Skip to main content

Table 2 Baseline data from Cohort 2, comprising 80 patients treated according to Specialised Relative Insulin Nutrition Tables protocol

From: Untangling glycaemia and mortality in critical care

 

Cohort 2

Survivors

Non-survivors

p Value

Number of subjects

80

63

17

 

Age, years

66 [54–75]

65 [49–74]

73 [57–76]

0.50

Sex, M/F

51/29

41/22

10/7

0.78

APACHE II score

21 [17–27]

21 [16–27]

21 [17–28]

0.60

First-day SOFA score

7 [4–8]

6 [4–8]

8 [6–8]

0.11

 Cardiac

3 [1–4]

3 [1–4]

4 [2–4]

 

 Pulmonary

3 [2–4]

3 [2–4]

4 [2–4]

 

 Hepatic

0 [0–0]

0 [0–0]

0 [0–1]

 

 Renal

0 [0–0]

0 [0–0]

0 [0–0]

 

 Coagulation

0 [0–1]

0 [0–1]

0 [0–1]

 

ICU length of stay, h

180 [136–371]

214 [142–405]

142 [108–159]

<0.01

SPRINT duration, h

155 [109–301]

161 [126–332]

110 [102–151]

0.01

Diabetes mellitus type 1/type 2, % of total

5/10 (15)

4/8 (12)

1/2 (3)

1.00

Cohort BG, mmol/L

5.7 [5.0–6.7]

5.8 [5.1–6.8]

5.6 [4.9–6.5]

<0.01a

Per-patient BG, mmol/L

5.8 [5.3–6.2]

5.9 [5.4–6.2]

5.4 [5.2–6.0]

0.11

Per-patient % BG, 4.4–8 mmol/L, % of all BG

84.7 [73.6–91.7] (81.3)

84.7 [74.0–91.7] (81.5)

83.3 [71.5–94.8] (80.7)

0.98

Per-patient % BG <4 mmol/L, % of all BG

1.4 [0.0–2.8] (2.5)

1.4 [0.0–2.8] (2.0)

1.4 [0.0–5.6] (4.2)

0.31

Number of patients with BG <2.2 mmol/L

0

0

0

 

BG measurements per day

15.1 [13.8–16.3]

15.1 [13.4–16.7]

15.4 [14.7–15.9]

0.66

Per-patient median insulin, U/h

3 [2–3]

3 [3–3]

3 [2–3]

0.40

Per-patient median feed, g/h

3.3 [1.9–4.8]

3.5 [1.9–4.6]

2.8 [2.1–5.6]

0.80

  1. Data are given as median [IQR] unless otherwise indicated. p Values were computed using Fisher’s exact test and rank-sum tests where appropriate
  2. Abbreviations: APACHE Acute Physiology and Chronic Health Evaluation, BG Blood glucose, ICU Intensive care unit, SOFA Sequential Organ Failure Assessment, SPRINT Specialised Relative Insulin Nutrition Tables protocol
  3. aEquivalence, as explained in ‘Analyses and statistics’ subsection under ‘Methods’