The Prevalence of Anti-SARS-CoV-2 Antibodies After COVID-19 is Similar in Kidney Transplant Recipients to Non-Transplanted Patients, and the Clinical Severity is a Predictor of Seroconversion

ABSTRACT

Purpose: To evaluate the seroconversion rate in kidney transplant recipients (KTR), compared to two non-transplanted groups of patients, and identify predictors of seroconversion in COVID-19 convalescent patients. Method(s) A retrospective cohort study enrolled RT-PCR COVID-19 diagnosed patients (Mar/20 and Oct/2020) of three groups 601 KTR, 211 health care workers (HCW), and 170 non-transplanted inhabitants (INH) in a countryside city in the state of Sao Paulo - Brazil. At least 14 days after diagnosis, all survivors underwent antibody testing by chemiluminescent microparticle immunoassay (titter expressed in RLU). The primary outcome was seroconversion. The group-adjusted multivariable model for the probability of seroconversion was built by generalized linear mixed models with binary logistic regression and the discrimination performance by AUC-ROC. Result(s) Several differences were observed among groups regarding demographic data and COVID-19 clinical presentation. Of note, KTR were older (54.0 years old vs. 37.0 in HCW vs. 42.0 in INH, P<0.001), more frequently had comorbidities (P<0.001), and severe COVID-19 (P<0.001). Compared to HCW and INH, respectively, admission to ICU (44.9% vs. 0% vs. 1.8%), MV requirement (32.3% vs. 0% vs. 1.8%), and death (28.8% vs. 0% vs. 1.2%) were significantly more frequent in KTR (P<0.001). On the other hand, the seroconversion rate was not different among survivors 76.2% for KTR, 74.9% for HCW, and 82.2% for INH (P=0.35). The IgG anti-SARS-CoV-2 was slightly higher among INH 5.8 RLU vs. 5.4 for KTR and 4.4 for HCW (P=0.009). Seroconversion was associated with a shorter time between infection and blood sample collection (OR for each day= 0.986;P<0.001) and increased by 64% if the fever was a COVID-19 symptom (OR=1.737;P=0.017), 78% if the cough was present (OR=1.785;P=0.005) and 98% if the ventilatory support was required (OR=1.981;P=0.017). This predictive model achieved an AU-ROC of 0.730 (P<0.001). Conclusion(s) As expected, the rates of clinical deterioration to ICU admission, MV requirement, and death were significantly higher among KTR. However, among the survivors, KTR had a similar rate of seroconversion, associated with clinical severity parameters and a shorter time of blood sample collection.