Table 1. Characteristics of studies included in this analysis.
| Trial | Number(sex) | Age at enrollment(years) | Country(follow-up time) | Design | A1C variability | Mean A1C(%)SD | Renal Outcome | HR(95% CI) | Variable Adjustment | NOSScore |
| Type 1 | ||||||||||
| Kilpatrick 2008 (DCCT)[5] | 1441 (Male: 52.8%) | 27 ±7* | USA Canada (6.5 years) | Post-hoc RCT | aSD | mean:8.5$ aSD:0.73 | microalbuminuria development | 1.80 (1.37–2.42) | age,sex, diabetes duration, randomization treatment, prevention cohort and A1Cat the eligibility stage | 4/2/3 |
| Wadén 2009 (FinnDiane)[13] | 2107 (Male: 53.2%) | 36.4 ±11.8 | Finland (5.7 years) | Retrospective cohort | cSD | mean:8.5 cSD:0.78 | any increase in albuminuria or progression to ESRD | 1.92 (1.49–2.47) | sex, diabetes duration, systolic BP, lipid, ever smoking, number of A1Cmeasurements and mean A1C | 4/2/3 |
| Marcovecchio 2011[14] | 1232 (Male: 55.4%) | 9.2(5.7–11.7) # | UK (NR) | Retrospective cohort | aSD | mean:9.5 aSD:0.91 | microalbuminuria development | 1.31 (1.01–1.70)& | age at diagnosis, sex, chronologic age, mean A1C | 4/2/1 |
| Raman 2011[15] | 893 (Male: 46.9%) | 8.17±3.73 # | USA (7.0 years) | Retrospective cohort | cSD | NR | microalbuminuria development | 1.91 (1.37–2.66) | age, sex, race and mean A1c | 2/2/1 |
| Type 2 | ||||||||||
| Sugawara 2012 [6] | 812 (Male: 68.7%) | 54.9 ±10.4 | Japan (4.3 years) | Prospective cohort | aSD | mean:7.1 aSD:0.61 | microalbuminuria development | 1.35 (1.05–1.72) | age,sex, diabetes duration, systolic BP, BMI, lipid, smoking history and mean A1C | 4/2/1 |
| Hsu 2012[9] | 821 (Male: 46.1%) | 51.2±8.3# | China Taiwan (6.2 years) | Prospective cohort | aSD | mean:7.9 aSD:1.03 | microalbuminuria development | 1.19 (1.03–1.38) | age at diabetes onset, sex, education, diabetes duration, smoking status, waist circumference, lipid, BP and mean A1C | 4/2/3 |
| Rodríguez-Segade 2012 [16] | 2013 (Male: 47.7%) | 59.2±10.6 | Spain (6.6 years) | Retrospective cohort | aSD | mean:7.6 aSD:0.94 | development or progression of nephropathy | 1.37 (1.12–1.69) | age, sex, diabetes duration, BMI, retinopathy status, drug use, smoking status, lipid, cohort, number of A1Cmeasurements, A1Cat baseline and updated mean A1C | 4/2/2 |
| CV | 1.03 (1.01–1.04) | |||||||||
| Luk 2013 (The Hong Kong Diabetes Registry) [17] | 8439 (Male: 47%) | 57.6±13.2 | China Hongkong (7.2 years) | Prospective cohort | aSD | mean:7.5 aSD:0.8 | Incident chronic kidney disease: eGFR <60 ml/min/1.73 m2 | 1.16 (1.10–1.22) | age, sex, smoking, diabetes duration, BMI, waist circumference, BP, lipid, log urine ACR, estimated GFR, haemoglobin, drug use and mean A1C | 4/2/2 |
| ESRD | 1.53 (1.35–1.73) | |||||||||
| Lin 2013[18] | 3220 (Male: 51.4%) | 57.2±10.8 | China Taiwan (4.4 years) | Retrospective cohort | CV | NR | incident chronic kidney disease: eGFR<60 ml/min/1.73 m2 | 1.58 (1.19–2.11) | age, sex, lifestyle factors, hypertension, baseline drug use, hyperlipidemia, BMI, diabetes-related diseases, mean fasting blood glucose and mean A1C | 4/2/1 |
* In this trial, 16% of included individuals were adolescent (13–18 years); $, A1C at the study eligibility stage; #, Age at diagnosis; &, an erratum HR from the author; aSD, adjusted SD of A1C; cSD, crude SD of A1C; CV, the coefficient of variation of A1C; NR, not reported; BP, blood pressure; BMI, body mass index; A1C, glycated hemoglobin A1C; eGFR, estimated glomerular filtration rate; ACR, albumin-creatinine ratio; HR, hazard ratio.