Wolff-Parkinson-White Syndrome and Ventricular Hypertrophy : Events Predictors

Background: The PRKAG2 syndrome is classified as a glycogen storage disease, characterized by the presence of the Wolff-Parkinson-White syndrome (WPW), ventricular hypertrophy (VH) and conduction system disease (CSD). Objectives: Finding potential prognostic factors for events in individuals affected by this disease and describing the clinical characteristics. Methods: Sixty individuals were monitored from March 2005 to March 2015, being divided into two groups: Group 1 (G1) patients with WPW, VH or both; and Group 2 (G2) asymptomatic patients, with normal physical examination, electrocardiogram and echocardiography. It included the performance of medical history, physical examination, electrocardiogram and echocardiogram. Holter and electrophysiological study were performed when necessary. Results: G1 was made of 18 out of the 60 patients selected. Of these, 11 (61.1%) had VH related to WPW, 6 (33.3%) had isolated WPW and 1 (5.6%) patient had isolated VH. The mean age was 27.0±16.0 years and 32 (53.3%) were male. Only the patients in Group 1 had isolated events: 3 (17.0%) cardiac arrests, 2 (11.0%) sudden deaths, 6 (33.0%) pacemaker implants, 4 (22.0%) transient ischemic attacks and 9 (50.0%) combined events. The events predictors in potential combined were: left atrium size (p=0.07) diabetes mellitus (p=0.05) and the atrioventricular blocks (p=0.019). Those factors did not have statistic significance when compared in the Cox regression analysis. Conclusions: In WPV patients with ventricular hypertrophy there was an association of diabetes mellitus, atrioventricular block and left atrium size with the main outcomes.


Introduction
The PRKAG2 1 syndrome was first described as a metabolic cardiomyopathy, characterized by the association of the triad: familiar-type Wolff-Parkinson-White syndrome (WPW), ventricular hypertrophy (VH) and conduction system disease (CSD).Those findings can be related or isolated.The WPW syndrome occurs in approximately 50.0-100% of the patients affected by the disease, in addition to frequent supraventricular arrhythmia, including the atrial fibrillation (AF).The CSD occurs in approximately 30.0-76.0%and VH, in 26.0-74.0% of the patients with the disease.Ten mutations related to the gene PRKAG2 2,3 were described so far.
The literature reports great polymorphism related to mutations in the gene PRKAG2, as well as cognitive alterations and DM in patients of advanced age 4 .From the electrophysiological perspective, the presence of accessory pathways (AP) of the nodo-fascicular or ventricular type, with anterograde decremental conduction, was described at first.The presence of multiple AP was not much reported in the literature so far [5][6][7] .In Brazil, Sternick et al. 8,9 described two families with short PR interval, Mahaim fibers and CSD and, recently, van der Steld et al. 10 described a new mutation in the gene PRKAG2 related to cardiac arrest, congestive heart failure and multiple AP.
Different from the hypertrophic cardiomyopathy (HCM), caused by a sarcomeric protein mutation, the PRKAG2 syndrome includes a deposit of glycogen in the vacuole 11 .The VH manifests itself after the second decade of life and the WPW syndrome shows alterations in the electrocardiogram (ECG) from the childhood.In the absence of CSD, the differential diagnosis is made mainly with the Pompe disease, the Danon disease and the mitochondrial myopathies vertically transmitted 12,13 .
The description of fatal and nonfatal events predictors not related to the PRKAG2 syndrome remains as a gap in the literature.
This study aimed at finding risk predicting factors in potential for cardiac events in patients of the association of the WPW syndrome, VH and CSD.Furthermore, describing the clinical, electrocardiographic, echocardiographic characteristics of a familiar cohort in patients of this disease.

Methods
Prospective

Statistical analysis
The categorical variables were expressed in percentages, and the continuous variables, in average±standard deviation, medians and interquartile interval (IQI) of not normal distribution, and compared in the chi-square test or Fisher's exact test.The Shapiro-Wilk test was used for testing the variables normality.
The survival and other outcomes were assessed with the comparative Kaplan-Meier curves, between G1 and G2.
In the G1 patients, the outcomes were compared in relation to the several variables.The log-rank test was used for the comparison between the groups.The multivariate analysis of Cox was performed for the independent predictors in potential.The analysis included the variables whose univariate model had significance ≤0.10.The variable remaining with significance of p<0.05 was considered as independent predictor.The results were expressed as hazard ratio with the respective confidence intervals of 95%.The statistic program SPSS version 14.0 (SPSS Inc, Chicago, IL, USA) and R 2.15.2 was used.

Results
Out of the total of 84 patients in the cohort, 24 were excluded, with 60 (71.4%) people studied remaining.The cohort is made of patients with mean age 27.4±16.The genogram showed a dominant autosomal hereditary disease.Out of the 84 patients in the cohort, 18 were from G1 (Figure 1).
The symptoms started in the childhood, being intensified as the age increased, as well as the complications.Three patients of WPW associated to VH (II.7,II.10 and II.15), reported spontaneous abortions and early fetal death up to the third month of life.
The asterisk identifies the children with deafness, physical hypo-development, speech and learning disabilities.In the first and second generations, patients (I.1) and (II.4) had sudden death preceded by palpitation There was no difference statistically significant regarding the variables age, gender, presence of VT, PW thickness to the echocardiogram.All the other variables assessed presented statistically significant difference, while comparing groups G1 and G2 (Table 1).
In group G1 the following symptoms prevailed: palpitation, dyspnea, dizziness, myalgia, chest pain, pre-syncope, syncope and the factors related: SAH, DM and smoking, which contributed to the increase in the number of events.
It was observed that group G1 had the shortest PR interval, the most enlarged QRS complexes, the longest QTc interval.Only G1 included AVB, as well as bradyand tachyarrhythmia.Regarding the ECO, group G1 had lower FEVE and larger IVS thickness than group G2.

Electrocardiographic findings
During the 10-year follow-up, G2 had bradycardia in 4 (9.5%)patients and RBBB in 1 (2.4%); all the other clinic, electrocardiographic and echocardiographic variables remained as normal.
There was a progressive standard of electrocardiographic changes in the children of G1, with short PR interval and delta wave with later development of morphology of intraventricular blocks including the enlargement of the QRS complexes QRS, tachy-and bradyarrhythmia.
Isolated outcomes found only in the G1 patients: SD, recovered CA, TIA/S and PM implants.
Combined outcomes: when one or more of the following outcomes occurred: SD, CA, TIA/S and/or PM.

Cardiac arrest recovery and documentation
There were three CR episodes in two patients of WPW associated to VH.The patient (II.10) had a CA episode at the age of 51, due to pre-excited AF, that degenerated into ventricular fibrillation (VF), during the positioning of the atrial electrode of the definitive artificial pacemaker (PM) implant, indicated due to the severe sinus bradycardia and cardiogenic shock (HR=20 bpm).Cardiorespiratory resuscitation procedures were followed, with respiratory ventilation and 300 J shock, with reversal to the sinus rhythm.No neurological consequences occurred.After that, the electrode was implanted to the right ventricle and the patient had a hemodynamic recovery.
Figure 2 shows the layout of the patient (III.27),male, who had his first CA episode at the age of 16, due to pre-excited AF, that degenerated into VF and was subject to successful cardiorespiratory reanimation.The patient remained with severe arterial hypotension and sinus bradycardia of 30 bpm.He recovered the hemodynamic stability after 24 hours, with the infusion of physiologic solution, atropine (0.04 mg/kg/dose) and intravenous amiodarone, in the dose of 10 mg/kg/day.
The second CA episode occurred with the same patient (III.27), at the age of 20, due to asystole following the use of intravenous amiodarone (10 mg/kg), for treating common atrial flutter 1:1 with ventricular rate of 300 bpm.
After a new cardiorespiratory reanimation, he underwent the implant of provisional PM, due to severe sinus bradycardia, removed 48 hours after the event.All the CA events were preceded by atrial tachyarrhythmia (AF and AFL) associated to the symptoms palpitation, presyncope and syncope.

Pacemaker implants
There were 6 (33.0%)PM implanted, being one dual chamber implantable cardioverter defibrillator, in patient (II.7), due to VH, obstruction in the left ventricle outflow tract, associated to VT and pre-excited AF.

Combined events (CE)
The presence of CE was found only in G1, in 9 (50.0%)patients of the group, and it can be said that the disease causes a great predisposition to the development of fatal or nonfatal events.

Univariate analysis
There was higher prevalence of outcomes with statistic significance in the patients from G1, compared to those in G2.
Figure 3 shows the Kaplan-Meier survival curves for the patients from G1 and G2, who survived free from the combined events, stressing their crossing.This shows the change in the survival and its impact between having the disease or not.The age considered for the statistical analysis, lower or higher than the median of 27 years, did not discriminate the higher incidence of combined outcomes (SD, CA, TIA/S, PM implants).No statistically significant difference was presented by the gender.
The factors related to the combined event found in the univariate analysis were: DM, with events occurring in average at the age of 27, ranging between 21-34 years, with a significance level (log-rank) of p=0.05,CI 95%: 23.400-26.600; the presence of CHB with events occurring in average at the age of 40.0, ranging between 25-56 years, log-rank (p=0.045) and CI 95%: 24.000-26.000;and the average size of the LA of 39.0 mm (p=0.07) and CI 95%: 15.12-33.938.The other variables had no statistic significance.
Some variables presenting biologic plausibility to the occurrence of CE had no statistical significance in this  cohort, through the univariate analysis.This was the case of AF, AFL and pauses, documented as CA physiopathologic mechanisms, but not revealing statistically significant data.

Multivariate analysis -Cox regression
The Cox model was used for determining the relative impact of the prognostic predictors.This model included all the variables with p<0.10 in the univariate analysis.The factors DM (p=0.201),CHB (p=0.180) and LA (p=0.475)lost their statistic significance.No factor analyzed remained as predictor independent from CE (Table 2).

Discussion
The WPW syndrome with structural heart disease is rare but even rarer is the association with CHB, loss of conduction through normal tract (VA node) and of AP, found in most of the patients from the G1 of this cohort.Although the genetic analysis not being presented, the phenotypic findings described confirm the suspect of the PRKAG2 syndrome, described by Gollob et al. 1 , as a myopathic degenerative process caused by the deposit of glycogen 1,8 .
In group G1, the disease had predominance in the female gender and all the patients were mixed-race people.The literature has no reference to the differences regarding gender and only one author mentioned the mutation of the gene PRAKG2 in mulattos 8 .
Only G1 had cardiac murmurs and symptoms like palpitation, dizziness and syncope, preceding the three CA documented.Those findings agree with the descriptions by Mehdirad et al. 14 , of the clinic and electrophysiological characteristics of the atrioventricular connections in the WPW syndrome associated to VH.
Other interesting fact was the detection of SHA and DM more frequently in G1 than in G2 and this was also observed by Gollob et al. 2 The population studied showed to the ECG, the presence of other AP besides Mahaim and predominance of short PR interval with delta wave in the childhood.
To the echocardiograph, there was a predominance of VH at the high and medium septum, similar to the data of Arad et al. 15 about the VH following the glycogen deposits and associated to the mutations of the genes PRKAG2 e LAMP2.
All the G1 patients with DM had an episode of TIA/S.CHB preceded one of the documented CA episodes.Those factors were predictors in potential to the occurrence of CE and are associated to poor prognosis.However, the Cox regression analysis failed to prove the prognosis independence, probably due to the small size of the sample and reduced number of events, considering the total number of people involved and/or determiners not identified, in addition to the inclusion of outcomes with different physiopathologies.
SAH was present in all patients with TIA/S, but was not statistically significant in the univariate analysis (p=0.279).Those findings were similar to the ones reported by Gollob et al. 2 , describing SAH associated to the mutation (R302Q) in the gene PRKAG2 for a 23 year-old patient, as well as Eosu et al. 4 , describing metabolic alterations as DM in advanced age and cognitive disabilities in patients with mutations in PRKAG2.
No description was found in the literature about this syndrome equal to those in this study regarding the CE predictors in potential, the predominance of the female gender, in addition to the electrocardiograph documentation of three CA episodes.
This study observed that the LA size had a more frequent association with the CE, with statistic significance.That factor is related to atrial tachyarrhythmia, in special the AF and AFL, which were frequent in patients of the WPW and VH association.The findings suggest that the LA size, the presence of atrial tachyarrhythmia, CHB, SAH and DM confer a poor prognosis relation in that disease.The data found are under Mehdirad et al. 14 who described metabolic alterations that affected the ionic channels and caused AF and CHB, related to the SD mechanisms in that disease.
As apparently incidental findings, reports by female patients of WPW associated to VH with history of spontaneous abortions, neonatal deaths were found.The presence of physical hypo-development, speech alterations, poor intellectual performance, deafness and autism was also observed and not found in the literature.In those cases, fate cannot be ruled out, but the authors suggest the phenotypic-genotypic investigation and anatomopathological examination, for the cause factor.
The deadly events were directly related to the presence of WPW, atrial tachyarrhythmia and asystole.VT was not included in the genesis of those events.Drug treatment to SAH, DM, of arrhythmias and invasive measures as pacemaker implants, RF ablation of arrhythmias are considered as beneficial measures, able to change the evolution of those patients.

Limitations and perspectives
A limitation of the study was the absence of genetic analysis of the people in the cohort, which is ongoing.The myocardial biopsy would be very useful in confirming the diagnosis.On the other hand, considering the risks inherent to the procedure, the option was for the clinic diagnosis.
The small sample size and the inclusion of outcomes with different physiopathologies made the multivariate analysis difficult.The patients are under family genetic advising in Hospital das Clínicas de Salvador.

Conclusions
In WPV patients with ventricular hypertrophy, there was an association of diabetes mellitus, atrioventricular block and left atrium size with the main outcomes.

Potential conflicts of interest
There are no relevant potential conflicts of interest.

Sources of Funding
This study had no external funding sources.

Academic Association
This manuscript is part of the Master's thesis of Lenises de Paula van der Steld, from Escola Bahiana de Medicina e Saúde Pública.
ABBREVIATIONS AND ACRONYMS

Figure 2
Figure 2 Layout of the cardiorespiratory arrest mechanism of the patient (III.27),16 years.Pre-excited atrial fibrillation.

Figure 3
Figure 3Curve of survival free from combined events (Kaplan-Meier) in groups G1 and G2 p < 0.0001; G1 = green; G2 = blue; CE -combined events The study was approved by the Research Ethics Committee from the Hospital São Rafael under no.03/10 and by the Research Ethics Committee from the Escola Bahiana de Medicina e Saúde Pública under no.165.803.All patients signed the Informed Consent Form, under the Resolution CNS 466/12.
study of familiar cohort, monitored in private hospital, serviced by the Unified Health System, in Salvador, BA, Brazil, from March 2005 to December 2015.The project was in compliance with the principles of the Helsinki Declaration.individuals who failed to answer the call, refused to participate in the study, failed to sign the Informed Consent Form and the patients of the Chagas disease were not included.The accessible population was made of four-generation family members, descending from a mix of Caucasian, ischemic attack/stroke (TIA/S), pacemaker implant (PM); Combined outcomes (CO): upon the existence of one or more out the following outcomes: SD, CA, TIA/S or PM. 2 -Independent variables: • Clinic: age, gender, palpitation, dizziness, myalgia, chest pain, pre-syncope, syncope, dyspnea, systemic arterial hypertension (SAH); • Electrocardiographic: PR interval, duration of QRS complexes, QTc, pauses ≥2.0s, SVT, ventricular tachycardia (VT), atrial fibrillation (AF), atrial flutter (AFL), atrial tachycardia (AT), pauses, AVB of I, II, III degree; • Echocardiographic: left atrium (LA) size, PW thickness and IVS, left ventricular ejection fraction (LVEF).

Table 1 Characteristics of the population studied, per groups
AVB -atrioventricular block; RBBB -right bundle branch block; LBBB -left bundle branch block ; DM -diabetes mellitus; AH -arterial hypertension; G1 -group of patients of the disease; G2 -group of patients apparently healthy; NE -no event; QTc -QT interval adjusted; SD -standard deviation