>0.005). Cataract (n=16) was the leading cause of both bilateral and unilateral blindness. Other causes of blindness were primary open glaucoma (n=4), vitreous haemorrhage (n=1) and diabetic macular oedema (n=2). Refractive error (n=8) was the leading cause of bilateral visual impairment. Other causes of bilateral visual impairment include primary open-angle glaucoma (n=3) and cataract (n=7). Unilateral visual impairment were caused by cataract (n=14), refractive error (n=10), primary open angle glaucoma (n=2) and diabetic macular oedema (n=2). Refractive error found in 154 (68.4%) patients was the commonest anterior segment manifestation and hypermetropia (n=128, 56.9%) was found to be commoner than myopia (n=26, 11.5%). Cataract was the second commonest anterior segment disorder (n=45, 20%). Other anterior segment disorders included ocular hypertension (n=21, 9.3%), iris new vessels (n=2, 0.9%) and primary open angle glaucoma (n=43, 19.1%). One hundred and twenty-eight (56.9%) of the patients knew that diabetes could cause eye complications but only 33 (25.8%) knew about specific eye complications associated with diabetes. There was no statistically significant difference between the males (n=62) and females (n=66) (P=0.195) on their awareness of the effect of diabetes on the eye. However, the age of the patients and their educational status significantly affected their awareness and knowledge of diabetic eye complications (p<0.005). There was also a positive correlation between age and educational status of the patients and their awareness of the effect of diabetes on the eye. The duration of diabetes did not have significant impact on the level of awareness of patients about eye complications of diabetes (P=0.066) Fifty-nine (26.2%) of the patients have had previous eye examinations and 36 (61%) of them went on self referral. Age, sex, educational status and awareness of the eye complications of diabetes did not significantly affect the proportion of those who have had an eye examination. However, the duration of diabetes has statistically significant effect (p=0.005) on those who have had an eye examination. The findings in this study underscores the need to improve patients’ education on diabetic eye complications and the adoption of collaboration between the physicians in the Endocrinology Unit and ophthalmologists aimed at improving visual screening of diabetic patients by the former for early detection of eye complications. It is necessary for these physicians to be retrained in the use of torch light and direct ophthalmoscope to detect simple ocular complications in diabetic patients and ensure prompt referral to the ophthalmologist.