31 / 48
VOLUME 12 NUMBER 2 • NOVEMBER 2015
73
SA JOURNAL OF DIABETES & VASCULAR DISEASE
RESEARCH ARTICLE
the need of insulin treatment. Patients not fitting into the clinical
features of either type were classified as undetermined and excluded
from the present analysis.
All patients completed a questionnaire on demographics and CV
history including smoking behaviour, presence of hypertension and
use of antihypertensive drugs. Clinic blood pressure was measured in
the supine position using a calibrated mercury sphygmomanometer
after at least half an hour of rest in a quiet room. The systolic BP
(SBP) was recorded at Korotkoff phase one and diastolic BP (DBP)
at phase five. Blood pressure was measured twice within an interval
of five to 10 minutes; the second measurement was taken as the
clinic BP.
Body weight was measured to the nearest 0.5 kg and height to
the nearest 0.5 cm. Body mass index was calculated as weight (kg)/
height (m
2
) and categorised according to the WHO physical status
interpretation.
15
Patients with a BMI of ≤ 18.5 kg/m
2
were regarded
as underweight, those with a BMI of 18.5–24.9 kg/m
2
as normal,
25.0–29.9 kg/m
2
as overweight and those ≥ 30 kg/m
2
as obese.
Waist and hip circumferences were measured in centimetres and the
waist-to-hip ratio (WHR) was calculated. Waist circumference (WC)
was measured to the nearest 0.1 cm, at the midway point between
the lowest rib and the iliac crest, at minimal respiration. Men with
WC above 102 cm (40 inches) and women with WC above 88 cm
(35 inches) were considered to have abdominal obesity.
16
Electrocardiogram
A resting, standard 12-lead ECG was obtained in all patients in a
supine position in a quite room. Recording was done at a speed of
25 mm/sec and calibration was standardised at 10 mm/mV on a
Schiller electrocardiograph Cardiovit AT-1 (Schiller AG, model No:
SHL41, Altgasse 68, CH-6341 Baar, Switzerland).
The rhythm was read from the rhythm strip obtained from lead
V1. The rhythm was classified as sinus rhythm, atrial fibrillation or
other. Left ventricular hypertrophy was diagnosed by Sokolow-Lyon
voltage criterion (the sum of the amplitude of the S wave in lead
V1 or V2 and R wave in lead V5 or V6, in both genders) and Cornell
voltage-duration product criterion (the sum of the amplitudes
of the R wave in lead aVL and S wave in lead V3, adding 6 mm
in women, and multiplied by the QRS duration). LV hypertrophy
was considered present if the Sokolow-Lyon voltage criterion was
above 35 mm,
17
or if the Cornell voltage-duration product criterion
was above 2 440 mm/ms. 18-21 Q waves, (code 1-1), ST-segment
deviation (codes 4-1 to 4-4), T-wave abnormalities (codes 5-1 to
5-4) and intraventricular conduction defects (codes 7-1 to 7-8)
were classified using the Minnesota codes.
22
Blood and urine analysis
Blood samples were drawn from the antecubital vein, centrifuged
and stored at −80°C until shipped to the central clinical laboratory
at the Haukeland University Hospital in Bergen, Norway. Samples
were analysed on a Modular Analytics SWA auto analyser (Roche
Diagnostics GmbH, 68298MannheimGermany) using kits produced
by Boehringer Mannheim (Mannheim, Germany). Serum and urine
creatinine were analysed using a modified kinetic method of the
Jaffé reaction, serum cholesterol was assessed on cholesterol oxidase
paminophenazone (CHOD-PAP) levels, and serum triglycerides
by the glycerol phosphate oxidase p-aminophenazone (GPOPAP)
method.
Capillary blood glucose (fasting or random), haemoglobin
(Hg) and HbA
1c
levels were measured in Dar es Salaam. Capillary
blood glucose was measured on a HemoCue AB glucose analyser
(Angelholm, Sweden), haemoglobin on a HemoCue analyser and
HbA
1c
was assessed by DCA 2000
®
+ (Bayer Corporation). The DCA
2000
®
+ was standardised against the DCCT method and verified
in 1996.
23
Quality control was maintained using standardised
solutions.
Urine analysis was done on two overnight specimens collected
on two separate days. Samples were screened for features of urinary
tract infection and excluded if present. Urine albumin concentrations
were determined using an automated immunoturbidity assay with
a sensitivity of 2.3 mg/l and inter- and intra-assay coefficients of
variation of 4.4 and 4.3%, respectively. A urinary albumin excretion
rate (AER) of ≤ 20 μg/min was categorised as normoalbuminuria,
20.1−200 μg/min as microalbuminuria and AER levels of > 200 μg/
min as macroalbuminuria.
Cardiovascular risk factors
The CV risk factors in this study were assessed and defined as
described by the 1999 WHO/ISH guidelines.
24
These included waist
circumference, hypertension, ECG-LVH, albuminuria, advanced age,
smoking, dyslipidaemia and albuminuria. Waist circumference was
chosen as the indicator variable for abdominal obesity as it has been
established to be a better predictor of cardiovascular health risks
than BMI.
25
Subjects were considered to be hypertensive when the
clinic blood pressure was ≥ 140 mmHg systolic and/or ≥ 90 mmHg
diastolic, or when using antihypertensive treatment.
26
Advanced
age was defined as age over 55 years for men and over 65 years
in women. Dyslipidaemia was defined as total cholesterol levels of
above 6.5 mmol/l and/or LDL cholesterol above 4.0 mmol/l, and/ or
HDL cholesterol below 1.0 mmol/l for men and below 1.2 mmol/l
for women. Albuminuria was considered present if AER ≥ 20 μg/
min. Diagnosis of ECG-LVH has been described in detail above.
Statistical analysis
This was performed using the statistical package for social sciences
(SPSS) software, version 13 for windows (SPSS, Inc, Chicago,
Illinois). Continuous data are reported as mean ± two standard
deviations and categorical data as proportions. For variables with
skewed distribution, which were age, diabetes duration, systolic
and diastolic BP, AER and creatinine clearance, the median ± range
was used. Patients were grouped according to type of diabetes.
Groups were compared using the unpaired t-test, Fisher exact
test and Mann-Whitney test where appropriate. A logistic regression
analysis test was used to quantify the association between covariates
and the presence of ECG-LVH. In regression analysis, results are
given per 10 mmHg higher systolic BP, per 5 mmHg higher diastolic
BP and per 40 msec longer QRS duration for meaningful clinical
interpretation. Results of regression analyses are given as odds ratio
(OR) with 95% confidence interval (CI). A two-tailed p-value less
than 0.05 was considered statistically significant in univariate and
multivariate analyses.
Results
Two hundred and seventy-one patients were enrolled. Of these,
54.3% were women; six patients did not complete the study and
were excluded, with 263 patients remaining. In 14 patients, the
type of diabetes could not be clearly determined and they were
excluded from the present study population, as were 12 patients
(six type 1 and six type 2) who did not have an ECG taken. In the