cumulative time in band: glycemic level, variability and patient outcome vs. mortality

Post on 01-Jan-2016

15 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

DESCRIPTION

CUMULATIVE TIME IN BAND: GLYCEMIC LEVEL, VARIABILITY AND PATIENT OUTCOME vs. MORTALITY. Sophie Penning, Matthew Signal, Jean-Charles Preiser , Geoffrey Shaw, Aaron Le Compte , Christopher Pretty, Thomas Desaive , J. Geoffrey Chase. Introduction. Results. - PowerPoint PPT Presentation

TRANSCRIPT

CUMULATIVE TIME IN BAND: GLYCEMIC LEVEL, VARIABILITY AND PATIENT OUTCOME vs. MORTALITY

Sophie Penning, Matthew Signal, Jean-Charles Preiser, Geoffrey Shaw, Aaron Le Compte, Christopher Pretty, Thomas Desaive, J. Geoffrey Chase

Negative outcomes of dysglycemia are associated with exposure and repetition to high glucose levels, indicating that metrics of exposure might influence outcome.

Introduction

This research addresses the key questions: Is glycemic control associated with improved hospital mortality regardless of how it

is achieved? Is there a metric of glycemic control performance or level that can be assessed in

real time to adequately discriminate between patient outcomes?

Objectives

Retrospective analysis of data from two glycemic control studies.

Glycemic performance metric: cumulative time in band – cTIB

Analysis by outcome glycemic performance rather than treatment group.

Methods

0 6 12 18 24 30 36 42 48 54 600

1

2

3

4

5

6

7

8

9

10

11

12

13 cTIB = 2/7 = 29%

cTIB = 19/25 = 76%

cTIB = 39/49 = 80%

BG (m

mol

/L)

Time (hours)

cTIB = 51/61 = 84%

“cTIB is the percentage of glucose measurements within a specified band from the start to

the present time”

“Yes, and it captures both

glycemic level and variability”

Days of stay

Odd

s ra

tio

Band4.0-7.0 mmol/L 5.0-8.0 mmol/L 4.0-8.0 mmol/L

ThresholdcTIB ≥ 50%

cTIB ≥ 80%cTIB ≥ 70%

cTIB ≥ 60%

“We investigated three different glycemic bands and four cumulative time in band

thresholds”“A higher threshold level indicates less tolerance

for dysglycemia and variability”

Yes! Glycemic control is associated with improved hospital mortality regardless of how it is achieved.

Yes! there is a metric of glycemic control performance or level that can be assessed in real time to adequately discriminate between patient outcomes.

ConclusionsLived Died

cTIB ≥ t N1 N2

cTIB < t N3 N4

“A higher OR means a better

chance of survival”

SPRINT Glucontrol All

Number of patients 784 933 1717

Percentage of males 61.2 63.2 62.3

Age of patients 65.0 [52.0 - 74.0] 65.2 [51.5 - 74.0] 65.0 [51.8 - 74.0]

APACHE 2 score 18.0 [15.0 - 24.0] 15.0 [11.0 - 21.0] 17.0 [13.0 - 23.0]

Cohort BG (mmol/L) 6.2 [5.3 - 7.4] 6.9 [5.8 - 8.4] 6.6 [5.6 - 8.1]

Per-patient median BG (mmol/L) 6.3 [5.6 - 7.5] 6.9 [6.1 - 8.2] 6.6 [5.8 - 7.9]

%BG in 4-7 mmol/L 66.8 49.8 56.4

Odds of living given cTIB ≥ tOdds of living given cTIB < t

N1 N4

N2 N3Odds ratio (OR) = =

0 12 24 36 48 60 72 84 96 108 120 132 1440

1

2

3

4

5

6

7

8

9

10

11

12

13

Time (hours)

4.0-7.0 mmol/L

0 12 24 36 48 60 72 84 96 108 120 132 1440

1

2

3

4

5

6

7

8

9

10

11

12

13

Time (hours)

5.0-8.0 mmol/L

0 12 24 36 48 60 72 84 96 108 120 132 1440

1

2

3

4

5

6

7

8

9

10

11

12

13

Time (hours)

4.0-8.0 mmol/L

BG (m

mol

/L)

Results

Band“A higher cumulative time

in band is a associated with a greater chance of

survival”

top related