Extracorporeal membrane oxygenation (ECMO) is a mechanical cardio-respiratory support for heart and lung failure in which conventional treatment is ineffective and is increasingly becoming a bridge to permanent solution such as recovery, transplantation, more durable device, or decision. Over the past decade, the indications for ECMO have expanded, and the use of ECMO has increased exponentially (1–3). Establishing veno-arterial (V-A) ECMO by percutaneous femoral artery and venous cannulation is easy to perform, minimally invasive, and widely used in clinically. However, femoral artery cannulation may result in acute limb ischemia (ALI) of the ipsilateral limb. Severe limb ischemia can render the arterial blood supply of the lower extremities unable to meet the basic physiological and metabolic needs at rest (4–8), which may require secondary surgical intervention, or amputation and even be life-threatening (9–11). In a systematic meta-analysis by Cheng et al., the cumulative incidence of limb ischemia in ECMO-assisted patients was 16.9% (12). Therefore, detection, prevention, and management of ECMO-related ALI are critical (12–14). Many prevention strategies have been proposed to avoid limb ischemia including the selection of small arterial return cannula and cannulation side selection, cannulation technique, and placement of a smaller cannula for anterograde or retrograde (ankle) distal perfusion (15). Recently, a new bidirectional femoral arterial cannula has been proposed and tested during cardiopulmonary bypass (16). With peripheral femoral cannulation, distal perfusion ipsilateral to the femoral artery cannulation is recommended by the ELSO Red Book and guidelines (1). Currently, the placement of the distal perfusion cannula (DPC) is the most common method for preventing and treating limb ischemia complications (17, 18). However, the timing of DPC application is not clearly defined. There is a lack of clear evidence-based medical recommendations as to whether the distal perfusion catheter should be placed prophylactically or only if acute limb ischemia develops (1, 18–21).

We reviewed the data of adult V-A ECMO-assisted patients at our hospital, as well as discussed the risk factors, clinical features, and summarized the prevention and treatment of limb ischemia complications in order to improve the prognosis of these patients.

As a powerful life-support device, V-A ECMO is being increasingly used in clinical practice. This technique is an invasive procedure and may lead to related complications. Limb ischemia complications are often an important factor affecting the effect of ECMO. Previous studies reported that the incidence of distal limb ischemia on the ipsilateral femoral artery cannulation in patients with peripheral V-A ECMO ranges from 10 to 50% (25, 26). This may be related to differences in ECMO indications, disease distribution, baseline characteristics, implantation techniques, definition of limb ischemia, detection tools, and timing of DPC insertion. In a study of Rastan et al. (24), 5.4% of patients with femoral artery cannulation experienced limb ischemia, and the use of DPC reduced limb ischemia and fasciotomy to <40%. In a cohort study of 75 patients, the overall rate of limb ischemia was 14.7%, and the rate of limb ischemia remained 4.6% in patients treated with conventional DPC (21). Limb ischemia complications have been reported to be associated with unsuccessful ECMO weaning and have been shown to be independent predictors of in-hospital mortality (27, 28). We reviewed the incidence of limb ischemia complications, clinical features, and effectiveness of salvage DPC placement in adult patients assisted by V-A ECMO, as well as compared the effect of limb ischemia on ECMO weaning rate, discharge rate, and length of hospital stay. Our study found that the ECMO weaning and discharge rates in the LI group were significantly lower than those in the nLI group, which is consistent with previous reports in the literature. There was no significant difference in length of hospital stay between the two groups, which may be related to a higher rate of recovery and fewer serious complications. Our center is an adult heart center, with most patients with cardiogenic shock. The total incidence of limb ischemia was 10.3%, and the incidence of severe limb ischemia requiring salvage with a DPC placement was 6.5%. The incidence was relatively low, which may be related to disease distribution and treatment strategies.

ALI associated with V-A ECMO results from relative or absolute insufficiency of arterial blood flow to distal tissues, which is related to multiple factors. Larger size arterial cannulations may result in reduced lumen area and even vascular damage, resulting in limb ischemia (29). However, this relationship is not correlated, and ECMO cannula size depends on body surface area and ECMO mode, and is based on a match between target flow and vessel diameter (30). There was no significant difference in the size of arterial cannulation between the two groups in this study. Because of the lack of collateral circulation and the smaller diameter of the femoral artery, women and younger patients are prone to limb ischemia. The peripheral atherosclerotic disease also increases the risk of plaque displacement and injury during intubation and cannulation (7, 13). Diabetes and chronic respiratory disease are independent risk factors for limb ischemia during V-A ECMO (5, 31), which may be associated with chronic endothelial injury (32). In this study, the proportion of diabetes in the limb ischemia group was higher than that in the non-limb ischemia group, and multivariate analysis indicated that limb ischemia was independently associated with diabetes. Danial et al. (32) found that limb ischemia was independently associated with the Sequential Organ Failure Score (SOFA score). This also suggests that the patient's ability to compensate for hypoperfusion may affect vascular function. In addition, before the establishment of V-A ECMO, most patients are in a state of shock, requiring large doses of vasoactive drugs to maintain, especially in the ECPR state that causes vasoconstriction and acidic metabolites, which may affect distal perfusion. Our study also found that the LVEF in the ischemia group was significantly lower, the cardiac function was worse, the proportion of ECPR was higher than that in the non-ischemic group, and the VIS-max was significantly higher than that in the non-limb ischemia group. Therefore, after the establishment of ECMO, limb ischemia can be improved by drug adjustment (including reducing vasopressors, increasing vasodilators, and adequate anticoagulation) in some patients (9). For LV venting, patients with V-A ECMO-assisted cardiogenic shock often require a combined IABP, which may improve patient outcomes (33), but bilateral femoral arterial cannulation may lead to an increased risk of ALI (34). Multivariate regression analysis in this study also found that use of IABP in V-A ECMO patients was an independent risk factor for limb ischemia, and the application of dual distal perfusion catheters may be a strategy for the treatment of such patients (35).

Considering the hazards caused by limb ischemia complications, all the patient on peripheral V-A ECMO should be closely monitored, especially unconscious patients. Clinical symptoms and signs are widely used in clinical practice as the primary assessment method. The classical description of patients with limb ischemia is grouped into a mnemonic device known as the 6 Ps, including: pain, pallor, paralysis, pulseless, paraesthesia, and paralysis (36). However, these signs may be deceptive for various reasons. The continuous flow of ECMO may render patients less pulsatile and make distal pulses less detectable. In patients with cardiogenic shock, the extremities often become colder due to a high dose of inotropes and pressors despite adequate perfusion. In addition, patients with endotracheal intubation and sedation and analgesia are unable to make immediate judgments. In these cases, a more sensitive tool to detect early distal-limb ischemia is warranted. Near-infrared reflectance spectroscopy (NIRS) technology can reflect local oxygen saturation (RSO2), and tissue oxygen saturation below 50% for a duration of more than 4 min is strongly associated with hypoperfusion (37, 38). Our center is mainly for awake and non-tracheal cannula ECMO patients, and the early clinical situation is easier to monitor. However, in special populations, the combined use of NIRS to measure bilateral tissue oxygen saturation (SmO2) is helpful for early identification. The technical aspects of the various NIRS systems are not consistent. We have been using near-infrared reflectance spectroscopy as an auxiliary assessment tool for limb hypoperfusion since January 2021 but more samples are still needed to further validate its clinical relevance. Patients were also monitored to detect NIRS tissue saturation (StO2) differentials of 15% or more between the cannulated limb and the contralateral limb for the diagnosis of cannula-related obstruction to flow. The literature search for this paper found no references for intervention or cautionary values specific to distal limb saturations.

Prevention and treatment of limb ischemia should be individualized. Choosing a smaller size cannulation provides considerable clinical support and reduces complications, which is feasible in clinical practice (9, 23, 29, 39). In addition, evaluation of vascular access and selection of cannulation sites play an important role in preventing limb ischemia (1, 40–42). Proper anticoagulation, drug optimization and adequate perfusion of distal limb are necessary for patients who have occurred limb ischemia complications (9). At present, DPC is still the most commonly used invasive technique for the prevention and treatment of limb ischemia. In a meta-analysis of 1,850 patients, prophylactic DPC placement was found to reduce the incidence of limb ischemia (43). Rastan et al. (24) found that the use of DPC can reduce the risk of lower extremity ischemia and surgical intervention to <40%, so it is recommended to routinely place DPC in patients with peripheral V-A ECMO. However, there are corresponding risks associated with catheter placement, which may lead to lower extremity-related complications. In a retrospective study of 84 adult patients with V-A ECMO, Tanaka et al. found that even with prophylactic DPC placement, limb ischemic complications occurred in 12% of patients (13). Some studies suggested that DPC should be used as a salvage measure for limb ischemia after clinical assessment (21). In addition, the types of DPC used by different centers also vary greatly. The size of DPC ranges from 5 to 14 Fr and the most commonly used types are central venous catheters and vascular guide sheaths (usually 6–8 Fr) (44). In our study, we adopted the regimen of drug adjustment combined with DPC. For patients who cannot be improved by medication treatment or severe limb ischemia, DPC was placed as a salvage intervention. The limb ischemia in the whole group was improved, and no adverse events occurred, indicating DPC as a salvage intervention is feasible. Therefore, in order to prevent and treat V-A ECMO-related limb ischemia, we summarize it as a “4S” scheme (Best Site selection, Match arterial cannula Size, Systematic evaluation, Salvage intervention). During the establishment of V-A ECMO, a reasonable cannulation strategy can reduce the occurrence of limb ischemia complications to a certain extent. After ECMO was established, adequate timely evaluation and placement of DPC as a salvage intervention according to clinical conditions.

This retrospective case-control study yielded some meaningful results for clinical guidance. However, there are still some limitations. First, this was a single-center retrospective study, and the distribution of disease are relatively single, which may cause selection bias. Second, no more objective indicators have been recorded and NIRS was not used in all patients to assess limb ischemia. In addition, there is a lack of short-and long term follow up on prognosis. Therefore, it is necessary to conduct a randomized controlled study to confirm the effectiveness of the placement of DPC as a salvage intervention.

In this study, limb ischemia complications were uncommon in patients receiving V-A ECMO assistance, but their occurrence may affect the ECMO weaning rate and survival rate. Diabetes, IABP, and VIS-max were independent risk factors for predicting limb ischemia complications in patients with V-A ECMO. It should be adequately evaluated during and after the establishment of V-A ECMO. DPC can be used as a measure of salvage for critical limb ischemic complications.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ethics Committee of Affiliated the First Hospital of Lanzhou University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

SH, AL, and MB: contributed to study concept and design. YW and WQ: acquisition and analysis of data. SH and AL: writing of the original manuscript and statistical analysis. MB and AL: revision and editing of the manuscript. BZ and CP: material, administrative support, and supervision. All authors approved the final version of the manuscript and agree to be responsible for all aspects of the work.

This work was supported by the Key Science and Technology Foundation of Gansu Province (No. 21YF5FA118) and the Natural Science Foundation of Gansu Province (No. 21JR7RA385).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.