The association between peripheral ischaemia, throm-bosis and malignancy has been well documented [1,2]. Bergan et al.  found malignancy to be the second leading cause of death in patients with peripheral vas-cular disease .
Armound Trousseau was the first to note microangi-opathy in superficial veins of patients with malignancy in the mid-1800s.
“I wish also to recall to your recollection some of the facts which prove that the same spontenous coagulation is common in cancerous patients, and gives rise to phlegmasiaalbadolens” .
He noted a high incidence of deep vein thrombosis (phlegmasiaalbadolens) in patients with occult malignancy. He suggested a “crisis of blood”, which fa-voured intravenous coagulation.
“… as I have told you, there exist a special crisis of blood, which irrespective of inflammation, favors intra-venous coagulation”.
This phenomenon was later called “Trousseau’s syn-drome” and has been shown to have multiple mechan-isms .
Subsequent research demonstrated that patients with malignancy are seven times more likely to develop venous thrombosis . Available data suggest that venous thrombosis in cancer patients is more common than arterial thrombosis. Khorana AA et al. showed that 5.4% and 1.5% of hospitalized patients with ma-lignancy developed venous and arterial thromboembolism respectively . Although less common, arterial thrombosis might lead to limb or arm ischaemia, cardiac arrest and strokes .
According to Khourana et al., about 1.5 – 3.1% of pa-tients with malignancy have limb ischaemia [5,12]. Previous studies have suggested that limb ischaemia might be suggestive of malignancy. El Sakkaet al.  also reported 11% of patients who presented with criti-cal limb ischaemia had malignancy. A figure of 12% was reported by Morris et al .
There is a considerable disparity in the literature about what constitutes limb ischaemia; therefore attention is drawn to Table 1 delineating the classification system proposed by the Rutherford team.
All the studies recorded a higher mortality in cancer patients compared to cancer-free patients.
Table 1. Rutherford scoring system .
The aetiology of these thrombotic events in cancer patient is most likely multifactorial. Kemik O et al. showed an increase in tumour necrosis factor alpha (TNF α), and in interleukins (IL) IL-1, IL-6, IL-8 and IL-10 in patients with malignancy . TNF-α and IL-1 cause an aberration in the production of thrombomulin and further fibrinolysis inhibitor PAI-1 production, leading to a higher risk of thrombosis . Increased factor VIII antigen, restocetin cofactorand shortened platelet half-life  in endometrial cancer, cervical cancer and myeloma respectively leads to platelet hyperactivity. Platelet aggregation is therefore inappropriately induced and causes thrombosis.
The hypercoagulability and prothrombotic state seen in cancer patients might also be iatrogenic. Routinely used chemotherapy regimes have been shown to in-crease the risk of thromboembolic events. Tamoxifen increases the 5-year risk of a thrombotic event in breast cancer patients . Khorana AA et al. also showed that thrombosis is the second leading cause of death in patients with active malignancy receiving am-bulatory chemotherapy .
Finally, compliance with prophylactic anticoagulation guidelines in cancer patients has been shown to be quite low  as patients with active malignancy are at a higher risk of adverse effect and drug interactions also have to be taken into consideration .
We reviewed six studies have looked at the outcome of cancer patients with limb ischaemia [15-18,20,21].
Three studies recorded an even higher mortality in pa-tients managed invasively inevitably suggesting a more palliative approach to treatment is ideal [16,20,21].
The other three studies [15-18] found no difference in mortality rates from one treatment modality to another and suggested patients with limb ischaemia and malignancy be managed similarly to cancer-free patients.
Most recently (2014), E Mouhayaret al.  studied the difference in outcome of 74 cancer patients with acute limb ischaemia who received either catheter based intervention, anticoagulation, “no therapy”, surgery, or amputations. Surgical interventions wereth-rombectomies, embolectomies ± bypass surgery. Ca-theter based interventions carried out were: catheter direct thrombolysis, arterial angioplasty ± stenting. The Rutherford staging criteria was used to defined ischae-mia ‘’based on the initial recordings by the treating physician’’. 18% of the subjects were classified as Ru-therford class I; this group received anticoagulation therapy with no limb loss and a 30day mortality of 8%. Class IIa (40%) had an amputation rate of 17%. 35% of the catheter based therapy performed failed and the 30 day mortality in this cohort of patients was 10%.
Class IIb (39%) had an amputation rate of 14% and 33% of the catheter based therapy failed in this group. 76% of patients in class IIb required surgery. The mor-tality rate at 30days in class IIb patients was 34%. Patients with class III ischaemia had an amputation rate of 50% and a 30day mortality of 50%. Cumulatively, amputation free survival rates were 80%, and 47% for 1month and 1year respectively .
This indicated that mortality increased with severity of ischaemia ± cancer prognosis and not treatment regi-men. However, only 2 patients were Rutherford class III and all the others were class IIb or below reflecting a relatively mild severity of ischaemia within their pa-tient population .
The scenario in the TOPAS trial published in 1998 by Kenneth Ouriel and Frank J Veith was somewhat similar. 544 patients, 64 with a history of malignancy were analysed. These patients were divided into two groups: surgical and thrombolyt-ic . In the first phase of the trial, patients with acute limb ischaemia, defined as lower limb ischaemia for 14days of less were analysed. Randomisation was done after arteriographic documentation of occlusion. Standard thrombolytic treatment was 400IU/min for 4hours followed by 2000IU/min of infusion. Detailed outcomes of different therapies within both groups were not given however the 1-year amputation free survival for patients with a history of malignancy was 50.6% . They found 12% of patients with limb ischaemia had a malignancy. Similar to 11% in El Sakkaet al.  and 12% in Morris-stiff et al. .
Tsang et al. looked at acute limb ischaemia in 419 patients. 16 of these patients had cancer and of these, 8 had a prior diagnosis and 4 patients malignancies were incidental findings after presenting with limb ischaemia. 1year survival was 44% for patients with invasive treatment .
El Sakkaet al.  published 2005 reported high mor-tality in 22 patients with critical or acute limb ischae-mia and cancer. When compared to 20.6% in the can-cer free patient population, 50% of patients with ma-lignancy managed by either thrombolysis alone, an-gioplasty, and surgical bypass died within 6months. “Critical limb ischaemia was defined as chronic ischaemia rest pain, ulcers or gangrene attributed to objectively proven arterial occlusive disease and acute leg ischaemia as a sudden decrease or worsening in limb perfusion causing a potential threat to extremity viability” . Magnetic resonance angioplasty, digital subtraction arteriography were performed if indicated. 11 of the 22 cancer patients were managed with anti-coagulation only, 1 bypass, and 2 with thrombolysis and angioplasty. Angioplasty alone and thromboembolectomy therapies were carried out on 4 patients each. Only one of the eight cancer patients who underwent an operation was alive 1 year postoperatively. They also showed that acute limb ischaemia in cancer patients was more prevalent than chronic .
M Javidet al.  published in 2007; studied arterial thrombosis in 20 patients with cancer. 22 limbs and 1 arm ischaemia presented with acute ischaemia. 1 of the 20 patients had presented with limb ischaemia 6years before, 3 had a past medical history of leg ischaemia, 2 patients had intermittent claudication, 2 had evidence of thrombosis in both legs, and 1 underwent an angioplasty 18months previously. 4 patients where managed with “palliation” and 8 conservatively. Conservative management involved anticoagulation, analgesia and glyceryltrinitrate patches. The other 8 patients were managed surgically 5 thromboembolectomies and 3 bypass procedures. Some procedures failed and had to be repeated. 7 out of 9 thromboebolectomies failed, 4 of who failed within 24 hours of the procedure. 3 bypasses were carried out and the patients with the only successful bypass died within 6months. 83% of the patients with thrombosis and malignancy died within a year. Only one patient received a below-knee amputation and died 5months later .
G. Morris-stiff et al.  published in 2010 compared the surgical outcomes of 14 patients with malignancy and acute limb ischaemia to that of 102 cancer free patients. A strictly surgical study who only included patients whom “anembolectomy was attempted on a naï?ve artery were included”. The patients were ma-naged according to guidelines, irrespective of their cancer status. 29% of the patients with malignancy required an amputation and similarly a high recurrence rate of 29% was recorded. The 60-day mortality was 100% in the cancer patients’ cohort and 35% in cancer-free patients. Patients with cancer also showed a high recurrence rate and higher need for amputations due to failed surgeries .
The findings and recordings in the TOPAS trial and E Mouhayar et al.  seem to be most accurate. The number patients recruited and the length of their stu-dies also adds strength to their findings.
Vascular surgery secondary to peripheral ischaemia in cancer patients with very poor prognosis from the ma-lignant burden has very limited success rates. Man-agement in this cohort of patients is hugely individua-listic as prognosis is different for each patient and has to influence management.
Ultimately, the management of this heterogenous group of patients should be individualised as much as possible. Factors such as age, comorbidities, tumour stage and grade and predicted course are all important. Of course the duration, complexity and recovery time of the proposed revascularisation procedure are key issues. Once these factors are addressed and accounted for can an individualised management plan be formulated.
Where possible, when local facilities and expertise allow, consideration should be given to the least inva-sive treatment modality. This will obviously depend on the extent and degree of ischaemia but a prompt simple intervention under local anaesthesia should be preferred over complex reconstrucion. The rationale behind this is that these patients have been shown to have severely truncated lifespan and should not be exposed to procedures with a long recovery or healing time (see Diagram 1).
Finally, the complexities of limited lifespan with a ma-jor amputation and the combined impact on quality of life have not been examined, probably because of the relative frailty of this subgroup of patients. The ulti-mate decision to amputate or palliate such patients remains a very difficult consultation with the patient and family.
2. Trousseau A. Phlegmasiaalbadolens.Clinique Medi-cale de l’HotelDieu de Paris.1865, 3:654–712. Translated by the New Sydenham Society, Lon-don as ‘Lectures on Clinical Medicine delivered at the Hotel Dieu de Paris’, 281-95, 1872.
7. Kemik O, Sumer A, Kemik AS, Hasirci I, Purisa S et al. The relationship among acute-phase response proteins, cytokines and hormones in cachectic patients with colon cancer. World Journal of Surgical Oncology. 2010, 8: 85.
9. S. J. Scialla, S. F. Speckart, M. J. Haut, D. B. Kim-ball, “Alterations in platelet surface sialyltransferase activity and platelet aggregation in a group of cancer patients with a high incidence of thrombosis,”Cancer Research. 1979, 61(39):
11. Hernandez RK, Sørensen HT, Pedersen L, Jacobsen J, Lash TL..Tamoxifen treatment and risk of deep venous thrombosis and pulmonary embolism: a Danish population-based cohort study. Cancer. 2009, 115(19): 4442-4449.
12. Khorana AA, Francis CW, Culakova E, Kuderer NM, Lyman GH. Thromboembolism is a leading cause of death in cancer patients receiving outpatient chemotherapy. Journal of ThrombHaemost. 2007, 5: 632-634.
13. Hikmat Abdel-Razeq, Faisal Albadainah, ShadiHijjawi, Asem Mansour, ImadTreish. Venous thromboembolism (VTE) in hospitalized cancer patients: prophylaxis failure or failure to prophylax!.Journal of Thrombosis and Thrombolysis. 2011, 31(1): 107-112.
14. Hutten BA, Prins MH, Gent M, Ginsberg J, Tijssen JG, Büller HR. Incidence of recurrent thromboembolic and bleeding complications among patients with ven-ous thromboembolism in relation to both malignancy and achieved international normalized ratio: a retros-pective analysis. Journal of Clinical Oncolology. 2002, 18: 3078-3083.
19. Rutherford RB, Baker JD, Ernst C, Johnston KW, Porter JM et al. Recommended standards for reports dealing with lower extremity ischemia: Revised ver-sion. Journal Vascular Surgury. 1997, 26: 517-538.