In this study involving 1,521 Chinese type 2 diabetic patients, 47.1% patients had an eGFR less than 90 mL/min/1.73 m2
and 28.9% had albuminuria. The frequency of CKD was 31.0%, of which 6.5% had renal insufficiency. Our result is in keeping with the result of Shanghai Diabetic Complications Study in which 29.6% Shanghai Chinese diabetic patients had CKD and 26.2% had albuminuria [23
]. Kong et al. [24
] also reported that 15.6% Hong Kong Chinese type 2 diabetic patients had CKD. A community-based study of CKD among type 2 diabetic Chinese patients in Taiwan showed the prevalence of CKD was 38.0% [25
]. However, Lu et al. [26
] reported the prevalence of CKD was as high as 63.9% in downtown Shanghai, China. The marked variation in the prevalence of CKD might be due to the difference in study cohort characteristics and difference in GFR evaluation methods [25
], while some studies did not include albuminuria in the evaluation [24
]. Nevertheless, all the above studies suggested that CKD was common in Chinese population with type 2 diabetes.
Hypertension is a well recognized risk factor for diabetic patients to develop CKD. From the result of our study, hypertension is a prevalent condition in diabetic patients, particularly those with CKD. Up to 66.5% and 81.8% diabetic patients with CKD stage 1 to 2 and stage 3 to 4, respectively, have hypertension. Moreover, hypertension imposes the strongest risk, when compared to other risk factors identified, associated with CKD. United Kingdom Prospective Diabetes Study (UKPDS) has demonstrated that every 10 mm Hg reduction in systolic BP was associated with a 13% reduction in the risk of microvascular complications [27
]. Our result supports that BP control is of crucial importance in diabetes management.
Duration of diabetes was the other risk factor associated with CKD. On the other hand, as an indicator of the average level of glucose, HbA1c was not associated with CKD in the present study. UKPDS study indicated a reduction of 1% in HbA1c was associated with a 37% decrease in microvascular endpoints [28
]. However, in the glucose control and vascular complications in veterans with type 2 diabetes (VADT) study, patients in the intensive arm after a mean follow-up of 5.6 years did not show any benefit regarding changing serum creatinine or GFR values and only a minor effect on albuminuria levels was observed [29
]. The same was observed in the intensive blood glucose control and vascular outcomes in patients with type 2 diabetes (ADVANCE) trial [30
]. The group in the intensive arm after an average follow-up of 5 years only showed a minimal reduction in the number of cases with new-onset microalbuminuria compared to the standard therapy group (23.7% vs. 25.7%) and no effect was observed in the serum creatinine values [30
]. The patients in VADT and ADVANCE study had an longer duration of diabetes (11.5 years and 8.0 years, respectively) compared to those in UKPDS study, in which all patients were newly diagnosed diabetes. This may be one of the major reasons causing the different results among these three studies and also demonstrated the importance of duration of diabetes in the incidence of CKD. In addition, the 10-year follow-up of the UKPDS study has shown that the effects of intensive glucose therapy on diabetes complications could be seen many years later [31
]. Therefore, in order to prevent or delay progression of kidney disease in diabetes, earlier intervention in the disease course seemed to be far more important.
Anemia is common among those with diabetes. Ahmed et al. [32
] reported in a large multiethnic cohort study with 65,696 diabetic patients that the prevalence of anemia was 28.0% in Asians and 33.6% in Whites. In our study, 24.7% subjects with type 2 diabetes had anemia. The patients with renal insufficiency had the lowest level of hemoglobin, and almost 60% of these subjects had anemia. It has been reported that anemia may contribute to the progression of kidney disease in diabetes [33
]. A more recent study indicated that anemia was the risk factor for both renal prognosis and survival in patients with diabetes [34
]. Our study also showed anemia being an independent associated factor with CKD in type 2 diabetic patients. Anemia in diabetic patients with CKD may mainly result from iron and erythropoietin deficiencies and hypo-responsiveness to the actions of erythropoietin [35
]. In addition, renin-angiotensin system (RAS) inhibitors could cause a reversible decrease in hemoglobin concentration in patients with diabetes and CKD. Mohanram et al. [36
] reported that long-term administration of losartan in a dose of 50 to 100 mg once daily in patients with type 2 diabetes was expected to have lower hemoglobin by about 10 g/L. In the present study, although more angiotensin converting enzyme inhibitors/angiotensin receptor blockers were prescribed in patients with anemia as compared those without anemia (data not shown), whether it is one of the reasons of anemia is worth exploring in the future.
Assessment of the association between GFR and urinary albumin excretion was scarcely reported in Chinese type 2 diabetic patients. In this study, we found that in subjects with a urinary ACR less than 90 mg/g, the average level of eGFR maintained relatively steadily but decreased rapidly in those with an urinary ACR more than 90 mg/g. This finding suggested that when patients presented with urinary ACR more than 90 mg/g, their kidney function might deteriorate very rapidly in the future. This finding was similar to that reported by Jia et al. [23
] about patients with urinary ACR of around 100 mg/g, and that study used the same GFR estimating formula as ours. However, increases in AER and decreased GFR do not always occur in parallel. A substantial proportion of normoalbuminuria may present with a reduced GFR in diabetic patients [7
]. Using eGFR alone as a screening test for CKD in diabetes is inadequate. Many patients with diabetes and CKD may have elevated or high normal GFRs, particularly in the early years after diagnosis. Therefore markers of kidney damage are required to detect early stages of CKD while eGFR alone may be able to detect CKD only at a later stage such as stage 3 or above [37
]. Our study also showed that even in patients with renal insufficiency, 32.3% subjects presented with normoalbuminuria. These results suggested albuminuria could imply, but not necessarily confirm, the onset of a decline in GFR. Therefore, both parameters (GFR and urinary albumin) should be measured in assessing the severity of CKD in diabetes patients.
There were several limitations in our study. Firstly, our study is a cross-sectional study that precluded the exploration of a causal relationship between risk factors and the development of CKD. Secondly, we used MDRD to estimate GFR rather than the measurement GFR using iothalamate clearance to determine kidney function. However, a number of studies have demonstrated that eGFR is a reliable and commonly used method to assess kidney function [11
]. Thirdly, the urinary ACR was checked only once. Because of potential variability in the UAE, ideally 3 specimens being collected within a 3 to 6 month period was recommended. Fourthly, when analyzing the association between eGFR and albuminuria, we had not taken into account the potential influence on the use of anti-hypertensive and/or lipid modifying agents, such as RAS inhibitors and statins [38
]. Finally, since all our subjects were recruited from urban area, the study data may or may not be applicable to other diabetic patients in mainland China, not to mention the generalizability of our results to other populations that need to be interpreted with caution. Interestingly, Xu et al. [40
] recently reported that among Chinese type 2 diabetic patients living in rural area, the rate of kidney diseases (including albuminuria) was 41.3%, which is similar if not even more prevalent than our figure being derived from urban area.
In conclusion, our study demonstrated that CKD and albuminuria were highly prevalent in the Chinese population with type 2 diabetes. Hypertension, duration of diabetes and anemia were the main risk factors associated with CKD. Screening programs using a combination of eGFR and the urinary ACR are necessary, especially among those diabetic patients at the middle or late phase of CKD, while urinary ACR only can be accepted for early cases with short history of disease. Strategies to prevent or delay progression of kidney disease in diabetes should be carried out at the early stages of CKD.