Thromboembolic Disease after Foot and Ankle Surgery

INTRODUCTION

Thromboembolic disease is a complication after orthopaedic surgery and has been documented after hip and knee surgery. ⁴ , ⁵ The incidence of deep vein thrombosis (DVT) has been reported after hip and knee surgery in more than 60% of patients, with up to 13% of these patients developing pulmonary emboli (PE). ⁵ Thromboprophylaxis in this group of patients has been shown to reduce the risk of DVT and fatal PE. ⁵ A meta-analysis by Freedman et al. ⁴ determined the risk of DVT after selective total hip replacement to be between 39% and 74%, and the risk of fatal PE between 0.19% and 3.4%. However, the incidence of DVT in foot and ankle surgery has not been high as this. ⁶ , ⁷ Mizel et al. ⁶ reported a prospective, multicenter study in which 15 orthopaedic surgeons completed a questionnaire on every surgical episode involving the foot and ankle over a 1-year period. They found a 0.22% incidence of DVT and a 0.15% incidence of PE. The other study by Solis and Saxby ⁷ was a single-surgeon prospective study that used screening duplex ultrasound (at the first postoperative visit) to diagnose lower limb thrombi in all patients who had surgery about the foot and ankle during the study period. They found a 3.5% prevalence of DVT, all of which were clinically silent, and no PE. Both studies concluded that routine prophylaxis is not indicated in patients having foot or ankle surgery, because the complication from anticoagulation of patients was higher than thromboembolic events in patients who did not receive anticoagulation medication.

A review article ² concluded that more prospective studies are required but that based on the current literature patients with risk factors for DVT should receive prophylaxis before and after foot and ankle surgery. ²

MATERIALS AND METHODS

A retrospective analysis was conducted to identify the incidence of thromboembolic disease in the practices of two foot and ankle surgeons (LG and HKS). The data was analyzed with the purposes of identifying risk factors that may predispose patients having foot and ankle surgery to thromboembolic events and to determine, if possible, a subgroup of patients that should be treated prophylactically with anticoagulation.

The clinical records of 608 patients who had surgical procedures about the foot and ankle between January, 1999, and April, 2000, were reviewed. The patients were from two orthopaedic foot and ankle surgeons (LG and HKS) affiliated with teaching hospitals, and all patients who had surgery during the study period were included and followed in the clinic until discharged from care. Six patients seen during the study period were lost to followup and were not included in the study. The study group consisted of 316 men and 286 women for a total of 602 patients, with a mean age of 42.9 (range 7 to 91) years.

The study parameters included the presenting symptoms of the thromboembolic complications, medical history to identify potential risk factors of connective tissue disease, diabetes and previous DVT, as well as concurrent use of medications such as corticosteroids, nonsteroidal anti-inflammatories, hormone replacement therapy, and oral contraceptives. A history of recent air travel, 2 weeks before surgery, also was noted.

The site of surgery was grouped into ankle or foot procedures. Elective and trauma patients were included in the study. The use of a tourniquet was assessed as a risk factor as was the type of anesthetic used. The use of thromboprophylaxis was recorded, and postoperative plaster immobilization and weightbearing status were evaluated as risk factors.

All thromboembolic complications were diagnosed clinically at postoperative followup. A clinical diagnosis of DVT was made if the calf was more swollen than expected or if the patient complained of calf pain or demonstrated tenderness on palpation of the calf. DVT was confirmed with duplex ultrasound images and PE was confirmed with ventilationperfusion scans. No clinically diagnosed DVT had a negative Doppler. The patients in this study were not screened for silent DVT with ultrasound. The average followup was 4.4 (range 1 to 15) months.

RESULTS

Twenty-four (4%) patients developed a postoperative DVT. There were 10 men and 14 women with a mean age of 48.8 (range 20 to 74) years. Eight (1.3%) patients had a PE diagnosed by ventilation-perfusion lung scan.

The distribution of presenting symptoms of the thromboembolic complications included calf pain (15 of 24 patients), chest pain (five patients), lower leg swelling (two patients), and shortness of breath with hemoptysis (two patients).

Connective tissue diseases, such as rheumatoid arthritis, systemic lupus erythematosis, and scleroderma were the most frequent comorbidities in these 24 patients followed by a history of DVT and diabetes. A history of rheumatoid arthritis and previous DVT or PE were significant (p = 0.04 and 0.02, respectively) in the development of thromboembolic complications. Three patients had a history of recent air travel.

A tourniquet was used in over 90% of patients (23 of 24). The tourniquet was applied to the thigh in each case. An average pressure of 286 (range 250 to 300) mmHg and a mean duration of 62 (range 5 to 120) minutes were recorded. The odds ratio for use of the tourniquet was 1.79, it was not a statistically significant predictor for thromboembolic complications in this patient population (p = 0.87).

Most patients (572 or 95%) in the study group received a general anesthetic, with spinal and regional blocks used for the remainder. The type of anesthetic was not an independent predictor for thromboembolic complications but the sample size for regional and local anesthetic was very small, making interpretation difficult.

Patients who received prophylaxis for DVT were those deemed to be at risk: those with a history of DVT or malignancy, those using oral contraceptives, and obese patients. Low-molecular weight heparin was used for perioperative prophylaxis and was given on the day of surgery as a single dose for day-only cases or was continued until discharge from hospital. The average duration of prophylaxis was 2.8 days (range 1 to 3 days). A daily 40-mg subcutaneous dose was used. One hundred and eighty-five patients (31%) received prophylaxis. Univariate analysis showed that prophylaxis was a risk factor for the development of thromboembolic complications, which is an unexpected finding given the literature which clearly shows that prophylaxis is beneficial. ⁴ No patient developed a wound hematoma or postoperative bleeding as a result of prophylaxis.

Weightbearing status alone was not an independent predictor, although partial weightbearing was statistically significant (p = 0.003) as a risk factor for DVT formation. Multivariate logistic regression analysis (MLRA) of the data showed that a recent history of air travel and the use of a short-leg cast were independent predictors of DVT and PE formation in this patient group.

DISCUSSION

We found a 4% clinical prevalence of thromboembolic complications after foot and ankle surgery, which is somewhat higher than previously reported. ⁶ The current study reviewed the same parameters as those by Mizel et al.; ⁶ however, they also performed a cost analysis for a 6-week course of low-dose adjusted warfarin for each patient. They calculated a cost of $165 per patient and a complication of prophylaxis rate of 1.5% to 4%. On the basis of these figures they concluded that routine prophylaxis was not justified.

Clinical diagnosis of DVT often is difficult and equivocal. Patients presenting with symptoms or signs of thromboembolic complications were assessed clinically by the treating surgeon and investigated, as was done in the study of Mizel et al. ⁶ As shown by Solis and Saxby, ⁷ screening of all patients may be the only way to accurately ascertain the true incidence of DVT after foot and ankle surgery because all the patients in their study had silent thrombi.

We found no association between thromboembolic disease and medication used, the type of surgery, the type of anesthetic, or the use of a tourniquet. There are no current studies in the literature that demonstrate a decreased incidence of DVT when using an ankle block. The prospective study by Solis and Saxby ⁷ found tourniquet time, hindfoot surgery, and advancing age to be associated with DVT formation. The surgeons involved in our study use a thigh tourniquet that allows greater exposure for harvesting bone graft from the proximal tibia, as well as, allowing both accuracy of alignment and greater limb maneuverability.

A positive association (MLRA) was found in patients with a recent history of air travel; however, no Doppler examinations were done on these patients before surgery. Recent studies have failed to demonstrate a statistical link between air travel and DVT formation in otherwise healthy individuals. ¹ , ³ Despite these studies, Doppler examination may be prudent in patients requiring foot and ankle surgery after air travel to ascertain the presence of a thrombus preoperatively and start appropriate treatment before surgery to decrease the risk of subsequent PE.

An interesting relationship found was an inverse relationship between thromboembolic disease and prophylaxis with low-molecular-weight heparin. This result could erroneously be interpreted as showing prophylaxis as being a risk factor for DVT. This finding may suggest, however, that the at-risk group for DVT was accurately predicted but that the prophylactic regimen was suboptimal. This risk group may benefit from a home course of self-administered low-molecular-weight heparin, which is cost-effective and easy to perform.