Differences are measured between participants who had experienced sepsis (n = 16) and all other participants (n = 196). Data points indicate differences; error bars, 95% CI.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Cheung YT, Eskind A, Inaba H, et al. Association of Bacteremic Sepsis With Long-term Neurocognitive Dysfunction in Pediatric Patients With Acute Lymphoblastic Leukemia. JAMA Pediatr. 2018;172(11):1092–1095. doi:10.1001/jamapediatrics.2018.2500
Although long-term survival in children with acute lymphoblastic leukemia (ALL) now exceeds 90%, survivors are at risk for treatment-related neurocognitive dysfunction that can persist into adulthood.1,2 We hypothesized that bacteremic sepsis during treatment for ALL contributes to long-term neurocognitive dysfunction and tested this hypothesis in a cohort study, using a propensity score–weighted model to adjust for potential confounders.3
In a prospective cohort study, survivors of pediatric ALL who had received risk-adapted chemotherapy from July 8, 2000, through November 25, 2010, without hematopoietic cell transplant or cranial irradiation underwent neurocognitive testing a median of 7.7 years after diagnosis of ALL (range, 5.1 to 12.5 years), with final follow-up completed on October 12, 2014.1,2 Episodes of bacteremia during treatment were retrospectively identified from institutional databases, and sepsis during each episode was categorized by review of medical records. Sepsis was defined as bacteremia plus urgent intervention for stabilization or severe sepsis according to consensus definitions.4,5 Because the participants were immunocompromised, systemic inflammatory response syndrome criteria were not obligatory.4 This study was prospectively approved by the institutional review board of St. Jude Children’s Research Hospital, Memphis, Tennessee, and written consent was obtained from participants or their legally authorized representatives.
Each neurocognitive outcome variable, obtained from performance-based research testing, was expressed as an age-adjusted z score against population norms. Propensity for sepsis was estimated with a multinomial logistic regression model, including sex, self-reported race, age at diagnosis of ALL, leukemia risk category, and age at the time of testing. We estimated propensity score–adjusted z scores using generalized linear modeling, with age at diagnosis as a covariate and inverse propensity score weighting and calculated the difference between the mean values for participants with and without sepsis. Statistical tests were performed using SAS software for Windows (version 9.4; SAS Institute, Inc) with α set at 0.05.
A total of 212 pediatric patients with ALL (104 girls [49.1%] and 108 boys [50.9%]; median age, 5.0 years [interquartile range, 3.2-8.8]) were enrolled. During ALL therapy, 16 participants (7.5%) had bacteremic sepsis, and 45 (21.2%) had bacteremia without sepsis. Baseline characteristics were similar for participants with a history of sepsis and the other participants; although those with a history of sepsis were older (median age, 7.8 years [IQR, 3.8-14.7 years] vs 4.9 years [IQR, 32-8.4 years]), the difference was not statistically significant (P = .07) (Table).
After adjustment for demographic and treatment-related confounders, including age, sex, race, and leukemia risk category, survivors with a history of sepsis performed worse than other participants in multiple neurocognitive domains, including executive function (spatial planning [difference, 0.78; 95% CI, 0.57-1.00], visuospatial working memory [difference, 0.52; 95% CI, 0.26-0.78], and verbal fluency [difference, 0.38; 95% CI, 0.14-0.62]) and attention (response time variability [difference, 0.63; 95% CI, 0.30-0.95], and vigilance [difference, 0.54; 95% CI, 0.24-0.85]) (Figure). Covariates, including sex, race, leukemia risk category, age at diagnosis, and age at testing, were well-balanced after adjustment. Unadjusted analyses showed similar results, and outcomes for participants who had bacteremia without sepsis were comparable to those of patients without bacteremia.
This study is, to our knowledge, the first publication to evaluate the association of sepsis during treatment for ALL with long-term neurocognitive function. Patients who had experienced bacteremic sepsis during therapy had significantly poorer neurocognitive function at long-term follow-up compared with other participants. Studies in other populations have suggested a similar association, but the mechanism remains controversial.3 In the population of children with cancer, these mechanisms might be augmented by increased blood-brain barrier permeability to neurotoxic chemotherapy drugs.3
The study has some limitations. Although outcome differences were observed for many measures, not all attained statistical significance because of the relatively small number of exposed study participants. A pragmatic sepsis definition enabled the identification of significant organ dysfunction, but the study-specific definition hampers interstudy comparisons.4 Last, because of potential confounding by unidentified variables, the gap between exposure and outcome assessment, and lack of information about baseline neurocognitive function, a causative association cannot be proven.
The finding that sepsis was associated with impaired long-term neurocognitive function has practice-changing implications for cancer survivorship. Prevention of infection, early recognition and appropriate management of sepsis, and preemptive neurocognitive interventions should be prioritized, because these might prevent or ameliorate neurologic damage.2,4,6 Future research could aim to validate this finding independently and evaluate mechanisms of brain injury.
Accepted for Publication: June 6, 2018.
Corresponding Author: Joshua Wolf, MBBS, Department of Infectious Diseases, St. Jude Children’s Research Hospital, 262 Danny Thomas Pl, Mailstop 320, Memphis, TN 38105 (email@example.com); Kevin R. Krull, PhD, Department of Epidemiology and Cancer Control, St. Jude Children’s Research Hospital, 262 Danny Thomas Pl, Mailstop 735, Memphis, TN 38105 (firstname.lastname@example.org).
Published Online: September 24, 2018. doi:10.1001/jamapediatrics.2018.2500
Author Contributions: Drs Krull and Wolf had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Cheung, Eskind, Hudson, Krull, Wolf.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Cheung, Eskind, Hudson, Krull, Wolf.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Cheung, Eskind, Wolf.
Obtained funding: Hudson, Krull.
Administrative, technical, or material support: Cheung, Eskind, Inaba, Pui, Krull, Wolf.
Supervision: Krull, Wolf.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported in part by grant MH085849 from the National Institute of Mental Health (Dr Krull), grant CA195547 (Dr Hudson) and Cancer Center Core grant CA21765 from the National Cancer Institute, and American Lebanese Syrian Associated Charities (ALSAC).
Role of the Funder/Sponsor: The funders/sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Meeting Presentations: This study was presented in part at the Annual Meeting of the Pediatric Academic Societies; April 12, 2017; San Francisco, California; and at ID Week, Infectious Diseases Society of America; October 19, 2016; New Orleans, Louisiana.
Additional Contributions: Cara Kimberg, PhD, Cynthia Jones, MA, Deborah Stewart, MA, and Adrienne Studaway, MEd, St. Jude Children’s Research Hospital, administered the neurocognitive tests, for which they were compensated. Joycelynn Butler, MS, and Tina Culley, BS, St. Jude Children’s Research Hospital, extracted data, for which they were compensated. Keith A. Laycock, PhD, ELS, St. Jude Children’s Research Hospital, provided scientific editing of the manuscript, for which he was compensated. Elaine Tuomanen, MD, Elisa Margolis, MD, Cheng Cheng, PhD, and Li Tang, PhD, St. Jude Children’s Research Hospital, provided constructive criticism of the manuscript, for which they were not compensated.