8 / 62
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 31, No 1, January/February 2020
6
AFRICA
specialising in diabetes. The cases were all patients treated for
hypothyroidism including those with SCH, and the controls
were clinically and biochemically confirmed euthyroid under
follow up between 1 January and 1 July 2016. The aim of the
study was to compare the prevalence and severity of micro- and
macrovascular complications, and indices of glycaemic control
between the cases and controls.
Diabetic kidney disease (DKD) was defined as an estimated
glomerular filtration rate (eGFR)
<
60 ml/min determined by
the CKD-epidemiology collaboration equation (CKD-EPI) and/
or a urine albumin/creatinine ratio (ACR)
>
3 mg/mmol in the
absence of other causes of kidney disease. DR was defined as
the presence of aneurysms, bleeds, exudates and new vessel
formation on retinal examined by an ophthalmologist; DPN
as the presence of symptoms, loss of 128-Hz sensation and/
or abnormal 10-gm monofilament; and CVD as the presence
of major adverse cardiovascular events (MACE: coronary
angioplasty, stent, coronary artery bypass grafting, myocardial
infarction or cerebrovascular accident) and amputation (surgical
removal of any part of a lower limb due to diabetic causes).
SCH was defined as a thyroid stimulating hormone (TSH) level
>
4 mIU/l with a normal T4 and T3 level, and overt hypothyroidism
as a T4 less than the normal range (7.6–16.1 pmol/l) and TSH
>
4 mIU/l. Thyroxine was given to all cases to maintain the T4
and TSH level within the normal range. Diabetes was defined
according to the American Diabetes Association criteria.
7
Cases
were excluded if they were receiving amiodarone or lithium, or
had had previous thyroid surgery or ablation therapy.
The study was approved by the University of Cape Town,
Research Ethics Committee (331/2017).
Statistical analysis
Descriptive statistics were used to summarise total cohort
characteristics. For purposes of analysis, the cohort was divided
into black (Africans) and non-black (whites and other race
groups). Median with interquartile range was used to summarise
continuous variables, and frequency and percentages were used
to summarise categorical variables. Differences in continuous
variables between cases and control patients were compared
using a Wilcoxon rank sum test, while categorical variables were
compared using Pearson chi-squared test or Fisher’s exact test.
Logistic regression was used to determine associations,
magnitude and direction between the dichotomous T2DM
outcomes (DKD, CVD, DPN and retinopathy) and
hypothyroidism, adjusted for
a priori
selection of confounders
and covariates. Highly skewed continuous variables were
log transformed prior to entering into the model. Linear
regression was used to assess associations between eGFR and
a priori
selection of covariates. Goodness-of-fit and influential
observations were assessed after fitting each model. All analyses
were performed using Stata software (Version 14.2, Stat Corp,
College Station, TX).
Results
We identified 310 subjects, of whom 162 were controls and 148
were cases. All the hypothyroid cases were receiving thyroxine.
The overall demographics of the population are shown in Table
1. The ethnic breakdown was predominantly white (84%), black
(13%) and other races (3%), and hypertension was present in
83% of the hypothyroid group and 79% of the controls.
Therewere significant differences in the baseline characteristics
between the two groups. There were more females in the
hypothyroid group (60.8 vs 39.2%,
p
=
0.001) and fewer blacks
(10.8 vs 21.4%,
p
=
0.021) compared to controls. In addition the
mean age of the patients with hypothyroidism and duration of
diabetes was 65 vs 58 years (
p
<
0.001) and 13 vs 10 years (
p
=
0.001), respectively. T4 levels were slightly higher in the cases (12
vs 13.1 pmol/l,
p
=
0.004), but there was no difference in TSH
level.
In respect of diabetic control, the cases had better glycated
haemoglobin (HbA
1c
) levels (6.9 vs 8%,
p
<
0.001) and used fewer
hypoglycaemic medications (
p
=
0.001) (Table 2). There were
differences in use of hypoglycaemic agents with more patients
in the control group receiving dipeptidyl peptidase-4 (DPP4)
inhibitors (40.1 vs 26.4%,
p
=
0.04), incretin mimetics (GLP
agonists) (13 vs 6.1%,
p
=
0.01), and a trend towards more insulin
use (51.9 vs 41.9.
p
=
0.08). There were no significant differences
in the use of metformin and sulphonylureas.
Regarding components of the metabolic syndrome, waist
circumference was not available, but in the cases, high-density
lipoprotein (HDL) cholesterol was significantly higher (1.1 vs
1 mmol/l,
p
=
0.001) and triglyceride levels were significantly
lower (1.9 vs 2.1 mmol/l,
p
=
0.034). There was no difference
in low-density lipoprotein (LDL) cholesterol, but all subjects
received statin therapy unless contra-indicated.
Microvascular complications tended to occur more frequently
in the hypothyroid group. The eGFR was significantly lower in
Table 2. Use of hypoglycaemic drugs in total group, cases and controls
No. of hypoglycaemic
drugs,
n
(%)
Overall
Controls
Cases
p-
value
1
101 (32.6)
38 (23.5)
63 (42.6)
0.001
2
122 (39.4)
66 (40.7)
56 (37.8)
3
67 (21.6)
46 (28.4)
21 (14.2)
4
20 (6.5)
12 (7.4)
8 (5.4)
Metformin,
n
(%)
254 (81.9)
135 (83.3)
119 (80.4)
0.503
Sulphonylurea,
n
(%)
93 (30.0)
52 (32.1)
41 (27.7)
0.399
GLP agonist,
n
(%)
30 (9.7)
21 (13.0)
9 (6.1)
0.041
DPP4 inhibitor,
n
(%)
104 (33.6)
65 (40.1)
39 (26.4)
0.01
Insulin,
n
(%)
146 (47.1)
84 (51.9)
62 (41.9)
0.08
Table 1. Demographics of the total group, cases and controls
Variable
Total group
(
n
=
310)
Controls
(
n
=
162)
Cases
(
n
=
148)
p
-value
Age (years)
62 (54–71)
58 (52–67)
65 (58–75)
<
0.001
Gender, male (%)
146 (47.1)
98 (60.5)
48 (32.4)
<
0.001
Race, non-black (%)
261 (84.2)
129 (79.6)
132 (89.2)
0.021
Duration of T2DM (years) 11 (7–18)
10 (5–16)
13 (9–19)
0.001
BMI (kg/m
2
)
34 (30–41)
34 (29–40)
34 (30–41)
0.370
HbA
1c
(%)
7.4 (6.3–9.1) 8.0 (6.7–9.6) 6.9 (6.1–8.7)
<
0.001
TSH (mIU/ml)
1.6 (1.0–2.5) 1.6 (1.2–2.2) 1.6 (0.8–3.1)
0.973
T4 (pmol/l)
12.3 (11.–15) 12.0 (10–13) 13.1 (11–16.5)
0.004
Total cholesterol (mmol/l) 4.5 (3.7–5.4) 4.5 (3.7–5.4) 4.4 (3.7–5.3)
0.776
Triglycerides (mmol/l)
2.0 (1.3–2.8) 2.1 (1.4–3.2) 1.9 (1.3–2.6)
0.034
HDL-C (mmol/l)
1.0 (0.9–1.2)
1 (0.8–1.1)
1.1 (0.9–1.3)
0.001
LDL-C (mmol/l)
2.6 (2.0–3.2) 2.7 (1.9–3.3) 2.5 (2.0–3.2)
0.766
eGFR (ml/mim)
71 (55–88)
75 (58–90)
66 (52–82)
0.001
Urine ACR (mgm/mmol)
1.8 (0.7–4.9) 1.6 (0.7–5.2) 2.0 (0.8–4.9)
0.717