I am pleased to have this opportunity to address some questions that have
arisen regarding lung volume reduction surgery (LVRS) and the Lung Volume
Reduction Clinical Trial supported by the National Heart, Lung, and Blood
Institute (NHLBI) and the Health Care Financing Administration (HCFA). Today, I
would like to tell you exactly why the NHLBI got involved with the LVRS issue;
why a randomized trial is the only valid way to determine the benefits and risks
of LVRS to patients; why such information is critical to patient decision-
making; and why the approach adopted by the NHLBI is the fastest, most
efficient, and most accurate way to provide that information.
Why is the NHLBI Involved?
In early 1995, members of the medical community discussed with the NHLBI
widespread concerns that hundreds of patients were receiving LVRS despite the
fact that its effectiveness had never been established. The only results we had
were preliminary, nothing was known about the long-term outcome, and no
scientific evidence was available to indicate which patients might benefit from
the operation.
In response to these concerns, the NHLBI convened a workshop in September
1995 to elicit the views of experts in pulmonary medicine, thoracic surgery,
physiology, outcomes assessment, quality-of-life evaluations, and statistics on
LVRS. They concluded that although LVRS was a promising procedure, it "must be
evaluated in a scientific, coordinated, and cooperative fashion." They
recommended that the NHLBI "develop a mechanism for funding data collection and
analysis from a multicenter study" and that "a randomized study with a
controlled nonsurgical arm ideally should be undertaken to evaluate the
procedure critically." Thus, by the time HCFA approached us, the NHLBI was
already firmly convinced that a critical evaluation of LVRS was scientifically
indicated and essential.
The preponderance of professional opinion supported the NHLBI in this view.
Moreover, the American Thoracic Society, the medical section of the American
Lung Association, adopted a position paper on LVRS in May 1996, stating "the
conduct of clinical trials, preferably in a controlled, randomized fashion, is
urgently needed."
Why a Randomized Trial?
A randomized clinical trial is the only way to answer the question of
whether LVRS helps or harms patients. In the absence of such a trial, we have no
idea whether this operation represents an improvement over current treatments
and whether it has unacceptable risks. To date, LVRS has been performed on a
very select group of patients (only 10 to 30 percent of patients referred for
the procedure are considered suitable for the surgery), and we do not know how
they would have fared without surgery. Simply comparing patients before and
after treatment ignores the possibility that factors other than the operation
may have come into play. Only a randomly selected control group allows
evaluation of the true effect of the treatment.
Our experience has demonstrated the importance of a control group time and
time again. It was impressively illustrated in the NHLBI Cardiac Arrhythmia
Suppression Trial (CAST). That study was designed to assess the effects of
antiarrhythmic drugs, since heart arrhythmias are associated with a several-fold
increase in death rate after heart attack. Although the drugs did, indeed,
suppress the arrhythmias, they were also associated with higher death rates
compared to placebo. Results were so dramatic that the trial was terminated
early. The surprising increase in mortality would not have been apparent if the
investigators had not included a control group that had received a placebo.
Experts on clinical trials overwhelmingly agree that control groups are the
best way to avoid incomplete or misleading conclusions. A control group allows
for reliable detection of subtle or subjective, but nevertheless important,
treatment effects such as quality of life. If the effect of the treatment is
large, then a control group allows us to detect it even faster and with more
certainty. Thus, constructing a rigorous trial design with a control group
ultimately serves patients better.
It has been argued that having a nonsurgical control group is unethical
because it denies some patients access to LVRS. However, one must bear in mind
that LVRS is a treatment of unproven efficacy. On the other hand, the medical
therapy that is included in the trial design is the best established therapy
currently available for severe emphysema, and all patients in the trial will
receive it.
Why Do Patients Need This Information?
History is full of promising treatments, medical and surgical-,-that when
evaluated under controlled conditions failed to live up to expectations of
benefit and, in some cases, caused harm. Can we ever forget thalidomide? And in
the area of surgery, consider the recent example of extracranial-intracranial
bypass. Based on comparison of clinical outcomes among nonrandomized cohorts
of patients, this operation was believed to prevent strokes; however, to the
surprise of many, a large multicenter randomized trial revealed that it did not
prevent stroke, but only exposed patients to the discomfort and risks of major
surgery.
We consider the LVRS trial to be not only appropriate, but necessary to give
both patients and their physicians the information they need to make informed,
intelligent decisions about treatment of end stage emphysema. How can physicians
competently and ethically advise their patients about LVRS when they have no
sound scientific basis for doing so?
Although determining efficacy and setting guidelines for which patients are
likely to ertain group of patients is highly likely to
benefit from the surgery. The DSMB has the power to recommend early termination
of the trial if the data show a clear benefit or risk from the intervention, as
was the case in CAST. We require this periodic review by an independent,
scientific peer group in all of our clinical trials to ensure their safety,
fairness, and scientific validity.
Because of the unusual demand to have this procedure evaluated quickly, I
have charged the investigators to have a protocol ready by June 1997, just one
year after the release of the request for proposals. This is a rapid timetable,
indeed, but I have every reason to believe it will be met.
Conclusion
End-stage emphysema is a devastating chronic illness, and its victims are
desperate. LVRS is a palliative treatment that has emerged without adequate
scientific validation of its efficacy. This trial represents our best
opportunity to provide it. If we fail to conduct the trial, we will be failing
in our mission to serve the American people. Billions of health care dollars
may be spent, tens of thousands of patients may be exposed to the risks of an
unproven treatment, and the value of LVRS will forever remain in doubt.
I will be pleased to answer any questions the Committee may have.
