Atherectomy for peripheral arterial disease


Peripheral arterial disease is a narrowing or blockage of the arteries in the legs. People with this condition can experience pain on walking, pain at rest, or leg ulceration due to poor blood supply. Treatment options are: surgery, using a blood vessel or graft to bypass the narrowed or blocked section of the artery; balloon angioplasty, when a deflated balloon is passed into the narrowing at the end of a wire, then blown up to stretch the artery; and stenting (used in addition to balloon angioplasty), which holds open the balloon-stretched section for extra support. A final option, less commonly used, is a technique called atherectomy. This treatment cuts or grinds away the fatty deposition (atheroma) within the artery that is causing the narrowing or occlusion.

Key results

In this review, we compared atherectomy with the other treatment options described above. We also looked within the two groups to assess whether using drug-releasing balloons or stents impacted on participants' outcomes. We identified seven studies with a total of 527 participants.

Six trials compared atherectomy against balloon angioplasty (372 participants, 427 treated lesions). We found no clear difference between the procedures when examining artery patency at six and 12 months, risk of death, initial procedure failure rates, need to re-treat the artery, risk of forming clots (embolisation), complication rates or risk of amputation. We found that atherectomy was associated with lower rates of emergency stenting during the procedure and lower balloon inflation pressures when compared with balloon angioplasty alone. We found no difference in results depending on whether the balloons were drug-releasing or not.

One study compared atherectomy against balloon angioplasty and primary stenting (155 participants and 155 treated lesions). This study did not report primary patency. We found no clear difference between the treatment arms in risk of death, complication rates, cardiovascular events and the need to re-treat the artery. This study found no initial procedure failure events,

We did not find any studies that compared bypass surgery against atherectomy.

Certainty of the evidence

Overall, our certainty in the evidence is very low, which means we do not have confidence that our results show the true effect of the treatments. We downgraded our certainty in the evidence because the studies were at high risk of bias (lack of blinding of participants or assessors, several outcomes were not reported and a number of the participants did not complete the studies); the trials were all small; and their results were inconsistent.


In conclusion, we have found no clear difference in effect on patency, mortality or cardiovascular event rates when comparing atherectomy against balloon angioplasty with or without stenting. The limited evidence available does not support a significant advantage of atherectomy over conventional balloon angioplasty or stenting.

Authors' conclusions: 

This review update shows that the evidence is very uncertain about the effect of atherectomy on patency, mortality and cardiovascular event rates compared to plain balloon angioplasty, with or without stenting. We detected no clear differences in initial technical failure rates or TVR, but there may be reduced dissection and bailout stenting after atherectomy although this is uncertain. Included studies were small, heterogenous and at high risk of bias. Larger studies powered to detect clinically meaningful, patient-centred outcomes are required.

Read the full abstract...

Symptomatic peripheral arterial disease (PAD) has several treatment options, including angioplasty, stenting, exercise therapy, and bypass surgery. Atherectomy is an alternative procedure, in which atheroma is cut or ground away within the artery. This is the first update of a Cochrane Review published in 2014.


To evaluate the effectiveness of atherectomy for peripheral arterial disease compared to other established treatments.

Search strategy: 

The Cochrane Vascular Information Specialist searched the Cochrane Vascular Specialised Register, Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, Cumulative Index to Nursing and Allied Health Literature (CINAHL) and Allied and Complementary Medicine (AMED) databases, and the World Health Organization International Clinical Trials Registry Platform and trials registers to 12 August 2019.

Selection criteria: 

We included all randomised controlled trials that compared atherectomy with other established treatments. All participants had symptomatic PAD with either claudication or critical limb ischaemia and evidence of lower limb arterial disease.

Data collection and analysis: 

Two review authors screened studies for inclusion, extracted data, assessed risk of bias and used GRADE criteria to assess the certainty of the evidence. We resolved any disagreements through discussion. Outcomes of interest were: primary patency (at six and 12 months), all-cause mortality, fatal and non-fatal cardiovascular events, initial technical failure rates, target vessel revascularisation rates (TVR; at six and 12 months); and complications.

Main results: 

We included seven studies, with a total of 527 participants and 581 treated lesions. We found two comparisons: atherectomy versus balloon angioplasty (BA) and atherectomy versus BA with primary stenting. No studies compared atherectomy with bypass surgery. Overall, the evidence from this review was of very low certainty, due to a high risk of bias, imprecision and inconsistency.

Six studies (372 participants, 427 treated lesions) compared atherectomy versus BA. We found no clear difference between atherectomy and BA for the primary outcomes: six-month primary patency rates (risk ratio (RR) 1.06, 95% confidence interval (CI) 0.94 to 1.20; 3 studies, 186 participants; very low-certainty evidence); 12-month primary patency rates (RR 1.20, 95% CI 0.78 to 1.84; 2 studies, 149 participants; very low-certainty evidence) or mortality rates (RR 0.50, 95% CI 0.10 to 2.66, 3 studies, 210 participants, very low-certainty evidence). One study reported cardiac failure and acute coronary syndrome as causes of death at 24 months but it was unclear which arm the participants belonged to, and one study reported no cardiovascular events.

There was no clear difference when examining: initial technical failure rates (RR 0.48, 95% CI 0.22 to 1.08; 6 studies, 425 treated vessels; very low-certainty evidence), six-month TVR (RR 0.51, 95% CI 0.06 to 4.42; 2 studies, 136 treated vessels; very low-certainty evidence) or 12-month TVR (RR 0.59, 95% CI 0.25 to 1.42; 3 studies, 176 treated vessels; very low-certainty evidence). All six studies reported complication rates (RR 0.69, 95% CI 0.28 to 1.68; 6 studies, 387 participants; very low-certainty evidence) and embolisation events (RR 2.51, 95% CI 0.64 to 9.80; 6 studies, 387 participants; very low-certainty evidence). Atherectomy may be less likely to cause dissection (RR 0.28, 95% CI 0.14 to 0.54; 4 studies, 290 participants; very low-certainty evidence) and may be associated with a reduction in bailout stenting (RR 0.26, 95% CI 0.09 to 0.74; 4 studies, 315 treated vessels; very low-certainty evidence). Four studies reported amputation rates, with only one amputation event recorded in a BA participant. We used subgroup analysis to compare the effect of plain balloons/stents and drug-eluting balloons/stents, but did not detect any differences between the subgroups.

One study (155 participants, 155 treated lesions) compared atherectomy versus BA and primary stenting, so comparison was extremely limited and subject to imprecision. This study did not report primary patency. The study reported one death (RR 0.38, 95% CI 0.04 to 3.23; 155 participants; very low-certainty evidence) and three complication events (RR 7.04, 95% CI 0.80 to 62.23; 155 participants; very low-certainty evidence) in a very small data set, making conclusions unreliable. We found no clear difference between the treatment arms in cardiovascular events (RR 0.38, 95% CI 0.04 to 3.23; 155 participants; very low-certainty evidence). This study found no initial technical failure events, and TVR rates at six and 24 months showed little difference between treatment arms (RR 2.27, 95% CI 0.95 to 5.46; 155 participants; very low-certainty evidence and RR 2.05, 95% CI 0.96 to 4.37; 155 participants; very low-certainty evidence, respectively).