Analysis of Severe Illness After Postvaccination COVID-19 Breakthrough Among Adults With and Without HIV in the US

Key Points Question In 2021, among fully vaccinated people with breakthrough COVID-19 illness, was the risk of severe illness higher for people with HIV (PWH) compared with people without HIV (PWoH)? Findings In this cohort study of 3649 patients with breakthrough COVID-19, there was no overall difference in risk of severe disease between PWH and PWoH. PWH with CD4 cell count less than 350 cells/μL had a 59% increased risk of severe breakthrough illness compared with PWoH. Meaning Although vaccinations effectively reduce the risk of severe COVID-19 illness in both PWH and PWoH, these findings suggest that PWH with moderate or severe immune suppression (CD4 cell count <350 cells/μL) could be at higher risk of severe breakthrough infection compared with PWoH, and PWH with moderate immune suppression should be considered for additional vaccine dosages and other risk-reduction measures.

Current US Centers for Disease Control and Prevention (CDC) guidelines recommend risk reduction behaviors (ie, mask wearing), an additional COVID-19 primary series vaccine dose, and a second booster dose for PWH with "advanced or untreated HIV infection." 21,22PWH with partially recovered CD4 cell counts (>200 cells/μL) and moderate immune suppression are not currently recommended for an additional or second booster dose.Our objective was to determine whether HIV infection was associated with increased severe COVID-19 illness among fully vaccinated adults with a breakthrough SARS-CoV-2 infection and to identify the factors associated with severe COVID-19 breakthrough illness among PWH, including level of immune suppression and HIV viral replication.

Statistical Analysis
Study entry was the date of observed breakthrough COVID-19.Individuals were followed to the first date of severe breakthrough COVID-19 illness (outcome) or date of death, disenrollment from the health system, 28 days after breakthrough COVID-19, or December 31, 2021.
Severe COVID-19 breakthrough illness monthly incidence rates (IRs) per 100 person-years and 95% CIs were calculated by HIV status.Severe COVID-19 breakthrough cumulative incidence was estimated from the date of breakthrough through day 28; estimates were stratified by HIV status, and among PWH, CD4 cell count, and viral suppression.Using the same timescale, cumulative incidence was estimated by HIV status for each vaccine type and for those who received an additional vaccine dose (Ն28 days after primary series completion and before breakthrough COVID- 19).
Log-rank tests were calculated to test for differences in cumulative incidence, and risk differences were estimated with SE and Wald 95% CI.
A discrete time-to-event analysis using a complimentary log-log model estimated the unadjusted and adjusted hazard ratios (aHRs) with 95% CI for severe COVID-

IRs and Cumulative Incidence of Severe COVID-19 Breakthrough Illness
The IRs of severe COVID-19 breakthrough illness were similar among PWoH (138 cases per 100 person-years; 95% CI, 118-160 cases per 100 person-years) vs PWH (117 cases per 100 person-years; 95% CI, 92-145 cases per 100 person-years) and were stable over time (Figure 1 and eTable 5 in the Supplement).There were slight fluctuations reflecting the bimodal distribution of the Delta and Omicron variant waves (eFigure 3 in the Supplement).
g Cancer and pregnancy were measured as close to the date fully vaccinated as possible within the window of January 1, 2020, to the date fully vaccinated.
h AIDS was defined using clinical diagnosis codes of AIDS-defining conditions.

Factors Associated With Risk of Severe COVID-19 Breakthrough Illness, by HIV Status
In unadjusted models, PWH with CD4 cell count less than 350 cells/μL had a higher risk of severe infection than PWoH (HR, 2.00; 95% CI, 1.26-3.02).In adjusted analyses, there was no difference in severe COVID-19 breakthrough illness risk among PWH vs PWoH (aHR, 1.02; 95% CI, 0.76-1.35).PWH with a CD4 cell count less than 350 cells/μL had a 59% increased risk of severe breakthrough illness compared with PWoH (aHR, 1.59; 95% CI, 0.99-2.46;P = .049)(Table 2).Stratified by HIV status, PWH and PWoH had an increased risk of severe COVID-19 breakthrough illness with increasing age and history of COVID-19 (before being fully vaccinated), and decreased risk among those with an additional vaccine dose.
Among PWH, severe COVID-19 breakthrough risk increased with decreasing CD4 cell count (compared with CD4 cell count Ն500 cells/μL) (Table 2).Female PWH had a nearly 3-fold increased severe COVID-19 breakthrough risk vs male PWH.Increased risk associated with non-Hispanic Black race and Hispanic ethnicity and comorbidities ranged from 12% to 52%.Having a cancer diagnosis was associated with nearly 2-fold increased risk of severe COVID-19 breakthrough illness.

Limitations
This study has limitations.Our findings may not be generalizable to all PWH, as our study population had a greater proportion of men (88.9%) than found in the US population of PWH, and those with higher barriers to accessing health care (who may be at greater risk for COVID-19) were less likely to be included in our study population.Health care utilization may also differ by HIV status, which could impact inclusion with recognized breakthrough and decisions for hospitalization (outcome).Other outcome data, including mechanical ventilation and death (particularly if death occurred out of hospital) may be underascertained.All discharge diagnoses were reviewed by clinicians to increase specificity in our classification of COVID-19 hospitalization, but discharge coding can be influenced by many factors including reimbursement practices.Our findings may not be generalizable to all SARS-CoV-2 variants; however, our study was conducted while Alpha, Delta, and Omicron variants were circulating.In addition, on the basis of the timing of additional primary series doses and booster recommendations in 2021, the majority of additional doses received were timed more similarly to a booster dose than an additional dose in the primary vaccination series.

Figure 1 .
Figure 1.Incidence Rates of Severe COVID-19 Breakthrough per 100 Patient-Years in People With HIV (PWH) and People Without HIV (PWoH) by Month 1000

Figure 3 .
Figure 3. Cumulative Incidence of Severe COVID-19 Breakthrough Illness by HIV Status and Primary Vaccination Series Type Among People With HIV (PWH) and People Without HIV (PWoH)

Table 1 ).
19 breakthrough illness risk by HIV status.Adjustment factors included sex, race, ethnicity, age, additional vaccine dose following primary series, prior COVID-19, obesity, diabetes, hypertension, ESKD, SOT, RA, SLE, cancer, and cohort.Among PWH, prior AIDS diagnosis, HIV viral suppression, and CD4 cell count were evaluated as risk factors for severe COVID-19 in a subgroup analysis.Sensitivity analyses excluded participants without first or second diagnosis code rankings.Analyses were conducted with R statistical software version 4.1.2(R Project for Statistical Computing).Two-sided P < .05 was considered significant.

JAMA Network Open | Public Health Severe
Illness After Postvaccination COVID-19 Breakthrough Among Adults With and Without HIV

Table 1 .
Characteristics of Patients at Date of SARS-CoV-2 Breakthrough Infection

Table 1 .
Characteristics of Patients at Date of SARS-CoV-2 Breakthrough Infection (continued) Death was defined as dying while hospitalized for COVID-19 or within 30 days following discharge or within 30 days of COVID-19 diagnosis among those not hospitalized.
a Mechanical ventilation was measured during the dates of hospital admission and discharge.bcAge was categorized in 5-year increments for descriptive purposes.dOtherrace includes Asian, Native Hawaiian or other Pacific Islander, American Indian or Alaska Native, and 1 or more race.eBody mass index, diabetes, hypertension, end-stage kidney disease, organ or tissue transplantation, rheumatoid arthritis, and systemic lupus erythematosus were measured as close to the date fully vaccinated as possible within the window of October 1, 2015, to the date fully vaccinated.
Figure 2. Cumulative Incidence of Severe COVID-19 Breakthrough Illness Among People With HIV (PWH) and People Without HIV (PWoH)

Table 2 .
Crude HRs and aHRs and 95% CIs of Severe SARS-CoV-2 Breakthrough Illness Among PWoH, women had a reduced risk of severe COVID-19 breakthrough illness compared with men (aHR, 0.48; 95% CI, 0.16-1.12)although the difference was not significant (Table2).There was no observed difference in risk by race, obesity, diabetes, RA, SLE or having a cancer diagnosis, but there was a 4.51-fold increased risk with hypertension and 2.53-fold increased risk with ESKD and 2.49-fold increased risk with SOT.Sensitivity analyses excluding patients without diagnosis code rankings did not qualitatively change findings.
a Adjusted for age (categorized into a 4-level group to assess a dose-response relationship among older people [ie, aged Ն55 years] who are known to be at greater risk), sex, race and ethnicity, primary vaccination series type, COVID-19 before full vaccination, comorbidities before full vaccination, immune suppressive conditions at full vaccination, and cohort.b Adjusted for the covariates in the table and cohort; 184 PWH (15.1%) were excluded because of missing CD4 or HIV RNA measurements.c The total number of severe COVID-19 illness events was 238.d Other race includes Asian, Native Hawaiian or other Pacific Islander, American Indian or Alaska Native, and 1 or more race.e Body mass index is calculated as weight in kilograms divided by height in meters squared.f There were no severe COVID-19 breakthrough illnesses among PWH with systemic lupus erythematosus, and it was therefore excluded from the multivariable model.JAMA Network Open | Public Health Severe Illness After Postvaccination COVID-19 Breakthrough Among Adults With and Without HIV JAMA Network Open.2022;5(10):e2236397. doi:10.1001/jamanetworkopen.2022.36397(Reprinted) October 13, 2022 9/15

Ventilation and Death in Severe COVID-19 Breakthrough Illness
22vere Illness After Postvaccination COVID-19 Breakthrough Among Adults With and Without HIV severe COVID-19 breakthrough illness; however, immune dysfunction with HIV may alter this effect.35,38,39Additionally,amongthefemalePWH,a high proportion had obesity, which likely influenced the increased risk of severe breakthrough illness and warrants further investigation.Despite recommendations for additional COVID-19 vaccine doses being based in part on CD4cell count,22we identified that the proportion of PWH who received additional doses was low, varied little by CD4 cell count, and was likely associated with clinical decision-making and patient preference.Prior COVID-19 has been found to reduce the risk of subsequent COVID-19 illness, breakthrough rates remained the same as PWoH.In unadjusted analyses, moderate to severe immune suppression from HIV itself was associated with increased risk of severe COVID-19 breakthrough, as were comorbidities and a recent cancer diagnosis among PWH.Mechanical ventilation and death were rare among both PWH and PWoH with breakthrough COVID-19 and were more likely to be experienced by those older than 55 years with 1 or more comorbidity, highlighting the need for targeted risk reduction measures among older and comorbid adults.