Hospital Pharmacy - February 2020 - 29

29

Bredhold et al
Table 1. Baseline Characteristics.

Age, years, mean (SD)
Gender (% males)
Race (% Caucasian)
Weight in kg, mean (SD)
Admitting team, n (%)
Medical critical care
Trauma/surgical
Cardiovascular
Mechanical ventilation, n (%)
Corticosteroids, n (%)
Midodrine, n (%)
Renal replacement therapy, n (%)
MAP at vasopressin initiation, mm Hg, mean (SD)
Maximum norepinephrine dose, µg/min, mean (SD)
Time from maximum norepinephrine to d/c, mean in hours (SD)
Additional vasopressor(s), n (%)
At initiation of vasoactive therapy, mean (SD)
APACHE II
SOFA
At discontinuation of first agent, mean (SD)
APACHE II
SOFA

NDF group (n = 34)

VDF group (n = 52)

P value

55 (16)
50
85.3
80.95 (20.2)

59 (12.6)
63.5
90.4
86.02 (29.4)

.192
.22
.37
.13
.10

26 (76.5)
5 (14.7)
3 (8.8)
21 (61.8)
19 (55.9)
2 (5.9)
6 (17.6)
72.91 (18)
34.3 (22)
33.5 (73.3)
5 (14.7)

42 (80.7)
9 (17.3)
0 (0)
47 (90.4)
33 (63.5)
2 (3.8)
8 (15.4)
67.81 (14.2)
41.3 (23.1)
70 (27.7)
15 (28.8)

.001
.48
.65
.90
.147
.135
.001
.13

19 (6)
7.8 (2.9)

22.2 (5.5)
8.7 (2.1)

.016
.120

12.8 (4.75)
6.2 (2.51)

15.9 (5.9)
7.4 (2.1)

.014
.025

Note. NDF = norepinephrine discontinued first group; VDF = vasopressin discontinued first group; MAP = mean arterial pressure.

Table 2. Results.
NDF group (n = 34) VDF group (n = 52) P value
ICU length of stay, days, mean (SD)
Dose of norepinephrine at which vasopressin was initiated, µg/min, mean (SD)
Dose of norepinephrine at which vasopressin was discontinued, µg/min, mean (SD)
Hypotension within 24 hours of discontinuation of first agent, n (%)

9.4 (6)
27 (22)
-
17 (50)

11.1 (8)
31.1 (18)
11.9 (14)
31 (60)

.313
.11
-
.38

Note. NDF = norepinephrine discontinued first group; VDF = vasopressin discontinued first group; ICU = intensive care unit.

Table 3. Regression Coefficients When Predicting for ICU
Length of Stay.
Predictor
Constant
Vasopressor (first)
Mechanical ventilation
Renal replacement therapy
Steroid use
APACHE II

B (SE)

β

P value

1.58 (0.26)
−0.12 (0.15)
0.72 (0.19)
0.53 (0.19)
−0.25 (0.14)
0.004 (0.01)

-
−0.08
0.41
0.28
−0.17
0.03

<.001
.422
<.001
.006
.076
.769

Note. ICU = intensive care unit.

performed to try to identify variables that may predict for a
longer ICU length of stay (see Table 3). This linear regression underwent log transformation due to the presence of
outliers. When controlling for VDF, we analyzed mechanical
ventilation use, renal replacement therapy, steroid use, and

APACHE II score at initiation of vasoactive agents as potential predictors of ICU length of stay. The results of the linear
regression showed that the only predictors of ICU length of
stay were mechanical ventilation use (P < .001) and renal
replacement therapy use (P = .006), which are routinely
found to be predictors of an increased ICU length of stay.
The regression showed that neither drug to be utilized last
would predict for ICU length of stay.

Discussion
In adults who are recovering from septic shock, we found
no statistical difference in the sequence of norepinephrine
and vasopressin discontinuation on ICU length of stay,
although we did not meet power. While the difference of
around 1.7 days may seem clinically important, more of
the patients in the VDF group had mechanical ventilation
use, which was shown to be a predictor of a longer ICU
Hospital Pharmacy - February 2020

Table of Contents for the Digital Edition of Hospital Pharmacy - February 2020
