Diabetic kidney disease (DKD) is one of the most common complications of diabetes (1), characterised by albuminuria, reduced glomerular filtration rate (GFR) or both (2). The estimated GFR (eGFR) was used to determine kidney function stage and was calculated based on serum concentration of creatinine, cystatin C or both using given formulas (3–6). A decline in eGFR is represented as a one-directional process, and once initiated, it progresses to end-stage kidney disease (ESKD), albeit at widely differing individual rates; the abnormal urinary albumin excretion can regress, stay the same or progress (3). The global percentage of prevalent ESKD patients with diabetes increased from 19.0% in 2000 to 29.7% in 2015, with the most rapid increase rates and highest average rates (44.1%) observed in the Western Pacific Region (7). According to some recent years’ reports, diabetes was the third cause (27.1%) for CKD onset, the leading cause (40.5%) for CKD progression and the most common cause (45%) for ESKD (8, 9). In the general population, the overall prevalence of chronic kidney disease (CKD) ranged from 9.9% to 16.8% in the Zhejiang and Henan provinces, and the prevalence of DKD was less than 5% (10–12). In patients with type 2 diabetes mellitus (T2DM), the prevalence of CKD increased greatly and ranged from 27.1% to 52.3% in different Chinese studies (10, 13–17).

To prevent the onset and progress of CKD in diabetes, the modifiable risk factors, including tobacco and alcohol use, physical activity, stress, body mass index (BMI), haemoglobin A1C (HbA_1c) level, blood pressure (BP), blood lipids, GFR and albuminuria, need to be controlled (18). The lower prevalence of albuminuria was observed over time (from 20.8% in 1988–1994 to 15.9% in 2009–2014) among United States adults with diabetes, and was attributable to the lowering of mean blood glucose, blood pressure and lipid levels with a higher rate of prescribed glucose and lipid lowering medications and renin-angiotensin-aldosterone system (RAAS) inhibitors (19).The mean decrease of HbA_1c(from 8.1% to 7.6%), systolic BP (SBP, from136.3 to 130.1 mmHg), diastolic BP (DBP, from76.2 to 68.9 mmHg), low-density lipoprotein cholesterol (LDL-C, from3.55 to 2.66 mmol/L) and triglycerides (TG, from2.25 to 1.61 mmol/L) was considerable; while the usage of RAAS inhibitors (from 24.4% to 56.2%) and lipid lowering medications (from 17.0% to 51.8%) increased significantly (19).The HbA_1c, BP and LDL-C/TG control (i.e. ABC control) in Chinese T2DM patients with CKD were still poor in recent years reports, with a mean HbA_1c of 8.3–9.2%, SBP ≥ 135 mmHg, DBP ≥ 80 mmHg, LDL-C ≥ 3.1 mmol/L and TG ≥ 2.2 mmol/L (13, 15, 17). Given the projected increase in the number of people diagnosed with diabetes and complicated with CKD, poor ABC control will induce a significant increase in the number of people with ESKD over the next few decades and will impose a heavy burden on our healthcare system.

More data from different regions are still needed to understand the prevalence and the metabolic control of CKD in Chinese patients with T2DM. This study aimed to evaluate the prevalence of CKD in Chinese adults with T2DM in primary care, and the association of ABC control at follow up (FU) with the progress and regression of CKD. The results would provide us important epidemiological information and be useful for understanding the present situation and long-term direction for diabetes care.

The prevalence of CKD in this study (29.6%) was similar to that in a Hong Kong (29.7%) and Jiangsu (31.0%) study, with relatively different prevalence of impaired eGFR (5.8%, 11.6% and 6.5%) and albuminuria (27.1%, 23.6% and 28.9%) (14, 24). The prevalence of albuminuria was often higher than that of reduced eGFR in T2DM patients (13, 14, 19, 24, 25), however, a pretty high prevalence of albuminuria (45.3%) with much lower prevalence of reduced eGFR (6.3%) was also reported in central Chinese urban population with T2DM (12). All the variables associated with CKD risk, including diabetes duration, HbA_1c, BP and TG level etc, as well different equation used to calculate eGFR, would contribute to the different CKD prevalence (26). The similar CKD prevalence in this study and in Hongkong and Jiangsu study might arise from similar baseline ABC control, with median/mean HbA_1c of 7.6%, 7.2%, and 7.56%, SBP of 130, 134 and 129 mmHg, DBP of 77, 74 and 77 mmHg, LDL-C of 2.77, 2.3 and 2.87mmol/L, and TG of 1.63, 1.6 and 1.47 mmol/L (11,39). The prevalence of CKD would increase to around 53% in polyclinics or hospitalized patients with diabetes duration ≥10 years and worse ABC control (3, 17). A relentless focus on improving ABC control would be important for reducing CKD prevalence.

Patients in this study were involved in a chronic shared care model since 2016, receiving comprehensive risk factor assessments, screening for diabetic complications, monitoring ABC control and adjusting treatment regularly. Patients with T2DM enrolled in this kind of chronic, integrated or shared care models were more likely to receive care that was patient centred and collaborative, compared with patients who received routine care regardless of the setting, and would be beneficial to improving ABC control and delaying the complications of diabetes (27–30). Thus we found SBP, DBP, HbA_1c, LDL-C level decreased at FU, and the proportion of patients reaching targets of BP ≤ 130/80 mmHg, HbA_1c<7.5%, LDL-C<2.60 mmol/L increased from baseline to FU and LV. The increased usage of insulin, RAS inhibitors, SGLT2i and lipid lowering for low, moderate and high/very high risk group indicated intensified and personalized treatment adjustment. Compared to baseline value, the median/mean HbA_1c (7.4%), SBP (127 mmHg), DBP (75 mmHg), LDL-C (2.64 mmol/L) and TG (1.60 mmol/L) at LV was closer to that in Hongkong population with T2DM enrolled in a management programme implemented since 2009, although better in low risk group and worse in high/very high risk group (14). However, an increase in the prevalence of CKD (38.6%), impaired eGFR (8.0%) and albuminuria (36.4%) was still found after 30 (21, 38) months follow up. We wonder if there were any space for improvement in respect of modifiable risk factors discussed here, i.e., ABC control.

The increased CKD prevalence in this study mainly resulted from the increased prevalence of albuminuria and the 21.8% of patients progressing to higher CKD outcome risk. The significant longer diabetes duration (10.0 vs 7.7 years), lower rate of HbA1c <7.5% at FU (39.6% vs 52.2%) and LV (40.1% vs 55.1%), as well TG< 1.7mmol/L at LV (46.8% vs 53.6%) in progress group than in regress group might lead to CKD outcome progress instead of regression. After multivariable adjustment, the mean and last BP ≤ 130/80 mmHg, HbA_1c<7.5% and TG< 1.7 mmol/L was associated with CKD risk progress in different models, indicating the importance of more aggressive ABC control in clinical practice. In the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, 3.7 years of intensive glycemic control targeting HbA_1c<6.0% reduced the incidence of the composite kidney outcome in the follow-on 7.7 years and primarily driven by a reduction in incident macroalbuminuria, indicating the importance of long-term intensive glycemic control (31). Aiming to a 30% decrease in UACR during a 2-year baseline period would lead slow progression of CKD and 17% risk reduction of ESKD (32). In a study focus on effects of BP on renal outcomes, a follow-up SBP > 130 mmHg was associated with an increase hazard ration (2.33) for renal outcomes with referent to SBP ≤ 130 mmHg (33). Among a subset of ACCORD-BP trial, UACR was 30% lower in the intensive BP control group (SBP<120 mm Hg) at year 2 compared to a less intensive BP target (SBP <140 mm Hg) (34). Thus, more efforts should be paid on improving ABC control at FU although it might be more difficult in daily clinical practice than in prospective intervention study. Compared to United states and Singapore study, the usage of RAAS inhibitors (56.2%, 59.5% vs 40.5%) and lipid lowering medications (51.8%, 81.2% vs 51.2%) in this study might also have space to increase (19, 26).

There were several limitations in this study. First, it was uncertain that the CKD in our subjects was due to diabetic nephropathy, since they might also have abnormal kidney structures or other diseases. However, the term ‘diabetic nephropathy’ has been considered outdated and is avoided in clinical practice with CKD recommendations (19). Second, we used creatinine-based GFR estimates, which have been criticised for lack of accuracy in patients with diabetes, especially within the normal or high range of GFR (6). Nevertheless, creatinine-based eGFR is still widely adopted in current clinical practice. Third, the dose of ACEI/ARB was not recorded and analysed in this study, since titrate to the max tolerated doses or even more aggressive doses would effectively reduce urinary protein excretion rates, and the efficacy would be enhanced in terms of better BP control and better HbA_1c control (35, 36). Therapeutic inertia (TI) in initiation or intensification of ACEI/ARB treatment might exist, since a 40.3% prevalence of TI was reported in Hongkong study (37), it probably also existed in our clinical practice and need further investigation.

In conclusion, this study provided us information about the prevalence of CKD and metabolic risk factors management at baseline and follows up, as well the association of ABC control and CKD outcome risk. Patients with T2DM had intensified and personalized treatment adjustment in this share care management model. Since 21.8% of patients progressing to higher CKD outcome risk, attention should be paid on long term and better ABC control.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

This study was conducted in accordance with the Declaration of Helsinki and approved by the ethics committee of Beijing Ruijing Diabetes Hospital.

Conception and design of the work: LA; Data collection: LA, QY, HT, XL, DW, QT, HX, YH, XZ, and SZ Supervision: LA, YL and JL; Analysis and interpretation of the data: LA, YL and JL; Statistical analysis: LA Drafting the manuscript: LA; Critical revision of the manuscript: LA, YL and JL; Approval of the final manuscript: all authors.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.