AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
Page 1 of 48
1
Title
2
Respiratory electrodialysis: a novel, highly efficient, extracorporeal CO2 removal
3
technique
4 5
Authors
6
Alberto Zanella1, Luigi Castagna1, Domenico Salerno1, Vittorio Scaravilli1, Salua Abd
7
El Aziz El Sayed Deab1, Federico Magni1, Marco Giani1, Silvia Mazzola2, Mariangela
8
Albertini2, Nicolò Patroniti1,3, Francesco Mantegazza1, Antonio Pesenti1,3
9 10
Institutional affiliations
11
(1) Dipartimento di Scienze della Salute, Università degli Studi di Milano Bicocca,
12
Monza, Italy
13
(2) Dipartimento di Scienze Veterinarie e Sanità Pubblica, Università degli Studi di
14
Milano, Italy
15
(3) Dipartimento di Anestesia e Rianimazione, Ospedale San Gerardo, Monza, Italy
16 17
Reprint/Negotiation/Corresponding Author
18
Professor Antonio Pesenti, M.D.
19
Dipartimento di Scienze della Salute, Università degli Studi di Milano Bicocca,
20
via Cadore 48 20052 Monza, Italy
21
Tel.: +39 0392333291, Fax: +39 0392332297
22
E-mail:
[email protected]
23 24
Authors’ contributions
1
Copyright © 2015 by the American Thoracic Society
AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
1
All authors provided substantial contribution to the concept, design, data acquisition,
2
analysis and interpretation of findings. Prof. Pesenti first advanced the original idea of
3
applying the electrodialysis technique in an extracorporeal carbon dioxide removal
4
circuitry. Dr. Zanella together with Dr. Castagna and Dr. Salerno developed the
5
Respiratory Electrodialysis technology. Dr. Zanella, Dr. Castagna and Dr. Scaravilli
6
drafted the manuscript and all authors contributed substantially to revisions. All
7
authors give their approval for the final version submitted for publication.
8 9
Equipment support: membrane lungs used for the experiments were provided by
10
Maquet, Rastatt - Germany. Fresenius Medical Care, Bad Homburg – Germany,
11
supplied all the hemofilters.
12 13
Running head: Respiratory Electrodialysis
14 15
Description number: 4.7 Mechanical Ventilation: Applications
16 17
Total word count: 3084
18 19
At a Glance Commentary
20 21
Scientific Knowledge on the Subject: extracorporeal CO2 removal (ECCO2R) has
22
been suggested for the treatment of patients with acute and chronic respiratory
23
failure. The current ECCO2R technology, although perfected compared to the past,
24
still has room for improvement.
2
Copyright © 2015 by the American Thoracic Society
Page 2 of 48
AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
Page 3 of 48
1
What this study adds to the field: we developed Respiratory Electrodialysis, an
2
innovative extracorporeal CO2 removal technique that selectively modulates pH and
3
electrolyte concentration and highly enhances CO2 removal by applying an electrical
4
field to blood. Respiratory Electrodialysis, requiring a minimally invasive approach,
5
could greatly affect the way we treat patients suffering from respiratory failure and
6
other conditions.
7 8
Keywords: Extracorporeal Circulation, Carbon Dioxide
9 10
Online data supplement: this article has an online data supplement, which is
11
accessible from this issue's table of contents online at www.atsjournals.org.
12
3
Copyright © 2015 by the American Thoracic Society
AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
1
Abstract
2
Rationale:
3
We developed an innovative, minimally invasive, highly efficient, extracorporeal CO2
4
removal (ECCO2R) technique called Respiratory Electrodialysis (R-ED).
5
Objectives: To evaluate the efficacy of R-ED in controlling ventilation compared to
6
conventional ECCO2R-technology.
7
Methods: Five mechanically ventilated swine were connected to a custom-made
8
circuit optimized for R-ED, consisting of a hemofilter, a membrane lung, and an
9
electrodialysis cell. Electrodialysis regionally modulates blood electrolyte
10
concentration to convert bicarbonate to CO2 prior to entering the membrane lung,
11
enhancing membrane lung CO2 extraction. All animals underwent 3 repeated
12
experimental sequences, consisting of 4 steps: Baseline (1 h), conventional ECCO2R
13
(2 h), R-ED (2 h), and Final NO-ECCO2R (1 h). Blood and gas flow were 250 mL/min
14
and 10 L/min, respectively. Tidal volume was set at 8 mL/kg and respiratory rate was
15
adjusted to maintain arterial pCO2 at 50 mmHg.
16
Measurements and Main Results: During R-ED, chloride and H+ concentration
17
increased in blood entering the membrane lung, almost doubling CO2 extraction
18
compared to ECCO2R (112 ± 6 vs. 64 ± 5 mL/min, p < 0.001). Compared to baseline,
19
R-ED and ECCO2R reduced minute ventilation by 50% and 27%, respectively.
20
Systemic arterial gas analyses remained stable during the experimental phases. No
21
major complication occurred, but an increase in creatinine level.
22
Conclusions: In this first in-vivo application, we proved electrodialysis feasible and
23
effective in increasing membrane lung CO2 extraction. R-ED was more effective than
24
conventional ECCO2R-technology in controlling ventilation. Further studies are
25
warranted to assess safety profile of R-ED, especially as regards to kidney function.
4
Copyright © 2015 by the American Thoracic Society
Page 4 of 48
AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
Page 5 of 48
1
Introduction
2
Endotracheal intubation and mechanical ventilation are the conventional treatments
3
for respiratory failure, which unfortunately are often associated with severe
4
complications (i.e. ventilator-associated pneumonia, ventilator-induced lung injury,
5
diaphragmatic dysfunction) (1–3). Extracorporeal CO2 removal (ECCO2R) can
6
partially replace lung gas exchange function, sparing patients from the detrimental
7
consequences of intubation and mechanical ventilation, possibly reducing patient
8
mortality (4–9). Recent applications of CO2 removal have been aimed at low
9
invasiveness, through low extracorporeal blood flows (10, 11). With current
10
technology this approach may be limited by a relatively low amount of CO2 removal
11
(12–15). We aimed at significantly increasing the amount of CO2 removed from a
12
given blood flow, so as to achieve a greater decrease of ventilation, and expand the
13
spectrum of potential clinical applications (16–18).
14
We present here a technique that can remove up to 50% of total CO2 production from
15
250 ml/min of blood flow and, by a possible scaling up, remove total CO2 production
16
from about 500 ml/min.
17
Membrane lungs can only remove dissolved CO2 from blood. This gaseous form
18
represents only a small part of the total blood CO2 content, while the majority is
19
chemically combined with water to form bicarbonate ions. The former and the latter
20
are in a chemical equilibrium that can be altered by shifts in acid-base status.
21
Specifically, the lower the pH, the higher the partial pressure of carbon dioxide
22
(pCO2) (19).
23
In order to exploit bicarbonate for gas exchange, we developed an innovative lung
24
support technique, called Respiratory Electrodialysis (R-ED), by combining a
25
hemofilter, a membrane lung, and an electrodialysis unit. By applying electrodialysis
5
Copyright © 2015 by the American Thoracic Society
AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
1
to hemodiafiltrate, the pH and the electrolyte concentration are selectively modulated
2
in specific sections of the extracorporeal circuitry.
3
Blood is regionally acidified, bicarbonate is exchanged with chloride, and the partial
4
pressure of carbon dioxide is increased, leading to facilitated membrane lung CO2
5
removal.
6
Here, we describe the first successful in-vivo application of the electrodialysis
7
technique. We report how R-ED, with minimally invasive extracorporeal support,
8
could largely reduce the ventilatory needs of healthy swine when compared with a
9
standard ECCO2R-technology.
10
6
Copyright © 2015 by the American Thoracic Society
Page 6 of 48
AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
Page 7 of 48
1
Methods
2
After anesthesia and instrumentation, 5 healthy Yorkshire swine (52 ± 2 kg) were
3
mechanically ventilated with inspired oxygen fraction of 50%, positive end-expiratory
4
pressure of 5 cmH2O and tidal volume of 8 ml/kg. Respiratory rate was adjusted
5
throughout the whole experiment to maintain an arterial pCO2 of 50 mmHg. Via a
6
double-lumen 14-Fr catheter inserted into the right external jugular vein, the animals
7
were connected to a custom-made extracorporeal circuit (see online data supplement
8
for details). Unfractionated heparin was continuously infused to achieve an activated
9
clotting time twice the baseline.
10
This circuit was composed of a blood circuit and an electrodialysis circuit (Figure 1,
11
panel A). In the blood circuit blood flowed (250 mL/min) through a hemofilter and a
12
pediatric membrane lung. In the electrodialysis circuit, hemodiafiltrate was generated
13
by the hemofilter (500 mL/min), flowed through the electrodialysis acid chamber, and
14
returned to the hemofilter. Prior to the electrodialysis acid chamber, 40 mL/min of
15
hemodiafiltrate was diverted towards a calcium filter, into the electrodialysis base
16
chamber, and finally returned after a safety filter to the reinfusion lumen of the
17
catheter. To prevent calcium precipitation inside the electrodialysis cell, a fraction of
18
the alkaline hemodiafiltrate (40 mL/min) was recirculated from the outlet of the
19
electrodialysis base chamber prior to the calcium filter. Thus, calcium precipitated
20
before electrodialysis to be trapped by the calcium filter. The electrodialysis unit was
21
customized with a bipolar, an anionic, and a bipolar membrane (see Figure 1, panel
22
B), see online data supplement.
23
All animals underwent 3 repeated experimental sequences, consisting of 4 steps
24
each:
25
- Baseline (1 h): no extracorporeal blood treatment.
7
Copyright © 2015 by the American Thoracic Society
AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
1
- ECCO2R (2 h): gas flow 10 L/min, electrodialysis circuit excluded.
2
- R-ED (2 h): gas flow 10 L/min, electrodialysis circuit functioning and electrodialysis
3
cell supplied by an electric current of 10 Amperes. Calcium gluconate was infused
4
to maintain arterial concentration of ionized calcium within the normal range.
5
- Final NO-ECCO2R (1 h): same as Baseline.
6
R-ED and ECCO2R steps were randomized. Experimental sequences were followed
7
by a 1 h equilibration period.
8
At the end of each step, hemodynamic and ventilatory parameters were recorded;
9
samples from the femoral artery, pulmonary artery and extracorporeal circuit were
10
collected for gas analyses and electrolytes; expired CO2 concentration of membrane
11
lung and natural lung were measured to compute membrane lung CO2 extraction,
12
natural lung CO2 extraction, total CO2 production (V’CO2), and alveolar ventilation. At
13
the end of Baseline and R-ED steps, plasma-free hemoglobin concentration was
14
measured. After instrumentation and at the end of the experiment, blood was
15
sampled for biochemistry measures (see online data supplement).
16
Statistical Analysis
17
Data are presented as mean ± standard deviations unless otherwise stated. Different
18
steps were compared with one-way analysis of variance (ANOVA) for repeated
19
measurements or Kruskal-Wallis test, when appropriate. Two-way ANOVA was used
20
to compare samples along the extracorporeal circuit during ECCO2R and R-ED.
21
Tukey test was used for post-hoc multiple comparisons. A p-value < 0.05 was
22
deemed statistically significant. Analyses were performed using JMP 11.0 (SAS,
23
Cary, NC, USA).
8
Copyright © 2015 by the American Thoracic Society
Page 8 of 48
AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
Page 9 of 48
1
Results
2
R-ED reduced minute volume by half compared to Baseline and Final NO-ECCO2R,
3
and almost doubled the membrane lung CO2 removal when compared with ECCO2R
4
achieving a supplemental 31% minute ventilation reduction (see Figure 2, and Table
5
1). Variations in alveolar ventilation perfectly resembled variations in minute
6
ventilation. Due to the study design, changes in minute ventilation were achieved by
7
modification of respiratory rate; while tidal volume and plateau pressure were stable
8
throughout the whole experiment. This reduction of ventilatory need was secondary
9
to a significant increase in the ratio between membrane lung CO2 removal and V’CO2
10
(see Figure 2 panel B, see also Table E1 in online data supplement). Indeed, during
11
ECCO2R, membrane lung CO2 removal was 64 ± 5 mL/min (i.e. 28 % of V’CO2);
12
while, during R-ED, membrane lung CO2 removal reached 112 ± 6 mL/min (i.e. 50%
13
of V’CO2). V’CO2 was constant during all four steps. Therefore, natural lung CO2
14
removal was significantly lower during R-ED compared with ECCO2R (113 ± 24 vs.
15
161 ± 35 mL/min, respectively, p < 0.001). No alteration in arterial pH and pCO2 was
16
detected during the whole experiment (see table 2).
17
During R-ED, higher levels of membrane lung CO2 removal were achieved by
18
selective modulation of the acid-base balance and the electrolyte concentration in the
19
extracorporeal blood (see Figure 3, see also Table E2 in online data supplement).
20
The concentration of chloride ions increased across the hemofilter (from 99 ± 4 to
21
108 ± 4 mMol/L, p < 0.05). Chloride ions partially replaced bicarbonate ions (from
22
34.6 ± 1.8 to 30.0 ± 1.8 mMol/L, p < 0.05) (see Figure 3 panel A and B), leading to a
23
significant reduction in pH (from 7.37 ± 0.02 to 6.77 ± 0.04, p < 0.05) (see Figure 3
24
panel C), which was associated with a 255% increase in pCO2 (from 61.0 ± 3.1 to
25
216.7 ± 15.5, p < 0.05) (see Figure 3 panel D). The subsequent passage of blood 9
Copyright © 2015 by the American Thoracic Society
AJRCCM Articles in Press. Published on 06-June-2015 as 10.1164/rccm.201502-0289OC
1
across the membrane lung led to a significant reduction of pCO2 and bicarbonate
2
ions, which restored pH back to physiological values. Notably, alterations in the acid-
3
base balance and electrolyte concentration in the extracorporeal blood due to R-ED
4
was limited to specific sections of the extracorporeal circuitry. Indeed, inlet and mixed
5
venous blood showed no differences during R-ED and ECCO2R (see Figure 3, see
6
also Table E2 and E3 in online data supplement).
7
Electrolyte concentration and pH of the hemodiafiltrate during R-ED are shown in
8
Figure 4 (see also Table E4 in on line data supplement). Passage of hemodiafiltrate
9
through the electrodialysis acid chamber increased chloride concentration (from 110
10
± 6 to 115 ± 6 mMol/L, p < 0.05), leading to a pH reduction (from 7.07 ± 0.28 to 6.60
11
± 0.25, p < 0.05). Passage of hemodiafiltrate through the electrodialysis base
12
chamber reduced chloride concentration (from 83 ± 8 to 60 ± 5 mMol/L, p < 0.05),
13
leading to a rise in pH (from 11.01 ± 0.4 to 12.54 ± 0.34, p < 0.05). As expected,
14
hemodiafiltrate entering the calcium filter was extremely alkaline (i.e. 11.00 ± 0.52),
15
which led to calcium precipitation inside the calcium filter 2.8 ± 0.7 to 0.3 ± 0.1 mg/dL,
16
electrodialysis base inlet pre-filter and electrodialysis base inlet post-filter,
17
respectively, p
