Long COVID Lung Damage and Immune Dysfunction Across the Lifespan: Jerry Krishnan, MD, PhD

In an interview with The American Journal of Managed Care^® (AJMC^®) at the American Thoracic Society International Conference 2026 in Orlando, Jerry A. Krishnan, MD, PhD, associate vice chancellor for population health sciences at the University of Illinois Chicago, discussed findings presented during the session “NIH RECOVER and C4R Cohorts: Pulmonary Complications and Immune Dysfunction in Long COVID Across the Lifespan.”

In this Q&A, Krishnan highlights how long COVID, heterogeneity, respiratory complications, and immune dysfunction are shaping future research and clinical care.

This transcript was lightly edited for clarity.

AJMC: The broader session highlights persistent respiratory symptoms, declines in lung function, and both restrictive and obstructive defects in patients with long COVID. From your perspective, what mechanisms appear to be driving these ongoing pulmonary complications?

Krishnan: First, there is a subset of individuals who, following the acute infection with SARS-CoV-2, end up essentially developing COVID pneumonia, and depending on the host response to that pneumonia, could then end up developing interstitial lung disease in people with long COVID. Scarring of the lungs then leads to a number of functional deficits, including the need for supplemental oxygen, poor exercise tolerance, and so forth.

The fortunate news is that it looks like a minority of individuals after COVID go on to develop virus-induced interstitial lung disease. Another type of interstitial lung disease that can be developed is also the need for mechanical ventilation, and any lung injury that can occur as part of the supportive care, as people are recovering from severe pneumonia. That's one type of lung disease that I would say is a lot less common than we had originally worried about.
We're also noticing that as people develop tissue lung disease, in many cases, it actually improves over time, over 6 to 12 months or even longer. There are, however, individuals who continue to require supplemental oxygen because of diffusion defects or enough structural damage to the lung that they can't adequately oxygenate the blood.

A second type of lung disease that can develop is really airway hyperreactivity, cough, and wheezing, almost like an asthma-like phenomenon, and that part is much less well characterized right now, but we do have several or many instances in which patients will describe chronic cough, and it seems to come from airway hyperactivity or aberrant repair of their respiratory epithelium. Sometimes it happens in individuals with pre-existing asthma who went into remission; they then develop a SARS-CoV-2 infection, and then it almost brings the asthma back in those individuals. That pattern is a little less well-described, but clinically, we're noticing it.

A third phenomenon that's again much less common, fortunately, with the current viral strains is really homeovascular disease, including thromboembolic disease and pulmonary embolism. Again, much less common now than the acute phase, when we really were worried that this was going to be a major component of long COVID.

AJMC: Long COVID is often described as highly heterogeneous, with patients experiencing very different symptom patterns and disease trajectories. How is the RECOVER initiative helping researchers better phenotype or classify patients with long COVID?

Krishnan: This is probably one of the most important contributions of the RECOVER (NCT05172024) initiative because of the size of the cohort study, over 15,000 adults and a similar number of children. Then we had pregnant women. Also, as part of the program, about 30,000 individuals have been well characterized and followed for multiple years, and what we're noticing is that this is a recent publication that came out looking at symptoms over time.

And there are multiple distinct trajectories of symptoms, so either they are persistently abnormal and highly symptomatic from the onset of long COVID, meaning right after the acute infection and complications resolved within a few months, or they go on to continue to have really disabling and impairing respiratory symptoms, all the way to individuals who had relatively minimal or no symptoms after a relatively mild case of COVID. And then months later, they start to develop abnormal symptomatology.

What we're noticing is that there are multiple distinct trajectories. We're recognizing now that long COVID is not just a continuation of problems people may have had during the acute phase; you can also have various trajectories, all the way to a delayed onset. These are based on symptoms.
What's less well characterized are the cardiopulmonary or pulmonary aspects of those trajectories.

There are now multiple manuscripts under development to look at that very carefully. And then I'm really pleased to say that the federal government, through the RECOVER initiative, has now funded an extension of the adult cohort study, where we're going to be following people even longer, starting in the next month or so, for the next several years. We're going to have roughly a 5- to 8- to 9-year trajectory that we will be able to pull together.

AJMC: Your work has focused extensively on asthma, chronic obstructive pulmonary disease (COPD), and health care delivery science. Are there parallels between long COVID-associated respiratory dysfunction and chronic obstructive airway diseases that clinicians should be paying closer attention to?

Krishnan: As clinicians, we always have to draw on our past experience to see if the pattern is repeating itself, and I would say that there are some unusual features of long COVID. They're kind of new ground for most of us as pulmonary or critical care physicians, but there are also some familiar patterns. Let me start with the familiar. The first is that we are now recognizing that asthma and COPD have multiple underlying endotypes, underlying biology related to pathogenic mechanisms, where there isn't only one type of asthma; there are multiple types of asthma. Think of TH2-driven or non-TH2-driven asthma.

Likewise, in COPD, we're realizing that there are also these different endotypes, and now there's interest in looking at imaging as well to help us understand COPD. Similarly, the common frame here is that long COVID, even on the respiratory side, is not one disease; it's multiple different conditions, again likely due to which viral strain was involved in the host response, your immune response to the viral infection, so heterogeneity is a common sort of platform here.

The second component, I would say, and this is the unusual part of long COVID, is that, although it's a respiratory virus, it's really creating a multisystem problem. The lungs become the portal of entry, but the problems with lung core really are systemic, and they can affect the brain with neurocognitive problems, the kidneys, the heart, and even cause autonomic neuropathy. But then the virus was able to travel and create damage in multiple locations, either through the viral particles themselves or maybe even viral persistence or just an immune dysregulation that the immune system forgot to shut off once the infection went away.

And now we have an autoimmune phenomenon, so that's the little bit of a difference I would say between asthma and COPD: there are multiple mechanisms at play, and it's not primarily a lung disease. In fact, some patients' primary symptoms are neurocognitive, fatigue, or arthralgia; they're not even pulmonary.

And then the last part I'll say is asthma and COPD. These are lung conditions that we've known about for decades. Long COVID is the newest sort of entrant into chronic disease, and we are very early in understanding the biology of it and how to treat it.

Currently, there are no proven effective treatments. Starting with some humility, listening to your patients and validating their symptoms and then indicating that their symptoms fit or do not fit the current long COVID definition of the US, which is essentially chronic symptoms for 3 or 4 more months after having had COVID, could be waxing and waning; it could be multisystem, but it's a very broad sort of definition right now because we're waiting for biomarkers and other diagnostic tests to help us.

Many patients will tell you that they feel unheard because most people don't know about long COVID, and that, as a pulmonologist, just because you don't find structural lung disease to explain their shortness of breath doesn't mean they don't have shortness of breath. There are many other nonpulmonary mechanisms that we're investigating.

AJMC: As clinicians continue to encounter patients struggling with long COVID symptoms years after infection, what practical strategies or multidisciplinary approaches do you believe are most important for improving patient care and quality of life right now?

Krishnan: This is really fundamentally one of the most vexing components of this: that long COVID is real. There are millions of Americans who have it, and many more millions across the world. We don't fully understand what causes it, and we don't have any proof of effective treatments. The question is, what can you do in the meantime?

Step one is to listen to your patients, take a history, and determine what their symptom pattern is. It meets the National Academies of Sciences, Engineering, and Medicine definition for long COVID, and then validates that individual concern that although your blood tests are normal or your chest x-ray is normal, your symptom pattern does indeed fit the definition of long COVID, and that's the first step is to recognize that not all illnesses have a biomarker and we're in the frontier here for long COVID, and having sort of a dialogue and building trust that you're here to work with them and then being honest with them about what we do and don't know, so building a therapeutic relationship is going to be essential.

The second is that because it's a multisystem disease, I tend to frequently rely on my colleagues to evaluate the patient for nonpulmonary problems, such as neurocognitive issues, or if they have dysautonomia, dizziness, and postexertional malaise. I may have the cardiologist see the patient to determine if they have evidence of dysautonomia. They may end up having psychiatric or mental health issues. Referring them to appropriate clinicians is key.

Perhaps as we think about the multisystem nature, having primary care involved and establishing a sort of multidisciplinary approach with primary care at its core, with different subspecialists that are assisting in the management of the patient, and then communicating among all the health care providers, is key.

Number one, diagnosing long COVID with the information we currently have as a clinical diagnosis.

Second, determine whether there may be other conditions that may be mimicking that person's long COVID symptoms, so they need appropriate evaluation, just like any other medical care, which might require referral to different clinicians.

Third, really identifying if there are support groups for patients is important. There are a lot of coping mechanisms and a sense of self and a sense of support that people need, so you can connect them to either local resources or, if you go to the NIH RECOVER website, there are a number of organizations that are helping to serve in that role, and they can identify local resources for patients.

And then, finally, of course, science is on it. There's an enormous effort underway. Do we have all the answers now? No, but are we making progress? Absolutely. And giving people a sense of hope, realistically, but hope, and then enrolling them in research studies, because the only path forward really is to understand this appropriately and more quickly, coming to terms with the harms and benefits of potential therapies, and so really, as much as you can, enroll them in research studies or give them the option or refer them to someone who is doing research in long COVID.