High-Caliber Femoral Dual-Lumen Cannula for ECCO₂R in Hypercapnic Respiratory Failure: Efficacy and Safety Evaluation

Abstract

Extracorporeal carbon dioxide removal (ECCO₂R) prevents intubation and facilitates extubation in patients with hypercapnic respiratory failure. However, low-flow systems increase shear stress and need full anticoagulation, increasing the risk of circuit-related complications. We assessed the safety and efficacy of a high-caliber dual-lumen cannula, designed for jugular veno-venous extracorporeal membrane oxygenation (VV ECMO), repurposed for femoral vein insertion, with the aim of achieving higher blood-flow rates and more efficient CO₂ clearance. We retrospectively analyzed 16 intensive care unit (ICU) patients (62 years; 81% chronic obstructive pulmonary disease [COPD]) treated with ECCO₂R using a large-caliber dual-lumen cannula (20–23 Fr) inserted via the femoral vein. Key outcomes included changes in gas exchange, ventilatory support status, and complications. Extracorporeal carbon dioxide removal enabled rapid CO₂ clearance, with arterial carbon dioxide pressure (PaCO₂) decreasing from 68 [62–95] to 49 [45–56] mm Hg at 2 h, and pH increasing from 7.20 [7.16–7.27] to 7.36 [7.33–7.41]. Fifty-six percent of patients avoided intubation, whereas all intubated patients were extubated during ECCO₂R. The median support duration was 5 [4–7] days. No hemolysis was documented. One bleeding episode and one clotting event occurred; no thrombotic or cannulation-related complications were observed. Femoral vein cannulation with a large-caliber dual-lumen cannula for ECCO₂R appears feasible and safe. This strategy may offer technical and clinical advantages over conventional ECCO₂R systems, warranting prospective investigation.

Extracorporeal carbon dioxide removal (ECCO₂R) is a form of extracorporeal respiratory support that selectively removes CO₂ from the bloodstream using low extracorporeal blood flows (BFs). In hypercapnic respiratory failure, particularly during chronic obstructive pulmonary disease (COPD) exacerbations or acute asthma, ECCO₂R improves acid-base status, reduces respiratory workload, and may prevent intubation.^1,2

Modern veno-venous ECCO₂R systems operate at very low BF rates (often <500 ml/min) through small dual-lumen cannulas (Fr 8–14) via the internal jugular or femoral vein,³ with a relatively small carbon dioxide removal (VCO₂) ranging between 70 and 120 ml/min. Such low-flow circuits are prone to blood stasis and thrombotic complications^2,4 due to high shear stress and non-laminar flow patterns in the circuit.

To maintain circuit patency, high levels of systemic anticoagulation are required, which in turn increases the risk of bleeding.⁵ Moreover, in several patients, a higher VCO₂ is needed for a substantial modulation of ventilation, requiring an extracorporeal BF of at least 1 L/min that is almost impossible to obtain with small-caliber catheters due to excessively negative drainage pressures and consequent hemolysis.

To overcome these limitations, at our center, we have adopted an alternative configuration based on the femoral insertion of a large-caliber dual-lumen cannula, originally designed for single-site jugular veno-venous ECMO. To obtain higher BF rates, limit drainage pressures, and reduce turbulence and thrombotic complications, we used cannulas of 19–23 Fr calibers, larger than those typically used for “conventional” ECCO₂R, into the femoral vein.

Femoral placement also simplifies the insertion procedure by eliminating the need for fluoroscopic or echocardiographic guidance and completely removing the risk of right ventricular injury, and it is often feasible without sedation or endotracheal intubation. This strategy may enhance patient comfort, improve tolerance during noninvasive ventilation (NIV), and facilitate physiotherapy and early mobilization.

The aim of the current study was to evaluate the safety and clinical performance of this approach in patients with hypercapnic respiratory failure.

Materials and Methods

Study Design

This retrospective observational study was conducted at the General Intensive Care Unit of the Foundation IRCCS Ca’ Granda Ospedale Maggiore Policlinico (Milan, Italy), an Italian tertiary referral center for respiratory failure and ECMO. The study was conducted following the World Medical Association Declaration of Helsinki ethical principles for medical research involving human subjects⁶ and the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for reporting observational studies.⁷ The study received approval from the Territorial Ethics Committee Lombardia 3 (Reference number 5778_12.03.2025_P). Written consent was waived due to the retrospective design of the study.

Patient Population

All adult patients (age > 18 years) with hypercapnic respiratory failure admitted to the intensive care unit (ICU) and treated with ECCO₂R via femoral dual-lumen cannulation between January 2018 and August 2024 were included in this study. Patients undergoing extracorporeal support primarily for hypoxemic respiratory failure were excluded.

Cannulation Procedure With Large-Caliber Dual-Lumen Cannula

Cannulation was performed with a sterile, percutaneous ultrasound-guided technique. First, the femoral vein was cannulated with a 5 Fr, 10 cm introducer (Terumo Radiofocus Introducer II, Terumo Corp., Japan); then, a 180 cm, 035 inch Amplatz Extra Stiff Wire Guide (Cook Medical, IN) was inserted via the introducer, and single-stage or progressive dilation was performed to facilitate cannula insertion. The right or left femoral vein was chosen according to the anatomy of the patient (Figure 1). At the time of cannulation, a bolus of unfractionated heparin was administered as per institutional guidance (eg, approximately 50–70 U/kg), followed by continuous intravenous infusion.

Figure 1.

Schematic representation of ECCO₂R cannulation and final cannula position as seen on an abdominal X-ray. ECCO₂R, extracorporeal carbon dioxide removal.

Clinical Management of Extracorporeal CO₂ Removal

At the time of ECCO₂R start, extracorporeal BF was progressively increased up to 0.8–1.5 L/min. Subsequently, gas flow was increased very slowly (approximately 0.5 L/min every 5–10 min) to avoid abrupt changes of pH and PaCO₂. Thus, gas flow was adjusted to obtain normal pH by reducing PaCO₂ while allowing for protective ventilation, whether invasive, noninvasive, or spontaneous.

In patients on invasive mechanical ventilation (IMV) at ECCO₂R start, consistently with hemodynamic stability, sedation was progressively reduced, and extubation was evaluated. Spontaneous or NIV, such as early physiotherapy, was always preferred over IMV.

Anticoagulation was carried out following institutional guidance: unfractionated heparin was administered with an activated partial thromboplastin time ratio (aPTTr) target of 1.5–2.0. Laboratory coagulation tests and complete blood count were performed every 8 h. Daily antithrombin levels, hemolysis indicators (ie, carboxyhemoglobin [CO-Hb], haptoglobin, circulating plasma free-hemoglobin), and measurement of circuit pressures (ie, drainage, trans-membrane lung) were performed.

Data Collection

We systematically collected data from a dedicated institutional registry of all patients undergoing ECCO₂R with femoral dual-lumen cannulation. Baseline data included age, sex, body mass index (BMI), Simplified Acute Physiology Score II (SAPS II) at ICU admission, primary diagnosis, and the date of hospital and ICU admission. Additionally, we recorded the date of initiation and termination of MV and ECCO₂R.

Clinical and physiological variables were recorded daily, including ventilation settings, ECCO₂R parameters, hemodynamic data, laboratory values, and any adverse events, such as bleeding or thrombosis. Data regarding respiratory physiotherapy and patient mobilization were also collected, including details on the frequency and duration of mobilization sessions. The duration of IMV, length of ICU stay, and in-hospital mortality were systematically documented.

Statistical Analysis

Descriptive statistics were used to summarize the data. Continuous variables are presented as median [interquartile range], whereas categorical variables are expressed as absolute numbers and percentages. Changes over time were assessed using the Friedman test for repeated measures of continuous variables and Cochran’s Q test for categorical variables. When appropriate, pairwise comparisons were performed using the Wilcoxon signed-rank test. Analyses were conducted using Stata (StataCorp, College Station, TX) and R (R Foundation for Statistical Computing, Austria, version 4.1.2).

Results

Study Population

This study included 16 patients admitted to the ICU for hypercapnic acute respiratory failure from February 2018 to August 2024. Patients were mostly male (62%), 62 [54–65] years old, with a BMI of 28 [23–33] kg/m². The main etiology of the respiratory failure was COPD (81%), followed by asthma (19%), with a median SOFA score of 7 [5–8] and Acute Physiology and Chronic Health Evaluation II (APACHE II) of 13 [9–15] at baseline, as described in Table 1.

Table 1.

Characteristics of the Study Population, Pre-ECCO₂R Ventilation and Circuit Management

Parameters	Overall Population N = 16
Age (years)	62 [54–65]
Gender (%)  Female  Male	6 (38) 10 (62)
BMI (kg/m2)	28 [23–33]
Comorbidities (%)   Chronic lung disease   Hypertension   CKD   Immunocompromised   Diabetes	13 (81) 7 (44) 2 (13) 2 (13) 2 (13)
ARF etiology (%)   COPD   Asthma	13 (81) 3 (19)
SOFA at admission	7 [5–8]
APACHE II at admission	13 [9–15]
Pre-ECCO2R mechanical ventilation days (days)	0 [0–0]
Pre-ECCO2R noninvasive ventilation days (days)	0 [0–1]
ECCO2R and anticoagulation
Cannulation technique: percutaneous (%)	16 (100)
Cannula size (%)    20 Fr    23 Fr	8 (50) 8 (50)
Anticoagulation: UFH (%)	16 (100)
UFH bolus dose at cannulation (U/kg)	67 [56–71]
UFH dose (U/kg/h)	14 [11–19]
aPTTratio	1.6 [1.4–1.9]
ATIII (%)	95 [81–105]
Blood products during ECCO2R course   PRBC   Fresh frozen plasma   Platelets pool	0 [0–0] 0 [0–0] 0 [0 –0]

aPTT, activated partial thromboplastin time; BMI, body mass index; COPD, chronic obstructive pulmonary disease; ECCO₂R, extracorporeal CO₂ removal; PRBC, packed red blood cells; SAPSII, Simplified Acute Physiology Score II; SOFA, Sequential Organ Failure Assessment; UFH, unfractionated heparin.

At baseline, 44% of the patients were on NIV, 12% on high-flow nasal oxygen (HFNC), and 44% were on IMV (Table 2).

Table 2.

Time Course of Ventilation Variables, Blood Gases, ECCO₂R Operational Characteristic and Hemodynamics

	Parameters	Pre-ECCO2R N = 16	ECCO2R 2 Hours N = 16	ECCO2R Day 1 N = 16	ECCO2R Day 3 N = 14	ECCO2R Day 5 N = 11	p Value
Respiratory mechanics	Ventilation mode N (%)  HFNC  NIV  IMV	2 (12) 7 (44) 7 (44)	–	3 (19) 7 (44) 6 (37)	7 (50) 6 (43) 1 (7)	7 (64) 3 (27) 1 (9)	0.002 0.29 0.01
	Tidal volume (ml)	500 [425–525]	–	420 [400–550]	415 [325–550]	–	0.35
	Respiratory rate (bpm)	22 [17–28]	–	18 [15–21]	16 [14–19]	17 [16–18]	0.03
	PEEP (cm H2O)	6 [5–10]	–	9 [8–12]	10 [8–10]	8 [6–11]	0.02
	*Plateau pressure (cm H2O)	22 [20–25]	–	21 [19–22]	18 [18–19]	–	0.71
	*Driving pressure (cm H2O)	10 [7–12]	–	13 [9–15]	9 [8–10]	–	0.60
	*Compliance (ml/cm H2O)	51 [39–76]	–	46 [38–60]	50 [43–69]	–	0.60
Gas exchange	pH	7.20 [7.16–7.27]	7.36 [7.33–7.41]	7.41 [7.38–7.45]	7.42 [7.40–7.46]	7.42 [7.40–7.46]	0.0006
	PaCO2 (mm Hg)	68 [62–95]	49 [45–56]	48 [40–58]	47 [39–54]	49 [39–52]	0.0004
	PaO2/FiO2	170 [96–247]	163 [107–262]	184 [112–255]	153 [127–221]	159 [102–189]	0.81
Hemodynamics	Heart rate (bpm)	97 [88–105]	–	89 [80–95]	71 [67–85]	66 [57–93]	0.01
	MAP (mm Hg)	74 [70–83]	–	73 [66–89]	84 [77–90]	81 [65–87]	0.09
	Lactate (mg/dl)	1.2 [0.9–2.2]	1.5 [1.2–2.2]	1.2 [1.1–1.8]	1.2 [0.8–1.6]	1.2 [1.1–1.6]	0.95
	Vasoactive support (%)	8 (50)	–	9 (56)	1 (7)	0 (0)	0.00004
ECCO2R	Blood flow (L/min)	–	1.5 [1.2–2.0]	1.4 [1.3–1.6]	1.3 [1.2–1.6]	1.4 [1.3–1.8]	0.31
	Sweep gas (L/min)	–	4.5 [3.0–7.0]	5.0 [3.0–8.0]	3.0 [2.0–3.5]	4.0 [3.0–6.5]	0.26
	VCO2 ML (ml/min)	–	195 [163–228]	147 [130–168]	117 [101–149]	130 [108–156]	–
	Drainage (mm Hg)	–	−33 [−24 to −42]	−49 [−40 to −55]	−47 [−40 to −58]	−70 [−55 to −75]	0.57
Hemolysis	Platelets (103 mmc)	265 [218–290]	–	177 [146–232]	157 [144–237]	191 [119–305]	0.009
	D-dimer (mcg/ml)	2032 [464–2873]	–	1253 [289–2241]	1111 [245–1958]	1718 [851–2801]	0.20
	Haptoglobin (mg/dl)	147 [145–149]	–	205 [113–331]	249 [86–347]	312 [130–397]	–
	Free Hb (mg/dl)	–	–	9.8 [6.1–16.0]	8.2 [6.3–14.0]	9.0 [3.2–15.0]	–
	CO-Hb (%)	0.9 [0.3–1.6]	–	1.0 [0.6–1.5]	0.9 [0.5–1.3]	0.9 [0.5–1.3]	0.73

^*Only for patients on mechanical ventilation.

CO-Hb, carboxyhemoglobin; HFNC, high-flow nasal cannula; IMV, invasive mechanical ventilation; MAP, mean arterial pressure; ML, membrane lung; NIV, noninvasive ventilation; PCV, pressure control ventilation; PEEP, positive end-expiratory pressure; PSV, pressure support ventilation; VCV, volume control ventilation.

Before ECCO₂R start, the median respiratory rate was 22 [17–28] bpm, PaCO₂ 68 [62–95] mm Hg, pH of 7.20 [7.16–7.27], and PaO₂/FiO₂ ratio of 170 [96–247].

Extracorporeal carbon dioxide removal was instituted on the same day as the initial respiratory support. Invasively ventilated patients had a median compliance of the respiratory system of 51 [39–76] ml/cm H₂O, with a plateau airway pressure of 22 [20–25] cm H₂O and a driving pressure of 10 [7–12] cm H₂O (Table 2).

Cannulation

The bilumen cannula was placed in the right femoral vein using a percutaneous technique in all cases. The Avalon Elite dual-lumen cannula (Getinge, Rastatt, Germany) was used in all patients, with a 20 Fr cannula inserted in eight cases (50%) and a 23 Fr cannula in eight cases (50%). Extracorporeal carbon dioxide removal was performed using a Quadrox-i Adult hollow fiber membrane oxygenator (Getinge, Rastatt, Germany), with the Rotaflow or the Cardiohelp system consoles (Getinge, Rastatt, Germany). No complications related to cannulation were documented. Throughout the study period, no cases of cannula displacement, cannulation site bleeding, or cannulation site infection were observed in any patient.

Extracorporeal Carbon Dioxide Removal Performance and Impact on Gas Exchange

After 2 h of ECCO₂R support, PaCO₂ levels dropped to 49 [45–56] mm Hg and remained stable thereafter. Accordingly, arterial pH increased from 7.20 [7.16–7.27] before ECCO₂R support to 7.36 [7.33–7.41] and then stabilized around 7.40 from day 1 and throughout the subsequent days of ECCO₂R support. The amount of CO₂ removed by the extracorporeal circuit was 195 [163–228] ml/min 2 h after the start of ECCO₂R and decreased to 147 [130–168] ml/min on the first day, with a median sweep gas flow of 4.5 [3.0–7.0] L/min and a median extracorporeal BF of 1.5 [1.2–2.2] L/min. These parameters remained relatively stable during the following days and gradually decreased during the weaning phase. There was a significant change in PaCO₂ and pH levels over time (p < 0.05), as shown in Table 2 and Figure 2.

Figure 2.

Time course of pH, RR, PaCO₂, and membrane lung VCO₂ during ECCO₂R. The figure shows the time course of key physiological parameters during ECCO₂R. The top panel illustrates pH (green) and respiratory rate (RR, blue), whereas the bottom panel depicts PaCO₂ (red) and VCO₂ (orange). Data points represent medians with bars indicating interquartile ranges. ECCO₂R, extracorporeal carbon dioxide removal; PaCO₂, arterial carbon dioxide pressure; VCO₂, membrane lung carbon dioxide elimination.

The extracorporeal BF was achieved with a median drainage pressure of −33 [−24 to −42] mm Hg, which slightly decreased over the days (Table 2). The transmembrane pressure was 10 [6–15] mm Hg and remained stable throughout the ECCO₂R support period. We did not observe hemolysis: free Hb was always less than 20 mg/L, haptoglobin levels were normal, and CO-Hb remained consistently less than 2%, without significant variations.

A direct calculation of the circuit’s recirculation fraction (R/BF) was not feasible, as the value of cardiac output was not available in most of the patients. However, an estimate of R/BF may be obtained by interpolating available data using a mathematical model we previously developed.⁸ Given a median BF of 1.4 [1.3–1.6] L/min at day 1 of ECCO₂R support, oxygen saturation of the blood entering the membrane lung was 82 [79–87]%, and central venous oxygen saturation (ScvO2) of 67 [62–76]%, assuming a CO ranging from 4 to 6 L/min, we could estimate R/BF ranging from 40% up to 60% (see Figure S1, Supplemental Digital Content, https://links.lww.com/ASAIO/B679).

Outcomes and Complications

The median duration of ECCO₂R support was 5 [4–7] days, and the median duration of IMV during ECCO₂R support was 0 [0–3] days (Table 3). All invasively ventilated patients at admission were extubated on the first 2 days after ECCO₂R institution, and only one required reintubation due to worsening of the hypoxemic component of respiratory failure. None of the patients on NIV or high-flow nasal oxygen (HFNO) at ICU admission was intubated during the ICU stay. The median ICU length of stay was 9 [8–10] days. One patient (6%) died during the ICU stay (Table 3). The median hospital length of stay was 22 [14–36] days; one patient died on the ward, resulting in an overall in-hospital mortality of 13%.

Table 3.

Study Outcomes

Length of ECCO2R support (days)	5 [4–7]
Length of mechanical ventilation (days)	0 [0–3]
Patient extubated during ECCO2R	16 (100)
Patient reintubated during ECCO2R (%)	1 (6)
Tracheostomy (%)	0 (0)
ICU LOS (days)	9 [8–10]
ICU mortality (%)	1 (6)
Hospital LOS (days)	22 [14–36]
Hospital mortality (%)	1 (6)
Complications
ECCO2R complication (%)   Cannulation site infection   Bleeding event during ECCO2R course   Hemolysis   Thrombosis    ML clotting (circuit change)   Catheter displacement	0 (0) 1 (6) 0 (0) 0 (0) 1 (6) 0 (0)

ECCO₂R, extracorporeal CO₂ removal; ICU, intensive care unit; LOS, length of stay; ML, membrane lung.

All patients received continuous intravenous unfractionated heparin, with an average dose of 14 [11–19] U/kg/h. During treatment, the median aPTTr was 1.6 [1.4–1.9]. Two sequential bleeding events were recorded in a single patient, requiring 4 and 1 packed red blood cells (PRBC), respectively, and a transient UFH interruption. No other episodes of clinically relevant bleeding were observed, and the median number of PRBC units transfused was 0 [0–0]. No thrombotic events were observed during ECCO₂R support and during the ICU stay.

Discussion

This study investigates the safety and clinical efficacy of a novel approach to ECCO₂R based on the femoral insertion of a large-caliber dual-lumen cannula, with the aim of obtaining a relatively high extracorporeal BF and maximizing the amount of CO₂ extracted by the extracorporeal circuit.

Although the clinical efficacy of ECCO₂R in acute COPD exacerbations has been explored extensively in prospective clinical studies,^2,9–13 several patients require high levels of CO₂ removal (ie, >150 ml/min) to achieve clinical benefit. These VCO₂ targets usually necessitate BFs of 1–2 L/min,¹⁴ which are difficult to reach with small-caliber cannulas without inducing significant shear stress and associated complications.^15–17 For these reasons, low-flow systems combining renal and respiratory support may be technically and physiologically limited and often ineffective in achieving these goals.¹⁸ By employing a high-caliber (ie, 20–23 Fr) dual-lumen cannula in a femoral position, we were able to achieve high and stable BFs with drainage pressures always greater than −50 mm Hg, without clinically significant hemolysis and a low complication rate. This translated into a rapid and sustained reduction in PaCO₂, accompanied by improved arterial pH, facilitating early extubation and ICU discharge in the majority of cases. Only one patient required a transfusion due to bleeding at ECCO₂R initiation, and one experienced extracorporeal circuit clotting. Notably, all patients were successfully extubated during ECCO₂R support, with one case of reintubation due to hypoxemia ultimately resulting in ICU death.

Efficiency of Femoral Large-Caliber Dual-Lumen Cannulation in CO₂ Removal

Traditionally, ECCO₂R is performed using either two small-caliber cannulas or a smaller dual-lumen cannula (eg, 8–15 Fr), which limits BF to a maximum of 500 ml/min².¹³ Although this flow is often not sufficient for clinically meaningful VCO₂, it may increase the risk of hemolytic, thrombotic, or hemorrhagic complications.^15–17 The use of a large-caliber dual-lumen cannula in the femoral vein was driven by the goal of providing effective extracorporeal CO₂ removal while minimizing the risks of bleeding and thrombosis. Moreover, the use of smaller cannulas, which often results in elevated negative drainage pressures to maintain an adequate BF, can substantially increase shear stress. This mechanical stress may trigger ADAMTS13 activation, contributing to acquired von Willebrand syndrome, especially in the presence of systemic anticoagulation, and thereby heightening the risk of bleeding.⁴

The large-caliber dual-lumen cannula, originally designed for veno-venous ECMO, allows higher BF rates, potentially mitigating the aforementioned limitations. This cannula is geometrically engineered for insertion into the right internal jugular vein, allowing blood drainage from both the superior and inferior vena cava and reinfusion into the right atrium, directed towards the tricuspid valve. However, its use in the femoral vein places both the drainage and reinfusion ports within the inferior vena cava, increasing the risk of recirculation, which represents a critical issue when the extracorporeal circuit is needed to support oxygenation. On the contrary, even elevated recirculation rates have a negligible effect on CO₂ removal, which remains the primary goal in hypercapnic patients. The efficiency of CO₂ clearance, even in the presence of recirculation, can be attributed to the unique dynamics of gas exchange in ECCO₂R, where CO₂ elimination is achieved at significantly lower BF rates than those required for oxygenation. In fact, limiting oxygen delivery while removing CO₂ may preserve some hypoxic pulmonary drive and help prevent lung derecruitment associated with hyperoxia.¹⁹ A limitation of femoral cannulation for ECCO₂R is the inability to provide oxygenation support, which may become critical in patients with significant hypoxemia, as exemplified by one patient in our cohort who required reintubation. Although jugular cannulation allows for oxygenation support, it is associated with greater procedural complexity and a higher risk of complications. Therefore, the choice of cannulation site should weigh the procedural simplicity and lower risk of femoral access against the potential need for oxygenation support, based on the individual patient’s characteristics and respiratory condition.

In our series, the increase in PEEP observed after ECCO₂R initiation likely reflected both ventilatory strategy adjustments and underlying physiological changes. Reduced driving pressure was possible due to decreased need for alveolar ventilation, prompting PEEP elevation to maintain mean airway pressure. In addition, ECCO₂R lowers the respiratory quotient by reducing native carbon dioxide elimination rate by the membrane lung (VCO₂) clearance, which may reduce alveolar and arterial PO₂ per the alveolar gas equation.²⁰ This, together with potential reabsorption atelectasis from regional nitrogen washout, may predispose to lung derecruitment. Increasing PEEP in this context serves as an anticipatory measure to preserve lung recruitment and oxygenation.

Although our initial ECCO₂R median BF was 1.5 L/min, higher than in most studies on extracorporeal support for acute respiratory failure (eg, up to 0.55 L/min in the REST and VENT-AVOID trials^2,11), effective CO₂ removal was achieved within the first hours of ECCO₂R initiation and was maintained throughout the treatment (Figure 2).

Safety

Bleeding and other adverse events remain significant concerns with current ECCO₂R devices, as seen in studies such as SUPERNOVA and REST.^11,12 The requirement for full systemic anticoagulation in low-flow ECCO₂R systems increases the risk of hemorrhagic complications, with reported rates ranging from 2% to 50% among COPD populations.¹³ However, the use of a large-caliber dual-lumen cannula, which allows for higher BF rates compared to traditional small-caliber cannulas, may mitigate these risks by reducing the need for high systemic anticoagulation. In our cohort, anticoagulation with unfractionated heparin was managed according to an aPTT-based institutional protocol. The median aPTTr was maintained at 1.6 during extracorporeal therapy, consistent with the lower bounds of the targets reported in recent trials.^2,11

No cases of cannula displacement, insertion-site bleeding, or infection were observed. Packed red blood cell transfusions were largely unnecessary in this cohort, with a median requirement of zero units. Only one patient experienced bleeding at the initiation of ECCO₂R, requiring blood product transfusions. The same patient later required membrane lung (ML) replacement due to clot formation, highlighting the potential risks associated with prolonged support (ie, 17 days).

Thrombus formation is a well-recognized risk in low-flow ECCO₂R configurations due to the prolonged exposure of blood to the ML and extracorporeal circuit. However, the use of a large-caliber dual-lumen cannula allowed BF above 1.0 L/min, likely reducing the risk of thrombosis compared to systems operating at lower flow rates. Indeed, no thrombotic events were observed in our patient cohort during ECCO₂R support.

In vitro studies have demonstrated that centrifugal pumps are optimized for high-flow rates, with an increased risk of blood damage occurring when operated at low flows.²¹

Studies of high-resolution computational fluid dynamics showed that hydraulic efficiency critically decreases when rotary blood pumps operate at BF rates below 1 L/min, with a potentially increased risk of adverse effects due to exposure to high shear stress.⁴

During ECCO₂R in adults, centrifugal pumps operating at low-flow rates are associated with a significantly higher incidence of hemolysis compared to their use at higher flows, as seen in veno-venous ECMO.¹⁷ Conversely, high blood-flow rates (3.0–4.5 L/min) through small-diameter cannulas have been shown to markedly increase blood cell trauma.¹⁷

At last, by integrating in silico computational fluid dynamics predictions with clinical data from 580 patients treated with veno-venous ECMO, Lehle et al.¹⁷ demonstrated that a key factor in reducing the incidence of ECMO-related side effects is lowering circuit resistance, including optimizing cannula diameter: the lower the flow rate, the more pronounced the influence of circuit resistance on hemolysis. In our study, BF of 1.5 [1.2–2.0] was not associated with increased hemolysis parameters (see Figure S1, Supplemental Digital Content, https://links.lww.com/ASAIO/B679), underscoring the probable synergistic combination of high BF and reduced cannula resistance due to higher calibers.

The low incidence of hemorrhagic, thrombotic, and hemolytic complications observed in this study is quite interesting, given that higher BF rates, enabled by the use of a high-caliber cannula, not only improve CO₂ removal efficiency but also reduce mechanical stress on blood components. Although larger cannulas might be perceived as more invasive, their clinical impact includes improved BF, which reduces stasis and shear stress, major drivers of hemolysis and thrombosis in ECCO₂R. Furthermore, femoral placement may pose less interference with breathing efforts in spontaneously breathing patients with hypercapnic respiratory failure. By optimizing flow and reducing these risks, our approach may enhance patient safety while managing hypercapnia and avoiding escalation to IMV.

Limitations of the Study

Several limitations must be acknowledged. First, the retrospective design of this study precludes any causal inference. This work is intended to document clinical experience rather than to claim therapeutic superiority. Given its observational nature, the study aims to generate hypotheses regarding safety and efficacy rather than to establish definitive conclusions.

Second, the small sample size and the single-center design limit the generalizability of our findings. The absence of a control group and the limited sample size restrict the external validity of our findings. Further controlled studies are necessary to validate these preliminary observations. Importantly, key safety outcomes such as hemolysis are not binary events, but rather continuous phenomena that require prospective, high-quality data collection for accurate assessment.

Conclusions

Femoral vein cannulation with a large-caliber dual-lumen cannula for ECCO₂R in ICU patients with acute hypercapnic respiratory failure appears to be a feasible, safe, and effective strategy. This approach provides enhanced CO₂ removal without a significant increase in hemorrhagic, hemolytic, or thrombotic complications, and may therefore be considered even in patients with ARDS to facilitate ultraprotective ventilation. The findings provide a foundation for prospective studies to further investigate the impact of femoral dual-lumen cannulation on clinical outcomes in ECCO₂R support and emphasize the need for the development of appropriately sized dual-lumen cannulas specifically designed for femoral placement and ECCO₂R use.

Acknowledgments

The authors extend our heartfelt gratitude to the dedicated team of medical professionals, including nurses, residents, and students at the Adult Intensive Care Unit of Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico in Milan.

Supplementary Material

mat-72-255-s001.pdf ^{(392.1KB, pdf)}