Left ventricular (LV) distension is a common complication during venoarterial extracorporeal membrane oxygenation (VA-ECMO) [1], and LV distension may cause subsequent problems [2-4]. Therefore, several strategies have been devised to improve LV distension. An intraaortic balloon pump, transaortic ventilation via the pigtail catheter, a catheter-based transaortic axial flow pump, left atrial (LA) decompression by venting catheter insertion, and interatrial septostomy are possible solutions [2]. These procedures have been found to be efficacious [5,6]. However, there is no consensus on the appropriate method for LV decompression, and the flow rate required for proper decompression is unknown.

LA venting catheter insertion is a commonly performed procedure for LV decompression in patients receiving VA-ECMO. However, the LA venting procedure may cause procedure-related complications, especially when a large catheter is inserted, whereas a small catheter may lead to problems of inadequate LV decompression. Therefore, the size of the catheter for the procedure should be selected carefully.

We hypothesized that LA drainage with an 8-Fr catheter, which is smaller than the catheter usually inserted for LV decompression in adult patients, may effectively decompress the left heart for adult patients receiving VA-ECMO who show signs of LV distension. For patients who showed signs of LV distension, we performed left heart decompression by inserting an 8-Fr Mullins sheath in the LA and evaluated the final outcomes and efficacy of the procedure, focusing on the improvement of pulmonary edema and development of pulse pressure.

The left heart decompression procedure was performed in seven patients during VA-ECMO. The median age of the patients was 58 years (range, 22 to 70 years) and body surface area was 1.78 m2 (range, 1.54 to 2.12 m²). Three patients had acute myocardial infarction, three patients had fulminant myocarditis, and one patient had dilated cardiomyopathy. Four patients had extracorporeal cardiopulmonary resuscitation. The left heart decompression procedure was performed immediately after the start of ECMO in four patients. In one patient, the procedure was performed within 24 hours from the start of ECMO. Among all seven patients, two patients were not supported by a mechanical ventilator. Before the left heart decompression, four patients showed severe pulmonary edema and four showed decreased pulse pressure of <10 mm Hg (Table 1). The median duration of VA-ECMO in all the patients was 185 hours (range, 82 to 342 hours). All the patients were successfully weaned from the ECMO, although one patient died of septic shock three weeks after ECMO weaning. One patient underwent heart transplantation (Table 2).

After the left heart decompression procedure, four patients showed improvement of pulmonary edema within 3 days (Figure 1). Among these four patients, two showed marked improvement of pulmonary edema within 24 hours. Among the other three patients, one took 6 days to attain improvement of pulmonary edema, and the other two developed pneumonia; therefore, we could not estimate the effect of left heart decompression by the improvement of pulmonary edema from the chest radiograph. Four patients, including two with very weak pulse pressure (<5 mm Hg) and one without improvement of pulmonary edema, showed a noticeable increase in pulse pressure within 24 hours from the left heart decompression procedure. One patient experienced aggravation of pulmonary congestion and decreased pulse pressure due to obstruction of the decompression catheter (Table 2).

For the seven patients who received VA-ECMO and had LV distension, we performed a left heart decompression procedure using an 8-Fr Mullins sheath. We checked the efficacy of the procedure in all seven patients, all of whom were successfully weaned from VA-ECMO, including one who underwent heart transplantation. Four patients showed improvement of pulmonary edema within three days from the procedure, and four patients showed a noticeable increase in pulse pressure within 24 hours.

LV distension is a common event during VA-ECMO. The incidence of LV distension ranges from 12% to 68% [7-9]. LV distension results in subsequent complications of pulmonary edema and LV thrombus [10]. The functional recovery of the LV also is disturbed by LV distension [2]. Therefore, the prognosis of patients who develop such complications is worse than that of patients without LV distension during ECMO [11].

Although no guidelines or gold standard has been established for the management of LV distension during VA-ECMO, several methods have been used to improve LV distension. An intra-aortic balloon pump was used to reduce the LV afterload and improve coronary artery flow [12]. Building a venting flow from the left heart is another method. To create a drain directly from the LV, a transaortic catheter or axial flow pump can be used [13-15]. LA venting is also effective in this condition because LV distension is accompanied by considerable mitral regurgitation, sometimes with the impairment of the mitral apparatus. Interatrial septostomy [16,17] or insertion of a LA venting catheter through the interatrial septum and connecting the catheter to the venous catheter of ECMO [18,19] is performed for LA venting. However, each procedure is associated with the risk of additional bleeding, and fluoroscopy or transesophageal echocardiography guidance is required to avoid opening the chest cavity. LA venting strategies with interatrial septostomy or insertion of a large catheter in the left atrium via the interatrial septum may lead to iatrogenic atrial septal defects.

For patients receiving VA-ECMO who have LV distension, we tried to decompress the left heart with an 8-Fr Mullins sheath. The 8-Fr catheter is much smaller in size than the catheters generally inserted for LA venting in adult patients receiving VA-ECMO [18,20]. The use of smaller catheters resulted in fewer complications, with no procedure-related bleeding in our patients. The cost of the procedure was less than the other procedures, and handling of the catheter was also easy during and after the procedure in our patients. The only complication in our study was catheter obstruction due to thrombus in patient 4, who showed delayed improvement of pulmonary edema. We thought that the thrombus developed as a result of the low activated coagulation time, which was <120 seconds on that day. However, the efficacy of our procedure was proven by this case. The pulmonary edema of patient 4 was aggravated after the obstruction of the catheter.

When inserting the decompression catheter, the catheter should have a sufficient flow rate because if the flow rate is very low, the LV distension cannot be relieved properly. Although we did not measure the flow rate of the decompression catheter, we expected the flow rate to be approximately 300 ml/min, based on our experience [21]. Another article also reported that the flow rate of the 8-Fr LA venting catheter in humans was up to 270 ml/min [18]. A flow rate of approximately 300 ml/min might be enough for adult patients receiving VA-ECMO, since the flow rate to the left atrium from the bronchial artery, which is maintained when cardiac function is stopped, is 1% of the cardiac output [22]. However, the required flow rate for left heart decompression varies according to the condition of patients (e.g., remaining cardiac function and duration of VA-ECMO). We think that a flow rate of approximately 300 ml/min for left heart decompression may meet the flow requirement for left heart decompression in patients who are expecting LV functional recovery soon. Finally, our patients recovered from the LV distension and were weaned from ECMO successfully. We suggest that an 8-Fr catheter may be another option for left heart decompression when recovery of LV systolic function is expected.