HEMO-TAVI study is single-centre, observational and prospective cohort study carried out in University Hospital Reina Sofía (Córdoba Spain). Enrolment of the study participants occurred between June 2020 and September 2023. The general objective of HEMO-TAVI is to assess the impact of haemolysis in patients with AS undergoing TAVI. The aim of the present analysis is to describe the prevalence of pre-procedural haemolysis and to assess its association with peri-procedural and 30 days outcomes. The protocol was approved by the Clinical Research Ethic Committee of Córdoba (Spain) (reference code: 5049) and the study was conducted in accordance with institutional and Good Clinical Practice guidelines.

We enrolled consecutive patients with severe AS scheduled to undergo TAVI at our centre. The eligibility criteria are shown in table S1. Key exclusion criteria included: haematological disorders, other severe valvular heart disease associated with aortic stenosis, cardiac prosthesis wearer prior to TAVI implantation and aortic insufficiency without stenosis.

Pre-procedural evaluation included a detailed medical history, physical examination, electrocardiogram, laboratory tests, transthoracic echocardiography, and computed tomography angiography (CTA). Transthoracic echocardiography variables of interest were the size of the interventricular septum, the left ventricular ejection fraction, the estimated systolic pulmonary artery pressure using tricuspid regurgitation, and the presence of aortic regurgitation or mitral regurgitation. The CTA variables of interest were aortic annulus diameter, perimeter and area; sinus-sinus, sinotubular joint and ascending aorta diameters; left and right coronary artery heights and right and left femoral artery diameters.

Pre-procedural haemolysis was defined as LDH>220U/L and the presence of ≥2 of the following criteria: haemoglobin<13.8g/dL (male) or <12.4g/dL (female), haptoglobin<50mg/dL, reticulocyte fraction ≥2%, schistocytes in peripheral blood present. The definition was adapted from the Skoularigis et al. criteria for haemolysis in mechanical heart valve prostheses, using the LDH threshold proposed by Horstkotte.9,10

Peri-procedural and 30-day events of interest included all-cause mortality, acute myocardial infarction, cerebrovascular accident, cardiac tamponade, systemic embolism, vascular complications, severe aortic regurgitation, major bleeding, renal complications, pacemaker implantation and heart failure hospitalization. A composite endpoint was defined as the occurrence of any of these events.

Categorical variables were expressed as counts (percentages), and continuous variables as median and interquartile range. For between-group comparisons, the chi-square test or the Fischer test were used to compare categorical variables and the Wilcoxon rank sum test for continuous variables. Univariable and multivariable (adjusting for age and sex) logistic regression models were fitted to evaluate the association between IH and the combined endpoint. Additionally, univariable and multivariable (adjusting for age and sex) ordinal logistic regression models were used to evaluate the association between hemolysis and the number of events. Statistical analyses were performed using R software (version 4.4.2; R Foundation for Statistical Computing, Austria).

For the present study, 301 patients with severe AS scheduled to undergo TAVI were analyzed. The prevalence of pre-procedural haemolysis was 6%. The median age was 79.9 years (IQR: 75.8–84.5), and 49.2% of the patients were women. The clinical characteristics of the study population are summarized in Table 1. Regarding cardiovascular risk factors, arterial hypertension was present in 86.2% of patients, diabetes mellitus in 38.2%, dyslipidaemia in 63.4%, and obesity in 41.9%. Regarding other comorbidities, 7.6% of patients had chronic obstructive pulmonary disease, 26.2% had atrial fibrillation, and 23.6% had coronary artery disease, of which 22.3% had previously undergone percutaneous revascularization and 0.3% surgical revascularization. Additionally, 2.3% had peripheral arterial disease, and 55.1% had chronic kidney disease.

With respect to functional class according to the NYHA scale, 7.0% of patients were classified as NYHA I, 57.5% as NYHA II, 31.2% as NYHA III, and 4.3% as NYHA IV. Surgical risk was categorized as low in 1.0% of patients, intermediate in 15.6%, high in 76.4%, and prohibitive in 6.3%. The median haemoglobin value was 11.4 (9.7–12.0) g/dL in the Haemolysis group and 11.7 (10.5, 12.8)g/dL in the No Haemolysis group (p=0.162). Regarding treatment, 87.4% were taking aspirin, 22.6% direct oral anticoagulant, 3.6% vitamin K antagonist, 41.9% beta blockers, 72.7% renin–angiotensin system inhibitors, 7.14% statin, 63.2% diuretic and 30.2% iron. No significant differences were observed in baseline characteristics between the groups with and without haemolysis.

CTA and echocardiogram values in the overall population and according to haemolysis groups are shown in Table 2. Haemolysis was not associated with any of the CTA variables. With respect to echocardiographic data, interventricular septum size and mean gradient were associated with haemolysis. More specifically, we found a larger interventricular septum size [13(12–15)mm vs. 16(13–17)mm, p 0.012] and a higher mean gradient [47(42–58)mmHg vs. 55(49–65), p 0.043] in patients with haemolysis. We found no difference in the other echocardiographic variables, including left ventricular ejection fraction, aortic valve area and aortic or mitral regurgitation.

Analytical values in the overall population and according to haemolysis groups are shown in Table 3. We only found a higher LDH in the haemolysis group (189 vs. 261, p<0.001), with no differences in haemoglobin, mean corpuscular volume, red cell distribution width, creatinine, estimated glomerular filtration rate, bilirubin, ferritin or liver enzymes.

Overall, 68 (22.6%) patients experienced at least one adverse event of interest during the hospitalization or first 30-days after the procedure: 8 (44.4%) in the Haemolysis group and 60 (21.2%) in the No Haemolysis group (OR 2.97, CI 95% 1.09–7.87, p=0.028). After adjusting by age and sex, haemolysis continued to be associated with a higher risk for the combined endpoint (OR 2.23, CI 95% 1.12–8.25, p=0.025). The most frequent event was pacemaker implantation (14.3%). The distribution of the number of events according to the presence of haemolysis is shown in Fig. 1. The odds for having a higher number of complications were higher in patients with haemolysis (OR 3.11, CI 95% 1.15–8.01, p 0.020]. The association remained significant after adjusting by age and sex (ORadj 3.26 (CI 95% 1.19–8.47), p 0.016). When comparing each individual event between the group with and without haemolysis, we found a higher incidence of cardiac tamponade (1.1% vs 11.1%, p=0.030), without statistically significant differences in the rest of complications such as all-cause mortality, pacemaker implantation or heart failure hospitalization (Table 4).

Fig. 1.

Association between the number of periprocedural complications and hemolysis. Preprocedural hemolysis was associated with increased odds of experiencing a greater number of complications. OR: odds ratio. CI: confidence interval.