- Email: [email protected]
Do alterations in inter-ictal heart rate variability predict sudden unexpected death in epilepsy?
Epilepsy Research (2009) 87, 277—280
journal homepage: www.elsevier.com/locate/epilepsyres
SHORT COMMUNICATION
Do alterations in inter-ictal heart rate variability predict sudden unexpected death in epilepsy? R. Surges a,∗, C. Henneberger a, P. Adjei a, C.A. Scott a, J.W. Sander a,b, M.C. Walker a a b
Department of Clinical and Experimental Epilepsy, UCL Institute of Neurology, Queen Square, London, UK SEIN — Epilepsy Institute of the Netherlands Foundation, Heemstede, The Netherlands
Received 25 June 2009; received in revised form 2 August 2009; accepted 9 August 2009 Available online 10 September 2009
KEYWORDS Heart rate variability; Autonomic dysfunction; Sympathetic tone; Sudden unexpected death in epilepsy; Nocturnal seizures
Summary Reduced heart rate variability (HRV) may predispose to sudden unexpected death in epilepsy (SUDEP). We ascertained whether HRV predicts SUDEP in chronic epilepsy using a case—control design and investigated parameters of inter-ictal HRV in 14 patients (7 had died from SUDEP). No HRV parameter was associated with SUDEP. Thus, although altered HRV might be involved in SUDEP, HRV parameters are not clear-cut predictors for SUDEP. © 2009 Elsevier B.V. All rights reserved.
Introduction Heart rate variability (HRV) is the beat-to-beat variability of successive heartbeats and is modulated by a balanced parasympathetic—sympathetic autonomic activity (Stein and Kleiger, 1999). Impaired HRV is known to be associated with an increased risk of sudden cardiac death in an apparently healthy population (Stein and Kleiger, 1999). Pre-
Abbreviations: AED, antiepileptic drug; FLE, frontal lobe epilepsy; HF, high frequency; HRV, heart rate variability; LF, low frequency; RMSSD, root mean square of successive differences; SD, standard deviation; SDNN, standard deviation of the RR intervals; SUDEP, sudden unexpected death in epilepsy; TLE, temporal lobe epilepsy; VLF, very low frequency. ∗ Corresponding author. Tel.: +44 20 7837 3611x4135; fax: +44 20 7278 5616. E-mail address: [email protected] (R. Surges).
vious studies have demonstrated decreased HRV in chronic epilepsy patients and have suggested that this may play a role in the pathophysiology of SUDEP (Tomson et al., 1998; Ronkainen et al., 2005). Possible mechanisms include an impaired autonomic function with augmented sympathetic tone (Diehl et al., 1997; Hilz et al., 2002). We ascertained inter-ictal HRV in a matched-pair case—control design to identify a potential predictive role of HRV for SUDEP in medically refractory epilepsy patients.
Methods Patients with medically refractory focal epilepsy who underwent standard pre-surgical assessment at the National Hospital for Neurology and Neurosurgery between 1996 and 2004 and who later died of SUDEP were included. Living medically refractory focal epilepsy patients were matched as controls for admission date of video-EEG telemetry, age and gender. At the last follow up all control patients were still alive. This assessment was part of a continuing mortal-
0920-1211/$ — see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2009.08.008
278
R. Surges et al.
ity audit and was approved as such by the Joint Ethics Committee of NHNN and the Institute of Neurology. Video-EEG telemetry was performed using conventional scalp EEG recordings (10—20 system) or intracranial recordings (1 patient) at a sampling rate of 200 Hz. ECG was recorded from 2 channels with a modified lead I (adhesive electrodes placed below the clavicles of either side). Peri-ictal data of 12 patients (no. 1—12, Supplementary Online Material) were included in a previous project (Surges et al., in press). HRV was analysed blinded to the outcome in a 1-h interval (by C.H.) with an in-house written software using Matlab (The MathWorks, Natick, MA, USA). As HRV was previously shown to be different in chronic TLE and control patients with most important differences during night around 4 am and during the afternoon (Ronkainen et al., 2005), HRV was determined during 1-h intervals between 4—5 am and 4—5 pm of the last recording day. Occasional ECG artefacts were removed prior to analysis. RR intervals were determined semi-automatically using the first derivative of the voltage signal and resulting data were visually inspected to ensure accuracy of the algorithm. RR intervals, the standard deviation of the RR intervals (SDNN, in ms) and their root mean square of successive differences (RMSSD, in ms) were calculated. The power spectrum of the RR-interval time series was obtained by a fast Fourier transform after re-sampling (7200 samples). The cumulative power was determined in three frequency bands: very low frequency (VLF, 0.005—0.04 Hz), low frequency (LF, 0.04—0.15 Hz), and high frequency (HF, 0.15—0.4 Hz). LF is primarily a measure of sympathetic activity, whilst HF is primarily a measure
Table 1
of parasympathetic activity. Sympathovagal balance was estimated by calculating LF/HF ratio. Statistics were performed using conditional logistic regression analysis with STATA software (StataCorp LP, TX 77845, USA). P-Values < 0.05 were regarded as statistically significant. The inter-individual variability of HRV parameters is relatively large (Tomson et al., 1998; Ronkainen et al., 2005). Therefore, we assumed that a clear-cut clinical predictor for SUDEP should differ at least by two standard deviations (SD) of the means. Then, with an ˛ of 0.05, we would require at least 6 people in each group to have a 90% chance of detecting this difference. Data values are expressed as mean ± SD.
Results A total of 7 SUDEP and 7 living control patients were assessed (Table 1). Mean age/epilepsy duration were 34.0 ± 7.0 years/23.9 ± 6.8 years in controls and 35.0 ± 7.7 years/23.9 ± 7.5 years in SUDEP patients. Most patients were on more than one antiepileptic drug (AED). AED treatment was reduced in 11 patients during telemetry, but re-instituted to usual doses at least 12—48 h prior to HRV assessment (supplementary table). No statistically significant differences were found in the studied HRV parameters (Table 2).
Clinical characteristics.
Patient
Age/sex/epilepsy duration
Etiology/epilepsy
1/S 2/C
32/F/10 29/F/22
Left TLE with hippocampal sclerosis Cryptogenic left FLE
3/S 4/C
33/F/25 32/F/11
Right TLE epilepsy with postischemic lesion Right TLE with diffuse post-radiation and resection (medulloblastoma) defect
5/S 6/C
36/M/26 32/M/29
Left FLE with gangliocytoma/hamartoma Left TLE with hippocampal sclerosis
7/S 8/C
21/M/19 25/M/20
Cryptogenic left FLE Right TLE with hippocampal sclerosis
9/S 10/C
44/M/26 43/M/26
Right FLE with posttraumatic lesions Cryptogenic left FLE
11/S 12/C
43/M/34 44/M/30
Left TLE with hippocampal sclerosis Bilateral TLE with bilateral hippocampal sclerosis
13/S 14/C
36/F/27 33/F/29
Cryptogenic TLE (probably left) Left TLE with dysembryoplastic neuroepithelial tumor
S, SUDEP patient; C, control patient.
Table 2
HRV parameters. RR (ms)
SDNN (ms)
RMSSD (ms)
Total power (ms2 )
VLF (ms2 )
LF (ms2 )
HF (ms2 )
LF/HF ratio
4—5 am Control SUDEP P-Value
994 ± 153 936 ± 140 0.51
70 ± 25 74 ± 40 0.83
49 ± 26 51 ± 32 0.92
3233 ± 1872 3846 ± 2596 0.67
1545 ± 816 1983 ± 1478 0.59
941 ± 513 1277 ± 802 0.44
747 ± 724 586 ± 507 0.66
2.31 ± 1.46 2.71 ± 0.81 0.50
4—5 pm Control SUDEP P-Value
799 ± 137 823 ± 71 0.72
59 ± 17 67 ± 16 0.45
46 ± 23 38 ± 12 0.53
2318 ± 1132 2319 ± 1186 0.99
905 ± 421 1245 ± 498 0.31
913 ± 456 777 ± 595 0.70
500 ± 449 297 ± 178 0.38
2.78 ± 1.66 2.81 ± 1.17 0.97
Mean ± SD.
Do alterations in inter-ictal HRV predict SUDEP?
Discussion Here, we aimed to identify a potential HRV-related predictor of SUDEP in medically refractory epilepsy patients who underwent pre-surgical evaluation. The inter-individual variability of HRV parameters is relatively large (Tomson et al., 1998; Ronkainen et al., 2005) and we were able to assess only a limited number of patients. Our failure to detect any difference in inter-ictal HRV in SUDEP and control patients could therefore be due to small sample size. Nevertheless, our study was powered to detect a difference of more than two standard deviations from the population mean, a criterion that is frequently taken to indicate clinical significance. Our results could also be biased by some inherent disadvantages of using a retrospective design. We have assessed patients who underwent video-EEG telemetry and who later died suddenly. People with chronic epilepsy with severe alterations of autonomic function might die earlier than those with less severe or little changes in HRV. The former patients have therefore a lower chance of being admitted to video-EEG telemetry. It is also possible, that although poor seizure control is an established risk factor for SUDEP, we might have missed SUDEP patients with relatively wellcontrolled chronic epilepsy, but severe alteration of HRV. There is, however, some evidence that severe alterations of HRV are associated with pharmacoresistance (Mukherjee et al., 2009). Finally, people who died suddenly were matched with living patients with pharmacoresistant focal epilepsy patients. We cannot exclude that these control patients may die of SUDEP in the future, and thus have been misclassified as controls. To reduce inter- and intra-individual heterogeneity of HRV, we controlled and checked for parameters which are known to affect HRV. Patients were matched for age and gender which both alter HRV (Umetani et al., 1998). Other potential confounders include sleep stages and AED treatment (Vaughn et al., 1995). The amount of slow wave and REM sleep was similar in SUDEP and control patients over night (not shown), so that sleep-related changes in HRV are unlikely to influence our study. Studies on AED effects on HRV yielded conflicting results and make it difficult to estimate a potential AED effect (Kennebäck et al., 1997; Hennessy et al., 2001; Persson et al., 2003; Lossius et al., 2007; Hallioglu et al., 2008). A putative effect of AEDs should have only minor influence on the comparison between both SUDEP and control patients, as AED types were relatively equally distributed in both groups and usual treatment was re-instituted 12—48 h prior to assessment of HRV parameters. The differential impact of the epilepsy or lesion type on HRV is unclear, but e.g. a decrease in parasympathetic activity was reported in both TLE and frontal lobe epilepsy (FLE) (Ronkainen et al., 2005; Harnod et al., 2009). Likewise, the influence of side of epileptogenic lesion in TLE has yet to be elucidated in larger studies (Tomson et al., 1998; Adjei et al., in press). An impaired HRV in chronic epilepsy has been well established in various studies, whereas the relative influence of parasympathetic or sympathetic system is less certain (Tomson et al., 1998; Ronkainen et al., 2005; Harnod et al., 2009; Mukherjee et al., 2009). Some studies, however, have suggested an enhanced sympathetic tone or a relative increase in sympathetic activity (via a decrease of parasym-
279 pathetic tone) in chronic epilepsy patients and maybe in SUDEP patients (Diehl et al., 1997; Hilz et al., 2002; Eppinger et al., 2004; Harnod et al., 2009; Mukherjee et al., 2009). SUDEP occurs more frequently during sleep (Kloster and Engelskjøn, 1999; Monté et al., 2007). As an increase in cardiac sympathetic influence was shown to increase the risk of sudden death (Routledge et al., 2002), an enhanced sympathetic tone during sleep could accentuate potentially pro-arrhythmogenic effects (Tisdale et al., 1995), thereby increasing the risk of life-threatening cardiac events. Furthermore, a chronic increase in nocturnal sympathetic tone might lead to long-term modifications in sympathetic cardiac innervations (Druschky et al., 2001), thereby enhancing the risk of cardiac repolarization abnormalities. We did, however, not observe unequivocal alterations of HRV in medically refractory epilepsy patients who died from SUDEP. We have to stress that this assessment was designed with the aim to identify HRV-related predictors for SUDEP in medically refractory epilepsy patients. Because of the limited numbers, we cannot rule out that there might be small differences in HRV between epilepsy patients who will die of SUDEP and those who will not. Likewise, our finding of no significant difference in HRV parameters in medically refractory patients does not exclude the possibility that an enhanced relative sympathetic influence contributes to risk of SUDEP in other people.
Conflict of interests None of the authors has competing interests.
Acknowledgements This work was undertaken at UCLH/UCL which receives a proportion of funding from the Department of Health’s NIHR Biomedical Research Centres funding scheme. RS was supported by a stipend from the Deutsche Forschungsgemeinschaft. The work was supported by the Human Frontier Science Programme and the Wellcome Trust (UK).
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.eplepsyres. 2009.08.008.
References Adjei, P., Surges, R., Scott, C.A., Kallis, C., Shorvon, S., Walker, M.C., 2009. Do subclinical electrographic seizure patterns affect heart rate and its variability? Epilepsy Res. 87, 281—285. Diehl, B., Diehl, R.R., Stodieck, S.R., Ringelstein, E.B., 1997. Spontaneous oscillations in cerebral blood flow velocities in middle cerebral arteries in control subjects and patients with epilepsy. Stroke 28, 2457—2459. Druschky, A., Hilz, M.J., Hopp, P., Platsch, G., Radespiel-Tröger, M., Druschky, K., Kuwert, T., Stefan, H., Neundörfer, B., 2001. Interictal cardiac autonomic dysfunction in temporal lobe epilepsy demonstrated by [(123)I]metaiodobenzylguanidine-SPECT. Brain 124, 2372—2382.
280 Eppinger, N., Schaumann, R., Jokeit, H., Buettner, U.W., Kraemer, G., 2004. Reduced heart rate variability (HRV) in victims of sudden death in epilepsy (SUDEP). Epilepsia 45 (Suppl. 3), 65. Hallioglu, O., Okuyaz, C., Mert, E., Makharoblidze, K., 2008. Effects of antiepileptic drug therapy on heart rate variability in children with epilepsy. Epilepsy Res. 79, 49—54. Harnod, T., Yang, C.C., Hsin, Y.L., Wang, P.J., Shieh, K.R., Kuo, T.B., 2009. Heart rate variability in patients with frontal lobe epilepsy. Seizure 18, 21—25. Hennessy, M.J., Tighe, M.G., Binnie, C.D., Nashef, L., 2001. Sudden withdrawal of carbamazepine increases cardiac sympathetic activity in sleep. Neurology 57, 1650—1654. Hilz, M.J., Devinsky, O., Doyle, W., Mauerer, A., Dütsch, M., 2002. Decrease of sympathetic cardiovascular modulation after temporal lobe epilepsy surgery. Brain 125, 985—995. Kennebäck, G., Ericson, M., Tomson, T., Bergfeldt, L., 1997. Changes in arrhythmia profile and heart rate variability during abrupt withdrawal of antiepileptic drugs. Implications for sudden death. Seizure 6, 369—375. Kloster, R., Engelskjøn, T., 1999. Sudden unexpected death in epilepsy (SUDEP): a clinical perspective and a search for risk factors. J. Neurol. Neurosurg. Psychiatry 67, 439—444. Lossius, M.I., Erikssen, J.E., Mowinckel, P., Gulbrandsen, P., Gjerstad, L., 2007. Changes in autonomic cardiac control in patients with epilepsy after discontinuation of antiepileptic drugs: a randomized controlled withdrawal study. Eur. J. Neurol. 14, 1022—1028. Monté, C.P., Arends, J.B., Tan, I.Y., Aldenkamp, A.P., Limburg, M., de Krom, M.C., 2007. Sudden unexpected death in epilepsy patients: risk factors. A systematic review. Seizure 16, 1—7. Mukherjee, S., Tripathi, M., Chandra, P.S., Yadav, R., Choudhary, N., Sagar, R., Bhore, R., Pandey, R.M., Deepak, K.K., 2009. Cardio-
R. Surges et al. vascular autonomic functions in well-controlled and intractable partial epilepsies. Epilepsy Res. 85, 261—269. Persson, H., Ericson, M., Tomson, T., 2003. Carbamazepine affects autonomic cardiac control in patients with newly diagnosed epilepsy. Epilepsy Res. 57, 69—75. Ronkainen, E., Ansakorpi, H., Huikuri, H.V., Myllylä, V.V., Isojärvi, J.I., Korpelainen, J.T., 2005. Suppressed circadian heart rate dynamics in temporal lobe epilepsy. J. Neurol. Neurosurg. Psychiatry 76, 1382—1386. Routledge, H.C., Chowdhary, S., Townend, J.N., 2002. Heart rate variability—–a therapeutic target? J. Clin. Pharm. Ther. 27, 85—92. Stein, P.K., Kleiger, R.E., 1999. Insights from the study of heart rate variability. Annu. Rev. Med. 50, 249—261. Surges, R., Adjei, P., Kallis, C., Erhuero, J., Scott, C.A., Bell, G.S., Sander, J.W., Walker, M.C., in press. Pathological cardiac repolarization in pharmacoresistant epilepsy and its potential role in sudden unexpected death in epilepsy: a case—control study. Epilepsia. Tisdale, J.E., Patel, R., Webb, C.R., Borzak, S., Zarowitz, B.J., 1995. Electrophysiologic and proarrhythmic effects of intravenous inotropic agents. Prog. Cardiovasc. Dis. 38, 167—180. Tomson, T., Ericson, M., Ihrman, C., Lindblad, L.E., 1998. Heart rate variability in patients with epilepsy. Epilepsy Res. 30, 77—83. Umetani, K., Singer, D.H., McCraty, R., Atkinson, M., 1998. Twentyfour hour time domain heart rate variability and heart rate: relations to age and gender over nine decades. J. Am. Coll. Cardiol. 31, 593—601. Vaughn, B.V., Quint, S.R., Messenheimer, J.A., Robertson, K.R., 1995. Heart period variability in sleep. Electroencephalogr. Clin. Neurophysiol. 94, 155—162.
journal homepage: www.elsevier.com/locate/epilepsyres
SHORT COMMUNICATION
Do alterations in inter-ictal heart rate variability predict sudden unexpected death in epilepsy? R. Surges a,∗, C. Henneberger a, P. Adjei a, C.A. Scott a, J.W. Sander a,b, M.C. Walker a a b
Department of Clinical and Experimental Epilepsy, UCL Institute of Neurology, Queen Square, London, UK SEIN — Epilepsy Institute of the Netherlands Foundation, Heemstede, The Netherlands
Received 25 June 2009; received in revised form 2 August 2009; accepted 9 August 2009 Available online 10 September 2009
KEYWORDS Heart rate variability; Autonomic dysfunction; Sympathetic tone; Sudden unexpected death in epilepsy; Nocturnal seizures
Summary Reduced heart rate variability (HRV) may predispose to sudden unexpected death in epilepsy (SUDEP). We ascertained whether HRV predicts SUDEP in chronic epilepsy using a case—control design and investigated parameters of inter-ictal HRV in 14 patients (7 had died from SUDEP). No HRV parameter was associated with SUDEP. Thus, although altered HRV might be involved in SUDEP, HRV parameters are not clear-cut predictors for SUDEP. © 2009 Elsevier B.V. All rights reserved.
Introduction Heart rate variability (HRV) is the beat-to-beat variability of successive heartbeats and is modulated by a balanced parasympathetic—sympathetic autonomic activity (Stein and Kleiger, 1999). Impaired HRV is known to be associated with an increased risk of sudden cardiac death in an apparently healthy population (Stein and Kleiger, 1999). Pre-
Abbreviations: AED, antiepileptic drug; FLE, frontal lobe epilepsy; HF, high frequency; HRV, heart rate variability; LF, low frequency; RMSSD, root mean square of successive differences; SD, standard deviation; SDNN, standard deviation of the RR intervals; SUDEP, sudden unexpected death in epilepsy; TLE, temporal lobe epilepsy; VLF, very low frequency. ∗ Corresponding author. Tel.: +44 20 7837 3611x4135; fax: +44 20 7278 5616. E-mail address: [email protected] (R. Surges).
vious studies have demonstrated decreased HRV in chronic epilepsy patients and have suggested that this may play a role in the pathophysiology of SUDEP (Tomson et al., 1998; Ronkainen et al., 2005). Possible mechanisms include an impaired autonomic function with augmented sympathetic tone (Diehl et al., 1997; Hilz et al., 2002). We ascertained inter-ictal HRV in a matched-pair case—control design to identify a potential predictive role of HRV for SUDEP in medically refractory epilepsy patients.
Methods Patients with medically refractory focal epilepsy who underwent standard pre-surgical assessment at the National Hospital for Neurology and Neurosurgery between 1996 and 2004 and who later died of SUDEP were included. Living medically refractory focal epilepsy patients were matched as controls for admission date of video-EEG telemetry, age and gender. At the last follow up all control patients were still alive. This assessment was part of a continuing mortal-
0920-1211/$ — see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2009.08.008
278
R. Surges et al.
ity audit and was approved as such by the Joint Ethics Committee of NHNN and the Institute of Neurology. Video-EEG telemetry was performed using conventional scalp EEG recordings (10—20 system) or intracranial recordings (1 patient) at a sampling rate of 200 Hz. ECG was recorded from 2 channels with a modified lead I (adhesive electrodes placed below the clavicles of either side). Peri-ictal data of 12 patients (no. 1—12, Supplementary Online Material) were included in a previous project (Surges et al., in press). HRV was analysed blinded to the outcome in a 1-h interval (by C.H.) with an in-house written software using Matlab (The MathWorks, Natick, MA, USA). As HRV was previously shown to be different in chronic TLE and control patients with most important differences during night around 4 am and during the afternoon (Ronkainen et al., 2005), HRV was determined during 1-h intervals between 4—5 am and 4—5 pm of the last recording day. Occasional ECG artefacts were removed prior to analysis. RR intervals were determined semi-automatically using the first derivative of the voltage signal and resulting data were visually inspected to ensure accuracy of the algorithm. RR intervals, the standard deviation of the RR intervals (SDNN, in ms) and their root mean square of successive differences (RMSSD, in ms) were calculated. The power spectrum of the RR-interval time series was obtained by a fast Fourier transform after re-sampling (7200 samples). The cumulative power was determined in three frequency bands: very low frequency (VLF, 0.005—0.04 Hz), low frequency (LF, 0.04—0.15 Hz), and high frequency (HF, 0.15—0.4 Hz). LF is primarily a measure of sympathetic activity, whilst HF is primarily a measure
Table 1
of parasympathetic activity. Sympathovagal balance was estimated by calculating LF/HF ratio. Statistics were performed using conditional logistic regression analysis with STATA software (StataCorp LP, TX 77845, USA). P-Values < 0.05 were regarded as statistically significant. The inter-individual variability of HRV parameters is relatively large (Tomson et al., 1998; Ronkainen et al., 2005). Therefore, we assumed that a clear-cut clinical predictor for SUDEP should differ at least by two standard deviations (SD) of the means. Then, with an ˛ of 0.05, we would require at least 6 people in each group to have a 90% chance of detecting this difference. Data values are expressed as mean ± SD.
Results A total of 7 SUDEP and 7 living control patients were assessed (Table 1). Mean age/epilepsy duration were 34.0 ± 7.0 years/23.9 ± 6.8 years in controls and 35.0 ± 7.7 years/23.9 ± 7.5 years in SUDEP patients. Most patients were on more than one antiepileptic drug (AED). AED treatment was reduced in 11 patients during telemetry, but re-instituted to usual doses at least 12—48 h prior to HRV assessment (supplementary table). No statistically significant differences were found in the studied HRV parameters (Table 2).
Clinical characteristics.
Patient
Age/sex/epilepsy duration
Etiology/epilepsy
1/S 2/C
32/F/10 29/F/22
Left TLE with hippocampal sclerosis Cryptogenic left FLE
3/S 4/C
33/F/25 32/F/11
Right TLE epilepsy with postischemic lesion Right TLE with diffuse post-radiation and resection (medulloblastoma) defect
5/S 6/C
36/M/26 32/M/29
Left FLE with gangliocytoma/hamartoma Left TLE with hippocampal sclerosis
7/S 8/C
21/M/19 25/M/20
Cryptogenic left FLE Right TLE with hippocampal sclerosis
9/S 10/C
44/M/26 43/M/26
Right FLE with posttraumatic lesions Cryptogenic left FLE
11/S 12/C
43/M/34 44/M/30
Left TLE with hippocampal sclerosis Bilateral TLE with bilateral hippocampal sclerosis
13/S 14/C
36/F/27 33/F/29
Cryptogenic TLE (probably left) Left TLE with dysembryoplastic neuroepithelial tumor
S, SUDEP patient; C, control patient.
Table 2
HRV parameters. RR (ms)
SDNN (ms)
RMSSD (ms)
Total power (ms2 )
VLF (ms2 )
LF (ms2 )
HF (ms2 )
LF/HF ratio
4—5 am Control SUDEP P-Value
994 ± 153 936 ± 140 0.51
70 ± 25 74 ± 40 0.83
49 ± 26 51 ± 32 0.92
3233 ± 1872 3846 ± 2596 0.67
1545 ± 816 1983 ± 1478 0.59
941 ± 513 1277 ± 802 0.44
747 ± 724 586 ± 507 0.66
2.31 ± 1.46 2.71 ± 0.81 0.50
4—5 pm Control SUDEP P-Value
799 ± 137 823 ± 71 0.72
59 ± 17 67 ± 16 0.45
46 ± 23 38 ± 12 0.53
2318 ± 1132 2319 ± 1186 0.99
905 ± 421 1245 ± 498 0.31
913 ± 456 777 ± 595 0.70
500 ± 449 297 ± 178 0.38
2.78 ± 1.66 2.81 ± 1.17 0.97
Mean ± SD.
Do alterations in inter-ictal HRV predict SUDEP?
Discussion Here, we aimed to identify a potential HRV-related predictor of SUDEP in medically refractory epilepsy patients who underwent pre-surgical evaluation. The inter-individual variability of HRV parameters is relatively large (Tomson et al., 1998; Ronkainen et al., 2005) and we were able to assess only a limited number of patients. Our failure to detect any difference in inter-ictal HRV in SUDEP and control patients could therefore be due to small sample size. Nevertheless, our study was powered to detect a difference of more than two standard deviations from the population mean, a criterion that is frequently taken to indicate clinical significance. Our results could also be biased by some inherent disadvantages of using a retrospective design. We have assessed patients who underwent video-EEG telemetry and who later died suddenly. People with chronic epilepsy with severe alterations of autonomic function might die earlier than those with less severe or little changes in HRV. The former patients have therefore a lower chance of being admitted to video-EEG telemetry. It is also possible, that although poor seizure control is an established risk factor for SUDEP, we might have missed SUDEP patients with relatively wellcontrolled chronic epilepsy, but severe alteration of HRV. There is, however, some evidence that severe alterations of HRV are associated with pharmacoresistance (Mukherjee et al., 2009). Finally, people who died suddenly were matched with living patients with pharmacoresistant focal epilepsy patients. We cannot exclude that these control patients may die of SUDEP in the future, and thus have been misclassified as controls. To reduce inter- and intra-individual heterogeneity of HRV, we controlled and checked for parameters which are known to affect HRV. Patients were matched for age and gender which both alter HRV (Umetani et al., 1998). Other potential confounders include sleep stages and AED treatment (Vaughn et al., 1995). The amount of slow wave and REM sleep was similar in SUDEP and control patients over night (not shown), so that sleep-related changes in HRV are unlikely to influence our study. Studies on AED effects on HRV yielded conflicting results and make it difficult to estimate a potential AED effect (Kennebäck et al., 1997; Hennessy et al., 2001; Persson et al., 2003; Lossius et al., 2007; Hallioglu et al., 2008). A putative effect of AEDs should have only minor influence on the comparison between both SUDEP and control patients, as AED types were relatively equally distributed in both groups and usual treatment was re-instituted 12—48 h prior to assessment of HRV parameters. The differential impact of the epilepsy or lesion type on HRV is unclear, but e.g. a decrease in parasympathetic activity was reported in both TLE and frontal lobe epilepsy (FLE) (Ronkainen et al., 2005; Harnod et al., 2009). Likewise, the influence of side of epileptogenic lesion in TLE has yet to be elucidated in larger studies (Tomson et al., 1998; Adjei et al., in press). An impaired HRV in chronic epilepsy has been well established in various studies, whereas the relative influence of parasympathetic or sympathetic system is less certain (Tomson et al., 1998; Ronkainen et al., 2005; Harnod et al., 2009; Mukherjee et al., 2009). Some studies, however, have suggested an enhanced sympathetic tone or a relative increase in sympathetic activity (via a decrease of parasym-
279 pathetic tone) in chronic epilepsy patients and maybe in SUDEP patients (Diehl et al., 1997; Hilz et al., 2002; Eppinger et al., 2004; Harnod et al., 2009; Mukherjee et al., 2009). SUDEP occurs more frequently during sleep (Kloster and Engelskjøn, 1999; Monté et al., 2007). As an increase in cardiac sympathetic influence was shown to increase the risk of sudden death (Routledge et al., 2002), an enhanced sympathetic tone during sleep could accentuate potentially pro-arrhythmogenic effects (Tisdale et al., 1995), thereby increasing the risk of life-threatening cardiac events. Furthermore, a chronic increase in nocturnal sympathetic tone might lead to long-term modifications in sympathetic cardiac innervations (Druschky et al., 2001), thereby enhancing the risk of cardiac repolarization abnormalities. We did, however, not observe unequivocal alterations of HRV in medically refractory epilepsy patients who died from SUDEP. We have to stress that this assessment was designed with the aim to identify HRV-related predictors for SUDEP in medically refractory epilepsy patients. Because of the limited numbers, we cannot rule out that there might be small differences in HRV between epilepsy patients who will die of SUDEP and those who will not. Likewise, our finding of no significant difference in HRV parameters in medically refractory patients does not exclude the possibility that an enhanced relative sympathetic influence contributes to risk of SUDEP in other people.
Conflict of interests None of the authors has competing interests.
Acknowledgements This work was undertaken at UCLH/UCL which receives a proportion of funding from the Department of Health’s NIHR Biomedical Research Centres funding scheme. RS was supported by a stipend from the Deutsche Forschungsgemeinschaft. The work was supported by the Human Frontier Science Programme and the Wellcome Trust (UK).
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.eplepsyres. 2009.08.008.
References Adjei, P., Surges, R., Scott, C.A., Kallis, C., Shorvon, S., Walker, M.C., 2009. Do subclinical electrographic seizure patterns affect heart rate and its variability? Epilepsy Res. 87, 281—285. Diehl, B., Diehl, R.R., Stodieck, S.R., Ringelstein, E.B., 1997. Spontaneous oscillations in cerebral blood flow velocities in middle cerebral arteries in control subjects and patients with epilepsy. Stroke 28, 2457—2459. Druschky, A., Hilz, M.J., Hopp, P., Platsch, G., Radespiel-Tröger, M., Druschky, K., Kuwert, T., Stefan, H., Neundörfer, B., 2001. Interictal cardiac autonomic dysfunction in temporal lobe epilepsy demonstrated by [(123)I]metaiodobenzylguanidine-SPECT. Brain 124, 2372—2382.
280 Eppinger, N., Schaumann, R., Jokeit, H., Buettner, U.W., Kraemer, G., 2004. Reduced heart rate variability (HRV) in victims of sudden death in epilepsy (SUDEP). Epilepsia 45 (Suppl. 3), 65. Hallioglu, O., Okuyaz, C., Mert, E., Makharoblidze, K., 2008. Effects of antiepileptic drug therapy on heart rate variability in children with epilepsy. Epilepsy Res. 79, 49—54. Harnod, T., Yang, C.C., Hsin, Y.L., Wang, P.J., Shieh, K.R., Kuo, T.B., 2009. Heart rate variability in patients with frontal lobe epilepsy. Seizure 18, 21—25. Hennessy, M.J., Tighe, M.G., Binnie, C.D., Nashef, L., 2001. Sudden withdrawal of carbamazepine increases cardiac sympathetic activity in sleep. Neurology 57, 1650—1654. Hilz, M.J., Devinsky, O., Doyle, W., Mauerer, A., Dütsch, M., 2002. Decrease of sympathetic cardiovascular modulation after temporal lobe epilepsy surgery. Brain 125, 985—995. Kennebäck, G., Ericson, M., Tomson, T., Bergfeldt, L., 1997. Changes in arrhythmia profile and heart rate variability during abrupt withdrawal of antiepileptic drugs. Implications for sudden death. Seizure 6, 369—375. Kloster, R., Engelskjøn, T., 1999. Sudden unexpected death in epilepsy (SUDEP): a clinical perspective and a search for risk factors. J. Neurol. Neurosurg. Psychiatry 67, 439—444. Lossius, M.I., Erikssen, J.E., Mowinckel, P., Gulbrandsen, P., Gjerstad, L., 2007. Changes in autonomic cardiac control in patients with epilepsy after discontinuation of antiepileptic drugs: a randomized controlled withdrawal study. Eur. J. Neurol. 14, 1022—1028. Monté, C.P., Arends, J.B., Tan, I.Y., Aldenkamp, A.P., Limburg, M., de Krom, M.C., 2007. Sudden unexpected death in epilepsy patients: risk factors. A systematic review. Seizure 16, 1—7. Mukherjee, S., Tripathi, M., Chandra, P.S., Yadav, R., Choudhary, N., Sagar, R., Bhore, R., Pandey, R.M., Deepak, K.K., 2009. Cardio-
R. Surges et al. vascular autonomic functions in well-controlled and intractable partial epilepsies. Epilepsy Res. 85, 261—269. Persson, H., Ericson, M., Tomson, T., 2003. Carbamazepine affects autonomic cardiac control in patients with newly diagnosed epilepsy. Epilepsy Res. 57, 69—75. Ronkainen, E., Ansakorpi, H., Huikuri, H.V., Myllylä, V.V., Isojärvi, J.I., Korpelainen, J.T., 2005. Suppressed circadian heart rate dynamics in temporal lobe epilepsy. J. Neurol. Neurosurg. Psychiatry 76, 1382—1386. Routledge, H.C., Chowdhary, S., Townend, J.N., 2002. Heart rate variability—–a therapeutic target? J. Clin. Pharm. Ther. 27, 85—92. Stein, P.K., Kleiger, R.E., 1999. Insights from the study of heart rate variability. Annu. Rev. Med. 50, 249—261. Surges, R., Adjei, P., Kallis, C., Erhuero, J., Scott, C.A., Bell, G.S., Sander, J.W., Walker, M.C., in press. Pathological cardiac repolarization in pharmacoresistant epilepsy and its potential role in sudden unexpected death in epilepsy: a case—control study. Epilepsia. Tisdale, J.E., Patel, R., Webb, C.R., Borzak, S., Zarowitz, B.J., 1995. Electrophysiologic and proarrhythmic effects of intravenous inotropic agents. Prog. Cardiovasc. Dis. 38, 167—180. Tomson, T., Ericson, M., Ihrman, C., Lindblad, L.E., 1998. Heart rate variability in patients with epilepsy. Epilepsy Res. 30, 77—83. Umetani, K., Singer, D.H., McCraty, R., Atkinson, M., 1998. Twentyfour hour time domain heart rate variability and heart rate: relations to age and gender over nine decades. J. Am. Coll. Cardiol. 31, 593—601. Vaughn, B.V., Quint, S.R., Messenheimer, J.A., Robertson, K.R., 1995. Heart period variability in sleep. Electroencephalogr. Clin. Neurophysiol. 94, 155—162.