Clinical and economic impact of vagus nerve stimulation therapy in patients with
drug-resistant epilepsy
☆
Sandra L. Helmers
a,
⁎
, Mei Sheng Duh
b
, Annie Guérin
b
, Sujata P. Sarda
b
, Thomas M. Samuelson
b
,
Mark T. Bunker
c
, Bryan D. Olin
c
, Stanley D. Jackson
c
, Edward Faught
a
a
Emory University School of Medicine, Atlanta, GA, USA
b
Analysis Group, Inc., Boston, MA, USA
c
Cyberonics, Inc., Houston, TX, USA
abstractarticle info
Article history:
Received 12 May 2011
Revised 13 July 2011
Accepted 17 July 2011
Available online 26 August 2011
Keywords:
Vagus nerve stimulation
Costs
Resource utilization
Drug-resistant epilepsy
Refractory epilepsy
Complex partial seizures
Cost savings
We evaluated long-term medical and economic benefits of vagus nerve stimulation (VNS) therapy in drug-
resistant epilepsy. A pre–post analysis was conducted using multistate Medicaid data (January 1997–June
2009). One thousand six hundred fifty-five patients with one or more neurologist visits with epilepsy
diagnoses (ICD-9 345.xx, 780.3, or 780.39), one or more procedures for vagus nerve stimulator implantation,
one or more antiepileptic drugs (AEDs), and 6 or more months of continuous Medicaid enrollment pre- and
post-VNS were selected. The pre-VNS period was 6 months. The post-VNS period extended from implantation
to device removal, death, Medicaid disenrollment, or study end (up to 3 years). Incidence rate ratios (IRRs)
and cost differences ($2009) were estimated. Mean age was 29.4 years. Hospitalizations decreased post-VNS
compared with pre-VNS (adjusted IRR=0.59, P b 0.001). Grand mal status events decreased post-VNS
compared with pre-VNS (adjusted IRR = 0.79, P b 0.001). Average total health care costs were lower post-VNS
than pre-VNS ($18,550 vs $19,945 quarterly, P b 0.001). VNS is associated with decreased resource utilization
and epilepsy-related clinical events and net cost savings after 1.5 years.
© 2011 Elsevier Inc. All rights reserved.
1. Introduction
Epilepsy is a common neurological disorder affecting approxi-
mately 0.5 to 2% of the population in the United States [1] and imposes
a substantial economic burden on the health care system. Numerous
cost-of-illness studies have quantified the direct and indirect costs
associated with epilepsy in the United States [2–5] and other
countries [6–11]. One study reported that annual direct medical
costs of epilepsy per patient in the United States range from $1620
to $52,558, depending on disease severity [12]. The existing cost
estimates probably underestimate the true costs as most data predate
the year 2000.
Uncontrolled seizures are associated with many detrimental effects,
including cognitive and memory impairment, high rates of depression,
reduced lifetime income, increased health care resource utilization,
higher risk of accidental injuries, and much higher mortality [13–18].
Antiepileptic drugs (AEDs) are the first step toward treating recurrent
seizures. Despite treatment with AEDs, 20 to 30% of patients still
suffer from uncontrolled seizures [19]. A subset of these drug-resistant
patients can have good clinical outcomes from epilepsy surgery
[20]. Unfortunately a substantial number of patients are not good
surgical candidates or are found to be poor surgical candidates after
presurgical evaluation [21]. Such patients may benefitfromvagus
nerve stimulation (VNS), a therapy approved by the US Food and Drug
Administration (FDA) in 1997 as adjunctive treatment for patients
12 years or older with complex partial seizures resistant to AEDs. The
efficacy of VNS (VNS Therapy, Cyberonics, Inc., Houston, TX, USA) in
reducing seizure frequency has been demonstrated in clinical trials
and observational studies [21–25].
Payers are increasingly demanding evidence that the costs of
drugs and technologies are justified by outcomes. Very few studies
have evaluated the effect of VNS on health care utilization and costs.
A previous study reported an annual cost savings of $3000 when
comparing 18 months before and after VNS implantation among 43
patients in Sweden [26]. Another study reported that the average
annual direct medical costs decreased from $4826 to $2496 for 25
patients who underwent VNS in Belgium [27]. The cost estimates in
both studies were reported in 1999 US dollars. In 2007, the average
quarterly resource utilization for 12 months before implantation was
compared with that 48 months after implantation in 138 patients
treated in the United States, and the investigators found that use of
Epilepsy & Behavior 22 (2011) 370–375
☆
Partial results from the study were presented as a poster at the 64th Annual
Meeting of the American Epilepsy Society, San Antonio, TX, December 3–7, 2010, and as
a platform presentation chosen among the top 5% showcased in Scientific Program
Highlights Plenary Session at the 63rd Annual Meeting of the American Academy of
Neurology, Honolulu, HA, April 9–16, 2011.
⁎ Corresponding author at: Emory University School of Medicine, Woodruff
Memorial Research Building, 101 Woodruff Circle, Ste 6000, Atlanta, GA 30322, USA.
Fax: + 1 404 727 3157.
E-mail address: sandra.helmers@emory.edu (S.L. Helmers).
1525-5050/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.yebeh.2011.07.020
Contents lists available at ScienceDirect
Epilepsy & Behavior
journal homepage: www.elsevier.com/locate/yebeh