Regarding a maximum of 311 customers subscribed to the study, 27 was excluded because they came across no less than one of your exclusion standards revealed in earlier times ten . The remaining 284 patients (123 people) had been as part of the next analyses. Their properties are given into the Table step 1. All round average period of brand new patients is 68 decades, their average Bmi twenty five.0 kg/yards 2 , as well as their average duration of all forms of diabetes 14 ages. The latest median CPR quantity try step one.eight ng/mL additionally the average Cv are filed given that twenty seven.8.

We first constructed a scatter plot of CPR versus CV to visualize the potential relationship between these two parameters. For a better understanding of the results, Fig. 1 shows the women seeking women for sex log-linear relationship between CPR and CV. The prediction curve, constructed using this scatter plot, showed a marked increase in CV with low CPR and relatively low CV with high CPR. CV had significant negative correlation with CPR (? = ? 0.39, P < 0.0001). The prediction formula was estimated to be the following Eq. (1).

CV also showed significant negative correlation with BMI, FPG, and eGFR (Table 2). In addition, CV had significant positive correlation with the duration of diabetes. The CV in patients on insulin was significantly greater than that of non-insulin patients. The CV in patients treated with an alpha-glucosidase inhibitor (?-GI) or dipeptidyl peptidase-4 (DPP-4) inhibitor was significantly smaller than in patients not treated with these drugs. No relationships were identified between CV and other factors, such as HbA1c, age, and other antidiabetic treatments. We assessed the multicollinearity in the variables shown to be associated with CV using Spearman’s correlation analysis (Supplementary Table S1). There was a relatively strong correlation between CPR and BMI, with a correlation coefficient of 0.51. The absolute values of the correlation coefficient for other variables were lower. The variance inflation factor (VIF) for these parameters was less than 2.0 (Supplementary Table S2). Thus, there were no indications of multicollinearity between these variables. Low CPR was an independent predictive marker for high CV, according to multiple regression analysis (regression coefficient (?): ? 0.285, 95% confidence interval [CI] ? 0.234 to ? 0.092, P < 0.0001). Insulin use, non-use of an ?-GI or DPP-4 inhibitor, and low eGFR were also predictors of high CV (Table 3). We also performed the calculation after correcting CPR for the concomitant plasma glucose level using CPR index, calculated using the formula: 100 ? fasting CPR (ng/mL)/FPG (mg/dL) 21 . The findings were similar to those obtained using the fasting CPR (data not shown).

A log-linear relationship between CPR and LBGI was observed. LBGI showed a significant negative correlation with CPR (? = ? 0.25, P < 0.0001) as well as the relationship between CPR and CV. As shown in Supplementary Table S3, LBGI also exhibited a significant negative correlation with BMI, FPG, HbA1c, and eGFR. The LBGI in patients treated with insulin was significantly greater than that in patients not treated with insulin. No relationships were identified between the LBGI and other factors. According to multiple regression analysis, low CPR was an independent predictive marker for a high LBGI (?: ? 0.163, 95% CI ? 0.327 to ? 0.043, P = 0.0107). Insulin use, low FPG, low HbA1c, and low eGFR were also predictors of a high LBGI (Supplementary Table S4). In contrast, a HBGI was not significantly correlated with CPR (? = ? 0.11, P = 0.0742).

The 284 patients were allocated to three subgroups: low CPR (CPR < 1 ng/mL, n = 62), moderate CPR (1 ng/mL ? CPR < 2 ng/mL, n = 113), and high CPR (CPR ? 2 ng/mL, n = 109). The biochemical and anthropometric characteristics of each group are shown in Table 4. Age, BMI, the duration of diabetes, the type of antidiabetic treatment, and FPG were associated with the CPR concentration. The indices of GV (CV, SD, and MAGE), and the key CGM indices (LBGI, HBGI, TBR, TIR, and TAR), were also associated with the CPR concentration. The mean glucose profiles of the patients in each of the three CPR groups are shown in Fig. 2. In summary, the GV appeared to be more unstable in the low CPR group than in the moderate and high CPR groups. Supplementary Figure S3 shows the CV distribution among the three subgroups. The number of patients with CV ? 36%, defined as unstable GV, was significantly higher in the low CPR subgroup than in the medium and high CPR subgroups (Table 4).