Incidence of Cardiac Arrhythmias in Asymptomatic Hereditary Hemochromatosis Subjects with C282Y Homozygosity

Abstract

It is not well known whether systemic iron overload per se in hereditary hemochromatosis (HH) is associated with cardiac arrhythmias before other signs and symptoms of cardiovascular disease occur. In this study, we examined the incidence of cardiac arrhythmias in cardiac asymptomatic HH subjects (NYHA functional class I), and compared it to that in age/gender-matched normal volunteers. The 42 HH subjects and 19 normal volunteer control subjects recruited through the NHLBI-sponsored "Heart Study of Hemochromatosis" completed 48-hour Holter electrocardiography ambulatory monitoring at the baseline evaluation. The HH subjects were classified as newly diagnosed (Group A) and chronically treated subjects (Group B). All HH subjects had C282Y homozygosity, and the normal volunteers lacked any HFE gene mutations which are known to cause HH. Although statistically insignificant, the incidence of ventricular and supraventricular ectopy tended to be higher in the combined HH groups than the controls. Supraventricular ectopy was more frequently noted in Group B as compared to the controls (ectopy rate per hour; 11.1±29.9 vs. 1.5±3.5, P < 0.05 by Kurskal Wallis test). No examples of heart block, other than first degree atrioventricular node block, were seen in any of the subjects. The incidence of cardiac arrhythmias was not significantly reduced after 6 months of intensive iron removal therapy in Group A subjects. No life threatening arrhythmias were observed in our HH subjects. In conclusion, our data suggest that the incidence of cardiac arrhythmias is, at most, marginally increased in asymptomatic HH subjects. A larger clinical study is warranted to further clarify our observation.

INTRODUCTION

The association of cardiac arrhythmia with hereditary hemochromatosis (HH) has been well documented.^1,2 However, most of the literature describe cases with cardiac arrhythmias in the presence of heart failure due to iron overload.^3–7 Thus, it is difficult to determine whether cardiac arrhythmias detected in HH subjects are directly linked to systemic iron overload rather than to heart failure. Interestingly, Kaiser et al. have recently shown that cardiac arrhythmias do not increase in the iron overload animal model of the gerbil, even when the level of iron overload induces liver cirrhosis.⁸ This observation raises further questions regarding the link between cardiac arrhythmias and systemic iron overload in HH. As a result, we studied the incidence of cardiac arrhythmias in cardiac asymptomatic HH subjects with C282Y homozygosity, known to be the most common HFE gene mutation to cause HH. We compared these results to those in age/gender-matched normal volunteers who lacked HFE mutations.

METHODS

We recruited 43 HH patients and 21 age/gender-matched healthy volunteer control subjects in total to participate in an NHLBI Institutional Review Board-approved protocol, number 03-H-0282, during the period of September 2003 to August 2005. Subjects were recruited with the assistance of the NIH Clinical Center Patient Recruitment and Public Liaison Office according to the office protocol. All subjects provided written informed consent. Eligibility criteria for HH subjects included: 1) age 21 years or older, 2) NYHA functional class I, 3) HFE genotype showing homozygosity for the C282Y HFE gene mutation, 4) transferrin saturation > 60% or a serum ferritin > 400 µg/L at the time of original diagnosis, and 5) absence of significant end-organ damage secondary to HH. Subjects with HH were categorized into two groups. Group A consisted of newly-diagnosed subjects who had undergone less than three lifetime phlebotomy treatments. Group B contained subjects who had undergone standard phlebotomy treatment for at least six months and were documented to be in a stable maintenance phase of therapy, achieving a transferrin saturation < 45% and a serum ferritin < 30 µg/L during the course of therapy. The control subjects' eligibility criteria included 1) age 21 years or older, 2) NYHA functional class I, 3) absence of C282Y or H63D mutations in the HFE gene, 4) normal transferrin saturation and serum ferritin. Exclusion criteria included a pattern of excessive alcohol use, uncontrolled hypertension, diabetes requiring use of more than one oral hypoglycemic agent or insulin, tobacco use within the prior three months, and current use of beta-adrenergic blockers or calcium channel blockers. Group A subjects were contacted for a 6-month follow up repeat evaluation. HH subjects were treated according to the current therapeutic phlebotomy guidelines.^9,10,11 These HH subjects were known to have similar left ventricular systolic and diastolic function to the control subjects as previously reported.¹²

Forty-two HH subjects (22 in Group A and 20 in Group B) and 19 normal volunteer control subjects completed Holter ECG ambulatory monitoring. After proper chest wall skin preparation, leads were placed for 3 channel recordings: CM-V1, CM-V3 and CM-V5 positions. Recordings were made using Aria Holter monitors (Del Mar Reynolds, Issaquah, WA) for approximately 48 hours. The recordings were analyzed using the Impresario Symphony Holter Analyzer (Del Mar Reynolds/Spacelabs Healthcare, Issaquah, WA) and interpreted by a board-certified cardiologist blindly (YS, DR).

Continuous data are presented as mean values ± SD and are supplemented with associated medians and interquartile range for variables of cardiac arrhythmias. When normality was established with the Shapiro-Wilks test, ANOVA with the Turkey-Kramer Honestly Significant Difference (HSD) test as a post-hoc test was used to compare means between the three groups. An unpaired Student’s t-test was used to compare variables between the combined HH group and the control group. A χ²-test was used to compare categorical data between the combined HH group and the control group. A paired Student's t-test was used to compare the parameters before and after phlebotomy treatment in group A. When normality was in question, the Kruskal-Wallis test was used to compare among the three groups, and the Wilcoxon rank sum test was used to compare between the combined HH group and the control group. The parameters analyzed using the nonparametric approach are highlighted with an “#” in the tables.

RESULTS

Parameters of iron overload, i.e., ferritin, tranferrin saturation, and serum iron, were significantly elevated in Group A compared to the control group (Table 1) indicating significant iron overload in newly diagnosed HH subjects. Transferrin saturation was significantly higher in Group B compared to the control group. Left ventricular ejection fraction calculated by the biplane Simpson method in transthoracic echocardiography, was ≥ 55% in all HH and the control subjects (data now shown). The recording time was shortened in 4 cases in group A, 7 cases in group B and 1 case in the control group due to subjects’ non-compliance, mechanical device failure, and accidental lead removal. In these cases, the test was not repeated. As a result, the recording time was slightly shorter in Group B than in Group A. In the Holter symptom log, no symptom associated with cardiac arrhythmias was entered for any subject. The incidence of ventricular ectopy and supraventricular ectopy tended to higher in the combined HH subjects than in the control group; however, these differences did not reach statistical significance (Table 2). The number of total beats and ectopy rate of supraventricular ectopy in HH subjects were significantly higher than those of normal volunteer controls (Table 2). Non-sustained ventricular tachycardia was noted only in the HH groups; but the frequency was not statistically significant as compared to the control group. We have previous reported that the myocardial iron level measured with T2* by cardiac magnetic resonance imaging was comparable among these 3 groups.¹³ In cases of newly diagnosed HH subjects (Group A) who could be assessed at 6-month follow up (n=20, 2 subjects were lost in contact), the incidence of cardiac arrhythmias was not significantly reduced at 6 months (Table 3) by the intensive phlebotomy therapy, although the iron level did markedly decreased {ferritin; 1189±899 vs. 280±464 (µg/dL), transferrin saturation; 76±18 vs. 48±27 (%), serum iron 191±43 vs. 140±74 (µg/dL) before and after 6 months of phlebotomy, P < 0.01 for all.} No sinus node or more than first degree atrioventricular node block were noted in both HH subjects and the control subjects (Table 3).

Table 1.

Clinical characteristics of hereditary hemochromatosis and control subjects.

Variables	Controls n=19	Group A n=22	Group B n=20	HH N=42
Age (years)	48 ± 9	48 ± 11	52 ± 9	50 ± 10
Female	7(37%)	6(27%)	7(35%)	13(31%)
Ferritin (µg/L)#	103 ± 78	1164 ± 886†¶	127 ± 269	670 ± 843†
Transferrin sat (%)	25 ± 10	76 ± 19†¶	40 ± 21†	58 ± 27†
Serum iron (µg/dL)	85 ± 27	189 ± 49†¶	109 ± 53	150 ± 63†
Hemoglobin (g/dL)	14.1 ± 1.1	14.6 ± 1.6	14.1 ± 0.9	14.4 ± 1.4
Hematocrit (%)#	41.6 ± 3.5	44.4 ± 4.5	42.0 ± 2.9	43.3 ± 4.0
Alanine aminotrasferase (IU/mL)#	27 ± 13	51 ± 24†¶	28 ± 17	40 ± 24
Aspartate aminotransferase (IU/mL)#	25 ± 26	35 ± 10†¶	26 ± 10	31 ± 11
Glucose (mg/dL)	99 ± 11	97 ± 14	95 ± 13	96 ± 14
Creatinine (mg/dL)	0.9 ± 0.2	0.9 ± 0.1	0.9 ± 0.2	0.9 ± 0.2

HH=subjects with hereditary hemochromatosis homozygous for C282Y, Group A=newly diagnosed HH subjects, Group B=chronically treated HH subjects. See Methods for statistical analyses.

^†P < 0.05 vs. Group C, and

^¶P < 0.05 vs. Group B. Data are expressed as mean ± SD.

^#Normality was not observed.

Table 2.

Characteristics of cardiac arrhythmias in hereditary hemochromatosis (HH) and control subjects.

Groups	Controls n=19	Group A n=22	Group B n=20	HH n=42
Age	48±9	48±11	52±9	50±10
Recording time (min)#	2690±600	2618±561¶	2331±711	2481±645
Average heart rate (bpm)	75±9	78±7	75±9	77±8
Minimum heart rate (bpm)	52±6	51±7	50±8	51±8
Maximum heart rate (bpm)	138±20	139±18	129±14	134±17
Ventricular ectopy:
Total beats#	20±49 [1,5]	284±814 [2,35]	39±80 [4,16]	167±598 [3,24]
Ectopy rate (beats per hour)#	0.41±1.02 [0.02,0.70]	5.93±16.92 [0.05, 0.85]	0.83±1.70 [0.09,0.32]	3.50±12.43 [0.08,0.50]
Single#	20±49 [1,5]	281±813 [2,35]	38±80 [2,17]	166±597 [2,24]
Nonsustained ventricular tachycardia	0	1	1	2
Sustained ventricular tachycardia	0	0	0	0
Supraventricular ectopy:
Total beats#	70±174 [11,36]	50±95 [7,50]	415±1248†§[44,76]	224±872 [19,72]
Ectopy rate (beats per hour)#	1.45±3.48 [0.23,0.70]	1.06±1.97 [0.15,1.05]	11.11±29.93†§[0.96,2.01]	5.85±21.05 [0.49,1.69]
Single#	73±171 [10,28]	44±90 [7,47]	140±432 [36,61]	90±305 [13,59]
Nonsustained suparventricular tachycardia	3	3	6	9
Sustained supraventricular tachycardia	0	0	0	0
Sinus node block	0	0	0	0
1st degree AV block	0	1	2	3
More than 1st degree AV block	0	0	0	0

Group A=newly diagnosed HH subjects, Group B=chronically treated HH subjects. The numbers of ventricular tachycardia and supraventricular tachycardia are an incidence during the ambulatory Holter electrocardiogram recording. See Methods for statistical analyses.

^†P < 0.05 vs. Controls,

^§P < 0.05 vs. Group A, and

^¶P < 0.05 vs. Group B. Data are expressed as mean ± SD. Medians and interquartile ranges are shown in [ ].

^#Normality was not observed.

Table 3.

Changes in cardiac arrhythmias before and after 6-month iron removal therapy in newly diagnosed hereditary hemochromatosis (HH) subjects.

Groups N	Before 20	After 20
Age	47±11	48±10
Recording time (min)	2663±518	2858±102
Average heart rate (bpm)	78±7	78±9
Minimum heart rate (bpm)	52±7	54±8
Maximum heart rate (bpm)	138±18	137±17
Ventricular ectopy:
Total beats#	312±850 [4,40]	230±604 [7,32]
Ectopy rate (beats per hour)#	6.52±17.67 [0.10,1.12]	4.26±10.82 [0.14,0.66]
Single#	310±849 [4,40]	205±521 [6,30]
Incidence of nonsustained ventricular tachycardia	1	1
Incidence of sustained ventricular tachycardia	0	0
Supraventricular ectopy:
Total beats#	55±99 [7,62]	23±21 [25,30]
Ectopy rate (beats per hour)#	1.15±2.05 [0.15,1.28]	0.56±0.51 [0.52,0.62]
Single#	47±93 [7,51]	26±24 [17,28]
Incidence of nonsustained supraventricular tachycardia	3	5
Incidence of sustained supraventricular tachycardia	0	0
Sinus node block	0	0
1st degree AV block	0	0
More than 1st degree AV block	0	0

Data are expressed as mean ± SD. Medians and interquartile ranges are shown in [ ].

^#Normality was not observed.

DISCUSSION

Our study shows that both supraventricular and ventricular ectopy tends to increase in the combined HH groups compared to those in age/gender-matched volunteer control subjects; however, only the number and rate of supraventricular ecotpy were statistically elevated in chronically treated HH subjects as compared to the volunteer control subjects. The incidence of cardiac arrhythmias is not reduced significantly in newly diagnosed HH subjects irrespective of a marked decrease of systemic iron level by phlebotomy therapy for 6 months.

Our report indicates that the arrhythmogenic effect of iron overload by HH in the asymptomatic stage is marginal at most. The acute reduction of iron overload at 6 months in newly diagnosed HH subjects (Group A) fails to reduce statistically the incidence of cardiac arrhythmias, and this observation further implies that systemic iron overload itself is weakly arrhythmogenic at best. Previous reports of cardiac arrhythmias in HH are for advanced stages of disease, often with concomitant heart failure symptoms.^3–7 There has been no previous systemic evaluation of cardiac arrhythmias in the asymptomatic phase of HH subjects with normal left ventricular function. Our observation questions a highly publicized view of the arrhythmogenicity of systemic iron overload. In addition, our previous observation of the lack of myocardial iron accumulation despite the high systemic iron levels¹³ may highlight the importance of organ specific iron accumulation susceptibility which in turn may dictate development of organ specific complications. By studying asymptomatic subjects, we may have preselected a group with HH who do not yet have myocardial iron overload, and thus do not suffer the same cardiac complications as those who have increased myocardial iron stores. A large clinical study is needed to better delineate the effect of iron overload per se on cardiac arrhythmias. The advancement of knowledge in this area promises to improve the cardiac management of iron overload.⁹

We have previously reported that levels of oxidative stress are persistently elevated in HH subjects, even when iron overload is improved. Given the well- documented association between arrhythmias and the level of oxidative stress,^14,15 our results suggest that the elevation of oxidative stress may be more closely associated with cardiac arrhythmias in HH subjects than systemic iron overload. The oxidative stress would not be normalized even though phlebotomy therapy achieves its goal of iron reduction.¹⁶ In addition, we have reported that complicated arrhythmias are more frequently observed in HH subjects during symptom-limited exercise tests,¹⁷ and exercise-induced cardiac arrhythmias may not correlate with arrhythmias noted on the Holter monitoring in our study. Nevertheless, our study has demonstrates that no life threatening arrhythmia was present on 48-hour Holter ECG ambulatory monitoring in HH subjects during regular daily activities.

The type of supraventricular tachycardia could not be further classified in this study due to the limited recording channels on Holter ECG recording. The more frequent early termination of recordings in HH subjects might cause an underestimation of cardiac arrhythmias in these groups. The subject number in this study is too small to detect subtle differences in the incidence of cardiac arrhythmia among the groups. The chance of coincidental statistical significance may have been increased due to multiple parameter comparisons. In addition, Holter data have reported to be a weak surrogate for life-threatening arrhythmias.^18,19 Thus, our study results cannot be extrapolated to estimate the risk of sudden cardiac death in this population.