Primary mitral regurgitation (MR) covers all etiologies in which intrinsic lesions affect one or several components of the mitral valve apparatus.1 Surgery is recommended in patients with chronic severe MR if they have any symptoms or in asymptomatic patients with left ventricular (LV) dysfunction.1 When guideline indications for surgery are reached, early surgery is associated with better outcomes, since the development of even mild symptoms by the time of surgery is associated with deleterious changes in cardiac function after surgery.2,3 However, difficulties in detecting early LV dysfunction, accurately assessing the severity of valve involvement, or recognizing early cardiac symptoms often make it difficult to determine the optimal timing of mitral valve surgery.4 In many patients, the development of symptoms is clear, but in others, symptoms are difficult to assess because of inactivity. In some patients, it may also be unclear whether symptoms are related to severe MR or to comorbidities.5,6

Increased levels of high-sensitivity C-reactive protein (hs-CRP) are associated with poorer functional capacity and adverse outcome in patients with LV systolic dysfunction.7 Adelaide et al. reported that CRP elevation is also associated with the presence and severity of MR and with diastolic dysfunction in acute myocardial infarction. This suggests that inflammation is related to ventricular remodeling processes, independently of LV systolic function.8 As with hs-CRP, recent findings suggest that there is also a relationship between chronic heart failure (CHF) and uric acid (UA) levels.9,10 Elevated serum UA levels can predict poor survival in severe and mild–moderate CHF9,10 and may also predict exercise intolerance and inflammatory activation.11 Nozari et al. showed that elevated UA and hs-CRP are associated with heart failure (HF) among patients with myocardial infarction (MI) and their serum levels could be a predictor for the occurrence of severe HF in patients with MI.12 In addition, Yazicioglu et al. have demonstrated that elevated serum UA in patients with dilated HF is related to severity of functional MR.13

Although the association between hyperuricemia and hs-CRP with chronic heart failure has been recognized, the role of these biomarkers in MR patients is less clear. We therefore prospectively investigated the association between hs-CRP, UA and organic MR. We also assessed whether hs-CRP and UA levels are correlated with symptoms and severity of MR, LV remodeling and outcomes at one-year follow-up.

Of the 200 patients with moderate, moderate to severe and severe MR included in the study, 54 were asymptomatic while 146 were symptomatic. Baseline demographic, clinical characteristics and laboratory parameters of patients with MR are listed and compared between asymptomatic and symptomatic patients in Table 1. Mean age was significantly higher in symptomatic than in asymptomatic MR patients (p<0.05). Male gender and hypertension were more prevalent in the symptomatic group than the asymptomatic group (p<0.05 for both). Treatment with angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin receptor blocker (ARBs) and diuretics was also more prevalent in symptomatic MR patients (all p<0.05).

Patients in the symptomatic group had lower LVEF, higher PAP, increased LVEDD and LVESD and larger LA diameter compared to the asymptomatic group (all p<0.05). Moderate to severe and severe mitral regurgitation was significantly more prevalent in the symptomatic group (p<0.001). Hemoglobin, total cholesterol, and low-density lipoprotein (LDL) cholesterol levels were significantly lower, while fasting glucose was higher, in the symptomatic group (all p<0.05).

hs-CRP was significantly elevated among symptomatic MR patients compared to asymptomatic patients (5.9±6.8 vs. 2.56±1.9 mg/l; p=0.004). Median hs-CRP levels were increased significantly with increase in NYHA class (NYHA class I: 1.50 mg/l [0.90–3.50], class II: 2.73 mg/l [2.59–6.57], class III: 5.20 mg/l [3.30–9.30], and class IV: 4.07 mg/l [4.03–7.92], p<0.001; Figure 1). A similar correlation was also found with MR severity, median hs-CRP levels being significantly increased with more severe MR (moderate: 2.62 mg/l [1.70–2.95], moderate to severe: 5.20 mg/l [3.30–5.6], and severe: 5.53 mg/l [4.50–9.30], p=0.002; Figure 2).

Figure 1.

High-sensitivity C-reactive protein levels according to NYHA functional class. Top of the bar represents median and whiskers represent the 25th and 75th percentile of concentrations. hs-CRP: high-sensitivity C-reactive protein.

P1: NYHA I vs. NYHA II; P2: NYHA I vs. NYHA III; P3: NYHA I vs. NYHA IV; P4: NYHA II vs. NYHA III; P5: NYHA II vs. NYHA IV; P6: NYHA III vs. NYHA IV.

p1<0.001, p2=0.001, p3<0.001, p4=0.008, p5=0.062, p6=0.794.

(0.1MB).

Figure 2.

High-sensitivity C-reactive protein levels according to degree of mitral regurgitation. Top of the bar represents median and whiskers represent the 25th and 75th percentile of concentrations. hs-CRP: high-sensitivity C-reactive protein.

P1: moderate vs. moderate to severe; P2: moderate vs. severe; P3: moderate to severe vs. severe.

p1=0.99, p2=0.004, p3=0.004.

(0.08MB).

UA levels were significantly elevated among symptomatic compared to asymptomatic patients (6.5±2.4 vs. 4.5±1.6 mg/dl; p<0.001). Mean UA levels increased significantly with NYHA class: 4.46±1.58 mg/dl for NYHA class I, 5.91±1.69 mg/dl for class II, 6.31±2.16 mg/dl for class III and 8.86±3.17 mg/dl for patients in class IV (p<0.001; Figure 3). Mean UA levels also increased significantly with increased severity of MR (moderate: 5.62±1.9 mg/dl, moderate to severe: 5.56±1.2 mg/dl, severe: 7.38±3.4 mg/dl, p<0.001; Figure 4). There was a significant correlation between UA level and LVEDD (r=0.40; p<0.001, Figure 5), LVESD (r=0.297; p=0.001) and LVEF (r=0.195, p=0.036), whereas hs-CRP was not correlated with these parameters.

Figure 3.

Uric acid levels according to NYHA functional class. Top of the bar represents mean and whiskers represent standard deviations. UA: uric acid.

P1: NYHA I vs. NYHA II; P2: NYHA I vs. NYHA III; P3: NYHA I vs. NYHA IV; P4: NYHA II vs. NYHA III; P5: NYHA II vs. NYHA IV; P6: NYHA III vs. NYHA IV.

p1=0.029, p2=0.010, p3<0.001, p4=0.840, p5<0.001, p6=0.001.

(0.07MB).

Figure 4.

Uric acid levels according to degree of mitral regurgitation. Top of the bar represents median and whiskers represent the 25th and 75th percentile of concentrations. UA: uric acid.

P1: moderate vs. moderate to severe; P2: moderate vs. severe; P3: moderate to severe vs. severe.

p1<0.001, p2<0.001, p3=0.123

(0.06MB).

Figure 5.

Relationship of uric acid levels with left ventricular diastolic diameter in patients with isolated and organic moderate or severe MR. LVDD: left ventricular diastolic diameter; UA: uric acid.

(0.08MB).

In addition, hsCRP levels were nearly significantly higher (5.5±2.5 vs. 4.2±3.6 mg/l, p=0.053), and UA levels were significantly higher (7.6±2.8 vs. 5.5±1.9 mg/dl, p<0.001) in patients who needed surgery than in those who did not according to current ESC guidelines on valvular heart disease.1

Mean follow-up was 437.1±133.9 (20–545) days. During follow-up 21 (10.5%) patients died and 26 (13%) underwent surgery. Increased UA levels were significantly associated with an increased risk of death during follow-up, along with male gender, increased serum creatinine, reduced LVEF, more severe MR and higher NYHA functional class (Table 2). In multivariate Cox proportional hazards analysis LVEF, NYHA class and UA levels were the only independent predictors of death (Table 2).

We found a significant increase in UA and hs-CRP levels in patients with higher degrees of MR and NYHA class. We also found that increased UA level, reduced LVEF and higher NYHA class were significantly associated with an increased risk of death during follow-up.

Surgery is indicated in patients who have symptoms due to chronic MR but no contraindication to surgery. The management of asymptomatic patients is controversial as there are no randomized trials to support any particular course of action.2,19 In some patients, including those who are elderly, physically inactive or obese, it is difficult to assess symptoms because of inactivity. On the other hand, in some patients, such as those with pulmonary disease, it may be unclear whether symptoms are related to severe MR or comorbidities.5,6 Noninvasive markers that show early changes in the cardiovascular system would therefore be helpful in assessing patients with MR.6 Several studies have reported the value of elevated BNP levels to identify asymptomatic MR patients at higher risk of developing heart failure, LV dysfunction or death in mid-term follow-up.5,20,21 A recent study showed a significant correlation between proadrenomedullin levels and degree of MR and NYHA functional class.22 It has been reported that patients with ischemic heart disease, LV systolic dysfunction and dilated cardiomyopathy with elevated hs-CRP have a worse prognosis.23,24 Two previous studies indicated that CRP predicts heart failure and death after MI.25,26 In our study, we found that hs-CRP levels were significantly higher among symptomatic than in asymptomatic patients. Additionally, hs-CRP levels were found to be significantly increased with higher degrees of MR and NYHA class. Hence, hs-CRP may be of value for identification of advanced stages of MR when other tools are not available or less informative. We also believe that hs-CRP may have a potential role for monitoring patient response to medication or surgery in follow-up. In agreement with our study, Arruda-Olson et al. reported that the presence of diastolic dysfunction and moderate or severe MR was associated with higher CRP levels. After further adjustment for infection within two weeks prior to MI, the association between CRP and diastolic dysfunction was no longer significant, but the association between CRP and MR remained.8 In contrast to studies in HF patients, hs-CRP was not correlated with LV dimensions or LVEF.

Pathways of UA production and metabolism may have a considerable impact on CHF pathophysiology and its clinical picture.27,28 It has been shown that patients with CHF are hyperuricemic compared to controls, independently of the effects of diuretics, renal impairment and other metabolic factors.9 UA levels were significantly correlated with LV systolic and diastolic dysfunction in patients with CHF.29,30 Hyperuricemia is a constant finding in CHF, its prevalence increasing with disease severity.31,32 Elevated serum UA may itself convey prognostic information in CHF.33 Anker et al. demonstrated an independent and graded relationship between serum UA levels and survival in patients with moderate to severe CHF.10 It has been shown that in patients with mild to moderate CHF, elevated serum UA levels are strongly related to death, and this relationship is independent of CHF severity and impaired renal function.33 In the same study, hyperuricemia was reported to predict exercise intolerance and was a marker of inflammatory activation in CHF. Serum UA levels increased significantly in parallel with CHF severity expressed as NYHA class.33 In our study, UA was significantly higher among symptomatic MR patients than in asymptomatic patients. This may help to identify asymptomatic patients in follow-up. We also found that mean UA levels increased significantly with NYHA class and higher degrees of MR. These findings suggest that UA is a potential biomarker for the assessment of therapeutic response to either medication or surgical intervention in follow-up. In agreement with previous studies in patients with heart failure, UA was significantly correlated with LVEDD, LVESD and LVEF. Therefore, we believe UA levels may be useful to assess the extent of LV remodeling and in the optimal timing of mitral valve surgery in certain subsets of patients.

As a limitation of the study, we cannot rule out the impact of diuretic dose on UA.

UA and hs-CRP levels can help to distinguish patients with asymptomatic moderate to severe MR from symptomatic patients. UA levels may be useful to assess the extent of LV remodeling and the optimal timing of mitral valve surgery in certain subsets of patients.

The authors have no conflicts of interest to declare.