Interactive tool
Calculate the Caprini score
Select the age band and tick all the factors present. The calculator applies the 2005version, the one adopted by the ACCP guidelines and the most widely validated.
How to use it
How to use it
When it applies
The model was developed and validated in surgicalpatients: general, vascular, urological, gynaecological, plastic and reconstructive, and orthopaedic surgery. For hospitalised medical patients the appropriate tool is the Padua score or the IMPROVE: applying Caprini to medical inpatients means using it outside the setting in which it was derived.
When to repeat it
- On admission, before surgery.
- After surgery, adding the items that the operation has made true: duration > 45 minutes, central venous access, immobilisation, bed rest > 72 hours. The preoperative score almost always underestimates the real risk.
- At every relevant clinical change during the hospital stay (infection, return to theatre, prolonged immobilisation).
Notes on the criteria
- Age: the bands are mutually exclusive; the calculator applies only one.
- Major surgery: an operation lasting > 45 minutes; the item can be added to prolonged laparoscopic or arthroscopic surgery only if they are separate procedures.
- Malignancy: present or previous. The score does not distinguish between a tumour under active treatment and one in remission, which is one of the recognised limitations of the model.
- Thrombophilia: each documented abnormality scores 3 points and the items are cumulative (for example factor V Leiden + antiphospholipid antibodies = 6 points).
- Items for women only: oral contraception or hormone replacement therapy, pregnancy or the puerperium within the last month, unexplained adverse obstetric history.
- The 5-point items (recent stroke, elective arthroplasty, hip, pelvis or leg fracture, acute spinal cord injury, multiple trauma) on their own place the patient in the highest risk class.
Caprini estimates the thromboticrisk. The decision on pharmacological prophylaxis comes from weighing it against the bleeding risk of the individual patient and the individual operation: in neurosurgery, eye surgery or with active bleeding, mechanical prophylaxis may remain the only option even at high scores.
Clinical reference
Guidelines and interpretation
| Score | Risk class | VTE risk without prophylaxis | Recommended strategy |
|---|---|---|---|
| 0 | Very low | < 0.5% | Early ambulation; no specific prophylaxis |
| 1 – 2 | Low | ~1.5% | Mechanical prophylaxis (intermittent pneumatic compression or graduated compression stockings) |
| 3 – 4 | Moderate | ~3.0% | Pharmacological prophylaxis (LMWH or unfractionated heparin) or mechanical if bleeding risk is high |
| 5 – 6 | High | ~6.0% | Pharmacological prophylaxis combined with mechanical prophylaxis |
| 7 – 8 | High | > 6% | Pharmacological + mechanical; consider extended duration |
| ≥ 9 | Highest | > 6%, up to ~11% | Pharmacological + mechanical, with prophylaxis extended to 30 days after major abdominal or pelvic cancer surgery |
The risk estimates up to 5 points are those of the 2012 ACCP guidelines (Gould et al.), which group all scores ≥ 5 into the high class (~6%). At higher scores the risk keeps rising: in the Pannucci series (plastic surgery, without pharmacological prophylaxis) a score > 8 was associated with VTE in 11.3% of cases.
Duration of prophylaxis
- General surgery at moderate to high risk: prophylaxis for the whole hospital stay and until independent walking is regained, usually 7–10 days.
- Major abdominal or pelvic cancer surgery: prophylaxis extended to 4 weeks after discharge, with a significant reduction in symptomatic VTE (ACCP and ESMO recommendation).
- Hip arthroplasty: 28–35 days. Knee arthroplasty: at least 10–14 days, extendable to 35.
- Hip fracture: 28–35 days.
- Bariatric surgery: individualised duration, with weight-adjusted doses.
Dosing considerations
In obese patients the standard enoxaparin dose of 4000 IU (40 mg) daily may be insufficient: more recent recommendations use a weight-proportional increase (for example 4000 IU every 12 hours or 6000 IU daily above 100–120 kg). In severe renal impairment (clearance < 30 mL/min) the LMWH dose should be reduced or unfractionated heparin preferred. The timing of the first dose relative to surgery and to any neuraxial anaesthesia follows the safety intervals recommended by the anaesthesiology societies (ESAIC, ASRA).
Recognised limitations
- The model was derived by consensus, not by multivariable analysis: the weights of the individual factors do not come from regression coefficients.
- Score distribution varies widely between specialties: in plastic surgery most patients fall between 3 and 6, in joint replacement surgery almost all exceed 8. Decision thresholds should be calibrated to the setting.
- Later versions of the model exist (for example the 2013version), which add and redefine some items; this calculator implements the 2005 version, the one adopted by the ACCP guidelines.
- A systematic review published in the Journal of Vascular Surgery: Venous and Lymphatic Disorders (2022) documented considerable heterogeneity in the risk categories used in the literature: when quoting a score, it is advisable to state which set of thresholds was used.
Bibliography
Sources
- Caprini JA., Dis Mon 2005 — Thrombosis risk assessment as a guide to quality patient care.
- Gould MK. et al., Chest 2012 (ACCP 9th ed.) — Prevention of VTE in nonorthopedic surgical patients: Antithrombotic Therapy and Prevention of Thrombosis.
- Bahl V. et al., Ann Surg 2010 — A validation study of a retrospective venous thromboembolism risk scoring method.
- Pannucci CJ. et al., J Am Coll Surg 2011 — Validation of the Caprini risk assessment model in plastic and reconstructive surgery patients (VTE risk of 11.3% with a score > 8 without pharmacological prophylaxis).
- Journal of Vascular Surgery: Venous and Lymphatic Disorders, 2022 — Systematic review of venous thromboembolism risk categories derived from Caprini score.
A quick-reference tool for healthcare professionals. It does not replace the original guidelines, the summaries of product characteristics or the specialist's clinical judgement. All calculations run in the browser: no patient data are transmitted to or stored on a server.