Guelho D, Paiva I, Batista C, Barros L, Carrilho F
Endocrinology, Diabetes and Metabolism Department Coimbra Hospital and University Centre Coimbra, Portugal -
Minerva Endocrinol. 2014 Jun;39(2):127-33.
The aim of this paper was to compare the predictive value of glycated hemoglobin (A1c) and glucose variability (GV) in the risk of hypoglycemia in patients with type 1 diabetes (T1D).
Analysis of continuous glucose monitoring performed in 130 T1D patients with a diabetes background of 17.1±8.6 years, in intensive insulin therapy (49.8±17.9 UI per day). Mean interstitial glucose (in mg/dL), GV (standard deviation of mean glucose, in mg/dL), time per day spent in hypoglycemia (interstitial glucose ≤70 mg/dL, in %), and episodes of asymptomatic or nocturnal hypoglycemia, (hypoglycemia between midnight and 8 a.m., in %), were assessed. Patients were divided into two groups: group I (N.=84) with A1C≤7.5% and group II (N.=46) with A1C>7.5%. A statistical analysis was performed using SPSS, version 21.0®.
Group I presented a significantly lower mean glucose (139.2±25.9 vs. 173.1±33.2 mg/dL, P<0.05) and GV (58.4±18.8 vs. 70.3±18.6 mg/dL, P<0.05) and more hypoglycemia time (7.65±7.04 vs. 5.35±5.64%, P<0.05). The number of patients with nocturnal hypoglycemia was not significantly different in both groups (8.7 vs. 5.8%, P>0.05). Hypoglycemia time was positively correlated with GV (r=0.23, P=0.01) and negatively with A1C and mean glucose (r=-0.23 and r=-0.36; P=0.01). In multivariate analysis, GV and mean glucose were associated with hypoglycemia time (β=0.22 and β=-0.15, P<0.01, respectively), independent of A1c, diabetes duration and insulin dose; nocturnal hypoglycemia was only associated with mean glucose (OR=0.9, P<0.05) and was associated with a 16-fold increased risk of asymptomatic hypoglycemia (OR: 16.9, P<0.01).
Patients with high HbA1c still remain at risk of hypoglycemia. Glucose variability independently predicts daily time spent in hypoglycemia. At night, hypoglycemia only correlates with mean glucose, suggesting that daily fluctuations are probably due to inadequate meals insulin coverage. The potential of GV for predicting hypoglycemia time supports the inclusion of measures of GV into a global diabetes strategy.
本文旨在比较糖化血红蛋白（A1c）和血糖变异性（GV）对1型糖尿病（T1D）患者低血糖风险的预测价值。
对130例T1D患者进行持续葡萄糖监测分析，这些患者糖尿病病程为17.1±8.6年，接受强化胰岛素治疗（每天49.8±17.9单位）。评估平均组织间液葡萄糖（单位：mg/dL）、GV（平均葡萄糖标准差，单位：mg/dL）、每天处于低血糖状态的时间（组织间液葡萄糖≤70mg/dL的时间占比）以及无症状或夜间低血糖发作次数（午夜至上午8点之间发生低血糖的次数占比）。患者分为两组：A1C≤7.5%的I组（n = 84）和A1C>7.5%的II组（n = 46）。使用SPSS 21.0®进行统计分析。
I组的平均葡萄糖水平（139.2±25.9 vs. 173.1±33.2mg/dL，P<0.05）和GV（58.4±18.8 vs. 70.3±18.6mg/dL，P<0.05）显著更低，低血糖时间更多（7.65±7.04 vs. 5.35±5.64%，P<0.05）。两组夜间低血糖患者数量无显著差异（8.7% vs. 5.8%，P>0.05）。低血糖时间与GV呈正相关（r = 0.23，P = 0.01），与A1C和平均葡萄糖呈负相关（r = -0.23和r = -0.36；P = 0.01）。多变量分析显示，GV和平均葡萄糖与低血糖时间相关（β = 0.22和β = -0.15，P均<0.01），独立于A1c、糖尿病病程和胰岛素剂量；夜间低血糖仅与平均葡萄糖相关（OR = 0.9，P<0.05），且与无症状低血糖风险增加16倍相关（OR：16.9，P<0.01）。
高HbA1c患者仍有低血糖风险。血糖变异性可独立预测每日低血糖时间。夜间，低血糖仅与平均葡萄糖相关，提示每日波动可能是由于餐时胰岛素覆盖不足。GV预测低血糖时间的潜力支持将GV测量纳入全球糖尿病管理策略。
