36 / 74
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 25, No 5, September/October 2014
230
AFRICA
observation was made regarding the proportions of patients with
hypertensive retinopathy among those with and without CKD
(89.3 vs 83.3%) (
χ
2
=
0.12,
p
=
0.73). The association of CKD with
hypertensive retinopathy was not significant for all retinopathy
stages combined (
χ
2
=
0.17,
p
=
0.68, OR
=
1.7, 95% CI: 0.4–6.6)
or for each retinopathy stage taken individually (
χ
2
=
0.03–2.82,
p
=
0.09–0.85). Compared to patients without hypertensive
retinopathy, those with stages 3 and 4 hypertensive retinopathy
were 3.3 and 13.3 times more likely to have CKD, respectively.
There were 85.7% of patients with hypertensive retinopathy
among those who suffered from stroke (28 patients, Table
3). This proportion was not significantly different from the
83.2% of patients with hypertensive retinopathy among those
without stroke (
χ
2
>
0.001,
p
=
0.99). Patients with hypertensive
retinopathy were as likely as those without retinopathy to have
stroke (18 vs 15.4%) (
χ
2
=
0.34,
p
=
0.56). No association was
found between stroke and hypertensive retinopathy regardless
of retinopathy stage (
χ
2
<
0.01,
p
=
0.96, OR
=
1.2; 95% CI:
0.4–3.8) and for individual retinopathy stages (
χ
2
=
0.02–1.06,
p
=
0.30–0.88).
A subset of data of 75 patients with complete documentation
was used to perform a multivariate logistic regression analysis
that included age, gender, BMI, alcohol consumption, smoking,
diabetes, arterial pressures (systolic, diastolic and pulse), current
blood pressure-lowering treatment, LVH, CKD and stroke as
candidate explanatory variables, and hypertensive retinopathy
as outcome variable after controlling for diabetes. The results,
shown in Table 4, indicate that CKD was the most significant
predictor of hypertensive retinopathy, with OR of 4.4 compared
to CKD-free patients. Age
>
50 years and smoking appeared to
decrease the risk of hypertensive retinopathy; the effects were
negligible but significant.
Discussion
Hypertension is an important cause of morbidity and mortality
in the general population in Western countries, and recent
surveys in sub-Saharan Africa have reported high prevalences
of hypertension ranging between 19 and 50% in both urban
and rural populations.
19,20
If left untreated, hypertension may
result in considerable damage to the cardiovascular, renal
and cerebrovascular systems, leading to such complications as
myocardial infarction, CKD and cerebrovascular accident.
While significant efforts have been invested to demonstrate
the benefits of antihypertensive treatment, it is critical for
better management to know both to what extent the various
hypertension-related TODs are interrelated, and the risk factors
for hypertension-related damage. Because studies in this regard
are limited in sub-SaharanAfrica, we investigated the relationship
between hypertensive retinopathy and LVH, CKD and stroke
among Congolese patients. We also assessed the determinants of
hypertensive retinopathy.
It has been hypothesised that both hypertension-related
retinal and renal vascular changes share common pathogenetic
mechanisms. As a result, earlier studies have consistently reported
an association between the presence of retinal vascular changes
associated with hypertension and lower GFR.
21-23
Surprisingly,
our results suggest otherwise, which may be ascribed to the small
study population.
Signs of hypertensive retinopathy have also long been
recognised as risk indicators of LVH, both in population- and
hospital-based studies.
24-26
For example, in the Chronic Renal
Insufficiency Cohort (CRIC) study,
26
there was an association
between severity of hypertensive retinopathy and the incidence
of any cardiovascular disease. Similarly, a follow up of the
National Health and Nutrition Survey (NHANES I) reported
an increased risk of cardiovascular disease in people with
hypertensive ocular fundus retinal vascular changes.
27
The lack of association between hypertensive retinopathy
and LVH found in our study echoes the findings of other earlier
studies.
28,29
While there is a general agreement on the association
between hypertensive retinopathy and all types of hypertensive
cardiovascular diseases,
26,27,30
our study only focused on LVH,
which may explain the lack of association. Overall, our findings
corroborate those of earlier studies that the risk of developing
LVH increases significantly with the severity of hypertensive
retinopathy.
Table 1. Association between hypertensive retinopathy
and left ventricular hypertrophy (LVH)
Retinopa-
thy grade
With LVH
(%)
Without
LVH (%)
OR
(95% CI)
Chi-
square
p
-value
0
7 (13.5)
12 (26.7)
1
–
–
1
19 (36.5)
21 (46.7) 1.6 (0.5–4.8)
0.24 0.62
2
10 (19.2)
5 (11.1) 3.4 (0.8–14.2) 1.91 0.17
3
13 (25.0)
5 (11.1) 4.5 (1.1–17.9) 3.34 0.07
4
3 (5.8)
2 (4.4)
2.6 (0.3–19.3) 0.18 0.67
OR: odd ratio, CI: confidence interval.
Table 2. Association between hypertensive retinopathy
and chronic kidney disease (CKD)
Retinopa-
thy grade
With CKD
(%)
Without
CKD (%)
OR
(95% CI)
Chi-
square
p
-value
0
3 (10.7)
10 (16.6)
1
–
–
1
9 (32.1)
26 (43.3)
1.2 (0.3–5.2)
0.04 0.85
2
2 (7.1)
13 (21.7)
0.5 (0.07–3.7)
0.03 0.86
3
10 (35.7)
10 (16.7)
3.3 (0.7–15.9)
1.4 0.24
4
4 (14.3)
1 (1.7)
13.3 (1.1–169.1) 2.8 0.09
OR: odd ratio, CI: confidence interval.
Table 3. Association between hypertensive
retinopathy and stroke
Retinopa-
thy grade
With
stroke (%)
Without
stroke (%)
OR
(95% CI)
Chi-
square
p
-value
0
4 (14.3)
22 (16.8)
1
-
-
1
11 (39.3)
56 (42.7) 1.1 (0.3–3.8)
0.04 0.85
2
6 (21.4)
12 (9.2)
2.8 (0.7–11.7) 1.1 0.30
3
4 (14.3)
33 (25.2) 0.7 (0.2–2.9)
0.02 0.88
4
3 (10.7)
8 (6.1)
2.1 (0.4–11.3) 0.2 0.70
OR: odd ratio, CI: confidence interval.
Table 4. Significant determinants of hypertensive retinopathy
Parameters
β
p
-value
OR (95% CI)
Constant
–0.88
0.23
0.41
Chronic kidney disease
1.49
0.018 4.4 (1.29–15.21)
Age
>
50 years
–1.46
0.046 0.23 (0.06–0.97)
Smoking
–2.02
0.035
0.1 (0.02–0.9)
OR: odd ratio, CI: confidence interval.