Venous thromboembolism (VTE) prevention measures are persistently underused, and their use is substantially lower in medical inpatients than in surgical inpatients. It is reported that anticoagulant prophylaxis causes a 0•6% absolute increase in major bleeding in patients who have had a stroke, whereas the degree of protection against VTE is debated, thus intermittent pneumatic compression devices (IPCs), which do not increase bleeding risk, represent an attractive alternative. An IPC includes inflatable sleeves that are wrapped around the legs and secured by Velcro. The sleeves can be applied to the calf alone, or the calf and thigh. They are inflated, one side at a time, to compress the legs at intervals. They are thought to reduce the risk of venous thrombosis by reducing stasis and stimulating release of intrinsic fibrinolytic factors. IPCs are often used in non-surgical patients, on the basis of indirect evidence from surgical populations, though it cannot be assumed that the benefits seen in surgical patients will apply to medical patients. Decades after IPCs were first reported to prevent VTE in surgical patients, researchers writing in the Lancet present the results of the randomised CLOTS 3 trial, the highest quality, and the largest, study to date of IPC use for VTE prophylaxis in non-surgical patients, which showed that they reduce the risk of VTE in immobilised medical inpatients who have had a stroke.
This multicentre UK study enrolled 2876 immobile patients (median age of 76 years; 67 to 84) with acute stroke from day 0 to day 3 of admission and randomised them to either IPC or no IPC. Background prophylaxis with additional methods was allowed, and anticoagulant prophylaxis or therapeutic anticoagulation was given to about a third of patients in both intervention groups. Also 118 (8%) patients allocated IPCs and 42 (3%) control patients used elastic stockings. A technician carried out a compression duplex ultrasound (CDU) of both legs at 7 to 10 days and, wherever practical, at 25 to 30 days after enrolment. Patients were followed up for 6 months. The primary outcome, DVT in the proximal veins detected on a screening CDU or any symptomatic DVT in the proximal veins, confirmed on imaging, within 30 days of randomisation, occurred in 122 (8.5%) of 1438 patients using IPC and 174 (12.1%) of 1438 patients allocated no IPC (absolute risk reduction 3.6%). Excluding the 323 patients who died before any primary outcome and 41 without any screening CDU, the adjusted odds ratio was 0.65 (95% CI 0.51—0.84; p=0.001). Deaths in the treatment period occurred in 156 (11%) of patients allocated IPC and 189 (13%) patients allocated no IPC (p=0•057). Skin breaks occurred more frequently in the IPC group than in the no IPC group (3% vs. 1%; p= 0.002), and a non-significant increase in falls with injury was noted in the IPC group.
The researchers acknowledge some limitations to their study such as: moderate adherence to IPC; imperfect masking of the technicians and no masking of caregivers or patients; some scheduled CDUs did not include the calf veins and some were missing; and there was no systematic screening for pulmonary emboli. All of these might mean the frequency of VTE was underestimated. Furthermore, as there was systematic screening for them, many patients found to have asymptomatic DVT were then treated with anticoagulants, which might bias the estimate of the effect of IPC. Other potential limitations included: lack of central verification of negative scans, use of selective source data verification and imbalance in the background use of graduated compression stockings, though the researchers consider these unlikely to have introduced bias or altered the external validity of the results. They therefore conclude that their study provides clear evidence that IPC is effective in reducing the risk of DVT in immobile patients who have had a stroke, including those patients with haemorrhagic stroke.
An accompanying Comment article notes that the rate of symptomatic (vs. asymptomatic) DVT was unexpectedly high, which could be attributed to the method of ascertainment; had more traditional methodology been used, more DVTs would have been classified as asymptomatic, and a smaller reduction in symptomatic DVT would probably have been reported. In addition, IPCs are hard to keep on patients and perfect adherence occurred in less than a third of patients; outside a trial, IPC adherence might be even worse, leading to less benefit. Nonetheless, the authors consider that the data from CLOTS 3 is convincing, with the benefit of IPCs in reducing VTE outweighing the risk of skin complications. More studies of IPCs in other medical inpatients are however needed.
In January 2010 NICE issued clinical guidance on reducing the risk of VTE in patients admitted to hospital, which recommended that the choice of mechanical VTE prophylaxis should be based on individual patient factors such as clinical condition, surgical procedure and patient preference, with a choice from anti-embolism stockings, foot impulse devices and IPCs. With regards to patients with stroke, it suggests offering prophylactic-dose LMWH (or UFH for patients with renal failure) if a diagnosis of haemorrhagic stroke has been excluded, AND the risk of bleeding (haemorrhagic transformation of stroke or bleeding into another site) is assessed to be low, AND the patient has one or more of: major restriction of mobility, previous history of VTE, dehydration, comorbidities (such as malignant disease). In addition, it advises that until the patient can have pharmacological VTE prophylaxis, consider offering a foot impulse or IPC device. In June 2010, NICE issued a quality standard covering the reduction in risk of VTE in adults admitted as hospital inpatients or formally admitted to a hospital bed for day-case procedures. In July 2010, the Government launched the VTE prevention programme in the NHS in England, which included a measure to ensure the NHS assesses 90% of admitted patients for the risk of VTE and treats those who need it.