5 Discussion
In this study we evaluated the relations between different indices of
glucose metabolism and the development of CKD in a large population of
middle-aged Chinese individuals from the REACTION study. The results
showed that 3 indices of glucose metabolism, 2h OGTT, HbA1c, and HOMA-IR
were significantly associated the development of CKD, independent of
potential confounding risk factors. Of the 3 indices, HOMA-IR exhibited
the best predictive ability. To the best of our knowledge, this is the
first and largest population-based cohort study to examine the best
index of glucose metabolism for predicting the development of CKD.
Because the only effective treatment for ESRD is transplantation,
controlling risk factors for the development of CKD, and screening
methods to determine persons at greater risk of developing CKD are
important for decreasing the number of patients develop ESRD. The
findings of the present study may assist in identifying persons who are
at risk of developing CKD and who may benefit from early interventions.
CKD is becoming a global health problem, and global deaths from kidney
disease have risen by 83% since 199021. Glucose metabolism
has been shown to be an important factor in the development of CKD12,22,23.
Our finding that an elevated 2h OGTT and an elevated HbA1c level are
independent risk factors for development of CKD is consistent with that
of prior studies23,24.
Gabir et al. 24 studies
5023 Pima Indian adults, and with a follow-up period of 10 years showed
that a 2h OGTT can predict the development of CKD. Markus et al.23 studies 7728
subjects with a median follow-up of 8.7 years; 871 (11.3%) developed
CKD and HbA1c was an independent predictors for the development of CKD.
However, there have been conflicting reports of the association of CKD
development and glucose metabolism. In a study in Germany, Schottker et
al. 25 followed 3,538
participants during for 8 years and found that pre-diabetes might not
contribute to the development of CKD, and that preventive efforts such
as regular exercise might reduce the risk of developing CKD26. An animal study
using rats showed that physical training increased insulin sensitivity
by enhancing muscle glucose uptake and glucose utilization via
glycolysis 27. A study
of hemodialysis patients showed that moderate physical training, using
of plasma insulin level for patients reduces by the 40%28. In addition, some
cross-sectional studies showed that neither glucose tolerance nor
insulin secretion were associated with CKD. Hanssen et al.29 studies 1796 persons
with normal glucose metabolism, 478 with pre-diabetes, and 669 with type
2 DM, and reported no association of CKD with 2h OGTT or HbA1c. However,
the follow-up was relatively short, and the results can be interpreted
as short or intermediate follow-up periods might not capture the
associations of 2h OGTT or HbA1c with CKD. It is possible that there
might be geographic or race variability with respect to the association
of 2h OGTT and HbA1c and the development of CKD, as study has shown that
both are predictors of insulin resistance, which is strongly associated
with the development of CKD30. Clinical
significance of each indicator is expected to be studied in the future
and further explored with long-term and multi-stage longitudinal
measures to better define the relationship.
The results of this study showed that HOMA-IR was the strongest
predictor of the development of CKD in middle-aged and elderly Chinese.
The HOMA-IR reflects a pathological state in which target tissues fail
to respond normally to the biological effects of insulin, and is
generally considered as an important influential factor for development
of CKD. A study using NHANES data showed that individuals with the
highest insulin levels had a 2.65 times greater risk for the development
of CKD (95% CI: 1.25-5.62)31. Ma et al.32 studied 3,237
middle-aged and elderly Chinese persons with a 3-year follow-up and
showed that an elevated HOMA-IR was associated with accelerated
progression of CKD. Huh et al.33 studied 6,065 Korean
persons without CKD at baseline, and over a follow-up period of 10 years
showed that insulin resistance was independent risk factor for
development of CKD. However, a prospective study of 73 non-diabetic
subjects with CKD showed that of incident CKD in Korean population.
However, a prospective study of 73 non-diabetic subjects with CKD showed
that HOMA-IR was not significantly different in patients with or without
renal endpoints 34. And
the findings of the study might not extend to our study since it was
conducted from one region of Turkey with small sample (n=73).
There are some theories as to why insulin resistance increases the risk
of developing CKD. First, normally, insulin binds to the insulin
receptor can activate insulin receptor substrate-1 (IRS-1), which can
phosphorylated phosphatidylinositol 3-kinase (PI3-K). Under insulin
resistance conditions, Impaired PI3-K lead to reductions in bioavailable
nitric oxide (NO) directly resulting in the development of endothelial
dysfunction and CKD 35.
Secondly, insulin resistance promotes CKD at the molecular level by
inflammation through endoplasmic reticulum (ER) stress, and is involved
in the pathophysiology of chronic kidney injury with tubulointerstitial
damage 36. Moreover,
insulin resistance can increase the levels of inflammatory, cytokines,
which can lead to basement membrane thickening, glomerular mesangial
expansion, and the loss of slit pore diaphragm integrity, ultimately
leading to glomerulosclerosis and tubule-interstitial injury37. Thirdly, insulin
resistance may cause overproduction of LDL-C and contribute to
hypertriglyceridemia, which can result in renal disease38. Triglyceride-rich
apolipoprotein B-containing lipoproteins promote the progression of
renal insufficiency 39.
Lastly, insulin resistance promotes CKD by worsening renal hemodynamics
through mechanisms such as activation of the sympathetic nervous system,
sodium retention, decreased Na+,
K+-ATPase activity, and increased GFR40,41.
The causes of insulin resistance are complex and multifactorial, and
involve genetic factors such as post-receptor signaling defects, an
unhealthy lifestyle that includes a lack of physical activity and poor
diet which can lead to obesity, obesity, medications, aging, metabolic
acidosis, oxidative stress, inflammation, vitamin D deficiency, uremic
toxicity, and anemia, as shown by previous human and animal studies.
Other co-morbid conditions that are strongly associated with insulin
resistance are hypertension, diabetes, and hyperlipidemia.
Several limitations of in this study need to be addressed. First, by
including only middle-aged and elderly Chinese subjects, the results
might not apply to different races or a population of younger
individuals. Secondly, our study population was predominantly female;
this was partially because we invited person ≥40 years old to
participate and females are predominant in this age range in China.
Third, as with any observational study all confounding factors that may
contribute to the development of CKD may not have been included in the
models. Fourth, the ”gold-standard” for documenting insulin resistance
is the euglycemic clamp test. However, the euglycemia clamp test is
time-consuming and requires trained personal so it is rarely used in
large epidemiological studies. HOMA-IR is a common method used to assess
insulin resistance in large epidemiological studies, and it is
relatively well-correlated with the euglycemia clamp technique (r=0.88)42. Fifth, we defined
the CKD based on the first measurements of eGFR and UACR; however, the
gold standard is two measurement results. This approach may have reduced
the accuracy of our results; however, the results of 1 measurement
correlate well with those of 2 measurements and use of 1 measurement is
common in large epidemiology studies43. Finally, the
follow-up rate in this study of 71% was relatively low; however, large
epidemiological investigations rarely studies can achieve follow-up rate
of ≥85%44,45.
It is worth noting that the follow-up rate of another REACTION study
with a defined 3-year follow-up in Shandong Province, China, from 2012
to 2015 was 77.8%, which was similar to that of our
study32.