| |
COVID-19 Disease Severity among People with HIV Infection or Solid Organ Transplant in the United States: A Nationally-representative, Multicenter, Observational Cohort Study
|
| |
| |
Download the PDF here
Jing Sun, Rena C. Patel, Qulu Zheng, Vithal Madhira, Amy L. Olex, Jessica Y. Islam, Evan French, Teresa Po-Yu Chiang, Hana Akselrod, Richard Moffitt, G. Caleb Alexander, Kathleen M. Andersen, Amanda J. Vinson, Todd T. Brown, Christopher G. Chute, Keith A. Crandall, Nora Franceschini, Roslyn B. Mannon, Gregory D. Kirk, National COVID Cohort Collaborative (N3C) Consortium
doi: https://doi.org/10.1101/2021.07.26.21261028
medrix.org
This article is a preprint and has not been peer-reviewed [what does this mean?]. It reports new medical research that has yet to be evaluated and so should not be used to guide clinical practice.
DISCUSSION
Our analysis of approximately 1.5 million U.S. patients, is among the largest and most representative multicenter cohort of COVID-19 patients assembled reflective of the U.S. pandemic to date 17,18. Both PWH or SOT had increased odds of severe COVID-19 outcomes, including requiring invasive ventilation/ECMO and death. The odds for severe disease were substantially driven by, though still independent of, sociodemographic variables and comorbidity burden. Among PWH, the degree of immunodeficiency and detectable HIV viremia were both independently associated with greater risk for severe COVID-19 disease. These data reveal that ISC patients are highly susceptible to severe COVID-19 and emphasize the need for identifying and implementing appropriate strategies for the prevention and management of COVID-19 disease among persons with immune dysfunction.
Table 2 showed that patients with HIV, SOT or both had higher odds of more severe disease. Among PWH, adjusted odds ratios (ORs) for ED visits, moderate hospitalization (without invasive ventilation/ECMO), severe hospitalization with invasive ventilation/ECMO, and death were 1.28 (95% CI: 1.27, 1.29), 0.81 (95% CI: 0.78, 0.86), 1.43 (95% CI: 1.43, 1.43), and 1.20 (95% CI: 1.19, 1.20), respectively, compared to people without ISC. Compared to the crude estimates, adjustment for sociodemographic and smoking (Table 2, model A) and comorbidity burden (model B) substantially attenuated the association, but all associations remained significant. Adjustment for study sites resulted in high precision for estimations and produced narrow CI (model A and B). We observed an inverse association for moderate hospitalization (WHO severity 4-6) among PWH in adjusted models A and B, despite a high crude prevalence of PWH compared to the general population (Figure 1). Evaluation of individual covariates in the adjusted models indicated that the reduced odds were mainly driven by high prevalence of smoking and non-White race/ethnicity among PWH.
"Among PWH with CD4>500 cells/mm3, detectable viremia is further associated with over 2-fold higher odds of hospitalization. Sensitivity analysis restricted to laboratory values prior to COVID-19 infection yielded consistent results."
ABSTRACT
Background - Individuals with immune dysfunction, including people with HIV (PWH) or solid organ transplant recipients (SOT), might have worse outcomes from COVID-19. We compared odds of COVID-19 outcomes between patients with and without immune dysfunction.
Methods - We evaluated data from the National COVID-19 Cohort Collaborative (N3C), a multicenter retrospective cohort of electronic medical record (EMR) data from across the United States, on. 1,446,913 adult patients with laboratory-confirmed SARS-CoV-2 infection. HIV, SOT, comorbidity, and HIV markers were identified from EMR data prior to SARS-CoV-2 infection. COVID-19 disease severity within 45 days of SARS-CoV-2 infection was classified into 5 categories: asymptomatic/mild disease with outpatient care; mild disease with emergency department (ED) visit; moderate disease requiring hospitalization; severe disease requiring ventilation or extracorporeal membrane oxygenation (ECMO); and death. We used multivariable, multinomial logistic regression models to compare odds of COVID-19 outcomes between patients with and without immune dysfunction.
Findings - Compared to patients without immune dysfunction, PWH and SOT had a greater likelihood of having ED visits (adjusted odds ratio [aOR]: 1.28, 95% confidence interval [CI] 1.27-1.29; aOR: 2.61, CI: 2.58-2.65, respectively), requiring ventilation or ECMO (aOR: 1.43, CI: 1.43-1.43; aOR: 4.82, CI: 4.78-4.86, respectively), and death (aOR: 1.20, CI: 1.19-1.20; aOR: 3.38, CI: 3.35-3.41, respectively). Associations were independent of sociodemographic and comorbidity burden. Compared to PWH with CD4>500 cells/mm3, PWH with CD4<350 cells/mm3 were independently at 4.4-, 5.4-, and 7.6-times higher odds for hospitalization, requiring ventilation, and death, respectively. Increased COVID-19 severity was associated with higher levels of HIV viremia.
Interpretation- Individuals with immune dysfunction have greater risk for severe COVID-19 outcomes. More advanced HIV disease (greater immunosuppression and HIV viremia) was associated with higher odds of severe COVID-19 outcomes. Appropriate prevention and treatment strategies should be investigated to reduce the higher morbidity and mortality associated with COVID-19 among PWH and SOT.
RESULTS
Patient characteristics at COVID-19 diagnosis
Of 1,446,913 adults with a diagnosis of COVID-19 (Table 1), the median age was 47 years (interquartile range [IQR]: 32-61), 55.0% were female, 52.8% were non-Hispanic White, 13.7% were non-Hispanic Black, and 16.9% were current or former smokers. We identified 8,270 PWH, 11,392 SOT, and 267 with both HIV and SOT. Compared to other COVID-19 patients, PWH and SOT patients were more likely to be older, male, non-Hispanic Black, and have more comorbid conditions (Table 1).
COVID-19 Disease Severity among PWH and/or SOT
Within 45 days of COVID-19 diagnosis, 67.9% of patients had asymptomatic to mild disease requiring outpatient care only. While only 3.4% were solely evaluated in the ED, 26.1% required hospitalization without invasive ventilation/ECMO, 0.9% required invasive ventilation or ECMO but did not die, and 1.7% died. Figure 1 demonstrated the crude prevalence of COVID-19 disease severity by ISC group. Table 2 showed that patients with HIV, SOT or both had higher odds of more severe disease. Among PWH, adjusted odds ratios (ORs) for ED visits, moderate hospitalization (without invasive ventilation/ECMO), severe hospitalization with invasive ventilation/ECMO, and death were 1.28 (95% CI: 1.27, 1.29), 0.81 (95% CI: 0.78, 0.86), 1.43 (95% CI: 1.43, 1.43), and 1.20 (95% CI: 1.19, 1.20), respectively, compared to people without ISC. Compared to the crude estimates, adjustment for sociodemographic and smoking (Table 2, model A) and comorbidity burden (model B) substantially attenuated the association, but all associations remained significant. Adjustment for study sites resulted in high precision for estimations and produced narrow CI (model A and B). We observed an inverse association for moderate hospitalization (WHO severity 4-6) among PWH in adjusted models A and B, despite a high crude prevalence of PWH compared to the general population (Figure 1). Evaluation of individual covariates in the adjusted models indicated that the reduced odds were mainly driven by high prevalence of smoking and non-White race/ethnicity among PWH.
Among SOT patients with COVID-19, adjusted ORs for ED visit, moderate hospitalization, severe hospitalization with invasive ventilation/ECMO, and death were 2.61 (95% CI: 2.58, 2.65), 2.00 (95% CI: 1.93, 2.08), 4.82 (95% CI: 4.78, 4.86), and 3.38 (95% CI: 3.35, 3.41), respectively, compared to people without ISC. Compared to the crude estimates, associations were substantially attenuated after adjustment for sociodemographic and smoking status (Table 2, model A) and comorbidity burden (model B), but remain statistically significant. Odds of severe outcomes among PWH with SOT were similar to that observed among SOT, although the small sample size might hinder precise risk estimates.
Evaluation among COVID-19 patients admitted to ED or hospital (N = 413,292) further confirmed that the odds of requiring invasive ventilation/ECMO during their hospitalization for PWH, SOT, and people with both were 1.62 (95% CI: 1.42, 1.85), 2.29 (95% CI: 2.13, 2.47), and 2.10 (95% CI: 1.38, 3.19), respectively, compared to patients without ISC. All associations were independent of sociodemographic characteristics, smoking, and comorbidity (Supplementary Table S-2).
HIV Markers and COVID-19 Severity
A total of 3,660 PWH with COVID-19 had available CD4 cell counts measurements: 22.9% were <350 cells/mm3, 16.4% were between 350-500 cells/mm3, and 60.7% were >500 cells/mm3. Of 1,424 PWH with available HIV viral load measurements, the majority of them (67.7%) had undetectable viremia (<50 copies/mL), 20.6% had low level viremia (50-1000 copies/mL), and 11.7% had high level viremia (>1,000 copies/mL). Demographic characteristics of individuals with available lab values and all PWH were similar (Supplementary Table S-3).
The associations between advanced HIV disease and COVID-19 outcomes were dose-dependent and independent of sociodemographic, smoking status, and comorbidity burdens (Figure 2. Panel A, B, C). Compared to CD4>500 cells/mm3, levels of <350 cells/mm3 were independently associated with greater odds of hospitalization (OR: 4.4, 95% CI: 3.6, 5.3), invasive ventilation (OR: 5.4, 95% CI: 3.2, 9), and death (OR: 7.6, 95% CI: 3.9, 14.9). Compared to undetectable viremia, viral load of 50-1000 or >1000 copies/mL were independently associated with greater odds of hospitalization (OR: 1.8 CI: 1.2-2.7 and OR: 3.5 CI: 2.2-5.5, respectively) and death (OR: 4.4 CI: 1.4-13.7 and OR: 7.3 CI: 2.1-25.7, respectively). Among PWH with undetectable viremia, lower CD4 cell counts (350-500 and <350 cells/mm 3) was associated with 2.9 and 6-times the odds of hospitalization (Table 3), respectively, compared to higher CD4 cell counts. Among PWH with CD4>500 cells/mm3, detectable viremia is further associated with over 2-fold higher odds of hospitalization. Sensitivity analysis restricted to laboratory values prior to COVID-19 infection yielded consistent results.
HIV Markers and COVID-19 Severity
A total of 3,660 PWH with COVID-19 had available CD4 cell counts measurements: 22.9% were <350 cells/mm3, 16.4% were between 350-500 cells/mm3, and 60.7% were >500 cells/mm3. Of 1,424 PWH with available HIV viral load measurements, the majority of them (67.7%) had undetectable viremia (<50 copies/mL), 20.6% had low level viremia (50-1000 copies/mL), and 11.7% had high level viremia (>1,000 copies/mL). Demographic characteristics of individuals with available lab values and all PWH were similar (Supplementary Table S-3). The associations between advanced HIV disease and COVID-19 outcomes were dose-dependent and independent of sociodemographic, smoking status, and comorbidity burdens (Figure 2. Panel A, B, C). Compared to CD4>500 cells/mm3, levels of <350 cells/mm3 were independently associated with greater odds of hospitalization (OR: 4.4, 95% CI: 3.6, 5.3), invasive ventilation (OR: 5.4, 95% CI: 3.2, 9), and death (OR: 7.6, 95% CI: 3.9, 14.9). Compared to undetectable viremia, viral load of 50-1000 or >1000 copies/mL were independently associated with greater odds of hospitalization (OR: 1.8 CI: 1.2-2.7 and OR: 3.5 CI: 2.2-5.5, respectively) and death (OR: 4.4 CI: 1.4-13.7 and OR: 7.3 CI: 2.1-25.7, respectively). Among PWH with undetectable viremia, lower CD4 cell counts (350-500 and <350 cells/mm 3) was associated with 2.9 and 6-times the odds of hospitalization (Table 3), respectively, compared to higher CD4 cell counts. Among PWH with CD4>500 cells/mm3, detectable viremia is further associated with over 2-fold higher odds of hospitalization. Sensitivity analysis restricted to laboratory values prior to COVID-19 infection yielded consistent results.
|
|
| |
| |
|
|
|
