The Caprini score: what the total counts and why risk levels differ

The model is a checklist. Each risk factor that is present adds its weight: one point for factors such as varicose veins, a BMI over 25, minor surgery or a medical patient at bed rest; two points for arthroscopic surgery, malignancy, major surgery over 45 minutes, central venous access or an immobilising cast; three points for a history of DVT or pulmonary embolism, a family history of thrombosis or an inherited or acquired thrombophilia; and five points for elective major lower limb arthroplasty, a hip, pelvis or leg fracture, a stroke, multiple trauma or an acute spinal cord injury with paralysis, the last four within the past month.

Age is scored in three bands — 41 to 60 years 1 point, 61 to 74 years 2 points, 75 years or over 3 points. The published table leaves the boundaries ambiguous, overlapping at 60 and omitting 75, so implementations have to choose a reading. This site scores 61 as the start of the middle band and 75 as the start of the top one.

Elective major lower extremity arthroplasty and a hip, pelvis or leg fracture within the last month are each worth 5 points before any other factor is counted. A 67-year-old having a knee replacement who is also obese and has a history of cancer reaches 10 points: 2 for age, 5 for the arthroplasty, 1 for the BMI and 2 for the malignancy.

The model does not say what to do about a total. The original publication, and the protocols built on it, group totals into risk levels that drive the choice of prophylaxis, and those groupings differ between versions of the model and between institutions. A level shown without the version and the protocol behind it would imply a precision the model does not have.

The model has also been revised since 2005, with factors added and weights changed, so a total from one version is not comparable with a total from another. Use the total to document the risk factors, and apply your hospital’s protocol to decide on prophylaxis, alongside the patient’s bleeding risk.

The Caprini score is a risk assessment, not a diagnostic test. It says nothing about whether a clot is already present; for suspected deep vein thrombosis, use a pre-test probability rule such as the Wells score. And its weights were assigned from the literature and clinical experience rather than fitted to outcome data, so it is not a calibrated estimate of an individual patient’s risk.