CLOTS 2009.
| Methods | Study: RCT Exclusion to post‐randomisation: 0 Losses to follow up: 69 missing data (41 in treatment group and 28 in control group) ‐ no CDU prior to death or 30 days DVT diagnosis: CDU (minimum of the popliteal and femoral veins) between day 7 and 10 and between day 25 and 30 Statistical analysis: odds ratio and NNT Scheduled treatment and follow‐up period: 30 days; clinical follow up at 6 months | |
| Participants | Country: UK, Italy and Australia Total number of participants: 2518 Total available for analysis: 2518 Age: 76 years (68 to 83) for both groups Sex: males 49.4% (620/1256) in the treatment group and 49.3% in the control group (622/1262) Immobilisation: yes Inclusion criteria: patients admitted with an acute stroke up to day 3 post‐admission Exclusion criteria: peripheral vascular disease, or with diabetic/sensory neuropathy, if clinicians judged GCS could cause skin damage Full intention‐to‐treat analysis: performed | |
| Interventions | Type: thigh‐length Tyco Healthcare TED GCS
Control: 1262
Treatment: 1256
Duration applied: night and day until death/discharge/mobile/refused Use of anticoagulants post randomisation: group allocated GCS • 117 post‐randomisation prophylactic dose heparin/LMWH prescribed • 78 post‐randomisation treatment dose heparin/LMWH prescribed • 186 post‐randomisation warfarin prescribed Use of anticoagulants post randomisation: group allocated 'avoid GCS' • 129 post‐randomisation prophylactic dose heparin/LMWH prescribed • 97 post‐randomisation treatment dose heparin/LMWH prescribed • 208 post‐randomisation warfarin prescribed |
|
| Outcomes | Any DVT Control: 224 Treatment: 205 P value: ns | |
| Notes | The primary outcome focused on proximal DVTs (popliteal or femoral) rather than any DVT. Randomising clinicians were allowed to elect prior to randomisation whether patients would have a second CDU at 25 to 30 days. The 6‐month outcomes have not yet been reported. The median delay from stroke onset to enrolment was 2 days but there was no trend towards more effect with earlier recruitment | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Low risk | A ‐ Adequate. Computer‐generated allocation based on within centre minimisation for prognostic factors (stroke onset; stroke severity; leg paresis; use of anticoagulants) plus random allocation |
| Allocation concealment? | Low risk | A ‐ Adequate. Central allocation (web and telephone) |
| Blinding? All outcomes | Low risk | A ‐ Adequate. GCS removed before CDU was performed Hard copies of positive scans were recorded for independent assessment |
| Incomplete outcome data addressed? All outcomes | Low risk | A ‐ Adequate. 69 missing data |