From the Division of Vascular Surgery (D.L.D., M.H.M., D.E.S.),
Department of Surgery, University of Washington, Seattle and The University of
Massachusetts Medical Center (B.S.C.), Worcester, Mass. Dr Dawson is currently
at Wilford Hall Medical Center, Lackland AFB, Tex.
Methods and Results—Study inclusion criteria included age
Conclusions—Cilostazol improved walking distances, significantly
increasing ICD and ACD. The data suggest cilostazol is safe and well
tolerated for the treatment of intermittent claudication.
Many agents have been tried for the treatment of intermittent
claudication, although few drugs have demonstrated efficacy in
adequately designed, placebo-controlled trials. Classes of drugs that
have been advocated or tested for treating claudication include
rheological agents, vasodilators, antiplatelet agents,
anticoagulants, prostaglandins and
prostaglandin analogs, "metabolic
enhancers," and others.2 Generally, all these
classes of drugs have been thought to increase skeletal muscle oxygen
delivery or increase the efficiency of oxygen utilization.
Cilostazol (Otsuka Pharmaceutical Co Ltd, Tokushima, Japan),
6-[4-(1-cyclohexyl-1H-tetrazol-5-yl)butoxy]-3,4-dihydro-2(1H)-quinolinone,
is a new compound that may be useful for treating chronic
arterial disease and symptoms of intermittent claudication.
The principal mechanisms of its action include inhibition of
platelet aggregation and vasodilation. Early preclinical reports
and small trials examining Japanese patients with lower-extremity
arterial disease have suggested that cilostazol increases
dermal blood flow,3 augments vasodilation with
reactive hyperemia,4 5 and has utility in
the treatment of cutaneous ulcers.6 Unpublished
data from phase II studies in Germany suggested benefit in the
treatment of patients with claudication, but very small sample sizes
and poor follow-up severely limited those
trials.7 8 9
The present study was performed to test the efficacy and safety of
cilostazol compared with placebo for amelioration of the symptoms of
intermittent claudication. The primary outcome measures used to assess
efficacy were pain-free walking distance (distance walked to the onset
of symptoms, or the initial claudication distance [ICD]) and the
maximum distance walked (absolute claudication distance [ACD]) on
standardized treadmill testing. Secondary outcome measures examined
included measurements of ankle pressures (at rest and after exercise),
subjective assessments of symptoms by the patient and physician, and
safety.
Patient Population
Patients with limb-threatening (grades II and III) chronic limb
ischemia,11 manifested by
ischemic rest pain, ulceration, or gangrene, were excluded.
Other exclusion criteria were lower-extremity surgical or endovascular
arterial reconstructions or sympathectomy
in the preceding 6 months, uncontrolled hypertension, inability to
complete the treadmill walking test for reasons other than
claudication, recent myocardial infarction (within 6 months), recent
deep vein thrombosis (within 3 months), severe concomitant diseases,
substance abuse, and gross obesity.
Concomitant use of drugs known to have a significant effect on
peripheral arteries was limited to antihypertensive agents,
including ACE inhibitors, ß-blockers, or calcium channel
blockers, or the occasional use of nitroglycerin.
Dosages of all concomitant medications were kept constant throughout
the study when feasible.
Other classes of concomitant medications that were specifically
disallowed included antiplatelet agents (including aspirin),
anticoagulants, vasoactive agents (papaverine, isoxsuprine, nylidrin,
cyclandelate, and niacin derivatives), hemorheological agents
(pentoxifylline), and nonsteroidal anti-inflammatory drugs (with the
exceptions of acetaminophen and diclofenac sodium, which
have minimal antiplatelet effects).
No specific counseling about diet, smoking cessation, or exercise was
offered during the study period. These factors were not specifically
controlled. Investigators neither encouraged nor discouraged lifestyle
changes until patients completed the trial.
Study Methods
Evaluation of walking performance was accomplished with
standardized treadmill testing. A constant speed of 3.2 km/h (2 mile/h)
and a fixed incline of 12.5% were used. The treadmill tests were
considered valid only if claudication symptoms were the reason the
subject had to stop walking. Brachial, anterior tibial, and posterior
tibial artery systolic pressures were measured with
continuous-wave Doppler ultrasound and cuff occlusion, and these
pressures were used to calculate ankle/brachial indices (ABI) as
follows:
In addition to assessment of walking distances and ankle/brachial
indices, patients were serially evaluated for new complaints, changes
in symptoms, and functional status. ECGs were obtained, and laboratory
testing, including hematologic studies, serum chemistries, and
urinalyses, were routinely performed as part of the assessment of
safety.
Statistical Analysis
The relationship of percent change to the natural logarithm of the
distance/baseline ratio is given by:
For analysis of the intention-to-treat population, log-rank
scores were obtained separately for patients from each study center.
Statistical comparisons of treatments for the combined centers were
based on 3 methods: Mantel-Haenszel, Wei-Lachin, and Fisher
tests.
Characteristics of the subjects in these treatment groups are
summarized in Table 1
Efficacy
Table 3
Geometric means were compared to detect differences between treatment
groups with less impact of extreme values. When either
intention-to-treat analysis was used or only those who
completed 12 weeks of treatment were considered, the differences in
log(distance/baseline) were significant (Table 4
There were no significant testing center or treatment-by-center
interactions that affected the observed changes in ICD or ACD.
At each subject's final study visit, both the subject and investigator
were asked to subjectively assess the response to treatment, even if
the subject withdrew early. These assessments are categorized in Table 5
Adverse Events
Laboratory Findings
Cilostazol Pharmacology
In a small, double-blind, crossover study,17
cilostazol was a better inhibitor of
thromboxane-stimulated platelet aggregation than either
aspirin or ticlopidine. This antiplatelet effect may be relevant,
because previous clinical studies have suggested that ticlopidine
improves walking distances and ankle pressures in patients with
claudication.18 19 20 Cilostazol is 10 to 30 times
more potent than aspirin in inhibiting aggregation induced by ADP,
collagen, epinephrine, or arachidonic
acid.14 Unlike aspirin, cilostazol does not
inhibit prostaglandin I2
(prostacyclin) synthesis. This may be important, because
endothelium-derived prostacyclin potentiates the
effects of the inhibition of platelet aggregation by
cilostazol.21 Prostaglandin
I2 has antithrombotic activity, inhibits
platelet aggregation, and relaxes vascular smooth muscle.
In addition to its antiplatelet effects, cilostazol acts as an
arterial vasodilator, probably through its direct action on
vascular smooth muscle. Intracellular cAMP blocks release of calcium
ions from intracellular storage granules within the smooth muscle
cells, thus inhibiting the function of contractile
proteins.22 It is unclear what role
arterial or arteriolar vasodilation plays in any effect
cilostazol has on patients with chronic occlusive arterial
disease. Vasodilator drugs have typically been unsuccessful for the
treatment of claudication.23 This is thought to
be because resistance beds are already maximally vasodilated in
ischemic limbs.24 Effects of cilostazol
on regional blood flow and tissue oxygen delivery are unknown.
Cilostazol also affects smooth muscle cell proliferation. The drug
inhibits replication and growth of rat vascular smooth muscle cells in
tissue culture. This effect may also be mediated through increased
levels of intracellular cAMP in smooth muscle
cells.25
In the present study, cilostazol decreased
triglycerides, LDL, and total cholesterol and
increased HDL cholesterol, as previously
described.26 27 28 29 Although the magnitude of the
lipid-altering effect is small, it adds to the potential clinical
utility of the drug. However, larger trials that control for other
factors that may affect lipid metabolism are needed to
confirm and better characterize any possible effect on plasma
lipids.
Considerations in the Analysis of Efficacy
In the present study, outcome measures were examined both by
analyses by intention-to-treat and by considering only those
patients who completed 12 weeks of double-blind treatment.
Analysis by intention-to-treat is a more rigorous comparison of
alternate treatment regimens. Any subject completing
For analysis of the intention-to-treat population,
log-rank scores were obtained separately for the patients from each
study center. Statistical comparisons of treatments for the combined
centers were based on 3 methods: Mantel-Haenszel, Wei-Lachin, and
Fisher.
The Mantel-Haenszel test for ordered contingency tables was used in
categorical analyses. It is a nonparametric test
that examines relationships among data in row-by-column tables. The
Mantel-Haenszel test is directed at a weighted linear combination of
differences between treatments in mean log-rank scores for the
respective centers. It is particularly useful when the pattern of
treatment differences is consistent across study centers. When
the response to therapy was categorized, a statistically significant
improvement was demonstrated in the cilostazol group.
Wei-Lachin is a 2-sample, nonparametric,
multivariate analysis. It compares values at
each visit as well as the change in walking distances over the course
of the trial. Thus, it is a more stringent statistical challenge than
ANOVA. The most rigorous analysis in any study of a therapeutic
drug or procedure is by intention-to-treat (last observation carried
forward) analysis. Because of this, the most pertinent
analysis is at the last visit (week 12). Both ICD and ACD
showed statistically significant increases at week 12 by Wei-Lachin
analysis. The other comparisons at weeks 2, 4, and 8 do not
show statistically significant changes. Because the early assessments
do not show significance, the overall multivariate test
is not significant.
The Fisher procedure is directed at the across-center sum of
(-2
Limitations of Current Study
This study was relatively short in duration. The mean ICD and ACD for
the cilostazol-treated subjects increased at each testing visit during
the 12-week double-blind treatment comparison in this study. This
suggests that the therapeutic response may be gradual in onset and
progressive. It is unknown whether treatment with a longer duration of
cilostazol therapy would yield additional benefit.
The only direct comparison that this trial made was between
cilostazol-treated patients and those treated with placebo. The
relative benefit of cilostazol versus other claudication therapies was
not examined as part of this trial. There was an effort to minimize the
effects of other variables, however. No changes in activity or risk
factor modification were recommended during the trial. The lack of
change in the pain-free and maximal walking distances in the control
group suggests that there were no significant confounding effects. The
characteristics of the 2 treatment groups were similar, and the
stratified randomization scheme made the treatment groups comparable
with respect to use of calcium channel blockers.
Comparison of Results With Those Reported for
Pentoxifylline
Pentoxifylline efficacy has been tested in at least 10 randomized,
placebo-controlled, blinded clinical
trials,12 30 31 32 33 34 35 36 37 1 of which was performed in the
United States.12 All but 1 of these
studies33 demonstrated benefit with
pentoxifylline Meta-analysis is not possible because of
significant differences in reporting, methodology, and outcome measures
used in different reports.
The US study of Porter et al12 was one of the
largest of the trials completed. Although 128 patients were randomized,
46 were withdrawn from that study and were not considered in the
analysis of efficacy. An intention-to-treat analysis
was not performed. During 24 weeks of treatment, there was a
statistically significant difference between the pentoxifylline-treated
and control groups, although the differences in percent change from
baseline ICD and ACD when considered at week 24 were not statistically
significant. In part, this reflects the pronounced placebo effect: 36%
increase in ICD, 25% increase in ACD. Pentoxifylline-treated patients
in the study had a 59% increase in ICD and a 38% increase in ACD.
In the present study, there was a mean 39% change from baseline
for ICD and 45% for ACD in the patients who completed 12 weeks of
treatment. Because the parallel, placebo-treated control group did not
change from baseline (3% increase in ICD, 5% decrease in ACD), it
appears that the treatment effect (relative to placebo) for cilostazol
may be greater than that for pentoxifylline.
Pentoxifylline has failed to consistently demonstrate important
clinical benefit in controlled clinical
trials.2 24 33 38 Cameron et
al24 reviewed the results of placebo-controlled
trials with pentoxifylline and found a negative correlation
between sample size and response. This may be because of a publication
bias that favors pentoxifylline. Investigators may tend to withhold
publication of negative results from small trials, but positive results
are published. Findings from large studies tend to be published even
with negative or less-favorable results. Publication bias may explain
why the largest trials of pentoxifylline
efficacy12 39 demonstrated less of an increase in
walking distances than seen in smaller studies. The studies that
reported the most-favorable responses were small, often with 20 or
fewer treated patients completing therapy.30 35
Seven of the 10 previous randomized, placebo-controlled trials of
pentoxifylline had fewer than 20 patients in the treatment group.
The overall clinical utility of pentoxifylline is also limited by drug
intolerance, costs of therapy,40 and
inconsistent clinical response.33 40 41
It has been suggested that pentoxifylline is most useful for patients
with claudication that is moderately severe and that little
symptomatic benefit is offered for patients with mild or
severe disease.38 39 41 42 The cost-effectiveness
of cilostazol therapy and the best target populations to treat are as
yet unknown.
Problems Associated With Claudication Studies
Of practical concern, there can be a difference between finding that a
claudication treatment yields results that are statistically
significant and finding that the effect is clinically significant, ie,
"is the patient better?" Given adequate sample size, a
statistically significant improvement in walking distance may be
demonstrated. However, if the patient's subjective assessment of his
or her symptoms is not affected, there is no reason to consider the
effect to be beneficial. On the other hand, therapy does not have to
completely eliminate symptoms to be considered useful. Claudication
therapy is primarily directed at palliating symptoms and providing an
improvement in patients' overall functional status, and modest
differences in walking ability may have little impact.
In the present study, participants' subjective assessments
corroborated the findings with treadmill testing. This suggests that
measured improvement in walking distances translates into real
functional benefit for some patients. However, although no patients
found that cilostazol worsened their walking, half reported no change
in performance. This suggests that there may be a subgroup of
patients with claudication who derive a clinical benefit. Clinical
features that might predict who would be expected to respond to therapy
with cilostazol have yet to be identified.
Although a trend toward lowering triglycerides and
increasing HDL cholesterol levels was observed in the
cilostazol-treated group, it is premature to consider these to be
therapeutic effects of cilostazol. Confirmation with additional
clinical trials that control for potentially confounding effects on
lipid metabolism is necessary. However, it is intriguing
that an agent that effectively treats claudication symptoms may also
have an effect that may slow progression of the underlying
atherosclerotic arterial disease.
Conclusions
This work was supported with funding from Otsuka America Pharmaceuticals, Inc, a US subsidiary of Otsuka Pharmaceuticals, Ltd, the manufacturer of cilostazol.
The views expressed herein are those of the authors and do not reflect the official policy of the Department of Defense or other departments of the US government.
Received December 10, 1997;
revision received April 7, 1998;
accepted April 20, 1998.
© 1998 American Heart Association, Inc.
Clinical Investigation and Reports
Cilostazol Has Beneficial Effects in Treatment of Intermittent Claudication
Results From a Multicenter, Randomized, Prospective, Double-blind Trial
Abstract
Top
Abstract
Introduction
Methods
Results
Discussion
References
Background—Cilostazol is a
new phosphodiesterase inhibitor that suppresses
platelet aggregation and also acts as a direct arterial
vasodilator. This prospective, randomized, placebo-controlled,
parallel-group clinical trial evaluated the efficacy of cilostazol for
treatment of stable, moderately severe intermittent
claudication. 
40
years, initial claudication distance (ICD) on treadmill (12.5%
incline, 3.2 km/h) between 30 and 200 m, and confirmation of
diagnosis of chronic lower-extremity arterial occlusive
disease. After stabilization and single-blind placebo lead-in, 81
subjects (62 male, 19 female) from 3 centers were randomized unequally
(2:1) to 12 weeks of treatment with cilostazol 100 mg PO BID or
placebo. Primary outcome measures included ICD and maximum distance
walked (absolute claudication distance, or ACD). Secondary outcome
measures included ankle pressures, subjective assessments of benefit by
patients and physicians, and safety. Treatment and control groups were
similar with respect to age, severity of symptoms, ankle pressures, and
smoking status. Statistical analyses used intention-to-treat
analyses for each of 77 subjects who had 1 treadmill test
after initiation of therapy. Comparisons between groups were based on
logarithms of ratios of ICD and ACD changes from baseline using ANOVA
test at last treatment visit. The estimated treatment effect showed a
35% increase in ICD (P<0.01) and a 41% increase in
ACD (P<0.01). There was no significant change in
resting or postexercise ankle/brachial indexes. Patients' and
physicians' subjective assessments corroborated the measured
improvements in walking performance observed in the
cilostazol-treated group.
Key Words: claudication • peripheral vascular disease • cilostazol • drugs
Introduction
Top
Abstract
Introduction
Methods
Results
Discussion
References
Although progress has
been seen in surgical and interventional techniques for the management
of arterial occlusive disease in recent years, these
procedures are only appropriate for patients with intermittent
claudication who have significant, lifestyle-limiting symptoms.
Initially, nonoperative treatment is appropriate for patients with
claudication, and for many this may suffice. Nonpharmacological
interventions, such as smoking cessation, exercise, and weight loss,
can improve walking performance1 and are
fundamental in the management of patients with claudication.
Pharmacological treatment also has a role. First, medical management of
diabetes, hypertension, and hyperlipidemia may slow
atherosclerosis progression. However, these
interventions have not been shown to affect symptoms or cause
regression of established peripheral arterial
disease. Second, drug therapy may be used to provide
symptomatic relief by improving pain-free and overall
walking distances.
Methods
Top
Abstract
Introduction
Methods
Results
Discussion
References
This study was designed according to the recommendations of the
Society for Vascular Surgery Ad Hoc Committee on Clinical Research,
which detail appropriate requirements for trials of medications for the
treatment of intermittent claudication.10
Participating study centers included the University of Washington
Medical Center, the University of Massachusetts Medical Center, and the
Oregon Health Sciences Medical Center. The study was conducted with the
approval of the Human Subjects Review committees of the respective
institutions, and written informed consent was obtained from each
participant before any study procedures were begun.
The study population included 81 patients with stable symptoms
of intermittent claudication secondary to chronic occlusive
arterial disease from atherosclerosis
(symptoms present for 6 months and not significantly changed
within the past 3 months). Clinical diagnoses of chronic occlusive
arterial disease were supported with objective criteria
from noninvasive vascular tests, including an ICD on the treadmill
between 30 and 200 m and a minimum postexercise drop in
Doppler-measured ankle systolic blood pressure of 20
mm Hg.
This study was a randomized, parallel, double-blind trial with
the administration of either cilostazol 100 mg BID or placebo BID.
After a 2-week baseline period for stabilization of concomitant
medications and entry treadmill testing, there was a 2- to 4-week
single-blind placebo lead-in phase, during which time each patient's
treadmill walking time (ICD) had to be within ±35% of the value at
the previous visit. Patients with stable treadmill walking
performance were then randomized 2:1 for treatment with
cilostazol or placebo, respectively, for a period of 12 weeks. The
active treatment group received cilostazol 100 mg PO BID. The control
group received an identical placebo BID. Randomization was stratified
by treatment center and patient's use of calcium channel blocker.
Subjects were then reevaluated 2, 4, 8, and 12 weeks after initiation
of therapy.
All statistical tests were 2-sided, and differences were
considered significant if the P value was <0.05. Using
methods described by Porter et al12 and Gillings
et al,13 we also analyzed data for ICD
and ACD in terms of logarithms of distance/baseline ratios. This
transformation reduces the impact of extreme values. It also allows the
results from each treatment group to be expressed in terms of percent
change from baseline.
Results
Top
Abstract
Introduction
Methods
Results
Discussion
References
There were a total of 81 patients from the 3 study centers entered
into the double-blind phase of the trial. All subjects had grade I
chronic limb ischemia (according to reporting standards of the
Society for Vascular Surgery/North American Chapter of the
International Society for Cardiovascular
Surgery11). Claudication was characterized as
"mild" (category 1) in 7 patients (8.6%), "moderate" (category
2) in 22 (27.3%), and "severe" (category 3) in 52 (64.2%). Before
randomization, eligibility for inclusion was confirmed, and concomitant
medical therapy was stabilized. All patients completed 2 to 4 weeks of
single-blind placebo administration before randomization. 
. The groups were
similar with respect to demographic features and major
cardiovascular risk factors. Eighty patients were
white; there was 1 black patient. Of the 81 subjects randomized, 4 did
not participate in the trial long enough to have 1 treadmill test
after initiation of treatment. The remaining 77 were considered
evaluable (by intention-to-treat analysis methods) and compose
the principal subject of this report. A total of 27 patients were
assigned to placebo, and 54 were assigned to treatment with cilostazol
100 mg PO BID. Sixty-six patients completed the trial, and
analyses of this group were separately performed to better
determine the actual treatment effect on walking performance
when continued therapy was administered. The reasons for subjects'
withdrawal after randomization are summarized in Table 2.
View this table:
[in a new window]
Table 1. Demographic and Study Population Characteristics at
Baseline
View this table:
[in a new window]
Table 2. Reasons for Withdrawal or Exclusion After
Randomization
The primary end point for the study was walking distance. Table 3 shows raw walking distances (ICD and
ACD) for the treatment and control groups at each visit after the
initiation of double-blind therapy. This table shows arithmetic mean
data for all subjects who completed 1 treadmill test after
randomization (n=77).
View this table:
[in a new window]
Table 3. Analysis of Raw Walking Distance Measurements, by
Intention to Treat (Last Valid Observation Carried
Forward) 
shows there was a significant improvement in subjects'
treadmill walking distances after 12 weeks of treatment, when
mean ICD and ACD were compared with baseline performance, in
the group treated with cilostazol. The mean ICD (arithmetic mean)
improved by 58% in the subjects treated with cilostazol, compared with
an increase of only 8.9% in the placebo group. These differences were
highly significant. Cilostazol-treated subjects increased mean ACD by
63%, whereas there was a 9.8% decrease in mean ACD in the placebo
group. These differences were also highly significant. No statistically
significant differences were demonstrated between the treatment and
control groups before week 12. ). As with the analysis of the
raw walking distances, the differences demonstrated between the
treatment and control groups were statistically significant after 12
weeks of treatment but not earlier. When patients were categorized by
the magnitude of their response to treatment, there was a significantly
more favorable response in the cilostazol-treated group (Figures 1 and 2).
View this table:
[in a new window]
Table 4. Percent Change in Walking Distances From
Baseline (Geometric Means)
View larger version (22K):
[in a new window]
Figure 1. Categorization of increase in raw ICD after 12
weeks of treatment.
View larger version (24K):
[in a new window]
Figure 2. Categorization of increase in raw ACD after 12
weeks of treatment. . Half of the patients treated with
cilostazol judged their walking performance to be "better"
or "much better"; none reported their symptoms to be worse.
Sixty-three percent of placebo-treated patients reported their symptoms
to be "unchanged," with similar proportions (19%) reporting either
improvement or worsening. Of note, patients' positive subjective
response rates were somewhat better when only the 44 patients who
completed 12 weeks of therapy with cilostazol were considered; 11 such
patients (25%) reported their symptoms to be "much better" and 14
(32%) said they were "better." Physicians' assessments were
similar. Considering all randomized patients, 48% of the
cilostazol-treated group were said to be "better" or "much
better" versus 22% of the placebo-treated patients. Both the
patients' and physicians' subjective assessments demonstrated a
statistically significant improvement in claudication symptoms in the
patients randomized to cilostazol.
View this table:
[in a new window]
Table 5. Subjective Assessment of Effect of Study Drug on
Ability to Walk
Reported serious adverse events include 1 death from myocardial
infarction in the placebo group and 6 hospitalizations of
cilostazol-treated patients (subclavian artery stenosis,
unstable angina, pneumonia [2 patients], myocardial infarction, and
transient ischemic attack). (The US Food and Drug
Administration defines a serious adverse event as an occurrence that is
fatal, life-threatening, disabling, or requires hospitalization; or a
drug overdose, congenital anomaly, or cancer.) Gastrointestinal
complaints were noted in 44% of the cilostazol-treated patients. The
most commonly reported side effects included diarrhea, loose stools,
flatulence, and nausea. Importantly, these symptoms were typically mild
and often self-limited, although they persisted in some patients.
Gastrointestinal complaints were recorded from 15% of the placebo
group. Headaches were the next most common type of complaint, occurring
in 20% of cilostazol-treated patients and 15% of placebo-treated
patients. In general, the drug was well tolerated, and most of the
reported side effects were managed symptomatically.
Neither treatment group demonstrated clinically significant shifts
in serum electrolytes, renal or liver function, coagulation profiles,
or peripheral blood cell counts. There was no evidence of
clinically significant changes in blood pressure, heart rate, or ECG
alterations associated with cilostazol treatment. Review of the
laboratory data seemed to indicate that the cilostazol-treated group
had potentially beneficial alterations in plasma lipids (Table 6). However, this trial did not
prospectively control for hyperlipidemia, concomitant
use of lipid-lowering agents, poor diabetes control, diet, or other
factors that might affect serum lipids.
View this table:
[in a new window]
Table 6. Summary of Plasma Lipid
Values
Discussion
Top
Abstract
Introduction
Methods
Results
Discussion
References
In this study, 12 weeks of cilostazol treatment resulted in a
significant improvement in treadmill walking performance. Both
ICD and ACD were significantly increased compared with baseline
performance, and in this carefully designed study, there was
essentially no change in walking performance observed in the
placebo-treated group. Patient and physician assessments of treatment
response suggested that the statistically significant measured
improvement was clinically relevant as well. Although gastrointestinal
side effects were common, these were typically mild and self-limited
and did not interfere with continued therapy.
Cilostazol has a number of actions that may be beneficial for the
treatment of patients with arterial disease, including
inhibition of platelet aggregation. Cilostazol inhibits type III
phosphodiesterase activity in platelets, thereby increasing
intracellular levels of cAMP by blocking its
hydrolysis.14 15 16 Increased intraplatelet
cAMP concentration inhibits thromboxane
A2 production and platelet
aggregation by inhibiting phospholipase and
cyclooxygenase.
A typical feature of many follow-up studies is the occurrence of
incomplete data for some of the subjects, either because some subjects
withdraw before completion of the study or because they occasionally
miss visits. An important consequence of such incomplete data is the
introduction of variation in the patient population across visits. This
can complicate interpretation across visits. One way to address this
problem is to maintain the intention-to-treat population by assigning
extended end-point values to missed visits, as was done in the current
study. With this method, if a value for a visit is missing, it is
replaced by the value from the nearest preceding visit. As a result,
the value remains missing only if all preceding visits are missing or
it corresponds to the first visit. 1 valid
treadmill test after randomization was included in the
intention-to-treat analysis. This was done by carrying the last
valid observation forward and using that value for the analyses
at later, missed visits. The disadvantage of this type of
analysis is that it tends to mask treatment effects that
increase over time. 
i log Pi), where log
denotes natural logarithm and P denotes the 1-sided P value
for percent change with cilostazol compared with placebo for the
i-th investigator. This method is sensitive to whether there
are 1 or more P values that are sufficiently small to
support the interpretation of the combined set as significant.
This trial was designed to be a pilot study; only a limited number
of patients participated. There were 81 subjects enrolled; of these, 77
were evaluable (completing 1 treadmill test after randomization), and
66 completed the trial. There are some problems and limitations that
are inescapable in a study with a relatively small sample population.
There may be type II statistical errors. Given the inherent variability
in the measurement of walking distances, we may have seen a significant
treatment effect earlier if there had been more patients. Also, there
is limited ability to perform analyses of selected subgroups
because sample sizes would be decreased further. However, the fact that
this trial demonstrated a positive effect despite the limited sample
size suggests that improvement in claudication symptoms is clinically
significant. Therapeutic responses that are statistically significant
only when large numbers of patients are studied tend to be less
important in clinical practice. Also, smaller studies are more subject
to patient-selection biases.
Although truly valid comparison awaits completion of trials that
prospectively examine the efficacy of cilostazol compared with other
therapies, it may be useful to compare the results of the current study
with the currently available medical therapy for intermittent
claudication. Pentoxifylline has been the most extensively evaluated
and most widely used drug for claudication, and it is the only agent in
the United States with Food and Drug Administration–approved labeling
for this indication.
Certainly patients in these trials undergo intensive surveillance,
and there may be extensive physician attention and counseling regarding
such factors as self-care, smoking cessation, and control of
diabetes. Because of this, trials must be designed as
placebo-controlled trials if useful conclusions are to be drawn. These
factors can have a large effect on treadmill walking
performance; as much as 149% improvement over baseline was
seen in the placebo group in 1 pentoxifylline
trial.31 Furthermore, treadmill testing has some
inherent variability with repeated
examinations.43
Cilostazol is an agent with pharmacological actions unlike other
drugs used for the treatment of chronic occlusive arterial
disease. This study represents a preliminary examination of its
efficacy for treatment of the symptoms of intermittent claudication.
Although this limited trial showed a statistically significant
improvement in treadmill walking performance after 12 weeks of
therapy with cilostazol 100 mg PO BID, and patients' subjective
assessment of symptoms corroborated this finding, additional studies
with larger numbers of participating claudication patients are needed
to confirm efficacy. The positive results of the current study are
encouraging, and larger trials have been initiated to answer some of
these questions.
Footnotes
Reprint requests to David L. Dawson, Lieutenant Colonel, USAF MC, Wilford Hall Medical Center, 959th MDOS/MMKG, 2200 Bergquist Dr, Suite 1, Lackland AFB, TX 78236-5300.
References
Top
Abstract
Introduction
Methods
Results
Discussion
References
This article has been cited by other articles:
 |
|
 |
|
  M. Sobel and R. Verhaeghe Antithrombotic Therapy for Peripheral Artery Occlusive Disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition) Chest, June 1, 2008; 133(6_suppl): 815S - 843S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. V. Milani and C. J. Lavie The role of exercise training in peripheral arterial disease Vascular Medicine, November 1, 2007; 12(4): 351 - 358. [Abstract] [PDF]
|
|
|
|
|
|
E. P Brass and W. R Hiatt Review of mortality and cardiovascular event rates in patients enrolled in clinical trials for claudication therapies Vascular Medicine, August 1, 2006; 11(3): 141 - 145. [Abstract] [PDF]
|
|
|
|
 |
|
 D. H. Kim Exercise and Peripheral Arterial Disease Ann Intern Med, May 2, 2006; 144(9): 699 - 699. [Full Text] [PDF]
|
|
|
|
 |
|
 A. T. Hirsch, Z. J. Haskal, N. R. Hertzer, C. W. Bakal, M. A. Creager, J. L. Halperin, L. F. Hiratzka, W. R.C. Murphy, J. W. Olin, J. B. Puschett, et al. ACC/AHA 2005 Guidelines for the Management of Patients With Peripheral Arterial Disease (Lower Extremity, Renal, Mesenteric, and Abdominal Aortic): Executive Summary A Collaborative Report From the American Association for Vascular Surgery/Society for Vascular Surgery, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, Society of Interventional Radiology, and the ACC/AHA Task Force on Practice Guidelines (Writing Committee to Develop Guidelines for the Management of Patients With Peripheral Arterial Disease): Endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation; National Heart, Lung, and Blood Institute; Society for Vascular Nursing; TransAtlantic Inter-Society Consensus; and Vascular Disease Foundation J. Am. Coll. Cardiol., March 21, 2006; 47(6): 1239 - 1312. [Full Text] [PDF]
|
|
|
|
|
|
S. P. Marso and W. R. Hiatt Peripheral Arterial Disease in Patients With Diabetes J. Am. Coll. Cardiol., March 7, 2006; 47(5): 921 - 929. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Geraghty Covered Stenting of the Superficial Femoral Artery Using the Viabahn Stent-Graft Perspectives in Vascular Surgery and Endovascular Therapy, March 1, 2006; 18(1): 39 - 43. [Abstract] [PDF]
|
|
|
|
 |
|
 F. Honda, H. Imai, M. Ishikawa, C. Kubota, T. Shimizu, M. Fukunaga, and N. Saito Cilostazol Attenuates Gray and White Matter Damage in a Rodent Model of Focal Cerebral Ischemia Stroke, January 1, 2006; 37(1): 223 - 228. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. D. Cunningham and M. S. Conte Lower-Extremity Arterial Disease Perspectives in Vascular Surgery and Endovascular Therapy, December 1, 2005; 17(4): 351 - 361. [Abstract] [PDF]
|
|
|
|
 |
|
 S. De Vivo, U. Palmer-Kazen, B. Kalin, and E. Wahlberg Risk Factors for Poor Collateral Development in Claudication Vascular and Endovascular Surgery, November 1, 2005; 39(6): 519 - 524. [Abstract] [PDF]
|
|
|
|
 |
|
 M.-J. Kim, K.-G. Park, K.-M. Lee, H.-S. Kim, S.-Y. Kim, C.-S. Kim, S.-L. Lee, Y.-C. Chang, J.-Y. Park, K.-U. Lee, et al. Cilostazol Inhibits Vascular Smooth Muscle Cell Growth by Downregulation of the Transcription Factor E2F Hypertension, April 1, 2005; 45(4): 552 - 556. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Y. Park, J. H. Lee, Y. K. Kim, C. D. Kim, B. Y. Rhim, W. S. Lee, and K. W. Hong Cilostazol Prevents Remnant Lipoprotein Particle-Induced Monocyte Adhesion to Endothelial Cells by Suppression of Adhesion Molecules and Monocyte Chemoattractant Protein-1 Expression via Lectin-Like Receptor for Oxidized Low-Density Lipoprotein Receptor Activation J. Pharmacol. Exp. Ther., March 1, 2005; 312(3): 1241 - 1248. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Casey, B. H. Tonnessen, W. C. Sternbergh III, and S. R. Money Medical Management of Intermittent Claudication Vascular and Endovascular Surgery, September 1, 2004; 38(5): 391 - 399. [Abstract] [PDF]
|
|
|
|
|
|
G. P. Clagett, M. Sobel, M. R. Jackson, G. Y. H. Lip, M. Tangelder, and R. Verhaeghe Antithrombotic Therapy in Peripheral Arterial Occlusive Disease: The Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy Chest, September 1, 2004; 126(3_suppl): 609S - 626S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. Sim, G. Merrill-Skoloff, B. C. Furie, B. Furie, and R. Flaumenhaft Initial accumulation of platelets during arterial thrombus formation in vivo is inhibited by elevation of basal cAMP levels Blood, March 15, 2004; 103(6): 2127 - 2134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. S. Aronow Management of Peripheral Arterial Disease of the Lower Extremities in Elderly Patients J. Gerontol. A Biol. Sci. Med. Sci., February 1, 2004; 59(2): M172 - 177. [Abstract] [Full Text] [PDF]
|
|