Neonatal seizures in the U.S. National Inpatient Population: Prevalence and outcomes

Pediatrics and Neonatology (2020) 61, 300e305 Available online at www.sciencedirect.com ScienceDirect journal homepage: http://www.pediatr-neonatol.com Original Article Neonatal seizures in the U.S. National Inpatient Population: Prevalence and outcomes Swetha Padiyar a,b,*, Leen Nusairat b, Amer Kadri b, Jalal Abu-Shaweesh b, Hany Aly b a Department of Neonatology, Metrohealth Medical Center/Case Western Reserve University School of Medicine, Cleveland, OH, USA b Cleveland Clinic Children’s, Cleveland, OH, USA Received Jun 22, 2019; received in revised form Oct 15, 2019; accepted Dec 20, 2019 Available online 28 December 2019 Key Words mortality; neonatal convulsions; neonatal ICU; neonatal seizures; prematurity Background: Seizures are the most frequent neurological manifestation in neonates. Prevalence of neonatal seizures has not been well described in relationship with gestational age (GA). Also, the impact of seizures on neonatal mortality has not been quantified. This study aims to determine 1) prevalence of neonatal seizures in all GA groups, 2) associated mortality in each GA group and 3) impact of seizures on length of stay (LOS) of survivors in each GA group. Methods: Data from the national Kids’ Inpatient Database (KID) for the years 2006, 2009 and 2012 was used in the study. All admitted infants with a documented GA were included in the study. All categorical variables were analyzed using Chi-square test, continuous variables were analyzed using t-test, and logistic regression analysis used to calculate odds ratio (OR) and 95% confidence intervals (CI). Results: A total of 10, 572,209 infants were included, of whom 4400 infants (0.04%) had seizures. The highest prevalence was at 24 weeks (0.12%). Overall mortality rate of patients with seizures was 4% with OR Z 2.24 (95% CI Z 1.90e2.65, p < 0.001). The correlation of seizure with mortality was significant after 33 weeks GA with greatest impact at 33e36 weeks GA (OR Z 46.38 (95% CI Z 26.86e80.08, p < 0.001). Seizures were associated with increased median LOS from 2 to 4 days (p < 0.001). Conclusion: The prevalence of seizures varies according to gestational age ranging from 0.02% to 0.12%. The highest prevalence is at 24 weeks GA. The greatest impact for seizures on mortality is at 33e36 weeks GA. * Corresponding author. Department of Neonatology, Metrohealth Medical Center, 2500 Metrohealth Drive, Cleveland, OH, 44109, USA. Fax: þ216 778 3252. E-mail address: swetha.padiyar@gmail.com (S. Padiyar). https://doi.org/10.1016/j.pedneo.2019.12.006 1875-9572/Copyright ª 2020, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BYNC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Neonatal seizures in US National Inpatient Population 301 Copyright ª 2020, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/ by-nc-nd/4.0/). 1. Introduction 2. Materials and methods Seizures in the newborn represent the most distinctive frequent manifestations of neurological dysfunction in the neonatal period. Neonatal seizures are clinically described as abnormal, stereotyped and paroxysmal dysfunctions in the central nervous system (CNS) occurring within the first 28 days after birth in full-term infants or before 44 weeks of gestational age in preterm infants.1 It is a relatively common occurrence in neonates with a heterogeneous etiology and most of the infants are acutely symptomatic. The etiologies are different in term and preterm infants. Hypoxicischemic encephalopathy (HIE), stroke, cerebral malformations and metabolic disorders are major causes of seizures in term infants. In preterm infants some of the major causes are intraventricular hemorrhage (IVH), periventricular leukomalacia (PVL) and infections.2,3 Seizures occur more often in the neonatal period especially in the first week of life than in other periods in the life span.4 It is estimated to occur in 0.1%e0.5% of newborns. Incidence is 1.5e3.5/1000 in full term infants and 10e130/1000 in preterm infants.5 The incidence increases as gestational age (GA) and birth weight of the infant decrease,6e10 and it is most common in the extremely low birth weight (ELBW) infant and varies from 9 to 58 per 1000 very low birth weight (VLBW) infants8e11 to 64 per 1000 ELBW infants.12 There are two gaps of knowledge in these reports: 1) the reports on seizure incidence did not specify a gestational age specific incidence rate; and 2) more importantly, they were based on local experiences and not necessarily representing national statistics. This study will address these two above-mentioned points. Factors that contribute to this high incidence include the neonate’s brain being more prone to seizures due to maturational factors, late gestational and birth-related injuries. Recognition of neonatal seizures is crucial and difficult because the signs can be subtle and/or absent. Neonatal seizures differ in their appearance, EEG characteristics, management and outcomes. Prognosis of seizures depends on the etiology. Various studies have shown that prolonged seizures lead to poor neurodevelopmental outcomes. Neonatal seizures are associated with unfavorable short- and long-term outcomes. More than half the survivors have considerable disability across a range of developmental domains, most frequently cerebral palsy, postneonatal epilepsy, and/or intellectual disability requiring lifelong therapies and social and financial support.13 The aim of this population-based study was to determine the prevalence and trend of neonatal seizures across all GA group neonates obtained from inpatient care database. We were also investigating the associated mortality with neonatal seizures in each GA group and the impact of seizures on length of stay (LOS) in each GA group. 2.1. Study design In this descriptive cohort study, we used the information from the National Kids’ Inpatient Database (KID). The KID is the only all-payer pediatric inpatient care database in the United States, containing data from two to three million hospital stays. It is published once in every three years. In this study, the KID data for the years 2006, 2009 and 2012 was used. Its large sample size is ideal for developing national and regional estimates and enables analyses of rare conditions, such as congenital anomalies, as well as uncommon treatments, such as organ transplants. The study included all infants admitted to the hospitals in the years 2006, 2009 and 2012 with a documented GA. A total of 10,572,209 infants were included in the study. Infants with undocumented GA, anencephaly, spina bifida and other congenital anomalies of the CNS were excluded from the study. Also, infants who were transferred to regional hospitals from tertiary centers were excluded to avoid duplication. The study included surviving and non-surviving babies. Information on neonatal seizures was obtained from the database by definition of seizures based on ICD-9 codes. 2.2. Statistical analysis Descriptive analysis was used for presenting the results, including mean  standard deviation (SD) for quantitative variables and frequency (percentage) for categorical variables. All categorical variables were analyzed using Chisquare test and continuous variables were analyzed using ttest. Logistic regression analyses were used to calculate adjusted odds ratio (aOR) and 95% confidence interval (CI) after controlling for factors that were significant in bivariate analyses. For the statistical analysis, the statistical software SPSS software, version 16.0 for Windows (SPSS Inc, Chicago, IL, USA) was used. P values of 0.05 or less were considered statistically significant. 3. Results Of the total of 21,604,237 admitted infants, 10,572,209 infants born <24 weeks to >38 weeks GA were included in the study (Fig. 1). A total of 4400 study infants (0.04%) from the 10,572,209 infants included in the study had neonatal seizures (Fig. 1). Infants with neonatal seizures had lower GA with prevalence of seizures decreasing as GA increases (Table 1). Highest prevalence of neonatal seizures (0.12%) was seen in infants at completed 24 weeks GA (Table 2) (Fig. 2). The overall mortality rate of patients with seizures was 4%. The highest mortality was seen in neonates 302 S. Padiyar et al Figure 1 Flow diagram of the study population. <24 weeks GA at 39.4% and mortality rate associated with neonatal seizures decreased as GA increased except for infants at 33e36 weeks completed GA group, which had a mortality rate of 11.9% (Table 3) (Fig. 2). Overall Mortality rate OR was 13.19 (95% CI Z 11.350e15.343, p < 0.001) (Table 3). After adjustment was made for significant factors (race, gender, birth asphyxia, meconium aspiration, respiratory distress syndrome (RDS), chorioamnionitis, necrotizing enterocolitis, sepsis, persistent pulmonary hypertension, IVH, mechanical ventilation) the aOR for mortality was 2.24 (95% CI Z 1.906e2.652, p < 0.001) (Table 3). The correlation of neonatal seizures with mortality was significant starting at 24 weeks gestation prior to adjustment for above-mentioned factors. After adjustment was made, correlation with mortality was significant beyond 33 weeks with the greatest impact at 33e36 weeks completed GA with aOR Z 8.535 (95% CI Z 4.527e16.094, p < 0.001) (Table 3). Median LOS for infants without seizures was 2 days vs. 4 days for infants with seizures (Table 2). Seizures were associated with increased LOS from 2 to 4 days (p < 0.001) (Table 2). 4. Discussion This study evaluated the prevalence of neonatal seizures between 2006 and 2012 from the KID Inpatient Database. In this large population-based cohort study, we report that neonatal seizures: (1) occurred in 4 in every 10,000 neonates born at < 24 weeks to > 38 weeks gestation; (2) were associated with an overall mortality rate of 4%. Of note, this mortality is all-cause mortality and not seizurespecific mortality; (3) the greatest impact of seizures on mortality occurs at 33e36 weeks gestational age; and (4) seizures were associated with a median LOS from 2 to 4 days. The incidence of neonatal seizures of very preterm infants was reported to be different than those in term infants. A review article from Vasudevan and Levene reported the incidence of neonatal seizures in term infants was approximately 1e3 per 1000 infants, whereas the incidence in very low birth weight or preterm infants was suspected to be approximately 10 times higher.11 This study found an overall incidence of neonatal seizures of 0.4/1000 live births with higher rates in preterm infants ranging from 0.2 to 1/1000 live births as compared to 0.4/ 1000 live births in term infants. Other population-based studies have reported higher rates than ours. In a retrospective cohort study on neonates born to residents of Fayette County in Kentucky, USA, a crude rate of neonatal seizures was reported as 3.5 per 1000 live births.14 In this study, seizures were identified through clinical details or an EEG record. In a US nationwide study based on International Classification of Diseases, ninth revision, clinical modification (ICD-9-CM) codes in medical records, an incidence of 2.84/1000 live births was found. The drawback of this study was that the validity of ICD codes was not tested and several nonepileptic episodes could have been misdiagnosed as neonatal seizures.8 Another study done in the state of California, USA based on ICD-9-CM codes during birth admissions reported an incidence of 0.95/1000 live births.13 The prevalence rate reported in this present study is in line with the above-mentioned study. In comparison to population based studies, clinical-based studies tend to find a higher incidence of neonatal seizures because infants included in these studies are affected by CNS injuries which would predispose them to seizures.2,7,15e24 Not many studies include preterm infants but when considered, a higher incidence of seizures is reported in this particular subset of patients10,18e21,25 and for infants with low birth weight.8e10,20,25 In this Neonatal seizures in US National Inpatient Population Table 1 303 Characteristics of the study population and associated diagnosis affecting prevalence of neonatal seizures. Female sex Gestational age - 24 weeks - 25e28 weeks - 29e32 weeks - 33e36 weeks - 37 weeks Race (%) - White - Black - Hispanic - Other/Asian or Pacific Islander/Native American/missing Birth asphyxia Meconium aspiration Respiratory distress syndrome (RDS) Chorioamnionitis Necrotizing enterocolitis (NEC) Sepsis Persistent pulmonary hypertension Intraventricular hemorrhage (IVH) Mechanical ventilation - None - Noninvasive - Invasive Total N Z 10,572,209 Seizure present N Z 4440 (0.04%) Seizure absent N Z 10,567,769 Pvalue 5,133,006 (48.6%) 2018 (45.5%) 5,130,988 (48.6%) <0.001 <0.001 26,502 (0.3%) 51,684 (0.5%) 127,130 (1.2%) 641,904 (6.1%) 9,724,990 (92.0%) 33 (0.7%) 57 (1.3%) 47 (1.1%) 126 (2.8%) 4178 (94.1%) 26,469 (0.3%) 51,627 (0.5%) 127,083 (1.2%) 641,778 (6.1%) 9,720,812 (92.0%) 4,587,145 1,127,155 1,736,930 3,120,979 1791 (40.3%) 662 (14.9%) 649 (14.6%) 1339 (30.2%) 4,585,354 1,126,493 1,736,281 3,119,640 10,603 (0.1%) 40,394 (0.4%) 221,980 (2.1%) 21,102 (0.2%) 8506 (0.1%) 232,430 (2.2%) 25,788 (0.2%) 33,093 (0.3%) 85 (1.9%) 41 (0.9%) 176 (4.0%) 16 (0.4%) 26 (0.6%) 650 (14.6%) 77 (1.7%) 147 (3.3%) 10,518 (0.1%) 40,353 (0.4%) 221,804 (2.1%) 21,086 (0.2%) 8480 (0.1%) 231,780 (2.2%) 25,711 (0.2%) 32,946 (0.3%) 10,204,402 (96.5%) 128,024 (1.2%) 239,784 (2.3%) 3295 (74.2%) 58 (1.3%) 1087 (24.5%) 10,201,107 (96.5%) 127,966 (1.2%) 238,697 (2.3%) <0.001 (43.4%) (10.7%) (16.4%) (29.5%) sample population, prevalence of neonatal seizures increased with a decrease in gestational age at birth with the highest prevalence occurring at  24 weeks completed gestational age (12/10,000 live births). This finding is consistent with several other studies8e10,13,25,26 and confirms the increased risk of neonatal seizures in preterm infants as their immature brain is prone to seizures in response to any kind of injury. The data in this study showed that mortality associated with neonatal seizures ranged from 3.3% to 39.4% with an overall mortality rate of 4%. Highest mortality rate was Table 2 (43.4%) (10.7%) (16.4%) (29.5%) <0.001 <0.001 <0.001 0.026 <0.001 <0.001 <0.001 <0.001 <0.001 observed in infants born at < 24 weeks gestational age at 39.4% (p < 0.001) (Table 2). Although some studies have reported a higher mortality rate than we found in this study, the variations may be explained by differences in study populations, outcomes and diagnosis of seizures. Scher et al. reported a mortality of 58% in preterm infants with EEG-confirmed neonatal seizures.2 Ronen et al. studied 26 preterm infants with neonatal seizures and reported mortality of 42%.25 Pisani et al. evaluated 51 preterm infants with EEG confirmed seizures and found a mortality rate of 34%,27 which is similar to our study’s results of Prevalence of seizure and length of stay according to gestational age. Prevalence rate of seizure (%) Seizure present LOSa Seizure absent LOSa Pvalue 24 weeks 0.123 58 (0e108) 0.446 2528 weeks 2932 weeks 3336 weeks 37 weeks Total 0.110 61 (22e81) 0.075 28 (17e40) 0.036 0.020 97 (34 e126) 81 (51 e105) 44 (18 e107) 7 (2e25) 4 (2e10) 0.108 0.043 0.042 4 (2e10) 4 (2e11) 2 (2e3) 2 (2e3) <0.001 <0.001 a 0.037 Analysis included only infants who survived through hospital discharge. Data are presented in median (interquartile range). 304 S. Padiyar et al 0.3% 48 36 24 12 0.0% 0 < 24 weeks 25-28 weeks 29-32 weeks Seizure prevalence rate 33-36 weeks ≥37 weeks Seizure mortality Odds' ra o Figure 2 Odds’ ratio (OR) for seizure mortality increased at 25e28 weeks gestational age (GA) and highest OR was at 33e36 weeks GA. Seizure mortality OR was not significant in infants <33 weeks GA. mortality rate of 27.3%. However, when their study was limited to infants born <30 weeks gestational age, the mortality rate increased to 41%. In the present study, we found the correlation of seizure with mortality was significant starting at 24 weeks of gestation with greatest impact at gestational age of 33e36 weeks (OR Z 46.38 (95% CI Z 26.862e80.082, p < 0.001) (Table 2). Particularly, the data shows that higher seizure burden is associated with a higher mortality and increased length of hospital stay. Seizures were associated with increased median length of stay from 2 days to 4 days (Table 2). Hence it is very important to detect neonatal seizures early and characterize the seizures with potential for improving the overall outcome including long-term neuro-developmental outcomes and better seizure control. Table 3 This study has strengths and limitations. A major strength is that this is a large population-based cohort study and our sample represents the general population and encompasses the full spectrum of neonatal seizures and minimizes selection bias. Secondly, the study investigated a combination of demographic and clinical details of neonatal seizures. However, the current study has some limitations as well. One of the limitations is that the data for the study relied on ICD-codes that are not necessarily inclusive for all diagnoses of neonatal seizures. The study did not offer other diagnoses that could have caused seizures. Therefore, it is not clear if all types of seizures would have the same outcome. Another limitation is that the study does not classify the seizure types in different gestational ages and associated outcomes. Patients with Prevalence and multivariable logistic regression of seizure and mortality according to gestational age. Gestational Age (GA) Mortality rate with seizures Mortality rate without seizures P- value Unadjusted seizure Mortality OR (95% CI) Pvalue Adjusted seizure Mortality aOR (95% CI)a Pvalue 24 weeks 13 (39.4%) 0.004 0.004 3 (6.4%) 1977 (1.6%) 0.037 15 (11.9%) 1828 (0.3%) <0.001 0.981 (0.458 e2.101) 1.933 (0.972 e3.843) 1.827 (0.516 e6.465) 8.535 (4.527 e16.094) 0.961 11 (19.3%) 37 weeks 136 (3.3%) 8192 (0.1%) <0.001 Total 177 (4.0%) 33,161 (0.3%) <0.001 0.364 (0.180 e0.737) 2.562 (1.325 e4.952) 4.341 (1.350 e13.961) 46.381 (26.862 e80.082) 39.836 (33.528 e47.331) 13.196 (11.350 e15.343) 0.005 2528 weeks 2932 weeks 3336 weeks 16,771 (63.4%) 4393 (8.5%) a 0.005 0.014 <0.001 0.060 0.350 <0.001 <0.001 2.957 (2.441 e3.581) <0.001 <0.001 2.248 (1.906 e2.652) <0.001 Adjustment was made for significant factors (race, gender, birth asphyxia, meconium aspiration, RDS, chorioamnionitis, necrotizing enterocolitis, sepsis, persistent pulmonary hypertension, IVH, mechanical ventilation). Neonatal seizures in US National Inpatient Population diseases in the exclusion criteria are associated with increased occurrence of neonatal seizures and including these patients might provide more accurate analysis. However, for practical and clinical purposes we selected the exclusion criteria to help clinicians estimate the impact of seizures in premature neonates to help facilitate counseling of parents and considering outcomes in caring for sick premature infants. 5. Conclusion Neonatal seizures are common in the newborn period and can be associated with significant morbidity and mortality. Prevalence of neonatal seizures varies by gestational age ranging from 0.02% to 0.12% and is inversely proportional to gestational age. The greatest impact of seizures on mortality is at 33e36 weeks of gestation. It is interesting to report that neonatal seizures were not associated with significant increase in mortality in premature infants with GA < 24 weeks, possibly because these infants have the highest mortality regardless of seizures. To the best of our knowledge, this is the first study to report such finding. It is important to consider these findings when counseling parents of premature infants and considering outcomes and/or futility of care in sick premature infants. Further studies are needed to assess the contributions of different diagnoses to the incidence of seizures in neonates. Declaration of Competing Interest The authors have no conflicts of interest relevant to this article. References 1. Han JY, Moon CJ, Youn YA, Sung IK, Lee IG. Efficacy of levetiracetam for neonatal seizures in preterm infants. BMC Pediatr 2018;18:131. 2. Scher MS, Aso K, Beggarly ME, Hamid MY, Steppe DA, Painter MJ. Electrographic seizures in preterm and full-term neonates: clinical correlates, associated brain lesions, and risk for neurologic sequelae. Pediatrics 1993;91:128e34. 3. Kohelet D, Shochat R, Lusky A, Reichman B, Israel Neonatal Network. Risk factors for seizures in very low birthweight infants with periventricular leukomalacia. J Child Neurol 2006; 21:965e70. 4. Annegers JF, Hauser WA, Lee JR, Rocca WA. Incidence of acute symptomatic seizures in Rochester, Minnesota, 1935-1984. Epilepsia 1995;36:327e33. 5. Bassan H, Bental Y, Shany E, Berger I, Froom P, Levi L, et al. Neonatal seizures: dilemmas in workup and management. Pediatr Neurol 2008;38:415e21. 6. Iwami H, Isayama T, Lodha A, Canning R, Abou Mehrem A, Lee SK, et al. Outcomes after neonatal seizures in infants less than 29 weeks’ gestation: a population-based cohort study. Am J Perinatol 2019;36:191e9. 305 7. Holden KR, Mellits ED, Freeman JM. Neonatal seizures. I. Correlation of prenatal and perinatal events with outcomes. Pediatrics 1982;70:165e76. 8. Lanska MJ, Lanska DJ. Neonatal seizures in the United States: results of the national hospital discharge survey, 19801991. Neuroepidemiology 1996;15:117e25. 9. Lanska MJ, Lanska DJ, Baumann RJ, Kryscio RJ. A populationbased study of neonatal seizures in Fayette County, Kentucky. Neurology 1995;45:724e32. 10. Saliba RM, Annegers JF, Waller DK, Tyson JE, Mizrahi EM. Incidence of neonatal seizures in Harris county, Texas, 19921994. Am J Epidemiol 1999;150:763e9. 11. Vasudevan C, Levene M. Epidemiology and aetiology of neonatal seizures. Semin Fetal Neonatal Med 2013;18:185e91. 12. Davis AS, Hintz SR, Van Meurs KP, Li L, Das A, Stoll BJ, et al. Seizures in extremely low birth weight infants are associated with adverse outcome. J Pediatr 2010;157:720e5. e12. 13. Glass HC, Shellhaas RA, Wusthoff CJ, Chang T, Abend NS, Chu CJ, et al. Contemporary profile of seizures in neonates: a prospective cohort study. J Pediatr 2016;174:98e103. e1. 14. Pisani F, Facini C, Pelosi A, Mazzotta S, Spagnoli C, Pavlidis E. Neonatal seizures in preterm newborns: a predictive model for outcome. Eur J Paediatr Neurol 2016;20:243e51. 15. Dennis J. Neonatal convulsions: aetiology, late neonatal status and long-term outcome. Dev Med Child Neurol 1978;20: 143e8. 16. Goldberg HJ. Neonatal convulsions -a 10 year review. Arch Dis Child 1983;58:976e8. 17. Tudehope DI, Harris A, Hawes D, Hayes M. Clinical spectrum and outcome of neonatal convulsions. Aust Paediatr J 1988;24: 249e53. 18. Airede KI. Neonatal seizures and a 2-year neurological outcome. J Trop Pediatr 1991;37:313e7. 19. Sheth RD, Hobbs GR, Mullett M. Neonatal seizures: incidence, onset, and etiology by gestational age. J Perinatol 1999;19: 40e3. 20. Kumar A, Gupta A, Talukdar B. Clinico-etiological and EEG profile of neonatal seizures. Indian J Pediatr 2007;74:33e7. 21. Al-Zwaini EJ, Al-Ani MM, Mengal AH. The epidemiology of clinical neonatal seizures in Ramadi city. Neurosciences (Riyadh) 2007;12:170e2. 22. Mwaniki M, Mathenge A, Gwer S, Mturi N, Bauni E, Newton CR, et al. Neonatal seizures in a rural Kenyan district hospital: aetiology, incidence and outcome of hospitalization. BMC Med 2010;8:16. 23. Sadeghian A, Damghanian M, Shariati M. Neonatal seizures in a rural Iranian district hospital: etiologies, incidence and predicting factors. Acta Med Iran 2012;50:760e4. 24. Kuti BP, Oseni SB, Owa JA. Pattern, etiological factors and determinants of mortality among sick newborns with seizures in Ilesa, Nigeria. J Pediatr Neurosci 2015;10:227e34. 25. Ronen GM, Penney S, Andrews W. The epidemiology of clinical neonatal seizures in Newfoundland: a population-based study. J Pediatr 1999;134:71e5. 26. Arpino C, Domizio S, Carrieri MP, Brescianini DS, Sabatino MG, Curatolo P. Prenatal and perinatal determinants of neonatal seizures occurring in the first week of life. J Child Neurol 2001; 16:651e6. 27. Pisani F, Barilli AL, Sisti L, Bevilacqua G, Seri S. Preterm infants with video-EEG confirmed seizures: outcome at 30 months of age. Brain Dev 2008;30:20e30.