Comparison of Incidence and Investigation of In-Hospital Pulmonary Embolism In Patients Receiving Dalteparin Prophylaxis Following Elective Knee Replacement at Two Hospitals

The incidence of in-hospital pulmonary embolism (PE)in patients who received dalteparin prophylaxis following elective total knee replacement (TKR) at a tertiary care centre was reported to be surprisingly high at 4.6% (Kerr et al, Thromb Haemost 2010, 103:123-128).

To compare the incidence and investigation of in-hospital PE between two centres that used dalteparin prophylaxis following elective TKR.

Charts of consecutive patients who underwent elective TKR and received dalteparin prophylaxis between January and June 2008 were retrospectively reviewed. A standardized case report form was used to abstract data on patient age, gender, type of surgery, risk factors for venous thromboembolism (VTE)(i.e. active cancer, prior history of VTE, hormonal therapy), results of imaging studies performed for suspected VTE, and in-hospital death. Electronic records were reviewed for 90 days after the date of surgery to capture subsequent studies for VTE performed at the same institution. Because the original study identified postoperative day 3 (POD#3) as the day that the majority of investigations were ordered for PE, a subgroup of 100 patients were randomly selected at each site, and data were abstracted on vital statistics which may have raised suspicion of PE: requirement for at least 2 litres of oxygen to keep oxygen saturation ≥ 92% or heart rate > 100 beats per minute or systolic blood pressure < 100 mmHg (These values were selected based on data from the original study).

The protocol at both hospitals dictated that the first dose of dalteparin (5,000 units subcut) was to be given on POD#1. At Hospital A only, 54 patients (12%) received dalteparin 2,500 units subcut 6 hours after surgery (personal practice of one surgeon).

None of the in-hospital PEs were fatal. The incidence of VTE (non-fatal and fatal) after discharge is less reliable because patients may have presented for medical attention at centres other than the hospital where they underwent surgery or died at home.

In the POD#3 substudy, the proportion of patients investigated for in-hospital PE was 11% and 3%, at Hospital A and Hospital B, respectively and the proportion of patients investigated for PE who were diagnosed with PE was 54% (6/11) and 0, respectively. The proportion of patients with at least one vital statistic which may have raised suspicion of PE on POD#3(hypoxia, tachycardia, hypotension) was similar at the two sites (45% vs 43%).

The higher incidence of in-hospital PE in one of two hospitals using the same method of thromboprophylaxis following elective TKR was likely the result of a more aggressive approach to investigation. The clinical importance of detecting these emboli is uncertain.

Crowther:Pfizer: Consultancy; Leo Pharma: Research Funding; Boehringer Ingelheim: Research Funding.