- Email: [email protected]
The effect of vagus nerve stimulation therapy on body mass index in children
Epilepsy & Behavior 19 (2010) 50–51
Contents lists available at ScienceDirect
Epilepsy & Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / ye b e h
The effect of vagus nerve stimulation therapy on body mass index in children Sujay Kansagra a, Nour Ataya b, Darrell Lewis a, William Gallentine a, Mohamad A. Mikati a,⁎ a b
Division of Pediatric Neurology, Department of Pediatrics, Duke University Medical Center, Durham, NC, USA Department of Pediatrics, American University of Beirut Medical Center, Beirut, Lebanon
a r t i c l e
i n f o
Article history: Received 8 May 2010 Accepted 14 June 2010 Available online 2 August 2010 Keywords: Epilepsy Vagus nerve stimulator Body mass index Weight
a b s t r a c t The effects of vagus nerve stimulation on weight in individuals with epilepsy are not fully characterized. A retrospective review was performed of all pediatric patients who underwent placement of a vagus nerve stimulator at Duke University Medical Center. Baseline body mass index (BMI) percentile was compared with percentile on follow-up visits. We studied 23 patients who had undergone VNS placement during the period 2001–2009. Baseline BMI percentile was 61.7± 34.3. We had a power of 81% to detect a difference of 20 in BMI percentile from baseline to last follow-up. At the 1-year follow-up (mean = 345 ± 112 days) and last follow-up (mean 4.2 ± 2.4 years) the average BMI percentile was 61.6± 31.88 and 56.09 ± 30.83, respectively. There was no significant difference in BMI percentile as compared with baseline at the 1-year and last follow-up visits (P = 0.992 and 0.681, respectively). Long-term pediatric VNS therapy did not have a major clinically significant effect on BMI percentile during an average follow-up of more than 4 years. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Changes in weight are frequent adverse side effects during treatment of epilepsy with antiepileptic drugs (AEDs). These side effects are at best a nuisance to patients and, at worst, a reason to discontinue the medication completely. Recently, another treatment option for intractable epilepsy, the vagus nerve stimulator (VNS), has been investigated to determine its potential role in weight change. The vagus nerve innervates the proximal stomach, pylorus, and duodenum. In animal studies, vagotomy leads to diminished satiety [1]. Likewise, stimulation of the vagus nerve in canine and rat models has been shown to decrease weight [2–4]. Therefore, the vagus nerve is thought to play a role in mediating hunger. There are reports of both weight loss and no change in weight in adult patients following VNS placement for refractory epilepsy, but these studies looked only at changes in weight as opposed to the more accurate body mass index (BMI) percentile [5,6]. As both weight and BMI undergo continuous changes not only in childhood, but also throughout adulthood, studies that look at only weight or BMI changes are not necessarily informative. There are no studies that examine the effects of VNS on weight in children, nor are there any studies in either adults or children that investigate BMI percentile changes during vagus nerve stimulation. The aim of our study was to
⁎ Corresponding author. Fax: +919 681 8943. E-mail address: [email protected] (M.A. Mikati). 1525-5050/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.yebeh.2010.06.012
determine the long-term effects of VNS placement on BMI percentile in a pediatric population. 2. Methods We performed an institutional review board-approved, retrospective, observational cohort study in which we reviewed the data on all pediatric patients who underwent VNS placement at Duke University Medical Center from 2001 to 2009. Inclusion criteria consisted of placement of a VNS for drug-resistant epilepsy, age between 1 and 21 years, and subsequent follow-up at our center. The following were obtained prior to surgery and at follow-up visits: age, gender, weight, height, and seizure type (partial vs primary generalized). In addition, the numbers of AEDs that reduce and increase weight were recorded at the baseline and last follow-up appointments. On the basis of prior studies, topiramate and zonisamide were counted as AEDs that reduce weight, and valproate, carbamazepine, and gabapentin were considered AEDs that increase weight [7–10]. BMI was calculated and normalized on the basis of established percentiles given age and gender. Baseline BMI percentile was compared with percentile on follow-up visits using a paired sample t test. Power was calculated using the G*Power 3 program (Heinrich Heine University, Düsseldorf, Germany). Bivariate analysis was performed to find variables that may be significantly associated with individual change in BMI percentile at follow-up. Variables studied included age, gender, baseline BMI percentile, number of days to last follow-up visit, number of AEDs that reduce weight at baseline and follow-up, number of AEDs that increase weight at baseline and follow-up, and seizure type. Bivariate analysis
S. Kansagra et al. / Epilepsy & Behavior 19 (2010) 50–51
51
Table 1 Bivariate analysis correlating variables with change in BMI percentile at last follow-up visit. Variable
P value
Baseline BMI percentile Sex Age at baseline (years) Days from baseline to last follow-up appointment Number of AEDs that increase weight at baseline visit Number of AEDs that decrease weight at baseline visit Number of AEDs that increase weight at last follow-up visit Number of AEDs that decrease weight at last follow-up visit Seizure type
0.035a 0.833 0.065 0.058 0.832 0.576 0.096 0.291 0.612
a
Fig. 1. Average (± standard deviation) BMI percentile prior to and after placement of a vagus nerve stimulator.
was performed using the SPSS Version 17 program (SPSS Inc., Chicago, IL, USA). 3. Results Of the 25 children who met the inclusion criteria, 23 patients had documented baseline and follow-up height and weight data that allowed them to be studied (male/female =11/12). Age ranged from 2 to 21 years (mean 8.7 ± 4.7 years). Baseline BMI percentile was 61.7 ±34.3 (n = 23). At follow-up closest to the 1-year time point (mean = 345 ± 112 days) and last follow-up (mean 4.2 ±2.4 years), the average BMI percentile was 61.6 ± 31.88 (n = 20) and 56.09 ± 30.83 (n = 23), respectively (see Fig. 1). There was no significant difference in BMI percentile as compared with baseline at the 1-year and last follow-up visits (P = 0.992 and 0.681, respectively). Given the above baseline BMI and SD, we had a power of 81% to detect a difference of 20 in BMI percentile from baseline to last follow-up visit. On bivariate analysis, the BMI at baseline was negatively correlated with change in BMI percentile; that is, the patients with higher baseline BMI percentiles were more likely to lose weight and those with lower baseline BMI percentiles were more likely to gain weight (P = 0.035) (see Table 1). Sex, age at baseline, duration of therapy, seizure type, number of AEDs that decrease weight at baseline, number of AEDs that decrease weight at last follow-up, number of AEDs that increase weight at baseline, and number of AEDs that increase weight at last follow-up were not found to be significantly associated with change in BMI percentile at last follow-up (P values ≥ 0.065). 4. Discussion Animal studies have indicated that stimulation of the vagus nerve leads to decrease in weight, although human studies have been inconsistent [1–6]. This is the first study to examine BMI changes in the pediatric population after VNS placement, as well as the first study to evaluate for weight change using BMI percentile. By expressing BMI as a percentile based on age and gender using a standardized scale, we were able to normalize data to assess for changes following VNS placement. From our results, it does not appear that vagus nerve stimulation is associated with major changes in BMI in pediatric patients. However,
Statistically significant negative correlation.
given our small sample size, and the power we had in our study that allowed detection of only major changes in BMI percentile (20%), we cannot rule out less remarkable changes. Additionally, our research is subject to the biases of an observational study in which changes in antiepileptic medications were introduced and the patients were managed during follow-up according to clinical judgment. Interestingly, despite these limitations, our regression analysis did reveal a statistically significant negative correlation between baseline BMI percentile and BMI percentile at last follow-up. This may represent a regression toward the mean phenomenon. Additional studies with larger numbers of patients for longer periods are needed to comprehensively address this question, given the encouraging results we obtained and the supporting data from animal models. In conclusion, unlike what has been reported with many antiepileptic drugs, we found, in this first study looking at BMI percentile and first study in children, that VNS therapy did not have major clinically significant effects on BMI during follow-up for up to an average of more than 4 years.
Acknowledgments The authors thank Dr. Pedro Weisleder, Dr. Rose-Mary Boustany, Dr. Robert Delong, Dr. Fawn Leigh, Dr. Kevan van Landingham, and Dr. Richard Morse for their care and dedication to the patients in this study.
References [1] Smith GP, Jerome C, Cushin BJ, Eterno R, Simansky KJ. Abdominal vagotomy blocks the satiety effect of cholecystokinin in the rat. Science 1981;213:1036–7. [2] Roslin M, Kurian M. The use of electrical stimulation of the vagus nerve to treat morbid obesity. Epilepsy Behav 2001;2(suppl 2):S11–6. [3] Laskiewicz J, Krolczyk G, Zurowski G, Sobocki J, Matyja A, Thor PJ. Effects of vagal neuromodulation and vagotomy on control of food intake and body weight in rats. J Physiol Pharmacol 2003;54:603–10. [4] Ziomber A, Juszczak K, Kaszuba-Zwoinska J, et al. Magnetically induced vagus nerve stimulation and feeding behavior in rats. J Physiol Pharmacol 2009;60:71–7. [5] Burneo JG, Faught E, Knowlton R, Morawetz R, Kuzniecky R. Weight loss associated with vagus nerve stimulation. Neurology 2002;59:463–4. [6] Koren MS, Holmes MD. Vagus nerve stimulation does not lead to significant changes in body weight in patients with epilepsy. Epilepsy Behav 2006;8:246–9. [7] El Yaman SH, Mroueh SM, Sinno DD, Mikati MA. Long-term patterns of weight changes during topiramate therapy: an observational study. Neurology 2007;69: 310–1. [8] Faught E, Ayala R, Montouris GG, Leppik IE. Randomized controlled trial of zonisamide for the treatment of refractory partial-onset seizures. Neurology 2001;57:1774–9. [9] Easter D, O'Bryan-Tear CG, Verity C. Weight gain with valproate or carbamazepine: a reappraisal. Seizure 1997;6:121–5. [10] DeToledo JC, Toledo C, DeCerce J, Ramsay RE. Changes in body weight with chronic, high-dose gabapentin therapy. Ther Drug Monit 1997;19:394–6.
Contents lists available at ScienceDirect
Epilepsy & Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / ye b e h
The effect of vagus nerve stimulation therapy on body mass index in children Sujay Kansagra a, Nour Ataya b, Darrell Lewis a, William Gallentine a, Mohamad A. Mikati a,⁎ a b
Division of Pediatric Neurology, Department of Pediatrics, Duke University Medical Center, Durham, NC, USA Department of Pediatrics, American University of Beirut Medical Center, Beirut, Lebanon
a r t i c l e
i n f o
Article history: Received 8 May 2010 Accepted 14 June 2010 Available online 2 August 2010 Keywords: Epilepsy Vagus nerve stimulator Body mass index Weight
a b s t r a c t The effects of vagus nerve stimulation on weight in individuals with epilepsy are not fully characterized. A retrospective review was performed of all pediatric patients who underwent placement of a vagus nerve stimulator at Duke University Medical Center. Baseline body mass index (BMI) percentile was compared with percentile on follow-up visits. We studied 23 patients who had undergone VNS placement during the period 2001–2009. Baseline BMI percentile was 61.7± 34.3. We had a power of 81% to detect a difference of 20 in BMI percentile from baseline to last follow-up. At the 1-year follow-up (mean = 345 ± 112 days) and last follow-up (mean 4.2 ± 2.4 years) the average BMI percentile was 61.6± 31.88 and 56.09 ± 30.83, respectively. There was no significant difference in BMI percentile as compared with baseline at the 1-year and last follow-up visits (P = 0.992 and 0.681, respectively). Long-term pediatric VNS therapy did not have a major clinically significant effect on BMI percentile during an average follow-up of more than 4 years. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Changes in weight are frequent adverse side effects during treatment of epilepsy with antiepileptic drugs (AEDs). These side effects are at best a nuisance to patients and, at worst, a reason to discontinue the medication completely. Recently, another treatment option for intractable epilepsy, the vagus nerve stimulator (VNS), has been investigated to determine its potential role in weight change. The vagus nerve innervates the proximal stomach, pylorus, and duodenum. In animal studies, vagotomy leads to diminished satiety [1]. Likewise, stimulation of the vagus nerve in canine and rat models has been shown to decrease weight [2–4]. Therefore, the vagus nerve is thought to play a role in mediating hunger. There are reports of both weight loss and no change in weight in adult patients following VNS placement for refractory epilepsy, but these studies looked only at changes in weight as opposed to the more accurate body mass index (BMI) percentile [5,6]. As both weight and BMI undergo continuous changes not only in childhood, but also throughout adulthood, studies that look at only weight or BMI changes are not necessarily informative. There are no studies that examine the effects of VNS on weight in children, nor are there any studies in either adults or children that investigate BMI percentile changes during vagus nerve stimulation. The aim of our study was to
⁎ Corresponding author. Fax: +919 681 8943. E-mail address: [email protected] (M.A. Mikati). 1525-5050/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.yebeh.2010.06.012
determine the long-term effects of VNS placement on BMI percentile in a pediatric population. 2. Methods We performed an institutional review board-approved, retrospective, observational cohort study in which we reviewed the data on all pediatric patients who underwent VNS placement at Duke University Medical Center from 2001 to 2009. Inclusion criteria consisted of placement of a VNS for drug-resistant epilepsy, age between 1 and 21 years, and subsequent follow-up at our center. The following were obtained prior to surgery and at follow-up visits: age, gender, weight, height, and seizure type (partial vs primary generalized). In addition, the numbers of AEDs that reduce and increase weight were recorded at the baseline and last follow-up appointments. On the basis of prior studies, topiramate and zonisamide were counted as AEDs that reduce weight, and valproate, carbamazepine, and gabapentin were considered AEDs that increase weight [7–10]. BMI was calculated and normalized on the basis of established percentiles given age and gender. Baseline BMI percentile was compared with percentile on follow-up visits using a paired sample t test. Power was calculated using the G*Power 3 program (Heinrich Heine University, Düsseldorf, Germany). Bivariate analysis was performed to find variables that may be significantly associated with individual change in BMI percentile at follow-up. Variables studied included age, gender, baseline BMI percentile, number of days to last follow-up visit, number of AEDs that reduce weight at baseline and follow-up, number of AEDs that increase weight at baseline and follow-up, and seizure type. Bivariate analysis
S. Kansagra et al. / Epilepsy & Behavior 19 (2010) 50–51
51
Table 1 Bivariate analysis correlating variables with change in BMI percentile at last follow-up visit. Variable
P value
Baseline BMI percentile Sex Age at baseline (years) Days from baseline to last follow-up appointment Number of AEDs that increase weight at baseline visit Number of AEDs that decrease weight at baseline visit Number of AEDs that increase weight at last follow-up visit Number of AEDs that decrease weight at last follow-up visit Seizure type
0.035a 0.833 0.065 0.058 0.832 0.576 0.096 0.291 0.612
a
Fig. 1. Average (± standard deviation) BMI percentile prior to and after placement of a vagus nerve stimulator.
was performed using the SPSS Version 17 program (SPSS Inc., Chicago, IL, USA). 3. Results Of the 25 children who met the inclusion criteria, 23 patients had documented baseline and follow-up height and weight data that allowed them to be studied (male/female =11/12). Age ranged from 2 to 21 years (mean 8.7 ± 4.7 years). Baseline BMI percentile was 61.7 ±34.3 (n = 23). At follow-up closest to the 1-year time point (mean = 345 ± 112 days) and last follow-up (mean 4.2 ±2.4 years), the average BMI percentile was 61.6 ± 31.88 (n = 20) and 56.09 ± 30.83 (n = 23), respectively (see Fig. 1). There was no significant difference in BMI percentile as compared with baseline at the 1-year and last follow-up visits (P = 0.992 and 0.681, respectively). Given the above baseline BMI and SD, we had a power of 81% to detect a difference of 20 in BMI percentile from baseline to last follow-up visit. On bivariate analysis, the BMI at baseline was negatively correlated with change in BMI percentile; that is, the patients with higher baseline BMI percentiles were more likely to lose weight and those with lower baseline BMI percentiles were more likely to gain weight (P = 0.035) (see Table 1). Sex, age at baseline, duration of therapy, seizure type, number of AEDs that decrease weight at baseline, number of AEDs that decrease weight at last follow-up, number of AEDs that increase weight at baseline, and number of AEDs that increase weight at last follow-up were not found to be significantly associated with change in BMI percentile at last follow-up (P values ≥ 0.065). 4. Discussion Animal studies have indicated that stimulation of the vagus nerve leads to decrease in weight, although human studies have been inconsistent [1–6]. This is the first study to examine BMI changes in the pediatric population after VNS placement, as well as the first study to evaluate for weight change using BMI percentile. By expressing BMI as a percentile based on age and gender using a standardized scale, we were able to normalize data to assess for changes following VNS placement. From our results, it does not appear that vagus nerve stimulation is associated with major changes in BMI in pediatric patients. However,
Statistically significant negative correlation.
given our small sample size, and the power we had in our study that allowed detection of only major changes in BMI percentile (20%), we cannot rule out less remarkable changes. Additionally, our research is subject to the biases of an observational study in which changes in antiepileptic medications were introduced and the patients were managed during follow-up according to clinical judgment. Interestingly, despite these limitations, our regression analysis did reveal a statistically significant negative correlation between baseline BMI percentile and BMI percentile at last follow-up. This may represent a regression toward the mean phenomenon. Additional studies with larger numbers of patients for longer periods are needed to comprehensively address this question, given the encouraging results we obtained and the supporting data from animal models. In conclusion, unlike what has been reported with many antiepileptic drugs, we found, in this first study looking at BMI percentile and first study in children, that VNS therapy did not have major clinically significant effects on BMI during follow-up for up to an average of more than 4 years.
Acknowledgments The authors thank Dr. Pedro Weisleder, Dr. Rose-Mary Boustany, Dr. Robert Delong, Dr. Fawn Leigh, Dr. Kevan van Landingham, and Dr. Richard Morse for their care and dedication to the patients in this study.
References [1] Smith GP, Jerome C, Cushin BJ, Eterno R, Simansky KJ. Abdominal vagotomy blocks the satiety effect of cholecystokinin in the rat. Science 1981;213:1036–7. [2] Roslin M, Kurian M. The use of electrical stimulation of the vagus nerve to treat morbid obesity. Epilepsy Behav 2001;2(suppl 2):S11–6. [3] Laskiewicz J, Krolczyk G, Zurowski G, Sobocki J, Matyja A, Thor PJ. Effects of vagal neuromodulation and vagotomy on control of food intake and body weight in rats. J Physiol Pharmacol 2003;54:603–10. [4] Ziomber A, Juszczak K, Kaszuba-Zwoinska J, et al. Magnetically induced vagus nerve stimulation and feeding behavior in rats. J Physiol Pharmacol 2009;60:71–7. [5] Burneo JG, Faught E, Knowlton R, Morawetz R, Kuzniecky R. Weight loss associated with vagus nerve stimulation. Neurology 2002;59:463–4. [6] Koren MS, Holmes MD. Vagus nerve stimulation does not lead to significant changes in body weight in patients with epilepsy. Epilepsy Behav 2006;8:246–9. [7] El Yaman SH, Mroueh SM, Sinno DD, Mikati MA. Long-term patterns of weight changes during topiramate therapy: an observational study. Neurology 2007;69: 310–1. [8] Faught E, Ayala R, Montouris GG, Leppik IE. Randomized controlled trial of zonisamide for the treatment of refractory partial-onset seizures. Neurology 2001;57:1774–9. [9] Easter D, O'Bryan-Tear CG, Verity C. Weight gain with valproate or carbamazepine: a reappraisal. Seizure 1997;6:121–5. [10] DeToledo JC, Toledo C, DeCerce J, Ramsay RE. Changes in body weight with chronic, high-dose gabapentin therapy. Ther Drug Monit 1997;19:394–6.