18 / 40
RESEARCH ARTICLE
SA JOURNAL OF DIABETES & VASCULAR DISEASE
60
VOLUME 16 NUMBER 2 • NOVEMBER 2019
Bivariate logistic regression analysis of the association between
age and gender of the diabetic subjects, as well as serum levels of
the major types of serum AGEs with endothelial dysfunction (serum
NO levels less than the first quartile) revealed that only higher serum
levels of CML were significantly associated with higher crude odds
of endothelial dysfunction [COR (95% CI), 1.910 (0.655–0.893)
(
p
< 0.05) (Table 3).
Discussion
As expected, serum levels of TIAGEs, CML and CEL were found
to be significantly higher in the diabetic patient group compared
with the non-diabetic control group. However, serum FAGE levels
of diabetic patients were not significantly different from those of
non-diabetic controls. This observation might be attributed to the
nature of the control group used in the study.
High serum FAGE levels, in particular high serum levels of
pentosidine, the most abundant fluorescent AGE in plasma
and tissues, have been associated with the development
and progression of osteoporosis in diabetic and non-diabetic
menopausal women.
18,19
Whether the high level of pentosidine
observed in the cited studies was the cause or product of
osteoporosis is currently not clear. It is possible that our patient
control group, which was recruited from orthopaedic wards
at DGMAH, may have included non-diabetic postmenopausal
women with osteoporosis-related fractures. While this likelihood
was not verified in the current study, it might explain the observed
high levels of FAGE in the non-diabetic control group.
Previous studies reported in the literature have used circulating
levels of NO, ET-1 and PAI-1, among others, as surrogate markers
of endothelial dysfunction in vivo.
3,4,20
According to these previous
studies, serum levels of NO and its metabolites are expected
to be decreased, while serum levels of both ET-1 and PAI-1 are
expected to be increased in conditions associated with endothelial
dysfunction, such as type 2 diabetes mellitus. Therefore the
findings of significantly reduced NO levels and significantly higher
serum levels of both ET-1 and PAI-1 are in perfect agreement with
the results of these previous studies. However, these findings
should be interpreted with caution, since these circulating markers
of endothelial dysfunction may come from sources other than the
vascular endothelium.
4,20
The observation in this study that serum NO levels were
negatively and significantly correlated with the age of the study
subject is in agreement with the well-documented observation that
endothelial function decreases with advanced age.
21,22
The findings
that serum levels of both TIAGEs and CML were negatively and
significantly correlated with serum NO levels and positively and
significantly correlated with serum levels of ET-1 were also not
unexpected, since high levels of some serum AGEs are known to
promote endothelial dysfunction through their interaction with
RAGE on the surface of the vascular endothelial cell.1 The finding
that serum CML level was the only parameter in this study that
was significantly associated with increased odds of developing
endothelial dysfunction suggests that serum CML is the major type
of serum AGEs that interacts with RAGE to promote endothelial
dysfunction.
Limitations
There are several limitations that should be taken into consideration
when interpreting results of this study. Firstly, the sample size
was small and study subjects were recruited from a single health
institution, therefore the findings could not be generalised beyond
the study samples. Secondly, the study was cross-sectional and
therefore cause and effect relationships could not be inferred from
the results. Thirdly, the possible confounding effect of exogenous
dietary and smoking-related AGEs on serum AGE levels was not
addressed. Fourthly, the control group selected for this study might
have confounded the results, particularly those of the FAGEs. Fifthly,
we did not concurrently measure serum AGE levels and circulating
markers of endothelial dysfunction of other South African race
groups for comparison purposes.
Despite these limitations, we believe that the results of this study
are of great interest in that they are the first to describe the status
of serum AGE levels among black South African patients with type
2 diabetes, as well as the association between serum AGE levels
and endothelial dysfunction in black South African patients with
type 2 diabetes mellitus.
Conclusions
The results of this study showed that serum AGE levels were
significantly higher in type 2 diabetes patients than in non-diabetic
black South Africans, and with the exception of CEL were not
influenced by gender. In addition, serum FAGE levels appeared to be
positively associated with increasing age of the subjects in the non-
diabetic controls, but not in in the diabetic subjects. Furthermore,
the findings of this part of the thesis showed that serum TIAGEs,
CML, CEL, ET-1 and PAI-1 levels were significantly elevated, whereas
serum levels of NO were significantly reduced in black South African
patients with type 2 diabetes compared to those in non-diabetic
control subjects. Moreover, the findings indicated that serum TIAGE
and CML levels, but not CEL and FAGE levels were correlated with
endothelial dysfunction in black South African patients with type 2
diabetes mellitus. However, only serum CML levels were associated
with a higher odds of developing endothelial dysfunction in these
black South African type 2 diabetes patients.
Acknowledgements
We acknowledge the contribution of the nursing and medical
personnel as well as the phlebotomists at the diabetes clinic of
Dr George Mukhari Academic Hospital. We are grateful for the
research funding obtained from the National Research Foundation
(grant no. TP1407187704).
Table 3.
Bivariate logistic analysis of the association between gender,
age and the major types of serum AGEs with endothelial dysfunction
(less than the first quartile of NO levels)
Parameters
COR
95% CI
p
-value
Age
0.600
1.372–2.62
0.460
Gender
1.040
0.996–1.12
0.296
TIAGEs (μg/ml)
0.348
0.014–8.916
0.523
CML (ng/ml)
1.910
0.655–0.893
0.013*
CEL (ng/ml)
1.172
0.963–1.638
0.112
FAGEs (Au)
0.991
0.882–1.038
0.141
COR: crude odds ratio; CI: confidence interval; TIAGEs: total immunogenic
advanced glycation end-products; CML: N
ε
-carboxymethyl-lysine; CEL:
N
ε
-carboxyethyl-lysine; FAGEs: fluorescent advanced glycation end-products;
*Significant at
p
< 0.05.