From the Special Topic of
Chronic Obstructive Pulmonary Disease (COPD)
According to our Special Topics analysis of COPD
research over the past decade, the work of Dr. Bartolome
Celli ranks at #5 by papers and #6 by cites, based on 83
papers cited a total of 3,255 times. Seven of these papers
are included in the top 20 papers lists for the topic.
Essential Science IndicatorsSM from
Reuters, his record can be found in the field of
Dr. Celli is on the faculty of the Pulmonary and
Critical Care Medicine Division of Brigham and Women's
Hospital in Boston, Massachusetts.
In the interview below, he talks with ScienceWatch.com
correspondent Gary Taubes about his highly cited COPD
Your most-cited paper is the 2004
European Respiratory Journal paper, "Standards for the
diagnosis and treatment of patients with COPD: A summary of the ATS/ERS
position paper" (Celli BR, et al., 23: 932-46, June 2004).
How did that paper come about and why do you think it's been so
I was chosen by the American Thoracic Society to represent and draft
guidelines for COPD diagnosis and treatment. That's not a research paper,
per se, like the BODE index paper that is number two on the list.
It is a guidelines paper and that's why it's so frequently cited. It's the
work of a large group of individuals working together. I was the chair of
the group and put it together along with Dr. William McNee from Scotland.
Okay, so let's talk about the 2004 New England
Journal of Medicine article on the BODE index (Celli BR, et
al., "The body-mass index, airflow obstruction, dyspnea, and
exercise capacity index in chronic obstructive pulmonary disease,"
350: 1005-12, 4 March 2004). Tell us about how that research
evolved and what you were trying to accomplish.
When I started my career in the mid-1970s, there was a fatalistic attitude
toward the disease. It was truly a nihilistic attitude. It was thought that
very little could be done for patients with COPD (emphysema and chronic
bronchitis); it was felt that COPD was self-inflicted, resulting from
smoking and the only therapy was to stop smoking.
I felt this was not the case and that, like many illnesses and diseases,
there were many things that could be done. So we embarked on different
areas of research to see what could be done, as other people were also
doing at the time.
"COPD is a preventable and treatable disease."
Much of our research was based on the belief that if we can't change the
disease itself, at least we can change some of the problems around the
disease and give people a better outcome. That led us to place a lot of
emphasis into the investigation of prognostic indicators of outcome.
That implies that the existing prognostic
indicators were inadequate. Is that true?
Yes. The way therapies were being measured 20 years ago is by how well they
would bronchodilate patients, how well they would improve lung function.
And we began to realize that many other dimensions that impacted on outcome
independent of how affected lung function may be.
We began a series of studies around 1990 exploring the value of measuring
exercise capacity, shortness of breath, and other variables that could be
used as surrogate markers of outcome. This all led us to the concept that
maybe we could create an index that would be used clinically to predict
That is how we came to the paper that may be the most important paper we
have published—the 2004 New England Journal of Medicine
article on the BODE index—and it took 10 years to reach that goal.
So what exactly is the BODE index and what does it
It integrates body-mass index (BMI, what the B stands for), degree of
obstruction (the O), dyspnea (the D), which is shortness of breath, and
then finally exercise capacity (the E).
That was a significant shift in the paradigm of this field, when we pushed
the agenda stating that the final outcome of patients with COPD was not
only related to how bad their lungs were but that the intensity of other
extra-pulmonary domains also had an effect. It opened ways to perhaps
intervene in those areas and improve outcome even if we don't change lung
What is the relationship with BMI?
We don't know why it happens, but people who have low BMI, people who are
very thin, do very poorly if they also have COPD. Being a little overweight
protects you from death, which is very interesting and somewhat
And that research took 10 years?
We started that work in 1994 and it was published in 2004.
Did you start with the goal of developing a
prognostic index, or did that idea evolve with the research
In the early 1990s, we already had some indications that there were other
independent predictors of outcome in patients with COPD that were
independent from the degree of airflow obstruction, but it was in 1994 that
we thought we might be able to incorporate several of them into an index. I
guess it took longer than 10 years, if you think about it like that.
In the early '90s we did some six-minute walking distance studies; by that
time some other investigators had looked at BMI, and it began to brew in
our heads that maybe we could put it all together. Not until 2003 did we
have the data that confirmed it could be combined into one index. We then
applied the index to a larger validating cohort that we had followed over
those six years and confirmed the value of the index.
Why does this kind of research take a
Well, this was a prospective study looking at mortality as an outcome. Very
few studies had looked at mortality and what predicted it—very few.
How many subjects did you have in this
We had in the end something like 859 for that first paper. Now we're up to
And now that the BODE index is published and other
people have confirmed its value, how is it actually used in the
clinic? If it is, that is.
Let me tell you where its use is accepted. It is accepted as a way of
staging of patients for transplants. People with a very high BODE index,
higher than 7, are candidates for lung transplants. We hope we will extend
the use of BODE to everyday clinical practice.
The way you implement it is you see the patient; you measure the degree of
obstruction with aspirometry, the BMI, their endurance, by seeing how far
they can walk in six minutes, and degree of shortness of breath, using the
Modified Medical Research Council Scale. That gives you a number from zero
to ten, and with that number, you stratify risk. The higher the number, the
worse the patient will do.
So the international lung transplant community uses it to stratify people
at high risk as needing a transplant. The people with a high number are
those who are likely to die if they don't get a lung transplant. They're
most in need.
Were there potential predictors that didn't pan
out and didn't make it into the BODE?
Absolutely. This is the way it's done: you first derive the index on a
cohort. In that cohort, we explored over 24 factors that were thought to
predict outcome: albumin, hemoglobin, many different lung functions, etc.
These 24 potential factors were accumulated as possibilities between 1990
and 1994. We put together an initial cohort of 207 patients in this phase.
The instrumental person on that was Dr. Claudia Cote, the second author.
She collected those 207 patients at a VA Hospital where the informatics
were very good. Then we tested them prospectively on the rest of the cohort
in Venezuela, Spain, and the US.
"When I started my career in the mid-1970s, there was a
fatalistic attitude toward the disease..."
In the end, the ones that panned out were the four that made this very nice
acronym—BODE. We were lucky to have an acronym that is practical and
meaningful: it BODES well for you or BODES badly for you. Sometimes in this
business you need a little luck. You put in the effort, but sometimes it
helps to have the wind blow your way.
Why did you choose Venezuela and Spain for the research?
First of all because I'm Venezuelan and I trained two Venezuelan doctors
and then I trained several Spanish doctors who went back to Spain. We had
access to their clinics, so it was natural that we could cooperate with
How has your research evolved in the half a dozen years since that
BODE paper was published?
After the BODE paper, we embarked on a series of studies attempting to see
if BODE could also be a surrogate marker of ulterior disease outcome. Our
thought was if you have a disease that lasts 30 years, and you have to wait
10 years just to see the outcome of an intervention, could something else
take the place of that final outcome and be used as a surrogate marker?
We published a series of papers evaluating the effect of rehabilitation on
the BODE index at six months, the effect of lung volume reduction therapy
and pharmacotherapy. All of these studied showed that the BODE index could
be a surrogate marker of death down the road. It could be very useful.
In two of those—surgery and rehabilitation—BODE at six months
predicts outcomes two and five years later. We have preliminary data, which
we've so far published only as an abstract, and pharmacotherapy also
reflected in that.
What's next for your research?
Right now we're working on biomarkers of COPD, specifically the use of
blood or tissue biomarkers that might predict a patient's final outcome.
The idea is if we could just draw blood without having to make any other
measurements, that could tell us who's at risk of developing a bad problem,
and then we could begin treatment much earlier.
What would you rate as your most difficult or trying professional
My most difficult was probably in the early 1990s, just following a career
path in research with very little funding. And I was doing this when I was
not very junior, but middle attending level. That is hard, very hard. It's
like trying to write a novel only to have the editors keep rejecting it.
Clinical studies do not receive a lot of support from funding agencies. You
have to work your tail off to make a living and be able to continue your
research. That transition was very hard. But perseverance won in the end.
How would you define the ultimate goal of your research?
I would like to be able to say that over the course of my career I helped
patients with COPD.
How much has the treatment of COPD improved over the years that
you've been working on the disease?
Significantly. We've found that oxygen therapy prolongs life or prevents
mortality. We've found that pharmacotherapy can change the rate of decline
of lung function, that pulmonary rehabilitation improves dyspnea and
exercise capacity; that it prevents exacerbations and improves the quality
of life, and we've shown that lung volume reduction surgery for some
patients improves all of those as well as longevity.
We've come a long way. And what once was a negative attitude toward therapy
is now much more positive. People do feel this disease is treatable, and
that's reflected by companies which are now making products for COPD
What were the greatest challenges in performing and presenting your
The greatest challenge is that the field is under-funded. Let me rephrase
that to give it the right flavor: it's the least well-funded, biggest
killer in the United States by any institute that funds research. Getting
funding to conduct studies in COPD has been very hard and it continues to
be so. It's gotten a little better, but it still remains the least-funded
largest killer in the United States. It's the fourth largest killer not
just in the United States, but also in the rest of the world.
Why do you think that is?
It's still this feeling that COPD is self-inflicted. It's due to
cigarettes. It's also a disease of poorly educated, poor people. You don't
see a lot of executives smoking now. It's not like a heart attack that
anyone can get. It's not like stroke or cancer. Those things anyone can
get. Getting emphysema from smoking is a blue-collar disease, and I think
that's why it's so poorly funded.
If you performed your research again, or published your paper
again, what, if anything, would you do differently and why?
I would have liked to have had more numbers—2,000 patients instead of
859. That would have given us a little more precision. I don't think it
would have changed anything else. I think the index is pretty good and our
results have been corroborated.
What would you like to convey to the general public about your
I'd like to say that COPD is a preventable and treatable disease. That's
the message for the general public. For the researchers and the younger
people in medicine, I'd like to say that having faith in one's potential is
very useful. It's essential, actually.
Final question: Are you satisfied with your career, and what you've
accomplished in the years that you've been working on COPD?
Absolutely, I have had a ball. I've made a living and I hope I've been able
to help others. You can't beat that. In addition, together with my wife, we
have built a nice family. We put four kids through college and have five
grandchildren. I don't have a house on Cape Cod. I'm not a member of a
club. I don't have horses or a boat, but I wouldn't have had time to use
them if I had. I've had a very nice life. I like what I do.
Bartolome Celli, M.D.
Pulmonary and Critical Care Medicine Division
Brigham and Women's Hospital
Boston, MA, USA
Bartolome Celli's current most-cited paper in Essential Science
Indicators, with 763 cites:
Celli BR, et al., "Standards for the diagnosis and treatment of
patients with COPD: a summary of the ATS/ERS position paper," Eur.
Resp. J. 23 (6): 932-46, June 2004. Source:
Essential Science Indicators from
Read an interview with
Jørgen Vestbo—coauthor of the paper above.
KEYWORDS: COPD, GUIDELINES, DIAGNOSIS, TREATMENT, BODE INDEX, PROGNOSTIC
INDICATORS, LUNG FUNCTION, SURROGATE MARKERS, EXERCISE CAPACITY, BODY MASS
INDEX, DYSPNEA, DEGREE OF OBSTRUCTION, EXTRA-PULMONARY DOMAINS, PROSPECTIVE
STUDY, MORTALITY, LUNG TRANSPLANT PATIENTS, COHORT, BIOMARKERS, FUNDING,