Daniel Musher, MD, on the Etiology of Community-Acquired Pneumonia

In this podcast, Daniel M. Musher, MD, answers our questions on his systematic review of the etiology of community-acquired pneumonia and briefly discusses his extensive research on community-acquired pneumonia.

Additional Resources: 

For more information on community-acquired pneumonia, visit our CABP Resource Center.

Daniel M Musher, MD, is a distinguished service professor of infectious disease, and a professor of microbiology and molecular virology at Baylor College of Medicine, as well as the former chief of the Veterans Affairs Medical Center in Houston, Texas. 



Leigh Precopio:  Hello everyone, and welcome to another installment of Podcasts360, your goto resource for medical news and clinical updates. I'm your moderator, Leigh Precopio, with Consultant360.

Communityacquired pneumonia, or CAP, is a leading cause of death in older adults. To successfully treat this infection healthcare practitioners must begin with a solid understanding of its causes.

To learn more about the etiology of CAP, I am joined by Daniel M Musher, MD, who is a distinguished service professor in the infectious disease department at Baylor College of Medicine, and former chief of the infectious disease section at the Veterans Affairs Medical Center in Houston, Texas.

Thank you for taking the time to answer my questions today, Dr Musher. Your article notes that concerns about treatment recommendations and antimicrobial stewardship prompted this systematic review of the literature on the etiology of communityacquired pneumonia. To begin, could you discuss why this is an important objective to research?

Daniel Musher:  Pneumonia is probably the most common infection that leads people to be hospitalized. And it remains a serious disease. There’s still substantial morbidity and mortality. People who have pneumonia, the older they are the more there are lingering after-effects, which can be explained on a very complicated molecular basis. It’s called epigenetic changes, and I had an article on that in the New England Journal of Medicine last year.

The point is that there's mortality. There's illness. And people who get over pneumonia, they get over the acute phase, they’re often sick for months. And as I say, the older you are, the longer that lasts. So, I think it's important to study pneumonia. I work at a Veterans Hospital. I see older adults. I'm older. I've got friends who are older. And I have seen the effects of pneumonia. I think it's a tremendously important disease.

That's about pneumonia. You asked me why is it important to study etiology? Well, it seems to me only logical that if the disease is that important, we should know as much about it as we can, and the cause of the disease is about as important as anything understanding it. So therefore, I've studied the etiology of pneumonia.

Leigh Precopio:  In one of your most recent literature reviews, your team found that the most common cause of CAP was Haemophilus influenzae, followed closely by Streptococcus pneumoniae, then Staphylococcus aureus, and gramnegative bacilli. Is this a result that surprised you or did you anticipate this?

Daniel Musher:  I was not surprised by the results of the literature review. Streptococcus pneumoniae, in the presecond World War data, was overwhelmingly the most common cause of pneumonia. In the 1920s and '30s, 90% or more of all pneumonias were caused by Streptococcus pneumoniae, which is called pneumococcus.

In those days, they didn't do appropriate cultures that would have detected Haemophilus influenzae, but still everybody agreed that it was pneumococcus, far and away the most common cause. That number started to decline after the second World War because attention began to be paid to other organisms which hadn't been paid before. And actually, Abers, Bartlett and I wrote an article on the decline in pneumococcus as the principal cause of pneumonia. We published that article in Clinical Infectious Diseases about two and a half years ago.

In the studies that I have done most recently, interestingly Haemophilus just edges out pneumococcus. Basically, they are the two most prominent bacteria that cause pneumonia, leading to hospitalization. They're followed by Staph aureus. Those three are far ahead of the gramnegative organisms.

Now what's important about that is the literature and the guidelines committee overemphasizes the role of gramnegative rods, especially Pseudomonas. The CDC published a major study in 8 hospitals of the causes of pneumonia. And they published that one in 2015. I published a 1 hospital, or a 1site study, exactly the same study 2 years before. Their study is bigger and includes more kinds of people. My study is better because it was tightly controlled. If you've got 1 investigator at 1 place, and he's totally committed to it, it's going to be better data.

And the yield in both studies was not very good. But pneumococcus was the most common of the bacterial pneumonia. Haemophilus was the second.  In their study, Pseudomonas caused 1% of pneumonia. Just 1%. I keep coming back to that organism that’s a gramnegative bacillus. The other gramnegative bacilli caused just a few percent. In my study it was similar.

Basically then, the principal organisms are pneumococcus, Haemophilus, followed by Staph aureus, followed by Moraxella. Then normal respiratory flora, in my most recent study, are the ones that are just behind pneumococcus, and Haemophilus, and Staph.

The data are slightly shifting. Those results have, I think, major implications for infection control and for confining the use of antimicrobials. My field has become so ingrained to think about Pseudomonas. I can actually give you a long talk because I'm old enough, I remember when they started talking about Pseudomonas and why. But it doesn't matter. Pseudomonas is not an important organism causing communityacquired pneumonia. All the young doctors talk about “covering for Pseudomonas, and covering for gramnegative rods,” and it's just not necessary. So that's the implication of this kind of research.

Leigh Precopio:  Could you discuss why Streptococcus pneumoniae has been demonstrated to be a less important cause of pneumonia in the United States than it has been in other countries?

Daniel Musher:  There are two reasons for that. The first is that the American public has been very good about taking the socalled pneumonia vaccine, which is a vaccine that is directed against pneumococcus. About 68% of all the United States adults have had a dose of pneumococcal vaccine somewhere along the way, and it's a fairly effective vaccine.

The second reason is that cigarette smoking in American adults has definitely declined whereas it hasn't budged in Europe. I don't know about other countries, but I suspect it hasn't been other countries either. There's seems to be a very direct relationship between cigarette smoking and pneumococcal pneumonia. That's been demonstrated in mice studies in this country.

I think those two factors, the pneumococcal vaccine uptake by older adults, and the decrease in smoking by US adults, have been responsible for the decline in pneumococcal pneumonia in the United States.

Also, of course, we were the first country to adopt a widespread vaccine for children. The vaccine for children prevents colonization. If the organisms aren't present in little children, they don't spread to adults. Some of the most common pneumococci, they've always caused pneumonia, have declined greatly in the population. Other countries are forced now giving the conjugate pneumococcal vaccine to infants and toddlers, but we were at the forefront of that. We began in 2000, so that also had a big effect.

Leigh Precopio:  Could you discuss other treatment options that may improve antibiotic stewardship and reduce potential antibiotic resistance, such as the routine use of sputum gram stain and culture?

Daniel Musher:  I would love to address that question. In the study that I published in the winter at the end of 2020 in Open Forum Infectious Diseases, I studied patients who are admitted to the hospital and were able to provide what's called a good quality sputum sample within, let's say, 18 hours of admission. Which is before antibiotics had an effect on what was present in the sputum.

In those people I was able to identify a cause, an etiologic agent, in 95% of cases. No one has ever come anywhere near that. I mentioned that big CDC study, they established the diagnosis in 45% of the cases. In my big study, I established an etiologic diagnosis in 55% of cases. That's a huge difference between 50/50 and 95%.

I did it by examining gram stains, and then I did quantitative studies. I actually counted the numbers of bacteria in the sputum to show that what you count and what you obtain as the number matches the gram stain. No one's ever done that either.

The gram stain is extremely reliable. And that is if it's done well, and if it's read properly. There's a lot of subjectivity in those readings. Of course, I pride myself on being an expert in that and the people in my lab have become expert, not my lab, but my hospital lab, just because I'm there all the time I keep showing them. I can't say that every other laboratory or even, I can't say that most other laboratories can do that. Nonetheless, the gram stain is distinctive. It takes 5 minutes to do it. That's all it takes.

If a patient is in the emergency room, and has a pneumonia, and coughs up a sample. If that sample can be carried off to the laboratory and somebody can just gram stain it and read it – which our laboratory can do, all day and in the evening hours – then you have a result back in half an hour. There's absolutely no adverse consequences of waiting a half an hour to give an antibiotic. None. Even an hour or 2 makes no difference.

Then if you look at the gram stain and it has pneumococci, you treat with amoxicillin. That's a narrowspectrum antibiotic. If you're giving it intravenously, you could treat with penicillin, but you got to give that every 4 hours. You could treat with ampicillin, you got to give every 6 hours. The reason we use ceftriaxone which has a broader spectrum is that you give it once a day. But still, that has less impact on the body's bacterial flora than some of the other things that people choose. A gram stain can give you a diagnosis very promptly.

Now the guidelines committee, of which I am a member, staunchly refused to recognize what I've just told you. The guidelines committee never discussed the etiology of pneumonia. They just kept putting it off. They submitted the first draft of the guidelines document with a recommendation that microbiological studies, that means a gram stain and culture, not be done in patients admitted for pneumonia unless they were admitted to an ICU or had risk factors for Pseudomonas. That's all that ridiculous emphasis on Pseudomonas. It was over my vigorous opposition, I didn't get through to them, but I think the Infectious Disease Society board wanted them to put in a section on etiology, so they came back and asked me to write it. Then they modified their wording so it doesn't say that gram stain and culture shouldn't be done. It just doesn't say gram stain and culture should be done. They don't recommend it at the present time. I just think it's a dreadful error. They say it doesn't affect the outcome. It doesn't affect the outcome because they're creating with broadspectrum antibiotics, and if they can narrow them that would affect the outcome. That's a very complicated question.

Leigh Precopio:  What other knowledge gaps exist regarding the etiology or empiric of treatment of CAP?

Daniel Musher:  The use of other agents that alter the immunologic response to infection would certainly be an area that would be ripe for investigation. For example, the guidelines do recommend the addition of azithromycin, which is a macrolide, to ceftriaxone for the standard treatment of pneumonia.

Now they recommend azithromycin for two reasons. First, because it does treat certain socalled atypical bacteria like Mycoplasma and chlamydia. These organisms cause pneumonia in younger adults, generally not serious, and almost never in older adults.

Once again, the CDC study showed these organisms cause no more than 3 or 4% of all pneumonias, and they had patients of all age. In my hospital, it almost doesn't occur.

The other reason azithromycin is added is because it is called an immunomodulating drug. It suppresses the immune response. Once you treat the causative organism, the immune response is basically getting in the way of healing, because the symptoms that we have are not just a result of bacteria being there in the lungs. They're a result of the immune response to those bacteria. That's what causes the fever, and the chills, and the sputum production. The immune response is what makes you sick. It’s an inflammatory response. If you give an effective antibiotic that is killing the bacteria, then you can suppress that immune response a little bit, the patient will probably do better because the immune response will cause the prolongation of the symptoms. There are data that suggests that there is a better outcome if azithromycin is given, along with ceftriaxone to defeat communityacquired pneumonia. It's not because of its antibacterial effect. It's because of this antiinflammatory effect.

There are other agents that could be given. There's a whole host of them. In the case of COVID19, there are specific antiinflammatory agents that are given that sre directed against certain white blood cells. And they do reduce the mortality from COVID. That's a whole area of research.

It's difficult to do such research. It’s difficult because pneumonia has many causes. And I would think any drug that would be used would be more relevant to pneumonia due to certain causes than pneumonia due to other causes.

You got to begin by looking at which is the etiologic agent of the pneumonia, and then figuring out which group you want to treat with your new drug. And nobody seems to be interested in doing the studies on the etiology of the pneumonia. I just don't think they're going to find any beneficial effects from any of these drugs, except in a very broad sense, but I think that's a big area for future research.

Leigh Precopio:  Thank you for answering all my questions today, Dr. Musher.

Daniel Musher:  Nice talking to you. Take care now.


Submit Feedback