Abraham Verghese, MD; Linda Geng, MD, PhD


June 23, 2022

This transcript has been edited for clarity.

Abraham Verghese, MD: Welcome to another episode of Medicine and the Machine. I'm Abraham Verghese. It's my great pleasure today to introduce a colleague from Stanford, Linda Geng.

Dr Geng is a clinical assistant professor. She's had an amazing career for the short time she's been in the ranks. She did her MD and PhD at the University of Washington and then came to Stanford for her residency and chief residency. Very early on, she established her interest in rare diseases and difficult diagnostic puzzles. When COVID came along, she naturally gravitated toward caring for people with the puzzling, lingering symptoms that some people experience after acute COVID infection. Linda, welcome to our podcast. It's a great pleasure to have you here.

Linda Geng, MD, PhD: Thank you so much for having me. It's an honor and pleasure for me.

Verghese: I remember when you were my chief resident and I asked you about your PhD research. I recall you telling me about a unique experience with a patient with a rare disorder that happened to be precisely what you were researching. Tell us about that.

Geng: My PhD work focused on trying to understand the molecular mechanisms of a rare inherited muscular dystrophy called FSHD, or facioscapulohumeral muscular dystrophy. It was a sandwich-type MD-PhD program. You did 2 years of medical school, then you went off to the graduate school to do your PhD, and then you returned to the clinical world for the last 2 years.

So I had been in the depths of doing basic research and bench work just before I returned to medical school as a third-year medical student. I was on my neurology rotation and one of the residents said," You're about to see a patient who has an unusual and rare condition that I didn't even know about. I hope you can learn a lot from this case.

I walked into the room and I looked at her chart. When I introduced myself, I said, it's really a privilege to be able to meet you. I just spent multiple years trying to research and understand your condition.

When she heard this, she broke down in tears and started to tell me about her long diagnostic journey and how challenging it had been and even after a diagnosis, the challenge of being able to find clinicians who were knowledgeable about her condition. It had been an isolating, tough journey for her. But it culminated in meeting someone who had dedicated years of her life trying to understand and help patients with her condition.

This experience greatly impacted me because there's the love of science and the love of discovery. But when we can apply it to patients and their suffering and help improve their lives, that is what brings meaning to the work we do.

Verghese: It must have been deeply meaningful to the patient. Have we made advances in what I was taught was a fairly irreversible, debilitating myopathy that gradually significantly impairs the patient's life? Has the research led to a therapeutic breakthrough?

Geng: This is collaborative work, internationally as well. For decades, even though the mutation and the genomic changes were known, the molecular mechanisms were poorly understood. Collectively we've made great strides and provided the foundations for therapeutic interventions, which are currently being studied. Advancements have been made.

It's been a great opportunity to work in this area. It drew me to challenging clinical problems, where teamwork and the merging of clinical medicine and scientific inquiry and discovery have enlivened my passions, in particular, for patients who have long journeys and finding ways to help shorten them.

Verghese: I remember having a discussion with you when you were in your chief residency. You knew you wanted to study rare diseases. There already is a discreet undiagnosed disease network pathway, which is focused on the genome and so on. At the time, I believe you recognized that your interests might be broader than that.

Explain how your path evolved into what is essentially a new specialty that you helped define. It existed, but you put parameters around it in a seminal The New England Journal of Medicine article. Tell us about how that evolved and about your understanding of consultative medicine.

Geng: After that first impactful moment, I had further poignant moments with other patients — those who suffered diagnostic delays, those who had mysterious symptoms or puzzling conditions. It was clear that there was this bucket of heterogeneous patients who didn't have the same ailment, but they all shared common obstacles and challenges in their encounters with the healthcare system. If there wasn't a clear diagnosis or something obvious in terms of fitting into a subspecialty or specialty realm, they got lost in the system. Their journeys were prolonged and very complex, posing a huge burden in terms of patient suffering as they try to navigate the healthcare system.

But also, there wasn't a clear home for their reevaluation in our fast-paced modern healthcare system. It became increasingly evident that it wasn't just patients with rare conditions, although that definitely was the initial draw for me. It's a huge collective group of patients. Each disease may be rare, but when you think of it collectively, they are quite common.

There are also patients with atypical presentations of conditions, complex multisystem issues, or multiple conditions all presenting in one patient and even novel, poorly understood and not well-defined conditions, such as long COVID and this post-COVID phenomenon.

These patients faced similar challenges in the healthcare system. Where do they go? They have prolonged medical diagnostic journeys, and often fall through the cracks. It was clear that there was a gap in care and a need to improve our approach to patients who suffer from these puzzling and complex conditions. That's when I started to develop a broader approach as well as an interest in what we've now defined as consultative medicine. It's a focused reevaluation and a generalist approach to a second opinion and a diagnostic evaluation.

Verghese: Centers like yours did exist, most notably at the Mayo Clinic, where one could come in and in a couple of days be seen by a "quarterback" internist. After studying your records, you'd be scheduled to see who you needed to see in the course of 2 or 3 days. That kind of efficiency is hard to match. What you have set up is terribly important and fills such a need.

I don't know that you are always successful. Sometimes it's a laborious process of coming to terms. But can you give us an anecdote of the kinds of patients you look back on with great satisfaction because you clearly made an impact in their diagnostic journey?

Geng: One of the first patients I saw stands out because she made me realize that we have to be able to change our approaches or think about ways — for diagnostic safety and diagnostic excellence — to catch these cases earlier in terms of evaluation and the time needed to seek second opinions.

A woman in her 50s had suffered severe abdominal pain for 4 years. This was quite a long journey with lots of diagnostic testing, including an exploratory laparoscopy. She got to the point of having her abdomen opened up just to see what was going on. And even after that, she had no clear answer.

At some point, she said, I got sick of going to the emergency room because it was always the same; they would give me pain medications and send me home. She saw a vascular surgeon who was going to operate on her for another putative diagnosis, but the vascular surgeon and the patient were both reluctant to do this again.

After a discussion with the patient and working as a team, we decided to look at the history again. It turns out that 4 years earlier, she had been prescribed a combination medication that included an angiotensin-converting enzyme (ACE) inhibitor. And ultimately, what was causing her episodic, unusual abdominal pain, sometimes with edema of her small bowel, was this ACE inhibitor.

She had a rare condition called ACE inhibitor–induced angioedema. It took a while to dig deep and understand that, but we worked as a team to come up with ideas.

Again, it's about spending time with the patient to review all the details and all the history. As they often say, the mystery's in the history. After she stopped the ACE inhibitor and until now, she has been abdominal pain-free.

How do we improve that process? How could we prevent this kind of 4-year journey for some of our patients? And not every patient gets a satisfying answer.

For me, one draw to long COVID is that we don't know exactly what's going on. There is a lot of work to be done in terms of helping patients, validating their suffering, and thinking of pathways to help them manage the way they live as we further our research and understanding of these puzzling conditions.

Verghese: I have to admire you. I know you have two young children, and you were managing to juggle two children and your career — homeschooling, I imagine, when the schools were closed because of COVID. Clearly, early on, you identified that your clinic was probably a natural place to study patients who were having lingering symptoms after COVID infection. How often do people continue to have symptoms long after we assume the infection has cleared? And has it actually cleared?

Geng: I work with incredible people who have been a wonderful support network as well as amazing colleagues who work together with me as a team. The teamwork is so important when we're tackling challenging problems.

Even before COVID, when we looked at several years' worth of patients who came to the consultative medicine clinic, about one fifth of them had preceding viral infections or viral-type infections that resulted in puzzling symptoms. Clearly this was an issue. We know that postviral and postinfectious phenomena are poorly understood and tend to be a neglected area of medicine. This was a major motivation for me. When the pandemic hit, we knew that this would be a huge public health problem and indeed, it is.

In terms of the rate of long COVID, you'll see different reports of the estimated numbers. Part of the reason for the variability is related to the types of studies, the populations being studied, and the period of time during the pandemic of the different strains, et cetera. We're coming to a consensus that a proportion of those who have COVID infection followed by persistent symptoms that we currently define as long COVID is in the range of 10%-30%.

There's a great dashboard on the American Academy of Physical Medicine and Rehabilitation website that shows the estimated rates of long COVID cases. If we go by the estimate of about 30% of survivors of COVID infection, well over 20 million Americans have or have had long COVID. That's a huge number. The latest UK National Statistics estimate that 2% of their population have likely had or have long COVID. So we're talking millions and millions of people and likely a global phenomenon in the wake of COVID infection. It's a huge problem.

Verghese: How do we define it? The list of symptoms of long COVID are fairly nonspecific and once people recognize that there is such an entity, there's a suggestibility factor. How do you define long COVID? Or do you just take all comers who have bothersome symptoms?

Geng: It's a big challenge in our field right now as we study long COVID. Even the estimates about the rates of long COVID vary. Also, because we don't have a great definition, there are varying definitions.

A couple of the accepted definitions are based on timing. The Centers for Disease Control and Prevention, for example, says it's symptoms that persist after 4 weeks. The National Institutes of Health calls it post-acute sequelae of SARS-CoV-2 infection (PASC) and also gives that 4-week time frame. The World Health Organization and some of the other health agencies define it with a 3-month time frame. So, there's variability in the duration of symptoms that we would consider long COVID.

You also bring up a great point in terms of the clinical criteria and manifestations. The list of the symptoms are in the hundreds based on review papers.

This is where a comprehensive evaluation comes in — when you suspect that a patient's lingering symptoms could be attributed to COVID. It's important that clinicians and those at specialized centers evaluating patients with presumed long COVID, that we do a thorough evaluation when patients present with nonspecific symptoms (such as fatigue) that could have multiple other causes.

Currently, it's a diagnosis of exclusion, where the timing matches one of these definitions and patients have lingering symptoms that are not clearly explained by another cause. That is very important because even in our long COVID clinics — ours and others around the country — there are definitely instances when we diagnose other conditions that have caused patients' symptoms. And it doesn't necessarily end up being long COVID.

Verghese: What sort of generalizations or truisms can you share from your experience with long COVID so far? What is your sense of the common symptoms? What is the natural history, and can we affect it?

Geng: One thing is certain: Long COVID is heterogeneous. The spectrum of the symptoms as well as the manifestations are highly variable among the patients. It's pretty well accepted that long COVID is likely an umbrella term that encompasses many subtypes within this phenomenon of post-COVID effects.

That may include post-ICU and post-acute COVID organ damage. But it also encompasses another group, people who may have had mild symptoms to start with and yet still have lingering symptoms. It's not clear to us whether they also exhibit certain subtypes. There is clear overlap with other syndromes, such as chronic fatigue syndrome (CFS), myalgic encephalomyelitis (ME), postural orthostatic tachycardia syndrome (POTS), and functional gastrointestinal disorders.

Clearly, some issues come about under the umbrella of long COVID and are reminiscent of other overlapping similar and possibly related conditions. That's where the exciting discovery is right now in research about the pathophysiology and the mechanisms. Not only could we understand and shed light on long COVID, but potentially we could be shedding light on many other poorly understood conditions that have existed before COVID but can be triggered by viral infections, such as CFS and POTS.

As our understanding of the subgroups and clinical phenotypes expands, we can be more conclusive in our studies to understand the pathophysiology and the treatments that can target those pathways in the clinical subgroups. These efforts are taking place in parallel, given the urgency and the public health impact of long COVID. Ideally, we would understand the underlying mechanisms and then be able to apply the right interventions.

Right now, we have to push forward in full force on all fronts — the clinical phenotype, the understanding of underlying mechanism, the pathobiology. At the same time, we have to push forward with intervention studies, and I think you'll be seeing those very soon in the pipeline. Clinical trials are underway. But we need rigorous practice-changing trials that will leverage the knowledge we have about related conditions and the hypotheses that already exist about long COVID and be able to apply those to the right populations and measure the right things. The other challenge in doing this research is that we don't have clear biomarkers to study and follow. It will be clinical outcomes at this point.

Verghese: I would worry about those clinical trials. They're so critical, and yet, as you said, it's such a heterogeneous group that it's difficult to compare one trial to another. What is going on at the viral level? What is happening with the immune system at that point?

Geng: Great question, and it's certainly an active area of research. There are multiple leading hypotheses about potential causes and mechanisms of long COVID. In the acute setting, we know that there can be variability in viral load as well as immune reactions, including cytokine storms and inflammatory responses. In the post-COVID period, we see more variability. We're still trying to understand that.

One of the leading ideas is that immune dysregulation — potentially autoimmunity, or persistent inflammation — occurs. That's a very attractive hypothesis and has gained a lot of traction in the field. There's also the idea that there may be persistent viral fragments. It's not clear if these are actively replicating viral reservoirs, but that's another interesting leading hypothesis.

Growing data show that viral particles remain in some individuals. The question is whether that is what's causing the long COVID symptoms. And if so, how do we intervene? How do we show causation and, essentially, affect treatment? With our current available oral antiviral agents, there's obviously a clear role and a good opportunity to study that question.

There are other leading hypotheses but it may be that in any given individual, it's a combination of a couple of pathways. Perhaps in another individual, it's another combination. All of these need to be studied and further evaluated, both mechanistically and therapeutically.

Verghese: You happen to be working side by side in the same clinic with the ME/CFS group. The idea that viral infections can cause a prolonged post-viral syndrome isn't new. We know that happens with Epstein-Barr virus. But do you see pathophysiologic similarities and similarities in the expression of symptoms?

Geng: Yes. If we look at the Institute of Medicine (IOM) criteria from 2015 for ME/CFS, the cardinal symptoms that define that condition based on the diagnostic criteria are debilitating fatigue that lasts for more than 6 months, post-exertional malaise — that's a cardinal symptom — and unrefreshing sleep, in addition to brain fog or orthostatic intolerance, or both.

If you look at the most common symptoms being reported by long COVID patients, those top the list. In our own clinical cohort, about half of our patients would meet the criteria for ME/CFS if we consider greater than 6 months' duration and many of those factors, particularly if severe and debilitating, impact the patient's function in multiple domains. So yes, we definitely see significant similarities.

This is from our clinic, and there are similar data from others as well. I suspect that this pattern will hold and there will be a significant overlap.

But the question remains, does that mean they share the same underlying pathophysiology? I suspect they do. We know that CFS and other postviral fatigue syndromes can be triggered by viruses that were around before COVID.

Could there be common mechanisms at play? I think that probably is the case. For example, cytokine abnormalities have been shown in ME/CFS and, similarly, some cytokines, in particular IL-6 [interleukin 6] and others, have been shown to be elevated post-COVID. More research still needs to be done in this area, but my suspicion is we'll probably be finding some shared pathways.

Verghese: A game changer could be the oral medications that are now available. Paxlovid, for example — anecdotal reports indicate that it can impact long COVID. But we're hearing, more in social media than in real publications, reports of a rebound after the 5-day course of Paxlovid. What is your take on that whole issue? Can you connect that to long COVID symptoms?

Geng: Paxlovid was a game changer as an oral medication that had good data in terms of reducing the risk for severe disease and hospitalization. That is still true. For high-risk patients, it is the best we have right now, especially for outpatient treatment.

In terms of a rebound, the CDC put out an official health advisory at the end of May. So it is a recognized phenomenon. They say that after recovery from an acute infection, perhaps from day 2 after recovery to day 8, that if you have either recurrent symptoms or another positive test, you may be experiencing a rebound after a Paxlovid treatment course of 5 days.

But based on what I know so far, most people who rebound don't go on to have severe disease or severe symptoms. Still, it raises a public health issue about whether these patients are potentially contagious again and could be spreading the virus. If so, the recommendation is to go back into isolation.

In terms of what data are out there — and this is important — it's a lot of anecdotes. Ultimately, we need big studies to be able to draw good conclusions.

A study from Mayo Clinic looked at about 400 patients who were treated with Paxlovid for high-risk COVID infection. They found a fairly low rate of rebound of I believe less than 1%. This is similar to the numbers released by Pfizer; about 1%-2% experience a rebound.

Anecdotally, there's been a lot of talk among my own colleagues about the rebound they are seeing. Maybe more studies will show different numbers. Currently, according to the data that is out there, the rate of rebound is relatively low, and when it does occur, it tends to be relatively mild. More studies may be able to show how much of the anecdotal evidence holds up and whether the rate is higher than what is being reported now.

In terms of long COVID, it's that interesting question about whether antiviral therapy might impact long COVID symptoms if the hypothesis about persistent viral reservoirs holds true. Paxlovid would definitely be the most obvious choice in terms of ease of use, availability, and oral preparation.

Verghese: Do we know what percentage of long COVID patients have measurable viral replication activity?

Geng: The data are not available yet to understand that fully. Even the accumulating data that show persistence in viral shedding or presence of viral particles show only that they are either RNA or protein particles. To my knowledge, no one has shown that these are actively replicating viral reservoirs. Whether or not antiviral therapies would eliminate them is very much an active area of research.

I also haven't seen large studies or repeated studies showing how well the presence of these viral particles correlates with symptoms. There are two things that we need to address here. If they're present in some patients, do they account for the symptoms? And then, are they actively replicating? Or would antiviral therapy target those viral reservoirs? These are important unanswered questions.

Verghese: What makes this so confounding is that we already know that different strains are probably associated with different incidences of long COVID. We have Paxlovid studies with one strain being compared with results in real-life with what is now probably a new strain. Long COVID is hard to define because we have so many confounding variables. I'm glad it's in your good hands to do clinical trials in this area. I know you're actively involved in some of these trials. I hope that in a couple of years, we'll have some better understanding of this, but then the target keeps moving.

Geng: A study just came out in The Lancet, I believe, that suggested that the risk for long COVID after Omicron is lower than the risk for long COVID after Delta. But if we think about how many cases of Omicron we've seen, we are going to see a big wave of long COVID.

Verghese: It's been a pleasure to have you here. I'm so glad someone like you and colleagues like you across the country are focusing on patients who don't neatly fall into the comfortable specialty or primary care buckets because it's terribly important. I read in your paper that roughly 20%-30% of primary care patients don't neatly fall into a diagnostic bucket. And their symptoms are puzzling.

With COVID and the aftereffects of COVID infection, we're going to see more and more patients who require this kind of sorting out. I'm glad you're doing this and I hope more people will take up this challenge. You helped to define the field of consultative medicine by calling it what it is. To me, that moment was as significant as when Wachter and Goldman wrote the paper defining hospital medicine. I believe this needs a name and a movement behind it. It needs reimbursement, because you can't do this in a vacuum.

Any closing thoughts for us, Linda?

Geng: It's been a pleasure and a true honor to be able to talk about this with you. What drew me to long COVID is that it is an example of what many patients are suffering from — having complex multisystem or complex conditions that elude standard diagnosis or clear subspecialty domains.

They tend to fall through the cracks of our healthcare system. We need to leverage great teamwork and the multidisciplinary approach that I'm seeing across the country for long COVID. This is great, but we also need to enlist the resources that are needed to promote it.

I'm hoping that we can learn things from long COVID that apply to other conditions that are puzzling and also develop these models of care that can be applied more broadly. We need teamwork and multidisciplinary care, time, and the resources in terms of reimbursement and funding, and to be able to integrate clinical work and research. These are all key and will change the way we deliver care and advance the care of these patients.

Verghese: One of the courses you're teaching at Stanford is about medicine on TV vs real-life. Medicine in real life doesn't work the way Dr House or the other brilliant television diagnosticians practice it. It's not a one-person operation. It's a team effort. I'm glad you're on the Stanford team.

You've been listening to Medicine and the Machine. My guest today was Linda Geng. Please join us again.

This podcast is intended for US healthcare professionals only.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


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.