Rheumatic mitral stenosis (MS) is frequently seen in developing countries and causes significant morbidity and mortality [¹]. Percutaneous mitral balloon valvuloplasty (PMBV) has become an effective and safe procedure for symptomatic or hemodynamically significant MS with favorable valve anatomy [²]. This procedure is highly successful with a low complication rate and significant short and long term improvement in hemodynamics and symptoms [³,⁴]. P wave dispersion (PWD) is an electrocardiographic (ECG) marker of nonuniform and heterogeneous atrial conduction with ECG leads of different orientation [⁵]. It can be defined as the difference between maximum and minimum P wave duration. Previous investigations have shown that Pmax and PWD are increased in patients with rheumatic MS and decreased with PMBV [⁵,⁶]. In addition the prolongation of intraatrial and interatrial conduction time and the inhomogeneous propagation of sinus impulses are well known electrophysiologic characteristics of the atrium prone to fibrillate. Prolonged P-wave duration and increased PWD have been reported to carry an increased risk for atrial fibrillation (AF) [⁵]. Therefore, we aimed to investigate the immediate and long term effects of successful PMBV on PWD in severe rheumatic MS patients with sinus rhythm (SR) and to analyse the restenosis, AF and redoballoon valvuloplasty rate.

In this study, we enrolled 41 patients. The mean patient age was 44.1±10,86 years and 95% (n:39) of the patients were female. The median Wilkins echo score in all patients was 8 (range: 6-12). All patients’ mitral regurgitation grade was <2/4. Baseline characteristics were shown in [Table 1]. All patients underwent successful PMBV. We grouped patients as 1 (prePMBV), 2 (1-3 days after PMBV), 3 (long term after PMBV) according to the time of PMBV. All patients had normal LV systolic function. There were no significant differences with respect to left ventricle enddiastolic (LVEDD) and endsystolic diameter (LVESD), left ventricle ejection fraction (LVEF) and Pmin between three groups. Statistically significant improvement in LA diameter, MVA, PAPs, max mitral gradient, mean mitral gradient, Pmax and PWD were achieved in all patients (p<0.001), ([Table 1]). A comparison of changes in electrocardiographic and echocardiographic parameters between three groups are displayed in [Table 1]. Mean pre-PMBV MVA (MVA 1) was 1.07±0.19 cm2 and significantly improved to 1,91±0,35 cm2 after PMBV (MVA 2) (p<0.001). At long term the mean MVA (MVA 3) was significantly decreased to 1,63±0,40 (p<0,001). Mean pre-PMBV mean gradient was 14,51±6,22mmHg, which significantly decreased to 5,76±3,04 mmHg (p<0,001) within 72 hours after PMBV. At long term the mean gradient significantly increased to 7,29±3,92 mmHg (p=0,004). Median PAPs 1 was 45 (30-120) mmHg and significantly decreased to 35 (25-120) mmHg (p:<0,001) ; at long term the mean PAPs was 35 (20-115) mmHg indicating nonsignificantly difference (p=1,0). The median LA 1, LA 2, LA 3 diameter was 44 (34-69), 42(33-54),42 (33-54) cm (p<0,001,p<0,001,p=0,673). Pmax 1 and PWD 1 decreased 1-3 days after PMBV (Pmax 2;PWD 2); (p<0,001) but the difference at 3 year was insignificant (Pmax 3; PWD 3) (p=1,0 for Pmax; p=0,961 for PWD) ([Figure 1],[Table 1]). Pmin didn’t differ between groups (p=0,092). Median PWD 1 was 45 (30-68 ms, PWD 2 was 35 (20-59) ms indicating significantly difference (p<0,001).

Figure 1 Box-plot representations for PWD and Pmax between three groups. PWD 1/Pmax 1:pre PMBV, PWD 2/Pmax 2:1-3 days after PMBV, PWD 3/Pmax 3:3 years after PMBV.

---

---

In correlation analysis, negative correlation was detected between MVA 1 and PWD 1 (r=-0.097, p=0.047). PWD 2 was positively correlated with echo score , PAPs 1, LA 1 and LA 2. Furthermore, PWD 3 was also correlated with LA 1, LA 3, PAPs 3, MVA 3 positively. In long term follow up; 4 AF, 12 restenosis occurred and 2 patients underwent mitral valve replacement with mechanical prosthesis, 3 patients underwent redoballoon. We therefore analyzed patients, who had composite endpoints (atrial fibrillation, restenosis, redoballoon, mitral valve surgery). Fourteen composite endpoints were seen. Restenosis occured at an average of 3,92±2,15 year. Composite endpoint was associated with LA 1 (p = 0.02), MVA 2 (p= 0.019), mean gradient 2 (p= 0.028), PWD 3 (p < 0.001) and Pmax 3(<0,001) ([Table 2], [Figure 2]). Additionally the incidence of the AF endpoint was associated with PWD 2 (p=0,019) and PWD 3 (p=0,010). By receiver operating characteristic (ROC) curve analysis, increased PWD was significantly associated with adverse cardiac events (area under ROC curve [AUC]:0,905, %95 confidence interval [CI]:0,813-0,996). The best cut-off value was defined as 39 ms for MACE (Accuracy = 90.5%, sensitivity = 85,7%, specificity = 81,5%) ([Figure 3]).

Figure 2 Box plot of ECG parameters comparing MACE – and MACE +.

---

---

Figure 3 ROC curve analysis for PWD. To determine the best cut-off value that discriminated between MACE+ and MACE-. The cut-off was defined as 39 ms [Accuracy = 90.5%, sensitivity = 85,7%, specificity = 81,5%]

---

---

This study has demonstrated that PWD 2 and PWD 3 were associated with AF and PWD 3 was an independent predictor of long term event. We also found that Pmax and PWD significantly decreased 1-3 days after PMBV as previously described by other studies [⁶,⁹,¹¹]. Three years later median Pmax and PWD didn’t change. MVA improved immediately after the procedure but later the mean MVA decreased significantly indicating the initiation of restenosis. The constancy of PWD lately, can be explained by the onset of restenosis. Greater ECG changes are achieved acutely and persist in the long term. However, patients with detoriated ECG parameters were interpreted as composite endpoint precursors in long term.

We found that PWD 1 was negatively correlated with MVA 1 and also PWD 3 was correlated with age, MVA 3, LA 1 and LA 3. These findings suggest that increasing mitral stenosis and larger LA volume are associated with more severe structural changes in the left atrium, leading to greater electrical inhomogeneity, non-uniform conduction velocities within the atrial myocardium, which manifests on the ECG as increased P-wave dispersion. So if PWD is longer than before, we can think about small MVA, larger left atrium requiring therapy.

Rheumatic MS still remains a major health problem in developing countries. A common arrhythmic complication encountered in patients with MS is AF. Indicators of electromechanical heterogeneity along the atria reflect the pathological changes within the atrium and also indicate an increased risk for AF development [¹⁰]. For patients with MS, disorganization of the atrial muscle bundles may be present. Structural changes in LA due to MS cause inhomogeneous electrical properties, abnormal conduction velocities and local dispersion of refractoriness within the atrial myocardium. This electrophysiological characteristic results in increased PWD and Pmax on electrocardiographic measurements. It has been reported that patients with rheumatic MS have increased PWD compared with control subjects [¹¹-¹³]. Demirkan et al. [⁶] have shown in a study including 30 patients who were eligible for PMBV, that there was statistically significant decrease in atrial electromechanical delay (AEMD) with P-max and PWD in the early period after PMBV (in 72 h). In another study, it was shown that only successful PMBV was associated with a decrease in Pmax and PWD [¹⁴]. Similar to our results; successful PMBV procedure results in a decrease in PWD. The decline of PWD immediately after PMBV was not supported by Beig et al [¹⁵]; they found statistically significant decrease in inter and intraatrial electromechanical delays. Prolongation of P-wave duration and increased PWD were correlated with increased AF risk according to previous studies [⁵,¹⁶]. The correlation between the presence of intraatrial conduction abnormalities and the induction of paroxysmal AF have been well documented [¹⁷]. All these findings suggest that mechanical dilation of the mitral valve with a balloon also reduces susceptibility to AF even in the early period [¹⁸]. In our study finding an association between AF and PWD 2 / PWD3 support these results.

Our study is the first study that has long term follow-up for PWD in patients with MS. Consistent with previous studies, we documented a significant decrease in P-max and PWD after successful PMBV. The supplementary finding of this study is that after PMBV, late term PWD was associated with AF and identified as an independent predictor of the composite endpoint. By ROC curve analysis, increased PWD was significantly associated with major advers cardiac events. The 39 ms PWD cut-off level was of significant diagnostic value for MACE by the ROC curve analyses ([Figure 3]). In the light of these information, following MS patients with atrial conduction properties can be useful for restenosis and AF prediction. Further prospective randomized studies are required to confirm our results.

All the authors declare no conflict of interest.

This study has demonstrated that Pmax and PWD significantly decreased within three days after PMBV, furthermore, late term PWD was associated with AF and identified as an independent predictor of the composite endpoint. Patients without any decrease in Pmax and PWD long term after PMBV may potentially be at greater risk for atrial fibrillation and restenosis. By electrocardiographicly P wave indices, immediate and long term effects of PMBV on relieving mitral valve obstruction can be reflected and also AF and restenosis can be predicted.