Sodium bicarbonate as a local adjunctive agent for limiting platelet activation, aggregation, and adhesion within cardiovascular therapeutic devices

Abstract

Cardiovascular therapeutic devices (CTDs) remain limited by thrombotic adverse events. Current antithrombotic agents limit thrombosis partially, often adding to bleeding. The Impella® blood pump utilizes heparin in 5% dextrose (D5W) as an internal purge to limit thrombosis. While effective, exogenous heparin often complicates overall anticoagulation management, increasing bleeding tendency. Recent clinical studies suggest sodium bicarbonate (bicarb) may be an effective alternative to heparin for local anti-thrombosis. We examined the effect of sodium bicarbonate on human platelet morphology and function to better understand its translational utility. Human platelets were incubated (60:40) with D5W + 25 mEq/L, 50 mEq/L, or 100 mEq/L sodium bicarbonate versus D5W or D5W + Heparin 50 U/mL as controls. pH of platelet-bicarbonate solutions mixtures was measured. Platelet morphology was examined via transmission electron microscopy; activation assessed via P-selectin expression, phosphatidylserine exposure and thrombin generation; and aggregation with TRAP-6, calcium ionophore, ADP and collagen quantified; adhesion to glass measured via fluorescence microscopy. Sodium bicarbonate did not alter platelet morphology but did significantly inhibit activation, aggregation, and adhesion. Phosphatidylserine exposure and thrombin generation were both reduced in a concentration-dependent manner—between 26.6 ± 8.2% (p = 0.01) and 70.7 ± 5.6% (p < 0.0001); and 14.0 ± 6.2% (p = 0.15) and 41.7 ± 6.8% (p = 0.03), respectively, compared to D5W control. Platelet aggregation via all agonists was also reduced, particularly at higher concentrations of bicarb. Platelet adhesion to glass was similarly reduced, between 0.04 ± 0.03% (p = 0.61) and 0.11 ± 0.04% (p = 0.05). Sodium bicarbonate has direct, local, dose-dependent effects limiting platelet activation and adhesion. Our results highlight the potential utility of sodium bicarbonate as a locally acting agent to limit device thrombosis.

Full text

Vol:.(1234567890)
Journal of Thrombosis and Thrombolysis (2023) 56:398–410
https://doi.org/10.1007/s11239-023-02852-4
1 3
Sodium bicarbonate asalocal adjunctive agent forlimiting platelet
activation, aggregation, andadhesion withincardiovascular
therapeutic devices
KaitlynR.Ammann1,2,3· ChristineE.Outridge2· YanaRoka‑Moiia1,2,3· SamiMuslmani2· JunDing4·
JosephE.Italiano5· ElisaTomat6· ScottCorbett4· MarvinJ.Slepian1,2,3,7
Accepted: 12 June 2023 / Published online: 11 July 2023
© The Author(s) 2023
Abstract
Cardiovascular therapeutic devices (CTDs) remain limited by thrombotic adverse events. Current antithrombotic agents
limit thrombosis partially, often adding to bleeding. The Impella® blood pump utilizes heparin in 5% dextrose (D5W) as an
internal purge to limit thrombosis. While effective, exogenous heparin often complicates overall anticoagulation management,
increasing bleeding tendency. Recent clinical studies suggest sodium bicarbonate (bicarb) may be an effective alternative to
heparin for local anti-thrombosis. We examined the effect of sodium bicarbonate on human platelet morphology and function
to better understand its translational utility. Human platelets were incubated (60:40) with D5W + 25mEq/L, 50mEq/L, or
100mEq/L sodium bicarbonate versus D5W or D5W + Heparin 50U/mL as controls. pH of platelet-bicarbonate solutions
mixtures was measured. Platelet morphology was examined via transmission electron microscopy; activation assessed via
P-selectin expression, phosphatidylserine exposure and thrombin generation; and aggregation with TRAP-6, calcium iono-
phore, ADP and collagen quantified; adhesion to glass measured via fluorescence microscopy. Sodium bicarbonate did not
alter platelet morphology but did significantly inhibit activation, aggregation, and adhesion. Phosphatidylserine exposure
and thrombin generation were both reduced in a concentration-dependent manner—between 26.6 ± 8.2% (p = 0.01) and
70.7 ± 5.6% (p < 0.0001); and 14.0 ± 6.2% (p = 0.15) and 41.7 ± 6.8% (p = 0.03), respectively, compared to D5W control.
Platelet aggregation via all agonists was also reduced, particularly at higher concentrations of bicarb. Platelet adhesion to
glass was similarly reduced, between 0.04 ± 0.03% (p = 0.61) and 0.11 ± 0.04% (p = 0.05). Sodium bicarbonate has direct,
local, dose-dependent effects limiting platelet activation and adhesion. Our results highlight the potential utility of sodium
bicarbonate as a locally acting agent to limit device thrombosis.
Keywords Thrombosis· Platelets· Sodium bicarbonate· Platelet activation· Platelet aggregation· Platelet adhesion·
Blood-contacting devices
* Marvin J. Slepian
slepian@arizona.edu
1 Department ofMedicine, University ofArizona, 1501 N
Campbell Ave, Tucson, AZ85724, USA
2 Arizona Center forAccelerated Biomedical Innovation,
University ofArizona, Tucson, AZ, USA
3 Sarver Heart Center, University ofArizona, 1501 N
Campbell Ave, Tucson, AZ85724, USA
4 ABIOMED, Inc., Danvers, MA, USA
5 Department ofSurgery, Boston Children’s Hospital, Harvard
Medical School, Boston, MA, USA
6 Department ofChemistry andBiochemistry, University
ofArizona, Tucson, AZ, USA
7 Department ofBiomedical Engineering, University
ofArizona, 1501 N Campbell Ave, Tucson, AZ85724, USA
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

399
Sodium bicarbonate asalocal adjunctive agent forlimiting platelet activation, aggregation,…
1 3
Highlights
• In the local micro-environment, sodium bicarbonate
can influence platelet function.
• Sodium bicarbonate led to reduction in platelet phos-
phatidylserine exposure, thrombin generation rate, ago-
nist-mediated aggregation, and adhesion to glass.
• The effect of sodium bicarbonate on platelet function
is concentration dependent. Higher concentrations of
sodium bicarbonate lead to more significant effects.
• Sodium bicarbonate may be a potential adjunctive or
alternative agent to traditional anti-thrombotic agents.
Introduction
Current therapy for many forms of advanced cardiovascular
disease is increasingly reliant on the use of implantable car-
diovascular therapeutic devices [1–4]. While these devices
have demonstrated restorative hemodynamic function and
efficacy, they remain limited as foreign bodies with signifi-
cant intrinsic thrombogenicity [5, 6]. Similarly, currently
in use are a wide range of anti-thrombotic agents i.e., anti-
coagulants and anti-platelet drugs, employed concurrently
with devices to mitigate device-related thrombosis [7].
Despite the use of these agents, significant thrombosis-
related limitations and adverse events remain, which include
device-associated thrombosis, thromboembolic-related com-
plications, and bleeding [8].
A prominent device in widespread use today for cardio-
vascular hemodynamic support is the catheter-based micro-
axial blood pump Impella® (ABIOMED, Danvers, MA).
Propelling blood via a high-speed rotating impeller, the
Impella® system relies on internal perfusion with a “purge
solution” to effectively lubricate high-speed rotating ele-
ments, prevent blood ingress, limit protein denaturation and
deposition in purge gaps, and limit purge gap thrombosis [9,
10]. Standard purge fluid contains heparin (25–50U/mL)
to directly inhibit thrombin generation and clot formation
locally in the purge gaps. While heparin has demonstrated
efficacy in reducing purge gap thrombosis and for maintain-
ing purge gap patency, the overall net delivery of heparin
to a given patient resulting from purge fluid infusion and
exit into the bloodstream may differ due to device varia-
tion in the purge gaps [11]. Concurrently with device use,
systemic heparin administration is also required to achieve
systemic anticoagulation, e.g. for management of acute
coronary syndromes. This combination of systemic heparin
administration combined with a variable degree of purge
fluid heparin delivery further complicates anticoagulation
management, potentially increasing bleeding risk [12, 13].
As such, a need exists for an alternative to heparin in purge
fluid to reduce risk of variable, additive heparin administra-
tion for Impella® patients, thereby simplifying anticoagula-
tion management overall.
Recently, in an attempt to limit excessive use of antico-
agulants, associated with bleeding and unpredictable clot-
ting times, clinicians demonstrated that the rate of bleeding
and rate of supratherapeutic anticoagulation is lower with
sodium bicarbonate added to the purge solution and is effec-
tive at ensuring purge patency and limiting device throm-
bosis, without evidence of systemic consequence [14, 15].
These observations have motivated our group to take a fresh
look at the effect of sodium bicarbonate on platelets, and
its impact as an adjunctive agent limiting platelet-mediated
thrombogenicity.
Sodium bicarbonate is a therapeutic buffering agent
often administered to patients with acid/base disorders [16,
17]. Prior studies have shown that sodium bicarbonate is
effective as an inhibitor of blood coagulation invitro and
as a means of limiting thrombosis in catheter access locks
invivo [18]. Further, sodium bicarbonate has been shown
to exhibit anti-fouling and anti-bacterial properties, limit-
ing protein denaturation, adsorption and biofilm formation
on foreign surfaces, contributors to clot formation [19, 20].
As such, this combination of anti-thrombotic and anti-dep-
osition/anti-fouling properties makes sodium bicarbonate
an attractive candidate as a purge fluid heparin alternative.
Generically, this raises the possibility of sodium bicarbonate
acting as a local anti-thrombotic agent, in a defined loca-
tion or space of a device, yet without significant systemic
consequence, as it will be rapidly diluted and buffered with
systemic blood exposure. Despite the promising properties
and clinical observations related to sodium bicarbonate, lim-
ited data exists on the direct effect of bicarbonate on platelet
function. In the present study, we hypothesize that sodium
bicarbonate, at acceptable pharmacologic levels and reason-
able pH for local use, will have a dose-dependent effect in
limiting platelet activation, aggregation, and adhesion to
non-biological materials. As a first step we examined the
effect of sodium bicarbonate on platelet morphology to
identify any signs of significant activation via transmis-
sion electron microscopy (TEM). Second, we examined
the effect of sodium bicarbonate on platelet activation,
i.e. platelet P-selectin exposure, phosphatidylserine expo-
sure and thrombin generation rate. Third, we examined the
effect of sodium bicarbonate on agonist-mediated platelet
aggregation. Finally, we examined the anti-adhesive effects
of sodium bicarbonate on fluorescent-labeled platelet adhe-
sion to a non-biological glass surface.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

400 K.R.Ammann et al.
1 3
Methods
Blood collection andplatelet preparation
Human whole blood was obtained via venipuncture from
healthy adult volunteers in accordance with University of
Arizona IRB-approved protocol (#1810013264). Blood
donors provided written informed consent and verbally
confirmed to have abstained from caffeine or alcohol con-
sumption prior to blood collection, nor taken non-steroidal
anti-inflammatory medications for 2weeks prior. Blood was
collected directly into ACD-A solution (85mM trisodium
citrate, 78mM citric acid, 111mM dextrose) to a final con-
centration of 10% v/v ACD-A in blood.
Platelet-rich plasma (PRP) was obtained via centrifu-
gation of ACD-A anticoagulated whole blood at 300g for
15min. Gel-filtered platelets (GFP) were separated from
PRP via gel-filtration in Sepharose 2B columns. Platelet
counts in PRP and GFP were measured with a Z1 Coul-
ter Particle Counter (Beckman Coulter Inc.). PRP and GFP
were diluted to relevant concentrations for experimentation
with a modified Tyrode’s buffer solution with 0.1% (w/v)
bovine serum albumin (pH 7.4). All active blood compo-
nents were used for experimentation within 8h of collection
from the human blood donor.
Purge solution preparation andpH measurement
Purge solutions were prepared in 5% dextrose (w/v) in water
(D5W; Baxter International). Sodium bicarbonate (8.4% in
water; Baxter International) was diluted in D5W to a final
concentration of 25mEq/L, 50mEq/L, or 100mEq/L, as
reasonable and consistent with clinical use. Heparin solu-
tion (1000U/mL) prepared with heparin sodium salt (Sigma
Aldrich) was diluted in D5W to a final concentration of
50U/mL. Prior to experimentation, platelet samples (GFP
or PRP) were mixed at a 60:40 volumetric ratio with D5W
purge solution alone or D5W purge solution formulated
with heparin (50U/mL) or sodium bicarbonate (25, 50, or
100mEq/L). Platelet-purge solution samples were incubated
for 15min at 37°C immediately prior to experimentation.
The pH of both sodium bicarbonate and heparin contain-
ing purge solutions were measured using a micro-electrode
and pH meter (Orion Versa Star Pro; Thermo Scientific)
at room temperature. pH measurements were also repeated
following 60:40 mixing of the purge solution with either
GFP or PRP.
Platelet morphology: transmission electron
microscopy
GFP samples (100,000 platelets/µL) were fixed with 2.5%
formaldehyde/2.5% glutaraldehyde mixture in 0.1M sodium
cacodylate buffer (pH 7.4; Electron Microscopy Sciences)
for 30min at room temperature. Fixed platelets were pel-
leted via centrifugation at 5000g × 5min and pellets were
stored in 0.2M sodium cacodylate buffer (pH 7.4; Electron
Microscopy Sciences) at 4°C until preparation for TEM.
Prepared platelet pellets were fixed with 1% osmium
tetroxide/1.5% potassium ferrocyanide for 60min followed
by washing in water and 0.2M maleate buffer (pH 5.15;
Electron Microscopy Sciences) and staining with 1% (v/v)
uranyl acetate in maleate buffer for 60min. Samples were
dehydrated via 10-min incubations in increasing concentra-
tions of ethanol (50%, 70, 90%, 100%) and infiltrated in a
1:1 mixture of TAAB resin and propylene oxide overnight at
4°C. Samples were embedded in TAAB Epon and polymer-
ized at 60°C for 48h. Ultrathin 60-nm sections of platelet
samples were cut using a microtome (Reichert Ultracut-S;
Leica Microsystems) and transferred onto copper grids fol-
lowed by staining with lead citrate. Sections were examined
with a Tecnai G2 Spirit BioTWIN transmission electron
microscope at an accelerating voltage of 80kV. Images
were captured with an AMT 2k CCD camera (Hamamatsu
ORCA-HR).
Platelet activation: flow cytometry
Recalcified GFP (20,000 platelets/µL, 1mM CaCl2) were
double-stained with fluorescein conjugated anti-CD41-
APC (clone MEM-6, Invitrogen) and annexin V-FITC
(eBioscience™) or anti-CD42a-FITC (clone GR-P, eBi-
oscience™) and anti-CD62P-APC (clone Psel.KO2.3,
eBioscience™) for 30min at room temperature fol-
lowed by fixation in filtered 2% (v/v) paraformaldehyde
in phosphate-buffered saline (PBS) for 20min at room
temperature. Fixed and stained platelets were then diluted
tenfold in PBS prior to analysis. Flow cytometry was
conducted utilizing a FACSCanto II flow cytometer (BD
Biosciences). Ten thousand events were captured within
a defined “platelet” stopping gate established based on
their forward versus side scatter characteristics (FSC-A/
SSC-A), as compared with standard polystyrene beads of
880nm and 1350nm size from the SPHERO™ Nano Fluo-
rescent particle standard kit (Spherotech, Lake Forest, IL).
Platelets were identified as CD41 + particles within this
gate. Activated platelets, as measured by surface P-selectin
(CD62P) expression, were identified as CD62P + plate-
lets and expressed as percentage of CD41 + population.
Similarly, platelet phosphatidylserine exposure was
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

401
Sodium bicarbonate asalocal adjunctive agent forlimiting platelet activation, aggregation,…
1 3
measured by bound annexin V and expressed as percent-
age of CD41 + population. All samples were analyzed in
duplicate, and data processed with FCS Express Software
(DeNovo Software, V3).
Platelet activation: thrombin generation rate
Thrombin generation rate, a parameter of platelet activa-
tion, was quantified using a modified prothrombinase assay,
as previously described [21]. In brief, following incubation
with purge solutions, platelet samples were re-calcified
(5mM CaCl2) and incubated with acetylated prothrombin
(0.2µM Factor IIa) and 0.1nM Factor Xa. Chromozyme-
TH (0.3mM; Roche Diagnostics GmbH) was utilized as a
chromogenic indicator of thrombin generation in the platelet
sample. Absorbance (405nm), corresponding to thrombin
generation, was kinetically recorded in a spectrophotometric
microplate reader (VersaMax; Molecular Devices) over a
7-min period. Slopes of the absorbance measurements were
recorded as thrombin generation rate.
Platelet aggregation
Platelet aggregation was measured utilizing a light-trans-
mittance aggregometer (PAP-8E; BioData Corporation).
Following incubation with purge solution, PRP samples
(100,000 platelets/µL) were re-calcified (1mM CaCl2) and
stimulated to aggregate with the addition of 4 separate ago-
nists: 32µM thrombin receptor activating peptide 6 (TRAP-
6; AnaSpec); 12µM calcium ionophore (A23187); 20µM
adenosine diphosphate (ADP); or 5µg/mL collagen. Imme-
diately following agonist addition, light transmittance was
recorded over a 10-min period. Samples were maintained
at 37°C and stirred at approximately 1000rpm during
aggregation recording. Area under the generated curve was
recorded as a measure of platelet aggregation.
Platelet adhesion: live fluorescent imaging
GFP samples (20,000 platelets/µL; 1 mL) were incu-
bated with 5µL fluorescent cytoplasmic membrane dye
(BioTracker 490; Sigma-Aldrich) for 15min at room tem-
perature. Immediately following staining, 20 µL of un-
fixed GFP were allowed to adhere to clean glass slides for
15min immediately prior to imaging on an upright micro-
scope (Nikon Eclipse Ni). Adhered platelets (Ex. 484/ Em.
501nm) were visualized via a 100 × objective oil-immersion
lens and captured with CMOS camera (Nikon DS-Qi2).
Three images were captured from each sample. Platelet
adhesion quantification was performed in MATLAB (Math-
works v.2019a) utilizing the imbinarize() and bwareopen()
functions for precise thresholding and segmentation of
adhered fluorescent platelets. Platelet adhesion was calcu-
lated as percent coverage per field of view.
Statistical analysis
Blood donations were taken from a pool of 12 human donors
(8M, 4F; Age 23–32years). Quantitative experiments were
repeated using blood from N ≥ 3 different donors. Assays
of each donor sample were performed in duplicate, apart
from fluorescent imaging which was performed in tripli-
cate. To account for inter-donor variability in platelet func-
tion, purge solutions were run in parallel for each donor and
results were quantified as a percent of the D5W negative
control. ANOVA with Tukey post hoc test was used to quan-
tify p-values and to identify significant differences between
purge solution with additive and D5W solution. All statis-
tical tests were performed at a significance level of 0.05.
All values are displayed as mean ± standard error, unless
otherwise stated. Data analysis was performed and graphed
using GraphPad Prism (v9.0.1).
Results
To elucidate the influence of sodium bicarbonate on platelet
function, human GFP or PRP were incubated with purge
solutions containing D5W alone, or D5W with the addition
of sodium bicarbonate (25, 50, and 100mEq/L) of 50U/
mL heparin. Platelet samples were mixed at a consistent
60:40 platelet: purge ratio to mimic the internal Impella®
device micro-environment where purge fluid and blood ini-
tially mix.
As a first step, we quantified the pH of our purge solu-
tion–platelet mixture. Subsequent platelet morphology via
transmission electron microscopy (TEM) was qualitatively
assessed. Platelet functionality was simultaneously inves-
tigated through platelet activation, i.e. platelet thrombin
generation rate, P-selectin (CD62P) expression, and phos-
phatidylserine exposure. Furthermore, platelet aggregability
and adhesivity were tested to examine potential alterations in
platelet ability to form thrombus in varying purge solution
environments.
pH range ofsodium bicarbonate solutions
The pH of our sodium bicarbonate starting stock solution
ranged between 8.60 and 8.68, whereas D5W and heparin
solutions had an average pH of 5.83 ± 0.07 and 6.12 ± 0.04,
respectively. Following mixing with GFP, the pH of the
platelet and D5W mixture was 7.48 ± 0.01 (Fig.1A). The
same mixture with 50 U/mL heparin resulted in a pH of
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

402 K.R.Ammann et al.
1 3
7.47 ± 0.02. GFP mixture with 25, 50, and 100 mEq/L
sodium bicarbonate solution yielded a pH of 7.68 ± 0.01,
7.83 ± 0.06, and 7.97 ± 0.06, respectively. A similar trend
was observed following mixing with PRP (Fig.1B). The
pH of PRP and D5W mixture was 7.51 ± 0.01 and the pH
of the PRP with heparin was 7.51 ± 0.01. The PRP mixture
with 25, 50, and 100mEq/L sodium bicarbonate solution
yielded a pH of 7.71 ± 0.01, 7.85 ± 0.02, and 8.17 ± 0.03,
respectively. In total, the operational range to which platelets
were exposed in subsequent studies ranged between a pH of
7.47 and 8.17.
Platelet morphology
TEM was employed to detect significant internal or external
morphological alterations of platelets following exposure to
various purge solutions (Fig.2). Platelets exposed to D5W—
i.e. a negative control without purge additives, were found
to have an overall normal morphology, with an average
2–3µm diameter and a typical discoid or ellipsoidal shape,
in agreement with previous reports of quiescent platelets
(Fig.2A) [22]. For D5W alone, occasional platelets were
observed to demonstrate a dilated open canalicular system
(OCS), suggestive of mild degranulation, though filopodia,
membrane extensions or evidence of overall shape change
were minimal. Membrane integrity of platelets within the
field of view remained intact, with little to no evidence of
overall platelet activation. Addition of heparin or sodium
bicarbonate (25, 50, 100mEq/L) to the D5W purge did not
lead to significant alterations in morphology, including OCS
dilation (Fig.2B–E). Notably, no significant differences in
platelet morphology between sodium bicarbonate-containing
purge solutions and heparin-containing purge solutions were
evident. Further, platelets in all purge solutions maintained
their membrane integrity with evident microtubular structure
proximate to the plasma membrane at high magnification,
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D
5W-100 Bicarb
6
7
8
9
pH
**
**
**
Gel-Filtered PlateletsPlatelet Rich Plasma
AB
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D5
W-100 Bicarb
6
7
8
9
pH
**
**
**
Fig. 1 Effect of Sodium Bicarbonate on pH. Human platelets derived
from ACD-A anticoagulated whole blood, were incubated (60:40)
with purge solution. Purge solutions were formulated in base solu-
tion of 5% dextrose in water (“D5W”) with additives 50U/mL hepa-
rin (“D5W + Heparin”), 25 mEq/L sodium bicarbonate (“D5W + 25
Bicarb”), 50 mEq/L sodium bicarbonate (“D5W + 50 Bicarb”), or
100 mEq/L sodium bicarbonate (“D5W + 100 Bicarb”). pH was
measured after 15 min of incubation with gel-A gel-filtered plate-
lets (GFP) or B platelet rich plasma (PRP) utilizing micro pH elec-
trode (Orion; PerpHecT ROSS). Values are reported as displayed as
mean ± standard error. Data were collected from N ≥ 3 human donors;
**indicates p-value < 0.01 in comparison to D5W
A. D5WB. D5W + Heparin
C. D5W + 25 Bicarb D. D5W + 50 Bicarb E. D5W + 100 Bicarb
Fig. 2 Effect of sodium bicarbonateon platelet morphology. Human-
gel-filtered platelets (GFP) derived from ACD-A anticoagulated
whole blood, were incubated (60:40) with purge solution. Purge
solutions were formulated in base solution of 5% dextrose in water
(“D5W”) with additives 50 U/mL heparin (“D5W + Heparin”),
25 mEq/L sodium bicarbonate (“D5W + 25Bicarb”), 50 mEq/L
sodium bicarbonate (“D5W + 50Bicarb”), or 100 mEq/L sodium
bicarbonate (“D5W + 100Bicarb”). Platelets were then pelleted via
high-speed centrifugation and prepared for transmission electronmi-
croscopy (JEOL1200X). Images were captured at 10000× magnifica-
tion. Scale bars (bottom right) in each image represent 1μm
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

403
Sodium bicarbonate asalocal adjunctive agent forlimiting platelet activation, aggregation,…
1 3
indicating absence of membrane alteration seen in classi-
cally activated platelets.
Platelet activation
As a measure of platelet degranulation, an indicator of
platelet activation, flow cytometric detection of P-selectin
(CD62P) was measured on the platelet membrane surface
(Fig.3A). Our results indicated no significant difference
in average CD62P + platelet populations after addition of
25mEq/L sodium bicarbonate (p = 0.82), 50mEq/L sodium
bicarbonate (p = 0.61), or heparin (p = 0.99). However, incu-
bation with 100mEq/L sodium bicarbonate solution led to
an average 7.9 ± 1.5% increase in CD62P + platelets, signifi-
cantly higher than levels seen with D5W alone (p = 0.001)
and D5W with heparin (p < 0.01). Sodium bicarbonate at
25 or 50mEq/L concentrations did not lead to significantly
different changes in P-selectin exposure compared to heparin
(p > 0.49).
As an alternative to platelet degranulation, phosphati-
dylserine exposure on the platelet membrane was measured
through binding of fluorescent-conjugated Annexin V. In
contrast to platelet activation measured by P-selectin expo-
sure, we found an inhibition in phosphatidylserine exposure
on platelets after incubation in bicarbonate-containing purge
solutions (Fig.3B). All concentrations of sodium bicarbo-
nate tested led to a significant decrease in platelet phosphati-
dylserine exposure, with an average decrease of 26.6 ± 8.2%
(p = 0.01), 42.3 ± 12.4% (p = 0.01), and 70.7 ± 5.6%
(p < 0.0001) for 25, 50, and 100mEq/L sodium bicarbonate,
respectively. The decrease in Annexin V + platelets was
positively correlated to the sodium bicarbonate concentra-
tion in the purge solution, with 100mEq/L sodium bicar-
bonate in D5W exhibiting the most dramatic decrease in
Annexin V + platelets. In contrast to sodium bicarbonate,
heparin in the purge solution did not lead to a significant
difference in the Annexin V + platelet population (p = 0.67).
When compared to heparin, sodium bicarbonate had a more
significant effect on phosphatidylserine exposure inhibi-
tion with an average difference of 18.1 ± 11.5% (p = 0.10),
34.5 ± 14.7% (p = 0.002), 62.9 ± 11.0% (p < 0.005) for 25,
50, and 100mEq/L sodium bicarbonate, respectively.
As a functional measurement of platelet activation,
thrombin generation rate was measured from platelets in
various purge solutions (Fig.3C). Sodium bicarbonate at
high concentrations led to a significant decline in platelet
thrombin generation rate versus D5W control, quantified as
an average decrease of 14.0 ± 6.3% (p = 0.15), 24.2 ± 9.6%
(p = 0.05), and 41.7 ± 6.8% (p = 0.03) for 25, 50 and
100mEq/L sodium bicarbonate solutions, respectively. The
decrease in thrombin generation was positively correlated
to the sodium bicarbonate concentration in the purge solu-
tion, with 100mEq/L sodium bicarbonate in D5W exhib-
iting the most dramatic decrease in thrombin generation
rate. As expected, heparin, an indirect thrombin inhibitor,
also led to a significant decrease (42.9 ± 7.1%; p < 0.001)
in thrombin detected in the assay. This was similar to the
level of decline exhibited from platelets after exposure to
100mEq/L sodium bicarbonate purge solution (p = 0.99).
Similarly, 25 and 50mEq/L sodium bicarbonate did not have
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D5
W-100 Bicarb
0
50
100
150
CD62P+ Platelets
(% of D5W)
**
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D5W-100 Bicarb
0
50
100
150
Thrombin Generation Rate
(% of D5W)
** **
*
**
BCA
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D5
W-100 Bicarb
0
50
100
150
AnnV+ Platelets
(% of D5W)
**
**
**
Fig. 3 Effect of sodium bicarbonate on platelet activation. Human
gel-filtered platelets (GFP) derived from ACD-A anticoagulated
whole blood, were incubated (60:40) with purge solution. Purge
solutions were formulated in base solution of 5% dextrose in water
(“D5W”) with additives 50 U/mL heparin (“D5W-Heparin”),
25 mEq/L sodium bicarbonate (“D5W-25 Bicarb”), 50 mEq/L
sodium bicarbonate (“D5W-50 Bicarb”), or 100 mEq/L sodium
bicarbonate (“D5W-100 Bicarb”). Flow cytometric detection (FAC-
ScantoII, BD BioSciences) of fluorescent platelet activation mark-
ers A P-selectin (CD62P); and B Annexin V binding were measured
from the platelet surface. GFP samples were simultaneously meas-
ured for C thrombin generation rate via colorimetric detection in a
modified prothrombinase assay. Values are reported as % of corre-
sponding measurement in D5W solution for each donor, displayed as
mean ± standard error. Data were collected from N ≥ 3 human donors,
performed in duplicate (n ≥ 6). *Indicates p-value < 0.05; **Indicates
p-value < 0.01 in comparison to D5W
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

404 K.R.Ammann et al.
1 3
a significantly different effect on thrombin generation rate as
compared to heparin (p > 0.09).
Platelet aggregation
To evaluate platelet aggregability in the purge solution
environment, platelet aggregation was stimulated via addi-
tion of defined agonists (Fig.4). Aggregation via TRAP-6
declined significantly with addition of either sodium
bicarbonate or heparin to the purge solution. Addition of
sodium bicarbonate led to a decrease in TRAP-6 induced
aggregation, with 25, 50, and 100mEq/L sodium bicar-
bonate concentrations resulting in an average decrease
of 28.9 ± 26.0% (p = 0.78), 43.9 ± 20.6% (p = 0.32), and
85.3 ± 6.6% (p = 0.01), respectively (Fig. 4A). In con-
trast, addition of heparin led to a significant 11.0 ± 3.4%
decline (p = 0.03) in aggregation. Despite the large decline
in average aggregation exhibited with sodium bicarbo-
nate solution, there was larger inter-donor variation in
the effect of lower sodium bicarbonate concentrations (25
and 50mEq/L) which led to statistical insignificance. As
such, there was no significant difference in TRAP-6 medi-
ated aggregation between platelets incubated in 25mEq/L
sodium bicarbonate versus heparin (p = 0.97) nor between
50mEq/L sodium bicarbonate versus heparin (p = 0.67).
In contrast, platelets in 100mEq/L sodium bicarbonate led
to a more significant decline in aggregation compared to
heparin (p = 0.04).
While TRAP-6 stimulates platelet aggregation via PAR1
receptor-binding, calcium ionophore aggregation is recep-
tor independent. Following exposure of platelets to 50 or
100mEq/L sodium bicarbonate, calcium ionophore-medi-
ated aggregation was significantly decreased (Fig.4B).
Sodium bicarbonate in D5W led to an average decrease
in aggregation of 27.0 ± 5.7% (p = 0.04) and 44.8 ± 16.9%
(p = 0.03) at concentrations of 50 and 100mEq/L, respec-
tively. However, low concentration of sodium bicarbonate
(25mEq/L) led to no significant change in aggregation, with
an average decline of 1.1 ± 10.2% (p = 0.99). In contrast to
TRAP-6-mediated aggregation, platelets exposed to hepa-
rin during calcium ionophore-mediated aggregation did not
exhibit statistically significant different aggregability, with
an average 1.0 ± 3.7% decrease (p = 0.99). Similarly, there
was no significant difference in calcium ionophore-induced
aggregation between 25mEq/L sodium bicarbonate and
heparin (p = 0.99); however, both 50mEq/L (p = 0.04) and
100mEq/L (p = 0.05) sodium bicarbonate had a more sta-
tistically significant inhibitory effect on calcium ionophore-
mediated aggregation as compared to heparin.
AB
CD
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D
5W-100 Bicarb
0
50
100
150
Area Under Curve
(% of D5W)
***
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D
5W-100 Bicarb
0
50
100
150
Area Under Curve
(% of D5W)
**
**
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D5
W-100 Bicarb
0
50
100
150
Area Under Curve
(% of D5W)
** *
100 200 300 400
-20
0
20
40
60
80
100
Time (s)
%Aggregation
ADP-Induced
Ca2+ Ionophore-
Induced
TRAP 6-Induced
Collagen-Induced
100 200 300 400
-20
0
20
40
60
80
100
Time (s)
%Aggregation
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D5
W-100 Bicarb
0
50
100
150
Area Under Curve
(% of D5W)
*
**
100 200 300 400
-20
0
20
40
60
80
100
Time (s)
%Aggregation
100 200 300 400
-20
0
20
40
60
80
100
Time (s)
%Aggregation
Fig. 4 Effect of purge solution on agonist-mediated platelet aggrega-
tion. Human platelet rich plasma (PRP), derived from ACD-A anti-
coagulated whole blood, was incubated (60:40) with purge solution.
Purge solutions were formulated in base solution of 5% dextrose in
water (“D5W”) with additives 50U/mL heparin (“D5W-Heparin”),
25mEq/L sodium bicarbonate (“D5W-25Bicarb”), 50mEq/L sodium
bicarbonate(“D5W-50Bicarb”), or 100 mEq/L sodium bicarbonate
(“D5W-100Bicarb”). PRP samples were stimulated to aggregate
via addition of agonists A 32μM TRAP-6; B 10μM calcium iono-
phore (A23187); C 20μM ADP; or D 5μg/mL collagen. Aggrega-
tion was measured over 10min with light-transmission aggregometer
(PAP-8E; Bio/Data Corporation) traces of aggregation over time are
displayed into prow; quantitative data are reported as area under the
curve (bottom row). Values are reported as % of aggregation meas-
ured in D5W solution for each donor, displayed as mean ± standard
error. Data were collected from N ≥ 4 human donors; *Indicates
p-value < 0.05;**indicates p-value < 0.01 in comparison to D5W
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

405
Sodium bicarbonate asalocal adjunctive agent forlimiting platelet activation, aggregation,…
1 3
In examining ADP induced aggregation, ADP as an ago-
nist of P2Y1 and P2Y12 receptors, neither sodium bicar-
bonate nor heparin had a major effect in modulating aggre-
gation (Fig.4C). Aggregation after platelet incubation in
100mEq/L sodium bicarbonate led to 3.9 ± 1.0%. decrease
(p = 0.21) as compared to D5 alone. Lower concentrations
of 25mEq/L sodium bicarbonate (p = 0.57) and 50mEq/L
sodium bicarbonate (p = 0.99) led to no significant difference
in ADP-mediated aggregation as compared to D5W solution
alone. Similarly, heparin had no significant effect on ADP-
induced aggregation (p = 0.18) with an average decrease of
5.1 ± 2.5% compared to D5W alone. Higher concentrations
of 50mEq/L sodium bicarbonate (p = 0.58) and 100mEq/L
sodium bicarbonate (p = 0.99) had no significant difference
in the impact on ADP-induced aggregation as compared to
heparin. However, heparin led to an average 10.9 ± 3.9%
lower aggregation as compared to 25mEq/L sodium bicar-
bonate (p = 0.04).
Collagen-mediated platelet aggregation occurs via bind-
ing to platelet surface GPVI or integrins, e.g., α2β1. Platelet
aggregation via collagen was not significantly influenced by
25mEq/L or 50mEq/L sodium bicarbonate with an average
change of 0.4 ± 4.1% (p = 0.99) and 4.6 ± 6.9% (p = 0.99),
respectively. Notably, 100mEq/L sodium bicarbonate did
show a statistically significant decrease in aggregation, cor-
responding to 42.2 ± 10.8% change (p = 0.001) as compared
to D5W alone (Fig.4D). Heparin led to a larger decline
in collagen-induced platelet aggregation, with an average
83.4 ± 3.2% decline (p < 0.0001). Cor respondingly, hepar in
led to significantly lower aggregation compared to sodium
bicarbonate with an average difference of 83.7 ± 5.8%
(p < 0.0001) vs. 25 mEq/L, 78.7 ± 8.8% (p < 0.0001) vs.
50mEq/L, and 41.1 ± 12.6% (p = 0.003) vs. 100mEq/L,
respectively.
Platelet adhesion
As an additional assessment of the effect of sodium bicarbo-
nate on platelet functionality in the purge solutions, platelet
adhesion to prototypic foreign surface—glass, was quanti-
fied (Fig.5). Lower concentrations of sodium bicarbonate
in the purge solution led to statistically non-significant
decreases in glass platelet surface coverage, correspond-
ing to average decrease of 0.04 ± 0.03% (p = 061) and
0.08 ± 0.04% (p = 0.22) for 25mEq/L and 50mEq/L sodium
bicarbonate, respectively. In contrast however, 100mEq/L
sodium bicarbonate in the purge solution led to a significant
reduction (0.11 ± 0.04%; p = 0.05) in platelet coverage on
the glass surface. Similar to the trends seen in many of the
aggregation assays, we found a significant decrease in the
platelet coverage in the heparin-containing purge solution
(0.07 ± 0.03%; p = 0.04). There were no significant differ-
ences in platelet adhesion between heparin and 25mEq/L
(p = 0.61), 50mEq/L (p = 0.99), or 100 mEq/L (p = 0.15)
sodium bicarbonate.
D5W
D5W-Heparin
D5W-25 Bicarb
D5W-50 Bicarb
D5W-100 Bicarb
0.00
0.05
0.10
0.15
0.20
0.25
% Area Coverage
*
*
D5W + 100 BicarbD5W + 50 BicarbD5W + 25 Bicarb
D5WD5W + Heparin
Fig. 5 Effect of sodium bicarbonate on platelet adhesion. Human gel-
filtered platelets (GFP), derived from ACD-A anticoagulated whole
blood, were incubated (60:40) with purge solution. Purge solutions
were formulated in base solution of 5% dextrose in water (“D5W”)
with additives 50U/mL heparin (“D5W-Heparin”), 25mEq/L sodium
bicarbonate (“D5W-25 Bicarb”), 50 mEq/L sodium bicarbonate
(“D5W-50 Bicarb”), or 100mEq/L sodium bicarbonate (“D5W-100
Bicarb”). Samples were incubated with cytoplasmic membrane stain
(BioTracker490, Invitrogen) and platelets were allowed to adhere to
glass microscope slides. Images of slides were taken at 100× magni-
fication and digitally analyzed for coverage area of platelets adhered
to the surface. Data are reported as % area of platelet coverage of the
image field of view. Scale bars (bottom right) in each image represent
25μm. Digitally magnified images of platelet singlets are displayed
as insets (upper right) with scale bars representing 3 μm. Values in
graph are displayed as mean ± standard error; *Indicates p < 0.05 in
comparison to D5W
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

406 K.R.Ammann et al.
1 3
Discussion
The use of heparin as an antithrombotic agent, while
mechanistically effective, adds complexity to clinical
anticoagulation management, stemming largely from its
intrinsic pharmacokinetic and pharmacodynamic profile
and issues related to under- and over-dosing [23, 24]. This
issue has particularly been brought to light in the setting
of heparin use for concomitant management of a given
patient's underlying cardiovascular prothrombotic disease,
as well as their therapeutic, but intrinsically thrombotic,
device. Recent clinical experience suggests that sodium
bicarbonate may be an effective agent to locally reduce
thrombosis. In the present study, we explored the possibil-
ity of using sodium bicarbonate as a heparin alternative
for local use and effect. Here we specifically examined
the effect of sodium bicarbonate on platelet morphology
and function. Platelet morphology was found to be largely
unchanged in the presence of sodium bicarbonate, at the
clinically acceptable levels tested, as compared to D5W
alone or D5W with heparin. Notably, we observed that
sodium bicarbonate exposure led to a reduction of sponta-
neous platelet activation, as well as to a reduction in ago-
nist-mediated platelet aggregation, particularly at higher
concentrations of sodium bicarbonate tested. Additionally,
sodium bicarbonate led to a coordinate decrease in platelet
adhesion to and coverage of a glass test surface. Overall,
our results suggest that in a local environment, such as in
a catheter purge gap or in a device intrinsic channel, where
bicarbonate concentration is maintained, sodium bicar-
bonate may offer an environment that decreases platelet
thrombogenicity. These observations offer insights into the
mechanism of action of sodium bicarbonate, and helps to
elucidate operational efficacy, reduced fouling and purge
patency when used with the Impella® catheter. These
observations also offer translational potential for use in
local environments of other devices.
Prior use of sodium bicarbonate as a means of altering
platelet function has largely been utilized for preservation
and storage of platelets and platelet concentrates [25, 26].
Addition of bicarbonate to stored platelets aids in buffering
the pH to maintain it within physiologic range [27]. Low
blood pH, or acidosis, is also known to significantly impair
coagulation [28]. Intravenous administration of sodium
bicarbonate is common therapy for re-neutralizing the pH of
the blood and thus can also prevent coagulopathy [17]. Prior
literature has demonstrated sodium bicarbonate’s inhibitory
effect on the conversion of fibrinogen to fibrin, a necessary
step in thrombus formation [29–31]. Wong etal. showed that
bicarbonate from both sodium and potassium salts led to an
increase of both thrombin clotting time and prothrombin
clotting time invitro as well as in clinical invivo cases.
Other prior studies of sodium bicarbonate have investigated
its potential to alter platelet metabolism; however, these
studies have shown no direct influence on platelet metabolic
activity [25, 26]. Instead, sodium bicarbonate is largely uti-
lized clinically for its indirect effect; its influence on pH
and free ionized calcium levels can have significant clini-
cal effects with particular impact on blood coagulation and
platelet function when administered intravenously [32, 33].
Sodium bicarbonate andplatelet activation
Platelet activation may be mediated by both biochemical
and mechanical means [34, 35]. Central in both of these
mechanisms is the role of calcium in activation, as well as
in the coagulation cascade [36, 37]. Here, we observe that
sodium bicarbonate significantly reduced platelet activation
as well as thrombin generation, consistent with effects on
both of these processes. While the exact operative biochemi-
cal mechanisms of sodium bicarbonate underlying these
effects remain undefined, a range of possibilities exist. These
include: (1) the alteration of local pH, particularly in a local
zone of a device (Fig.1), subject to limited buffering from
other means; (2) intracellular changes related to buffering
and carbon dioxide generation; and (3) influence of bicarbo-
nate on calcium availability necessary for platelet function
and protein conformation.
Sodium bicarbonate acts as a buffer to preserve physi-
ologic pH and counteract acidity, ultimately influencing
metabolic function [11]. In an aqueous environment sodium
bicarbonate dissociates to form sodium (Na+) and bicar-
bonate (HCO3
−) ions. Depending on the pH levels in dif-
ferent biological locales, bicarbonate consumes hydrogen
ions (H+) to form carbonic acid (H2CO3), that is ultimately
converted in the body to water (H2O) and carbon dioxide
(CO2) and excreted via the lungs. In our studies, we exam-
ined a reasonable level of sodium bicarbonate as was clini-
cally applicable to the concept here of developing a local
anti-thrombotic agent, meaning that the ionic efficacy and
pH would be maintained in a privileged region, without it
instantly being diluted or buffered by blood. It is understood
that if bicarbonate at the amounts used in this study, if added
to blood, would be instantly buffered. The concept here is
the contained regional effect. As such, it was important to
understand the outer boundary of pH as excessive alkalinity
in a supraphysiologic level may be caustic. Our goal was to
avoid that by operating within a reasonable range (7.47–8.17
pH units). As evidenced by TEM and other data, it did not
appear to have significant negative consequences. Prior stud-
ies have indicated that bicarbonate does not significantly
influence platelet metabolic activity; however, higher alka-
linity has been shown to reduce phosphatidylserine exposure
in blood cells which coincides with our findings [38].
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

407
Sodium bicarbonate asalocal adjunctive agent forlimiting platelet activation, aggregation,…
1 3
Another likely mechanism of bicarbonate influence on
platelet function is via decrease of free ionized calcium
levels [39]. Free ionized calcium (Ca2+) is a vital cofac-
tor in platelet activation and the overall coagulation path-
way. Additionally, intracellular cytosolic calcium eleva-
tion activates multiple platelet signaling events, including
those measured in the present study e.g., phosphatidylserine
exposure, platelet granule secretion, and thrombin genera-
tion. However, the source of calcium Ca2+ e.g., intracellu-
lar stores or extracellular influx through the plasma mem-
brane, may differ depending on the pathway of activation.
In prior work, our group and others have demonstrated that
phosphatidylserine exposure relies on a sustained level of
cytosolic Ca2+, necessitating extracellular influx of calcium
into the platelet [34, 37, 40]. Further, phosphatidylserine
exposure and thrombin generation are closely linked as
phosphatidylserine on the external membrane provides a
favorable site for prothrombinase complexes to bind, con-
verting prothrombin to thrombin. As expected, the Annexin
V binding and thrombin generation rate quantified in the
present study are positively correlated and addition of
sodium bicarbonate resulted in similar decreasing trends
of platelet activation [41, 42]. From that perspective, add-
ing sodium bicarbonate may function here to reduce levels
of free ionized calcium in the local milieu i.e., extracel-
lular Ca2+ available for platelet influx. In a similar fash-
ion, ɑ-granule secretion necessary for P-selectin exposure,
necessitates elevated cytosolic calcium concentration for
cytoskeleton reformation and granule centralization [41].
However, sodium bicarbonate addition to platelets in the
present study did not lead to inhibition of P-selectin expo-
sure. Unlike phosphatidylserine exposure, granular secre-
tion, particularly ɑ-granules containing P-selectin, is trig-
gered after release of Ca2+ from internal stores [43–45].
Therefore, our results suggest that sodium bicarbonate may
influence extracellular Ca2+ availability but does not appear
to impact internal stores of Ca2+ availability required for
ɑ-granule secretion and P-selectin exposure. Further inves-
tigation is required to fully establish the mechanistic influ-
ence of sodium bicarbonate on these different platelet acti-
vation pathways and Ca2+ availability.
Sodium bicarbonate andplatelet aggregation
Platelet aggregation was also reduced in the presence of
sodium bicarbonate. Aggregation can be stimulated by a
variety of agonists which bind to specific surface receptors,
leading to structural and functional changes of GPIIb/IIIa
(integrin ɑIIbβ3) on the platelet surface. Activated GPIIb/IIIa
can bind fibrinogen, allowing for agonist-stimulated platelets
to aggregate [46, 47].
Aggregation agonist TRAP-6 is an analog for thrombin,
binding to G protein-coupled receptor PAR-1 on the plate-
let membrane surface to stimulate aggregation. Collagen
interacts with constitutively expressed GPVI on platelets to
initiate platelet activation and aggregation. Despite different
surface receptors for “outside-in” signaling, both TRAP-6
and collagen lead to an internal signaling cascade within the
platelet that ultimately leads to granule secretion, release
of Ca2+ from internal stores from the platelet dense tubular
system, and activation of GPIIb/IIIa necessary for platelet
aggregation. After depletion of intra-platelet calcium stores,
calcium influx needed for sustained activation is dependent
upon extracellular Ca2+ availability. As previously men-
tioned, sodium bicarbonate lowers ionized calcium levels
and therefore can abrogate the aggregation process in these
cases. This is reflected in the present study in which aggre-
gation was decreased, particularly after addition of high
concentrations of sodium bicarbonate.
The mechanism of platelet activation mediated by cal-
cium ionophore, calcimycin (A23187) is receptor-independ-
ent. Calcimycin selectively binds to free ionized calcium,
allowing transport across the platelet plasma membrane,
effectively increasing intracellular Ca2+ necessary for
platelet activation and aggregation [48]. Our findings of
decreased ionophore-mediated aggregation with increas-
ing sodium bicarbonate concentration further supports our
results observed with platelet phosphatidylserine exposure,
all underpinning the idea that sodium bicarbonate reduces
extracellular Ca2+ availability for influx into the platelet.
ADP stimulates platelet aggregation through interac-
tion with G protein-coupled receptors P2Y1 and P2Y12.
While ADP in the present study is added exogenously to
stimulate aggregation, ADP is also released by activated
platelets through δ-granule secretion. While other agonists
investigated in the present study showed clear indications
of dose-dependent response to sodium bicarbonate, ADP
had a less significant response. It is possible that further
secretion of ADP from δ-granules in this case created a
more robust, sustained aggregation response. Prior work
has also demonstrated ADP as an important mediator in
calcium signaling between adhered and aggregated plate-
lets and therefore may have further complicated Ca2+ influx
affected by sodium bicarbonate [49]. Alternatively local
buffering effects of sodium bicarbonate may have limited
effects on the P2Y1 and P2Y12 purinergic receptors, which
are highly integrated into the platelet membrane [50]. Over-
all, as higher acidity has been shown to impair aggregation
and coagulation, it is unlikely that the influence of sodium
bicarbonate on increasing alkalinity is the main mechanism
at play for reducing agonist-mediated aggregation in the
present study [51–53].
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

408 K.R.Ammann et al.
1 3
Sodium bicarbonate andplatelet adhesion
toforeign surface
Glass as a model surface for cell and platelet adhesion is
well established and characterized [54–57]. Platelet adhe-
sion to a non-biologic glass surface examined in this study
exhibited similar trends as seen with platelet activation by
phosphatidylserine exposure. These results were expected
as platelet activation typically precedes adhesion to a sur-
face. As such, this is a calcium-dependent process and can
similarly be inhibited with reduction in free ionized calcium
and is supported by previous reports [54, 58]. Further, phos-
phatidylserine exposed on the platelet membrane surface is
negatively charged, altering electrostatic interaction with the
glass surface. Because of evident diminished phosphatidyl-
serine exposed in the presence of bicarbonate, it is possible
that the increased negative charge prevented platelet interac-
tion and adhesion to the glass surface.
Study limitations
Experiments were performed under static conditions. It is
possible that non-physiological shear stresses in the high-
speed environment of mechanical circulatory support may
further influence platelet function and the effect of sodium
bicarbonate on the observed parameters. Furthermore, stud-
ies were performed with ambient levels of carbon dioxide,
below levels observed invivo. As part of the bicarbonate
buffering system, the addition of carbon dioxide could
further influence the effect of sodium bicarbonate on pH
and local biochemistry observed in the present study. Non-
physiological shear and carbon dioxide level as experimental
variables were beyond the scope of this project but are a
subject of future study.
Conclusions
Sodium bicarbonate, when used in a local micro-environ-
ment and with clinically acceptable dosing, appears to have
direct effects on platelet function and coagulation. Under
the conditions tested, sodium bicarbonate appears to sig-
nificantly reduce platelet activation, aggregation, and for-
eign surface adhesion, with negligible alterations to platelet
morphology. This modulation of platelet function occurs in
a dose-dependent manner, with more significant changes
observed with higher concentrations of sodium bicarbonate.
Our results support the utility of sodium bicarbonate as a
local antithrombotic agent, for use in geographically defined
and constrained regions and spaces, such as in the purge
gaps of the Impella®. To be clear, the elevated pH of the
bicarb solutions (8.60–8.68pH) tested limit the practicality
of its use for systemic antithrombotic purposes, not to men-
tion whole blood large volume dilution and buffering effects
which would extinguish its effects. Operating in a defined
space, the effects of sodium bicarbonate appear to be other-
wise dominant over other blood clotting effects that would
otherwise occur in the larger systemic circulating blood vol-
ume. Sodium bicarbonate is a comparatively safe option for
use both as an anti-fouling strategy for maintaining device
patency, as well as for use in the purge gap microenviron-
ment of the Impella® to limit thrombogenicity. Our results
are encouraging and mechanistically supportive of use of
sodium bicarbonate as a potential alternative to heparin in
as a local anti-thrombotic agent with cardiovascular thera-
peutic devices.
Acknowledgements We thank Vladimir Gilman PhD for guidance and
review of results and discussion of this paper. Electron microscopy
imaging and services were performed in the Harvard Medical School
Electron Microscopy Facility.
Funding This study was supported by the Arizona Center for Acceler-
ated Biomedical Innovation (ACABI) of the University of Arizona and
by an unrestricted educational grant from Abiomed.
Data availability Data that support the findings of this study are avail-
able for the corresponding author upon request.
Declarations
Conflict of interest Authors KRA, CEO, YRM, SM, ET and JEO re-
port no conflict of interest. MJS reports receiving grant support for
research. JD, SC are employees of Abiomed.
Open Access This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article's Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
the article's Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
References
1. Schmidt T, Abbott JD (2018) Coronary stents: history, design, and
construction. J Clin Med 7:126. https:// doi. org/ 10. 3390/ JCM70
60126
2. Chowdhury M, Quintero PA (2020) Cardiogenic shock and
advanced heart failure therapies. Handb Inpatient Cardiol. https://
doi. org/ 10. 1007/ 978-3- 030- 47868-1_ 12
3. Batsides G, Massaro J, Cheung A, etal (2018) Outcomes of
Impella 5.0 in Cardiogenic Shock. 13:254–260. https:// doi. org/
10. 1097/ IMI. 00000 00000 000535
4. Khan F, Okuno T, Malebranche D etal (2020) Transcatheter aortic
valve replacement in patients with multivalvular heart disease.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

409
Sodium bicarbonate asalocal adjunctive agent forlimiting platelet activation, aggregation,…
1 3
Cardiovasc Interv 13:1503–1514. https:// doi. org/ 10. 1016/J. JCIN.
2020. 03. 052
5. Padera RF, Schoen FJ (2020) Cardiovascular medical devices:
heart valves, pacemakers and defibrillators, mechanical circula-
tory support, and other intracardiac devices. Biomater Sci. https://
doi. org/ 10. 1016/ B978-0- 12- 816137- 1. 00067-2
6. Labarrere CA, Dabiri AE, Kassab GS (2020) Thrombogenic and
inflammatory reactions to biomaterials in medical devices. Front
Bioeng Biotechnol 8:123. https:// doi. org/ 10. 3389/ FBIOE. 2020.
00123
7. Gurbel PA, Shah P, Desai S, Tantry US (2018) Antithrombotic
strategies and device thrombosis. Cardiol Clin 36:541–550.
https:// doi. org/ 10. 1016/J. CCL. 2018. 06. 008
8. Vieira JL, Ventura HO, Mehra MR (2020) Mechanical circulatory
support devices in advanced heart failure: 2020 and beyond. Prog
Cardiovasc Dis 63:630–639. https:// doi. org/ 10. 1016/J. PCAD.
2020. 09. 003
9. Ramzy D, Anderson M, Batsides G etal (2021) Early outcomes of
the first 200 US patients treated with Impella 5.5: a novel tempo-
rary left ventricular assist device. Innovations (Phila) 16:365–372.
https:// doi. org/ 10. 1177/ 15569 84521 10133 29
10. Ammann KR, Ding J, Gilman V etal (2023) Sodium bicarbo-
nate alters protein stability and blood coagulability in a simulated
Impella purge gap model. Artif Organs. https:// doi. org/ 10. 1111/
AOR. 14497
11. Lookabill SK, Dulaney AR, Shepherd G, Kerns WP (2022)
Sodium Bicarbonate. Crit Care Toxicol Diagn Manag Crit Poi-
soned Patient. https:// doi. org/ 10. 1007/ 978-3- 319- 17900-1_ 169
12. Reed BN, Didomenico RJ, Erin Allender J etal (2019) Survey of
anticoagulation practices with the impella percutaneous ventricu-
lar assist device at high-volume centers. J Interv Cardiol. https://
doi. org/ 10. 1155/ 2019/ 37913 07
13. Beavers CJ, DiDomenico RJ, Dunn SP etal (2021) Optimizing
anticoagulation for patients receiving Impella support. Pharma-
cotherapy 41:932–942. https:// doi. org/ 10. 1002/ PHAR. 2629
14. Bergen K, Sridhara S, Cavarocchi N etal (2022) Analysis of
bicarbonate-based purge solution in patients with cardiogenic
shock supported via impella ventricular assist device. Ann Phar-
macother. https:// doi. org/ 10. 1177/ 10600 28022 11241 56/ ASSET/
IMAGES/ LARGE/ 10. 1177_ 10600 28022 11241 56- FIG2. JPEG
15. Al-Ayoubi AM, Bhavsar K, Hobbs RA etal (2022) Use of sodium
bicarbonate purge solution in impella devices for heparin-induced
thrombocytopenia. J Pharm Pract 2022:1–4. https:// doi. org/ 10.
1177/ 08971 90022 10890 78/ ASSET/ IMAGES/ LARGE/ 10. 1177_
08971 90022 10890 78- FIG1. JPEG
16. Sabatini S, Kurtzman NA (2009) Bicarbonate therapy in severe
metabolic acidosis. J Am Soc Nephrol 20:692–695. https:// doi.
org/ 10. 1681/ ASN. 20071 21329
17. Adeva-Andany MM, Fernández-Fernández C, Mouriño-Bayolo
D etal (2014) Sodium bicarbonate therapy in patients with meta-
bolic acidosis. Sci World J. https:// doi. org/ 10. 1155/ 2014/ 627673
18. El-Hennawy AS, Frolova E, Romney WA (2019) Sodium bicar-
bonate catheter lock solution reduces hemodialysis catheter loss
due to catheter-related thrombosis and blood stream infection: an
open-label clinical trial. Nephrol Dial Transplant 34:1739–1745.
https:// doi. org/ 10. 1093/ NDT/ GFY388
19. Pratten J, Wiecek J, Mordan N etal (2016) Physical disruption of
oral biofilms by sodium bicarbonate: an invitro study. Int J Dent
Hyg 14:209–214. https:// doi. org/ 10. 1111/ IDH. 12162
20. Corral LG, Post LS, Montville TJ (1988) Antimicrobial activity
of sodium bicarbonate. J Food Sci 53:981–982. https:// doi. org/ 10.
1111/J. 1365- 2621. 1988. TB090 05.X
21. Jesty J, Bluestein D (1999) Acetylated prothrombin as a substrate
in the measurement of the procoagulant activity of platelets:
elimination of the feedback activation of platelets by thrombin.
Anal Biochem 272:64–70. https:// doi. org/ 10. 1006/ ABIO. 1999.
4148
22. Neumüller J, Meisslitzer-Ruppitsch C, Ellinger A etal (2013)
Monitoring of platelet activation in platelet concentrates using
transmission electron microscopy. Transfus Med Hemotherapy
40:101. https:// doi. org/ 10. 1159/ 00035 0034
23. Hemker HC (2016) A century of heparin: past, present and future.
J Thromb Haemost 14:2329–2338. https:// doi. org/ 10. 1111/ JTH.
13555
24. Hirsh J, Fuster V (1994) Guide to anticoagulant therapy. Part 1:
heparin. Am Heart Assoc Circ 89:1449–1468. https:// doi. org/ 10.
1161/ 01. CIR. 89.3. 1449
25. Radwanski K, Min K (2013) The role of bicarbonate in platelet
additive solution for apheresis platelet concentrates stored with
low residual plasma. Transfusion 53:591–599. https:// doi. org/ 10.
1111/J. 1537- 2995. 2012. 03745.X
26. Gulliksson H (2014) Platelet storage media. Vox Sang 107:205–
212. https:// doi. org/ 10. 1111/ VOX. 12172
27. Martini WZ, Dubick MA, Pusateri AE etal (2006) Does bicarbo-
nate correct coagulation function impaired by acidosis in swine?
J Trauma Inj Infect Crit Care 61:99–106. https:// doi. org/ 10. 1097/
01. TA. 00002 15574. 99093. 22
28. Ramaker AJDWR, Meyer P, Van Der Meer J etal (2009) Effects of
acidosis, alkalosis, hyperthermia and hypothermia on haemosta-
sis: results of point of care testing with the thromboelastography
analyser. Blood Coagul Fibrinolysis 20:436–439. https:// doi. org/
10. 1097/ MBC. 0B013 E3283 2DC327
29. Wong DW, Mishkin FS, Tanaka TT (1980) The effects of bicar-
bonate on blood coagulation. JAMA 244:61–62. https:// doi. org/
10. 1001/ JAMA. 1980. 03310 01004 7028
30. Wong DW (1980) Effect of sodium bicarbonate on invitro conver-
sion of fibrinogen to fibrin. J Pharm Sci 69:978–980. https:// doi.
org/ 10. 1002/ JPS. 26006 90832
31. Vig S, Chitolie A, Bevan D etal (2021) Thromboelastography.
Hematology 6:205–213. https:// doi. org/ 10. 1080/ 10245 332. 2001.
11746 573
32. Chew DJ, Leonard M, Muir W (1989) Effect of sodium bicarbo-
nate infusions on ionized calcium and total calcium concentrations
in serum of clinically normal cats. Am J Vet Res 50:145–150
33. Rastegar M, Levine BS, Felsenfeld AJ (2014) Metabolic acidosis-
induced hypercalcemia in an azotemic patient with primary hyper-
parathyroidism. Clin Kidney J 7:299–302. https:// doi. org/ 10. 1093/
CKJ/ SFU041
34. Roka-Moiia Y, Walk R, Palomares DE etal (2020) Platelet acti-
vation via shear stress exposure induces a differing pattern of
biomarkers of activation versus biochemical agonists. Thromb
Haemost 120:776–792. https:// doi. org/ 10. 1055/S- 0040- 17095 24/
ID/ JR190 519- 11
35. Slepian MJ, Sheriff J, Hutchinson M etal (2017) Shear-mediated
platelet activation in the free flow: perspectives on the emerging
spectrum of cell mechanobiological mechanisms mediating car-
diovascular implant thrombosis. J Biomech 50:20. https:// doi. org/
10. 1016/J. JBIOM ECH. 2016. 11. 016
36. Li Z, Delaney MK, O’Brien KA, Du X (2010) Signaling during
platelet adhesion and activation. Arterioscler Thromb Vasc Biol
30:2341–2349. https:// doi. org/ 10. 1161/ ATVBA HA. 110. 207522
37. Singh S, Dodt J, Volkers P etal (2019) (2019) Structure functional
insights into calcium binding during the activation of coagulation
factor XIII A. Sci Reports 91(9):1–18. https:// doi. org/ 10. 1038/
s41598- 019- 47815-z
38. Lu DCY, Wadud R, Hannemann A etal (2021) Pathophysiological
relevance of renal medullary conditions on the behaviour of red
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

410 K.R.Ammann et al.
1 3
cells from patients with sickle cell anaemia. Front Physiol 12:326.
https:// doi. org/ 10. 3389/ FPHYS. 2021. 653545/ BIBTEX
39. Hughes WS, Aurbach GD, Sharp ME, Marx SJ (1984) The effect
of the bicarbonate anion on serum ionized calcium concentration
invitro. J Lab Clin Med 103:93–103
40. Roka-Moiia Y, Miller-Gutierrez S, Palomares DE etal (2021)
Platelet dysfunction during mechanical circulatory support: ele-
vated shear stress promotes downregulation of αIIbβ3and GPIb
via microparticle shedding decreasing platelet aggregability. Arte-
rioscler Thromb Vasc Biol 41:1319–1336. https:// doi. org/ 10. 1161/
ATVBA HA. 120. 315583
41. Heijnen H, van der Sluijs P (2015) Platelet secretory behaviour:
as diverse as the granules … or not? J Thromb Haemost 13:2141–
2151. https:// doi. org/ 10. 1111/ JTH. 13147
42. Harper MT, Poole AW (2010) Protein kinase Cθ negatively regu-
lates store-independent Ca 2+ entry and phosphatidylserine expo-
sure downstream of glycoprotein VI in platelets. J Biol Chem
285:19865–19873. https:// doi. org/ 10. 1074/ jbc. M109. 085654
43. Harper AGS, Mason MJ, Sage SO (2009) A key role for dense
granule secretion in potentiation of the Ca2+ signal arising from
store-operated calcium entry in human platelets. Cell Calcium
45:413–420. https:// doi. org/ 10. 1016/J. CECA. 2009. 02. 003
44. Lopez E, Bermejo N, Berna-Erro A etal (2015) Relationship
between calcium mobilization and platelet α- And δ-granule
secretion. A role for TRPC6 in thrombin-evoked δ-granule exo-
cytosis. Arch Biochem Biophys 585:75–81. https:// doi. org/ 10.
1016/J. ABB. 2015. 09. 012
45. Varga-Szabo D, Braun A, Nieswandt B (2009) Calcium signaling
in platelets. J Thromb Haemost 7:1057–1066. https:// doi. org/ 10.
1111/J. 1538- 7836. 2009. 03455.X
46. Rivera J, Lozano ML, Navarro-Núñez L, Vicente García V (2009)
Platelet receptors and signaling in the dynamics of thrombus for-
mation. Haematologica 94:700. https:// doi. org/ 10. 3324/ HAEMA
TOL. 2008. 003178
47. Zhou Y, Yasumoto A, Lei C etal (2020) Intelligent classification
of platelet aggregates by agonist type. Elife 9:1–15. https:// doi.
org/ 10. 7554/ ELIFE. 52938
48. Verma A, Bhatt AN, Farooque A etal (2011) Calcium ionophore
A23187 reveals calcium related cellular stress as “I-bodies”: an
old actor in a new role. Cell Calcium 50:510–522. https:// doi. org/
10. 1016/J. CECA. 2011. 08. 007
49. Nesbitt WS, Giuliano S, Kulkarni S etal (2003) Intercellular cal-
cium communication regulates platelet aggregation and thrombus
growth. J Cell Biol 160:1151–1161. https:// doi. org/ 10. 1083/ JCB.
20020 7119/ VIDEO-1
50. Zhan C, Yang J, Dong XC, Wang YL (2007) Molecular modeling
of purinergic receptor P2Y12 and interaction with its antagonists.
J Mol Graph Model 26:20–31. https:// doi. org/ 10. 1016/J. JMGM.
2006. 09. 006
51. Scharbert G, Franta G, Wetzel L, Kozek-Langenecker S (2011)
Effect of pH levels on platelet aggregation and coagulation: a
whole blood invitro study. Crit Care. https:// doi. org/ 10. 1186/
CC9866/ METRI CS
52. Engström M, Schött U, Romner B, Reinstrup P (2006) Acidosis
impairs the coagulation: a thromboelastographic study. J Trauma
61:624–628. https:// doi. org/ 10. 1097/ 01. TA. 00002 26739. 30655.
75
53. Watts SE, Tunbridge LJ, Lloyd JV (1983) Storage of platelets for
tests of platelet function: effects of pH on platelet aggregation and
liberation of β-thromboglobulin. Thromb Res 29:343–353. https://
doi. org/ 10. 1016/ 0049- 3848(83) 90046-4
54. George JN (1972) Direct assessment of platelet adhesion to glass:
a study of the forces of interaction and the effects of plasma and
serum factors, platelet function, and modification of the glass sur-
face. Blood 40:862–874. https:// doi. org/ 10. 1182/ BLOOD. V40.6.
862. 862
55. Broberg M, Eriksson C, Nygren H (2002) GPIIb/IIIa is the main
receptor for initial platelet adhesion to glass and titanium surfaces
in contact with whole blood. J Lab Clin Med 139:163–172. https://
doi. org/ 10. 1067/ mlc. 2002. 121604
56. Curtis AS (1964) The mechanism of adhesion of cells to glass: a
study by interference reflection microscopy. J Cell Biol 20:199.
https:// doi. org/ 10. 1083/ JCB. 20.2. 199
57. Rajaraman R, Rounds DE, Yen SPS, Rembaum A (1974) A scan-
ning electron microscope study of cell adhesion and spreading
invitro. Exp Cell Res 88:327–339. https:// doi. org/ 10. 1016/ 0014-
4827(74) 90248-1
58. White GC, Scarborough DE, Brinkhous KM (1983) Morphologi-
cal study of early phases of platelet adhesion to foreign surfaces:
effect of calcium. Ann N Y Acad Sci 416:351–362. https:// doi.
org/ 10. 1111/J. 1749- 6632. 1983. TB351 98.X
Publisher's Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com