Patient data were acquired by searching for electronic discharge notes coded with a diagnosis of valve disease according to the International Classification of Diseases, Ninth Revision (ICD-9) of all patients admitted to the cardiology ward in the Cardiology Department of Santa Maria University Hospital between January 1, 2014 and October 3, 2015 (see Appendix 1 for details).

The search identified 391 patients, whose discharge notes were then manually reviewed. Only patients with significant VHD and those with prior heart valve interventions were selected. Accordingly, only patients who underwent echocardiographic assessment during the hospitalization under study (index hospitalization) were included.

Significant VHD was defined on the basis of clinical plus imaging criteria (mainly echocardiography),15,16 and included moderate to severe valvular stenosis and/or regurgitation of any cardiac valve (or multiple valves) that was responsible in any way for the index hospitalization. Left ventricular ejection fraction (LVEF) was considered to be reduced if <50%.17

In accordance with the above criteria, 66 patients were excluded for presenting only mild valvular disease; additionally, 38 patients were also excluded because of insufficient information available in the system (patients admitted electively to undergo selective examinations only or patients with hospital stay less than 24 hours). The final study population included 287 patients.

Discharge notes of these patients were carefully manually reviewed. In addition to characterization of VHD (valves involved, disease severity and etiology) on the basis of all the available information, the following data were collected and analyzed: demographics (gender, age, ethnicity, place of birth), admission and discharge dates, type of admission to the cardiology ward (elective or urgent), main reason for hospital admission, length of hospital stay (LOS), major cardiovascular risk factors (systemic hypertension, diabetes, dyslipidemia, smoking), presence and type of comorbidities, history of chronic HF and New York Heart Association (NYHA) functional class, presence of advanced atrioventricular (AV) block, bundle branch block (BBB) or atrial fibrillation (AF) (chronic or new-onset), presence of pulmonary hypertension (assessed invasively or by echocardiography), coronary artery disease (CAD) (assessed by coronary angiography), coronary artery bypass grafting (CABG) during the index admission, implanted cardiac devices (during the index hospitalization or in the past), valvular intervention (during the index hospitalization or in the past), hospital admissions due to VHD in the 12 months preceding the index hospitalization, mortality (and cause of death) during the index hospitalization, and destination after discharge. Various echocardiographic parameters and indices (assessed during the index hospitalization before any new valvular intervention) were also collected, including cardiac chamber size and measures of systolic LV and right ventricular (RV) function.

In addition to the descriptive data, we looked for associations between the type and etiology of VHD and demographic, clinical, electrocardiographic and echocardiographic features, in order to improve characterization of the profiles of hospitalized patients with VHD.

The statistical analysis was performed with Microsoft Excel 2013® and IBM SPSS® version 22.0. A descriptive analysis of the different variables was performed and possible differences between genders were assessed. Quantitative variables are expressed as means ± standard deviation, and qualitative variables as absolute or relative frequencies (percentages). The Student's parametric t test for independent variables was used to compare quantitative variables as a function of the presence and type of VHD (mitral, aortic, mitral-aortic or tricuspid valve disease). Categorical variables were analyzed using the chi-square test and Fisher's exact test, as appropriate. Possible associations between type of VHD and demographic, clinical, electrocardiographic and echocardiographic features were analyzed using Pearson's correlation test.

A p-value of <0.05 for a confidence interval of 95% was considered statistically significant.

The study population included 142 females (49.5%) and 145 males (50.5%), mean age 74.86±13.39 years (range 23 to 93), with no difference between genders. Most patients were born in Portugal and 11% were originally from Portuguese-speaking African countries.

The majority of patients were admitted through the emergency room or by transfer from other departments or hospitals (66.9%), while 33.1% were elective admissions. Fifty-six patients (19.51%) were hospitalized for causes not directly related to VHD, primarily with acute coronary syndrome (ACS) (71.4% of these cases and 16.4% of all patients with VHD), and valvular disease was a secondary diagnosis.

Regarding all patients with a diagnosis of VHD, decompensated HF was the most frequent reason for admission (84 patients, 29.3%). This proportion was even higher in the group of 231 patients admitted for VHD (35.5%). Eighty-four (36.4%) were admitted electively for valve intervention, 70 (24.4%) because of a dysrhythmia (AF in 9.1%), 15 (5.2%) in the context of a syncopal episode (four of which were due to complete AV block), and seven (2.4%) due to infective endocarditis.

Forty patients (13.9%) had been hospitalized for reasons related to VHD in the 12 months preceding the index hospitalization.

Etiologies and types of valvular disease are shown in Figure 1. Most patients (67.9%) had degenerative aortic and/or mitral valve disease, followed in frequency (15.3%) by significant functional mitral and/or tricuspid regurgitation (TR) (secondary to left and/or right ventricular dilatation due to ischemic heart disease or dilated cardiomyopathy). Rheumatic etiology was diagnosed in 8.7% of patients. Other etiologies included infectious endocarditis (3.8%), mitral valve prolapse (2.8%), and congenital VHD (0.7%).

Figure 1.

Etiologies of valvular heart disease. IE: infectious endocarditis; MR: mitral regurgitation; MVP: mitral valve prolapse; TR: tricuspid regurgitation; VHD: valvular heart disease.

(0.07MB).

Most patients with rheumatic valve disease were women (84%, p<0.001), with no significant gender differences observed in other etiologies. Age correlated directly with degenerative etiology (r=0.473; p<0.001), and inversely with rheumatic disease (r=-0.36; p<0.001).

Aortic valve disease was documented in most cases (181 patients, 63.1%), of whom 89% (56% of all patients) had AS, and 27.1% (17.1% of all patients) had aortic valve regurgitation. Mitral disease was diagnosed in 146 patients (50.9%), mostly MR (90.4%, 46% of all patients); mitral stenosis was observed in 15% of these patients (7.7% of all patients). Concomitant aortic and mitral valve disease was present in 22.3% of patients.

There were no documented cases of tricuspid valve stenosis, but significant TR was diagnosed in 100 patients (34.8%), and was associated with mitral valve disease in 69 patients (69.7% of patients with TR, 24% of all patients). Isolated moderate to severe TR was observed in 20 cases (6.97%).

Moderate pulmonary valve regurgitation was infrequent (13 patients), and was diagnosed in the context of predominant mitral valve disease and TR; it did not require targeted therapeutic intervention in any patient.

A significant proportion of patients admitted had chronic HF (39.4%), and 47.4% had chronic AF. Systemic hypertension was the most frequent cardiovascular risk factor observed, present in 219 (76.3%) patients, while 177 (61.7%) patients had more than one of the classical cardiovascular risk factors. Multiple comorbidities were found in 53% of patients. The most common was CAD (n=81; 28.21%), followed by chronic kidney disease (n=60; 20.9%), chronic anemia (n=76; 26.5%), chronic pulmonary disease (n=43; 15%) and malignancies (n=39; 13.6%).

Coronary angiography was performed in 105 (36.6%) of patients, mostly in the context of ACS, but also due to a previous history of CAD or as a standard procedure before valve intervention. Significant CAD was documented in 55 (52.4%) of cases. Coronary revascularization was performed in 32 (11.1%) patients, by a percutaneous approach in 28 and by CABG in four.

AF was diagnosed in 136 (47.4%) patients, and was significantly associated with aortic valve disease (p=0.014), mitral valve disease (p<0.001) or TR (p<0.001). Advanced AV block was also frequent, with 29 patients (10.1%) having complete AV block at presentation (16 patients with AS and 13 with MR; AS and MR coexisted in five patients). Complete AV block was the primary reason for pacemaker implantation in the index hospitalization. However, it was not a frequently documented cause of syncope on admission. Other causes of syncope at presentation were sinus node dysfunction and ventricular tachycardia.

Reduced (<50%) LVEF was documented in 33.4% of patients and was associated with mitral disease (p=0.003). In 12.2% of patients RV longitudinal systolic function assessed by tricuspid annular plane systolic excursion (TAPSE) was also compromised, coexisting with LV systolic dysfunction in 22 (62.9%) patients.

LV dilatation (24%) and LV hypertrophy (54.7%) were both associated with aortic valve disease, but only the former with mitral disease. The left atrium was dilated in 79.4% of patients, but there was no significant association with aortic or mitral valve disease. RV dilatation was present in 16.4% of cases and right atrial dilatation in 47%; both conditions were associated with the presence of severe TR.

Forty-eight (16.7%) patients had a cardiac device in place prior to the index admission: a pacemaker (PM) in 30 (10.5%), an implanted cardioverter-defibrillator (ICD) in 10 (3.5%), and a cardiac resynchronization therapy (CRT) device in eight (2.7%) patients. Severe chronic TR was documented in 62.5% of these patients, and was associated with mitral valve disease in 21 (43.8%; significant chronic MR in 50% of these) and with aortic valve disease in 10 patients (20.8%; moderate or severe AS in 41.7%).

In addition, in 55 patients (19.2%) a device was implanted for the first time during the index hospitalization: a PM in 51 (17.8%) patients, an ICD in two (0.7%), and a CRT pacemaker in another two.

Disease of the aortic valve showed no gender predominance, but correlated directly with age (p<0.001), degenerative etiology (p<0.001) and presence of HF (p=0.004), AF (p=0.014), LV dilatation (p=0.003), and LV hypertrophy (p<0.001). There was no significant association between aortic valve disease and rheumatic etiology, reduced LVEF, left atrial (LA) dilatation, AV block or bundle branch block (p=NS for all).

Mitral disease was more frequent in women than in men (58.9% vs. 41.1%, p=0.002), and was associated directly with rheumatic etiology (p=0.011), ischemic heart disease (p<0.001), HF (p=0.001), AF (p<0.001), LV dilatation (p<0.001) and reduced LVEF (p=0.003). No significant association was found between mitral valve disease and LA dilatation.

The presence of moderate-to-severe TR did not differ between genders (p=0.702), and correlated directly with age (p<0.001) and with the presence of previously implanted cardiac devices (p<0.001). The latter association was observed only when TR coexisted with aortic or mitral disease (of any etiology). A significant association was also found between TR and mitral disease (p<0.001), right chamber dilatation (p<0.001), AF (p<0.001) and pulmonary hypertension (p<0.001), but not with chronic pulmonary disease.

Thirty-one patients (10.8%) had previously undergone valve intervention (repair or replacement), most frequently SAVR (58.1%), followed by mitral valve repair (38.7%) and TAVI (12.9%). No patient had been previously submitted to percutaneous mitral valve intervention.

During the index hospitalization, 118 patients (41.1%) underwent valve intervention. Percutaneous procedures were the most frequently performed (74.6%), mainly TAVI (65.3%); surgical replacement was performed in 12.7% of patients (aortic valve in 12 patients, mitral valve in four), and surgical repair in 5.9% of cases (mitral valve in four patients and aortic valve in three). Percutaneous mitral valve interventions were performed in 11 (9.3%) patients.

Regarding the 31 patients who had previously undergone valve intervention, most were admitted with decompensated HF (48.4%), 16.1% were admitted electively for another valve procedure, and 6.5% presented with infectious endocarditis. Furthermore, 35.5% of this population underwent valve intervention again, 25.8% on the same valve. In-hospital mortality was 12.9% in this high-risk population.

Of note, 4 patients underwent CABG during the index hospitalization, and 2 also underwent SAVR.

Overall, mean LOS was 11.98±14.33 days. Most patients were discharged home from hospital (84.3%) and 5.2% were transferred to another hospital (in their area of residence) for continued care. Overall in-hospital mortality was 9.8% and cardiovascular mortality was 8.7%. Most deaths (33.3%) occurred in the context of ACS, followed by advanced refractory HF and cardiogenic shock due to VHD (29.6%), and to severe refractory pulmonary hypertension (related to valve disease and/or pulmonary embolism) in 7.4%. Other causes of death were infectious endocarditis of prosthetic valves (7.4%) or devices (3.7%), acute complications after prosthetic valve implantation (surgical or percutaneous) (7.4%), and pericardial effusion (3.7%).