The 8-Year Experience of the Florence Referral ECMO Center and Retrieval Team for Acute Respiratory Failure

Author’s Accepted Manuscript The 7-year experience of Florence Referral ECMO center and retrieval team for acute respiratory failure Giovanni Cianchi, Chiara Lazzeri, Manuela Bonizzoli, Stefano Batacchi, Morena Cozzolino, Marco Ciapetti, Pasquale Bernardo, Andrea Franci, Marco Chiostri, Adriano Peris

PII: DOI: Reference:

www.elsevier.com/locate/buildenv

S1053-0770(17)30544-X http://dx.doi.org/10.1053/j.jvca.2017.06.018 YJCAN4202

To appear in: Journal of Cardiothoracic and Vascular Anesthesia Cite this article as: Giovanni Cianchi, Chiara Lazzeri, Manuela Bonizzoli, Stefano Batacchi, Morena Cozzolino, Marco Ciapetti, Pasquale Bernardo, Andrea Franci, Marco Chiostri and Adriano Peris, The 7-year experience of Florence Referral ECMO center and retrieval team for acute respiratory failure, Journal of Cardiothoracic and Vascular Anesthesia, http://dx.doi.org/10.1053/j.jvca.2017.06.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The 7-year experience of Florence Referral ECMO center and retrieval team for acute respiratory failure 1*

Giovanni Cianchi,1Chiara Lazzeri, , 1Manuela Bonizzoli, 1Stefano Batacchi, Morena Cozzolino , 1Marco Ciapetti, 2Pasquale Bernardo, 1 Andrea Franci, 1Marco Chiostri 1Adriano Peris,

1

1

Intensive Care Unit and Regional ECMO Referral centre, Emergency Department, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy;

2

Intensive Cardiac Care Unit, Heart and Vessels Department,Azienda OspedalieroUniversitaria Careggi, Florence, Italy;

Short title: A 7-year experience of an ECMO referral Center

Address correspondence to: Giovanni Cianchi, MD Intensive Care Unit and Regional ECMO Referral centre Emergency Department, Viale Morgagni 85, 50134 Florence, Italy Tel: +39-55-7947823; Fax: +39-55-7947821 e-mail:[email protected]

ABSTRACT

Objective: Many ECMO centers for respiratory failure and ECMO mobile teams were instituted during the H1N1 pandemic. Data on transportation are scarce and heterogeneous.

We

therefore described the experience of our referral ECMO center for severe respiratory failure from 2009 to 2016 and gave a comprehensive report of

transfers performed by our mobile ECMO team.

Design observational retrospective study

Setting an intensive care unit (ECMO referral center) in a teaching hospital

Participants 160 consecutive patients with ARDS refractory to conventional treatment requiring VV ECMO.

Intervention VV ECMO implantation.

Measurements and main results In our series, the transferred patients represented the 57%, with annual percentages ranging from 28% to 90%,

over the years

. No adverse event was observed during transportation. A progressive increase in SAPS

values and in the use of norepinepnhrine were detectable (p=0.048 and p=0.037, respectively) as well as in neuromuscolar blockers use (p=0.004). Dual lumen cannula was more frequently used in recent years (p<0.001).The overall mortality rate was 40% (64/160) with no differences over the years nor between transferred and local patients. BMI (body mass index) and pre ECMO neuromuscular blockers and SAPS were independent predictors for early mortality (when adjusted for age). Conclusions: The workload of our referral center and our mobile team did not change, documenting that severe respiratory failure requiring VV-ECMO support is still a clinical need. No difference in mortality rate was detectable during this period nor between transferred and local

patients, who were managed by the same team.

Re: MS # JCVA-D-17-00215

INTRODUCTION Veno-Venous Extracorporeal circulation support (VV-ECMO) can be a lifesaving procedure for treatment of pulmonary failure refractory to conventional treatments in adult patients

(1,2)

.

Management of patients with severe respiratory failure refractory requiring ECMO support is often restricted to specialized centers with particular expertise, since successful use of ECMO requires considerable experience and resource utilization

(3)

. In peripheral hospitals, transferring patients to

centers with better technical resources can be the only alternative (4-6). . Most ECMO centers for respiratory failure and ECMO mobile teams were instituted during the H1N1 pandemia (1,3). Data are scarce and heterogeneous for number consistency, composition of ECMO mobile team, kind of transportation (ground vs air) and distance (6-14). The present investigation was aimed at describing the experience of our referral ECMO center for severe respiratory failure from 2009 to 2016 and to give a comprehensive report of transfers performed by our mobile ECMO team (which was instituted during the H1N1 pandemic) during this period.

METHODS

Between October 2009 and December 2016, 160 consecutive patients with ARDS refractory to conventional treatment and supported by VV ECMO were managed at our Center, which is a referral one. Among these patients, 91 patients (57%) were implanted with VV ECMO at a primary care hospital and transferred to our center on ECMO support by our mobile ECMO team. Twelve patients were included in a previously preliminary experience (15).

The Careggi Teaching Hospital had started, since H1N1 influenza pandemic, a collaboration with the ICUs of 12 district hospitals in Tuscany in a project for centralization of acute lung injury/ARDS patients who require (or may require) ECMO treatment. In 2008 and spring 2009, preliminary meetings were organized to inform the peripheral hospitals' ICU staff and Administrations about the availability of the new ECMO program. During the H1N1 influenza A pandemic, the knowledge of ECMO treatment rapidly spread among the medical community and the Regional Ministry of Health issued indications to transfer all patients affected by severe respiratory failure related to influenza to Careggi Hospital.

Patient selection Referral of patients to our center is initiated by telephone contacts: the intensivist in charge collects clinical information and evaluates entry and exclusion criteria for ECMO. Parameters that were adopted to quickly detect patients suitable for extracorporeal treatment in the peripheral hospitals have been previously described

(15)

. The time from decision to go until

departure is between 30 and 90 minutes, 24 h per day, 7 days per week, all year round. If the patient is judged eligible, the referring center is asked to make relatives fully aware of the procedure and transportation and check the availability of blood products.

Mobile ECMO Team The ECMO team consists of an anesthesiologist expert in critical care (the intensivist), a cardiac surgeon, a cardiologist and a perfusionist, all trained on ECMO technique and management. According to our activation protocol, ECMO team can be summoned within one hour, with 24 hour coverage. The intensivist is the on-call ECMO consultant. In most cases, the intensivist primes the process on the basis of clinical and radiological findings, summons the full ECMO team and stays in charge of the team and supervises the clinical decisions. The cardiologist's main task (CL and PB) is to evaluate cardiac function in the pre-ECMO phase and to guide the correct positioning of

ECMO cannulas by transesophageal ultrasonography. Furthermore, the cardiologist is directly involved in selecting patients with cardiac failure suitable for ECMO treatment. The cardiac surgeon, in addition to actively participating to the clinical decision making process, is responsible for selecting and inserting the cannulas and starting the extracorporeal circulation, with the assistance of the perfusionist. Ground transportation is performed by means of a specially equipped ambulance (Figure1) , equipped with extended oxygen capacity (6400 liters) and two power inverters, which provide 1800+800 watts at 220 volts. This ambulance is loaded with a an ICU portable monitor, infusion pumps, suction equipment, and additional oxygen . Until 2015 an ICU ventilator was loaded aboard to permit high quality ventilation during ambulance trip (Servo I, Maquett, Rastatt, Germany) while a portable ventilator was used for in-hospital transport only (Oxylog 3000, Drägerwerk AG, Lubec, Germany). Eventually, a high performance portable ventilator (Hamilton –T1, Hamilton Medical AG, Bonaduz, Switzerland) became available and replaced the ICU ventilators in most cases. The ambulance is equipped with a self-loading-heavy duty stretcher (Power-PRO XT, Stryker EMS, Portage, MI USA) and a custom-made steal frame is mounted on the stretcher to accommodate all medical equipment (including oxygen cylinders) for in-hospital transportation. However due to Italian Road Safety Regulations, all medical devices have to be firmly fixed to the ambulance cabin, and the stretcher frame has to be removed during ambulance trip. An additional vehicle transports the team and the rest of the equipment, included a a portable ultrasound, equipped with probes for transthoracic, transesophageal and vascular examination and a fibroscope. All the equipment material for cannulation, circuit set-up, and transportation is prepacked in 3 backpacks. Therefore the ECMO team is completely autonomous for any needing and does not have to rely on the material of the referral hospital. Once at the referral hospital, the ECMO crew receives clinical handover and re-evaluates the patient. Re-evaluation is a stepwise process, which includes, in progression, optimization of ventilator therapy, recruiting maneuvers andh fibroscopy when needed. The aim of this procedure is

to assess the safety of a conventional transport: if judged unsafe, the ECMO cannulation procedure is started.

The requirement of ECMO initiation was decided on the basis of the Italian Ministry of Health criteria, as previously described (15-17). However, there is some flexibility as to who is deemed suitable for ECMO, and each case is judged on an individual basis and the final decision is

taken with the advice of an on-call ECMO consultant, at Careggi Hospital,when needed.

The Italian Ministry of Health criteria are as follows: 1) SaO2 < 85% for at least 1 hour; 2) Oxygenation Index >25 for at least 6 hours

after ventilation optimisation.

Oxygenation Index: Mean airway pressure (cmH2O) * FiO2 * 100/PaO2 ; 3) PaO2/FiO2 < 100 with PEEP ≥ 10cmH2O for at least 6 hours after ventilation’s optimization ; 4) Hypercapnia with pH < 7.25 ; 5) SvO2 (central

venous oxygen saturation) < 65% with hematocrit >30 and under vasoactive drugs infusion. The parameters are referred to a condition of lung protective ventilation (tidal volume:4-6 ml/Kg of predicted body weight; plateau pressure ≤ 30 cmH20; PEEP >lower inflection point of the curve

pressure-volume).

ECMO for respiratory failure

Patients deemed suitable for ECMO treatment were evaluated on site by the ECMO team. Depending on clinical condition, the transfer was performed on conventional ventilation or, alternatively, ECMO treatment was initiated at the peripheral hospital and maintained during transportation.

We preferentially adopted a high flow technique (5-6 litres per minute of blood flow), to maximize the opportunity of providing protective ventilation, aiming to achieve a plateau pressure below 28 cm H2O and PEEP 2 cmH2O above the lower inflection point of the quasistatic pressure volume curve, regardless the delivered tidal volume (in any case less than 6 ml/kg). As consequence

of extracorporeal ventilation, respiratory frequency of the ventilator was reduced to 4-10/min to improve lung protection strategy

(18-19)

. Inspired oxygen fraction was reduced to 0.5 or lower,

whenever possible. A recruitment maneuver was performed at least once a day, and ventilation with an intermittent high pressure breath ("sigh") was adopted to improve lung aeration (20).

Equipment

The ECMO circuit consisted of a Rotaflow Maquet Centrifugal Pump (Maquet, Rastatt, Germany) and a polymethylpentene membrane oxygenator (Quadrox-D Oxygenator, Maquet, Rastatt, Germany), connected with biocoated tubes (Bioline Coating. Maquet, Rastatt Germany). For V-V ECMO two types of cannulas were used. At the beginning, a femoral-femoral cannulation was preferentially employed, with an inflow cannula ranging from 19 to 21 Fr (Bio-Medicus Medtronic, Minneapolis, MN, US) and an outflow multistage cannula, ranging from 21 to 25 Fr (HLS Venous Cannula; Maquet, Rastatt, Germany ). In case of jugular-femoral cannulation, the inflow cannula was an HLS Arterial Cannula (Maquett, Rastatt, Germany), 23 gm length, 17-21 Fr. Since July 2009, Avalon Elite™ Bi-Caval Dual Lumen Catheters have become available (Maquel, Rastatt, Germany). These specially designed dual lumen cannulas, inserted in the right internal jugular vein, permit both drainage and reinfusion of blood. Dual lumen cannulas are provided in different diameters ranging from pediatric to adult size. For adults the 27 Fr and 31 Fr are recommended. As an high flow ECMO technique was the main choice of our center, the larger 31 French cannulas were preferably used, except in very short and lean patients Heparin therapy was titrated by bedside measurement of activated partial thromboplastin time (aPTT) with Hemochron (Hemochron Jr. Sign. plus, ITC Europe, Milan, IT) every two hours.

Cannulation procedure Ultrasonography was thoroughly used to verify patency of the vessels selected for

cannulation and adequacy of transthoracic and trans-esophageal windows for cannulation monitoring. The first cannulation choice was the jugular approach for a dual lumen cannula. When a trans-esophageal examination could not be obtained a fermoral-femoral approach was preferred. A femoral-femoral approach was also chosed in case of right internal jugular vein thrombosis or anatomical abnormalitiese (i.e ipoplasia). The intensivist and the cardiac surgeon took care of the cannulation procedure and the insertion technique was percutaneous with Seldinger technique. Vascular cannulation was under direct ultrasound assistance. For dual lumen cannula, the guide-wire advancement through superior vena cava and right atrium was monitored by transesophageal bi-caval view, while progression in inferior vena cava was followed by sub-costal transthoracic view. Also cannula advancement was likewise monitored with cardiac ultrasound. In order to verify the correct final position of the cannula, its tip was measured 6-8 cm deep in the inferior vena cava from atrio-caval junction. After ECMO start a doppler flow image, from the outflow port of the cannula, was searched in the right atrium, properly directed towards the tricuspid valve and cannula position was consequently adjusted. For femoral-femoral approach, a clear view of the two guide-wire was obtained in inferior vena cava before cannula insertion and eventually the inflow cannula was advanced until its tip was visualized in the right atrium. Taking into account the skin insertion, the withdrawal cannula was inserted about 8-10 cm less than the inflow cannula, in order to prevent recirculation due to the tips of the two cannulas too close.

Study population Baseline characteristics were collected for all patients and the Sequential Organ Failure Assessment (SOFA) and the simplified acute physiology score (SAPS II) scores were calculated, respectively

(21-22)

. The presence

of

comorbidities (that is the presence of diabetes, hypertension and/or chronic obstructive

pulmonary disease and/or chronic renal disease) was determined by taking the patients’ history from family members. The term “heart disease”

indicates the presence of chronic heart failure and/or chronic ischemic heart disease and/or previous cardiac surgery.

All participants (or their kins) signed a written informed consent for storing their clinical data. The study is a retrospective analysis of data

and the study design was approved by our Institutional Review Board.

Statistical analysis

Data have been stored in a dedicated data-base and analyzed with SPSS for Windows 20.0 (SPSS Inc., Chicago, IL). A p value less than 0.05 has been considered statistically significant. Categorical variables are reported as frequencies and percentages; continuous variables are reported as mean ± standard deviation (SD).

For continuous variables, between-groups comparisons have been performed with

Student’s t-test or ANOVA (followed by Bonferroni post-tests if overall p was significant) or by means of Kruskal-Wallis H test. Categorical variables have been compared with chi-square. Univariate analysis (chi-squared or Fisher’s exact test for categorical data; Student’s t test or Mann-Whitney U

test for continuous data) was used to identify candidate variables for multivariate analysis which included those variables that resulted statistically

significant at univariate analysis.

Multivariable logistic regression was performed in order to identify predictors of ICU mortality. Hosmer-

Lemeshow goodness-of-fit test and Nagelkerke pseudo-R2 are reported.

RESULTS

The characteristics of the study population are depicted year by year in Table 1. The average distance of travel was 128.6 ± 139.8 km (range 5-407 km) and transportation was via road in all cases. No adverse event was observed during transportation. In the overall population, the transferred patients represented the 57%, with annual percentages ranging from 28% to 90%. In our population no differences were detectable in age and gender during the study period as well as in ICU length of stay and in duration of mechanical ventilation while a progressive increase in SAPS values and in the use of norepinepnhrine were detectable (p=0.048 and p=0.037,

respectively). Neuromuscular blockers were used in 51.3% of all patients,

and their use

significantly increased from 2009 to 2016 (p=0.005) while pronation was adopted since 2013. Dual lumen cannula was progressively more used (129/160, 80.6%), especially in recent years (p<0.001). Primary lung injury was the more frequent cause of disease (136/160,85%). The overall ICUmortality rate in our population was 40% (64/160) and no differences were detectable over the years during our study period. Table 2 shows the comparison between patients transferred on ECMO support and those in whom the device was implanted at our Center (local patients). No differences were observed between the two subgroups expect for the fact that transferred patients were submitted to a longer period of pre ECMO mechanical ventilation (p<0.001) and were more frequently given neuromuscular blockers (p=0.010). Mortality rates were comparable between subgroups. As shown in Table 3, non-survivors were older (p=0.045), leaner (p<0.001) and had higher values of SOFA and SAPS II (p=0.027 and p=0.003, respectively). Pre ECMO neuromuscular blockers were more frequently used in non survivors (p=0.010) who showed a longer period of ECMO support (p=0.004). At multivariate regression analysis (Table 4), BMI (body mass index) and pre ECMO neuromuscular blockers and SAPS were independent predictors for early mortality (when adjusted for age).

DISCUSSION

In the present investigation, we describe the eight year experience of our ECMO center for adult severe respiratory failure, including the activity of our ECMO mobile team. We further analysed preECMO management (comprising NMB use, pronation, and pre ECMO mechanical ventilation duration) over the study period and the changes in the type of cannulation, taking into account that since 2009 we performed all ECMO implantations guided by echocardiography.

Previous papers on ECMO mobile teams' activities for severe respiratory failure in adults are quite heterogeneous. While some investigations reports small numbers of patients

(7,23,24)

and/or

included mixed population (also pediatric patients (12), or those supported with veno-arterial ECMO (25,26)

(4,15)

), others comprised only adult patients supported by VV-ECMO all affected by H1N1 infection

. More recently, larger series of patients retrieved from peripheral centers have been described.

Broman et al

(13)

reported the four year Stockholm experience (2010-2013) of 322 transports (both

adult and pediatric patients), mainly by air (62%) and concluded that long- and short-distance interhospital transports on ECMO can be performed safely. Vaja et al

(14)

retrospectively analyzed 102

adult patients who were cannulated at their referring hospitals (2010-2014) and transported by their mobile team ( Leicestershire, UK). Transportation was via road in most cases (77%) and the 93% of patients were managed by VV-ECMO (using a dual-lumen cannula in the 70%). These Authors came to similar conclusions, that transportation of these patients was safe according to their experience, thanks to a well-trained dedicated ECMO mobile team. Different health care systems have their own strategies for how to organize and staff an ECMO retrieval organization

(4,8,27)

and geographic discrepancies were reported in the composition

of mobile ECMO teams, with European centers much more likely to incorporate an anesthesiologist in the mobile ECMO process, compared to centers in North America (28). According to our data, the introduction of our mobile ECMO during the H1N1 pandemic has allowed during the following years access to our referral Center of adult patients with severe respiratory failure coming from peripheral hospitals (also from out of our Region) who could benefit from VV-ECMO support. In keeping with previously reported experiences

(13,14)

, also in our series (which included only adult

patients treated with VV-ECMO) cannulation could be achieved in all case at the peripheral hospitals and on ECMO transportation was safe, with no adverse events. The peculiarity of our mobile ECMO team is that all implantation was guided with echocardiography which also allowed the assessment of left and right ventricle functions before the procedure as well as estimation of right ventricle dimensions and systolic pulmonary arterial pressure. We documented that

echocardiographic evaluation pre-ECMO implantation can help in risk stratifying these patients

(29)

,

since RV dilatation is an independent factor for early death, as we documented in a previous paper (29)

. In one case, the detection of severe left ventricle dysfunction (which the patient developed in

the few hours while our mobile ECMO team was arriving at the referring hospital) led to venoarterial ECMO (and not VV-ECMO) implantation. The patient was then safely transferred to our Center on VA-ECMO and eventually converted to VV-ECMO as soon as the cardiac function recovered (this patient was not included in our study population). We observed that the use of dual lumen cannula progressively increased over the study period both at our Center and at peripheral hospitals. According to our experience, echocardiographic guidance makes dual lumen cannula positioning feasible and safe since it allows the visualization of the wire and finally the cannula in the inferior vena cava. More importantly, the correct positioning can be visualized real time by color Doppler as the flow directed towards the tricuspid valve. Our series included quite a large number of adult patients with severe respiratory failure requiring VV-ECMO support, most of whom retrieved by our mobile ECMO team from peripheral hospitals, and consecutively admitted to our Referral Center over a 8 year period. Our findings allow us to reflect on potential changes in the management of these patients over this study period in our Region (since we included also patients managed at and transferred by peripheral hospitals). Firstly, the pre ECMO management of these patients changed over the years, as inferred by the increasing use of neuromuscular blockers and the introduction of pronation since 2013. Transferred patients receive more NMBA and are more frequently pronated pre ECMO, as rescue therapies are used more frequently while tacking decision to refer and while waiting for ECMO team arrival. On the contrary for local patients, for whom the ECMO resource is immediately available, ECMO is instituted

expeditiously,

and

pronation

and

NMBA

are

employed

eventually.

Finally, the workload of our referral Center and our mobile team did not change overtime thus documenting that severe respiratory failure requiring VV-ECMO support is still a clinical need.

Interestingly, mortality rate of the transferred patients is comparable to that of patients in whom VV-ECMO was implanted at our Center. This finding can be related to the fact that in our Regional scenario, peripheral centers properly managed patients with respiratory failure in the initial phase and timely referred the most severe ones. Furthermore, the management of these patients was performed at our referral center by the same team who has become over the years more experienced and skilled. Finally the mortality rate remained unchanged over the study period at our Center, despite the fact that more serious patients were treated as shown by the increasing values of SAPS II and the increasing use of norepinephrine and pre ECMO NMB. Indeed SAPS II and pre ECMO NMB were independent predictors for in-ICU mortality in our series together with BMI, as we recently reported (30).

Limitations of the study This is a single center experience. However, our investigation reports quite a large series of adult patients with severe respiratory failure requiring VV-ECMO, most of whom were transferred by our mobile ECMO team. Despite geographic differences in health care organization, our data strongly suggest that the availability of a mobile ECMO team makes ECMO treatment effectively available to ARDS patients refractory to conventional treatment, independently of the hospital they are firstly admitted and confirm that transportation on ECMO support is safe.

No conflict of interest

AKNOWLEDGEMENTS: none

REFERENCES

1.

Biscotti M, Agerstrand C, Abrams D, et al. One Hundred Transports on Extracorporeal Support to an Extracorporeal Membrane

Oxygenation Center. Ann Thorac Surg 2015 ;100(1):34-9.

2. Allen S, Holena D, McCunn M, et al. A review of the fundamental principles and evidence base in the use of extracorporeal membrane oxygenation (ECMO) in critically ill adult patients. J Intens Care Med. 2011;26:13–26.

3.

Sherren PB, Shepherd SJ, Glover GW, et al. Capabilities of a mobile extracorporeal membrane oxygenation service for severe respiratory

failure delivered by intensive care specialists. Anaesthesia 2015; 70; 707-714

4. Beurtheret S, Mordant P, Paoletti X, et al. Emergency circulatory support in refractory cardiogenic shock patients in remote institutions: a pilot study (the cardiac-RESCUE program). Eur Heart J. 2013;34:112–20.

5. Gariboldi V, Grisoli D, Tarmiz A, et al. Mobile extracorporeal membrane oxygenation unit expands cardiac assist surgical programs. Ann Thorac Surg. 2010;90:1548–52.

6. Haneya A, Philipp A, Foltan M,

et al. Extracorporeal

circulatory systems in the interhospital

transfer of critically ill patients: experience of a single institution. Ann Saudi Med. 2009;29:110–4.

7.

Isgrò S, Patroniti N, Bombino M, et al. Extracorporeal membrane oxygenation for interhospital transfer of severe acute respiratory distress syndrome patients: 5-year experience. Int J Artif Organs. 2011;34:1052 – 60.

13

8. Lunz D, Philipp A, Judemann K, et al. First experience with the deltastream DP3 in venovenous extracorporeal membrane oxygenation and air-supported inter-hospital transport. Interact Cardiovasc Thorac Surg. 2013;17:773 – 7.

9. Lucchini A, De Felippis C, Elli S, et al. Mobile ECMO team for inter-hospital transportation of patients with ARDS: a retrospective case series. Heart Lung Vessel. 2014;6:262 – 73.

10.Foley DS, Pranikoff T, Younger JG, et al. A review of 100 patients transported on extracorporeal life support. ASAIO J. 2002;48:612 – 9.

11.Bryner B, Cooley E, Copenhaver W, et al. Two decades ’ experience with interfacility transport on extracorporeal membrane oxygenation. Ann Thorac Surg. 2014;98:1363 – 70.

12.Linden V, Palmer K, Reinhard J, et al. Inter-hospital transportation of patients with severe acute respiratory failure on extracorporeal membrane oxygenation – national and international experience. Intensive Care Med. 2001;27:1643 – 8.

13.Broman LM, Holzgraefe B, et al. The Stockholm experience: interhospital transports on extracorporeal membrane

oxygenation . Critical

Care 2015; 19:278

14.Vaja R, Chauhan I, Joshi V, et al. Five-year experience with mobile adult extracorporeal membrane

oxygenation in a tertiary

referral center. Journal of Critical Care 2015; 30: 1195–1198.

15.Ciapetti M, Cianchi G, Zagli G, et al. Feasibility of inter-hospital transportation using extracorporeal membrane oxygenation (ECMO) support of patients affected by severe swineflu(H1N1)-related ARDS. Scand J Trauma Resusc Emerg Med.2011; 19:32

16.Bonizzoli M, Lazzeri C, Cianchi G, et al . Serial Lactate Measurements as a Prognostic Tool in Venovenous Extracorporeal Membrane Oxygenation Support. Ann Thorac Surg. 2016 Sep 22. pii: S0003-4975(16)30850-5. doi: 10.1016/j.athoracsur.2016.06.087.

17.Cianchi G, Bonizzoli M, Pasquini A, et al. Ventilatory and ECMO treatment of H1N1induced severe respiratory failure: results of an Italian referral ECMO center BMC Pulmonary Medicine 2011, 11:2

18.Marini JJ1, Hotchkiss JR, Broccard AF. Bench-to-bedside review: microvascular and airspace linkage in ventilator-induced lung injury.Crit Care. 2003 Dec;7(6):435-44.

19. Gattinoni L, Tonetti T, Cressoni M, et al. Ventilator-related causes of lung injury: the mechanical power. Intensive Care Med. 2016 Oct;42(10):1567-75.

20.Valente Barbas CS. Lung recruitment maneuvers in acute respiratory distress syndrome and facilitating resolution. Crit Care Med 2003, 31:S265-271.

21.Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology score (SAPS II) based on a European / North American multicenter study. JAMA. 1993;270:2957-63

22.Vincent JL, Moreno R, Takala J, et al on behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996;22:707–10.

23.Javidfar J, Brodie D, Takayama H, et al. Safe transport of critically ill adult patients on extracorporeal membrane oxygenation support to a regional extracorporeal membrane oxygenation center. ASAIO J. 2011;57:421 – 5.

24.Hodgson CL, Hayes K, Everard T, et al. Long-term quality of life in patients with acute respiratory distress syndrome requiring extracorporeal membrane oxygenation for refractory hypoxaemia. Critical Care 2012, 16:R202

25. Dalmau MJ, Beiras-Fernandez A, Kelle H, et al. Initiation of an Inter-hospital ECMO Transfer Program for Patients Suffering From Severe Acute Heart and/or Pulmonary FailurRev Esp Cardiol. 2014;67(4):324–330.

26.Lindskov C, Jensen RH, Sprogoe P, et al.Extracorporeal membrane oxygenation in adult patients with severe acute respiratory failure Acta Anaesthesiol Scand 2013; 57: 303–311.

27.Forrest P, Ratchford J, Burns B, et al.Retrieval of critically ill adults using extracorporeal membrane oxygenation: an Australian experience. Intensive Care Med. 2011;37:324 – 830.

28.Nwozuzu A, Fontes ML, Schonberger RB. Mobile Extracorporeal Membrane Oxygenation Teams: The North American Versus the European Experience J Cardiothorac Vasc Anesth 2016; 30(6): 1441-1448.

29.Lazzeri C, Cianchi G, Bonizzoli M, et al.Right ventricle dilation as a prognostic factor in refractory ARDS requiring VV-ECMO. Minerva Anestesiol. 2016 Mar 9

30.Lazzeri C, Bonizzoli M, Cianchi G, Batacchi S, Terenzi P, Cozzolino M, Bernardo P, Peris A.Body mass index and echocardiography in refractory ARDS treated with veno-venous extracorporeal membrane oxygenation. J Artif Organs. 2017 Mar;20(1):50-56

LEGEND FOR FIGURE

Figure 1: A: Image of the ambulance. B: Self-loading stretcher and metal frame for mounting medical equipment. C and D: details of medical equipment installation in the ambulance cabin in trip configuration.

Table 1 Characteristics of the study population year by year throughout the study period.

2009 2010 2011 2012 2013 2014 2015 2016 All patients n.160 Transferred patients, n (%) Age (yrs), mean±SD

11

18

20

7 (64)

5 (28)

9 (45)

51 ± 16.8

52 ± 15.8

52 ± 14.5

18

28

29

17

11 (61) 11 (39) 26 (90) 14 (82) 55 ± 13.4

46 ± 16.9

50 ± 13.4

47 ± 13.7

p value

19 0.251 8 (42) (Pearson’s R2 0.21) 55±14.0 0.378 (A)

Males, n (%)

10 (91) 14 (78) 14 (70) 13 (72) 18 (64) 20 (69) 10 (59) 10 (53) 0.432 (χ)

SOFA

11.5 ± 3.6

10.5 ± 2.9

10.6 ± 1.9

10.7 ± 3.2

10.9 ± 3.7

10.7 ± 3.1

11.6 ± 3.1

8.5 ± 5.2

SAPS II

46.7 ± 20.1

41.5 ± 17.4

34.8 ± 13.4

52.3 ± 24.9

46.2 ± 21.0

37.4 ± 17.4

51.0 ± 17.2

45.2 ± 0.048 (A) 19.4

Pre ECMO mechanical ventilation (hrs), median (IQR)

72 (48- 48 (12- 24 (24- 21 (12- 21 (12- 48 (24- 84 (48- 24 (1284) 96) 48) 48) 96) 96) 108) 48)

Pre ECMO Pronation, n (%)

0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 2 (7.1) 2 (6.9) 5 (29.4) 2 (10.5) 0.009 (χ)

Pre ECMO NMBA (n)

5 10 9 (52.9) (45.5) (50.0)

Pre-ECMO norepinephrine, n (%)

1 (9)

6 (33)

8

17

18

16

22

25

13

12

H1N1

7

16

8

3

6

4

6

6

Bacterial

1

1

6

11

14

15

7

4

4

2

2

6

0

2

10 (55.6)

10 (35.7)

20 (69.0)

0.287 (A)

0.002 (KW)

14 4 (21.1) 0.005 (χ) (82.4)

11 (55) 12 (67) 16 (57) 15 (52) 10 (59) 13 (68) 0.037 (χ)

Cause of lung injury Primary (n)

Other Secondary (n.)

3

1

2

2

6

4

4

7

Septic shock

0

0

2

2

2

2

2

1

0

1

0

0

2

1

1

3

0

0

0

0

2

1

1

3

Trauma Cystic fibrosis

N/A

N/A

other P/F < 100

3

0

0

0

0

0

0

0

8

12

14

14

18

22

17

12

0.217 (χ)

Cannulation (n.) Dual lumen cannula (%)

5 (45) 10 (55) 12 (60) 16 (89) 26 (93) 26 (90) 16 (94) 18 (95)

Femur femoral/femurjugular (%)

6 (55)

8 (45)

8 (40)

2 (11)

2 (7)

3 (10)

ECMO (days) , 9.5 (6median (IQR) 11)

10 (717)

8 (1124.5)

10 (519)

9.5 (313)

10 (816)

1 (6)

0.001 (χ)

1 (5)

22 (11- 11 (3.50.066 (A) 35) 24)

Ventilation 13.5 19 (12- 19 (11- 18.5 (9- 13 (3.5- 16 (12- 23 (15- 20 (10(days) , median 0.323 (A) (10-26) 26) 28) 31) 23) 23) 35) 40) (IQR) ICU LOS 17 (15- 20 (14- 21 (14- 19 (9- 17 (5.5- 17 (12- 24 (16(days) , median 20) 29) 30) 34) 26) 28) 35) (IQR)

19 (942)

0.706 (A)

ICU mortality, n (%)

8 (42)

0.883 (χ)

4 (36)

6 (33)

9 (45)

7 (39) 12 (43) 11 (38) 10 (59)

(A): one-way ANOVA; (χ): chi-squared; IQR: interquartile range (25th-75th pct); (KW): KruskalWallis; N/A: not available

SAPS II: simplified acute physiology score; SOFA: simplified organ failure assessment; NMBA: neuromuscular blocker agents; ECMO: extracorporeal membrane oxygenation; P/F= PaO2/FiO2; ICU: intensive Care Unit; LOS: length of stay

Table 2 Comparison between transferred and local patients. p values

Local patients

Transferred patients

69

91

51.8 ± 16.9

50.1 ± 13.1

0.468 (t)

Males, n (%)

45 (71%)

64 (70%)

0.605 (χ)

BMI

26.7 ± 7.2

28.9 ± 8.2

0.082 (t)

Smoking

8

10

0.654 (χ)

Diabetes

8

14

Hypertension

12

13

Heart disease

7

8

COPD

6

4

Other

19

12

SOFA, mean±SD

10.8 ± 3.6

10.5 ± 3.2

0.601 (t)

SAPS II, mean±SD

44.6 ± 19.8

43.2 ± 19.4

0.647 (t)

Pre ECMO mechanical ventilation (hrs) median (IQR)

24 (12-72)

48 (24-96)

<0.001 (U)

2 (3)

9 (10)

0.088 (χ)

27 (40)

55 (60)

0.010 (χ)

60

71

0.566 (χ)

H1N1

28

28

Bacterial

24

35

Other

8

8

9

20

Septic shock

3

8

Trauma

2

6

Cystic fibrosis

2

5

Other

2

1

49 (71.0)

68 (74.7)

0.731 (χ)

Norepinephrine (ug/kg/min), mean±SD

0.51± 0.58

0.32 ± 0.23

0.069 (t)

Dobutamine (ug/kg/min) mean±SD

1.60± 1.52

2.43 ± 0.73

0.262 (t)

Number Age (yrs), mean±SD

Pre-existing conditions

Pre ECMOPronation, n (%) Pre ECMO NMBA, n (%) Cause of lung injury (n.) Primary

Secondary

P/F <100

0.570 (χ)

Vasopressors

Cannulation (n.) Dual lumen cannula

56

74

Femur femoral/femur- jugular

13

17

0.979 (χ)

ECMO (days) median (IQR)

10 (4.5-16.5)

11 (8-19)

0.192 (U)

Ventilation (days) median (IQR)

14.5 (8-28)

16 (11-28)

0.258 (U)

ICU LOS (days) median (IQR)

20 (8-34.5)

19 (13.5-29.5)

0.715 (U)

30 (43.5)

37 (40.7)

0.720 (χ)

Mortality, n (%)

COPD: chronic Obstructive Pulmonary Disease; ECMO: extracorporeal membrane oxygenation; ICU: intensive care unit: SAPS II: simplified acute physiology score; SOFA: simplified organ failure assessment; BMI: body mass index; LOS: length of stay; P/F= PaO2/FiO2.; NMBA:

neuromuscular blocker agents IQR: interquartile range T= Student's t test; χ=chi-square; U: Mann-Whitney

Table 3 Comparison between survivors and non survivors p values

Survivors

Non Survivors

93 (58.1)

67 (41.9)

Age (yrs), mean±SD

48.8 ± 15.1

53.6 ± 14.2

0.045 (t)

BMI

29.9 ± 9.1

25.3 ± 4.5

<0.001 (t)

Males, n (%)

62 (66%)

47 (70%)

0.768 (χ)

Smoking

10

8

0.927(χ)

Diabetes

11

11

Hypertension

10

15

Heart disease

8

7

COPD

5

5

Other

14

17

SOFA, mean±SD

10.1 ± 3.3

11.3 ± 3.4

0.027 (t)

SAPS II, mean±SD

39.9 ± 18.8

49.2 ± 19.4

0.003 (t)

Pre ECMO mechanical ventilation (hrs) median (IQR)

36 (15-72)

48 (18-96)

0.236 (U)

4 (4)

7 (11)

0.123 (χ)

40 (43.0)

42 (63.6)

0.010 (χ)

74

57

0.297 (χ)

H1N1

36

20

Bacterial

30

29

Other

8

8

19

10

Septic shock

6

5

Trauma

4

4

Cystic fibrosis

6

1

Other

3

0

70 (75.3)

47 (70.1)

0.589 (χ)

Dual lumen cannula

72

58

0.143(χ)

Femur femoral/femur- jugular

21

9

0.31± 0.28

0.50 ± 0.52

Number (%)

Pre-existing conditions

Pre ECMOPronation, n (%) Pre ECMO NMBA, n (%) Cause of lung injury (n.) Primary

Secondary

P/F <100, n (%)

0.484(χ)

Cannulation (n.)

Vasopressors Norepinephrine (ug/kg/min) n.73, mean±SD

0.054 (t)

Dobutamine n.13 (ug/kg/min) mean±SD

2.28± 1.23

1.53 ± 1.05

0.312 (t)

9.5 (6-13)

15 (8-25)

0.004 (U)

Ventilation (days) median (IQR)

15.5 (11-24)

18 (10-33.5)

0.481 (U)

ICU LOS (days) median (IQR)

19 (13-29)

18.5 (10-34)

0.763 (U)

ECMO (days) median (IQR)

ECMO: extracorporeal membrane oxygenation; ICU: intensive care unit: SAPS II: simplified acute physiology score; SOFA: simplified organf failure assessment; BMI: body mass index; LOS: length of stay; NMBA: neuromuscular blocker agents; IQR: interquartile range; t= Student's t test; χ=chi-square test; U: Mann-Whitney U test

Table 4 Multivariate regression analysis Unadjusted OR

95%CI

Wald

p

BMI (1Kg/m2 step)

0.91

0.86-0.96

11.256

<0.001

Pre ECMO NMBA

2.32

1.21-4.43

6.469

0.011

SOFA(1 unit step)

1.12

1.01-1.23

4.717

0.030

SAPS (1 unit step)

1.03

1.01-1.04

8.326

0.004

Age (1 year step)

1.02

1.00-1.05

3.947

0.047

Pre ECMO ventilation (1 hour step)

1.006

0.998-1.014

2.017

0.155

ECMO support (1 day step)

1.02

1.00-1.04

2.971

0.085

Adjusted OR

95%CI

p

BMI (1Kg/m step)

0.89

0.83-0.95

11.166

<0.001

Pre ECMO NMBA

2.28

1.11-4.65

5.076

0.024

SAPS (1 unit step)

1.02

1.00-1.04

4.602

0.032

Age (1 year step)

1.02

0.99-1.04

1.967

0.161

2

Hosmer-Lemeshow χ2 7.81, p=0.917 Nagelkerke pseudo-R2 0.25 BMI: body mass index; ECMO: extracorporeal membrane oxygenation, SAPS: simplified acute physiology score; NMBA: neuromuscular blocker agents