COVID-19 in pediatric kidney transplantation: The Improving Renal Outcomes Collaborative

2.1 IROC registry

IROC was formed in 2016 with 15 founding centers. Since that time, 19 additional centers have joined the collaborative on a rolling basis. Currently, over 50% of all pediatric KTs in the United States are performed at IROC-participating centers. IROC centers complete structured training in quality improvement and maintain site-specific infrastructure to report patient-level clinical variables to the IROC registry. The parent IROC study was approved by the Institutional Review Board (IRB) at Cincinnati Children's Hospital Medical Center under a master reliance agreement that was approved by each participating center. This agreement allows Cincinnati Children's Medical Center to act as the IRB of record for the registry-specific data analyzed in this study. This COVID-19 study was approved by the Cincinnati Children's IRB.

Clinical data for pediatric KT patients at each of the IROC sites are routinely uploaded into a central IROC data registry. These patient-level data include demographics, clinical visit data, laboratory data, medications, hospitalization-related data, biopsy data, and rejection events/outcomes. These elements are processed and disseminated to centers as statistical process control charts, population management measures, and previsit planning forms to support ongoing pediatric KT care, quality improvement activities at the institutional level, and for benchmarking across the network. Demographic data in the registry included patient age, sex, and race (although race data were incomplete for some patients). Available medical data included cause of end-stage kidney disease leading to transplant, date of KT (including prior if multiple transplants), donor type (living or deceased), current nonimmunosuppression medications (e.g., ACE-inhibitors), current immunosuppression drug regimen, and history of rejection episodes with grade of rejection stratified according to Banff criteria.^{12, 13} IROC-based research studies are permitted to use these registry data as covariates as in the current study.

2.2 COVID-19 data collection tool

To collect specific information regarding COVID-19 testing, results, and clinical outcomes, a Research Electronic Data Capture (REDCap) data collection tool was created and distributed to IROC centers (Figure 1). REDCap is a secure, web-based software platform designed to support data capture for research studies.^{14, 15} Entry of COVID-19 testing information into the REDCap data collection tool was voluntary for IROC centers during the study period. Centers were encouraged to enter all patients tested rather than only those who tested positive. During the study period of April 6 to September 3, 2020, centers entered patient-level data into REDCap for COVID-19 testing results, indication(s) for testing, symptoms at the time testing (if applicable), known history of lung disease, and the provided treatment(s) for confirmed or suspected COVID-19. Indications for testing were not mutually exclusive, thus centers can report multiple indications for testing. Additional outcome questions included the highest level of care the patient required (e.g., outpatient, inpatient wards, intensive care) and endpoints for both the allograft and patient following COVID-19 disease, such as acute kidney injury (AKI), rejection, respiratory failure, and/or death.

2.3 Data linkage

Each patient in the IROC registry is assigned both a unique center ID and patient ID. The center ID and IROC registry ID(s) were entered along with COVID-19 testing data in the REDCap data collection tool. Patient clinical data, including immunosuppressive and antihypertensive agent(s), were available from the IROC registry and linked to the COVID-19 testing data using patient-level, secure identifiers. Individual centers were contacted to provide follow-up for missing data, if available, on individual patients from either the IROC registry or the REDCap data collection tool.

2.4 Statistical analysis

Linked data from the REDCap collection tool and IROC registry were analyzed to provide descriptive information on the patients entered in the database, including the number of patients tested, indication for testing, the incidence of COVID-19, and clinical outcomes. If a patient was tested and entered into REDCap tool multiple times, the individual was analyzed once. If subsequent testing revealed a positive test, or a test with symptoms was reported, those entries were prioritized. To estimate the incidence of COVID-19-positive testing using patients with a documented test result, we analyzed a subgroup of patients from centers that entered all COVID-19 testing data during the study period. Ages were calculated based on demographic data reported in the IROC registry relative to the date that testing data were entered into the REDCap database. Continuous variables were summarized using medians with interquartile ranges (IQR) and tested for differences between groups using Wilcoxon rank-sum test to accommodate the nonnormally distributed variables. Categorical variables were analyzed by Fisher's exact test due to the small number of patients with a positive test. p < .05 were considered significant results. Analyses were conducted using SAS version 9.4 (SAS Institute, Inc).

3.1 Indications for COVID-19 testing

Twenty-two IROC centers that collectively care for 2732 patients submitted COVID-19 testing data from 281 patients between April 6, 2020 and September 3, 2020 (see Figure 2). All centers reported using SARS-CoV-2 PCR for testing. Patient characteristics and demographics are detailed in Table 1 and COVID-19 testing indications and symptoms are detailed in Table 2. When evaluating testing indication for all tested patients, 105 (37.4%) displayed symptoms consistent with COVID-19, 24 (8.5%) had close contact with a confirmed case of COVID-19, and five (1.8%) had close contact with a person under investigation or displaying symptoms consistent with COVID-19. Most patients (55.9%) were tested per hospital policy, since many hospitals during this time required universal testing for all admitted patients, even without COVID-19 symptoms, as part of any admission or preprocedure anesthesia evaluation.

3.2 Incidence of COVID-19

To estimate the incidence of a positive COVID-19 test in our cohort, we excluded any center that did not enter all their tested transplant patients during the study period. Thirteen IROC centers caring for a total of 1686 patients entered data on all tested patients (n = 229). The overall estimated incidence of COVID-19 at these centers was 0.6% (10/1686). The incidence of COVID-19 in tested patients at these centers was 4.4% (10/229). Five of the 10 positive patients (50%) from this subanalysis were asymptomatic.

3.3 Characteristics of COVID-19-positive kidney transplant patients

Twenty-four patients (8.5%) tested positive for COVID-19 out of all those entered in the REDCap registry (n = 281). Treatment and outcomes data for the entire cohort are detailed in Table 3. The median (IQR) age for the positive patients was 18.6 years (14.3–20.6 years), and the range was 4.4–24 years, which was higher than the median age for negative patients of 14.3 (7.3–18.3). The COVID-19-positive patients had a median (IQR) time since transplant of 2.9 years (1.8–8.7) years. Twenty percent of our cohort were in their first year posttransplant (60/281), and three of those 60 (5%) patients tested positive for COVID-19. Over half of all positive patients were male (15/24, 62%). The top three primary kidney diseases reported were congenital anomalies of the kidney and urinary tract (CAKUT), glomerulonephritis, and nephrotic syndrome/FSGS.

When analyzing all positive patients, 15 of 24 (63%) had any symptoms associated with COVID-19, including cough (33.3%), fever (29.2%), vomiting (16.7%), diarrhea (12.5%), rhinorrhea (8.3%), and shortness of breath (8.3%). Fifteen of the 113 (13.3%) symptomatic patients tested positive compared to nine of the 168 (5.4%) asymptomatic patients testing positive (p = .03). Only one (4.2%) COVID-19-positive patient had a history of any lung disease (childhood asthma), and 15 (62.5%) had a diagnosis of hypertension, with 93% of those patients on antihypertensive medication.

3.4 Treatment and outcomes in COVID-19-positive patients

Nineteen of the positive patients (79.2%) required supportive care only. Four (16.7%) had a reduction in their immunosuppression regimen, three of which were temporary reductions in mycophenolate mofetil dosing. Sixteen of the 24 positive patients (66.7%) were treated as outpatients, six (25%) were treated as inpatients in a non-ICU setting, and two (8.3%) required ICU admission. Upon further review with the submitting center, the first ICU patient was admitted per hospital policy in the early days of the pandemic out of an abundance of caution rather than the presence of critical illness. The second ICU patient was admitted with concern for shock that rapidly improved and was transferred to the inpatient wards within 48 h. This patient had concomitant pyelonephritis at the time of COVID-19 diagnosis. Of the six remaining hospitalized COVID-19 patients, one was asymptomatic and detected by hospital universal screening policy, the other five had symptoms requiring inpatient management.

Twenty of the patients (83.3%) had no transplant-related complications, two patients (8.3%) had AKI at the time of testing that resolved prior to discharge, one (4.2%) was diagnosed with T cell–mediated rejection at the time of COVID-19 diagnosis, and one (4.2%) was diagnosed with antibody-mediated rejection at the time of COVID-19 diagnosis. The cause of AKI in one patient was directly related to a supratherapeutic tacrolimus level that required holding the tacrolimus for 36 h. The second patient with AKI was admitted for a kidney biopsy secondary to elevated creatinine without apparent cause. The patient's biopsy did not show evidence of rejection and the creatinine normalized without specific intervention. This patient's COVID-19 test resulted positive after the procedure was completed.

The patient with T cell–mediated rejection had no COVID-19 symptoms but was screened per hospital policy prior to a diagnostic biopsy for elevated creatinine and presence of donor-specific antibodies (biopsy showed Banff 1A T cell–mediated rejection and no antibody-mediated rejection). The patient had no symptoms at the time of biopsy. The patient was treated with methylprednisolone and intravenous immunoglobulin (IVIG) for the rejection episode and donor-specific antibodies and subsequently developed fever and cough that required supplemental oxygen while admitted. Chest x-ray demonstrated patchy airspace densities with prominent central pulmonary vascularity and interstitial prominence. This patient received dexamethasone and over the subsequent week showed clinical improvement, becoming afebrile and weaning off oxygen. The patient was discharged home on maintenance immunosuppression and the creatinine returned to baseline.

The patient with antibody-mediated rejection had longstanding donor-specific antibodies, a history of antibody-mediated rejection, and a history of poor immunosuppression adherence. This patient was receiving IVIG for treatment of donor-specific antibodies and was tested for COVID-19 during an infusion visit. After monitoring for 20 days, this patient remained asymptomatic while the COVID-19 test remained positive. A biopsy was done to follow-up treatment of donor-specific antibodies, which showed severe, active antibody-mediated rejection. This patient was treated with IV methylprednisolone, plasmapheresis, rituximab, bortezomib, and IVIG. Neither of the two patients who experienced rejection were on reduced immunosuppression due to their COVID diagnosis.

None of the patients who tested positive required dialysis or experienced graft failure, and none of them experienced respiratory failure, required intubation, or died.