HIV Increases Risk for Severe COVID-19

Heather Boerner

July 19, 2021

Editor's note: Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

HIV increases the risk for severe COVID-19 by 6% and the risk of dying of COVID-19 in the hospital by 30%, according to a report from the World Health Organization (WHO) on COVID-19 outcomes among people living with HIV. The study primarily included people from South Africa, but also some data from other parts of the world, including the U.S.

However, the report, presented at the 11th IAS Conference on HIV Science (IAS 2021), couldn't answer some crucial questions clinicians have been wondering about since the COVID-19 pandemic began. For example, was the increase in COVID risk for result of the presence of HIV or because of the immune compromise caused by untreated HIV?

The report didn't include data on viral load or CD counts, both used to evaluate the health of a person's immune system. On effective treatment, people living with HIV have a lifespan close to their HIV-negative peers. And effective treatment causes undetectable viral loads which, when maintained for 6 months or more, eliminates transmission of HIV to sexual partners.

What's clear is that in people with HIV, as in people without HIV, older people, men, and people with diabetes, hypertension, or obesity had the worst outcomes and were most likely to die from COVID-19.

For David Malebranche, MD, MPH, an internal medicine doctor who provides primary care for people in Atlanta, Georgia, and who was not involved in the study, the WHO study didn't add anything new. He already recommends the COVID-19 vaccine for all of his patients, HIV-positive or not.

"We don't have any information from this about the T-cell counts [or] the rates of viral suppression, which I think is tremendously important," he told Medscape Medical News. "To bypass that and not include that in any of the discussion puts the results in a questionable place for me."

The results come from the WHO Clinical Platform, which culls data from WHO member country surveillance as well as manual case reports from all over the world. By April 29, data on 268,412 people hospitalized with COVID-19 from 37 countries were reported to the platform. Of those, 22,640 people are from the US.

A total of 15,522 participants worldwide were living with HIV, 664 in the US All U.S. cases were reported from the New York City Health and Hospitals system, Henry Ford Hospital in Detroit, and BronxCare Health System in New York City. Almost all of the remaining participants lived in South Africa —14,682 of the 15,522, or 94.5%.

Of the 15,522 people living with HIV in the overall group, 37.1% of participants were male, and their median age was 45 years. More than 1 in 3 (36.2%) were admitted with severe or critical COVID-19, and nearly one quarter — 23.1% — with a known outcome died. More than half had one or more chronic conditions, including those that themselves are associated with worse COVID-19 outcomes, such as hypertension (in 33.2% of the participants), diabetes (22.7%), and BMIs above 30 (16.9%). In addition, 8.9% were smokers, 6.6% had chronic pulmonary disease, and 4.3% had chronic heart disease.

After adjusting for those chronic conditions, age, and sex, people living with HIV had a 6% higher rate of severe or critical COVID-19 illness. When investigators adjusted the analysis additionally to differentiate outcomes based on not just the presence of comorbid conditions but the number of them a person had, that increased risk rose to 13%. HIV itself is a comorbid condition, though it wasn't counted as one in this adjusted analysis.

It didn't matter whether researchers looked at risk for severe outcomes or deaths after removing the significant co-occurring conditions or if they looked at number of chronic illnesses (aside from HIV), said Silvia Bertagnolio, MD, medical officer at the World Health Organization and co-author of the analysis.

"Both models show almost identical [adjusted odds ratios], meaning that HIV was independently significantly associated with severe/critical presentation," she told Medscape.

As for death, the analysis showed that, overall, people living with HIV were 30% more likely to die of COVID-19 compared with those not living with HIV. And while this held true even when they adjusted the data for comorbidities, people with HIV were more likely to die if they were over age 65 (risk increased by 82%), male (risk increased by 21%), had diabetes (risk increased by 50%), or had hypertension (risk increased by 26%).

When they broke down the data by WHO region — Africa, Europe, the Americas — investigators found that the increased risk for death held true in Africa. But there were not enough data from the other regions to model mortality risk. What's more, when they broke the data down by country and excluded South Africa, they found that the elevated risk for death in people living with HIV did not reach statistical significance. Bertagnolio said she suspects that the small sample sizes from other regions made it impossible to detect a difference, but one could still be present.

One thing conspicuously absent from the analysis was information on viral load, CD4 T-cell count, progression of HIV to AIDS, and whether individuals were in HIV care. The first three factors were not reported in the platform, and the fourth was available for 60% of participants but was not included in the analysis. Bertagnolio pointed out that, for those 60% of participants, 91.8% were on antiretroviral treatment (ART).

"The majority of patients come from South Africa, and we know that in South Africa, over 90% of people receiving ART are virologically suppressed," she told Medscape. "So we could speculate that this effect persists despite the use of ART, in a population likely to be virally suppressed, although we cannot assess this with certainty through the data set we had."

A much smaller study of 749 people living with HIV and diagnosed with SARS-CoV-2, also presented at the conference, found that detectable HIV viral load was significantly associated with a slightly higher risk of severe outcomes (P < .039), but CD4 counts less than 200 cells/mm3 was not (P = .15).

And although both Bertagnolio and conference organizers presented this data as proof that HIV increases the risk for poor COVID-19 outcomes, Malebranche isn't so sure. He estimates that only about half his patients have received the COVID-19 vaccine. But this study is unlikely to make him forcefully recommend a COVID-19 vaccination with young, otherwise healthy, and undetectable people in his care who express particular concern about long-term effects of the vaccine. He also manages a lot of people with HIV who have undetectable viral loads and CD4 counts of up to 1200 but are older, with diabetes, obesity, and high blood pressure. Those are the people he will target with stronger messages regarding the vaccine.

"The young patients who are healthy, virally suppressed, and doing well may very much argue with me, 'I'm not going to push it,' but I will bring it up on the next visit," he said. The analysis "just helps reinforce in me that I need to have these conversations and be a little bit more persuasive to my older patients with comorbid conditions."

Bertagnolio has disclosed no relevant financial relationships. Malebranche serves on the pre-exposure prophylaxis (PrEP) speakers bureau for Gilead Sciences and has consulted and advised for ViiV Healthcare. This study was funded by the World Health Organization.

International AIDS Society Conference on HIV Science 2021: Abstract 2498 and Abstract 1099. Presented July 14, 2021.

Heather Boerner is a science journalist and author who has been covering HIV for a decade. Her book, Positively Negative: Love, Sex, and Science's Surprising Victory Over HIV, came out in 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.