The Cardiovascular Manifestations and Management Recommendations for Ogden Syndrome

Abstract

The NatA complex is composed of the NAA10, NAA15, and HYPK sub-units. It is primarily responsible for N-terminal acetylation, a critical post-translational modification in eukaryotes. Pathogenic variants within NAA10 cause Ogden Syndrome (OS), which is characterized by varying degrees of intellectual disability, hypotonia, developmental delay, and cardiac abnormalities. Although the cardiac manifestations of the disease have been described extensively in case reports, there has not been a study focusing on the cardiac manifestations and their recommended clinical cardiac management. In this study, we describe the cardiac manifestations of OS in a cohort of 85 probands. We found increased incidence of structural and electrophysiologic abnormalities, with particularly high prevalence of QT interval prolongation. Sub-analysis showed that male probands and those with variants within the NAA15-binding domain had more severe phenotypes than females or those with variants outside of the NAA15-binding domain. Our results suggest that an OS diagnosis should be accompanied by full cardiac workup with emphasis on echocardiogram for structural defects and EKG/Holter monitoring for electrophysiologic abnormalities. Additionally, we strongly recommend that the use of QT-prolonging drugs be followed up with routine electrophysiological monitoring or consultation with a pediatric cardiologist. We hope this study guides clinicians and caregivers treating patients with OS and moves the field toward a standardized diagnostic workup for patients with this condition.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00246-025-03877-7.

Introduction

Ogden syndrome (OS), (OMIM #300855) is an X-linked neurodevelopmental disorder characterized by postnatal growth failure, severely delayed psychomotor development, variable dysmorphic features, and hypotonia. Many patients also have cardiac malformations or arrhythmias [1]. Ogden syndrome is causing by a mutation in the NAA10 gene and the associated NatA complex. N-α-terminal acetylation is a highly conserved form of co-translational protein modification, whereby an acyl group is transferred from Acetyl-CoA to the α-amino group at the N-terminus of a protein in order to modify the half-life, structure, or localization of the final product [2–8]. The most prevalent of N-α-terminal acetyltransferases is the NatA complex, which targets around 40% of the human proteome [2, 9]. The NatA complex is composed of three distinct sub-units: NAA10, NAA15, and HYPK [10, 11]. NAA10 functions as the catalytic domain, NAA15 as an auxiliary domain, and HYPK as a regulatory subunit [12–14].

NAA10 is considered an essential human gene as knockouts of it are incompatible with life [15]. NAA10’s transcript is 235 amino acids (AA) in length and includes a NAA15 interaction domain and a Gcn5-related N-acetyltransferase (GNAT) domain [11]. Variants in NAA10 lead to a distinct clinical phenotype characterized primarily by intellectual disability, hypotonia, variable developmental delay, dysphagia, difficulty feeding, and variable ophthalmic manifestations known as NAA10-related neurodevelopmental syndrome or Ogden Syndrome (OS) [16–19]. OS is an X-linked ultra-rare genetic syndrome. The first two families were sequenced and reported in 2011 [20]. In addition to the effects on the nervous, gastrointestinal, and ophthalmic systems, NAA10 variants lead to various cardiac manifestations that have been previously described in case reports, such as cardiomyopathy [21–25] and prolonged QT intervals [22, 26, 27]. Variants in the autosomal gene NAA15 can also present with variable developmental delay and cardiac manifestations, these individuals tend to have much less severe manifestations than those with OS [16, 18, 19, 28–32].

Specific cardiomyocyte models of OS disease have been developed through the creation of patient derived iPSC cell lines of the p.S37P (severe) and p.Y43S (milder) variants. These studies suggested there may exist specific NAA10-related arrhythmias secondary to increased calcium conductance through L-type voltage gate calcium channels [33]. Cell lines from a patient with the p.R4S were also developed and showed similarly increased calcium conductance, as well as sodium and potassium conductance dysfunction and cytoarchitectural and contractile abnormalities, as reported in a preprint [34].

There have been several case reports and manuscripts describing the genotype–cardiac phenotype relationship of singular or familial cases of OS. However, there has not yet been an overarching study focusing on a large cohort of individuals with OS and their cardiovascular manifestations of disease. Through a more thorough exploration of the cardiac manifestations of OS, we hope to provide direct guidance to clinicians and caregivers treating a child with OS. Additionally, we hope to expand the current understanding of NAA10 variant clinical manifestations.

Methods

Participant Data

The subject population is composed of a group of probands diagnosed with OS. Probands and their caregivers signed Institutional Review Board-approved consent forms and HIPAA forms to allow for their information to be used in the study, with approval of protocol #7659 for the Jervis Clinic by the New York State Psychiatric Institute—Columbia University Department of Psychiatry Institutional Review Board, followed by recent transfer to and re-approval by the Nathan Kline Institute Institutional Review Board (in 2024). The scope of the project was explained in lay language, and there was no financial compensation offered for participation.

Thorough medical histories were collected on all probands. Clinical exome sequencing was used to confirm the pathogenicity of each variant. All data were organized based on a unique identifier system, “OS_XXX,” that functions as an internal registry of individuals with Ogden Syndrome. This registry and its key are known only to the research team. Narrative histories were annotated and probands were characterized as having the presence or absence of cardiac pathology based on an official diagnosis from their associated medical records. Structural abnormalities were characterized discretely as either present or absent without grading.

Pathologies were grouped grossly into 6 distinct categories for organizational purposes: structural, valvular, myocardial, vascular, and cardiac electrophysiologic. Structural abnormalities include patent foramen ovale (PFO), atrial septal defect (ASD), ventricular septal defect (VSD), patent ductus arteriosus (PDA), bicuspid aortic valve, aortic root dilatation, ascending aorta dilation, and tetralogy of Fallot. Patients with tetralogy of Fallot were not considered to have, for example, VSD, to prevent overcounting. Valvular abnormalities include tricuspid regurgitation (TR), mitral regurgitation (MR), pulmonic regurgitation, aortic stenosis, and mitral stenosis. Myocardial pathologies include hypertrophic cardiomyopathy (HOCM) and ventricular hypertrophy of unknown etiology (ventricular hypertrophy). Vascular abnormalities include persistent left superior vena cava (PLSVC), pulmonary hypertension, pulmonary artery stenosis, and coronary fistula. Electrophysiologic abnormalities include bradycardia, shortened PR interval, prolonged QT interval, supraventricular tachyarrhythmias (SVT), atrial fibrillation, ventricular fibrillation, and ventricular tachycardia. Additional cardiac pathology observed and included in total pathology counts but not in the broader categories include syncopal episodes, dextrocardia, cardiac arrest, myocardial infarction, apical cyst, and heart failure. Patients with lesions that are known to cause one another, such as VSD and MR, were noted as co-occurring, but not grouped in order to not invoke the third-cause fallacy.

Analysis

Descriptive statistics were compiled for each proband. EKG parameters analyzed included rate, rhythm, and length of the PR, QRS, and QT intervals in milliseconds (ms). QTc for each EKG was calculated via the Bazett formula [35]. Mean rate, PR, QRS, QT, and QTc values were used for probands with greater than one EKG reading. Additionally, due to incomplete records, not all electrophysiological data included all components of the EKG such that certain probands were included in calculations for only the rate, only PR interval and rate, etc. Current ages were calculated by the probands’ date of birth subtracted from the date of their last diagnostic study. For those probands who are deceased, their age at the time of death was calculated. Murmurs were graded based off their description in patient cardiologist clinical reports or based off descriptive data from the echocardiography reports using the Recommendations for Noninvasive Evaluation of Native Valvular Regurgitation from the American Society of Echocardiography [36]. Patients without echocardiography reports that included quantitative descriptions or without characterization in their clinical cardiology reports were noted as pathologic and included in total counts. Murmurs categorized as mild or clinically insignificant via echocardiography or cardiologist notes were not included in total calculations.

Analyses were repeated to stratify by proband sex and by the presence of a variant within the NAA15 interaction domain (AA 1–58). One-tailed unequal variance t-tests were calculated comparing the prevalence of each category of cardiac pathology and total number of pathologies. Two-tailed unequal variance t-tests were calculated comparing the rate and length of rate, PR, QRS, QT, and QTc values. Chi-square statistics were calculated comparing each group’s presence or absence of ever having a recorded QTc interval greater than 440 ms. in males or greater than 460 ms. in females. Fisher’s exact test values were calculated comparing each group’s proportion of probands with at least one cardiac pathology and across all discrete counts of cardiac pathologies and for the sex distribution of probands with variants within the NAA15 interaction domain. All alpha values were set to 0.05. The null hypothesis was that there is no differences between groups. Our proposed alternate hypothesis is that male probands and probands with variants within the NAA15 interaction domain have more cardiac pathologies and more aberrant EKG parameters.

Sub-analysis was performed by further splitting probands into three groups—male, female Arg83 Cys, and female non-Arg83 Cys. The three groups were compared with two-sample unequal variance t-tests to isolate the effect of the greater number of probands with Arg83 Cys variants in this cohort. This was done to identify if Arg83 Cys presents with a unique cardiac phenotype, as they tend to present with a more severe neurodevelopmental phenotype than those with other variants [37] and to being the most common variant present in females. A Bonferroni correction was made to account for multiple hypothesis testing with a modified α of 0.02.

Lastly, correlation analysis was performed to uncover correlations EKG rate, PR, QRS, QT, and QTc values, and QTc length with the age of EKG acquisition. For probands with greater than three EKGs over time (OS_198, OS_106, OS_107, OS_108, OS_109, OS_127, OS_138, OS_137, OS_178, OS_182, and OS_152), the QTc analysis was repeated. P-values were calculated for each correlation coefficient and α was set at 0.05. A Bonferroni correction was applied to the individual EKG analysis to account for the multiple hypothesis testing with a modified alpha of 0.005 (n = 11 probands with greater than 3 recorded EKGs). Simple linear regressions for analyses performed over time were calculated to aid in data visualization.

Results

There were 85 probands included in the study—65 female and 20 males. The average age of the cohort was 16.5 years old (SD = 17.2). The average age of the females was 16.1 (SD = 12.1) and males was 17.9 (SD = 27.8). Ten deceased individuals were also included in the study, 9 males and 1 female. Three families were also included in the cohort. The first was a mother and her two sons—all with p.Tyr43Ser variants. The second was a mother and her two sons—all with p.Ile72 Thre variants. The third family is a mother and son, both with p.Arg116Gln variants. Probands were from 16 different countries, primarily the United States (n = 41) and the United Kingdom (n = 13). There were 25 unique variants across the probands with Arg83 Cys being the most prevalent (n = 33/85). A complete breakdown of the variants can be seen summarized in Supplemental Table 1. Further demographic breakdown by country is available in Supplemental Table 2.

The probands in this study underwent various types of medical testing to discern cardiac structure and function. The average age at the time of first documented cardiac screening test was 6.1 years old (SD = 9.2). From the cohort, 66 probands received a baseline echocardiogram (echo) or electrocardiogram (EKG) post-diagnosis. Abnormal EKG or echo readings were present in 46.48% (n = 33/71). Of the probands with abnormal EKG or echo readings, 60.61% (n = 20/33) received a follow-up Holter monitor or Cardiac MRI. For probands who received advanced cardiac testing, 75.0% (n = 15/20) were still alive at the time of the data freeze.

There were 161 total cardiac anomalies accounted for from the cohort—83 identified in the females and 78 in the males. Structural abnormalities (n = 49, males (M) = 22, females (F) = 27) and electrophysiologic abnormalities (n = 49, M = 21, F = 28) were equally most prevalent. Vascular pathologies were the least prevalent (n = 11, M = 7, F = 4). A complete quantitative count of the cardiac pathologies by group and by sex can be found pictorially in Fig. 1. The counts of each individual cardiac pathology can be seen in Table 1. The average number of pathologies present per proband was 1.9 (SD = 2.2). Males, on average, had 3.9 (SD = 2.7) different cardiac abnormalities, while females had 1.3 (SD = 1.7). There were 13 probands with variants within the NAA15 interaction domain of NAA10 (males = 9, females = 4) and 72 without (males = 11, females = 61). There was a significant difference in the distribution of males and females who had variants within the NAA15 interaction domain (p = 0.0003). NAA15 interacting variants had an average of 3.9 (SD = 2.5) cardiac abnormalities versus 1.5 (SD = 2.0) in non-interacting variants. This difference was significant (p = 0.003). Of note, hypertrophic obstructive cardiomyopathy was more prevalent in the probands with variants outside of the NAA15 interaction domain (n = 0.08, SD = 0.3) versus those with a variants within it (n = 0, SD = 0; p = 0.007). The average number of cardiac pathologies present per group and by pathology is present in Table 2 and Fig. 2. It should be noted that only a single occurrence of MR (n = 5) co-occurred with VSD (n = 9). It never co-occurred with PDA (n = 7). Additionally, two occurrences of TR (n = 6) occurred with pulmonary hypertension (n = 3).

Fig. 1.

Discrete number of cardiac pathology by sex

Table 1.

Count of cardiac pathology by group

	Patent foramen ovale (PFO)	Atrial septal defect (ASD)	Ventricular septal defect (VSD)	Patent ductus arteriosus (PDA)	Bicuspid aortic valve	Aortic root dilatation	Ascending aorta dilation	Tetralogy of Fallot	Structural
Total Count	9	16	9	7	3	1	2	2	49
Male	4	7	5	5	1	0	0	0	22
Female	5	9	4	2	2	1	2	2	27
NAA15	4	3	3	4	1	0	0	0	15
No NAA15	5	13	6	3	2	1	2	2	34

	Tricuspid regurgitation (TR)	Mitral regurgitation(MR)	Pulmonic Regurgitation	Aortic stenosis	Mitral stenosis	Valvular
Total Count	6	5	2	2	1	16
Male	2	1	0	0	0	3
Female	4	4	2	2	1	13
NAA15	1	1	0	0	0	2
No NAA15	5	4	2	2	1	14

	Hypertrophic cardiomyopathy (HOCM)	Ventricular hypertrophy	Myocardial	Apical cyst
Total Count	6	15	21	1
Male	3	9	12	1
Female	3	6	9	0
NAA15	0	7	7	0
No NAA15	6	8	14	1

	Persistent left superior vena cava (PLSVC)	Pulmonary hypertension	Pulmonary vessel stenosis	Coronary fistula	Vascular
Total Count	1	3	6	1	11
Male	0	3	4	0	7
Female	1	0	2	1	4
NAA15	0	1	3	0	4
No NAA15	1	2	3	1	7

	Bradycardia	Shortened PR	Prolonged QT	Supraventricular tachyarrhythmias	Atrial fibrillation	Ventricular fibrillation	Ventricular tachycardia	Electrophysiological
Total Count	5	2	28	5	4	2	3	49
Male	1	1	9	2	3	2	3	21
Female	4	1	19	3	1	0	0	28
NAA15	1	0	7	0	4	2	1	15
No NAA15	4	2	21	5	0	0	2	34

	Cardiac arrest	Myocardial infarction	Heart failure	Syncopal episode	Dextrocardia	Total
Total Count	8	1	1	3	1	161
Male	7	1	1	2	1	78
Female	1	0	0	1	0	83
NAA15	5	1	0	1	1	51
No NAA15	3	0	1	2	0	111

Table 2.

Proportion of Probands with Specific Cardiac Pathology by Group

	Patent foramen ovale (PFO)	Atrial septal defect (ASD)	Ventricular septal defect (VSD)	Patent ductus arteriosus (PDA)	Bicuspid aortic valve	Aortic root dilatation	Ascending aorta dilation	Tetralogy of Fallot	Structural
Total (SD)	0.11 (0.31)	0.19 (0.39)	0.11 (0.31)	0.08 (0.27)	0.04 (0.18)	0.01 (0.11)	0.02 (0.15)	0.02 (0.15)	0.58 (1)
Males	0.2 (0.4)	0.35 (0.5)	0.25 (0.4)	0.25 (0.4)	0.05 (0.2)	0 (0)	0 (0)	0 (0)	1.1 (1.3)
Females	0.08 (0.3)	0.14 (0.3)	0.06 (0.2)	0.03 (0.2)	0.03 (0.2)	0.02 (0.1)	0.03 (0.2)	0.03 (0.2)	0.42 (0.8)
p value	0.1	0.04*	0.04*	0.02*	0.4	0.2	0.08	0.08	0.02*
NAA15	0.31 (0.5)	0.23 (0.4)	0.23 (0.4)	0.31 (0.5)	0.08 (0.3)	0 (0)	0 (0)	0 (0)	1.15 (1.4)
No NAA15	0.07 (0.3)	0.18 (0.4)	0.08 (0.3)	0.04 (0.2)	0.03 (0.2)	0.01 (0.1)	0.03 (0.2)	0.03 (0.2)	0.47 (0.9)
p value	0.05	0.3	0.1	0.03*	0.3	0.2	0.08	0.08	0.06

	Tricuspid regurgitation	Mitral regurgitation	Pulmonic Regurgitation	Aortic stenosis	Mitral stenosis	Valvular
Total	0.07 (0.26)	0.06 (0.24)	0.02 (0.15)	0.02 (0.15)	0.01 (0.11)	0.19 (0.52)
Males	0.1 (0.3)	0.05 (0.2)	0 (0)	0 (0)	0 (0)	0.15 (0.4)
Females	0.06 (0.2)	0.06 (0.2)	0.03 (0.2)	0.03 (0.2)	0.02 (0.1)	0.2 (0.6)
p value	0.3	0.4	0.08	0.08	0.2	0.3
NAA15	0.08 (0.3)	0.08 (0.3)	0 (0)	0 (0)	0 (0)	0.15 (0.4)
No NAA15	0.07 (0.3)	0.06 (0.2)	0.03 (0.2)	0.03 (0.2)	0.01 (0.1)	0.19 (0.5)
p value	0.4	0.4	0.08	0.08	0.2	0.4

	Hypertrophic cardiomyopathy (HOCM)	Ventricular hypertrophy	Myocardial	Apical cyst
Total	0.07 (0.26)	0.18 (0.38)	0.25 (0.43)	0.01 (0.11)
Males	0.15 (0.4)	0.45 (0.5)	0.6 (0.5)	0.05 (0.2)
Females	0.05 (0.2)	0.09 (0.3)	0.14 (0.3)	0 (0)
p value	0.1	0.003*	0.0003*	0.2
NAA15	0 (0)	0.54 (0.5)	0.54 (0.5)	0 (0)
No NAA15	0.08 (0.3)	0.11 (0.3)	0.19 (0.4)	0.01 (0.1)
p value	0.007*	0.006*	0.02*	0.2

	Persistent left superior vena cava (PLSVC)	Pulmonary hypertension	Pulmonary vessel stenosis	Coronary fistula	Vascular
Total	0.01 (0.11)	0.04 (0.19)	0.07 (0.26)	0.01 (0.11)	0.13 (0.4)
Males	0 (0)	0.16 (0.4)	0.2 (0.4)	0 (0)	0.35 (0.7)
Females	0.02 (0.1)	0 (0)	0.03 (0.2)	0.02 (0.1)	0.06 (0.2)
p value	0.2	0.04*	0.04*	0.2	0.04*
NAA15	0 (0)	0.08 (0.3)	0.23 (0.4)	0 (0)	0.31 (0.6)
No NAA15	0.01 (0.1)	0.03 (0.2)	0.04 (0.2)	0.01 (0.1)	0.1 (0.3)
p value	0.2	0.3	0.08	0.2	0.1

	Bradycardia	Shortened PR	Prolonged QT	Supraventricular tachyarrhythmias (SVT)	Atrial fibrillation	Ventricular fibrillation	Ventricular tachycardia	Electrophysiological
Total	0.06 (0.24)	0.02 (0.15)	0.33 (0.47)	0.06 (0.24)	0.05 (0.21)	0.02 (0.15)	0.04 (0.18)	0.58 (0.79)
Males	0.05 (0.2)	0.05 (0.2)	0.45 (0.5)	0.1 (0.3)	0.15 (0.4)	0.1 (0.3)	0.15 (0.4)	1.05 (1)
Females	0.06 (0.2)	0.02 (0.1)	0.29 (0.5)	0.05 (0.2)	0.02 (0.1)	0 (0)	0 (0)	0.43 (0.7)
p value	0.4	0.3	0.1	0.2	0.06	0.08	0.04*	0.008*
NAA15	0.08 (0.3)	0 (0)	0.54 (0.5)	0 (0)	0.31 (0.5)	0.15 (0.4)	0.08 (0.3)	1.15 (0.9)
No NAA15	0.06 (0.2)	0.03 (0.2)	0.29 (0.5)	0.07 (0.3)	0 (0)	0 (0)	0.03 (0.2)	0.47 (0.7)
p value	0.4	0.08	0.06	0.01*	0.02*	0.08	0.3	0.02*

	Cardiac arrest	Myocardial infarction	Heart failure	Syncopal episode	Dextrocardia	Total
Total	0.09 (0.29)	0.01 (0.11)	0.01 (0.11)	0.04 (0.18)	0.01 (0.11)	1.9 (2.2)
Males	0.35 (0.5)	0.05 (0.2)	0.05 (0.2)	0.1 (0.3)	0.05 (0.2)	3.9 (2.7)
Females	0.02 (0.1)	0 (0)	0 (0)	0.02 (0.1)	0 (0)	1.3 (1.7)
p value	0.003*	0.2	0.2	0.1	0.2	0.0002*
NAA15	0.38 (0.5)	0.08 (0.3)	0 (0)	0.08 (0.3)	0.08 (0.3)	3.9 (2.4)
No NAA15	0.04 (0.2)	0 (0)	0.01 (0.1)	0.03 (0.2)	0 (0)	1.5 (2.0)
p value	0.02*	0.2	0.2	0.3	0.2	0.0003*

*Denotes significance with α < 0.05

Fig. 2.

Number of cardiac pathologies by sex and by NAA15 interaction status

Females were further stratified into Arg83 Cys variants (R83 C; n = 32) and non-R83 C variants (n = 33). These groups were compared to each other and to the males to determine if the R83 C variant serves as an effect modifier. The average prevalence in R83 C female probands of cardiac pathology was 2.0 (SD = 2.0) versus 0.73 in non-R83 C females (SD = 1.1). This difference was statistically significant (p = 0.001). There was also a significant difference between the males and both R83 C females (p = 0.015) and non-R83 C females (p = 0.00001). The average prevalence of structural cardiac pathology in R83 C females was 0.72 (SD = 1.17) compared to 0.21 (SD = 0.55) in non-R83 C females. There was also a significant difference between the males and the non-R83 C females (p = 0.01). The average prevalence in R83 C female probands of myocardial pathology was 0.22 (SD = 0.42), while non-R83 C females had 0.06 (SD = 0.24). There was a significant difference between the males and R83 C (p = 0.01) and the non-R83 C groups (p = 0.0003). The average prevalence of electrophysiologic pathology in R83 C female probands was 0.66 (SD = 0.75) and 0.21 (SD = 0.48) in non-R83 C females. This was a statistically significant difference (p = 0.01). There was also a significant difference between the males and the non-R83 C group (p = 0.002). The remainder of the findings were not statistically significant. A summary of the cardiac pathologies present in the R83 C and non-R83 C groups can be found in Supplemental Table 4. The p-values for comparisons between the groups can be found in Table 3.

Table 3.

Cardiac pathology by males vs. Arg83 Cys females vs. other variant females

p Values	Structural	Valvular	Myocardial	Vascular	Electrophysiologic	Total
M vs R83 C	0.31	0.28	0.01*	0.12	0.14	0.015*
M vs Other	0.01*	0.78	0.0003*	0.09	0.002*	0.00001*
R83 C vs Other	0.03	0.38	0.07	0.62	0.01*	0.002*

*Denotes significance with α < 0.02

Out of 85 probands, 39 had electrophysiological data. Of the 39 with electrophysiological data, 11 had more than three reported EKGs. A total of 166 EKGs were compiled with an average age of acquisition of 22.7 (SD = 16.7) years old. The average and standard deviation of the age of EKG acquisition, rate, PR, QRS, QT, and QTc values for the cohort and probands with more than 3 recorded EKGs can be seen in Table 4.

Table 4.

Average EKG parameters by cohort and individual proband

Identifier	Variant	EKG Count	Age (SD)	Rate (SD) n = 33	PR (SD) n = 30	QRS (SD) n = 31	QT (SD) n = 32	QTc (SD) n = 38	Corr Coeff (p-val)
OS_198	p.Cys17Gly	21	7.62 (0.5)	92 (16.1)	127 (12.5)	81 (4.8)	384 (37.9)	472 (23.6)	− 0.3 (0.11)
OS_106	p.Ser37Pro	17	0.76 (0.4)	127 (16.3)	102 (8.82)	66 (4.7)	300 (25.0)	434 (20.6)	0.2 (0.29)
OS_107	p.Tyr43Ser	47	44.1 (3.46)	7 (14.7)	154 (50.0)	85 (6.5)	437 (60.4)	464 (41.2)	− 0.1 (0.61)
OS_108	p.Tyr43Ser	24	23.94 (3.89)	76 (12.5)	127 (87.5)	101 (14.4)	448 (56.5	503 (28.3)	0.43 (0.02)
OS_109	p.Tyr43Ser	27	24.6 (3.4)	91 (29.7)	186 (103.2)	119 (13.2)	419 (64.1)	498 (34.4)	0.44 (0.02)
OS_127	p.Arg83 Cys	6	4.39 (1.9)	111 (18.2)	91 (7.2)	62 (2.6)	316 (17.7)	443 (19.7)	0.38 (0.09)
OS_138	p.Arg83 Cys	6	0.04 (0.01)	142 (13.3)	114 (29.6)	70 (13.0)	302 (34.1)	439.(18.5)	0.36 (0.11)
OS_137	p.Arg83 Cys	3	1.5 (0.09)	134 (3.7)	148 (4.0)	68 (11.3)	311 (6.6)	465 (6.8)	− 0.21 (0.46)
OS_178	p.Arg83 Cys	7	31.1 (1.2)	105 (11.12)	81 (57.5)	83 (4.8)	370 (22.8)	482 (30.5)	− 0.87 (0.0003)*
OS_182	p.Leu126 Arg	4	1.2 (0.5)	116 (11.9)	123 (12.5)	85 (5.4)	355 (14.2)	492 (5.5)	− 0.5 (0.09)
OS_152	p.Phe128Leu	3	3.3 (0.9)	159 (23.5)	65 (46.2)	56 (0.9)	270 (20.1)	443 (9.8)	1 (0.002)*
	Total	213	11.6 (11.8)	106 (25.5)	120 (25.4)	77 (14.0)	358 (52.9)	460 (31.6)	0.18 (0.35)

All probands were included in the total EKG count. The mean values for the cohort utilized the mean values for probands with greater than 3 EKGs to minimize skew. Individual probands were only included in the table if they had greater than 3 EKGs administered. Age is the age at time of EKG acquisition in years. Rate and the PR, QRS, QT, and QTc intervals are in milliseconds. N is equal to the number of probands with electrophysiological data that were included in calculations. Corr Coeff is the correlation coefficient calculated between QTc interval and age of EKG acquisition

*Denotes significance with α < 0.005

Of the 32 probands with QT and QRS interval data, 7 had variants in the NAA15 interaction domain. The average QT interval in those variants was 395 ms (SD = 45.7 ms). The average QT interval in those with variants outside of the NAA15 interaction domain was 349 ms (SD = 47.7 ms). This difference was not statistically significant (p = 0.056). Additionally, there were no statistically significant differences in the average heart rates (p = 0.25), PR (p = 0.18), QRS (p = 0.14), or QTc (p = 0.48) intervals between probands with variants within the NAA15 interaction domain and those with variants outside of it. Lastly, there were no statistically significant differences between male and female mean electrophysiological data. A summary of the mean EKG parameters and their p-values broken down by sex, NAA15 interaction domain, and living status are recorded in Supplemental Table 4.

When assessing whether probands ever had a recorded prolonged QTc interval, 85.7% (n = 6/7) of the probands with variants within the NAA15 interaction domain had at least one. For probands with variants outside of the NAA15 domain, only 51.6% (n = 16/31) probands had a recorded prolonged QTc. This was a statistically significant difference (p = 0.049). Of the males, 88.9% (n = 8/9) had at least one prolonged QTc value, compared to only 48.3% of the females. This was also statistically significant (p = 0.015).

The average QTc in this cohort was 473 ms (SD = 38.7 ms). OS_178 had a strongly negative relationship between the age of EKG acquisition and QTc interval (r = − 0.87). This was a statistically significant correlation (p = 0.0003). Conversely, OS_152 had a strongly positive relationship between age of acquisition and QTc interval (r = 1.0; p = 0.002). These relationships over time can be seen graphically in Fig. 3a for the cohort as a whole and Fig. 3b for each individual proband. The remainder of the correlation coefficients for the cohort and each individual proband and their p-values are present in Table 4.

Fig. 3.

QTc length over time. A QTc (ms) versus age (years) for the entire cohort (n = 166 EKGs). B QTc (ms) versus age (years) for probands with greater than 3 EKGs (n = 11 probands) taken. A line of best fit was added to 2 A) to aid in data visualization

Out of 85 probands, 79 provided medication records. Of these probands, 40 were treated for noncardiac comorbidities, including seizures, behavioral disorders, intellectual disabilities, and/or infections. 24.05% (n = 19/79) were prescribed antiepileptics, 5.06% (n = 4/79) were on antipsychotics, 6.33% (n = 5/79) were on anxiolytics, 11.39% (n = 9/79) were on antihypertensives or stimulants, 10.13% (n = 8/79) were on antihistamines, antacids, or anti-inflammatory medication, and 12.66% (n = 10/79) were on antimicrobial, antibiotic, or antifungal medication. A table of all medications and their classes can be found in Supplemental Table 5. None of the probands taking QT-prolonging medications presented with evidence of prolonged QTc intervals at the time of their EKGs.

Ten deceased individuals were included in this cohort. The average age at time of death was 9.0 (SD = 4.3) years with a range from 0.3 to 64 with 7/10 passing away from cardiac arrest. Half of this group had prolonged QT intervals recorded. Of these patients, 3/10 were prescribed β-blocking medication and 1/10 were prescribed dihydropyridine calcium channel blockers. One of the patients was prescribed ranitidine. A more thorough clinical history of the decreased probands can be found in Supplemental File 2.

Only 6 valvular lesions (n = 16) were able to be graded based on quantitative data from echocardiography reports or cardiologist clinical reports. Two probands had mild TR and one had moderate TR. One proband had moderate MR and another had moderate to severe MR. One patient had severe MS. The remainder of the echocardiography reports showed benign results or did not comment on murmurs.

Discussion

Cardiac issues are present across males and females with Ogden Syndrome regardless of genetic variant. This phenomenon has also been observed in the previous literature. Cheng et al. [30] reported 22 patients, associated with heterozygous or hemizygous mutations in the NAA10 gene. Patients ranged from 4 months to 15 years of age, except for one 34-year-old woman. A third of patients had cardiovascular problems, such as long QT syndrome, atrial septal defect, and pulmonary hypertension. Saunier et al. [38] reported 11 unrelated females and a male and female sibling pair with Ogden syndrome. Cardiac conduction abnormalities included long QT in two probands and right bundle branch block in three [38]. In our study, structural and electrophysiologic pathology are the most prevalent. One explanation for the structural pathology seen in OS could be due to NatA playing a role in the proper formation of the heart. NAA10 is expressed relatively consistently within the cardiac tissue of mice during development [39] and has been shown to play a role in neuronal development [40]. Thus, dysfunction could lead to the structural malformations seen in the disease. However, human cardiac organoid studies identifying and comparing the expression of NAA10 at different stages of development have not been performed.

The incidence of electrophysiological abnormalities could be due to decreased NAA10 activity, which may contribute to altered cardiomyocyte ion channel and conductance development [34]. For example, increased calcium conductance is seen in the p.R4S, p.S37P, and p.Y43S variants of disease [33, 34]. Calcium is traditionally associated with mediating excitation-coupling and cardiac contractility [41] and increased calcium conductance can precipitate long QT syndromes [42]. In addition to increased calcium conductance, OS has been associated with increased voltage-gated sodium conductance and decreased potassium repolarizing current [34]. Impaired repolarization is another mechanism by which long QT syndrome can develop, as is seen when applying potassium blocking drugs to myocardial tissue [43]. Due to the prevalence of prolonged QT intervals in patients with OS, it should be recommended that clinicians proceed with caution before prescribing any QT-prolonging therapies. These circumstances also warrant additional EKG monitoring and regular follow-up. Furthermore, given the lethality of ventricular fibrillation secondary to prolonged QT intervals, there should be a decreased threshold for the implantation of cardioverter defibrillation devices in this population [22]. These recommendations are extrapolated from the studies performed on cardiomyocytes with p.R4S, p.S37P, and p.Y43S variants.

In this cohort, males with OS had more severe mortality and morbidity than females, as evidenced by the greater average prevalence of myocardial, electrophysiologic, and total cardiac pathologies and greater number of cardiac arrests. This finding could be due to skewed X-inactivation in the females leading to more expression of functional NAA10, and thus a less severe phenotype [44, 45]. However, the worse cardiac presentation in the males contrasts with the neurodevelopmental and ophthalmic findings of the disease, where females tended to have more severe phenotypes [16]. While additional studies need to be performed to ascertain the pathophysiology of OS, providers should give caregivers of children with OS anticipatory guidance surrounding myocardial and arrhythmogenic disease. Additionally, it is recommended that asymptomatic female carriers of OS alleles be screened for cardiac pathology to aid in early identification of cardiomyopathy [38, 46].

Interestingly, when controlling for the most prevalent variant, Arg83 Cys, the difference between total pathologies remained. There also appeared to be a stratification of the differences between the male, Arg83 Cys, and non-Arg83 Cys groups. This is different than what has been observed in the neurodevelopmental symptoms of OS, in which there were no differences in adaptive functioning between probands with Arg83 Cys variants and those with non-Arg83 Cys variants [37]. Mechanistic studies have shown that Arg83 Cys variants have reduced NAA10 activity, suggesting that variations in the precise level of NAA10 activity may contribute to significant differences in observed phenotypic pathologies [38].

NAA15-related neurodevelopmental syndrome is a disease that is related to OS in that both associated proteins are related to the NatA complex. NAA15-related neurodevelopmental syndrome has also been associated with hypertrophic obstructive cardiomyopathy (HOCM), suggesting NAA15 plays a role in NatA’s function in cardiac homeostasis [31]. Characterization of cardiac ion channels of iPSC’s modeling OS variants p.S37P and p.Y43S both exhibited a long QT phenotype and the increased prevalence of electrophysiologic abnormalities in this group further supports this assertion. Our findings suggest providers should lower their threshold for referring patients with OS NAA15 interacting variants to pediatric cardiology for workup if symptomatic.

Several medications prescribed to probands in our cohort are associated with QT interval prolongation. These include levetiracetam, risperidone, lisdexamfetamine, methylphenidate, famotidine, ciprofloxacin, erythromycin, or ofloxacin [38]. As patients with Ogden Syndrome are predisposed to cardiac issues, caution should be taken when prescribing these and similar medications. Alternative medications with lower incidence of QT interval prolongation are recommended to avoid the exacerbation of cardiac pathologies. In instances where QT-prolonging medications are necessary, there should be a low threshold for cardiac screenings, follow-up, and specialist referrals.

Lastly, due to the retrospective nature of this study, the ultra-rare nature of the disease, and the relative paucity of structural and biochemical data regarding NAA10, there are several limitations to this study. For example, while this study represents one of the largest cohorts of OS studied to date, the retrospective design leads to inconsistencies in the tests ordered between patients. This can be most clearly observed in the number of EKGs ordered, as some patients had nearly 40 times more EKGs done than others. The uneven administration of EKGs across our cohort lends to decreased reliability of our findings. A future study that is performed prospectively and longitudinally, ideally from birth, would help increase the internal validity of the findings. The retrospective design of the study also limited our results due to inconsistencies or incomplete record keeping to the inability to grade valvular and other structural lesions. Thus, these abnormalities were only compared by the presence as opposed by severity which can lead to an over-reporting of the severity of the cardiac manifestations in these cases. While this may contributed to the characterization of the genotype–phenotype associations present in OS, it also leads to results that may be less clinically significant. In addition, small sample size due to the ultra-rare nature of the disease acted as a limitation to our study. For example, both the male group and the NAA15 interaction group had worse cardiac outcomes than the female and non-NAA15 interaction groups. However, our cohort included a significantly greater number of males within it that have NAA15 variants. Due to the small overall number of probands, it is difficult to determine whether sex or variant type had a greater effect on cardiac outcomes in the disease. Increasing the cohort size would allow for subtle differences that were otherwise hidden to be elucidated. Lastly, the lack of biochemical and structural data regarding NAA10 and NatA function hindered our ability to provide mechanistic explanations for our clinical observations. While a handful of studies exist that helped showcase, for example, electrophysiology of cardiomyocytes in iPSC lines of several OS variants, further work must be done to characterize the NAA10 structure and perform additional electrophysiological and biochemical studies in-vitro to aid in understanding NAA10’s function in cardiac development.

Conclusion

There is a high incidence of structural and electrophysiological abnormalities in OS. These issues are especially prevalent in males and in individuals with variants within the NAA15-binding domain of NAA10. Patients with OS should be evaluated comprehensively for cardiac structural and electrophysiologic abnormalities upon diagnosis and also during the follow-up by echocardiography, EKG, and Holter monitoring. Furthermore, caution should be exercised in prescribing QT-prolonging drugs to these patients. This is especially important given the high prevalence of comorbidities such as intellectual disability and seizures in this patient population, which often rely on therapies associated with QT prolongation. Given the ultra-rare nature of this disease, further prospective work with larger cohorts is required to more accurately determine whether sex or variant type has a greater impact on overall cardiac pathology and function.

Ethical Approval

Both oral and written patient consent were obtained for research and publication, with approval of protocol #7659 for the Jervis Clinic by the New York State Psychiatric Institute Institutional Review Board.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contribution

GJL conducted all virtual interviews with participants, with data curation conducted by EM, RP, and RO. RM, RP, and GJL were responsible for project conception. RM was responsible for data analysis. The first draft of the manuscript was written by RM and RP, with critical revision performed by GJL and RO at several points.

Funding

This work is supported by New York State Office for People with Developmental Disabilities (OPWDD) and NIH NIGMS R35-GM-133408.

Data Availability

All data are deidentified to protect subject privacy, and the underlying data cannot be shared due to these same privacy restrictions. No datasets were generated or analysed during the current study.

Declarations

Conflict of Interest

The authors declare no competing interests.