Valvular heart disease (VHD) is a clinical umbrella term referring to any structural cardiovascular condition provoked by a malfunction of any of the four valves of the heart. This abnormal functioning can be caused by congenital defects, damage related to infections, inflammatory processes, and multiple or individual modifiable risk factors besides ageing.1,2 Although the initial stages of the disease usually remain asymptomatic, when mild to moderate alterations occur, VHD eventually leads to significant impairment in patients’ well-being due to dyspnoea, fatigue, arrhythmia, and palpitations related to heart failure. In addition, VHD is associated with a higher risk of morbi-mortality in comparison to matched populations without VHD, particularly, if suboptimal treatment is provided.3,4

Recent research conducted in Europe outlined a VHD incidence rate of 16.3 cases per 10000 person-years, being mitral regurgitation the most common condition (8.2%; 95% confidence interval [95%CI], 8.0–8.4), followed by aortic stenosis (7.2%; 95%CI, 7.0–7.3) and aortic regurgitation (5.0%; 95%CI, 4.9–5.2).5 This epidemiological burden has been also highlighted in the US using an indirect approach based on the International Classification of Diseases 9th Revision (ICD-9) and 10th Revision (ICD-10).6,7 It is clear that educational programmes, prevention strategies, technological advances in diagnosis along with innovations widening the opportunity to treat VHD patients,8 are crucial for reducing the clinical impact of VHD. However, considering the increasing life expectancy and the progressive ageing of our societies,9 VHD imposes a challenge to national healthcare services that have been recently affected by the severe acute respiratory syndrome coronavirus-2 pandemic.10

Previous research in Spain has shed light on aortic stenosis incidence and its management with a special focus on transcatheter aortic valve replacement or surgical valve replacement, showing an increase in the annual number of procedures in our country during the last years, while observing positive outcomes such as reduced in-hospital mortality.11,12 However, despite their clinical importance, little attention has been paid to mitral regurgitation and tricuspid regurgitation to date. Consequently, we designed this research to characterize the magnitude of the burden of disease associated with VHD (in relation to aortic, mitral, and tricuspid valves procedures) and to gain knowledge about patient characteristics, and how VHD is being managed at the National Health Service and the results being obtained. To this end, public data available from the Hospital Discharge Records in the Registry of specialized healthcare activity-minimum data set (RAE-CMBD), regulated by the Royal Decree 69/2015 of February 6 (BOE-A-2015-1235), is a useful source to recognize the casuistry related to heart valve pathology in terms of its diagnosis, the procedures performed, and the results obtained. This is the case because of the mandatory nature of the administrative registry, which is integrated into the Health Information System of the National Health Service and is required for all health centres that maintain an inpatient regimen. In addition to its high standardization, there are multiple examples of the use of this data source for the detailed analysis of the reality of healthcare in our country.11–14 This reinforces its use as a valid resource for research despite the limitations of its retrospective nature. Some of these inherent limitations in the database stem from the existence of centres and autonomous communities (AC) which, despite the mandatory character, have not provided the data to the RAE-CMBD during the years 2016–2017, or which have made non-systematic provisions.

In this publication, we describe the procedures performed at the public hospitals (including private hospitals publicly funded, as known as subsidized centres) in our country between 2016 and 2019 for aortic, mitral, and tricuspid valve disease and the outcomes observed in terms of mortality. Additionally, patients’ length of stay (LoS) is outlined as a core driver for hospital burden and costs.

The analysis has been carried out following the reference methodological guidelines SAMPL (Statistical Analyses and Methods in the Published Literature)15 and STROBE (Strengthening the Reporting of Observational Studies in Epidemiology).16

Following a formal request, the Ministry of Health of Spain authorizes the use of RAE-CMBD data for research. No authorization from the Ethics Committee is required since the administrative database is completely anonymized, and no patient medical records are reviewed. This anonymization and data processing carried out by the Ministry itself is covered in Royal Decree 69/2015, of February 6, regulating the Registry of Specialized Healthcare Activity, and complies with the requirements of Organic Law 3/2018, of December 5, on the Protection of Personal Data and Guarantee of Digital Rights, which incorporates the General Data Protection Regulation of the European Union.

Source for the analysis

Standardized public data was extracted from RAE-CMBD for the last available years at the time of the study (2016–2019). All available cases aged 18–99 years, in which a heart valve procedure had been carried out during a hospital stay in a public and subsidized centre were included in the analysis (n=57878). In accordance with the study's objective, all accessible records were incorporated to ensure a comprehensive assessment of the implications of VHD management within the healthcare system, extending beyond the influence of patients’ clinical characteristics on such management. This global VHD scope includes percutaneous and surgical procedures for aortic, mitral, and tricuspid valves, both isolated and concomitant. Procedures, whether concomitant with other coronary or aortic procedures or not, have also been included, without exclusion of cases involving endocarditis or post-myocardial infarction rupture of the mitral valve.

Procedures were grouped for in-depth analysis following the clinical criteria for the management of VHD. This group classification was made using the ICD-10 procedure codes (see “Methods” in the supplementary data). Regarding data management, in the situation of several records associated with the same case (variable “recoded CIP SNS”), the consecutive episodes were consolidated (see “Methods” in the supplementary data). Finally, all hospitals with less than 20 procedures per year were excluded to avoid incorporating into the sample centres that may potentially compromise certain aspects of the management of VHD analysis. In the specific case of La Rioja, all procedures from the autonomous community were excluded since only six procedures were conducted in that region. Fig. 1 shows the flow chart with the entire sample selection process and the loss of observations in each step. In all these final sample, days of stay, stays in the intensive care unit (ICU), exitus, and complications of all the episodes have been computed jointly.

Fig. 1.

Flowchart of the sample selection process. *Exclusion of hospitals with less than 20 interventions per year and events transferred to other hospitals without data from the second admission. Hosp, hospital.

This article provides detailed representative description of the in-hospital procedures for the management of VHD in Spain, responding to the need for information of the different agents involved in the national healthcare system to monitor the quality of healthcare provisions. While conventional approaches typically scrutinize outcomes within valvular and etiological subgroups, this publication directs its attention to the complete findings, aiming to better understand the magnitude of the VHD in term of procedures. Overall, this research showed a growing trend of the number of VHD procedures between 2016 and 2019, with a 14.9% increase during this period. This substantial increase aligns with UK VHD diagnoses projections based on the OxVALVE population cohort study3 and the cumulative number of mitral and aortic procedures recently reported by Tung et al.5 This tendency, which has been highlighted by previous research with data up to 2018,12 might be partly explained by the expansion of transcatheter approaches for aortic valve replacement with widened candidate populations.1 Also, newer transcatheter developments for mitral and tricuspid valves might explain this progression to a lesser extent.8,22–24 However, since our cohort included cases up to 2019, this factor should be further explored in the near future, once these innovations have been fully incorporated into the healthcare system and hospitals have recovered their activity since the pandemic situation. In any case, this growth in VHD procedures can be justified considering recent epidemiological research and the ageing of our population.

With respect to the complexity of cases treated, a-CCI worsened during the study period, in agreement with the severity observed in recent research.11,12 Given that the mean age has also increased over the study period, one might think that this increase in a-CCI could be attributed to the older profile of the patients. However, upon observing a rise in case complexity within each age group, a direct relationship cannot be conclusively established. Despite the complex risk profile of patients, VHD procedures appear reasonably safe with low rates of major complications (AMI, atrial fibrillation, cerebrovascular complications, and others) in all groups analyzed over the years. However, it should be noted that the permanent pacemaker implant rate increased in 2019 with respect to previous years (up to 9.27% permanent pacemaker implants in 2019). It is also possible that the low incidence of postoperative atrial fibrillation was due to the undercoding of this complication. Although these rates are consistent with recent research, further monitoring to understand this specific finding should be carried out in future research. Furthermore, an adjusted national rate of 73.2 in-hospital deaths per million inhabitants was observed. It should be noted that patients with endocarditis were not excluded. Excluding patients with endocarditis has the effect of reducing the mortality in the analyzed sample. Specifically, when endocarditis is excluded the in-hospital mortality rate was observed to decrease significantly (P-value <.001) from 5.85% to 5.25%. This reduction can be attributed to the higher mortality rate of 23.72% observed in patients with endocarditis. However, according to evidence from recent observational data with median follow-up post-surgery of up to 54 months,25 this rate of endocarditis would probably rise after discharge. Furthermore, patients who underwent procedures for ischaemic heart disease or procedures involving the ascending or descending thoracic aorta were not excluded. All these factors would contribute to a significant reduction (P-value <.001) in overall mortality to 4.99% during the entire studied period.

Regarding hospitalizations, the LoS was 15 days on mean and 64% of patients required admission to ICU. The frequency of ICU admissions exhibited an upward trend throughout the analyzed period. Furthermore, these admitted patients presented a significantly longer mean LoS, extended by approximately 3 days. Additionally, a noticeable increase of nearly 0.5 days in the mean duration of ICU stays has been observed over the years examined. Taking these findings into account, and despite a mild decrease in the LoS (−4.11%), the utilization of resources remains significant. A detailed analysis by valve affectation etiologies and procedures would provide detailed information about these trends.

Finally, the mean cost per patient dropped from €24801 in 2016 to €24015 in 2019. This finding was surprising considering that the costs increased in the study period by 11.23%. This is most probably explained by the progressive increase in the annual proportion of transcatheter interventions conducted. Another factor to consider in the impact on costs is the proportion of patients with specific DRGs. Given that mean costs in this database are contingent upon the quantity of cases categorized under each DRG, this information holds significance in comprehending the composition of the mean cost itself. However, conducting a comprehensive analysis of these factors is imperative to reliably attribute causality to the observed changes in costs. Nonetheless, the feasibility of conducting such an exhaustive analysis is called into question due to the inherent composition of the database.

This article provides a comprehensive overview of VHD procedures in Spain, highlighting the significant increase (14.9%) in procedures from 2016 to 2019. This growth aligns with UK VHD projections and may be linked to transcatheter advancements. Besides, the study reflects the rising complexity of cases and an increased rate of permanent pacemaker implants in 2019. Although a causal effect cannot be determined by this study, there is an increase in hospitalizations, especially related to ICU admissions, which leads to greater utilization of resources and costs. Fortunately, in-hospital mortality rates decreased. Finally, it is important to know that this study provides useful information from the hospital perspective, but the outlined burden of disease only captures the resources corresponding to the procedures, so the global burden of disease needs complementary studies to provide a complete analysis.

This study received financial support from Edwards Lifesciences. The company was involved in shaping the research scope and methodology. The analysis and interpretation of the results were conducted independently.

Following a formal request, the Ministry of Health of Spain authorized the use of RAE-CMBD data for research. No authorization from the Ethics Committee was required since the administrative database is completely anonymized, and no patient medical records are reviewed. This anonymization and data processing carried out by the Ministry itself is covered in Royal Decree 69/2015, of February 6, regulating the Registry of Specialized Healthcare Activity, and complies with the requirements of Organic Law 3/2018, of December 5, on the Protection of Personal Data and Guarantee of Digital Rights, which incorporates the General Data Protection Regulation of the European Union. Furthermore, this work has considered the potential variables of sex and gender in accordance with the SAGER guidelines. STROBE guidelines were followed.

No artificial intelligence tool was used to prepare this article.

All authors were involved in the initial design of the study. Analyses were performed by O. Martínez-Pérez, J. Cuervo, M. Carnero-Alcázar, and M. Álvarez-Bartolomé. All authors aided in interpreting the results and have made a significant contribution to the final manuscript.

P. González is employed by Edwards Lifesciences. O. Martínez-Pérez and J. Cuervo are members of Axentiva Solutions, which received consultancy fees from Edwards Lifesciences. I. Cruz-González and M. Carnero-Alcázar received consultancy fees from Medtronic, Edwards Lifesciences, and Abbott Laboratories. M. Álvarez-Bartolomé, A. Pardo-Sanz and J.L. Zamorano received consultancy fees from Edwards Lifesciences. M. Carnero-Alcázar received consultancy fees from Atricure and is the is the Secretary-General of SECCE. I. Cruz-González received consultancy fees from Boston Scientific and Biosensors. J.L. Zamorano received consultancy fees from Bayer, Pfizer, Daiichi, and Novartis.