National Trends in Utilization and In-Hospital Outcomes of Surgical Aortic Valve Replacements in Spain, 2001-2015

Objective The aims of this study were to examine the incidence and in-hospital outcomes of surgical aortic valve replacement (SAVR) and to identify factors associated with in-hospital mortality (IHM) among patients according to the type of implanted valve used in SAVR. Methods We performed a retrospective study using the Spanish National Hospital Discharge Database, 2001-2015. We included patients who had SAVR listed as a procedure in their discharge report. Results We identified 86,578 patients who underwent SAVR (52.78% mechanical and 47.22% bioprosthetic). Incidence of SAVR coding increased significantly from 11.95 cases per 100,000 inhabitants in 2001 to 17.92 in 2015 (P<0.001). Age and comorbidities increased over time (P<0.001). There was a significant increase in the frequency of concomitant coronary artery bypass grafting (CABG) and in the use of pacemaker implantation. The use of mechanical SAVR decreased and the use of bioprosthetic valves increased over time. IHM decreased over time (from 8.13% in 2001-05 to 5.39% in 2011-15). Patients who underwent mechanical SAVR had higher IHM than those who underwent bioprosthetic SAVR (7.44% vs. 6%; P<0.05). Higher IHM rates were associated with advanced age, female sex, comorbidities, concomitant CABG, and the use of mechanical SAVR (OR 1.67; 95% CI 1.57-1.77). Conclusion The number of SAVRs performed in Spain has increased since 2001. The use of mechanical SAVR has decreased and the use of bioprosthetic valves has increased over time. IHM has decreased over time for both types of valves and despite a concomitant increase in age and comorbidities of patients during the same period.


INTRODUCTION
Aortic valve replacement is the treatment of choice in severe symptomatic aortic valve disease. Surgical aortic valve replacement (SAVR) reduces mortality, provides symptom relief and increases quality of life at subsequent follow-up [1] . Furthermore, continuous improvements in SAVR techniques and new technologies have recently been developed to facilitate the procedure and reduce operative times [2] .
The range of available prostheses changed significantly during the last decades in favor of biological valve versus mechanical valve replacements [3] . A study using the National Inpatient Sample (NIS) found an increase in the use of bioprosthetic valves from 37.7% in 1998-2001 to 63.6% in 2007-2011 [4] . Biological valves are increasingly implanted in younger patients due to a higher durability enabled by improved anticalcification treatment and the adverse events associated with mechanical prostheses [4,5] .
Previous studies have provided insight into changes in patient demographics, risk factors and outcomes of SAVR [2] . Therefore, it is necessary to study changes in practice over time and to establish outcomes after SAVR to help inform decision making for high-risk patients [3] . However, valve surgery trends and in-hospital outcomes nationwide are less often reported [6] .
Using the Spanish National Hospital Discharge Database (SNHDD), this study aimed to: i) examine trends in the incidence, characteristics and in-hospital outcomes of SAVR hospitalizations from 2001 to 2015; ii) compare clinical variables among patients according to the type of implanted valves in the discharge report; and iii) identify factors associated with in-hospital mortality (IHM) among patients according to the type of implanted valves.

METHODS
This retrospective observational study was performed using the SNHDD. Details of the SNHDD design and description are available online [7] .

Trends in Mechanical SAVR Hospitalizations
Incidence rates of hospital admissions with a mechanical SAVR-coded procedure decreased significantly from 8.94 cases per 100,000 inhabitants in 2001 to 7.23 in 2015 (Supplementary Figure 1B). The mean age increased slightly, but significantly, from 65.05 years to 65.5 years, and the percentage of female patients showed a significant decrease over the study period (37.84% in 2001-05 vs. 36.09% in 2011-15) ( Table 2).
Comorbidity increased significantly over the study period. The prevalence of chronic conditions and use of procedures among patients who underwent mechanical SAVR agreed with the results found in the total sample, except for the prevalence of concomitant CABG, which showed a significant decrease (18.91% to 17.73%; P<0.001).
The crude IHM decreased significantly from 8.45% to 5.9% over the study period.

Trends in Bioprosthetic SAVR Hospitalizations
Supplementary Figure 1C shows a significant and constant increase in the hospitalization rates over time (from 3.01 cases per 100,000 inhabitants in 2001 to 10.7 cases in 2015) for patients who underwent bioprosthetic SAVR.
The evolution over time of the demographic and clinical variables analyzed was similar to those described for the total To identify variables associated with IHM as a binary outcome among patients who underwent SAVR, we performed three logistic regression analyses, one for each type of SAVR. The variables included in the multivariable models were those with significant results in the bivariable analysis and those considered relevant in other investigations. Estimates were the odds ratio (OR) with a 95% confidence interval (95% CI).
All statistical analysis was performed with Stata version 10.1 (Stata, College Station, Texas, USA). Statistical significance was set at P<0.05 (two-tailed).

Trends in SAVR Hospitalizations
We found that the incidence of SAVR coding increased from 11.95 cases per 100,000 inhabitants in 2001 to 17.  Table 1). We detected a significant increase in comorbidity according to the CCI over time. The most common associated comorbidities for hospitalized patients who underwent SAVR were coronary artery disease (32.56%), atrial fibrillation (32.31%) and diabetes (22.9%). The frequency of all conditions analyzed increased over time (P<0.001), with the exception of chronic obstructive pulmonary disease (COPD), which showed a slight decrease over the study period (9.1% to 8.88%; P=0.011) ( Table 1).
There was a significant increase in the frequency of concomitant coronary artery bypass grafting (CABG) (from     Factors associated with a lower risk of death for both types of SAVR included type 2 diabetes mellitus (T2DM) and atrial fibrillation. Patients who underwent mechanical SAVR and received a pacemaker had lower risk of death (OR 0.7; 95% CI 0.56-0.86).
Finally, after adjusting for study variables, mechanical SAVR was associated with a significantly higher IHM (OR 1.67; 95% CI 1.57-1.77) than bioprosthetic SAVR in our study.

DISCUSSION
In the past 15 years, there has been a substantial increase in the number of heart valve surgeries in Spain. This increase is in line with the trend observed in other European countries and in the United States [3,6,10,11] . A retrospective study found that the number of procedures increased from 47.5 to 88.9 per 100,000 Medicare beneficiaries between 2009 and 2015 [11] . A large study based on national registry data in the Netherlands showed more than twice the number of SAVRs were reported in 1995 compared with 2010 [10] . The authors concluded that this trend can be partly attributed to an increased prevalence of valvular heart disease sample. The IHM decreased significantly from 7.45% to 5.01% (P<0.001) ( Table 3).

Factors Associated with IHM (Table 4)
The time trend analysis showed a significant decrease in IHM from 2007-09 to 2013-2015 in patients who underwent mechanical SAVR and from 2010-12 to the last period in patients who underwent bioprosthetic SAVR.
Regardless of the type of valve implanted, factors that increased IHM included advanced age, female sex, renal disease, cerebrovascular disease, congestive heart failure and coronary artery disease. Patients with peripheral vascular disease who underwent mechanical SAVR had 1.23 times the risk of dying compared to patients without this comorbidity (OR 1.23; 95% CI 1.12-1.36). The presence of pulmonary hypertension increased the risk of death in patients who underwent bioprosthetic SAVR (OR 1.26; 95% CI 1.08-1.47).
In both types of SAVR, concomitant CABG and emergency room admission were factors associated with a higher risk of death (Table 4). As expected, a substantial reduction in the rate of mechanical valves implanted was seen, and we found an increase in the use of bioprosthetic valves. A study using NIS found an increase in bioprosthetic valve use from 37.7% in 1998-2001 to 63.6% in 2007-2011 [4] . This trend has also been reported in other registries [3,6,10,16] and suggested improved durability of biological prostheses, fewer neurological and functional complications and avoidance of permanent anticoagulation [12,17] . Lastly, technological advances such as the valve-in-valve transcatheter procedure provided new alternatives for reoperations in biological prostheses [10] .
As found in other studies, patients with coronary artery disease, atrial fibrillation and T2DM are more likely to receive bioprosthetic than mechanical valves [4] . Age plays a major role in bioprosthetic valve selection for patients with and without comorbidities, specifically in coronary artery disease patients and partly to an increased proportion of diseased patients that are diagnosed. Because the mean age of the patients has risen, both are likely to have played an important role [3,10] .
Previous studies have found that after the introduction of transcatheter aortic valve replacement (TAVR), a slight decline in the number of SAVRs was observed over time [11][12][13] . Silaschi et al. [12] indicated that this may be explained by a substantial number of patients being eligible for both types of procedures and the consequence that this leads to a reduction in surgically treated patients. However, in a single high-volume Canadian center, the surgical volume of isolated SAVR increased since the introduction of TAVR [14] . In our study, we cannot analyze the effect of TAVR over the trends of SAVR because, although TAVR was introduced in Europe in 2008, adoption trends in Spain seemed to have a plateau between 2008 and 2012 in comparison with other European countries [15] . following SAVR was associated with lower IHM [25] . However, Greason et al. [24] found a significant association of early pacemaker implantation with death [HR 1.49 (95% CI 1.20-1.84)].
We have found longer LOHS in our country compared to previous studies [6,16,19] . We believe that differences in patient baseline characteristics (comorbidities and age) and in the use of concomitant procedures, such as CABG, could partly explain this result. It is also remarkable that SNHDD only provides the date of admission and the date of discharge, so the LOHS in our investigation includes pre and postoperative time. This fact may contribute to our longer LOHS, when compared to other investigations, which only analyzed postoperative time [16] .
There are some points that should be taken into consideration when interpreting the results of the present study. Our data source was SNHDD, an administrative database that contains discharge data for hospitalizations in Spain and uses the information the physician has included in the discharge report [7] . Coding practices, as well as coding errors, may differ between individual physicians and institutions. Thus, our results are subject to several potential biases, including differences in the capture of adverse outcomes across hospitals along the study period.
The SNHDD only provides the patient's vital status at discharge, but not the cause of death, so this relevant information is not included in our study. Furthermore, for confidentiality reasons, hospital names and characteristics are not included in the SNHDD, so this information could not be analyzed. Not all hospitals may have the same expected result, and the more specialized centers may have better outcomes, as described earlier [26] . However, we think this would not affect our main results because we included all Spanish hospitals, beside their characteristics, over the 15-year period.
Finally, there is a lack of information on previous cardiac surgery (PCS) that may be a risk factor for operative mortality in patients undergoing aortic valve replacement. However, previous studies have found that, after propensity matching of PCS patients with first-time surgery, previous surgery was not a predictor of operative mortality nor long-term survival in patients undergoing isolated aortic valve replacement [27] .

CONCLUSION
The results of this study provide a comprehensive overview of valve surgery trends and outcomes in Spain. The number of SAVRs performed in Spain has increased since 2001. The use of mechanical SAVR has decreased and the use of bioprosthetic valves has increased over time. IHM has decreased over time, despite a concomitant increase in age and comorbidity of patients over the same period. Higher IHM rates in patients were associated with increased age, female sex, presence of comorbidities, and concomitant CABG. Remarkably, patients who underwent mechanical SAVR had higher IHM than those who underwent bioprosthetic SAVR. requiring revascularization [4] . We observed that valve choice is influenced by age, with most patients aged 70 to 79 years receiving bioprosthetic valves, and patients with a lower mean age of 65.27 years receiving mechanical valves.
We found that IHM of all types of SAVR has significantly decreased over the last 15 years. Siregar et al. [10] found that IHM for SAVR with or without CABG decreased significantly from 3.5% in 2007 to 2.4% in 2010. A similar trend was found for operative mortality in most other studies, which could reflect a combination of improved overall health care, healthier aging and gradual improvements in cardiac surgery over time [3,16] .
In our study, aortic valve replacement in patients with a bioprosthetic valve was associated with lower IHM than in patients with a mechanical valve, which is consistent with observational evidence. In the general population, bioprosthetic valves are associated with lower IHM compared with mechanical prostheses, which come at the cost of slightly higher rates of inhospital complications [4] . Du et al. [5] found that the risk of death at the time of surgery was 60% higher for mechanical-valve recipients compared with bioprosthetic valve recipients. Isaacs et al. [4] found higher IHM among patients who received mechanical valves (5.2%), compared with bioprosthetic valves (4.4%).
As we expected, we found an increased risk of IHM with increasing age. Using the NIS database, Agarwall et al. [18] reported that patients with advanced age and high surgical risk have increased IHM and the incidence of adverse neurologic events after SAVR.
Female sex is a factor associated with IHM in patients who underwent SAVR. In agreement with this finding, a study using the NIS data found that IHM was significantly higher in women than in men (5.6% vs. 4%) [19] . This could be related to the onset mechanism for cardiovascular disease, a delay in the presentation of valve problems and/or a later referral of women to cardiothoracic surgery [20] .
Chronic kidney disease, peripheral vascular disease, cerebrovascular disease and congestive heart failure are risk factors of IHM. However, the presence of T2DM and atrial fibrillation are associated with a lower risk of death for both types of SAVR. Halkos et al. [21] found that diabetes was not associated with IHM (OR 0.86; 95% CI 0.49-1.50) after SAVR. The lower IHM in T2DM patients undergoing SAVR compared to nondiabetic patients might be multifactorial. Obesity is more prevalent in T2DM patients undergoing SAVR and this may have contributed to the decrease of IHM, as previously mentioned [22] .
We found a significantly higher IHM in patients who underwent bioprosthetic or mechanical SAVR performed with CABG than in those patients who did not undergo this procedure. Analysis of the German Aortic Valve Registry study showed that IHM was 1.7% for the SAVR group and 3.3% for the SAVR+CABG group, and this outcome measure remained constant between 2011 and 2015 in both groups [6] .
Several previous studies have shown increased preoperative risk/comorbidities and rate of postoperative complications in patients requiring permanent pacemaker after SAVR [23,24] . We found that pacemaker implantation reduces the risk of IHM after mechanical SAVR. In agreement with this finding, a study using NIS data found that permanent pacemaker implantation Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published