Clinical Characteristics and Outcomes of COVID-19 in Solid Organ Transplant Recipients

A Cohort Study

Zohra S. Chaudhry; Jonathan D. Williams; Amit Vahia; Raef Fadel; Tommy Parraga Acosta; Rohini Prashar; Pritika Shrivastava; Nadeen Khoury; Julio Pinto Corrales; Celeste Williams; Shunji Nagai; Marwan Abouljoud; Milagros Samaniego-Picota; Odaliz Abreu-Lanfranco; Ramon del Busto; Mayur S. Ramesh; Anita Patel; George J. Alangaden


American Journal of Transplantation. 2020;20(11):3051-3060. 

In This Article


Study Setting

The study was performed at the Henry Ford Health System (HFHS) a quaternary-care academic institution that comprises 5 hospitals in southeast and south-central Michigan. The Henry Ford Transplant Institute housed in HFHS annually performs approximately 300 organ transplants including kidney, liver, heart, lung, pancreas, intestinal, and multiorgan transplants.

Study Population

A confirmed case of COVID-19 was defined as a patient with a positive reverse transcription-polymerase chain reaction (RT-PCR) assay for SARS-CoV-2 in a nasopharyngeal sample tested by the Michigan Department of Health and Human Services or the HFHS centralized clinical microbiology laboratory.

Cases: SOT recipients with the confirmed diagnosis of COVID-19 from March 20, 2020 through April 18, 2020, were eligible for inclusion if they were 18 years of age or older. SOT recipients managed as outpatients were subsequently excluded from the outcome analyses to avoid potential selection bias.

Controls: A convenience sample of 100 consecutive nontransplant patients hospitalized with the confirmed diagnosis of COVID-19 from March 20, 2020 onwards were eligible for inclusion if they were 18 years of age or older.

The Henry Ford Hospital (HFH) COVID-19 severity scoring system was used to risk stratify patients on presentation to the hospital as mild, moderate, or severe COVID-19. Mild disease was defined as patients who had normal chest radiography and SpO2 of ≥94% without the need for supplemental oxygen. Moderate disease patients were those who had abnormal chest radiography, SpO2 of <94% and needing between 1 and 5 liters/min supplemental O2. Patients with severe disease were defined by abnormal chest radiography, SpO2 of <94% and requiring ≥6 liters/min of O2.

Study Design

This was a retrospective cohort study examining epidemiologic, laboratory, and clinical characteristics for adverse outcomes comparing SOT recipients and nontransplant controls hospitalized for COVID-19. The study was approved by the institution's institutional review board (#13739) with waiver of consent.

Both cases and controls received standard care, comprised of supplemental oxygen, high-flow nasal cannula support, mechanical ventilation, antibiotics, antiviral agents, immunomodulating medications, vasopressor support, and renal replacement therapy, as determined by the primary team.

Immunosuppression Management. The management of immunosuppression was at the discretion of the transplant team. In general the approach was to decrease overall immunosuppression. Typically, antimetabolites were the first to either be withdrawn or dose reduced with consideration for modification of calcineurin inhibitors (CNIs).

Specific COVID-19 Treatment (COVID-19 Protocol). HFHS deployed a COVID-19 team during the time of outbreak in an effort to standardize management. Laboratory tests ordered included baseline serum C-reactive protein (CRP), lactate dehydrogenase (LDH), ferritin, D-dimer, creatine phosphokinase (CPK), high-sensitivity troponin, and procalcitonin. Additionally, the protocol also defined the use of antiviral agents and adjuvant immunomodulation therapies in the treatment of patients with COVID-19.

COVID-19-positive SOT recipients and controls with moderate to severe disease as per our severity criteria along with QTc interval of <500 msec received hydroxychloroquine (HCQ) 400mg twice daily orally for the first 24 hours followed by 200mg twice a day for 4 days in patients as the antiviral agent. Adjuvant therapy included an early short course of corticosteroids (methylprednisolone 1mg/kg in 2 divided doses for 3–7 days).[7]

Data Collection

Data were abstracted from the electronic medical record (EMR) and recorded in a standardized electronic case report form. Data included patient demographics, clinical symptoms and signs, laboratory and radiologic results at the time of presentation. Patient data were censored on April 26, 2020.

Study Definitions

The National Early Warning Score (NEWS) and quick sequential organ failure (qSOFA) were calculated to evaluate baseline illness severity based on vital signs obtained in the emergency department.[8,9]

Outcome Measures

Primary End Point. The primary composite end point was escalation to ICU from a general practice unit (GPU), progression to respiratory failure requiring mechanical ventilation, or in-hospital all-cause mortality. This composite end point has been utilized to measure outcomes in COVID-19 pneumonia previously.[2,7]

Secondary End Points. Secondary end points included development and severity of acute respiratory distress syndrome (ARDS), acute kidney injury (AKI) requiring renal replacement therapy, median duration of ventilation, and length of hospital stay (LOS). ARDS was diagnosed and classified according to the Berlin Definition.[10] AKI and chronic kidney disease (CKD) were diagnosed according to the Kidney Disease: Improving Global Outcomes definition.[11]

Statistical Analysis

Continuous variables were reported as median and interquartile range (IQR) and compared using the Mann-Whitney test or t test, as appropriate. Categorical data were reported as number and percentage and compared using the chi-square test or Fisher's exact t test as appropriate. No imputation was made for missing data points. The sample size was derived from all eligible consecutive hospitalized patients during the study period. A 2-sided α ≤0.05 was considered statistically significant. Bivariate and multivariable logistic regression analysis was planned a priori to test the association between the individual composite end point components and SOT status. Clinically observed risk factors for mortality were fit a priori into a multivariate logistic regression model examining key exposures and outcomes. Survival curves were modeled using Kaplan-Meier estimation censoring data at the end of follow-up. Statistical analysis was performed using SAS 9.4 (SAS Institute, Cary, NC).