Article type
Year
Abstract
Introduction:
Objectives: To evaluate the efficacy of cinnarizine, cyclandelate, isoxsuprine, nafronyl, pentoxifylline, xanthinol nicotinate, and buflomedil on walking distance (painfree and maximal) in intermittent claudication.
Methods: We searched Medline, the IPA database, and the Cochrane Library. We addressed distributing companies and key authors. All references in the retrieved articles were searched for relevant clinical trials. Data selection: We selected randomized placebo-controlled trials (RCT) in Fontaine stage II, with oral medication, and with as outcome walking distance (PFWD and/or MWD), analyzed by standardized exercise test. The 36 available trials were then subjected to a systematic quality assessment by the 4 authors. The main reasons to further reject trials were : too short duration between baseline and outcome assessment (<12 weeks); too small sample size (< 30 patients); failure to report details on variability of results (no standard deviation or confidence limits given). Data extraction: For each trial, the mean initial and final walking distance was extracted for the active substance and for placebo, with the standard deviation. Data synthesis: The difference between active product and placebo in gain between initial and final walking distance with their confidence intervals were calculated, using an approximative method (Altman DG, Gardner MJ. Statistics with confidence). For cinnarizine. cyclandelate. isoxsuprine. and xanthinol nicotinate the evidence is completely lacking. For buflomedil. we retrieved 6 published RCTs of which 2 were accepted after quality assessment, showing a marginally positive effect. We identified in a previous meta-analysis (with a possible conflict of interest) 5 additional, non-published negative trials. It was impossible to retrieve these trials as the authors of this meta-analysis and the company did not react adequately to our repeated requests for additional information. For naftidrofuryl. we retrieved 9 RCTs (6 acceptable of which 5 with a positive result). The existence of negative unpublished trials could not be excluded. This risk of publication bias and the great heterogeneity of the results within and between the trials precluded a (technically feasible) meta-analysis. For pentoxifylline. a literature base of 18 RCTs dwindled after careful quality assessment into only 2 acceptable studies, both with neutral results.
Results:
Discussion: There is little evidence of efficacy for vasoactive medication in PAOD. For cinnarizine, cyclandelate, isoxsuprine, and xanthinol nicotinate, their prolonged reimbursement, utilization and registration as medicines for PAOD is hardly justified. For pentoxifylline, it is unlikely that further clinical studies for this drug in PAOD could reverse this failure to provide positive evidence of efficacy. For buflomedil, the marginally positive results are undoubtly undermined by publication bias. For naftidrofuryl, further well executed, double-blind RCTs with sufficient sample size are needed to confirm the efficacy of this drug. Resources should rather be allocated to prevention and rehabilitation.
Objectives: To evaluate the efficacy of cinnarizine, cyclandelate, isoxsuprine, nafronyl, pentoxifylline, xanthinol nicotinate, and buflomedil on walking distance (painfree and maximal) in intermittent claudication.
Methods: We searched Medline, the IPA database, and the Cochrane Library. We addressed distributing companies and key authors. All references in the retrieved articles were searched for relevant clinical trials. Data selection: We selected randomized placebo-controlled trials (RCT) in Fontaine stage II, with oral medication, and with as outcome walking distance (PFWD and/or MWD), analyzed by standardized exercise test. The 36 available trials were then subjected to a systematic quality assessment by the 4 authors. The main reasons to further reject trials were : too short duration between baseline and outcome assessment (<12 weeks); too small sample size (< 30 patients); failure to report details on variability of results (no standard deviation or confidence limits given). Data extraction: For each trial, the mean initial and final walking distance was extracted for the active substance and for placebo, with the standard deviation. Data synthesis: The difference between active product and placebo in gain between initial and final walking distance with their confidence intervals were calculated, using an approximative method (Altman DG, Gardner MJ. Statistics with confidence). For cinnarizine. cyclandelate. isoxsuprine. and xanthinol nicotinate the evidence is completely lacking. For buflomedil. we retrieved 6 published RCTs of which 2 were accepted after quality assessment, showing a marginally positive effect. We identified in a previous meta-analysis (with a possible conflict of interest) 5 additional, non-published negative trials. It was impossible to retrieve these trials as the authors of this meta-analysis and the company did not react adequately to our repeated requests for additional information. For naftidrofuryl. we retrieved 9 RCTs (6 acceptable of which 5 with a positive result). The existence of negative unpublished trials could not be excluded. This risk of publication bias and the great heterogeneity of the results within and between the trials precluded a (technically feasible) meta-analysis. For pentoxifylline. a literature base of 18 RCTs dwindled after careful quality assessment into only 2 acceptable studies, both with neutral results.
Results:
Discussion: There is little evidence of efficacy for vasoactive medication in PAOD. For cinnarizine, cyclandelate, isoxsuprine, and xanthinol nicotinate, their prolonged reimbursement, utilization and registration as medicines for PAOD is hardly justified. For pentoxifylline, it is unlikely that further clinical studies for this drug in PAOD could reverse this failure to provide positive evidence of efficacy. For buflomedil, the marginally positive results are undoubtly undermined by publication bias. For naftidrofuryl, further well executed, double-blind RCTs with sufficient sample size are needed to confirm the efficacy of this drug. Resources should rather be allocated to prevention and rehabilitation.