Skip to main content

Table 3 Analyses of blood glucose (BG) levels and insulin therapy across groups with different severities of sepsis

From: The impact of the severity of sepsis on the risk of hypoglycaemia and glycaemic variability

BG and insulin analyses

Sepsis n = 47

Severe sepsis n = 83

Septic shock n = 61

p valuesa

Mean BG level during sepsis (mg/dl; mean/SD)

132.0 (21.4)

142.4 (25.7)

140.2 (21.4)

0.0516b

Median BG level during sepsis (mg/dl; median/IQR)

133.7 (113.5 to 142.1)

142.1 (122.5 to 158.3)

133.5 (124.0 to 161.3)

0.0458 c

Mean morning BG level during sepsis (mg/dl; mean/SD)

124.4 (17.5)

131.0 (± 32.5)

135.1 (54.4)

0.7227b

Median SD during sepsis (mg/dl; median/IQR)

29.5 (19.2 to 35.9)

38.5 (26.0 to 52.1)

31.1 (25.2 to 46.9)

0.0090 c

Median morning BG level during sepsis (mg/dl; median/IQR)

126.3 (113.0 to 136.0)

122.5 (107.2 to 152.0)

121.8 (109.0 to 139.5)

0.9552c

Median number of BG measurements per day during sepsis (median/IQR)

7.0 (5.6 to 8.5)

7.0 (5.5 to 8.5)

8.7 (7.6 to 9.7)

0.0001 c

Absolute number of critical hypoglycaemic episodes *

1

6

7

 

Rate of critical hypoglycaemia per 100 hours of IIT (median/IQR)

0.17 (single episode)

0.81 (0.48 to 1.01)

0.37 (0.16 to 0.58)

0.3278c

Median time interval of BG measurements within 24 hour prior to the critical hypoglycaemic episode (hours; median/IQR)

4.1 (3.6 to 5.2)

2.0 (1.5 to 2.3)

2.5 (1.5 to 2.8)

0.1715c

Median time interval from the last BG measurement before to the critical hypoglycaemic episode itself (hours; median/IQR)

3.9 (single episode)

1.4 (0.7 to 3.1)

2.4 (1.3 to 4.3)

0.5619c

  1. * The median number of episodes per patients was 1 in each category.
  2. a Significance level p < 0.05; p values for comparisons between the different sepsis groups were calculated using ANOVA (b) and Kruskal–Wallis analysis of variance (c), as appropriate.
  3. IQR = interquartile range; IIT = intensive insulin therapy; SD = standard deviation.