Morbid obesity is not a contraindication to transport on extracorporeal support

Morbid obesity is not a contraindication to transport on extracorporeal support Abstract OBJECTIVES Extracorporeal membrane oxygenation (ECMO) transport has not been described in morbidly obese patients, a population that can pose significant challenges in obtaining vascular access, indexed flows and transport logistics. We sought to study the feasibility and safety of transporting obese and morbidly obese patients during extracorporeal support. METHODS We conducted a retrospective review of all patients transported to our institution while receiving ECMO from September 2008 to September 2016. Survival to decannulation and survival to discharge were the primary outcomes. Obesity and morbid obesity were defined as a body mass index of greater than 30 kg/m2 and greater than 40 kg/m2, respectively. RESULTS From 2008 to 2016, 222 patients were transported to our institution while receiving ECMO. Among these included patients, 131 were non-obese (interquartile range 22–27 kg/m2), 63 were obese (interquartile range 31–35 kg/m2) and 28 were morbidly obese (interquartile range 41–49 kg/m2), with 6 patients having a body mass index greater than 50 kg/m2 (range 52.3–79 kg/m2). Pre-ECMO arterial blood gases, disease severity indices, cannulation strategies and transport distances were similar between these 3 groups. There was no mortality of patients during transport, and survival to discharge was 66% (n = 87) in non-obese patients, 56% (n = 35) in obese patients and 82% (n = 23) in morbidly obese patients (P = 0.042). On multivariable logistic regression analysis, body mass index was not a predictor of in-hospital mortality (odds ratio 0.99, 95% confidence interval 0.95–1.03; P = 0.517). CONCLUSIONS Transport of morbidly obese patients receiving ECMO may be performed safely and with excellent results in the setting of a dedicated ECMO transport programme with well-established management protocols. Extracorporeal membrane oxygenation, Respiratory failure, Obesity INTRODUCTION Obesity, defined as a body mass index (BMI) greater than 30 kg/m2, is becoming increasingly common, and currently, more than one-third of Americans are obese [1]. Obese patients pose unique challenges to surgeons and intensivists. In addition to comorbidities that tend to accompany obesity, such as diabetes mellitus and hypertension, there appears to be increased morbidity, increased overall hospital and intensive care unit (ICU) lengths of stay (LOS) and longer durations of mechanical ventilator support [2]. The treatment of these patients with venovenous extracorporeal membrane oxygenation (VV-ECMO) and venoarterial ECMO (VA-ECMO; also known as extracorporeal life support) remains particularly problematic for several reasons: large body size causes difficulty in achieving sufficient indexed flows in the circuit, excess weight and tissue affect patient positioning and anatomical landmark identification and weight restrictions limit radiology scans for many obese patients. Despite these limitations, obesity and extreme obesity have not been found to be significant risk factors for hospital mortality in the setting of acute lung failure and are not contraindications for ECMO [3–5]. The ability to transport patients during ECMO is critical to provide advanced care or transplantation to patients with refractory cardiac or respiratory failure. However, the transportation of obese patients receiving ECMO has not been described in the literature and presents its own unique set of challenges. We hypothesized that patients on ECMO could be transported safely, regardless of the BMI. The purpose of this study was to evaluate the outcomes in transporting obese and morbidly obese patients. MATERIALS AND METHODS Patients A retrospective cohort study was conducted using clinical data of all patients transported to our institution on either VV-ECMO or VA-ECMO from September 2008 to September 2016. These included patients cannulated at outside hospitals by our mobile ECMO team or by outside hospital physicians as well as patients from our most recent ECMO transport publication. This study was approved by the Institutional Review Board of Columbia University Medical Center with waiver of informed consent. The use of ECMO, as described in this article, is off-label. Predictor variables The primary predictor of interest was obesity, defined as a BMI of ≥30 kg/m2. Other variables considered included morbid obesity (BMI ≥40 kg/m2), gender, age, indication for ECMO, cannulation team (outside hospital versus our institution), travel distance and Acute Physiology and Chronic Health Evaluation II (APACHE II) score. Primary outcomes The primary outcome was survival to discharge. The secondary outcomes included survival during transport, survival to decannulation, 30-day survival, LOS, the need for tracheostomy, post-cannulation arterial gases, and percent predicted ECMO flow. We calculate this by dividing a patient’s initial flow on ECMO by that patient’s predicted cardiac output for a cardiac index of 2.4 l/min/m2. We also conducted exploratory secondary analysis of outcomes in super obese patients (BMI >50 kg/m2). Extracorporeal membrane oxygenation management The details of our ECMO transport procedures have been previously described [5]. Indications for VV-ECMO include severe, refractory, potentially reversible hypoxaemic or hypercapnic respiratory failure that meets the following criteria: partial pressure of arterial oxygen (PaO2) to fraction of inspired oxygen ratio <80 mmHg, severe uncompensated respiratory acidosis (pH <7.15) or excessively high airway pressures >35–45 cmH2O, despite optimized ventilator management. Our VA-ECMO indications for refractory cardiogenic shock include evidence of end-organ hypoperfusion with a systolic blood pressure of <90 mmHg, a cardiac index of <2.0 l/min/m2 or the pulmonary capillary wedge pressure of >16 mmHg. Briefly, lung protective mechanical ventilation is employed after ECMO initiation at the outside hospital. Tidal volumes are reduced to maintain peak airway pressures less than 35 mmHg, respiratory rates decreased to less than 20 breaths/min and sweep gas flow rate is adjusted to maintain a pH between 7.35 and 7.45. Heparin is used for anticoagulation with a target activated partial thromboplastin time of 40–60 s. Our standard ECMO device is the Cardiohelp system (Maquet, Wayne, NJ, USA), though our first transports relied upon the Rotaflow pump and Quadrox oxygenator (Maquet). Although there were no differences in the performance, we found a significant difference in the ergonomics of transport using the lower profile Cardiohelp system. On arrival to the ICU, care is transitioned to the ICU team with consultation from the ECMO team. Patients in the ICU are managed with the ultraprotective lung ventilation as tolerated: volume assist–control ventilation with a set respiratory rate of 2–10 breaths/min, plateau airway pressures <20 cm of water, if possible, or at least <25 cm of water and with moderate levels of positive end-expiratory pressure, typically 10–15 cm of water. Additional drainage cannulas, typically a short 21- or 23-Fr Bio-Medicus cannula (Medtronic, Minneapolis, MN, USA), may be placed in the contralateral femoral vein if additional ECMO support is required upon arrival to the ICU. Statistical analysis All statistical analyses were performed with SPSS 20.0 (IBM SPSS Inc., Armonk, NY, USA) and R statistics software (R Core Team, Vienna, Austria). Clinical and demographic variables were presented using standard summary statistics, including mean and standard deviation or median and interquartile range (IQR) depending on the normality of distribution for continuous variables and frequencies and proportions for categorical variables. The distributions of continuous variables were tested with the Shapiro–Wilk test. To assess the predictor variables and major morbidity, the χ2 test or the Fisher’s exact test (if the expected value of a 4 × 4 table cell was <5) was used for categorical variables. The Student’s t-test, analysis of variance or the Mann–Whitney U-test for continuous variables using 2 sample means, 3 or more sample means and in cases where sample medians with skewed distributions were used, respectively. These criteria were prespecified prior to statistical testing. Variables with P-values ≤0.10 in univariable analyses were analysed in a multivariable logistic model with an a priori decision to automatically include BMI in this model. An additional multivariable model was built using forward stepwise selection, where variables were included if associated P-values met the significance criterion (P < 0.05) or if their inclusion changed the magnitude of the coefficients by ≥10%. RESULTS All 222 patients were successfully placed on VV-ECMO or VA-ECMO (extracorporeal life support) via percutaneous cannulation and transported to our centre alive. Of these included patients, 131 (59%) were non-obese, 63 (28%) had BMIs ≥30 to ≤39 and 28 (13%) patients with a BMI ≥40 kg/m2. There were no significant demographic differences between these 3 groups, although there was a trend towards younger ages in the morbidly obese cohort. Moreover, there was a trend towards a lower incidence of cardiogenic shock in the morbidly obesity group compared with the non-obese and obese groups. There were no significant inter-group differences in method or distance of transportation and in illness severity as measured by the APACHE II score, upon presentation to our institution. Baseline characteristics are reported in Table 1. Table 1: Characteristics and outcomes of ECMO patients stratified by BMI     BMI (kg/m2)     Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Age (years)  42 (27–55)  41 (26–56)  47 (33–56)  36 (27–47)  0.076  Male gender, n (%)  131 (59.0)  74 (56.5)  41 (65.1)  16 (57.1)  0.514  BMI (kg/m2)  28 (24–34)  25 (22–27)  34 (31–35)  44 (41–49)  <0.001  Cannulated by our team, n (%)  174 (78.4)  101 (77.1)  51 (81.0)  22 (78.6)  0.845  APACHE II score  29 (25–34)  30 (25–34)  28 (26–34)  28 (24–32)  0.871  Transported safely, n (%)  222 (100)  131 (100)  63 (100)  28 (100)  1.000  Median distance (miles)  16 (8–32)  15 (7–32)  17 (10–31)  16 (7–32)  0.377  Mode of transportation          1.000a   Ambulance, n (%)  218 (98.2)  128 (97.7)  61 (98.4)  28 (100)     Fixed-wing aircraft, n (%)  4 (1.8)  3 (2.3)  1 (1.6)  0 (0)        BMI (kg/m2)     Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Age (years)  42 (27–55)  41 (26–56)  47 (33–56)  36 (27–47)  0.076  Male gender, n (%)  131 (59.0)  74 (56.5)  41 (65.1)  16 (57.1)  0.514  BMI (kg/m2)  28 (24–34)  25 (22–27)  34 (31–35)  44 (41–49)  <0.001  Cannulated by our team, n (%)  174 (78.4)  101 (77.1)  51 (81.0)  22 (78.6)  0.845  APACHE II score  29 (25–34)  30 (25–34)  28 (26–34)  28 (24–32)  0.871  Transported safely, n (%)  222 (100)  131 (100)  63 (100)  28 (100)  1.000  Median distance (miles)  16 (8–32)  15 (7–32)  17 (10–31)  16 (7–32)  0.377  Mode of transportation          1.000a   Ambulance, n (%)  218 (98.2)  128 (97.7)  61 (98.4)  28 (100)     Fixed-wing aircraft, n (%)  4 (1.8)  3 (2.3)  1 (1.6)  0 (0)    Data are represented as median (interquartile range) for non-normally distributed data and mean ± standard deviation for normally distributed data unless otherwise specified. P-values represent statistical test results between the 3 BMI cohorts. a By the Fisher’s exact test. APACHE II: Acute Physiology and Chronic Health Evaluation II; BMI: body mass index; ECMO: extracorporeal membrane oxygenation. Cannulation strategies were not found to be significantly different between the BMI groups. Among the 60 patients placed on VA-ECMO (extracorporeal life support), 42 were cannulated in the bilateral femoral arteries, 14 were cannulated centrally, 2 via femoral vein to ipsilateral axillary artery and 2 via right internal jugular vein to the ipsilateral axillary artery. VV-ECMO patients were largely cannulated via the femoral vein to ipsilateral internal jugular vein (n = 132/162), with the remainder having a dual-lumen cannula placed in the right internal jugular vein. There was a non-statistically significant increase in the dual-lumen cannula use among VV-ECMO patients with lower BMI (24% of those <30 kg/m2 vs 11% of those 30–39 kg/m2 vs 9% of the patients with BMI ≥40 kg/m2; P = 0.119). There was no difference in drainage cannula sizes, and precannulation blood gases were comparable between the groups (Table 2). Following cannulation, there were significantly lower post-cannulation arterial blood oxygenation (PaO2) levels in the morbidly obese group [72 mmHg (IQR 60–118 mmHg) vs 111 mmHg (83–189 mmHg) in the obese cohort and 125 mmHg (86–277 mmHg) in the non-obese cohort, P = 0.001]. As the BMI increased, the initial flows on extracorporeal support were significantly lower than that predicted. Morbidly obese patients, on average, only achieved 71% (SD 15%) of their predicted flows compared with the non-obese patients who achieved 87% (SD 19%). Additional venous drainage cannulas were added to the circuits of 5 (8%) of the obese and morbidly obese patients more than 24 h after transport to improve the blood flow rate and optimize gas exchange. After the addition of these cannulas, flows were augmented by an average of 65%, achieving greater than 88% of predicted flow rates in all patients. Notably, there were no additional venous cannulas required in the non-obese patients. Table 2: ECMO results stratified by BMI     BMI (kg/m2)   ECMO characteristics  Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Indication for ECMO          0.308   Respiratory failure, n (%)  124 (55.9)  71 (54.2)  34 (54.0)  19 (67.9)  0.417   ARDS, n (%)  106 (85.5)  60 (84.5)  30 (88.2)  16 (84.2)     Other, n (%)  18 (14.5)  11 (15.5)  4 (11.8)  3 (15.8)     Cardiogenic shock, n (%)  52 (23.4)  33 (25.2)  17 (27.0)  2 (7.1)  0.092   Post-cardiac arrest, n (%)  28 (53.8)  20 (60.6)  7 (41.2)  1 (50.0)     Post-cardiotomy, n (%)  8 (15.3)  4 (12.1)  4 (23.5)  0 (0.0)     Heart failure, n (%)  7 (13.5)  3 (9.1)  3 (17.6)  1 (50.0)     Other, n (%)  9 (17.4)  6 (18.2)  3 (17.6)  0 (0.0)     Combined, n (%)  46 (20.7)  27 (20.6)  12 (19.0)  7 (25)  0.813  Venovenous ECMO, n (%)  162 (73.0)  95 (72.5)  45 (71.4)  22 (78.6)  0.784  Venoarterial ECMO (ECLS), n (%)  60 (27.0)  36 (27.5)  18 (28.6)  6 (21.4)  0.784  Pre-ECMO ventilation days  1 (0–4)  1 (0–4)  1 (1–3)  1 (0–5)  0.769  Drainage cannula (Fr)  23 (23–23)  23 (23–23)  23 (23–23)  23 (23–23)  0.346  Arterial cannula (Fr)  18 (15–20)  17 (15–19)  19 (15–22)  22 (19–22)a  0.117  Venous return cannula (Fr)  20 (20–22)  20 (20–22)  20 (20–20)  20 (20–22)  0.288  Additional venous drainage, n (%)  5 (2.2)  0 (0.0)  3 (4.8)  2 (7.1)  0.008b  ECMO duration (days)  8 (4–14)  8 (4–13)  9 (5–14)  9 (5–19)  0.599  Initial flow (l/min)  3.93 ± 0.85  3.83 ± 0.89  4.04 ± 0.77  4.07 ± 0.76  0.203  Predicted ECMO flow  4.38 ± 1.57  3.97 ± 1.42  4.70 ± 1.51  5.54 ± 1.38  <0.001  % Predicted flow  82 ± 18%  87 ± 19%  78 ± 14%  71 ± 15%  <0.001      BMI (kg/m2)   ECMO characteristics  Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Indication for ECMO          0.308   Respiratory failure, n (%)  124 (55.9)  71 (54.2)  34 (54.0)  19 (67.9)  0.417   ARDS, n (%)  106 (85.5)  60 (84.5)  30 (88.2)  16 (84.2)     Other, n (%)  18 (14.5)  11 (15.5)  4 (11.8)  3 (15.8)     Cardiogenic shock, n (%)  52 (23.4)  33 (25.2)  17 (27.0)  2 (7.1)  0.092   Post-cardiac arrest, n (%)  28 (53.8)  20 (60.6)  7 (41.2)  1 (50.0)     Post-cardiotomy, n (%)  8 (15.3)  4 (12.1)  4 (23.5)  0 (0.0)     Heart failure, n (%)  7 (13.5)  3 (9.1)  3 (17.6)  1 (50.0)     Other, n (%)  9 (17.4)  6 (18.2)  3 (17.6)  0 (0.0)     Combined, n (%)  46 (20.7)  27 (20.6)  12 (19.0)  7 (25)  0.813  Venovenous ECMO, n (%)  162 (73.0)  95 (72.5)  45 (71.4)  22 (78.6)  0.784  Venoarterial ECMO (ECLS), n (%)  60 (27.0)  36 (27.5)  18 (28.6)  6 (21.4)  0.784  Pre-ECMO ventilation days  1 (0–4)  1 (0–4)  1 (1–3)  1 (0–5)  0.769  Drainage cannula (Fr)  23 (23–23)  23 (23–23)  23 (23–23)  23 (23–23)  0.346  Arterial cannula (Fr)  18 (15–20)  17 (15–19)  19 (15–22)  22 (19–22)a  0.117  Venous return cannula (Fr)  20 (20–22)  20 (20–22)  20 (20–20)  20 (20–22)  0.288  Additional venous drainage, n (%)  5 (2.2)  0 (0.0)  3 (4.8)  2 (7.1)  0.008b  ECMO duration (days)  8 (4–14)  8 (4–13)  9 (5–14)  9 (5–19)  0.599  Initial flow (l/min)  3.93 ± 0.85  3.83 ± 0.89  4.04 ± 0.77  4.07 ± 0.76  0.203  Predicted ECMO flow  4.38 ± 1.57  3.97 ± 1.42  4.70 ± 1.51  5.54 ± 1.38  <0.001  % Predicted flow  82 ± 18%  87 ± 19%  78 ± 14%  71 ± 15%  <0.001  Data are represented as median (interquartile range) for non-normally distributed data and mean ± standard deviation for normally distributed data unless otherwise specified. P-values represent statistical test results between the 3 BMI cohorts. a There were 5 morbidly obese patients on venoarterial-ECMO: 1 was peripherally cannulated and 4 were centrally cannulated. b By the Fisher’s exact test. BMI: body mass index; ECLS: extracorporeal life support; ECMO: extracorporeal membrane oxygenation. The overall median duration of ECMO support was 8 days (IQR 4–14 days), with no significant differences between any of the 3 BMI groups (Table 3). Morbidly obese patients were noted to have longer ICU and overall hospital LOSs compared with the obese and the non-obese patients; however, these results did not achieve statistical significance. There were no differences in the rate of tracheostomy procedures performed between the 3 groups and no difference in the incidence of deep vein thrombosis at the cannulation sites (either the femoral vein or the internal jugular vein). Three non-obese patients exhibited signs of lower-limb ischaemia, requiring 2 patients to undergo fasciotomies and 1 patient to have a distal perfusion cannula inserted upon arrival to our ICU. No pulmonary emboli were noted in any of the patients studied. There did, however, appear to be a survival advantage, unadjusted for indication, in morbidly obese patients with 23 (82%) patients surviving to hospital discharge compared with 35 (56%) obese patients and 87 (66%) non-obese patients (P = 0.042). Table 3: Patient outcomes in ECMO transport stratified by BMI   BMI (kg/m2)   Outcomes  Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Hospital length of stay (days)  28 (12–47)  27 (13–46)  21 (10–46)  37 (24–61)  0.065  ICU length of stay (days)  18 (9–30)  18 (9–28)  15 (9–29)  24 (15–42)  0.126  Tracheostomy, n (%)  110 (49.5)  62 (47.3)  33 (52.4)  15 (53.6)  0.649  Deep vein thrombosis, n (%)  49 (22.1)  28 (21.3)  15 (23.8)  6 (21.4)  0.883  Survival to decannulation, n (%)  174 (78.7)  105 (80.2)  45 (71.4)  24 (85.7)  0.204  30-day survival  159 (71.9)  96 (73.3)  39 (61.9)  24 (85.7)  0.047  Survival to discharge, n (%)  145 (65.6)  87 (66.4)  35 (55.6)  23 (82.1)  0.042    BMI (kg/m2)   Outcomes  Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Hospital length of stay (days)  28 (12–47)  27 (13–46)  21 (10–46)  37 (24–61)  0.065  ICU length of stay (days)  18 (9–30)  18 (9–28)  15 (9–29)  24 (15–42)  0.126  Tracheostomy, n (%)  110 (49.5)  62 (47.3)  33 (52.4)  15 (53.6)  0.649  Deep vein thrombosis, n (%)  49 (22.1)  28 (21.3)  15 (23.8)  6 (21.4)  0.883  Survival to decannulation, n (%)  174 (78.7)  105 (80.2)  45 (71.4)  24 (85.7)  0.204  30-day survival  159 (71.9)  96 (73.3)  39 (61.9)  24 (85.7)  0.047  Survival to discharge, n (%)  145 (65.6)  87 (66.4)  35 (55.6)  23 (82.1)  0.042  Data are represented as median (interquartile range) for non-normally distributed data and mean ± standard deviation for normally distributed data unless otherwise specified. P-values represent statistical test results between the 3 BMI cohorts. BMI: body mass index; ECMO: extracorporeal membrane oxygenation; ICU: intensive care unit. In univariable analysis, age, APACHE II score, BMI, post-ECMO PaO2, cardiogenic shock as an aetiology for ECMO, cannulation by our mobile ECMO team and transport distance were found to have statistical associations with mortality. However, on multivariable logistic regression, only age and APACHE II score were significant predictors of in-hospital mortality when adjusted for the above variables. On forward stepwise selection, age and APACHE II score remained significant predictors of in-hospital mortality while cannulation by our mobile ECMO team appeared to be protective [odds ratio (OR) 0.392, 95% CI 0.181–0.852; P = 0.018] (Table 4). Table 4: Multivariable associations with in-hospital mortality   Stepwise model   Multivariable model  OR  95% CI  P-value  OR  95% CI  P-value  Age  1.035  1.010–1.061  0.005  1.033  1.009–1.057  0.006  APACHE II score  1.069  1.013–1.130  0.016  1.077  1.021–1.137  0.007  Cardiogenic shock aetiology  0.941  0.364–2.434  0.900        Body mass index  0.987  0.948–1.027  0.517        Transport distance (miles)  1.003  0.997–1.009  0.356        Cannulated by our team  0.572  0.212–1.543  0.270  0.392  0.181–0.852  0.018  Post-ECMO PaO2  1.002  0.999–1.004  0.166          Stepwise model   Multivariable model  OR  95% CI  P-value  OR  95% CI  P-value  Age  1.035  1.010–1.061  0.005  1.033  1.009–1.057  0.006  APACHE II score  1.069  1.013–1.130  0.016  1.077  1.021–1.137  0.007  Cardiogenic shock aetiology  0.941  0.364–2.434  0.900        Body mass index  0.987  0.948–1.027  0.517        Transport distance (miles)  1.003  0.997–1.009  0.356        Cannulated by our team  0.572  0.212–1.543  0.270  0.392  0.181–0.852  0.018  Post-ECMO PaO2  1.002  0.999–1.004  0.166        APACHE II: Acute Physiology and Chronic Health Evaluation II; CI: confidence interval; ECMO: extracorporeal membrane oxygenation; OR: odds ratio. Transport of super obese patients A cohort of 6 patients within the morbidly obese category qualified as super obese (BMI ≥ 50 kg/m2), with a median BMI of 60.1 kg/m2 (range 52.3–79.0 kg/m2). All patients were directly cannulated by our mobile ECMO team with no complications during transportation. Pre-ECMO initiation PaCO2 tended to be higher in super obese patients [median 76 mmHg (IQR 53–98)] and median post-ECMO PaO2 was noticeably lower than in other BMI cohorts [median 63 mmHg (IQR 53–103)]. Two (33%) of these patients required additional venous drainage cannulas for inadequate gas exchange, and all patients required tracheostomies for prolonged endotracheal intubation. These super obese patients also had considerably longer ECMO runs compared with patients of lower BMIs with a median duration of support of 18 days (IQR 12–20 days). The median ICU and total LOSs of super obese patients were nearly double that of morbidly obese patients [ICU LOS: 32 days (IQR 27–40 days) vs 17 days (IQR 9–36 days)]; total LOS: 60 days (40–82) vs 31 days (16–47). All super obese patients survived to decannulation and all but 1 survived to hospital discharge. DISCUSSION In this study of ECMO transport patient safety, obese patients were able to be transported to our institution safely, regardless of the BMI, and morbid obesity was not associated with worse outcomes. As global obesity incidence increases, it is likely that the need for ECMO and transportation during ECMO support for obese patients will also increase. Obese patients may be at a higher risk for respiratory failure and the development of acute respiratory distress syndrome (ARDS) [2, 6]. This was demonstrated in the 2009 influenza A (H1N1) pandemic, where nearly half of US adults requiring ICU management were obese as well as in a meta-analysis of 24 studies demonstrating obesity’s association with a significantly increased risk of ARDS [odds ratio 1.89, 95% CI 1.45–2.47; P < 0.001) [7, 8]. Although obesity is associated with a number of comorbidities, a growing body of literature suggests that obesity may be protective in critical illness, [9, 10] including in ARDS [2, 8, 11]. This ‘obesity paradox’ may be due to larger nutritional reserves in obesity and inflammatory modulation by endocrine signalling of adipose tissues [11, 12]. Morbid obesity is not a contraindication to ECMO [3–5]. Nevertheless, the logistical complexities of ECMO cannulation and transport are complicated by the large body habitus. A large pannus can hinder groin exposure, necessitating Trendelenburg positioning and modified equipment selection such as the use of a long spinal needle or a stiffer guide wire for percutaneous access. Patients with high BMIs also have difficulty remaining on ambulance gurneys and, depending on their weight, may require ambulances with hydraulic lifts. Although there were no complications during transportation, these patients, coupled with extensive equipment, made monitoring and medication adjustments more difficult as they took up significantly more space in the ambulances and forced staff to sit at the head of the patient rather than beside them. Furthermore, a larger body habitus oftentimes necessitated the ECMO device, itself, be placed at a higher height above the patient than usual—requiring greater negative inlet pressures exceeding −120 mmHg to maintain sufficient flows. In our experience, pushing inlet pressures below −110 mmHg results in more frequent chatter and suction events, which can lead to poor forward flow and inadequate oxygenation support. During the transportation of obese patients, chatter was often encountered and managed with crystalloid fluid boluses. Once at our institution, however, if higher inlet pressures resulted in repeated suction events despite repositioning, with patient intolerance of lower blood flow rates, an additional venous drainage cannula was added to reduce the inlet pressure and improve gas exchange. Although there were no transport complications in any of the obese patients, these patients—particularly those with morbid obesity (BMI ≥40 kg/m2)—were noted to have significantly lower PaO2 levels after cannulation. This was likely related, in part, to greater degrees of atelectasis. The comparable initial blood flow rates between obese and non-obese patients may have also contributed to the lower levels of arterial oxygenation seen with rising BMI, as evidenced by lower percentages of predicted extracorporeal flow. This is likely due to a smaller proportion of an obese patient’s cardiac index being supplied by the ECMO circuit. Morbidly obese patients presented several issues upon arrival to the ICU. Most notably, routine nursing care necessitated 1:1 critical care nursing coverage throughout much of their course, which increased staffing burdens, and these patients had considerably longer intensive care and overall hospital LOSs. This is consistent with the work of Swol et al. [13] who reported a mean ICU LOS of 31 days in 12 adult patients with acute lung failure treated with VV-ECMO. Limitations This single-centre retrospective study is subject to the limitations inherent in all observational studies. Our morbid obesity cohort had a trend towards being younger and having a lower incidence of cardiogenic shock as an indication for ECMO. While this may suggest a selection bias towards a healthier and fitter obese population, APACHE II scores and pre-ECMO initiation gases were comparable to all BMI cohorts. Similarly, neither cardiogenic shock nor age was found to be a significant predictor of mortality on multivariable analysis. Our small sample size of super obese patients precluded us from performing any statistical tests and, consequently, drawing any conclusions with this demographic. As many patients were referred from outside hospitals, we lacked sufficient follow-up data to conduct meaningful time-to-event analyses post-discharge. Finally, our data were limited only to those who survived long enough to be cannulated and were accepted for ECMO transport. CONCLUSION In summary, our results indicate that adult patients with obesity or morbid obesity can be safely transported to a tertiary care centre while receiving ECMO support and that obesity, regardless of its degree, may not be a risk factor for in-hospital mortality. A high BMI, by itself, should not preclude patients from being considered for ECMO transport. Conflict of interest: Daniel Brodie is currently on the medical advisory boards of ALung Technologies and Kadence. All compensation for these activities is paid to Columbia University. All other authors declared no conflict of interest. REFERENCES 1 Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA  2010; 303: 235– 41. Google Scholar CrossRef Search ADS PubMed  2 Akinnusi ME, Pineda LA, El Solh AA. Effect of obesity on intensive care morbidity and mortality: a meta-analysis. Crit Care Med  2008; 36: 151– 8. Google Scholar CrossRef Search ADS PubMed  3 Al-Soufi S, Buscher H, Nguyen ND, Rycus P, Nair P. Lack of association between body weight and mortality in patients on veno-venous extracorporeal membrane oxygenation. Intensive Care Med  2013; 39: 1995– 2002. Google Scholar CrossRef Search ADS PubMed  4 Kon ZN, Dahi S, Evans CF, Byrnes KA, Bittle GJ, Wehman B et al.   Class III obesity is not a contraindication to venovenous extracorporeal membrane oxygenation support. Ann Thorac Surg  2015; 100: 1855– 60. Google Scholar CrossRef Search ADS PubMed  5 Biscotti M, Agerstrand C, Abrams D, Ginsburg M, Sonett J, Mongero L et al.   One hundred transports on extracorporeal support to an extracorporeal membrane oxygenation center. Ann Thorac Surg  2015; 100: 34– 9; discussion 39–40. Google Scholar CrossRef Search ADS PubMed  6 Gong MN, Bajwa EK, Thompson BT, Christiani DC. Body mass index is associated with the development of acute respiratory distress syndrome. Thorax  2010; 65: 44– 50. Google Scholar CrossRef Search ADS PubMed  7 Louie JK, Acosta M, Samuel MC, Schechter R, Vugia DJ, Harriman K et al.   A novel risk factor for a novel virus: obesity and 2009 pandemic influenza A (H1N1). Clin Infect Dis  2011; 52: 301– 12. Google Scholar CrossRef Search ADS PubMed  8 Zhi G, Xin W, Ying W, Guohong X, Shuying L, Zhao Y-Y. ‘Obesity paradox’ in acute respiratory distress syndrome: a systematic review and meta-analysis. PLoS One  2016; 11: e1– 12. 9 Hogue CW, Stearns JD, Colantuoni E, Robinson KA, Stierer T, Mitter N et al.   The impact of obesity on outcomes after critical illness: a meta-analysis. Intensive Care Med  2009; 35: 1152– 70. Google Scholar CrossRef Search ADS PubMed  10 O’Brien JMJr, Welsh CH, Fish RH, Ancukiewicz M, Kramer AM; National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network. Excess body weight is not independently associated with outcome in mechanically ventilated patients with acute lung injury. Ann Intern Med  2004; 140: 338– 45. Google Scholar CrossRef Search ADS PubMed  11 Koch A, Sanson E, Voigt S, Helm A, Trautwein C, Tacke F. Serum adiponectin upon admission to the intensive care unit may predict mortality in critically ill patients. J Crit Care  2011; 26: 166– 74. Google Scholar CrossRef Search ADS PubMed  12 Stapleton RD, Dixon AE, Parsons PE, Ware LB, Suratt BT; NHLBI Acute Respiratory Distress Network. The association between BMI and plasma cytokine levels in patients with acute lung injury. Chest  2010; 138: 568– 77. Google Scholar CrossRef Search ADS PubMed  13 Swol J, Buchwald D, Dudda M, Strauch J, Schildhauer TA. Veno-venous extracorporeal membrane oxygenation in obese surgical patients with hypercapnic lung failure. Acta Anaesthesiol Scand  2014; 58: 534– 8. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Journal of Cardio-Thoracic Surgery Oxford University Press

Morbid obesity is not a contraindication to transport on extracorporeal support

Loading next page...
 
/lp/ou_press/morbid-obesity-is-not-a-contraindication-to-transport-on-uIZDYE033P
Publisher
Oxford University Press
ISSN
1010-7940
eISSN
1873-734X
D.O.I.
10.1093/ejcts/ezx452
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES Extracorporeal membrane oxygenation (ECMO) transport has not been described in morbidly obese patients, a population that can pose significant challenges in obtaining vascular access, indexed flows and transport logistics. We sought to study the feasibility and safety of transporting obese and morbidly obese patients during extracorporeal support. METHODS We conducted a retrospective review of all patients transported to our institution while receiving ECMO from September 2008 to September 2016. Survival to decannulation and survival to discharge were the primary outcomes. Obesity and morbid obesity were defined as a body mass index of greater than 30 kg/m2 and greater than 40 kg/m2, respectively. RESULTS From 2008 to 2016, 222 patients were transported to our institution while receiving ECMO. Among these included patients, 131 were non-obese (interquartile range 22–27 kg/m2), 63 were obese (interquartile range 31–35 kg/m2) and 28 were morbidly obese (interquartile range 41–49 kg/m2), with 6 patients having a body mass index greater than 50 kg/m2 (range 52.3–79 kg/m2). Pre-ECMO arterial blood gases, disease severity indices, cannulation strategies and transport distances were similar between these 3 groups. There was no mortality of patients during transport, and survival to discharge was 66% (n = 87) in non-obese patients, 56% (n = 35) in obese patients and 82% (n = 23) in morbidly obese patients (P = 0.042). On multivariable logistic regression analysis, body mass index was not a predictor of in-hospital mortality (odds ratio 0.99, 95% confidence interval 0.95–1.03; P = 0.517). CONCLUSIONS Transport of morbidly obese patients receiving ECMO may be performed safely and with excellent results in the setting of a dedicated ECMO transport programme with well-established management protocols. Extracorporeal membrane oxygenation, Respiratory failure, Obesity INTRODUCTION Obesity, defined as a body mass index (BMI) greater than 30 kg/m2, is becoming increasingly common, and currently, more than one-third of Americans are obese [1]. Obese patients pose unique challenges to surgeons and intensivists. In addition to comorbidities that tend to accompany obesity, such as diabetes mellitus and hypertension, there appears to be increased morbidity, increased overall hospital and intensive care unit (ICU) lengths of stay (LOS) and longer durations of mechanical ventilator support [2]. The treatment of these patients with venovenous extracorporeal membrane oxygenation (VV-ECMO) and venoarterial ECMO (VA-ECMO; also known as extracorporeal life support) remains particularly problematic for several reasons: large body size causes difficulty in achieving sufficient indexed flows in the circuit, excess weight and tissue affect patient positioning and anatomical landmark identification and weight restrictions limit radiology scans for many obese patients. Despite these limitations, obesity and extreme obesity have not been found to be significant risk factors for hospital mortality in the setting of acute lung failure and are not contraindications for ECMO [3–5]. The ability to transport patients during ECMO is critical to provide advanced care or transplantation to patients with refractory cardiac or respiratory failure. However, the transportation of obese patients receiving ECMO has not been described in the literature and presents its own unique set of challenges. We hypothesized that patients on ECMO could be transported safely, regardless of the BMI. The purpose of this study was to evaluate the outcomes in transporting obese and morbidly obese patients. MATERIALS AND METHODS Patients A retrospective cohort study was conducted using clinical data of all patients transported to our institution on either VV-ECMO or VA-ECMO from September 2008 to September 2016. These included patients cannulated at outside hospitals by our mobile ECMO team or by outside hospital physicians as well as patients from our most recent ECMO transport publication. This study was approved by the Institutional Review Board of Columbia University Medical Center with waiver of informed consent. The use of ECMO, as described in this article, is off-label. Predictor variables The primary predictor of interest was obesity, defined as a BMI of ≥30 kg/m2. Other variables considered included morbid obesity (BMI ≥40 kg/m2), gender, age, indication for ECMO, cannulation team (outside hospital versus our institution), travel distance and Acute Physiology and Chronic Health Evaluation II (APACHE II) score. Primary outcomes The primary outcome was survival to discharge. The secondary outcomes included survival during transport, survival to decannulation, 30-day survival, LOS, the need for tracheostomy, post-cannulation arterial gases, and percent predicted ECMO flow. We calculate this by dividing a patient’s initial flow on ECMO by that patient’s predicted cardiac output for a cardiac index of 2.4 l/min/m2. We also conducted exploratory secondary analysis of outcomes in super obese patients (BMI >50 kg/m2). Extracorporeal membrane oxygenation management The details of our ECMO transport procedures have been previously described [5]. Indications for VV-ECMO include severe, refractory, potentially reversible hypoxaemic or hypercapnic respiratory failure that meets the following criteria: partial pressure of arterial oxygen (PaO2) to fraction of inspired oxygen ratio <80 mmHg, severe uncompensated respiratory acidosis (pH <7.15) or excessively high airway pressures >35–45 cmH2O, despite optimized ventilator management. Our VA-ECMO indications for refractory cardiogenic shock include evidence of end-organ hypoperfusion with a systolic blood pressure of <90 mmHg, a cardiac index of <2.0 l/min/m2 or the pulmonary capillary wedge pressure of >16 mmHg. Briefly, lung protective mechanical ventilation is employed after ECMO initiation at the outside hospital. Tidal volumes are reduced to maintain peak airway pressures less than 35 mmHg, respiratory rates decreased to less than 20 breaths/min and sweep gas flow rate is adjusted to maintain a pH between 7.35 and 7.45. Heparin is used for anticoagulation with a target activated partial thromboplastin time of 40–60 s. Our standard ECMO device is the Cardiohelp system (Maquet, Wayne, NJ, USA), though our first transports relied upon the Rotaflow pump and Quadrox oxygenator (Maquet). Although there were no differences in the performance, we found a significant difference in the ergonomics of transport using the lower profile Cardiohelp system. On arrival to the ICU, care is transitioned to the ICU team with consultation from the ECMO team. Patients in the ICU are managed with the ultraprotective lung ventilation as tolerated: volume assist–control ventilation with a set respiratory rate of 2–10 breaths/min, plateau airway pressures <20 cm of water, if possible, or at least <25 cm of water and with moderate levels of positive end-expiratory pressure, typically 10–15 cm of water. Additional drainage cannulas, typically a short 21- or 23-Fr Bio-Medicus cannula (Medtronic, Minneapolis, MN, USA), may be placed in the contralateral femoral vein if additional ECMO support is required upon arrival to the ICU. Statistical analysis All statistical analyses were performed with SPSS 20.0 (IBM SPSS Inc., Armonk, NY, USA) and R statistics software (R Core Team, Vienna, Austria). Clinical and demographic variables were presented using standard summary statistics, including mean and standard deviation or median and interquartile range (IQR) depending on the normality of distribution for continuous variables and frequencies and proportions for categorical variables. The distributions of continuous variables were tested with the Shapiro–Wilk test. To assess the predictor variables and major morbidity, the χ2 test or the Fisher’s exact test (if the expected value of a 4 × 4 table cell was <5) was used for categorical variables. The Student’s t-test, analysis of variance or the Mann–Whitney U-test for continuous variables using 2 sample means, 3 or more sample means and in cases where sample medians with skewed distributions were used, respectively. These criteria were prespecified prior to statistical testing. Variables with P-values ≤0.10 in univariable analyses were analysed in a multivariable logistic model with an a priori decision to automatically include BMI in this model. An additional multivariable model was built using forward stepwise selection, where variables were included if associated P-values met the significance criterion (P < 0.05) or if their inclusion changed the magnitude of the coefficients by ≥10%. RESULTS All 222 patients were successfully placed on VV-ECMO or VA-ECMO (extracorporeal life support) via percutaneous cannulation and transported to our centre alive. Of these included patients, 131 (59%) were non-obese, 63 (28%) had BMIs ≥30 to ≤39 and 28 (13%) patients with a BMI ≥40 kg/m2. There were no significant demographic differences between these 3 groups, although there was a trend towards younger ages in the morbidly obese cohort. Moreover, there was a trend towards a lower incidence of cardiogenic shock in the morbidly obesity group compared with the non-obese and obese groups. There were no significant inter-group differences in method or distance of transportation and in illness severity as measured by the APACHE II score, upon presentation to our institution. Baseline characteristics are reported in Table 1. Table 1: Characteristics and outcomes of ECMO patients stratified by BMI     BMI (kg/m2)     Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Age (years)  42 (27–55)  41 (26–56)  47 (33–56)  36 (27–47)  0.076  Male gender, n (%)  131 (59.0)  74 (56.5)  41 (65.1)  16 (57.1)  0.514  BMI (kg/m2)  28 (24–34)  25 (22–27)  34 (31–35)  44 (41–49)  <0.001  Cannulated by our team, n (%)  174 (78.4)  101 (77.1)  51 (81.0)  22 (78.6)  0.845  APACHE II score  29 (25–34)  30 (25–34)  28 (26–34)  28 (24–32)  0.871  Transported safely, n (%)  222 (100)  131 (100)  63 (100)  28 (100)  1.000  Median distance (miles)  16 (8–32)  15 (7–32)  17 (10–31)  16 (7–32)  0.377  Mode of transportation          1.000a   Ambulance, n (%)  218 (98.2)  128 (97.7)  61 (98.4)  28 (100)     Fixed-wing aircraft, n (%)  4 (1.8)  3 (2.3)  1 (1.6)  0 (0)        BMI (kg/m2)     Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Age (years)  42 (27–55)  41 (26–56)  47 (33–56)  36 (27–47)  0.076  Male gender, n (%)  131 (59.0)  74 (56.5)  41 (65.1)  16 (57.1)  0.514  BMI (kg/m2)  28 (24–34)  25 (22–27)  34 (31–35)  44 (41–49)  <0.001  Cannulated by our team, n (%)  174 (78.4)  101 (77.1)  51 (81.0)  22 (78.6)  0.845  APACHE II score  29 (25–34)  30 (25–34)  28 (26–34)  28 (24–32)  0.871  Transported safely, n (%)  222 (100)  131 (100)  63 (100)  28 (100)  1.000  Median distance (miles)  16 (8–32)  15 (7–32)  17 (10–31)  16 (7–32)  0.377  Mode of transportation          1.000a   Ambulance, n (%)  218 (98.2)  128 (97.7)  61 (98.4)  28 (100)     Fixed-wing aircraft, n (%)  4 (1.8)  3 (2.3)  1 (1.6)  0 (0)    Data are represented as median (interquartile range) for non-normally distributed data and mean ± standard deviation for normally distributed data unless otherwise specified. P-values represent statistical test results between the 3 BMI cohorts. a By the Fisher’s exact test. APACHE II: Acute Physiology and Chronic Health Evaluation II; BMI: body mass index; ECMO: extracorporeal membrane oxygenation. Cannulation strategies were not found to be significantly different between the BMI groups. Among the 60 patients placed on VA-ECMO (extracorporeal life support), 42 were cannulated in the bilateral femoral arteries, 14 were cannulated centrally, 2 via femoral vein to ipsilateral axillary artery and 2 via right internal jugular vein to the ipsilateral axillary artery. VV-ECMO patients were largely cannulated via the femoral vein to ipsilateral internal jugular vein (n = 132/162), with the remainder having a dual-lumen cannula placed in the right internal jugular vein. There was a non-statistically significant increase in the dual-lumen cannula use among VV-ECMO patients with lower BMI (24% of those <30 kg/m2 vs 11% of those 30–39 kg/m2 vs 9% of the patients with BMI ≥40 kg/m2; P = 0.119). There was no difference in drainage cannula sizes, and precannulation blood gases were comparable between the groups (Table 2). Following cannulation, there were significantly lower post-cannulation arterial blood oxygenation (PaO2) levels in the morbidly obese group [72 mmHg (IQR 60–118 mmHg) vs 111 mmHg (83–189 mmHg) in the obese cohort and 125 mmHg (86–277 mmHg) in the non-obese cohort, P = 0.001]. As the BMI increased, the initial flows on extracorporeal support were significantly lower than that predicted. Morbidly obese patients, on average, only achieved 71% (SD 15%) of their predicted flows compared with the non-obese patients who achieved 87% (SD 19%). Additional venous drainage cannulas were added to the circuits of 5 (8%) of the obese and morbidly obese patients more than 24 h after transport to improve the blood flow rate and optimize gas exchange. After the addition of these cannulas, flows were augmented by an average of 65%, achieving greater than 88% of predicted flow rates in all patients. Notably, there were no additional venous cannulas required in the non-obese patients. Table 2: ECMO results stratified by BMI     BMI (kg/m2)   ECMO characteristics  Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Indication for ECMO          0.308   Respiratory failure, n (%)  124 (55.9)  71 (54.2)  34 (54.0)  19 (67.9)  0.417   ARDS, n (%)  106 (85.5)  60 (84.5)  30 (88.2)  16 (84.2)     Other, n (%)  18 (14.5)  11 (15.5)  4 (11.8)  3 (15.8)     Cardiogenic shock, n (%)  52 (23.4)  33 (25.2)  17 (27.0)  2 (7.1)  0.092   Post-cardiac arrest, n (%)  28 (53.8)  20 (60.6)  7 (41.2)  1 (50.0)     Post-cardiotomy, n (%)  8 (15.3)  4 (12.1)  4 (23.5)  0 (0.0)     Heart failure, n (%)  7 (13.5)  3 (9.1)  3 (17.6)  1 (50.0)     Other, n (%)  9 (17.4)  6 (18.2)  3 (17.6)  0 (0.0)     Combined, n (%)  46 (20.7)  27 (20.6)  12 (19.0)  7 (25)  0.813  Venovenous ECMO, n (%)  162 (73.0)  95 (72.5)  45 (71.4)  22 (78.6)  0.784  Venoarterial ECMO (ECLS), n (%)  60 (27.0)  36 (27.5)  18 (28.6)  6 (21.4)  0.784  Pre-ECMO ventilation days  1 (0–4)  1 (0–4)  1 (1–3)  1 (0–5)  0.769  Drainage cannula (Fr)  23 (23–23)  23 (23–23)  23 (23–23)  23 (23–23)  0.346  Arterial cannula (Fr)  18 (15–20)  17 (15–19)  19 (15–22)  22 (19–22)a  0.117  Venous return cannula (Fr)  20 (20–22)  20 (20–22)  20 (20–20)  20 (20–22)  0.288  Additional venous drainage, n (%)  5 (2.2)  0 (0.0)  3 (4.8)  2 (7.1)  0.008b  ECMO duration (days)  8 (4–14)  8 (4–13)  9 (5–14)  9 (5–19)  0.599  Initial flow (l/min)  3.93 ± 0.85  3.83 ± 0.89  4.04 ± 0.77  4.07 ± 0.76  0.203  Predicted ECMO flow  4.38 ± 1.57  3.97 ± 1.42  4.70 ± 1.51  5.54 ± 1.38  <0.001  % Predicted flow  82 ± 18%  87 ± 19%  78 ± 14%  71 ± 15%  <0.001      BMI (kg/m2)   ECMO characteristics  Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Indication for ECMO          0.308   Respiratory failure, n (%)  124 (55.9)  71 (54.2)  34 (54.0)  19 (67.9)  0.417   ARDS, n (%)  106 (85.5)  60 (84.5)  30 (88.2)  16 (84.2)     Other, n (%)  18 (14.5)  11 (15.5)  4 (11.8)  3 (15.8)     Cardiogenic shock, n (%)  52 (23.4)  33 (25.2)  17 (27.0)  2 (7.1)  0.092   Post-cardiac arrest, n (%)  28 (53.8)  20 (60.6)  7 (41.2)  1 (50.0)     Post-cardiotomy, n (%)  8 (15.3)  4 (12.1)  4 (23.5)  0 (0.0)     Heart failure, n (%)  7 (13.5)  3 (9.1)  3 (17.6)  1 (50.0)     Other, n (%)  9 (17.4)  6 (18.2)  3 (17.6)  0 (0.0)     Combined, n (%)  46 (20.7)  27 (20.6)  12 (19.0)  7 (25)  0.813  Venovenous ECMO, n (%)  162 (73.0)  95 (72.5)  45 (71.4)  22 (78.6)  0.784  Venoarterial ECMO (ECLS), n (%)  60 (27.0)  36 (27.5)  18 (28.6)  6 (21.4)  0.784  Pre-ECMO ventilation days  1 (0–4)  1 (0–4)  1 (1–3)  1 (0–5)  0.769  Drainage cannula (Fr)  23 (23–23)  23 (23–23)  23 (23–23)  23 (23–23)  0.346  Arterial cannula (Fr)  18 (15–20)  17 (15–19)  19 (15–22)  22 (19–22)a  0.117  Venous return cannula (Fr)  20 (20–22)  20 (20–22)  20 (20–20)  20 (20–22)  0.288  Additional venous drainage, n (%)  5 (2.2)  0 (0.0)  3 (4.8)  2 (7.1)  0.008b  ECMO duration (days)  8 (4–14)  8 (4–13)  9 (5–14)  9 (5–19)  0.599  Initial flow (l/min)  3.93 ± 0.85  3.83 ± 0.89  4.04 ± 0.77  4.07 ± 0.76  0.203  Predicted ECMO flow  4.38 ± 1.57  3.97 ± 1.42  4.70 ± 1.51  5.54 ± 1.38  <0.001  % Predicted flow  82 ± 18%  87 ± 19%  78 ± 14%  71 ± 15%  <0.001  Data are represented as median (interquartile range) for non-normally distributed data and mean ± standard deviation for normally distributed data unless otherwise specified. P-values represent statistical test results between the 3 BMI cohorts. a There were 5 morbidly obese patients on venoarterial-ECMO: 1 was peripherally cannulated and 4 were centrally cannulated. b By the Fisher’s exact test. BMI: body mass index; ECLS: extracorporeal life support; ECMO: extracorporeal membrane oxygenation. The overall median duration of ECMO support was 8 days (IQR 4–14 days), with no significant differences between any of the 3 BMI groups (Table 3). Morbidly obese patients were noted to have longer ICU and overall hospital LOSs compared with the obese and the non-obese patients; however, these results did not achieve statistical significance. There were no differences in the rate of tracheostomy procedures performed between the 3 groups and no difference in the incidence of deep vein thrombosis at the cannulation sites (either the femoral vein or the internal jugular vein). Three non-obese patients exhibited signs of lower-limb ischaemia, requiring 2 patients to undergo fasciotomies and 1 patient to have a distal perfusion cannula inserted upon arrival to our ICU. No pulmonary emboli were noted in any of the patients studied. There did, however, appear to be a survival advantage, unadjusted for indication, in morbidly obese patients with 23 (82%) patients surviving to hospital discharge compared with 35 (56%) obese patients and 87 (66%) non-obese patients (P = 0.042). Table 3: Patient outcomes in ECMO transport stratified by BMI   BMI (kg/m2)   Outcomes  Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Hospital length of stay (days)  28 (12–47)  27 (13–46)  21 (10–46)  37 (24–61)  0.065  ICU length of stay (days)  18 (9–30)  18 (9–28)  15 (9–29)  24 (15–42)  0.126  Tracheostomy, n (%)  110 (49.5)  62 (47.3)  33 (52.4)  15 (53.6)  0.649  Deep vein thrombosis, n (%)  49 (22.1)  28 (21.3)  15 (23.8)  6 (21.4)  0.883  Survival to decannulation, n (%)  174 (78.7)  105 (80.2)  45 (71.4)  24 (85.7)  0.204  30-day survival  159 (71.9)  96 (73.3)  39 (61.9)  24 (85.7)  0.047  Survival to discharge, n (%)  145 (65.6)  87 (66.4)  35 (55.6)  23 (82.1)  0.042    BMI (kg/m2)   Outcomes  Overall (n = 222)  <30 (n = 131)  ≥30 to ≤39 (n = 63)  ≥40 (n = 28)  P-value  Hospital length of stay (days)  28 (12–47)  27 (13–46)  21 (10–46)  37 (24–61)  0.065  ICU length of stay (days)  18 (9–30)  18 (9–28)  15 (9–29)  24 (15–42)  0.126  Tracheostomy, n (%)  110 (49.5)  62 (47.3)  33 (52.4)  15 (53.6)  0.649  Deep vein thrombosis, n (%)  49 (22.1)  28 (21.3)  15 (23.8)  6 (21.4)  0.883  Survival to decannulation, n (%)  174 (78.7)  105 (80.2)  45 (71.4)  24 (85.7)  0.204  30-day survival  159 (71.9)  96 (73.3)  39 (61.9)  24 (85.7)  0.047  Survival to discharge, n (%)  145 (65.6)  87 (66.4)  35 (55.6)  23 (82.1)  0.042  Data are represented as median (interquartile range) for non-normally distributed data and mean ± standard deviation for normally distributed data unless otherwise specified. P-values represent statistical test results between the 3 BMI cohorts. BMI: body mass index; ECMO: extracorporeal membrane oxygenation; ICU: intensive care unit. In univariable analysis, age, APACHE II score, BMI, post-ECMO PaO2, cardiogenic shock as an aetiology for ECMO, cannulation by our mobile ECMO team and transport distance were found to have statistical associations with mortality. However, on multivariable logistic regression, only age and APACHE II score were significant predictors of in-hospital mortality when adjusted for the above variables. On forward stepwise selection, age and APACHE II score remained significant predictors of in-hospital mortality while cannulation by our mobile ECMO team appeared to be protective [odds ratio (OR) 0.392, 95% CI 0.181–0.852; P = 0.018] (Table 4). Table 4: Multivariable associations with in-hospital mortality   Stepwise model   Multivariable model  OR  95% CI  P-value  OR  95% CI  P-value  Age  1.035  1.010–1.061  0.005  1.033  1.009–1.057  0.006  APACHE II score  1.069  1.013–1.130  0.016  1.077  1.021–1.137  0.007  Cardiogenic shock aetiology  0.941  0.364–2.434  0.900        Body mass index  0.987  0.948–1.027  0.517        Transport distance (miles)  1.003  0.997–1.009  0.356        Cannulated by our team  0.572  0.212–1.543  0.270  0.392  0.181–0.852  0.018  Post-ECMO PaO2  1.002  0.999–1.004  0.166          Stepwise model   Multivariable model  OR  95% CI  P-value  OR  95% CI  P-value  Age  1.035  1.010–1.061  0.005  1.033  1.009–1.057  0.006  APACHE II score  1.069  1.013–1.130  0.016  1.077  1.021–1.137  0.007  Cardiogenic shock aetiology  0.941  0.364–2.434  0.900        Body mass index  0.987  0.948–1.027  0.517        Transport distance (miles)  1.003  0.997–1.009  0.356        Cannulated by our team  0.572  0.212–1.543  0.270  0.392  0.181–0.852  0.018  Post-ECMO PaO2  1.002  0.999–1.004  0.166        APACHE II: Acute Physiology and Chronic Health Evaluation II; CI: confidence interval; ECMO: extracorporeal membrane oxygenation; OR: odds ratio. Transport of super obese patients A cohort of 6 patients within the morbidly obese category qualified as super obese (BMI ≥ 50 kg/m2), with a median BMI of 60.1 kg/m2 (range 52.3–79.0 kg/m2). All patients were directly cannulated by our mobile ECMO team with no complications during transportation. Pre-ECMO initiation PaCO2 tended to be higher in super obese patients [median 76 mmHg (IQR 53–98)] and median post-ECMO PaO2 was noticeably lower than in other BMI cohorts [median 63 mmHg (IQR 53–103)]. Two (33%) of these patients required additional venous drainage cannulas for inadequate gas exchange, and all patients required tracheostomies for prolonged endotracheal intubation. These super obese patients also had considerably longer ECMO runs compared with patients of lower BMIs with a median duration of support of 18 days (IQR 12–20 days). The median ICU and total LOSs of super obese patients were nearly double that of morbidly obese patients [ICU LOS: 32 days (IQR 27–40 days) vs 17 days (IQR 9–36 days)]; total LOS: 60 days (40–82) vs 31 days (16–47). All super obese patients survived to decannulation and all but 1 survived to hospital discharge. DISCUSSION In this study of ECMO transport patient safety, obese patients were able to be transported to our institution safely, regardless of the BMI, and morbid obesity was not associated with worse outcomes. As global obesity incidence increases, it is likely that the need for ECMO and transportation during ECMO support for obese patients will also increase. Obese patients may be at a higher risk for respiratory failure and the development of acute respiratory distress syndrome (ARDS) [2, 6]. This was demonstrated in the 2009 influenza A (H1N1) pandemic, where nearly half of US adults requiring ICU management were obese as well as in a meta-analysis of 24 studies demonstrating obesity’s association with a significantly increased risk of ARDS [odds ratio 1.89, 95% CI 1.45–2.47; P < 0.001) [7, 8]. Although obesity is associated with a number of comorbidities, a growing body of literature suggests that obesity may be protective in critical illness, [9, 10] including in ARDS [2, 8, 11]. This ‘obesity paradox’ may be due to larger nutritional reserves in obesity and inflammatory modulation by endocrine signalling of adipose tissues [11, 12]. Morbid obesity is not a contraindication to ECMO [3–5]. Nevertheless, the logistical complexities of ECMO cannulation and transport are complicated by the large body habitus. A large pannus can hinder groin exposure, necessitating Trendelenburg positioning and modified equipment selection such as the use of a long spinal needle or a stiffer guide wire for percutaneous access. Patients with high BMIs also have difficulty remaining on ambulance gurneys and, depending on their weight, may require ambulances with hydraulic lifts. Although there were no complications during transportation, these patients, coupled with extensive equipment, made monitoring and medication adjustments more difficult as they took up significantly more space in the ambulances and forced staff to sit at the head of the patient rather than beside them. Furthermore, a larger body habitus oftentimes necessitated the ECMO device, itself, be placed at a higher height above the patient than usual—requiring greater negative inlet pressures exceeding −120 mmHg to maintain sufficient flows. In our experience, pushing inlet pressures below −110 mmHg results in more frequent chatter and suction events, which can lead to poor forward flow and inadequate oxygenation support. During the transportation of obese patients, chatter was often encountered and managed with crystalloid fluid boluses. Once at our institution, however, if higher inlet pressures resulted in repeated suction events despite repositioning, with patient intolerance of lower blood flow rates, an additional venous drainage cannula was added to reduce the inlet pressure and improve gas exchange. Although there were no transport complications in any of the obese patients, these patients—particularly those with morbid obesity (BMI ≥40 kg/m2)—were noted to have significantly lower PaO2 levels after cannulation. This was likely related, in part, to greater degrees of atelectasis. The comparable initial blood flow rates between obese and non-obese patients may have also contributed to the lower levels of arterial oxygenation seen with rising BMI, as evidenced by lower percentages of predicted extracorporeal flow. This is likely due to a smaller proportion of an obese patient’s cardiac index being supplied by the ECMO circuit. Morbidly obese patients presented several issues upon arrival to the ICU. Most notably, routine nursing care necessitated 1:1 critical care nursing coverage throughout much of their course, which increased staffing burdens, and these patients had considerably longer intensive care and overall hospital LOSs. This is consistent with the work of Swol et al. [13] who reported a mean ICU LOS of 31 days in 12 adult patients with acute lung failure treated with VV-ECMO. Limitations This single-centre retrospective study is subject to the limitations inherent in all observational studies. Our morbid obesity cohort had a trend towards being younger and having a lower incidence of cardiogenic shock as an indication for ECMO. While this may suggest a selection bias towards a healthier and fitter obese population, APACHE II scores and pre-ECMO initiation gases were comparable to all BMI cohorts. Similarly, neither cardiogenic shock nor age was found to be a significant predictor of mortality on multivariable analysis. Our small sample size of super obese patients precluded us from performing any statistical tests and, consequently, drawing any conclusions with this demographic. As many patients were referred from outside hospitals, we lacked sufficient follow-up data to conduct meaningful time-to-event analyses post-discharge. Finally, our data were limited only to those who survived long enough to be cannulated and were accepted for ECMO transport. CONCLUSION In summary, our results indicate that adult patients with obesity or morbid obesity can be safely transported to a tertiary care centre while receiving ECMO support and that obesity, regardless of its degree, may not be a risk factor for in-hospital mortality. A high BMI, by itself, should not preclude patients from being considered for ECMO transport. Conflict of interest: Daniel Brodie is currently on the medical advisory boards of ALung Technologies and Kadence. All compensation for these activities is paid to Columbia University. All other authors declared no conflict of interest. REFERENCES 1 Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA  2010; 303: 235– 41. Google Scholar CrossRef Search ADS PubMed  2 Akinnusi ME, Pineda LA, El Solh AA. Effect of obesity on intensive care morbidity and mortality: a meta-analysis. Crit Care Med  2008; 36: 151– 8. Google Scholar CrossRef Search ADS PubMed  3 Al-Soufi S, Buscher H, Nguyen ND, Rycus P, Nair P. Lack of association between body weight and mortality in patients on veno-venous extracorporeal membrane oxygenation. Intensive Care Med  2013; 39: 1995– 2002. Google Scholar CrossRef Search ADS PubMed  4 Kon ZN, Dahi S, Evans CF, Byrnes KA, Bittle GJ, Wehman B et al.   Class III obesity is not a contraindication to venovenous extracorporeal membrane oxygenation support. Ann Thorac Surg  2015; 100: 1855– 60. Google Scholar CrossRef Search ADS PubMed  5 Biscotti M, Agerstrand C, Abrams D, Ginsburg M, Sonett J, Mongero L et al.   One hundred transports on extracorporeal support to an extracorporeal membrane oxygenation center. Ann Thorac Surg  2015; 100: 34– 9; discussion 39–40. Google Scholar CrossRef Search ADS PubMed  6 Gong MN, Bajwa EK, Thompson BT, Christiani DC. Body mass index is associated with the development of acute respiratory distress syndrome. Thorax  2010; 65: 44– 50. Google Scholar CrossRef Search ADS PubMed  7 Louie JK, Acosta M, Samuel MC, Schechter R, Vugia DJ, Harriman K et al.   A novel risk factor for a novel virus: obesity and 2009 pandemic influenza A (H1N1). Clin Infect Dis  2011; 52: 301– 12. Google Scholar CrossRef Search ADS PubMed  8 Zhi G, Xin W, Ying W, Guohong X, Shuying L, Zhao Y-Y. ‘Obesity paradox’ in acute respiratory distress syndrome: a systematic review and meta-analysis. PLoS One  2016; 11: e1– 12. 9 Hogue CW, Stearns JD, Colantuoni E, Robinson KA, Stierer T, Mitter N et al.   The impact of obesity on outcomes after critical illness: a meta-analysis. Intensive Care Med  2009; 35: 1152– 70. Google Scholar CrossRef Search ADS PubMed  10 O’Brien JMJr, Welsh CH, Fish RH, Ancukiewicz M, Kramer AM; National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network. Excess body weight is not independently associated with outcome in mechanically ventilated patients with acute lung injury. Ann Intern Med  2004; 140: 338– 45. Google Scholar CrossRef Search ADS PubMed  11 Koch A, Sanson E, Voigt S, Helm A, Trautwein C, Tacke F. Serum adiponectin upon admission to the intensive care unit may predict mortality in critically ill patients. J Crit Care  2011; 26: 166– 74. Google Scholar CrossRef Search ADS PubMed  12 Stapleton RD, Dixon AE, Parsons PE, Ware LB, Suratt BT; NHLBI Acute Respiratory Distress Network. The association between BMI and plasma cytokine levels in patients with acute lung injury. Chest  2010; 138: 568– 77. Google Scholar CrossRef Search ADS PubMed  13 Swol J, Buchwald D, Dudda M, Strauch J, Schildhauer TA. Veno-venous extracorporeal membrane oxygenation in obese surgical patients with hypercapnic lung failure. Acta Anaesthesiol Scand  2014; 58: 534– 8. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

Journal

European Journal of Cardio-Thoracic SurgeryOxford University Press

Published: Apr 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from
Google Scholar,
PubMed
Create lists to
organize your research
Export lists, citations
Read DeepDyve articles
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off