Audit of stroke care at the Institute of Neurology, National Hospital of Sri Lanka, 1994-1997.

OBJECTIVE: To audit the process of stroke care. DESIGN: Retrospective case record evaluation using an audit package designed by the Royal College of Physicians of London. SETTING: Institute of Neurology, National Hospital of Sri Lanka, Colombo. PATIENTS: 263 patients with stroke admitted over a period of 3 years. MEASUREMENTS: Documentation of 60 audit items related to 13 aspects of stroke care. RESULTS: The process of care was considered 'very good' for only 11 (18.3%), and 'good' for only 9 (15%) of the audit items. Care was 'average' for 5 (8.3%), 'poor' for 9 (15%) and 'very poor' for 26 (43.3%) of the items. CONCLUSIONS: Stroke care was suboptimal in many aspects. Care related to rehabilitation oriented neurological assessments, initiation of secondary preventive measures, rehabilitation planning and discharge planning were especially deficient. Competing interests: none declared. Some of the data reported in this paper have been presented at the Annual Scientific Sessions of the Sri Lanka Medical Association, 1998.

Glycaemic indices of different varieties of rice grown in Sri Lanka.

OBJECTIVE: Varieties of red raw rice are widely believed to have a better nutritional quality. The physiological effects of consuming different varieties of rice may not be so. The glycaemic index has been developed as an indicator of the physiological effect of foods. It is the glycaemic response of a 50 g carbohydrate portion of food expressed as a percentage of that of a standard. The objective of this study was to determine the glycaemic indices of different varieties of rice grown in Sri Lanka. DESIGN: Digestible carbohydrate content of 11 varieties of rice flour and bread were determined. Fasting blood samples followed by half-hourly samples for two hours were drawn after giving portions of either cooked rice or bread containing 50 g carbohydrate each. SETTING: Fibre mill in Gampaha district. SUBJECTS: Twenty-two fibre mill workers aged between 25 and 50 years. MEASUREMENTS: The area under the blood glucose curve (AUC) for varieties of rice for a subject was calculated. Average AUC of 3 values for bread were calculated. Glycaemic index of each variety of rice was determined from the above variables. RESULTS: Glycaemic indices of varieties of rice differ. The glycaemic indices of varieties of red raw rice varied between 56 and 73 and the variety Bg 350 had the lowest glycaemic index. There was no significant difference between mean glycaemic index of varieties of white raw and some varieties of red raw rice (p = 0.2). Parboiled varieties of red raw rice had a significantly lower glycaemic index than white raw rice (p = 0.04) and some of the red raw rice (p = 0.005). CONCLUSIONS: The glycaemic index cannot be predicted from the colour of the rice grain. Red parboiled varieties of rice and Bg 350 can be recommended for patients with diabetes.

The prevalence of gestational diabetes in a Sri Lankan antenatal clinic.

INTRODUCTION: Early diagnosis of gestational diabetes mellitus (GDM) is a prerequisite to reducing fetal and neonatal complications of GDM. OBJECTIVES: (a) To ascertain the prevalence of GDM in a Sri Lankan pregnant population. Using the 75 g oral glucose tolerance test (GTT) and WHO criteria. (b) To establish the predictive value of a 50 g glucose challenge test (GCT) compared to the GTT (c) To compare the outcome of pregnancy in GDM with 'non-diabetic pregnancy' (NDP) STUDY DESIGN: Prospective study on a cohort of pregnant women attending antenatal clinics. SETTING: Sri Jayawardenepura General Hospital (SJGH) RESULTS: Of the 721 patients, 131 (18%) had a positive GCT. 40 (5.5%) patients had GDM. If a one-hour GCT of 7.8 mmol/l was considered suspicious of GDM the sensitivity of the glucose challenge test was 63% and the specificity 84%. Statistically significant differences in the prevalence was found when the women were > 35 years [Relative risk (RR) = 3.87 (95% CI-2.06 to 7.27)] or the body mass index > or = 25. (RR = 2.45 (95 CI-1.30 to 4.61) Presence or absence of high parity, family history of diabetes or recurrent abortions had no significant impact on the prevalence of GDM. Mean birth weight was higher (p < 0.05) in GDM (3615 SD 103) than in NDP (2898 SD 143.6). The likelihood of having a caesarean section was higher (p < 0.01, Relative risk (RR) 2.50, 95% CI 1.56-3.95) in GDM when compared to NDP. A higher incidence of hydramnios (p < 0.01 RR 3.41 95% CI 1.44-8.05) was recorded in GDM when compared to NDP. CONCLUSION: The prevalence of GDM in the antenatal clinics at SJGH is 5.5%. Traditional risk factors did not predict GDM. GDM is associated with a higher risk of caesarean section, hydramnios and macrosomia. Hence screening for GDM should be performed in all pregnant women at 24 to 28 weeks of pregnancy using a GCT.

The prevalence of peripheral neuropathy in newly diagnosed patients with non-insulin-dependent diabetes mellitus.

INTRODUCTION: Neuropathy is considered to be a longterm complication of diabetes. AIMS: To determine the prevalence of peripheral neuropathy in 112 Sri Lankan patients with recently diagnosed non-insulin-dependent diabetes and a control population of 100 people. METHODS: A diagnosis of peripheral neuropathy was made using a clinical symptom score, clinical examination, quantitative sensory testing and electrophysiological studies. RESULTS: 9.8% of diabetic patients at diagnosis had peripheral neuropathy, 2.6% had foot ulcers, 7.1% had signs of neuropathy, abnormal vibration perception threshold and nerve conduction abnormalities and 15.1% had abnormal nerve conduction velocity without signs or symptoms of neuropathy. The peroneal nerve conduction velocity was higher in diabetic patients when compared to controls. There was an inverse correlation between nerve conduction velocity and fasting blood glucose in diabetic patients (p < 0/05). This association was stronger for peroneal nerve conduction velocity (r = -0.73) than for median nerve motor conduction velocity (r = -0.42). CONCLUSIONS: Symptomatic peripheral neuropathy is common among Sri Lankans with newly diagnosed non-insulin-dependent diabetes. This may be due to a later presentation of diabetes or a genetic predisposition.

Patient mixed biphasic insulin in a diabetic clinic.

INTRODUCTION: Improved glycaemic control is possible with the use of multiple injections of premixed insulin. These are expensive, and not available in state hospitals. OBJECTIVES: To study the cost, patient acceptance and efficacy of a patient mixed and administered combination of soluble and lente (biphasic) insulin administered twice a day. PATIENTS: A cohort of 25 patients with poor glycaemic control on a single dose of 100 units or more of lente insulin. 25 patients matched for age and glycaemic control were used as a control. SETTING: The diabetic clinic of the National Hospital Sri Lanka. METHOD: A prospective study of a cohort of patients. RESULTS: Mean fasting blood glucose decreased from 8.3 mmol/l (SD 3.1) to 6.9 mmol/l (SD 2.3, p < 0.01) and mean blood glucose levels declined from 12.3 mmol/l (SD 4.1) to 10.1 mmol/l (SD 4.7, p < 0.01) in the biphasic group. Total mean insulin dose fell from 80 units (SD 12) to 61 units (SD 11) in the biphasic group, but increased in the control group from 82 units (SD 16) to 91 units (SD 13.1). The diabetes well-being score in the biphasic group was 91.5 (SD 35.3), while the control group had a score of 63.7 (SD 21.3 p < 0.01). Mean glycosylated haemoglobin (HbA1c %) was 8.1 (SD 2.7) in the biphasic group compared to 9.2 (SD 3.3) in the control group. CONCLUSION: Patient mixed and administered biphasic insulin on a twice daily basis is feasible, acceptable to patients, results in better glycaemic control and affords better patient satisfaction.

An audit of structure, process and outcome of care of the diabetic clinic, National Hospital of Sri Lanka.

AIMS: To audit the structure, process and outcome of care. SETTING: The diabetic clinic, National of Hospital Sri Lanka (NHSL). METHODS: A previously validated MCQ paper of 10 questions which assessed knowledge of diabetes on insulin therapy, dietary management, management during acute illness and management of emergencies was administered to all patients. The function of the clinic was assessed using previously validated audit case record forms. MEASURES OF OUTCOME: Diabetes knowledge among patients, waiting times, bypassing of local institutions, availability of diagnostic equipment, screening activities and time spent for consultation. RESULTS: The clinic had a daily average attendance of 186 patients seen between 0800 to 1200 hours. A single medical officer spent 2.1 minutes for each patient. No screening was performed. There were no facilities to examine patients or for them to sit during consultation. The diabetes knowledge score was 15.1 (SD 3) from a maximum score of 40.43% had bypassed a local institution. Reasons for bypass included non-availability of drugs and the expectation of quality care at NHSL. Patients spent a mean of 1.5 (SD 0.7) hours travelling to the clinic and waited a mean of 1.56 (SD 0.4) hours to see the doctor and 1.3, (SD 0.12) hours to obtain drugs. CONCLUSIONS: The services of the diabetic clinic do not meet the standards expected of a clinic at a tertiary referral centre. Lack of planning and resources (space, manpower and management skills) can be identified as principal shortcomings.

Macrovascular disease in Sri Lankan kidney transplant recipients.

INTRODUCTION: Macrovascular diseases such as angina, myocardial infarction, stroke and peripheral occlusive arterial disease are common causes of mortality and morbidity in transplant patients. OBJECTIVES: To study the frequency and determinants if vascular disease in Sri Lankan patients with renal transplants. PATIENTS: A cohort of 237 renal transplant recipients (163 males (76%), mean age 35.09, SD 11.05 years). SETTING: Patients under follow up at the Faculty of Medicine Transplant program. METHOD: Prospective study of a cohort of patients. RESULTS: Macrovascular diseases developed in 42% of transplant recipients. 5.6% had transient ischaemic attacks, 2.8% strokes, 18.2% angina, 8.4% myocardial infarctions, 2.3% underwent lower extremity amputations because of vascular insufficiency, 7% peripheral vascular occlusive disease and 49.5% were hypertensive. Male sex (odds ratio of 2.87, 95% CI 1.33-6.85, p = 0.006), serum cholesterol greater than 6.2 mmol/l at transplant (odds ratio 8.60, 95% CI 2.53-38.9, p = 0.0001) at 5 years after transplant (odds ratio 10.25, 95% CI 1.22-22.9, p = 0.01) were associated with a higher risk of macrovascular disease. Pre-transplant diabetes was associated with a higher risk of macrovascular disease and lower extremity amputation (odds ratio 4.53, 95% CI 1.32-6.76, p = 0.007). CONCLUSIONS: Strategies to screen for and treat potential risk factors should form an integral part of the follow up for Sri Lankan and Asian kidney transplant recipients.

The prevalence of neuropathic foot ulceration in Sri Lankan diabetic patients.

OBJECTIVE: To determine the prevalence of diabetic neuropathy and neuropathic ulceration among patients with non-insulin dependent diabetes (NIDDM) attending a Sri Lankan diabetic clinic. SETTING: The diabetic clinic of the General Hospital Colombo (Sri Lanka National Hospital). RESEARCH DESIGN AND METHOD: Five hundred randomly selected diabetic patients (mean age 47.4 SD 17.59 years, 275 were males, mean duration of diabetes 5.29, SD 6.44 years) were screened for diabetic neuropathy using a neuropathy symptom score (NSS) neurological disability score (NDS) and pressure preception threshold using Semmes Weinstein monofilaments. RESULTS: Patients with neuropathy were older (mean 55.69 years SD 14.16) than those who did not (mean 47.1 years, SD 15.05 p = 0.001) and had diabetes for a longer period (mean 7.5, SD 8 years vs 4.8 SD 5.66, p = 0.002). 123 (30.6%, 95% CI 28-32%) patients had neuropathy according to the criteria used. 51(10.2%, 95% CI 8.2-12.2%) had a foot ulcer, a history of foot ulceration or a lower extremity amputation due to neuropathic ulceration. 26(5.1%, 95% CI 3.2-7%) patients had neuropathic foot ulcers at presentation and a further 14 (2.8%, 95% CI 1.4-4.2%) had developed an ulcer within one year of diagnosis of diabetes. 24 (4.8%, 95% CI 3-6.8%) had a history of lower extremity amputation. CONCLUSION: Neuropathic ulceration is a significant cause of morbidity in patients with NIDDM. One third of all NIDDM patients attending the clinic had a risk of foot ulceration. Preventive foot care programmes should be implemented at all diabetic clinics in Sri Lanka.

The work of a diabetes clinic: an audit.

OBJECTIVE: To audit outcome and process of care at the diabetic and medical clinics. DESIGN: Prospective interventional study with each patient acting as a retrospective control. SETTING: The Sri Jayawardenepura General Hospital (SJGH) diabetic clinic. PATIENTS: 250 consecutive patients with NIDDM transferred to the diabetic clinic from the general medical clinic. INTERVENTION: Participation in a structured scheme of health care delivery which included patient education dietary advice and screening for complications. MEASUREMENTS: Fasting blood glucose, patients' knowledge score, patient satisfaction, frequency of screening. RESULTS: A Significant improvement was seen after attendance at the diabetic clinic in fasting blood glucose (8.7 SD 1.3 vs 12.5 SD 3.2 mmol/l, p < 0.05), body mass index (26.8 SD 0.6 vs 27.4 SD 0.4 kg/m2, p < 0.05), admission rates (10% vs 21%, p < 0.05), patients knowledge score (76 SD 12 vs 42 SD 16%, p < 0.05) rates of screening for retinopathy (88% vs 6%, p < 0.05) and patient satisfaction in clinical care (87% vs 13%, p < 0.05). There was no significant improvement in cholesterol levels (5.7 SD 0.1 vs 5.8 SD 0.9 mmol/l). CONCLUSIONS: The diabetes clinic is a more efficient system and a more appropriate setting to care for patients with NIDDM.