Annual Meeting Abstracts: View

  • (Abst. 1.303), 2017
  • Comparative Effectiveness of ACTH Versus Other Therapies for Infantile Spasms: Counterfactual Calculations and Potential Effect of Practice Change.
  • Authors: Renee Shellhaas, University of Michigan; Kelly G. Knupp, University of Colorado Anschutz Medical Campus; Anup Patel, Nationwide Children's Hospital and The Ohio State University College of Medicine; Iván Sánchez Fernández, Boston Children’s Hospital, Harvard University Medical School, Boston, MA, United States;Hospital Sant Joan de Déu, Universidad de Barcelona, Barcelona, Spain; Tobias Loddenkemper, Boston Children's Hospital; Jason Coryell, Oregon Health Sciences University; Cynthia Keator, Cook Children's Hospital; Catherine J. Chu, MGH/Harvard; Ignacio Valencia, St. Christopher's Children's Hospital; Anne T. Berg, Ann & Robert H. Lurie Children's Hospital of Chicago; John Mytinger, Nationwide Children's Hospital; Shaun A. Hussain, David Geffen School of Medicine at UCLA; UCLA Mattel Children’s Hospital; Nilika Singhal, UCSF Medical School; Wendy Mitchell, University of Southern California; Children's Hospital Los Angeles; Douglas Nordli, Children’s Hospital Los Angeles, Keck School of Medicine of the University of Southern California; Chellamani Harini, Boston Children's Hospital; William D. Gaillard, Children’s National Medical Center, George Washington University School of Medicine and Health Sciences, Washington, DC, United States.; Courtney Wusthoff, Stanford Medical School; Elissa Yozawitz, Montefiore Medical Centre; Elaine C. Wirrell, Mayo Clinic Rochester; and Zachary Grinspan, Weill Cornell Medicine, New York, NY, USA
  • Content:

    Rationale: Effective treatment of infantile spasms (IS) improves cognitive and epilepsy outcomes. ACTH has the strongest evidence of efficacy. However, it has more side effects and is more expensive compared to alternatives. Our objective was to estimate the comparative effectiveness of ACTH versus other treatments, with a focus on the potential effects of changes in practice. Methods: We compared the effectiveness of ACTH to vigabatrin, oral corticosteroids, and other treatments using data from the National Infantile Spasms Consortium (NISC) database from 2012 – 2016. NISC contains prospectively collected clinical data from 23 US academic medical centers. Successful response required (1) no second medication for IS for 60 days, and (2) no spasms for 30 days after completing 30 days of treatment. We adjusted for selection bias via propensity score methods (weighting by odds of treatment), and accounted for potential medical center effects via generalized estimating equations. Adjustment variables included demographics (age at spasms onset, sex, race, ethnicity, insurance, distance from center, gestational age at birth), epilepsy history (etiology, prior non-spasm seizures, prior meds, time from spasms onset to treatment), neurologic history (head circumference, developmental delay, regression), and investigations (MRI, EEG). We did not adjust for dosing regimen in this analysis. We used these methods to estimate the counterfactual question: "What would happen if this infant had been treated with ACTH instead of the given medication?" Results: There were 358 infants with IS. The majority (312; 87%) received standard treatment (150 ACTH, 71 vigabatrin, and 91 oral corticosteroids). The remainder (46; 13%) were treated with nonstandard treatments (25 topiramate, 8 levetiracetam, 6 clobazam, 3 zonisamide, 2 ketogenic diet, 1 oxcarbazepine, and 1 rufinamide). Bivariate analyses found that infants were more likely to receive ACTH if the epilepsy etiology was unknown, the MRI was normal, the EEG showed hypsarrhythmia, or the infant was not already on an AED (Table).If infants who received nonstandard treatments had instead received ACTH, the response rate would improve from the observed 5% to an estimated 32% [95% CI 10, 67]. There was not a significant estimated difference for switching from vigabatrin to ACTH (observed 29% to estimated 37% [19, 60]) nor for switching from oral steroids to ACTH (observed 46% to estimated 44% [35, 54]) (Table). Conclusions: Infants with IS continue to receive initial treatment with nonstandard medications, despite limited effectiveness. For infants currently treated with nonstandard therapy, changing treatment to ACTH would lead to one additional responder for every four infants with IS.The population effect of practice change could be estimated if there were data on the pharmaco-epidemiology of IS treatment outside of academic centers. The present analysis is limited in that it does not account for differences in dosing regimens. Funding: Funded by The Pediatric Epilepsy Research Foundation.
  • Figures:
  • Figure 1