BACKGROUND: Peripheral arterial disease (PAD) is a common circulatory problem that can lead to reduced blood flow to the limbs, which may result in critical limb ischaemia (CLI), a painful manifestation that occurs when a person is at rest. The mainstay of treatment for CLI is surgical or endovascular repair. However, when these means of treatment are not suitable, due to anatomical reasons or comorbidities, treatment for pain is limited. Lumbar sympathectomy and prostanoids have both been shown to reduce pain from CLI in people who suffer from non-reconstructable PAD, but there is currently insufficient evidence to determine if one treatment is superior. Due to the severity of the rest pain caused by CLI, and its impact on quality of life, it is important that people are receiving the best pain relief treatment available, therefore interest in this area of research is high. OBJECTIVES: To compare the efficacy of lumbar sympathectomy with prostanoid infusion in improving symptoms and function and avoiding amputation in people with critical limb ischaemia (CLI) due to non-reconstructable peripheral arterial disease (PAD). SEARCH METHODS: The Cochrane Vascular Information Specialist (CIS) searched the Specialised Register (last searched 29 March 2017) and CENTRAL (2017, Issue 2). The CIS also searched clinical trials databases for ongoing or unpublished studies. SELECTION CRITERIA: Randomised controlled trials (RCTs), with parallel treatment groups, that compared lumbar sympathectomy (surgical or chemical) with prostanoids (any type and dosage) in people with CLI due to non-reconstructable PAD. DATA COLLECTION AND ANALYSIS: Three review authors independently selected trials, extracted data and assessed risk of bias. Any disagreements were resolved by discussion. We performed fixed-effect model meta-analyses, when there was no overt sign of heterogeneity, with risk ratios (RRs) and 95% confidence intervals (CIs). We graded the quality of evidence according to GRADE. MAIN RESULTS: We included a single study in this review comparing lumbar sympathectomy with prostanoids for the treatment of CLI in people with non-reconstructable PAD. The single study included 200 participants with Buerger's disease, a form of PAD, 100 in each treatment group, but only 162 were actually included in the analyses. The study compared an open surgical technique for lumbar sympathectomy with the prostanoid, iloprost, and followed participants for 24 weeks.Risk of bias was low for most evaluated domains. Due to the nature of the treatment, blinding of the participants and those providing the treatment would be impossible as a surgical procedure was compared with intravenous injections. It was not mentioned if blinded assessors evaluated the study outcomes, therefore, we judged subjective outcomes (i.e. pain reduction) to be at unclear risk of detection bias and objective outcomes (i.e. ulcer healing, amputation and mortality) at low risk of detection bias. We also rated the risk of attrition bias as unclear; 38 out of 200 (19%) participants were not included in the analysis without clear explanation (16 of 100 in the iloprost arm and 22 of 100 in the sympathectomy arm). The quality of evidence was low due to serious imprecision because the study numbers were low and there was only one study included.The single included study reported on the outcome of complete healing without pain or major amputation, which fell under three separate outcomes for our review: relief of rest pain, complete ulcer healing and avoidance of major amputation. We chose to keep the outcome as a singularly reported outcome in order to not introduce bias into the outcomes, which may have been the case if reported separately. The limited evidence suggests participants who received prostaglandins had improved complete ulcer healing without rest pain or major amputation when compared with those who received lumbar sympathectomy (RR 1.63, 95% CI 1.30 to 2.05), but as it was the only included study, we rated the data as low-quality and could not draw any overall conclusions. The study authors stated that more participants who received prostaglandins reported adverse effects, such as headache, flushing, nausea and abdominal discomfort, but only one participant experienced severe enough adverse effects to drop out. Five participants who underwent lumbar sympathectomy reported minor wound infection (low-quality evidence). There was no reported mortality in either of the treatment groups (low-quality evidence).The included study did not report on claudication distances, quality of life or functional status, ankle brachial pressure index (ABPI), tissue oxygenation or toe pressures, or progression to minor amputation, complications or provide any cost-effectiveness data. AUTHORS' CONCLUSIONS: Low-quality evidence from a single study in a select group of participants (people with Buerger's disease) suggests that prostaglandins are superior to open surgical lumbar sympathectomy for complete ulcer healing without rest pain or major amputation, but possibly incur more adverse effects. Further studies are needed to better understand if prostaglandins truly are more efficacious than open surgical lumbar sympathectomy and if there are any concerns with adverse effects. It would be of great importance for future studies to include other forms of PAD (as Buerger's disease is a select type of PAD), other methods of sympathectomy as well as data on quality of life, complications and cost-effectiveness.