Article Text
PDF
Abstract
Background Rituximab is associated with high infection rates, but studies of infections following rituximab in youth with childhood-onset SLE (cSLE) are limited. We conducted a retrospective longitudinal cohort study to assess the incidence of hospitalised infections following rituximab among children with cSLE and to assess changes in hospital-based rituximab administration over time.
Methods Youth ages 2–21 years with an International Classification of Diseases (ICD) code for SLE who received rituximab during admission to a Pediatric Health Information System hospital from 2009 to 2021 were included. Incidence rates for infections requiring hospitalisation over the 12 months following first rituximab administration were calculated. Rituximab use by year of hospital discharge was tabulated.
Results We identified 1567 children with cSLE who received rituximab. 219 children were admitted with an infection within 1 year after first rituximab administration, for an incidence rate of 140 cases per 1000 patient-years. Seven children (0.44%) died during a hospitalisation with an infection in the year following rituximab administration. The most common hospitalised infections were bacterial pneumonia, sepsis and cellulitis. 12 children were hospitalised with COVID-19, none of whom died. Hospitalisations with rituximab administered decreased from 2019 to 2021.
Conclusions In this cohort of patients with cSLE who received inpatient treatment with rituximab, we observed a 14% rate of hospitalisation with infection in the year following rituximab administration among youth with cSLE. Rituximab use declined during the COVID-19 pandemic. No fatalities with COVID-19 were observed. Given the lack of outpatient data, including doses of concomitant medications and disease activity measures, further research is needed to identify risk factors for infection following rituximab among children with cSLE.
- Infections
- Lupus Erythematosus, Systemic
- Lupus Nephritis
- Biological Products
- Therapeutics
Data availability statement
Data are available upon reasonable request. Deidentified data may be provided upon request pending approval of the Seattle Children’s Hospital IRB and the Children’s Hospital Association.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Introduction
Rituximab, an anti-CD20 B cell depleting monoclonal antibody, is associated with high rates of infection among adults with SLE and other conditions.1 2 An elevated risk of infection has also been shown among children receiving rituximab for autoimmune indications,3–5 but studies among children with childhood-onset SLE (cSLE) are limited. It is unknown how much risk is attributable to rituximab itself versus concomitantly administered immunosuppressants. Physician, patient and parent/caregiver concern for the immunosuppressive risks of rituximab may have increased since the emergence of SARS-CoV-2, as early studies showed high mortality among adults with rheumatological disease who contracted COVID-19 after receiving rituximab.6–8 However, the risk of severe COVID-19 among youth who received rituximab for cSLE remains poorly understood. We conducted a retrospective longitudinal cohort study to assess the incidence of hospitalised infections following first rituximab administration among children with cSLE. We also aimed to assess changes in hospital administration of rituximab over time, hypothesising that rituximab use declined after the start of the COVID-19 pandemic.
Methods
Children and adolescents ages 2–21 years with an International Classification of Diseases (ICD)-9 or ICD-10 code for SLE (710.0, M32*) and who received at least one dose of rituximab during admission to a Pediatric Health Information System (PHIS)-participating hospital from 2009 to 2021 were included. PHIS is a national administrative database with in-hospital billing data from over 50 free-standing children’s hospitals in the USA. These data include patient demographic data obtained from hospitals using standardised categories and payor types, discharge diagnosis codes, charges for intensive care unit (ICU) level of care (ICU ‘Flag’) and pharmacy administration data for drugs received during an inpatient or observation hospitalisation encounter.
Lupus organ involvement and disease severity were assessed using ICD-9 and ICD-10 codes for lupus nephritis, end-stage renal disease (ESRD), seizure, stroke and pulmonary haemorrhage at any time prior to first rituximab dose, and ICU admission during index hospitalisation for first rituximab dose. Infections were defined by ICD-9 and ICD-10 codes during an inpatient or observation encounter and were based on review of the literature. Codes that may have represented a manifestation of lupus itself (eg, undifferentiated fever) or were likely to represent a non-serious infection incidentally coded during hospitalisation for another indication were removed. Final codes included in the infection definition are listed in online supplemental table 1. In addition to a serious infection code, antibiotic use during the hospitalisation was also required for bacterial infections.
Supplemental material
The observation period for infections was 12 months following the index hospitalisation with first rituximab administration. Summary statistics were used to describe children with cSLE who received rituximab and the subset who developed an infection. Coadministered immunosuppressive medications received during index hospitalisation for first rituximab administration and any inpatient encounter during 12 months of follow-up were reported. Due to the longer duration of action of cyclophosphamide, exposure was defined as receipt of cyclophosphamide up to 3 months prior to index hospitalisation for first rituximab or up to 12 months after. Incidence rates for infections requiring hospitalisation over the 12 months following first rituximab administration were calculated using exposure time truncated at time of death, first hospitalised infection or 12 months after index hospitalisation. Survival analysis using a Cox proportional hazard model was performed to assess the impact of patient characteristics (age, sex, race, ethnicity, lupus nephritis, ESRD, ICU admission at index hospitalisation) and concomitantly administered cyclophosphamide on time to hospitalised infection following first rituximab administration. Kaplan-Meier curve and log-rank test were also used to assess the impact of concomitant cyclophosphamide use on time to readmission with infection.
Inpatient rituximab administration by year of hospital discharge was tabulated from 2009 to 2021 in admissions with all length of stay (LOS) and short (≤1 day) LOS in order to assess trends in probable planned admissions for infusion. To account for changes in the number of cSLE admissions over time, the total admissions with rituximab and the proportion of cSLE admissions with rituximab administered were calculated.
Results
We identified 8588 children with cSLE, of whom 1567 received ≥1 dose of rituximab during admission to a PHIS-participating hospital. Those hospitalised with infection were similar to the overall cohort in demographic and clinical characteristics, although children with ESRD were over-represented among those with infection (11% of those readmitted with infection vs 5% of overall) (table 1). Medication exposures were also similar between groups, although slightly higher rates of all drug exposures were seen in those readmitted with infection. Of the entire rituximab cohort, 39% (n=604) received cyclophosphamide, compared with 44% of those who were readmitted with a serious infection.
Characteristics of children with childhood-onset SLE who received rituximab and were hospitalised with infection
Of 1567 rituximab-treated children, 219 (14%, representing 339 hospitalisations) were readmitted with an infection within 1 year after the first rituximab administration, for an incidence rate of 140 cases per 1000 patient-years. Of the children readmitted with infection, 26% required ICU-level care. The median (IQR) time to hospitalisation with infection following rituximab was 1.83 (0.61–5.84) months. In a Cox proportional hazard model, only lupus nephritis (HR 1.94 (1.43, 2.63), p<0.001) was significantly associated with shorter time to readmission with serious infection (online supplemental table 2, online supplemental figure). Exposure to cyclophosphamide combined with rituximab was not associated with higher infection risk than rituximab (HR 1.05 (0.79, 1.39)).
Supplemental material
Supplemental material
The most common infections were bacterial pneumonia, sepsis and cellulitis (table 2). Twelve children were hospitalised with COVID-19 (4% of all hospitalised infections), of whom two required ICU-level care. There were no in-hospital deaths with COVID-19.
Hospitalisation outcome by type of infection
Seven children with cSLE (0.44%) died during hospitalisation with an infection in the first year following rituximab administration. Of these seven children, 71% (n=5) had multiple infection codes. The most common infections among those who died included sepsis (n=5), bacterial pneumonia (n=2), cellulitis (n=2) and systemic candidiasis (n=2). Other infections observed in cases of in-hospital mortality included cytomegalovirus, herpes simplex, Pneumocystis jirovecii pneumonia, other viral pneumonia and other mycoses (n=1 for each diagnosis).
The total number of hospitalisations with rituximab administered decreased from 324 in 2019 to 299 in 2020 and 264 in 2021. Among those hospitalisations with LOS ≤1 day, the number with rituximab administered decreased from 89 in 2019 to 60 in 2020 and 57 in 2021. The proportion of short LOS admissions with rituximab administered plateaued and decreased starting in 2016, while a decrease in both short and total LOS admissions with rituximab administered was observed starting in 2020 (figure 1).
Proportion of hospitalisations with rituximab administered by discharge year and length of stay (LOS).
Discussion
We observed high rates of infection after rituximab administration in a multicentre cohort of youth with cSLE. This is the largest cohort of children with cSLE who received rituximab in which the types and incidence of hospitalised infections have been evaluated. The overall rate of hospitalisation for infection in the year following rituximab among our cohort exceeded that previously reported in a large Medicaid-insured cohort of children with cSLE.9 As demonstrated in that cohort,9 and our previous investigations of children with cSLE using PHIS,10 bacterial pneumonia was the single largest contributing infection. Viral and fungal infections were also common, including influenza, herpes simplex virus and herpes zoster. Several prevalent infections were potentially vaccine-preventable, including bacterial pneumonia, influenza and herpes zoster. We observed one death from P. jirovecii pneumonia. Due to lack of outpatient medication data, we were unable to ascertain if this child had received Pneumocystis prophylaxis, although the use of prophylaxis is controversial in the management of paediatric and adult SLE and has not routinely been given following rituximab.11–14
Rituximab use declined during the COVID-19 pandemic. This may indicate more hesitancy regarding rituximab use in cSLE care. Despite the possibility of concern for severe COVID-19 leading to reductions in rituximab use, no fatalities during hospitalisations with COVID-19 were observed in our cohort. We also observed a decline in short LOS admissions with rituximab that preceded the pandemic, which may indicate a shift of more routine cSLE care towards infusion centre rather than inpatient settings. This trend may have increased following COVID-19.
Our study has several limitations. Concomitant use of other highly immunosuppressive medications administered inpatient, including methylprednisolone and cyclophosphamide,15 was common, which may have contributed to infection risk. Although other immunomodulator use was similar in those who did and did not have an infection, and cyclophosphamide was not an independent risk factor for hospitalised infection in our cohort in multivariable regression, a limitation of our approach is that PHIS does not contain outpatient prescription data. Therefore, concomitant immunomodulator use may have been underestimated, and there is a risk of ascertainment bias if oral medications usually received in the outpatient setting were observed at a higher rate among those with additional hospitalisations during the observation period. We were unable to conclusively assess the additional risk of rituximab itself by comparing children with similar background immunosuppressant use, including comparable glucocorticoid doses. This limitation highlights the need for more comprehensive data sources that include both inpatient and outpatient cSLE treatment data, and a future direction of this work includes linkages to other data sets to perform more robust pharmacoepidemiological studies of cSLE treatment.
As many children receive their first rituximab dose while inpatient and the subsequent doses at outpatient infusion centres, we were not able to determine the total number of doses received and we did not have laboratory data to confirm B cell depletion or assess hypogammaglobulinaemia, which are known risk factors for post-rituximab infection.1 5 Finally, while individual participants can be tracked through multiple hospitalisations at different PHIS-participating hospitals, we cannot rule out that children may have received rituximab prior to the index hospitalisation at non-participating hospitals or outpatient infusion centres not captured in our data set. It is possible that these data set limitations set may lead to overestimation of the infection risks of rituximab in youth with cSLE, or limit the generalisability of our findings, as the population of patients with cSLE who received rituximab while inpatient may include sicker patients than those who receive rituximab solely in outpatient settings.
Our results, including the prevalence of infections which are potentially preventable by prophylactic medication and vaccination, may guide the use of preventative care measures and help target quality improvement efforts to reduce the risk of readmission and death from infection in this high-risk population.16 Further research is needed to understand the additional risk of rituximab compared with conventional disease-modifying antirheumatic drugs and glucocorticoids alone, and to identify modifiable risk factors for infection following rituximab among children with cSLE.
Data availability statement
Data are available upon reasonable request. Deidentified data may be provided upon request pending approval of the Seattle Children’s Hospital IRB and the Children’s Hospital Association.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by the Seattle Children’s Hospital Institutional Review Board (protocol #00003955). The study was determined exempt due to minimal risk and use of secondary data only. PHIS data use was approved by the Children’s Hospital Association. This was an exempt study not requiring consent due to use of secondary data only.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
X @jordanrobertsMD
Presented at Data included in this manuscript were previously published as a conference abstract: Roberts J, Faino A, Bryan M, Cogen J, Morgan E. Serious infections following rituximab administration in children with systemic lupus erythematosus (abstract). Arthritis Rheumatol. 2023; 75 (suppl 9). https://acrabstracts.org/abstract/serious-infections-following-rituximab-administration-in-children-with-systemic-lupus-erythematosus/.
Contributors JER, AVF, JDC, MAB and EMM contributed to the study conception and design. Data analysis was performed by AVF and JER. AVF, JDC, MAB and EMM contributed to analysis of the results. The first draft of the manuscript was written by JER. She is the guarantor. All authors contributed to critically revising the manuscript and approved the final version.
Funding JER was supported by grants from the Lupus Foundation of America, the Childhood Arthritis and Rheumatology Research Alliance-Arthritis Foundation, and the Center for Clinical and Translational Research, Seattle Children’s Research Institute.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.