P-Wave Duration and Dispersion in Patients with Mitral Stenosis

Background: Mitral stenosis (MS) is related to prolonged inter- and intra-atrial electromechanical delays and increased P-wave dispersion. The objective of the current study was to investigate the correlation between the P-wave duration, P-wave dispersion (PWD), mitral stenosis (MS) and to explore the cut-off values for predicting the MS in the patients. Methods: We enrolled 62 patients with MS and sinus rhythm as test group, and 62 healthy subjects matched in age- and sex- were selected as control group. We conducted the 12-lead electrocardiogram and echocardiography for all the subjects. The maximum and the minimum P-wave duration and PWD were calculated. Univariate and multivariate logistic regression analyses were performed to demonstrate the correlation between P-wave duration and PWD and MS. The receiver operating characteristic (ROC) curve was drawn to detect the threshold of P-wave duration and PWD for predicting the MS. Results: There were signicant differences in the left atrial diameter (45.00±5.78 vs. 32.31±4.24 cm 2 ), pulmonary artery pressure (46.68±17.29 vs. 32.64±2.86 mm Hg), left ventricular end-diastolic diameter (47.57±4.80 vs. 45.58±5.04 cm), ejection fraction (63.10±3.05 vs. 65.13±2.56%), aortic root inside diameter (29.60±3.50 vs. 31.58±3.58) and pulmonary trunk (24.17±2.78 vs. 22.23±1.77) values between the test group and the control group. Besides, the test subjects had signicantly longer maximum P-wave duration (123.42±12.33 vs. 108.18±9.07) and larger P-wave dispersion (47.24±13.61 vs. 28.94±9.19). In the multivariate analysis, maximum P-wave duration (OR:1.221, 95% CI:1.126-1.324) and P-wave dispersion (OR:1.164, 95% CI:1.094-1.238)


Background
The rheumatic mitral stenosis (MS) is a highly prevalent, clinically acquired, and progressive valvular cardiac disorder. It commonly occurs in developing countries burdened with the weak economy, which results in a notable rise in morbidity and mortality rate [1] . The most common cause of aberrant atrial conduction is abnormal mitral valve structure or dysfunction, which is related to the increased risk of atrial brillation (AF). Protraction of conduction time in atrium and atrium and uneven sinus impulse transmission are the electrophysiological features associated with the AF [2,3] . Electrocardiographic features have long been evaluated with the help of electrocardiogram (ECG) and transesophageal ultrasound. However, the application of ECG characteristics for helping the diagnosis of MS is rarely studied. P-wave re ects the electrical activity of the atria, and the features of P-wave, especially P-wave duration, can be used as markers of changes in atrial conduction P-wave dispersion (PWD) refers to the differences leads of ECG. Previous study showed a novel electrocardiographic indication for heterogeneous and intermittent dissemination of sinus impulses from the ECG [4] . Furthermore, escalated PWD on surface ECG signi es the increased risk of AF and left atrial volume [5] . The relationship between the MS and P-wave indices remains unclear and demands further investigation.
The objectives of the current study were: (1) to compare the electromechanical prolongation in the atrium of MS patients and healthy individuals; (2) to investigate the association between the electromechanical prolongation in atrium and the PWD in MS patients and healthy individuals; (3) to establish the optimal threshold to test the Chinese subjects with MS.

Study subjects
We diagnosed 598 consecutive patients with mitral stenosis (MS) by echocardiography at the Shanghai Chest Hospital A liated to Shanghai Jiaotong University, from May 2016 to December 2019. The exclusion criteria were primary or secondary hypertension, coronary disease, type 2 diabetes, permanent or paroxysmal AF, severe aortic or mitral regurgitation, aortic stenosis, valvular intervention, thyroid dysfunction, respiratory diseases, bundle branch block, intraventricular conduction dysfunction, ventricular preexcitation, prior pacemaker implantation or irregular electrolytes. The sinus rhythm for each test subject was con rmed by testing on a 12-lead electrocardiogram (ECG) before the start of the study.
Besides, an echocardiographic evaluation was also conducted for each test subject. Our exclusion criteria resulted in a nal cohort of 62 patients (22 mild, 22 moderate, and 18 severe cases of MS). The healthy control group included in this study contained 62 individuals from our hospital without a history of cardiac diseases or the presence of cardiovascular hazard factors. The ethics committee of the Shanghai Chest Hospital A liated to Shanghai Jiaotong University approved our study.

Echocardiography
A couple of trained physicians, blinded to patient's characteristics, performed all the echocardiographic examinations by using an ultrasound echocardiography (Vivid E95, General Electric, USA) with a 3-MHz sensor. After 15 minutes, participants were assessed by ECG in the left recumbent position. The LA diameter and left ventricular end-systolic and end end-diastolic diameters were calculated. Additionally, the mitral valve area (MVA) was estimated by employing a mean pressure half-life and planimetric approach. MVA less than or equal to 1 cm 2 was marked as severe, 1 cm 2 < MVA ≤ 1.5 cm 2 was marked as moderate, and 1.5 cm 2 MVA < 2 cm 2 was marked as mild stenosis.

Twelve-Lead Ecg
We used a 12-lead ECG with a speed of 25 mm/s, and a voltage of 10 mm/mv to document the information in the test subjects under the supine position. In order to acquire accurate data, we calculated ECGs manually by using a magnifying function by the Nalong system. Before recording the ECG, the test subjects were rested for 15 min and con rmed for the presence of sinus rhythm by ECG. Two trained observers blindly completed the gauge of the P-wave duration. It was measured in milliseconds from all the 12 leads. The starting point of the P-wave was reported as the point where the original de ection of the P-wave crossed the equipotential line, and the endpoint of P-wave was reported as the outlet where the eventual de ection of P-wave crossed the equipotential line. The P-maximum was de ned as the longest duration, and the P-minimum was described as the shortest duration among the 12 leads. The Pwave dispersion (PWD) was described as the difference between the maximum P-wave duration and the minimum P-wave duration. Additionally, in this study, we also included test subjects with P waves in nine or more leads that could be precisely measured (6).

Statistical Methods
The continuous variables are presented as the mean ± standard deviation (SD), and categorical variables are presented as number and frequencies (%). The Student's t-test, Wilcoxon rank-sum test, Pearson Chisquare test, and Fisher exact test were employed for the comparison of the continuous and the categorical variables. Univariate and multivariate logistic regression analyses were performed to demonstrate the correlation between P-wave duration and PWD and MS. The receiver operating characteristic (ROC) curve was drawn to detect the threshold of P-wave duration and PWD for predicting the MS. P-value < 0.05 was represented as statistically signi cant. Data were analyzed using the SPSS Statistics software (version 22.0).

Baseline characteristics
The baseline descriptions of the test subjects and the control individuals are indicated in Table 1. 62 patients (15 males; mean age 55.63 ± 12.53) and an age-sex matched 55-year-old healthy control subject was enrolled in this study. Among the total test subjects included in this study, 22 had mild MS, 22 had moderate MS and 18 had severe MS. Both the test and the control groups were age, height, weight, BMI, heart rate, SBP, and DBP matched.

Univariate and multivariate logistic regression analysis of the P-wave duration and the P-wave dispersion in MS
In order to explore the parameters associated with the MS, we performed univariate and multivariate logistic regression analysis for further investigation. Univariate analysis showed that the P-wave duration (OR:1.173, 95% CI:1.106-1.244, P < 0.01), and P-wave dispersion (OR:1.150, 95% CI:1.096-1.2081, P < 0.01) in the test group (Table 3). Also, as per the multivariate analysis the occurrence of maximum Pwave duration (OR:1.221, 95% CI:1.126-1.324, P < 0.01), and P-wave dispersion (OR:1.164, 95% CI:1.094-1.238, P < 0.01) were related with the test group (Table 3). 3.4 The best threshold of the P-wave duration and the PWD for MS prediction using ROC curve Since maximum P-wave duration and PWD were closely related to MS, we quantitative analyzed the cutoff values of P-wave duration and PWD for predicting MS via ROC curves. Results showed that the optimal cut-off values were 119.50 ms and 42.50 ms for maximum P-wave duration and PWD, respectively. The AUCs for P-wave duration and PWD were 0.859 ms and 0.865 ms, respectively, indicating well predicting effects of the two factors ( Fig. 1)

Discussion
The major ndings of the current study were as follows: (1) MS patients exhibited a longer maximum Pwave and higher PWD as compared to the healthy individuals; (2) PWD and P-wave duration were strongly correlated with MS. (3) According to the ROC curve, it was found that the cut-off values for maximum P-wave duration and PWD for predicting the MS were 119.50 ms and 42.50 ms for PWD, respectively. These ndings may effectively contribute to the early detection of MS, which in turn may help the clinicians to impede the rate of MS progression.
PWD was rst de ned in 1998 by Dilaveris and his colleague as the difference between the maximum and the minimum P-wave duration in 12-lead ECG [3] . Numerous studies have established the relationship between the PWD and an array of cardiovascular disorders such as paroxysmal atrial brillation, hypertension, and MS [5,7−8] . In healthy subjects, it was found that the PWD was correlated with age [9] .
Also, the prolongation of P wave duration and the escalated PWD were related to increased risk of AF [3,5] . Therefore, PWD has been considered as a crucial indicator to predict the AF risk in MS patients with sinus rhythm [5] .
AF is a common clinical complication in MS patients. MS patients with AF were showed poor clinical prognosis, which was associated with the damage to the atrial contraction and fast ventricular rate [10] .
Furthermore, secondary injury of the mitral valve and the in ammation of the atrium can aggravate the dilation of the left atrium and the degree of myocardial brosis [11] . The resulting anatomical abnormalities will lead to electrical heterogeneity, heterogeneous transmission speeds, and heterogeneous refractory phases within the atrial myocardium. It may also be accompanied by the atrial wall brosis and discordance of the atrial bundle, which was manifested as the increase of the P-wave duration and the PWD in ECG [2,10] .
As compared to the healthy individuals, maximum P-wave duration, and PWD in MS patients were notably escalated, which suggested an increased risk of AF. In addition, a possibility of thromboembolism and mortality were signi cantly elevated in MS patients with AF. Hence, early detection of patients with MS by simple and acceptable method is a crucial step for the screen of MS patients in clinical practices. Previous studies showed some preliminary evidences on the relationship between ECG parameters and MS. Gholam and colleagues [12] found that the maximum P-wave duration was longer in patients with MS than that of matched controls, and the maximum P-wave duration was signi cantly correlated with the size of left atrium. More importantly, in a study with more than 3 years of long-term follow-up, unal et al. [13] revealed that the P-wave duration and PWD were increased with the development of MS further indicating the essential role of these two factors in the progression of MS. In this study, we revealed that in the absence of an echocardiographic evaluation, a prolonged P-wave duration and PWD would signi cantly contribute towards the early prognosis in MS patients. Also, a major advantage of this mode of prognosis is the low-cost of technology and the ease of evaluation.
Over the years, researchers have revealed the presence of increased sympathetic activity in MS patients [14,15] . Moreover, Tukek et al. [16] demonstrated that the enhanced sympathetic excitation results in a remarkable rise in PWD. Besides, previous studies have shown [14] the reduced sympathetic excitation after the mitral valve surgery and the bene cial activity of beta-blockers in MS patients [17] . A chronic administration of beta-blockers could result in the reduced maximum P-wave duration and PWD in MS patients by constraining the sympathetic activity [17] . Thus, these ndings suggest that enhanced sympathetic excitation could lead to the prolongation of maximum P-wave duration and PWD in MS patients.
We herein, for the rst time, determined the cutoff value for P-wave duration as 119.50 ms and P-wave dispersion as 42.50 ms, for the effective prognosis of MS in the Chinese population. Our unpublished data showed that a regular ECG evaluation of patients with an increased risk of mitral valve abnormalities could delay the progression of mitral stenosis. However, this study demonstrated that the P-wave duration and PWD might not be enough for the prognosis of the degree of mitral stenosis in the Chinese population. Therefore, an in-depth analysis and validation studies are required to determine the predictive e cacy of P-wave duration and the PWD in the MS population. Nevertheless, global access to the affected patient's clinical reports may help to unravel the mechanism of P-wave parameters and to access the increased risk of AF in MS patients.
This study, however, has certain limitations. Firstly, the sample size of the current research is comparatively small. Secondly, this study was a single-centered, observational study, which might have affected the generalizability. Thirdly, the ECG's P-wave calculations were performed manually rather than employing the computer-assisted P-wave calculations. Finally, a selection bias may have occurred due to the delay in recording the onset of paroxysmal AF in the affected individuals.

Conclusions
In summary, our results demonstrate that a longer P-wave duration and a higher PWD are correlated with the increased risk of MS progression. We should pay attention to the regular follow-up to impede the progression of mitral stenosis.

Declarations
Ethics approval and consent to participate The study was approved by the Shanghai Chest Hospital's local ethics committee. Written informed consent from all participants was obtained.
Consent for publication Not applicable.
Availability of data and materials The datasets used and/or analysed during the current study are deidenti ed and available from the corresponding author on reasonable request.
Competing interests The authors declare that they have no competing interests.
Funding Not applicable.

Authors' Contributions
Each author accomplishes the standard for the authorship. WXF and RGL planned and executed this study. WXF, JWL, LXC, WZL, XJW acquired the data. WXF drafted the manuscript. WXF performed the statistical analysis of data. All authors reviewed and accepted the nal version of this manuscript.