Mechanical circulatory support is an effective management option for refractory cardiogenic shock.1,2 Among percutaneous devices for mechanical circulatory support, use of venoarterial extracorporeal membrane oxygenation (VA ECMO) for cardiopulmonary failure is a growing trend increased by technical advances.3 Regarding the background to this incremental trend, VA ECMO has its own distinct advantages, including biventricular support, immediate bedside application, oxygenation, full circulatory support, and various applicable indications.1,4,5 Despite these benefits, however, VA ECMO has shown a somewhat high incidence of the following complications: lower limb ischemia, bleeding, infection, thrombosis, rise in afterload, and coagulopathy.6 Conventionally, a femoral arterial cannula of over 15 Fr is typically used in peripheral VA ECMO for full circulatory support.1,7 However, the large femoral arterial cannula used in peripheral VA ECMO is thought to be the potential cause of many vascular complications.8 In general, the choice of cannula size is dependent on sex, height, weight, underlying cause, native cardiac function, and peripheral vascular health. However, to date, limited data are available on the clinical impact of arterial cannula size in peripheral VA ECMO. Therefore, we investigated the clinical outcomes and procedure-related complications according to arterial cannula size in the patients who received peripheral VA ECMO with femoral artery cannulation.
Materials and Methods
Between January 2014 and April 2016, 257 patients who underwent ECMO were enrolled in a single-center registry. Among the patients registered, we only included patients who were placed on peripheral VA ECMO with femoral artery cannulation. Exclusion criteria were as follows: 1) patients <18 years of age, 2) patients with central aortic cannulation or axillary artery cannulation, or 3) patients who were still at ECMO implantation status by the time of enrollment. Finally, 165 patients were finally analyzed in this study. We classified patients into two groups according to arterial cannula size: the small cannula group (14–15 Fr, n = 87) and the large cannula group (16–21 Fr, n = 78). The flow of study population is shown in Figure 1. The Institutional Review Board of Samsung Medical Center approved this study and waived the requirement for written informed consent.
ECMO Implantation and Management
We previously reported ECMO practice in detail.4,9 The decision to implant ECMO for mechanical support was determined by an experienced team, and the ECMO was placed by cardiovascular surgeons or interventional cardiologists. Capiox Emergency Bypass System (Capiox EBS; Terumo, Inc., Tokyo, Japan) and Permanent Life Support (PLS; MAQUET, Rastatt, Germany) were used in our hospital. Pump speed was adjusted to obtain a cardiac index greater than 2.2 L/min/body surface area (BSA; m2), mean arterial pressure greater than 65 mm Hg, and central mixed venous saturation greater than 70%. Inotropes were discontinued or reduced to minimal doses within a few hours of achieving goal-directed flows. Echocardiography was performed daily to monitor cardiac function. If patients were hemodynamically stable and adequately oxygenated, they were considered for ECMO weaning when the flow rate was 1 L/min at least for 4 hours. Cannula sizes ranged from 14 to 21 Fr for the femoral artery and from 21 to 28 Fr for the femoral vein. The size of both arterial and venous cannula was selected according to the ECMO physician and the perfusionist at the scene. Although there was no institutional guideline for selection of cannula size, physicians generally prefer large cannula for a large patient or a patient who is expected to have ECMO for a prolonged period. Distal perfusion was routinely recommended but was left to the operator’s discretion. If hypoperfusion of the leg was suspected in patients without distal perfusion, as noted on physical examination and Doppler ultrasound of the femoral artery, an additional 7–9 Fr percutaneous catheter distal to the ECMO arterial cannula was placed into the superficial femoral artery and if not available, posterior tibial artery is exposed via a small incision and cannulated in a retrograde fashion. In the event of left ventricular distension refractory to conservative managements, left ventricular venting was performed by cardiovascular surgeons or interventional cardiologists.
Data Collection, Definitions, and Study Outcomes
Baseline characteristics, procedural characteristics, laboratory data, and clinical outcome data were collected through medical record review. Baseline characteristics included age, sex, body mass index (BMI), BSA, and comorbidities. Clinical presentations were classified into five categories: ischemic cardiomyopathy, nonischemic cardiomyopathy, septic shock, refractory arrhythmia, and all the other causes combined. Nonischemic cardiomyopathy was composed of myocarditis, dilated cardiomyopathy, stress-induced cardiomyopathy, valvular heart disease, and peripartum cardiomyopathy. According to the purpose of VA ECMO implantation, patients were divided into two groups: “bridge to recovery” and “bridge to heart or lung transplantation.” Extracorporeal cardiopulmonary resuscitation (ECPR) was defined as intention-to-treat with hemodynamic VA ECMO support during cardiac arrest, regardless of the interim return of spontaneous circulation. Laboratory findings, including creatinine and lactate, were collected just before VA ECMO insertion and at 24-hour intervals after VA ECMO insertion.
The primary outcome was survival to discharge. Secondary outcomes were a successful weaning rate and procedure-related complications, such as lower limb ischemia, cannula site bleeding, noncannula site bleeding, thrombotic events, coagulopathy, and cannula-related sepsis. Lower limb ischemia was defined as need for surgical management, not just distal perfusion catheterization, or having a dependent performance from 0 to 2 scale on functional ambulation classification because of lower limb ischemia at discharge.10 Functional ambulation classification was assessed at discharge by rehabilitation physicians. Cannula site bleeding was defined as need for surgical wound repair. Noncannula site bleeding included gastrointestinal bleeding and cerebral hemorrhage. Thrombosis events were identified by duplex ultrasonography and defined as systemic thrombosis including systemic arterial embolization and systemic venous thrombosis not ECMO circuit thrombosis and cannula thrombosis. Sepsis was defined as bacteremia induced by cannula-related infections.
Categorical variables were presented as percentages and were compared using Pearson’s χ2 or Fisher’s exact tests. Continuous variables were presented as mean ± standard deviation or as median and interquartile range and were compared with an independent t-test or a Wilcoxon rank sum test. The relation between BSA and arterial cannula size was analyzed with Jonckheere–Terpstra test for trend. We used Cochran–Armitage test for trend for categorical variables. All tests were two tailed, and p < 0.05 was considered to be statistically significant. Statistical analyses were performed using SPSS version 23 for Windows (IBM, Armonk, NY) and R version 3.4.3. (R Foundation for Statistical Computing, Vienna, Austria).
Baseline characteristics between the small cannula and the large cannula groups are shown in Table 1. Patients in the small cannula group were older than were those in the large cannula group (59.9 ± 15.7 vs. 49.8 ± 15.0 years; p < 0.01). Sex, BMI, BSA, comorbidities, and clinical presentations at VA ECMO implantation were not significantly different between the two groups. In both groups, the most frequent clinical presentation was ischemic cardiomyopathy. In terms of the purpose of implantation, VA ECMO was more deployed for bridge to recovery (87.9%) than for bridge to transplantation (12.1%). The patients who received VA ECMO for bridge to transplantation were more frequently included in the large arterial cannula group (17.9%) than in the small arterial cannula group (6.9%; p = 0.03).
Procedural Characteristics and Laboratory Findings
The procedural characteristics and laboratory findings are presented in Table 2. ECPR were similarly distributed between the two groups (54.0% in the small cannula group vs. 55.1% in the large cannula group; p = 0.89). There were no significant differences in procedural characteristics and laboratory findings during VA ECMO support except the location of ECMO implantation. Small arterial cannulas were more frequently used in intensive care units whereas large arterial cannulas were comparatively more inserted in the environs of the catheterization lab. Distal perfusion catheters were placed in 63 patients (38.2%). In the 63 patients who underwent distal perfusion catheterization, 48 patients simultaneously underwent distal perfusion catheterization at the time of primary cannulation. Distal perfusion catheters were similarly placed in the two groups.
Distribution of Femoral Arterial Cannula
The distribution of femoral arterial cannula size is shown in Figure 2. Femoral arterial cannulas between 15 and 17 Fr were mainly used. The mean of arterial cannula size was numerically larger in male than in female patients, but it was not statistically significant (15.9 ± 1.1 Fr in male patients vs. 15.6 ± 1.0 Fr in female patients; p = 0.08). The mean of arterial cannula size increased with BSA (p for trend < 0.01; Table 3).
Clinical Outcomes and Complications
Observed clinical outcomes and procedure-related complications are shown in Table 4. The survival to discharge was not significantly different between the two groups (51.7% in the small cannula group vs. 57.7% in the large cannula group; p = 0.44). The success rates of weaning from VA ECMO were also similar between the two groups (70.1% in the small cannula group vs. 64.1% in the large cannula group; p = 0.41). Among 111 patients successfully weaning from ECMO, 21 patients died during hospitalization. Causes of death were as follows: pump failure in eight patients (38.1%), neurologic causes in four patients (19.0%), and noncardiac causes including bleeding and sepsis in nine patients (42.9%). There was no significant difference in initial ECMO flow/BSA between the two groups (1.86 ± 0.42 vs. 1.98 ± 0.49 L/min/m2; p = 0.12) although small cannula group had a numerically low value. Duration of VA ECMO support was shorter in the small cannula group than in the large cannula group (2.6 [0.7–5.2] days in the small cannula group vs. 4.0 [1.3–7.8] days in the large cannula group; p < 0.01).
The number of patients who had any one of the collected complication events was significantly higher in the large cannula group (17.2% in the small cannula group vs. 32.1% in the large cannula group; p = 0.03). The incidence of lower limb ischemia was significantly higher in the large cannula group than in the small cannula group (4.6% in the small cannula group vs. 15.4% in the large cannula group; p = 0.02).
The trends of clinical outcomes (survival to discharge, successful weaning from ECMO, lower limb ischemia, and any complications) for each Fr of arterial cannulas are shown in Figure 3. The incidences of lower limb ischemia and any complications significantly increased with increasing cannulas size compared with 14 Fr cannula (p for trend = 0.03 for both), but survival to discharge and successful weaning of ECMO did not.
Among patients who received VA ECMO for “bridge to recovery” (n = 145) and among ECPR patients (n = 90), survival to discharge was not significantly different between the small and large cannula groups (see Tables 1 and 2, Supplemental Digital Content, http://links.lww.com/ASAIO/A315). However, having a large arterial cannula was still associated with significantly higher incidence of lower limb ischemia in both subgroups. Moreover, in the subgroup of ECPR patients, the rate of cannula-related sepsis events was significantly higher in the large cannula group than in the small cannula group.
In the current study, we compared the clinical outcomes and procedural complications between the small arterial cannula strategy (14–15 Fr) and the large arterial cannula strategy (16–21 Fr) in the patients who underwent peripheral VA ECMO. The major findings of this study were as follows: 1) The small arterial cannula strategy showed a similar survival to discharge and ECMO weaning success rate as compared with the large arterial cannula strategy; 2) The event rates of lower limb ischemia and any complications were significantly lower in the small cannula group as compared with the large arterial cannula group and increased with the size of arterial cannulas; and 3) The patients in the small cannula group experienced a shorter duration of ECMO support than did those in the large cannula group.
Although VA ECMO has been increasingly used for the management of cardiopulmonary failure, procedural-related complications occur with high frequency; and the handling of these complications is still challenging.6,11,12 Among such complications, vascular complications caused by percutaneous femoral artery cannulation were associated with a poor clinical outcome.13 A large-sized femoral arterial cannula, the presence of an underlying peripheral artery disease, and a long duration of ECMO support are potential risk factors for developing vascular complications, including lower limb ischemia and cannula site bleeding.14–17 The large femoral arterial cannula of over 15 Fr is routinely used in peripheral VA ECMO for providing full flow for circulatory support. However, there is no consensus concerning the optimal flow for adequate tissue perfusion. For that reason, we hypothesized that if circulatory support by a smaller arterial cannula offers adequate tissue perfusion, use of a smaller cannula could result in clinical outcomes similar to those resulting from the use of a conventionally large cannula while reducing the flow rate and vascular complications. Therefore, we investigated the clinical outcomes and complications according to the size of the arterial cannula.
As expected, the small arterial cannula strategy resulted in similar survival to discharge and successful weaning rate compared with the large cannula strategy. This finding is consistent with a previous retrospective study by Takayama et al.8 concerning the clinical outcomes of using a smaller arterial cannula. However, because the size of arterial cannula is a continuous variable, the design of this study that classified the size of arterial cannulas into the two groups may affected the study results. To confirm whether the study outcomes were diluted or exaggerated by dichotomous analysis, we investigated the trends of clinical outcomes (survival to discharge, successful weaning from ECMO, lower limb ischemia, and any complications) for each Fr of arterial cannulas. Similar to the analysis using arterial cannula size as a binary variable, survival to discharge and successful weaning from ECMO were not associated with the arterial cannula size. On the basis of our findings, full circulatory support using a conventionally large arterial cannula over 15 Fr would not be necessary, especially in Asian populations who have a relatively small BSA. With proper supportive care, the small cannula strategy could provide the optimal flow necessary for adequate tissue perfusion and could be an alternative to the conventionally large cannula strategy.
As we hypothesized, the small arterial cannula strategy showed a lower incidence in lower limb ischemia than the large cannula strategy and the incidence of lower limb ischemia increased with the size of arterial cannulas. However, it would be more accurate to analyze the incidence of lower limb ischemia based on the ratio of arterial cannula size to common femoral artery (CFA) diameter. Unfortunately, we did not measure the diameter of CFA in the study populations because of the nature of retrospective study. Previous studies showed that the diameter of CFA was related to sex and BSA.18,19 In both studies, the diameter of CFA was larger in men and patients with large BSA than women and patients with small BSA, respectively. While Sandgren et al.18 found an increasing CFA diameter with age, the study by Schnyder et al.19 did not. Although not all relevant factors were not corrected, we considered BSA as an estimate of CFA diameter and investigated the incidence of lower limb ischemia according to the ratio of arterial cannula size to BSA. The results are shown in Figure 4. The incidence of lower limb ischemia numerically increased when the ratio of arterial cannula size to BSA was greater than 11. From that finding, we could state that the large size of an arterial cannula would be a risk factor for the development of lower extremity ischemia and that vascular complications could be decreased by the small cannula strategy. Particularly when choosing the size of arterial cannula, it is better to avoid arterial cannula with a ratio of cannula size to BSA greater than 11. This result is different from the finding of the study by Takayama et al.,8 which showed fewer incidences of cannulation site bleeding in the small arterial cannula group and similar incidences of lower limb ischemia among the small and large cannula groups.8 The difference in the incidence of lower limb ischemia between Takayama and this study may be attributed to the lower procedure rates of surgical cut-down and simultaneous distal perfusion of our study. In a meta-analysis of 1,866 patients with either cardiogenic shock or cardiac arrest, the pooled estimate rates of complication were as follows: lower extremity ischemia (16.9%), major bleeding (40.8%), and significant infection (30.4%).6 Our study showed a comparatively lower incidence of complications (lower limb ischemia [9.7%], bleeding events [12.7%], and sepsis [3.6%]) than did the meta-analysis. Both studies’ use of different criteria to define what constitutes complications may be the cause of the differences in the complication rate. For instance, we defined cannula site bleeding as need for surgical wound repair.
In subgroup analysis (of the “bridge to recovery” patients and the ECPR patients), the small arterial cannula strategy consistently showed comparable clinical outcomes as compared with the large cannula strategy. In addition, the large arterial cannula strategy was consistently associated with a higher incidence of lower limb ischemia.
Our study had several limitations. First, the selection of arterial cannula size might be influenced by many factors, such as underlying demographics, patient’s condition, aim of ECMO implantation, and physician’s preference. This potential selection bias might lead to differences in patient baseline characteristics between the small and large cannula group. We performed subgroup analysis to overcome the influence on study outcome by aim of ECMO implantation. Second, the number of enrolled patients is not enough to draw definite conclusions from this study. Third, small cannula strategy was not verified in the following special conditions: patients with a large BSA and patients who received VA ECMO for “bridge to transplantation.” Because patients with BSA over 1.9 m2 comprised only 15.1% (25/165) of our study, it was impossible to demonstrate the feasibility of using a small cannula strategy in obese patients. Also, the safety of small cannula strategy in the “bridge to transplantation” patients was not evaluated because the patients in the “bridge to transplantation” group comprised only 12.1% (20/165) in our study. To ascertain the safety of the application of the small arterial cannula strategy for such groups, additional study in such conditions is necessary. Fourth, we did not have information regarding vascular complications such as limb problems from venous congestion or deep vein thrombosis because the data were collected only through retrospective medical chart reviews.
1. Werdan K, Gielen S, Ebelt H, Hochman JS. Mechanical circulatory support in cardiogenic shock
. Eur Heart J 2014.35: 156–167.
2. Cove ME, MacLaren G. Clinical review: Mechanical circulatory support for cardiogenic shock
complicating acute myocardial infarction. Crit Care 2010.14: 235.
3. Paden ML, Conrad SA, Rycus PT, Thiagarajan RR; Registry ELSO: Extracorporeal Life Support Organization Registry Report 2012. ASAIO J 2013.59: 202–210.
4. Park TK, Yang JH, Jeon K, et al. Extracorporeal membrane oxygenation
for refractory septic shock in adults. Eur J Cardiothorac Surg 2015.47: e68–e74.
5. Sheu JJ, Tsai TH, Lee FY, et al. Early extracorporeal membrane oxygenator-assisted primary percutaneous coronary intervention improved 30-day clinical outcomes in patients with ST-segment elevation myocardial infarction complicated with profound cardiogenic shock
. Crit Care Med 2010.38: 1810–1817.
6. Cheng R, Hachamovitch R, Kittleson M, et al. Complications
of extracorporeal membrane oxygenation
for treatment of cardiogenic shock
and cardiac arrest: A meta-analysis of 1,866 adult patients. Ann Thorac Surg 2014.97: 610–616.
7. Rupprecht L, Lunz D, Philipp A, Lubnow M, Schmid C. Pitfalls in percutaneous ECMO cannulation. Heart Lung Vessel 2015.7: 320–326.
8. Takayama H, Landes E, Truby L, et al. Feasibility of smaller arterial cannulas in venoarterial extracorporeal membrane oxygenation
. J Thorac Cardiovasc Surg 2015.149: 1428–1433.
9. Ha TS, Yang JH, Cho YH, et al. Clinical outcomes after rescue extracorporeal cardiopulmonary resuscitation for out-of-hospital cardiac arrest. Emerg Med J 2017.34: 107–111.
10. Holden MK, Gill KM, Magliozzi MR. Gait assessment for neurologically impaired patients. Standards for outcome assessment. Phys Ther 1986.66: 1530–1539.
11. Abrams D, Combes A, Brodie D. Extracorporeal membrane oxygenation
in cardiopulmonary disease in adults. J Am Coll Cardiol 2014.63(25 pt A): 2769–2778.
12. Stretch R, Sauer CM, Yuh DD, Bonde P. National trends in the utilization of short-term mechanical circulatory support: Incidence, outcomes, and cost analysis. J Am Coll Cardiol 2014.64: 1407–1415.
13. Tanaka D, Hirose H, Cavarocchi N, Entwistle JW. The impact of vascular complications
on survival of patients on venoarterial extracorporeal membrane oxygenation
. Ann Thorac Surg 2016.101: 1729–1734.
14. Bisdas T, Beutel G, Warnecke G, et al. Vascular complications
in patients undergoing femoral cannulation for extracorporeal membrane oxygenation
support. Ann Thorac Surg 2011.92: 626–631.
15. Zimpfer D, Heinisch B, Czerny M, et al. Late vascular complications
after extracorporeal membrane oxygenation
support. Ann Thorac Surg 2006.81: 892–895.
16. Aziz F, Brehm CE, El-Banyosy A, Han DC, Atnip RG, Reed AB. Arterial complications
in patients undergoing extracorporeal membrane oxygenation
via femoral cannulation. Ann Vasc Surg 2014.28: 178–183.
17. Foley PJ, Morris RJ, Woo EY, et al. Limb ischemia during femoral cannulation for cardiopulmonary support. J Vasc Surg 2010.52: 850–853.
18. Sandgren T, Sonesson B, Ahlgren R, Länne T. The diameter of the common femoral artery in healthy human: Influence of sex, age, and body size. J Vasc Surg 1999.29: 503–510.
19. Schnyder G, Sawhney N, Whisenant B, Tsimikas S, Turi ZG. Common femoral artery anatomy is influenced by demographics and comorbidity: Implications for cardiac and peripheral invasive studies. Catheter Cardiovasc Interv 2001.53: 289–295.