Prior Use of Calcium Channel Blockers Is Associated

With Decreased Mortality in Critically Ill Patients With

Sepsis: A Prospective Observational Study
Maryse A. Wiewel, MD1,2; Lonneke A. van Vught, MD1,2; Brendon P. Scicluna, PhD1,2;
Arie J. Hoogendijk, PhD1,2; Jos F. Frencken, MD6,7; Aeilko H. Zwinderman, PhD3;
Janneke Horn, MD, PhD4; Olaf L. Cremer, MD, PhD6; Marc J. Bonten, MD, PhD7,8;
Marcus J. Schultz, MD, PhD4; Tom van der Poll, MD, PhD1,2,5; on behalf of the Molecular
Diagnosis and Risk Stratification of Sepsis (MARS) Consortium

Objectives: Experimental studies suggest that calcium channel nel blockers and the outcome of patients admitted to the ICU with
blockers can improve sepsis outcome. The aim of this study was sepsis.
to determine the association between prior use of calcium chan- Design: A prospective observational study.
Setting: The ICUs of two tertiary care hospitals in the Netherlands.
1
Center for Experimental and Molecular Medicine, Academic Medical Patients: In total, 1,060 consecutive patients admitted with sepsis
Center, University of Amsterdam, Amsterdam, the Netherlands.
were analyzed, 18.6% of whom used calcium channel blockers.
2
The Center for Infection and Immunity Amsterdam, Academic Medical
Center, University of Amsterdam, Amsterdam, the Netherlands. Interventions: None.
3
Department of Clinical Epidemiology, Bioinformatics, and Biostatistics, Measurements and Main Results: Considering large baseline dif-
Academic Medical Center, University of Amsterdam, Amsterdam, the ferences between calcium channel blocker users and nonusers, a
Netherlands. propensity score matched cohort was constructed to account for
4
Department of Intensive Care, Academic Medical Center, University of differential likelihoods of receiving calcium channel blockers. Fif-
Amsterdam, Amsterdam, the Netherlands.
teen plasma biomarkers providing insight in key host responses
5
Division of Infectious Diseases, Academic Medical Center, University of
Amsterdam, Amsterdam, the Netherlands. implicated in sepsis pathogenesis were measured during the first 4
6
Department of Intensive Care Medicine, University Medical Center days after admission. Severity of illness over the first 24 hours, sites
Utrecht, Utrecht, the Netherlands. of infection and causative pathogens were similar in both groups.
7
Julius Center for Health Sciences and Primary Care, University Medical Prior use of calcium channel blockers was associated with improved
Center Utrecht, Utrecht, the Netherlands.
30-day survival in the propensity-matched cohort (20.2% vs 32.9%
8
Department of Medical Microbiology, University Medical Center Utrecht,
Utrecht, the Netherlands. in non-calcium channel blockers users; p = 0.009) and in multivari-
A complete list of members of the Molecular Diagnosis and Risk Stratifica- ate analysis (odds ratio, 0.48; 95% CI, 0.31–0.74; p = 0.0007).
tion of Sepsis Consortium appears in the Acknowledgements. Prior calcium channel blocker use was not associated with changes
Supplemental digital content is available for this article. Direct URL citations in the plasma levels of host biomarkers indicative of activation of
appear in the printed text and are provided in the HTML and PDF versions the cytokine network, the vascular endothelium and the coagulation
of this article on the journal’s website (http://journals.lww.com/ccmjournal).
system, with the exception of antithrombin levels, which were less
This research was performed within the framework of the Center for Trans-
lational Molecular Medicine (CTMM) (www.ctmm.nl), project Molecular decreased in calcium channel blocker users.
Diagnosis and Risk Stratification of Sepsis (grant 04I-201). The sponsor Conclusions: Prior calcium channel blocker use is associated with
CTMM was not involved in the design and conduction of the study; nor
reduced mortality in patients following ICU admission with sepsis.
was the sponsor involved in collection, management, analysis, and inter-
pretation of the data or preparation, review or approval of the article. Deci- (Crit Care Med 2017; 45:454–463)
sion to submit the article was not dependent on the sponsor. Key Words: calcium channel blockers; intensive care units;
Dr. Cremer’s institution received funding from Center for Translational mortality; sepsis
Molecular Medicine, project Molecular Diagnosis and Risk Stratification
of Sepsis. The remaining authors have disclosed that they do not have any
potential conflicts of interest.
For information regarding this article, E-mail: m.a.wiewel@amc.uva.nl

S
Copyright © 2017 by the Society of Critical Care Medicine and Wolters epsis is the consequence of a dysregulated host response
Kluwer Health, Inc. All Rights Reserved. to an infection, resulting in tissue injury and organ mal-
DOI: 10.1097/CCM.0000000000002236 function (1). Sepsis is a frequent indication for ICU

454 www.ccmjournal.org March 2017 • Volume 45 • Number 3

Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

admission and contributes significantly to morbidity and
mortality (2, 3). The outcome and host response in sepsis are
influenced by factors associated with the pathogen, such as
microbial load and virulence, and the host, including genetic
composition, age, and comorbidities (1).
Cardiovascular disease is among the most common comor-
bid diseases in sepsis patients (3, 4). In patients with cardiovas-
cular disease, calcium channel blockers (CCBs) are frequently
used to lower diastolic blood pressure, improve myocardial oxy-
gen supply, and to slow cardiac conduction (5, 6). CCBs block
L-type voltage gated myocytes, thereby promoting relaxation
of cardiac and smooth muscle cells by inhibition of calcium
influx through calcium channels and calcium release from
intracellular stores (7, 8). Calcium functions as an important
second messenger in cells, regulating many cellular processes
known to be disturbed in sepsis, including muscle cell con-
tractility, glycogen and protein turnover, and vascular smooth
muscle tone (9). Sepsis is associated with increased Ca2+ levels
in many cell types (9), and a sustained increase in intracellu-
lar Ca2+ levels can cause cytotoxicity and cell death through
multiple mechanisms, including mitochondrial dysfunc-
tion, nuclear injury, disruption of the cytoskeleton, increased
nitric oxide and proinflammatory cytokine production, and
enhanced apoptosis (9–11). Animal studies have suggested that
CCBs can improve mortality induced by endotoxemia or sepsis
by restoring intracellular calcium homeostasis (9, 12–15).
In spite of abundant literature on the role of intracellular
Ca2+ in cell function and the effect of CCBs in experimental
sepsis models (9–11), knowledge on the association between
chronic CCB use and sepsis outcome in humans is limited.
The aim of the present study was to determine the relationship
between prior use of CCBs and the presentation and outcome
of patients admitted to the ICU with sepsis. In addition, we
sought to investigate the association between prior CCB use
and the host response to sepsis by measuring a set of 15 host
biomarkers affording insight into key host responses impli-
cated in the pathogenesis of sepsis.
METHODS
Study Design, Patients and Definitions
This study was conducted as part of the “Molecular Diagno-
sis and Risk Stratification of Sepsis” (MARS) project, a pro-
spective observational study in the mixed ICUs of two tertiary
teaching hospitals (Academic Medical Center [AMC] in
Amsterdam and University Medical Center Utrecht [UMCU])
in the Netherlands between January 2011 and January 2014
(16, 17). Dedicated and trained physicians prospectively col-
lected the following data from all patients: demographics,
comorbidities, chronic medication use, ICU admission char-
acteristics (including the Acute Physiology and Chronic Health
Evaluation [APACHE] IV score), and daily physiologic mea-
surements, severity scores (including Sequential Organ Failure
Assessment [SOFA] scores), antibiotic use, and culture results.
Organ failure was defined as a score of three or greater on the
SOFA score, except for cardiovascular failure for which a score
Critical Care Medicine www.ccmjournal.org 455
Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Clinical Investigations
of one or more was used (18). Shock was defined as the use
of vasopressors (noradrenaline) for hypotension in a dose of
0.1 mcg/kg/min during at least 50% of the ICU day. Patients
were assessed daily for the presence of acute kidney injury
(AKI) and acute lung injury (ALI) using strict preset crite-
ria (19, 20). The plausibility of infection was post hoc scored
based on all available evidence and classified on a 4-point scale
(none, possible, probable, or definite) according to Center for
Disease Control and Prevention (21) and International Sepsis
Forum consensus definitions (22), as described in detail pre-
viously (16). Daily (at admission and at 6 am thereafter) left-
over plasma (obtained from blood drawn for patient care) was
stored within 4 hours at –80 °C. The Medical Ethical Commit-
tees of both study centers gave approval for an opt-out consent
method (institutional review board number, 10-056C) (16,
17). The Municipal Personal Records Database was consulted
to determine survival up to 1 year after ICU admission.
For the current analysis, we selected all patients included
in the MARS-study between January 2011 and July 2013 with
sepsis, diagnosed within 24 hours after admission, defined
as a “definite or probable” infection (16) combined with at
least one of general, inflammatory, hemodynamic, organ
dysfunction, or tissue perfusion parameters derived from
the 2001 International Sepsis Definitions Conference (23).
Readmissions and patients transferred from another ICU were
excluded, except for patients referred to one of the study cen-
ters on the day of admission.
Biomarker Measurements
All measurements were done in ethylenediaminetetraacetic
acid anticoagulated plasma obtained at admission (day 0) and
days 2 and 4. Assays are described in the online supplement
(Supplemental Digital Content 1, http://links.lww.com/CCM/
C300). Normal biomarker values were acquired from 27 age-
and gender-matched healthy volunteers, from whom written
informed consent was obtained.
Statistical Analysis
Data-analyses were performed in R (v3.1.1). Baseline charac-
teristics of study groups were compared with chi-square test for
categorical variables and t test or Wilcoxon signed rank test for
continuous variables. Mixed-effects models were used to ana-
lyze repeated measurements. Propensity score matching and
multiple binary logistic regression analyses were performed
as described in the online supplement (Supplemental Digital
Content 1, http://links.lww.com/CCM/C300). To account for
random effects of propensity matching, paired t test, Wilcoxon
signed rank test, McNemar test, and stratified log-rank test
were used in the propensity-matched cohort. p values below
0.05 were considered statistically significant.
RESULTS
Study Population
A total of 6,994 admissions were included in the MARS-study
from January 2011 until July 2013, of which 1,483 involved
Wiewel et al

an admission diagnosis of sepsis. One hundred twenty-
nine transfers from other ICUs and 250 readmissions were
excluded. Prior use of medication could not be traced in 44
cases. As a result, 1,060 patients were included for analysis, of
whom 197 patients (18.6%) used CCBs (Table 1). Amlodipine
was the most common CCB prescribed (54.8%), followed by
nifedipine (27.4%). Patients who used CCBs were older and
had a higher body mass index, whereas the gender ratio did
not differ between the two groups. CCB-users more frequently
had diabetes, hypertension, cerebrovascular disease, chronic
renal insufficiency, chronic obstructive pulmonary disease,
immune deficiency and peripheral vascular disease, and less
often metastatic malignancy. In accordance with these differ-
ences in comorbid conditions, CCB-users were more likely to
use a variety of other types of chronic medication, including
angiotensin converting enzyme (ACE) inhibitors, angiotensin
II receptor blockers (ARBs), beta-blockers, statins, insulin, oral
antidiabetic drugs, and corticosteroids. Continuation of CCBs
during ICU admission was dependent on the judgment of the
medical team; at least one dose in the first 4 days of admis-
sion was received by 10.7% of CCB-users, whereas during the
complete ICU admission, 21.8% of prior CCB-users received
a CCB.
Prior CCB Use and Severity and Cause of Sepsis
Considering the large differences between users and nonusers
of CCBs at baseline, we constructed propensity score matched
cohorts to correct for these dissimilarities (24). Twelve
patients (1.1%) could not be assigned a propensity score
due to missing data. In total, 173 of 197 CCB-users could be
matched to nonusers (Table 1) (Supplemental Fig. 1, Supple-
mental Digital Content 1, http://links.lww.com/CCM/C300).
The propensity score matched groups were similar with
regard to demographic distribution, comorbidity, and chronic
medication. The severity of disease at admission was similar
between CCB-users and nonusers, as reflected by APACHE IV

TABLE 1. Baseline Characteristics of Sepsis Patients Admitted to the ICU Stratified

According to Prior Use of Calcium Channel Blockers
Unmatched Cohort Propensity-Matched Cohort

CCBs No CCBs CCBs No CCBs

Variables (n = 197) (n = 863) p (n = 173) (n = 173) p

Demographics
  Age (yr), mean (sd) 65.9 (11.8) 60.5 (14.9) < 0.0001 65.4 (11.9) 65.8 (12) 0.77
  Gender, male (%) 118 (59.9) 522 (60.5) 0.93 103 (59.5) 101 (58.4) 0.91
  Race, white (%) 171 (86.8) 763 (88.4) 0.50 153 (88.4) 157 (90.8) 0.60
  Body mass index, kg/m , mean (sd)
2
27.3 (6.4) 25.7 (6.1) 0.002 27.2 (6.3) 27 (7.3) 0.91
  Admission type, medical (%) 151 (76.6) 633 (73.3) 0.15 130 (75.1) 131 (75.7) 0.51
Comorbidities
  Charlson score, median (IQR)a 2 (1–3) 1 (0–2) 0.0001 2 (1–3) 2 (1–3) 0.28
  Cerebrovascular disease (%) 34 (17.3) 67 (7.8) 0.001 25 (14.5) 30 (17.3) 0.54
  Chronic cardiovascular insufficiency (%) 9 (4.6) 30 (3.5) 0.52 9 (5.2) 9 (5.2) 1
  Chronic renal insufficiency (%) 53 (26.9) 102 (11.8) < 0.001 46 (26.6) 39 (22.5) 0.44
  Congestive heart failure (%) 12 (6.1) 40 (4.6) 0.47 11 (6.4) 10 (5.8) 1
  Chronic obstructive pulmonary disease (%) 42 (21.3) 117 (13.6) 0.006 39 (22.5) 39 (22.5) 1
  Diabetes mellitus (%) 75 (38.1) 147 (17) < 0.001 60 (34.7) 54 (31.2) 0.50
  Hematologic malignancy (%) 11 (5.6) 68 (7.9) 0.27 11 (6.4) 13 (7.5) 0.83
  Hypertension (%) 108 (54.8) 225 (26.1) < 0.001 88 (50.9) 92 (53.2) 0.72
  Immune deficiency (%) 57 (28.9) 173 (20) 0.0105 47 (27.2) 49 (28.3) 0.90
  Metastatic malignancy (%) 3 (1.5) 42 (4.9) 0.049 3 (1.7) 2 (1.2) 0.13
  Myocardial infarction (history of) (%) 25 (12.7) 75 (8.7) 0.10 20 (11.6) 22 (12.7) 0.87
  Nonmetastatic malignancy (%) 25 (12.7) 126 (14.6) 0.50 23 (13.3) 37 (21.4) 0.07
  Peripheral vascular disease (%) 38 (19.3) 94 (10.9) 0.0025 32 (18.5) 33 (19.1) 1

(Continued)

456 www.ccmjournal.org March 2017 • Volume 45 • Number 3

Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
 Clinical Investigations

TABLE 1. (Continued). Baseline Characteristics of Sepsis Patients Admitted to the ICU

Stratified According to Prior Use of Calcium Channel Blockers
Unmatched Cohort Propensity-Matched Cohort

CCBs No CCBs CCBs No CCBs

Variables (n = 197) (n = 863) p (n = 173) (n = 173) p

Chronic medication
  Angiotensin converting enzyme inhibitors and 104 (52.8) 212 (24.6) < 0.001 86 (49.7) 83 (48) 0.81
angiotensin II receptor blockers (%)
  Anticoagulants (%) 38 (19.3) 131 (15.2) 0.16 35 (20.2) 32 (18.5) 0.78
  Antiplatelet drugs (%) 14 (7.1) 51 (5.9) 0.64 75 (43.4) 75 (43.4) 1
  Beta-blockers (%) 106 (53.8) 233 (27) < 0.001 91 (52.6) 96 (55.5) 0.64
  Corticosteroids (%) 51 (25.9) 112 (13) < 0.001 41 (23.7) 40 (23.1) 1
  Insulin (%) 36 (18.3) 92 (10.7) 0.003 29 (16.8) 33 (19.1) 0.67
  Nonsteroidal anti-inflammatory drugs and 18 (9.1) 107 (12.4) 0.22 16 (9.2) 19 (11) 0.72
cyclooxygenase II inhibitors (%)
  Oral antidiabetic drugs (%) 51 (25.9) 87 (10.1) < 0.001 39 (22.5) 37 (21.4) 0.89
  Other antiarrhythmic drugs (%) 19 (9.6) 36 (4.2) 0.0045 16 (9.2) 12 (6.9) 0.54
  Statins (%) 111 (56.3) 240 (27.8) < 0.001 89 (51.4) 89 (51.4) 1
  Calcium channel blockers
  Amlodipine (%) 108 (54.8) — 93 (53.8) —
  Nifedipine (%) 54 (27.4) — 49 (28.3) —
  Barnidipine (%) 3 (1.5) — 1 (0.6) —
  Felodipine (%) 2 (1) — 2 (1.2) —
  Lercanidipine (%) 4 (2) — 4 (2.3) —
  Verapamil (%) 16 (8.1) — 14 (8.1) —
  Diltiazem (%) 11 (5.6) — 10 (5.8) —
Site of infection
  Pulmonary (%) 97 (49.2) 366 (42.4) 0.10 81 (46.8) 86 (49.7) 0.60
  Abdominal (%) 31 (15.7) 172 (19.9) 0.19 28 (16.2) 26 (15) 0.78
  Urinary tract (%) 27 (13.7) 82 (9.5) 0.09 26 (15) 23 (13.3) 0.65
  Other (%) b
23 (11.7) 142 (16.5) 0.11 21 (12.1) 19 (11) 0.75
  Coinfection (%) 19 (9.6) 101 (11.7) 0.44 17 (9.8) 19 (11) 0.73
Causative pathogens
  Gram-positive bacteria (%) 80 (40.6) 431 (49.9) 0.12 69 (39.9) 76 (43.9) 1
  Gram-negative bacteria (%) 114 (57.9) 501 (58.1) 0.59 95 (54.9) 110 (63.6) 0.70
  Yeast/fungi (%) 25 (12.7) 92 (10.7) 0.30 22 (12.7) 16 (9.2) 0.23
  Other (%) 24 (12.2) 109 (12.6) 1 23 (13.3) 29 (16.8) 0.68
  Unknown (%) 34 (17.3) 141 (16.3) 0.58 30 (17.3) 33 (19.1) 1

(Continued)

Critical Care Medicine www.ccmjournal.org 457

Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Wiewel et al

TABLE 1. (Continued). Baseline Characteristics of Sepsis Patients Admitted to the ICU

Stratified According to Prior Use of Calcium Channel Blockers
Unmatched Cohort Propensity-Matched Cohort

CCBs No CCBs CCBs No CCBs

Variables (n = 197) (n = 863) p (n = 173) (n = 173) p

Severity of disease in first 24 hr

  Acute Physiology and Chronic Health 80 (69–101) 80 (62–102) 0.30 81 (69–101) 85 (67–102) 0.93
Evaluation IV score, median (IQR)
  Sequential Organ Failure Assessment score, 7 (5.5–9) 7 (5–9) 0.67 7 (5–9) 7 (5–9) 0.66
median (IQR)c
  Organ failure (%) 168 (85.3) 727 (84.2) 0.63 148 (85.5) 143 (82.7) 0.82
  Shock (%) 58 (29.4) 301 (34.9) 0.16 53 (30.6) 60 (34.7) 0.49
  Acute lung injury (%) 58 (29.4) 233 (27) 0.52 52 (30.1) 42 (24.3) 0.28
  Acute kidney injury (%) 87 (44.2) 341 (39.5) 0.26 74 (42.8) 66 (38.2) 0.47
CCBs = calcium channel blockers, IQR = interquartile range.
Age not included in score.
a

Site of infection: “other” includes cardiovascular infection, mediastinitis, and skin infection.
b

Central nervous system not included in score, due to large number of sedated patients.
c

and SOFA scores, and the proportion of patients with organ
failure, shock, AKI, and ALI (Table 1). Sites of infection and
causative pathogens did not differ between users and nonus-
ers of CCBs (Table 1). In the matched cohort, 10.4% of prior
CCB-users received at least one CCB dose during the first 4
days after ICU admission, whereas 19.7% received a CCB dur-
ing their ICU stay.
Prior CCB Use and Sepsis Outcome
Prior use of CCBs was associated with a significantly improved
survival at day 30 after ICU admission in both the unmatched
and matched cohorts (Fig. 1). The beneficial association of
prior CCB use with survival remained at days 60 and 90 after
ICU admission (Table 2). Eleven patients were lost to follow-
up at 30 days, 13 at 60 days, and 16 at 90 days after ICU admis-
sion. ICU length of stay (LOS) and the proportion of patients
with organ failure or shock at any time during ICU stay were
similar in CCB-users and nonusers in both the unmatched and
matched cohorts (Table 2). Acute myocardial infarction was a
rare complication in both groups.
Sensitivity and Subgroup Analyses
The association between prior CCB use and 30-day mortal-
ity was also analyzed by multivariable logistic regression.
Through this analysis, we aimed to assess the association
between prior CCB use and reduced sepsis mortality in
important subgroups. In addition, we sought to establish the
relation between chronic use of different other chronic car-
dioprotective medication and sepsis mortality. Prior CCB use
was significantly associated with a reduced risk of mortality
at 30 days after admission (adjusted odds ratio [aOR], 0.48;
95% CI, 0.31–0.74; p = 0.0007) (Table 3). Prior use of beta-
blockers or ACE inhibitors/ARBs was not associated with
altered 30-day mortality.
Patients were enrolled in two ICUs; when analyzed sepa-
rately, the association between prior CCB use and reduced
30-day mortality appeared reproducible in both hospitals
(aOR, 0.46; 95% CI, 0.26–0.82 in AMC and aOR, 0.49; 95%
CI, 0.26–0.93 in UMCU). In a subgroup analysis only includ-
ing patients with septic shock, CCB use was associated with
a reduced risk of mortality (aOR, 0.31; 95% CI, 0.14–0.65),
whereas other cardioprotective medications were not. CCBs
can be divided into dihydropyridine and nondihydropyridines.
In an analysis comparing only dihydropyridine users (amlo-
dipine, nifedipine, barnidipine, felodipine, lercanidipine) with
CCB nonusers, preadmission dihydropyridine use was strongly
associated with an improved survival at day 30 (aOR, 0.51; 95%
CI, 0.32–0.79). A similar trend was seen for nondihydropyri-
dine CCBs (verapamil and diltiazem), although possibly due
to the relatively low number of patients statistical significance
was not reached (aOR, 0.30; 95% CI, 0.09–1.06; p = 0.06). The
association between prior CCB and reduced 30-day mortality
remained when patients who were administered CCBs during
admission were excluded from the analysis (aOR, 0.59; 95% CI,
0.37–0.94; p = 0.03).
Prior CBB Use and Host Response Biomarkers
To obtain insight into the possible effect of prior CCB use
on the host response to sepsis, we measured 15 biomarkers
indicative of activation and/or deregulation of key pathways
implicated in sepsis pathogenesis. As expected (25), sepsis
patients displayed a profound activation of the cytokine net-
work (elevated plasma levels of interleukin [IL]-6, IL-8, and
IL-10) (Fig. 2A), the vascular endothelium (elevated plasma
concentrations of soluble E-selectin, soluble intercellular
adhesion molecule-1 and angiopoietin-2, and reduced levels
of angiopoietin-1) (Fig. 2B) and the coagulation system (pro-
longed prothrombin time, elevated D-dimer levels, reduced

458 www.ccmjournal.org March 2017 • Volume 45 • Number 3

Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
 Clinical Investigations

of tumor necrosis factor-α,

interferon-γ, IL-1β, and IL-13
were undetectable or very low
in the vast majority of patients
and not different between
groups (data not shown).

DISCUSSION
In this prospective observa-
tional study in 1,060 consecu-
tive well-characterized sepsis
patients, CCB use prior to ICU
admission was associated with
Figure 1. Thirty-day Kaplan-Meier survival plots of sepsis patients stratified according to prior use of a calcium improved survival. Chronic
channel blocker (CCB) in the unmatched and propensity-matched cohorts.
CCB use did not modify the
severity of disease at admis-
levels of the anticoagulant proteins protein C and antithrom- sion, sites or causes of infection, or the occurrence of common
bin) (Fig. 2C). In the unmatched cohort, prior CCB use was sepsis complications.
associated with lower IL-6 and IL-8 concentrations during the One previous investigation reported on the association
first 4 days after ICU admission (p = 0.002 and 0.02, respec- between CCB use and the outcome of severe infection. In retro-
tively, versus CCB nonusers by mixed model analysis; Sup- spective review of the clinical records of 388 bacteremic infec-
plemental Table 1, Supplemental Digital Content 1, http:// tions due to aerobic gram-negative bacilli and Staphylococcus
links.lww.com/CCM/C300). In addition, the reductions in aureus, prior CCB use was not associated with an altered mor-
the plasma levels of the anticoagulant proteins protein C tality in a multivariate analysis (26). This earlier investigation
and antithrombin were blunted in prior CCB-users during differed from the present study in multiple ways, including its
this period (p = 0.01 and 0.001, respectively, by mixed model retrospective nature, the analysis of bacteremic patients only
analysis). Other host responses were not different between and the absence of information about the type of care (general
groups in the unmatched cohort. In the propensity-matched ward or ICU) (26).
cohort, prior CCB use only remained significantly associated Our finding of reduced mortality in critically ill sepsis
with an attenuated decline in antithrombin levels (p = 0.03 patients receiving CCBs is supported by experimental stud-
by mixed-effects model) (Fig. 2). The plasma concentrations ies reporting improved survival by administration of CCBs in

TABLE 2. Outcomes of Sepsis Patients Admitted to the ICU Stratified According to Prior
Use of Calcium Channel Blockers
Unmatched Cohort Propensity-Matched Cohort

CCBs No CCBs CCBs No CCBs

Variables (n = 197) (n = 863) p (n = 173) (n = 173) p

Length of stay ICU, median, days (IQR) 5 (2–11) 4 (2–9) 0.21 5 (2–11) 4 (1–9) 0.23
Organ failure during admission (%) 175 (88.8) 766 (88.8) 0.54 154 (89) 151 (87.3) 1
Shock during admission (%) 72 (36.5) 374 (43.3) 0.10 65 (37.6) 76 (43.9) 0.27
Acute lung injury during admission (%) 66 (33.5) 273 (31.6) 0.67 58 (33.5) 51 (29.5) 0.49
Acute kidney injury during admission (%) 105 (53.3) 405 (46.9) 0.12 90 (52) 80 (46.2) 0.33
Acute myocardial infarction during admission (%) 6 (3) 30 (3.5) 0.93 6 (3.5) 7 (4) 1
Mortality
  ICU mortality (%) 23 (11.7) 191 (22.1) 0.001 22 (12.7) 42 (24.3) 0.01
  Hospital mortality (%) 48 (24.4) 284 (32.9) 0.03 45 (26) 64 (37) 0.04
  30-d mortality (%) 38 (19.3) 252 (29.2) 0.005 35 (20.2) 57 (32.9) 0.009
  60-d mortality (%) 53 (26.9) 294 (34.1) 0.06 47 (27.2) 64 (37) 0.048
  90-d mortality (%) 58 (29.4) 326 (37.8) 0.03 52 (30.1) 71 (41) 0.04
CCBs = calcium channel blockers, IQR = interquartile range.

Critical Care Medicine www.ccmjournal.org 459

Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Wiewel et al

TABLE 3. Multiple Binary Logistic Regression Analysis of the Total Cohort to Evaluate the
Association Between Chronic Use of Cardioprotective Medication and 30-D Mortality in
Patients Admitted to the ICU With Sepsis
Variables OR 95% CI p

Calcium channel blockers 0.46 0.59–0.72 0.0007

Beta-blockers 1.14 0.79–1.65 0.48
Angiotensin converting enzyme inhibitors/angiotensin II receptor blockers 1.27 0.88–1.84 0.21
Acute Physiology and Chronic Health Evaluation IV score 1.03 1.02–1.04 < 0.0001
Age 1.02 1.01–1.04 0.0006
Gender 1.06 0.76–1.47 0.73
Race (white) 1.05 0.63–1.75 0.86
Weight 0.99 0.98–1.002 0.16
Cerebrovascular disease 0.77 0.45–1.31 0.33
Chronic cardiovascular insufficency 0.69 0.29–1.65 0.40
Congestive heart failure 0.89 0.42–1.86 0.74
Myocardial infarction (history of) 0.72 0.41–1.27 0.26
Chronic renal insufficiency 1.01 0.64–1.60 0.96
Chronic obstructive pulmonary disease 0.60 0.38–0.96 0.03
Diabetes mellitus 0.99 0.58–1.66 0.96
Hypertension 0.82 0.57–1.19 0.30
Immune deficiency 0.91 0.56–1.49 0.72
Hematologic malignancy 1.59 0.89–2.82 0.12
Nonmetastatic tumor 0.995 0.65–1.53 0.98
Metastatic malignancy 1.27 0.63–2.56 0.51
Antiplatelets 1.35 0.89–2.05 0.16
Statins 1.29 0.52–1.15 0.33
Oral antidiabetics 0.77 0.64–2.24 0.20
Antiarrhythmic drugs 1.20 0.67–2.69 0.57
Corticosteroids 1.34 0.77–2.18 0.41
OR = odds ratio.

animal models of endotoxemia and sepsis. Verapamil improved
survival in an Escherichia coli shock model in dogs (12), pro-
tected mice from endotoxin lethality (13) and enhanced sur-
vival in mice subjected to E. coli peritonitis (15). Diltiazem
decreased mortality during polymicrobial abdominal sepsis
after hemorrhagic shock (27), whereas nifedipine decreased
mortality caused by E. coli peritonitis (15). Several mecha-
nisms have been implicated in the protective effects of CCBs
in experimental sepsis, including restoration of cardiac func-
tion (12), inhibition of proinflammatory cytokine release (13,
28, 29), improvement of lymphocyte functions (27), reduc-
tion of immune suppression (27), and inhibition of muscle
protein breakdown (30). In addition, CCBs may restore endo-
thelial dysfunction by enhancing endothelial nitric oxide syn-
thase activity and improving the antioxidant capacity of the
endothelium by scavenging reactive oxygen species and reduc-
ing the availability of free radicals to react with nitric oxide
(31). Furthermore, CCBs may inhibit the induction of proco-
agulant activity by macrophages stimulated with endotoxin
(32). We studied the potential influence of CCBs on three key
host response systems implicated in sepsis pathogenesis (i.e.,
activation of the cytokine network, the vascular endothelium,
and the coagulation system) by measuring 15 biomarkers
indicative of these responses during the first 4 days after ICU
admission but did not find differences between propensity-
matched CCB-users and nonusers except for less reduction in
antithrombin levels relative to normal values in CCB-users. In
the unmatched cohort, CCB use was associated with attenu-
ated cytokine release and blunted reductions in the anticoag-
ulant proteins antithrombin and protein C, suggesting some

460 www.ccmjournal.org March 2017 • Volume 45 • Number 3

Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
 Clinical Investigations

Figure 2. Plasma host response biomarkers in sepsis patients on the first days of ICU admission stratified according to prior use of a calcium channel
blocker (CCB) in the propensity-matched cohort. A, Cytokines. B, Endothelial cell activation. C, Coagulation activation. Data are expressed as box-and-
whisker diagrams depicting the median and lower quartile, upper quartile, and their respective 1.5 IQR as whiskers (as specified by Tukey). Biomarker
levels were measured in admission plasma of CCB-users (n = 147) and nonusers (n = 138); day 2 plasma of CCB-users (n = 130) and nonusers
(n = 109); and day 4 plasma of CCB-users (n = 88) and nonusers (n = 77). Dashed lines represent median levels in 27 healthy volunteers, except for
prothrombin time, which represents the clinical laboratory upper reference value. Using mixed model analysis, overall antithrombin was higher in CCB-
users than nonusers (p = 0.03), angiopoietin-1 was lower in CCB-users on day 4 (p = 0.03). Note: soluble ICAM-1 is also derived from leukocytes.
* p < 0.05. ICAM-1 = intercellular adhesion molecule-1, PT max = maximal prothrombin time measured during first 24 hr after admission.

Critical Care Medicine www.ccmjournal.org 461

Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Wiewel et al
effect of CCBs in patients who also receive other cardioprotec-
tive and/or vasoactive drugs. Notably, CCBs can reduce intra-
cellular Ca2+ levels and thereby prevent the cellular toxicity
caused by a sustained increase in intracellular Ca2+ concentra-
tions (14, 33, 34). We speculate that chronic use of CCBs par-
tially prevents the activation of destructive processes caused
by prolonged elevations in intracellular Ca2+ levels that target
subcellular structures, including the plasma membrane, cyto-
skeleton, mitochondria, and nucleus (10, 11, 35), and thereby
provides some protection against sepsis lethality. Such effects
are not captured by the detailed measurements reported here;
subcellular studies are needed to provide insight herein. Some
of these CCB mediated effects might result in an improved car-
diac function (12, 36). Data on cardiac function measurements
were not collected during this study.
In the present study, prior use of ACE inhibitors, ARBs,
or beta-blockers was not associated with an altered outcome
of sepsis. Studies in patients hospitalized with pneumonia
reported improved outcomes in patients on ACE inhibitors
and ARBs (37–39). These investigations are different from
ours in that they only included pneumonia patients requiring
hospital admission, whereas our study was conducted in sepsis
patients with different infectious sources requiring ICU admis-
sion. In this same cohort of pneumonia patients, chronic beta-
blocker use was not associated with mortality (39). One study
reported on chronic beta-blocker use and sepsis outcome (40).
In a retrospective record linkage analysis of administrative
databases of Italian patients hospitalized with sepsis, patients
previously prescribed with beta-blockers had a lower 28-day
mortality than those previously untreated. Major differences
with the current study are the inclusion of sepsis patients based
on hospital discharge records versus prospective inclusion by
dedicated research physicians, and inclusion of all hospitalized
patients versus ICU patients. As a consequence of the latter,
mortality rates were relatively low in the earlier survey (21.6%)
(40). We found no association between prior use of statins or
antiplatelet drugs and mortality at day 30; previous studies
were not consistent with regard to these associations (41–46).
Our study has strengths and limitations. We studied a
large prospectively enrolled cohort in which patients were
methodically categorized and followed by committed research
physicians. We implemented propensity matching to enable
evaluation of the effect of chronic CCB use in an unbiased
manner, and the size of our study population enabled us to
perform propensity matching by many important covariates,
resulting in balanced and comparable groups. Nonetheless, it
is possible that a bias remained due to unmeasured confound-
ers. Another limitation involves the fact that this study was
performed in two centers in the Netherlands, limiting gener-
alizability of the results. Also, we are unable to provide data
on the association between chronic CCB therapy and sepsis
progression prior to ICU admission. Our observational study
does not provide insight into potential mechanisms that may
contribute to the association between CCB use and sepsis mor-
tality; in this respect, it is important to note that prior CCB use
did not modify ICU LOS or the occurrence of common sepsis
462 www.ccmjournal.org March 2017 • Volume 45 • Number 3
Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
complications such as AKI and ALI. Finally, our study does not
offer insight into the potential therapeutic effect of CCBs in
sepsis.
Here, we show a significantly reduced mortality in ICU
patients who were on chronic CCB treatment before develop-
ment of sepsis. The association between prior CCB use and
reduced sepsis mortality was consistent in sensitivity and sub-
group analyses. The mechanism by which chronic CCB use
may influence sepsis outcome was not revealed by sequential
measurements of host response biomarkers reflecting activa-
tion of the cytokine network, the vascular endothelium or the
coagulation system, and rather may involve partial prevention
of cellular toxicity related to sustained elevations in intracel-
lular Ca2+ levels.
ACKNOWLEDGMENTS
The authors acknowledge all members of the Molecular Diag-
nosis and Risk Stratification of Sepsis consortium for the
participation in data collection and especially acknowledge:
Friso M. de Beer, MD, Lieuwe D. J. Bos, PhD, Gerie J. Glas,
MD, Roosmarijn T. M. van Hooijdonk, MD, PhD (Depart-
ment of Care, Academic Medical Center, University of Amster-
dam), Mischa A. Huson, MD (Center for Experimental and
Molecular Medicine, Academic Medical Center, University of
Amsterdam), Peter M. C. Klein Klouwenberg, MD, PharmD,
PhD, David S. Y. Ong, MD, PharmD (Department of Intensive
Care Medicine, Julius Center for Health Sciences and Primary
Care and Department of Medical Microbiology, University
Medical Center Utrecht, Utrecht, the Netherlands), Laura R. A.
Schouten, MD (Department of Intensive Care, Academic Med-
ical Center, University of Amsterdam), Marleen Straat, MD,
Esther Witteveen, MD, and Luuk Wieske, MD, PhD (Depart-
ment of Intensive Care, Academic Medical Center, University
of Amsterdam).
REFERENCES
1. Angus DC, van der Poll T: Severe sepsis and septic shock. N Engl J
Med 2013; 369:840–851
2. Liu V, Escobar GJ, Greene JD, et al: Hospital deaths in patients with
sepsis from 2 independent cohorts. JAMA 2014; 312:90–92
3. Vincent JL, Rello J, Marshall J, et al; EPIC II Group of Investigators:
International study of the prevalence and outcomes of infection in
intensive care units. JAMA 2009; 302:2323–2329
4. Martin GS, Mannino DM, Eaton S, et al: The epidemiology of sepsis
in the United States from 1979 through 2000. N Engl J Med 2003;
348:1546–1554
5. James PA, Oparil S, Carter BL, et al: 2014 evidence-based guideline
for the management of high blood pressure in adults: Report from
the panel members appointed to the Eighth Joint National Committee
(JNC 8). JAMA 2014; 311:507–520
6. Fihn SD, Gardin JM, Abrams J, et al; American College of Cardiol-
ogy Foundation; American Heart Association Task Force on Practice
Guidelines; American College of Physicians; American Association
for Thoracic Surgery; Preventive Cardiovascular Nurses Association;
Society for Cardiovascular Angiography and Interventions; Society of
Thoracic Surgeons: 2012 ACCF/AHA/ACP/AATS/PCNA/SCAI/STS
Guideline for the diagnosis and management of patients with stable
ischemic heart disease: A report of the American College of Cardiology
Foundation/American Heart Association Task Force on Practice Guide-
lines, and the American College of Physicians, American Association

for Thoracic Surgery, Preventive Cardiovascular Nurses Association,
Society for Cardiovascular Angiography and Interventions, and Society
of Thoracic Surgeons. J Am Coll Cardiol 2012; 60:e44–e164
7. Braunwald E: Mechanism of action of calcium-channel-blocking
agents. N Engl J Med 1982; 307:1618–1627
8. Mason RP, Marche P, Hintze TH: Novel vascular biology of third-
generation L-type calcium channel antagonists: Ancillary actions of
amlodipine. Arterioscler Thromb Vasc Biol 2003; 23:2155–2163
9. Hotchkiss RS, Karl IE: Calcium: A regulator of the inflammatory
response in endotoxemia and sepsis. New Horiz 1996; 4:58–71
10. Dong Z, Saikumar P, Weinberg JM, et al: Calcium in cell injury and
death. Annu Rev Pathol 2006; 1:405–434
11. Clapham DE: Calcium signaling. Cell 2007; 131:1047–1058
12. Bosson S, Kuenzig M, Schwartz SI: Verapamil improves cardiac func-
tion and increases survival in canine E. coli endotoxin shock. Circ
Shock 1985; 16:307–316
13. Wyska E: Pretreatment with R(+)-verapamil significantly reduces mor-
tality and cytokine expression in murine model of septic shock. Int
Immunopharmacol 2009; 9:478–490
14. Song SK, Karl IE, Ackerman JJ, et al: Increased intracellular Ca2+: A
critical link in the pathophysiology of sepsis? Proc Natl Acad Sci U S
A 1993; 90:3933–3937
15. Bosson S, Kuenzig M, Schwartz SI: Increased survival with calcium
antagonists in antibiotic-treated bacteremia. Circ Shock 1986;
19:69–74
16. Klein Klouwenberg PM, Ong DS, Bos LD, et al: Interobserver
agreement of centers for disease control and prevention criteria for
classifying infections in critically ill patients. Crit Care Med 2013;
41:2373–2378
17. Scicluna BP, Klein Klouwenberg PM, van Vught LA, et al: A molecular
biomarker to diagnose community-acquired pneumonia on intensive
care unit admission. Am J Respir Crit Care Med 2015; 192:826–835
18. Kaukonen KM, Bailey M, Suzuki S, et al: Mortality related to severe
sepsis and septic shock among critically ill patients in Australia and
New Zealand, 2000-2012. JAMA 2014; 311:1308–1316
19. Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure—definition,
outcome measures, animal models, fluid therapy and information tech-
nology needs: The Second International Consensus Conference of
the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004;
8:R204–R212
20. Ranieri VM, Rubenfeld GD, Thompson BT, et al; ARDS Definition
Task Force: Acute respiratory distress syndrome: The Berlin Defini-
tion. JAMA 2012; 307:2526–2533
21. Garner JS, Jarvis WR, Emori TG, et al: CDC definitions for nosoco-
mial infections, 1988. Am J Infect Control 1988; 16:128–140
22. Calandra T, Cohen J: The International Sepsis Forum Consensus
Conference on Definitions of Infection in the Intensive Care Unit. Crit
Care Med 2005; 33:1538–1548
23. Levy MM, Fink MP, Marshall JC, et al: 2001 SCCM/ESICM/ACCP/
ATS/SIS International Sepsis Definitions Conference. Intensive Care
Med 2003; 29:530–538
24. Rubin DB: Estimating causal effects from large data sets using pro-
pensity scores. Ann Intern Med 1997; 127:757–763
25. Parlato M, Cavaillon JM: Host response biomarkers in the diagnosis of
sepsis: A general overview. Methods Mol Biol 2015; 1237:149–211
26. Liappis AP, Kan VL, Rochester CG, et al: The effect of statins on
mortality in patients with bacteremia. Clin Infect Dis 2001; 33:
1352–1357
27. Meldrum DR, Ayala A, Perrin MM, et al: Diltiazem restores IL-2, IL-3,
IL-6, and IFN-gamma synthesis and decreases host susceptibility to
sepsis following hemorrhage. J Surg Res 1991; 51:158–164
Critical Care Medicine www.ccmjournal.org 463
Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Clinical Investigations
28. Szabó C, Haskó G, Németh ZH, et al: Calcium entry blockers increase
interleukin-10 production in endotoxemia. Shock 1997; 7:304–307
29. Hotchkiss RS, Osborne DF, Lappas GD, et al: Calcium antagonists
decrease plasma and tissue concentrations of tumor necrosis factor-
alpha, interleukin-1 beta, and interleukin-1 alpha in a mouse model of
endotoxin. Shock 1995; 3:337–342
30. Hotchkiss RS, Karl IE: Dantrolene ameliorates the metabolic hall-
marks of sepsis in rats and improves survival in a mouse model of
endotoxemia. Proc Natl Acad Sci U S A 1994; 91:3039–3043
31. Brovkovych V, Kalinowski L, Muller-Peddinghaus R, et al: Synergis-
tic antihypertensive effects of nifedipine on endothelium: Concurrent
release of NO and scavenging of superoxide. Hypertension 2001;
37:34–39
32. Lo CJ, Garcia I, Cryer HG, et al: Calcium and calmodulin regulate
lipopolysaccharide-induced alveolar macrophage production of tumor
necrosis factor and procoagulant activity. Arch Surg 1996; 131:
44–50
33. Sayeed MM, Maitra SR: Effect of diltiazem on altered cellular calcium
regulation during endotoxic shock. Am J Physiol 1987; 253:R549–
R554
34. Meldrum DR, Ayala A, Chaudry IH: Mechanism of diltiazem’s immuno-
modulatory effects after hemorrhage and resuscitation. Am J Physiol
1993; 265:C412–C421
35. Duchen MR: Mitochondria and calcium: From cell signalling to cell
death. J Physiol 2000; 529 Pt 1:57–68
36. Zhu X, Bernecker OY, Manohar NS, et al: Increased leakage of sarco-
plasmic reticulum Ca2+ contributes to abnormal myocyte Ca2+ han-
dling and shortening in sepsis. Crit Care Med 2005; 33:598–604
37. Mortensen EM, Nakashima B, Cornell J, et al: Population-based study
of statins, angiotensin II receptor blockers, and angiotensin-convert-
ing enzyme inhibitors on pneumonia-related outcomes. Clin Infect Dis
2012; 55:1466–1473
38. Mortensen EM, Pugh MJ, Copeland LA, et al: Impact of statins and
angiotensin-converting enzyme inhibitors on mortality of subjects hos-
pitalised with pneumonia. Eur Respir J 2008; 31:611–617
39. Wu A, Good C, Downs JR, et al: The association of cardioprotec-
tive medications with pneumonia-related outcomes. PLoS One 2014;
9:e85797
40. Macchia A, Romero M, Comignani PD, et al: Previous prescription
of β-blockers is associated with reduced mortality among patients
hospitalized in intensive care units for sepsis. Crit Care Med 2012;
40:2768–2772
41. Ma Y, Wen X, Peng J, et al: Systematic review and meta-analysis on
the association between outpatient statins use and infectious dis-
ease-related mortality. PLoS One 2012; 7:e51548
42. Wan YD, Sun TW, Kan QC, et al: Effect of statin therapy on mortality
from infection and sepsis: A meta-analysis of randomized and obser-
vational studies. Crit Care 2014; 18:R71
43. Wiewel MA, de Stoppelaar SF, van Vught LA, et al; MARS Consor-
tium: Chronic antiplatelet therapy is not associated with alterations
in the presentation, outcome, or host response biomarkers during
sepsis: A propensity-matched analysis. Intensive Care Med 2016;
42:352–360
44. Tsai MJ, Ou SM, Shih CJ, et al: Association of prior antiplatelet agents
with mortality in sepsis patients: A nationwide population-based
cohort study. Intensive Care Med 2015; 41:806–813
45. Valerio-Rojas JC, Jaffer IJ, Kor DJ, et al: Outcomes of severe sepsis
and septic shock patients on chronic antiplatelet treatment: A histori-
cal cohort study. Crit Care Res Pract 2013; 2013:782573
46. Eisen DP, Reid D, McBryde ES: Acetyl salicylic acid usage and mor-
tality in critically ill patients with the systemic inflammatory response
syndrome and sepsis. Crit Care Med 2012; 40:1761–1767