Introduction:

Surgeons want to counsel their patients accurately about the risks of rare complications. This is difficult for venous thromboembolism (VTE), as most studies report rates of asymptomatic disease, which may not be clinically relevant. Generic guidelines promote chemoprophylaxis in foot and ankle surgery despite a relative dearth of evidence. We therefore calculated the rate of confirmed, symptomatic deep vein thrombosis and pulmonary embolism, after surgery or trauma to the foot or ankle, in our hospital.

Surgeons want to counsel their patients accurately about the risks of rare complications. This is difficult for venous thromboembolism (VTE), as most studies report rates of asymptomatic disease. We calculated the rate of confirmed, symptomatic deep vein thrombosis (DVT) and pulmonary embolism (PE) after elective lower limb surgery in our hospital.

We scrutinised referrals for venous Doppler ultrasound and computed tomography pulmonary angiography, identifying all cases of DVT or PE over an 18 month period. These patients were cross-referenced with our elective orthopaedic database and Healthcare Resource Group coding data.

Out of 1071 total hip replacements, there were three DVTs and two PEs, giving an incidence of 0.28% for DVT and 0.19% for PE. Out of 1351 total knee replacements, there were four DVTs and three PEs (incidence 0.29% and 0.22% respectively). Out of 1988 non-arthroplasty hip and knee procedures, there were no DVTs and two PEs (incidence 0.1%). For 1763 elective surgical foot procedures, there were five DVTs (incidence 0.28%), and no PEs.

Currently, Rivaroxaban is offered to patients undergoing hip and knee replacement surgery, but other patients do not receive anticoagulants routinely. Our low incidence of VTE supports this policy and is reassuring for surgeons.

Since NICE issued guidance on preventing venous thromboembolism (VTE), the use of chemoprophylaxis has increased dramatically in trauma and orthopaedics. However, enthusiasm is tempered by a lack of data regarding the true incidence of VTE in everyday practice. We investigated the epidemiology of VTE among ambulatory patients with lower limb injuries within our Trust.

We identified all patients who suffered pulmonary embolism (PE) or deep vein thrombosis (DVT) over an 18 month period, and cross-referenced them with our trauma database. All lower limb injuries were included, whether operated or not. Hip fractures routinely receive dalteparin and were excluded.

There were 11,594 new attendances or post-operative attendances in all fracture clinics over 18 months. Of these, 4530 had lower limb injuries and were immobilised. There were 21 DVTs and 7 PEs in these patients, an incidence of 0.43% and 0.14% respectively. Of note, three DVTs were in patients with Achilles tendon rupture.

The incidence of symptomatic VTE is low in a population of ambulant patients with lower limb injuries in casts, without chemical thromboprophylaxis. Prophylaxis for VTE would thus have a large number needed to treat. The costs and complications of chemoprophylaxis should also be considered before it is introduced universally.

Rivaroxaban, an oral direct factor Xa inhibitor was introduced for thromboprophylaxis at the Royal Cornwall Hospital for hip and knee arthroplasty surgery in October 2009.

Our aim was to investigate how safely Rivaroxaban could be implemented and how quickly its regular use was established.

We identified 140 patients from theatre logbooks who underwent elective total hip and knee joint replacements between October 2009 and March 2010. Patient notes, computer and DVT clinic records data were collected to determine the uptake of the new drug and the incidence of post-operative complications. We compared our chemical thromboprophylactic rates to those recorded at discharge in a 4-month period prior to our study in 2009. In addition we quantified the time needed before a newly introduced drug becomes established in clinical practice.

Patients were divided into two groups. Those who received Rivaroxaban were in group A (n=78, 55.7%) and those who received alternative or no chemical thromboprophylaxis constituted group B (n=62, 44.3%). All patients were prescribed TEDs stockings.

10.3% [8/78] of patients in group A suffered wound complications compared with 6.6% [4/62] of group B patients. Within group A we found that 41.1% (n=7) of the documented wound complications were wound ooze. DVTs occurred in both groups, 1 in group A and 2 in group B. 4 patients had postoperative haematemesis, 2 in each group.

Group A had 17 (22%) documented complications. A similar number (n=15, 24%) of patients in group B had recorded complications.

Our complication rates compared favourably to the RECORD 1–3 pooled study.

From January to April 2009, prior to introduction of Rivaroxaban, 51% of all elective hip and knee replacement surgery patients were receiving any chemical thromboprophylaxis on discharge. This increased to 83% following introduction of Rivaroxaban. During the first month of introduction of Rivaroxaban at our hospital, following NICE guideline, 28% of patients who qualified to receive the drug did. This improved to 95% by the time it had been in use by 3 months.

The data shows that there is no statistical significance in complications in thromboprophylaxis in elective total hip and knee replacement surgery between Group A and Group B (P-value 0.8941). This shows similar complication rates to the RECORD clinical study and concludes a safe introduction of the drug to our District General Hospital.

Patients in Group A had a reduced occurrence of thrombotic events, but an increase in cases of wound ooze when compared to group B.

Following the introduction of Rivaroxaban, it took 3 months for 95% of eligible patients for the drug to be NICE compliant. This demonstrated a 3 month run in time for the implementation of this new treatment regime in our hospital.

Rivaroxaban was introduced for thromboprophylaxis at the Royal Cornwall Hospital for hip and knee arthroplasty surgery in October 2009. We identified 140 patients from theatre logbooks who underwent elective joint replacement between October 2009 and March 2010. Patient notes, computer and DVT clinic records and WebPacs data were collected to determine the uptake of the new drug and the incidence of wound problems, DVTs and any other post-operative complications. In our sample 55.7% [78/140] patients received rivaroxaban. 10.3% [8/78] of patients on rivaroxaban suffered wound complications compared with 6.6% [4/62] of patients on alternative anticoagulation. Three patients suffered DVT's, 1 of whom was taking rivaroxaban. There were a further 6 patients, 4 on rivaroxaban, with leg swelling severe enough to merit investigation, all of whom had negative doppler scans. Bleeding events included 4 patients with postoperative haematemesis of which 2 were taking rivaroxaban. Five patients, all under different surgical operators of which 3 had taken rivaroxaban, developed stiff total knee replacements and were offered MUA or physiotherapy.

We carried out a comparison of the analgesic requirements, length of stay in hospital, complications and cost effectiveness of patients who had either a continuous disposable infusion pump or standard treatment for unicompartmental knee replacement.

This study began as an audit. We completed the audit loop with a prospective study after implementing our recommendations. The device is a single use disposable elastometric pump, set immediately after surgery to deliver a continuous flow of 0.5% bupivacaine at a rate of 2mls per hour for 48 hrs into the knee joint via a fine catheter with a fenestrated tip.

The case notes of all patients in the study were analysed to establish their total analgesic requirements for 48hrs following surgery. Complications and length of stay in hospital were recorded This has changed our clinical practice. Results of our study of 50 patients, 25 in each group (age and sex matched) confirmed that there was a significant reduction in opiate requirements after introduction of the pain pump. Length of stay in hospital was reduced from 5.7 to 3.9 days. Patients' requirements for NSAIDs were reduced. There were no significant complications in the pain pump group. The costs of the pump (£40) were offset by the reduction in analgesic requirements and shortened length of stay in hospital.

We conclude this form of analgesia is safe, effective and cost effective. Our department has now changed to using the pump for all unicompartmental knee replacements.