458. Predicting Outcomes of Rotavirus-Related Disease in Adults-Does the Modified Vesikari Score Work?
Session: Poster Abstract Session: Prevention and Treatment of Viral Infections
Thursday, October 3, 2013
Room: The Moscone Center: Poster Hall C
  • Rotavirus Severity #40935.pdf (161.1 kB)
  • Background: Data are lacking regarding grading the severity of viral gastroenteritis, a common illness in adults.  We assessed the utility of the modified Vesikari score (MVS), a scoring system used in children, and other factors predictive of outcomes in adults with rotavirus.

    Methods: Stools from adults >18 years during Feb – May 2006 – 2011 that were submitted for bacterial stool culture at Northwestern Mem. Hosp., Chicago, were tested retrospectively for rotavirus by EIA and were genotyped by RT-PCR.  Clinical data were abstracted to calculate the MVS (duration of vomiting and diarrhea, maximum number of vomiting episodes and diarrheal episodes/24hrs, and temperature).  The dehydration and treatment components of the scale were not included.  The MVS and other demographic and clinical factors were examined as risk factors for evidence of dehydration (BUN/creatinine (Cr) ratio of ≥20:1, urine spec. gravity ≥1.020, Cr elevation ≥1.5 x baseline), hospitalization, and hospital length of stay (LOS) ≥3 days. Immunocompromised adults were defined as having HIV, cancer, transplant, or chronic steroids. Univariate and multivariate analyses were performed to identify risk factors for outcomes.

    Results: 125 rotavirus-positive adults were identified. The MVS did not predict dehydration (7.29 vs 7.14, p = 0.7), hospitalization (7.37 vs 6.95, p = 0.4), or LOS ≥3 days (7.25 vs 7.19, p = 0.9).  In contrast, older age predicted dehydration (54 vs 40 yrs, p = 0.009) and hospitalization (57 vs 36.5 yrs, p < 0.001).  Older age (63.5 vs 40, p = 0.002) and being immunocompromised (53% vs 23%, p = 0.002) predicted LOS ≥3 days.  Unusual rotavirus genotypes (e.g., G2 and G12) did not correlate with any outcomes.  In multivariate analyses, older age remained predictive of hospitalization, and cancer predictive of LOS ≥3 days.

    Conclusion: The MVS was not predictive of these clinical outcomes in adults with rotavirus, whereas age predicted all outcomes and being immunocompromised predicted LOS ≥3 days. Although confirmatory data are needed, these data challenge the use of the modified Vesikari score in adults for predicting severe outcomes and other markers of disease severity should be considered.

    Evan J. Anderson, MD1, Jacqueline E. Tate, PhD2, Deanna Shippee, B.A.3, Melissa Davila, B.A.3, Bruce Larkin, BS3, Ben Katz, MD3,4, Gary Noskin, MD3, Andi L. Shane, MD, MPH, MSc5, Melissa Weinrobe, B.A.3, Ram Yogev, MD4 and Umesh D. Parashar, MBBS, MPH2, (1)Pediatrics and Medicine, Emory University School of Medicine, Atlanta, GA, (2)Centers for Disease Control and Prevention, Atlanta, GA, (3)Northwestern University Feinberg School of Medicine, Chicago, IL, (4)Infectious Diseases, Ann and Robert H Lurie Children's Hospital of Chicago, Chicago, IL, (5)Division of Pediatric Infectious Diseases, Emory University School of Medicine, Atlanta, GA


    E. J. Anderson, Merck: Grant Investigator, Research grant

    J. E. Tate, None

    D. Shippee, None

    M. Davila, None

    B. Larkin, None

    B. Katz, None

    G. Noskin, None

    A. L. Shane, The Gerber Foundation: Grant Investigator, Research grant
    Medscape, LLC: Speaker's Bureau, Speaker honorarium

    M. Weinrobe, None

    R. Yogev, None

    U. D. Parashar, None

    Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.