Abbreviations and Acronyms

ABI

ankle-brachial index

AG

atherosclerosis group

AOD

arterial occlusive disease

BDG

Buerger disease group

CA-PAD

chronic atherosclerotic peripheral arterial disease

CAD

coronary artery disease

DAG

diabetic angiopathy group

ETG

embolism/thrombosis group

IC

intermittent claudication

Sto₂

tissue oxygen saturation

TASC II

TransAtlantic Inter-Society Consensus for the Management of Peripheral Arterial Disease

Introduction

Chronic atherosclerotic peripheral arterial disease (CA-PAD) of the lower extremities is seen in 4% to 12% of the population aged 55 to 70 years. It increases up to 20% in the population older than 70 years.^1,2 Cilostazol was approved by US Food and Drug Administration in 1999, the European Medicines Agency in 2002, and in Spain in 2009 to improve walking distances among patients with intermittent claudication (IC) in CA-PAD, and today, it is widely used worldwide.^3,4 Based on the positive clinical results from its use in CA-PAD, cilostazol is now also used in patients with Buerger disease, diabetic angiopathy, or arterial embolism/thrombosis that have nonatherosclerotic etiologies. Studies on the use of cilostazol in patients with Buerger disease or diabetic CA-PAD and in those on hemodialysis and the combined use of cilostazol with aspirin and/or clopidogrel are present in the literature,^5–8 but to the authors' knowledge, no comparison of cilostazol's effectiveness in atherosclerotic and nonatherosclerotic arterial occlusive diseases (AODs) has yet been performed.^9,10

The aim of this study was to analyze the benefits of cilostazol in patients with nonatherosclerotic AOD and compare the results with use in patients with CA-PAD, the disease for which cilostazol was approved for use.

Patients and Methods

A single-center, prospective cohort study of patients with Rutherford grade 0 and 1 (category 0, 1, 2, 3) AOD was undertaken. Following exclusion of patients with Rutherford grades 2 and 3 (category 4, 5, 6) disease and chronic, limb-threatening ischemia, the remaining patients were started on cilostazol 100 mg twice daily in 2019 and 2020, then allocated to 1 of 4 groups according based on etiology. The patients were followed for 1 year at 3-month intervals. Patients without diabetes but with isolated CA-PAD were included in the atherosclerosis group (AG), which had a rate of target extremity per patient of 99/60. Beyond aggravation of atherosclerosis, chronic hyperglycemia causes low-grade inflammation, vasoconstriction, thrombotic abnormalities, arterial stiffness, endothelial dysfunction, and microvascular dysfunction, leading to distal artery disease. Therefore, patients with distal diabetic angiopathy, patients with CA-PAD who use insulin or oral antidiabetics, and those with high glucose and hemoglobin A_1c—even if they did not use medication—were allocated to the diabetic angiopathy group (DAG), which had a rate of target extremity per patient of 87/53. Patients who were started on cilostazol because of arterial embolism or thrombosis were allocated to the embolism/thrombosis group (ETG), which had a rate of target extremity per patient of 76/53. Patients with Buerger disease were allocated to the Buerger disease group (BDG), which had a rate of target extremity per patient of 45/28.

Patients' Rutherford clinical classification, TransAtlantic Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASC II) classification of the accompanying atherosclerotic lesions, ankle-brachial index (ABI) score (ERKA manual sphygmomanometer; Kallmeyer Medizintechnik GmbH & Co KG; and Huntleigh DMX Digital Doppler; Huntleigh Healthcare Ltd), distal tissue oxygen saturation (Sto₂) (INVOS 5100C Cerebral/Somatic Oximeter; Covidien), and maximum walking distance (Cardioline xr450m adult treadmill ergometer [Trento], at fixed incline of 12.5% and speed of 3.2 kph) measurements were performed by at least 2 authors in the cardiovascular surgery clinic. Extremities whose ABI scores were under 0.90 were chosen as target extremities. In addition, patients were informed and received recommendations about walking exercise, dietary changes, and smoking cessation.

Afterwards, patients were called to the control outpatient clinics every 3 months unless they had new or increasing symptoms. At the outpatient control clinics, 1 author collected data on the recommended walking exercise, dietary program, smoking cessation, and side effects related to cilostazol. Then, all authors evaluated the collected data jointly. Cilostazol dose reduction was made by joint decision, when necessary. At the 12-month control visit, at least 2 authors performed the final measurements and data collection at the cardiovascular surgery clinic. Interobserver reliability was ensured through communication and information sharing among the observers (the article's authors) at every stage of measurements and data collection and also at monthly meetings. Any disagreements were solved by consulting Bugra Harmandar, MD, who is also the head of the clinic, as an impartial observer.

The obtained data from the study were analyzed using SSPS software, version 20.0 (IBM Corporation). The Kolmogorov-Smirnov test was performed to test the normality of quantitative variables. Because no assumption of normal distribution was provided, the Wilcoxon test was used to compare paired samples, and the Kruskal-Wallis H test was used to compare independent groups. Continuous variables that follow normal distribution are shown as mean (SD), and continuous variables that do not follow normal distribution are shown as median (25th–75th percentile). To analyze the categorical data, the authors used the Pearson χ² test. The McNemar test was used to compare the number of vascular surgeries and wound counts before and after cilostazol use. Descriptive statistics for categorical variables are presented as number (%). P < .05 was considered statistically significant.

Results

In total, 307 target extremities in 194 patients were evaluated. Of the patients, 34 (18%) were female and 160 (82%) were male. The mean (SD) age was 65 (14) years, and the mean (SD) weight was 80 (15) kg. All patients had distal arterial lesions under the level of the knee. Patients' characteristics, Rutherford clinical classification, accompanying diseases, and smoking status are provided in Table I. In patient imaging studies, it was observed that even though the main occlusive disease etiology was different from atherosclerosis, a specific portion of the patients had accompanying multisegmental atherosclerotic lesions. In the BDG and DAG, these additional atherosclerotic lesions may have increased ischemia. In the ETG, the thrombus may have developed on a chronic atherosclerotic background and determined the occlusion level; it may also have been the reason for incomplete or unsuccessful embolectomy procedures. Therefore, the authors decided to include patients' additional atherosclerotic lesions and TASC II classifications to better evaluate the effects of the lesions cumulatively in all groups. Imaging methods used to diagnose AOD and TASC II classifications of the patients' arterial lesions are provided in Table II.

TABLE I Patient Characteristics, Rutherford Clinical Classification, Accompanying Disease, and Smoking Status

View Fullscreen

TABLE II Imaging Methods Used to Diagnose the AOD and TASC II Classification of the Patients' Arterial Lesion

View Fullscreen

The mean (SD) admission time of patients in the ETG to our hospital was 25.60 (15.40) hours. Embolism/thrombosis regions were the common femoral artery in 6 extremities, the superficial femoral artery in 19 extremities, the popliteal artery in 24 extremities, and the crural arteries in 27 extremities. Because of their previous cardiovascular disease, severe peripheral arterial lesions, applied vascular surgeries, and atrial fibrillation, cilostazol was combined with aspirin in 100 patients, with clopidogrel in 19 patients, with new oral anticoagulants in 7 patients, and with warfarin in 5 patients. Despite the authors' insistence, 21 patients in the AG, 9 patients in the DAG, 16 patients in the ETG, and 8 patients in the BDG continued smoking. According to the patients' feedback and controls, clinical improvements began to be seen at a mean (SD) of 4.62 (2.93) weeks and reached a maximum at a mean (SD) of 10.71 (5.78) weeks after cilostazol was started. Headache (n = 41 [21.1%), diarrhea (n = 25 [12.9%]), and nausea (n = 21 [10.8%]) were the most commonly seen early side effects, which continued until maximum improvements were reached. Edema (n = 40 [20.6%]), ecchymosis (n = 18 [9.3%]), and petechiae (n = 11 [5.7%]) were the most commonly seen late side effects after maximum improvements had been reached. Cilostazol doses were reduced to 50 mg twice daily in 5 patients because of diarrhea, in 3 patients because of nausea, and in 6 patients because of headache in the first month; and in 2 patients because of edema in the fourth month and in 1 patient because of ecchymosis in the sixth month. Measurement comparisons were performed at the end of mean (SD) 13.24 (1.84) months because of the COVID-19 pandemic. Maximum walking distance, ABI, and Sto₂ values increased significantly in all groups, but there was no significant difference among the groups regarding maximum walking distance increase (P = .441). Ankle-brachial index scores after cilostazol use in the ETG were significantly higher than in the DAG (P = .037). After cilostazol use, Sto₂ in the DAG and BDG were significantly lower than in the AG (P < .001). Differences in ABI in the BDG was significantly lower than in the DAG and ETG (P < .001). The Sto₂ difference in the BDG was significantly lower than in the AG, DAG, and ETG (P < .001) (Table III). Although vascular surgery counts decreased in all groups after cilostazol use, the decrease in the AG and ETG were statistically significant (P = .019 and P = .004, respectively). Although the wound counts decreased after cilostazol use in all groups, however, no statistically significant result was obtained (Table IV).

TABLE III Descriptive Statistics and Comparison Results of MWD, ABI, Distal Sto₂, ABI Difference, and Distal Sto₂ Difference in the AG, DAG, ETG, and BDG Before and After Cilostazol Use ^a

View Fullscreen

TABLE IV The Comparison of the AG, DAG, ETG, and BDG According to Vascular Surgery and Presence of Wounds, Before and After Cilostazol Use ^a

View Fullscreen

Discussion

As a result of this study, significant increases in maximum walking distance, ABI, and Sto₂ values were seen in all 4 groups (Table III). In the meta-analysis by Bedenis et al,² which included 15 clinical trials of 3,718 participants, the authors stated that “cilostazol improves both initial and absolute IC distances.” Similarly, Kaya et al¹¹ found a 28% to 63% improvement in IC distances; in 4 different studies, only cilostazol 100 mg twice daily was used. In addition, they reported that they found a significant increase in ABI, which they considered a moderate improvement in terms of clinical significance. In the study by O'Donnell et al,¹² however, the authors could not find any difference in ABI levels at 6 or 24 weeks of cilostazol treatment. Boezeman et al¹³ evaluated foot oxygenation with near-infrared spectroscopy after endovascular revascularization; the authors found a significant increase in ABI and a 20% increase in Sto₂ 4 weeks after endovascular revascularization. Komiyama et al¹⁴ evaluated tissue oxygenation with spatially resolved spectroscopy. They determined that the decrease in muscle Sto₂ at the end of exercise correlated with the severity of IC. Although no significant difference was found among the AG, DAG, and BDG in this study, the authors considered the significantly higher ABI and ABI difference (P = .037 and P < .001, respectively) found in the ETG to be the result of the embolectomy procedures performed in patients.

In a study of patients with diabetic angiopathy, Gardner et al¹⁵ found a 46.07% decrease in calf muscle Sto₂ 1 minute after exercise. The median calf muscle Sto₂, which was 51% during rest, decreased by as much as 9% during exercise in these patients. They also concluded that elevated glucose level plays a role in impaired peripheral circulation and vascular dysfunction in patients with symptomatic CA-PAD, which leads to decreased muscle Sto₂.¹⁵ Yong et al⁵ evaluated the effects of aspirin combined with cilostazol in patients with diabetes and Buerger disease. They concluded that higher levels of serum inflammatory factors caused by diabetes and Buerger disease trigger the occurrence of thrombus and lead to endothelial dysfunction.⁵ Moreover, Klein-Weigel and Richter¹⁶ reported that Buerger disease relates to tobacco use and pro- and anti-inflammatory cytokines. The significantly lower Sto₂ levels in the DAG and BDG (P < .001) obtained in the current study were also thought to be associated with high glucose levels, tobacco use, and increased inflammation. With the significantly lower ABI and Sto₂ differences (P < .001 and P < .001, respectively) in the BDG combined with the earlier results, the authors concluded that nonatherosclerotic AOD etiologies also benefit from cilostazol, but patients in the DAG and BDG seemed to benefit less.

de Donato et al¹⁷ reported that cilostazol was the only medical treatment that prevents restenosis in lower-limb revascularization and significantly reduces repeat revascularizations. In this study, vascular surgery counts also decreased significantly in the AG and ETG (P = .019 and P = .004, respectively), but repeat operation counts were high in the ETG because of embolectomy procedures performed on patients with significant atherosclerotic stenosis. Femoropopliteal bypass was the most commonly performed procedure as a repeat operation. Mii et al¹⁸ achieved a 92% wound healing rate in patients administered cilostazol after infrainguinal bypass at 1 year. In addition, the median wound healing time was shorter than in those who did not receive cilostazol (45 vs 78 days).¹⁸ Colak et al¹⁹ found faster wound healing, higher rates of response to treatment, and better symptom improvement in patients with CA-PAD and diabetic foot ulcer who received cilostazol than those who received aspirin. In this study, although wound counts decreased in all groups after cilostazol use, when the healed and new wounds were combined, no significant difference was observed among the groups in wound counts after cilostazol use.

According to the CASTLE study by Hiatt et al,²⁰ headache (10.5%), diarrhea (10.9%), and palpitations (5.3%) were the most common side effects of cilostazol. Edema, headache, and diarrhea most commonly led to drug discontinuation.²⁰ In the study by Kaya et al,¹¹ 11.2% of patients experienced cilostazol side effects, but they did not cause any patient to discontinue the treatment, although cilostazol doses were reduced in some patients. In this study, headache, diarrhea, and nausea were the most common early side effects, whereas edema, ecchymosis, and petechiae were the most common late side effects. All side effects were treated according to symptoms without discontinuation of cilostazol.

Conclusion

This study showed that cilostazol is an effective and well-tolerated agent in nonatherosclerotic etiologies of AOD as well as for CA-PAD. Patients with Buerger disease or diabetic angiopathy, however, seem to benefit less from cilostazol. Variable glucose levels, dietary disruption, and tobacco use contributed to this outcome because these conditions increase inflammation. Combined use of cilostazol with other anticoagulant or antiaggregant agents and closer monitoring of the patients may provide better results in these patients.