Background: Cardiogenic shock (CS) is the most severe form of acute heart failure with ongoing high morbidity and mortality. Percutaneous left ventricular assist devices (pLVAD), like the Impella^® system, actively unload the left ventricle, and provide a sufficient oxygen supply to all tissues, which might improve outcome. In the current study, we compared gender-specific differences in mortality and acute renal failure in CS patients treated with a micro-axial pLVAD.

Methods: Data were collected from Dresden Impella Registry, an ongoing registry including more than 720 patients since 2014. Among, a total of 319 male and 113 female patients received a micro-axial pLVAD due to refractory CS. A propensity score matched analysis using a logistic regression model based on 16 variables was used for adjustment of baseline characteristics. The matching was performed 1:2 using nearest neighbor method without replacement. Primary endpoint was the composite of all-cause mortality and requirement of renal replacement therapy (RRT) at 30 days. Secondary, each endpoint was analyzed alone.

Results: After the propensity score matched analysis, two cohorts of 90 male and 61 female patients resulted, which showed a well balancing of baseline characteristics (Figure 1). At time of admission, male patients were 68.4 ± 1.3 years old and had a body surface area (estimated by Mosteller formula) of 2.02 ± 0.01 m². Female patients were 69.1 ±1.5 years old (p=0.785) and BSA was calculated with 1.93 ± 0.02 m² (p=0.466). A cardiopulmonary resuscitation before pLVAD was performed in 47.8% of males and 45.9% of females (p=0.821), respectively. The comparison of cardiovascular risk factors, hemodynamic parameters, and characteristics of coronary artery disease showed no differences. The duration of left ventricular unloading was 49.7±6.0 h among male patients and 43.1±8.0 h among female patients (p=0.318).

No patient was lost at 30-day follow-up. There was no difference in the primary composite endpoint between male and female patients. 81.1% (n=73) of male and 68.9% (n=42) of female patients died or needed RRT (p=0.056, Figure 2A). All-cause mortality did not differ between both cohorts (♂ 74.4%, n=67 vs. ♀ 65.6%, n=40; p=0.284). However, RRT was required less frequently in female patients (♂ 49.1%, n=36 vs. ♀ 23.3%, n=12; p=0.007; Figure 2B). The duration of RRT showed no difference (♂ 98.3 ± 23.4 h vs. ♀ 123.4 ± 73.2 h, p=0.898). Furthermore, estimated glomerular filtration rate before pLVAD, number of treated coronary lesions, and volume of radiocontrast agent did not differ between males and females.

Hemodynamic parameters, such as mean arterial pressure, norepinephrine dosage, and serum lactate did not differ in clinical course between both cohorts. No differences were observed in length of hospital stay, or length of stay at intensive care unit. The occurrence of bleeding was comparable between males and females (severe TIMI bleeding ♂ 20.0%, n=18 vs. ♀ 18.0%, n=11; p=0.763).

Conclusion: In a well-balanced propensity score matched analysis, RRT was less frequently required in female patients, who received a micro-axial pLVAD during CS. All-cause mortality at 30 days showed no gender-specific difference. Further studies are needed to investigate whether less frequent RRT might improve outcome in larger cohorts.