Incidence of deep vein thrombosis and pulmonary embolism in Asian patients after direct anterior total hip arthroplasty

Abstract

Objective

Venous thromboembolism (VTE), comprising either deep vein thrombosis (DVT) and pulmonary embolism (PE), is one of the most significant complications after Total Hip Arthroplasty (THA). The incidence of VTE after Direct Anterior Approach (DAA) hip arthroplasty is not known. Preceding studies have shown that internal femoral rotation used in the posterior approach leads to “scissoring” of the femoral vein and complete occlusion of flow. In contrast, external femoral rotation, utilized in an anterior approach, does not lead to any limitation in venous flow. We hypothesize that the rate of VTE post-operatively after DAA THA is comparable, if not lower, to other conventional approaches.

Methods

Retrospective analysis of medical records was examined to identify the incidence of symptomatic DVT within 6 months post-surgery, yielding 196 cases of elective primary DAA after THAs between 2015 and 2019. Aspirin was administered post-operatively. Corresponding imaging was performed for cases with high clinical suspicion of either DVT or PE based on Well's criteria.

Results

There were no cases of symptomatic DVT or fatal PE within 1 year of THA. There was 1 case of non-fatal subsegmental pulmonary embolism, treated with warfarin, which made an uneventful recovery. The overall incidence of symptomatic DVT and PE after DAA THA was 0% and 0.5% respectively in our study population.

Conclusion

Cumulative rate of symptomatic DVT and PE was less than 1%, demonstrating that incidence of VTE after DAA THA is low. DAA is a viable approach in Asian patients undergoing THA, in view of lower incidence of post-operative VTE compared values cited in other conventional surgical approaches. Further large scale studies involving use of routine post-operative venography for assessment will further validate our results.

1. Introduction

Venous thromboembolism (VTE), comprising either deep vein thrombosis (DVT) and pulmonary embolism (PE), is one of the most significant complications after Total Hip Arthroplasty (THA).¹

The most common surgical approaches in THA are the posterior and lateral approaches, followed by the direct anterior approach (DAA).2, 3, 4, 5 There has been increased adoption of DAA as the defacto procedure for hip replacement surgery owing to its ability to avoid the need for muscle splitting, which runs a risk of partial denervation and detachment of the tendons as well as leading to shorter rehabilitation times.⁶ Despite the fact that there have been some studies evaluating incidence of DVT post THA, the operative approach is usually either unspecified or comprises of one of the more common approaches. To date, no study has evaluated the incidence of VTE post DAA and THA. We therefore aimed to evaluate the incidence of post-operative VTE in our Asian cohort. We hypothesized that patients receiving DAA after THA would develop less DVT as compared to other traditional approaches (e.g. posterior, direct lateral and transtrochanteric) due to its reduced risk of femoral vein occlusion during hip arthroplasty.3, 4, 7

2. Materials and methods

2.1. Demographics

The study was approved by the local Institutional Review Board and informed consent was obtained from all patients. Retrospective analysis of medical records was examined to identify the incidence of DVT within 6 months post-surgery, yielding patients who underwent elective primary DAA after THAs between 2015 and 2019. A total of 196 hips were identified after excluding cases of infection and prolonged immobilization.

There were 71 men and 125 women, with a mean age of 63.3 years and standard deviation of 10.9 years. Mean BMI was noted to be 25.8 kg/m2 with standard deviation of 5.3. The most prevalent diagnosis for THA was osteoarthritis of the hip, followed by avascular necrosis and dysplasia. The mean follows up duration for all patients exceeded 1 years.

2.2. Surgical methods

DAA THA for all patients included in this study was performed by a single fellowship trained surgeon with expertise in THA, in a high-volume tertiary hospital between 2015 and 2019. The patient was first positioned supine on the table. Both legs were cleaned and draped to facilitate limb length assessment. The incision for the DAA was marked from anterior superior iliac spine extending down obliquely pointing laterally towards the aspect of the patella. The tensor fascia lata iwas identified by the presence of its bluish tinge. The fascia over the tensor fascia lata was then incised and the TFL muscle was bluntly freed from its intermuscular septum which separates it from the rectus femoris. The ascending branches of the lateral femoral circumflex artery were identified and cauterized.

The femoral neck was visualized although partly covered by the reflected head of the rectus femoris. Retractors were placed on either side of the femoral neck and over the lateral aspect proximal femur. The capsule was further exposed by placing a retractor over the anterior acetabulum, retracting the rectus femoris medially. The capsule was excised, and the femoral neck was cut with a saw. Once the femoral head was removed, the acetabulum was exposed. Reaming was performed using a reamer 2 sizes smaller than the templated cup size. The acetabular reaming and cup placement were done with the assistance of an image intensifier. 1 or 2 screws are inserted to give a better fixation. The femoral preparation was next performed by initially releasing the posterior capsular in order for appropriate retractor placement to take place as well as delivering the proximal femur out from under the acetabulum. The femur was externally rotated and adducted by passing the leg under the contralateral leg in a figure of 4 position. Femoral broaching was performed until a good fit is achieved, taking into account the templated size. A Metal ball head was then inserted, and the hip checked for stability, impingement and limb length discrepancy. The wound was then washed with diluted povidone iodine and a cocktail of injections are given together with tranexamic acid into surrounding muscle. Post-operative aspirin was used for chemoprophylaxis.

2.3. Mode of detection of VTE

All patients were assessed daily for signs and symptoms of venous thromboembolism in an inpatient setting. The symptoms of venous thromboembolism that were assessed included calf swelling, redness, and pain. In patients with suspected DVT, a clinical assessment based on the Well's criteria probability scoring was performed. The Well's criteria is the most studied clinical prediction rule for the assessment of DVT. DVT was classified as unlikely if a Well's score of ≤1 is recorded while a Well's score ≥2 was observed, then DVT was estimated as likely to have occurred in the patient.^8,9

The following pre-test probability assessment criteria was undertaken in order to estimate if symptomatic DVT had occurred in the patient: paralysis, paresis or recent plaster immobilization of the lower extremities (1), patient has been bedridden for more than 3 days or has had major surgery within 12 days that has required general anesthesia (1), localized tenderness observed along the deep veins (1), entire swollen leg following DAA (1), calf swelling 3 cm > asymptomatic side (measured 10 cm below tibial tuberosity) (1), pitting edema which is confined to the symptomatic leg (1) and collateral superficial veins that are non-varicose (1) and a previous history of DVT (1).¹⁰ If the Well's probability was observed to be in the high scale, further imaging tests of the veins would be performed using the Doppler ultrasound method, which has been demonstrated to exhibit up to 97% sensitivity in detection of DVTs. An ultrasound was not performed for the detection of DVT for all patients given the fact that this study was performed in order to detect and treat symptomatic DVT and not asymptomatic DVT which is sometimes observed in patients.¹¹ It has been demonstrated in several studies that asymptomatic DVT does not require any major treatment interventions owing to fact that asymptomatic DVT is most likely to resolve itself and as such does not require any major medical intervention for resolution.¹²

All patients with post-operative doppler ultrasound conducted for possible DVT were reviewed to identify positive cases of DVT. In cases where post-operative DVT was identified, we proceeded to characterize the type (symptomatic, versus incidental), and location of DVT. Assessment of presence of concomitant pulmonary emboli was performed by screening through other relevant imaging such as CT pulmonary angiography (CTPA), 2D Echocardiography or MRI Scan (Magnetic Resonance Imaging).

In patients who presented with clinical suspicion of pulmonary embolism, such as chest pain, shortness of breath, and desaturation, CT pulmonary angiography was performed to detect the present of PE. Patients were discharged once they are able to achieve independent ambulatory status, usually on POD1 to 3. All patients were reviewed in clinic on POD 14 for signs and symptoms of venous thromboembolism and Doppler ultrasound performed for those with clinical suspicion. The clinical presentation of the symptoms of DVT such as asymmetrical swelling, warmth and pain in the extremity, when present frequently occurs 3–14 post-operative days due to prolonged immobilization which is most often required after DAA hence the decision to carry out a follow up post-operative day 14 to determine if patients presented with DVT.^13,14

3. Results

There were no patients with fatal PE within 1 year of THA. 0 cases of symptomatic DVT were detected. There was 1 case of non-fatal pulmonary embolism. The patient presented with shortness of [breath on post-operative day 7 and CT pulmonary angiography (CTPA) showed a small subsegmental acute PE in the posterior basal segment of the right lower lobe. The patient was started on warfarin by Hematology and made an uneventful recovery.

The overall incidence of symptomatic DVT and PE after DAA THA in this study was 0% and 0.5% respectively, which is comparably lower than traditional approaches to hip replacement therapy.

4. Discussion

This study is the first study to describe the incidence of venous thromboembolic disease in patients undergoing direct anterior approach hip arthroplasty. In this study, we demonstrated an extremely low post-operative incidence of less than 1% for symptomatic VTE (DVT and PE) in our Asian cohort.

Historically, the incidence of DVT after hip arthroplasty has been reported to be as high as 57%, based on a study by Stamatakis et al.¹⁵ With advancement in surgical technologies, and with adoption of chemoprophylaxis, it can be significantly reduced. This is backed by a study by Kim et al., involving a cohort of 150 Korean patients, where incidence was noted to be about 20% in the control group that was not given prior DVT prophylaxis while DVT was recorded in 12% of patients treated with aspirin.¹⁶ In the study by Kim et al., contrast venograms were performed prior to surgery and within a week after, increasing the detection rate of DVT. In a separate study. A similar study by Planes et al., where bilateral venography was performed between 12 and 15 days after total hip replacement in 745 consecutive patients, on a background of heparin prophylaxis found DVT incidence to be about 11%.

It is for these reasons that our protocol for thromboembolic disease prophylaxis included aspirin postoperatively, with use of anti-thromboembolic disease stockings as well as early ambulation. However, in this study, we focused on detection of clinically symptomatic VTE. This was done to maintain a cost-effective approach to treatment.

A similar preceding study by Kang et al., studying VTE incidence in an evaluation 861 patients who underwent 992 primary elective THAs demonstrated that the incidences of symptomatic DVT, symptomatic PE and fatal PE were noted to be 0.8%, 0.1% and 0% respectively.¹⁷ However, majority of cases in this study utilized the posterolateral approach. While DVT after conventional hip arthroplasty is well described, there are no studies that purely describe its incidence after direct anterior approach THA, and our study seeks to fill this gap, and shows that incidence of symptomatic VTE after DAA and THA is comparable to preceding studies covering other approaches. Through objective comparison in numbers, we infer that DAA is a non-inferior approach, as opposed to other conventional approaches of THA.

The pathophysiology behind DVT can be described using Virchow's triad, which encompasses the following three factors: hypercoagulability, stasis and endothelial injury.¹⁸ Specific to hip arthroplasty, femoral vein occlusion and damage to the femoral vein resulting from manipulation of the leg is a known risk factor.¹⁵

We propose that adopting the direct anterior approach confers benefit of reduced VTE risk over other conventional approaches, due to the nature of the operation. In a study by Stamatakis et al. studying intraoperative venography in patients undergoing THA showed severe distortion of the common femoral vein in all cases and observed almost total occlusion. Flow studies, as described by Malcolm et al.,⁷ concurs with this theory, showing that moderate degrees of flexion and adduction, together with dislocation of the hip, leads to obstruction of venous flow distal to the femoral sheath. It was also noted that internal femoral rotation utilized in the posterior approach caused “scissoring” of the femoral vein between the femur and pubic bone, predisposing patients to VTE. In contrast, external femoral rotation, utilized in an anterior approach, did not lead to occlusion of venous blood flow.⁷ In this aspect, it can be inferred that DAA may be more mechanically favorable as opposed to other approaches in reducing risk of post-operative DVT.

There exist several other differences between the DAA and the conventional total hip replacement methods such as posterior or lateral approaches which may confer risk reduction in terms of developing VTE. Several studies have reported that DAA showed superiority to the Posterior approach, due to faster rehabilitation times, less muscle pain and blood loss as well as less soft tissue damage due to the fact that DAA utilizes an intermuscular plane for the insertion of the hip replacement components.¹⁹ THA carried out using a conventional hip replacement lateral approach that involves muscle splitting which leads to greater pain for the patient in the early post-operative period, longer hospitalization and rehabilitation periods.²⁰ Preceding studies have also shown that incidence of DVT after orthopedic interventions is associated with prolonged immobilization. With DAA, patients may benefit from lower incidence of post-operative VTE due to shorter immobilization and faster periods of recovery.

Some previous Asian studies have recommended that it may not be cost effective to prescribe aggressive pre-operative prophylaxis, such as heparin, in patients undergoing THA who have no known risk factors for DVT and this recommendation appears consistent with our study considering the low rates of symptomatic VTE in our study cohort.¹ Current literature reports much lower incidence rates in the Asian population as compared to Western populations post-THA.^1,17,21,22 Despite wide variance in results, the general consensus is that incidence of DVT and PE is significantly lower in the Asian population as opposed to the western population with diet, genetics and other environmental factors accounting for the discrepancies in the incidence of DVT observed in these two populations.23, 24, 25 In light of the low incidence of DVT and PE, we agree that the classic recommendation for pharmacological treatment such as unfractionated heparin (UFH), low molecular weight heparin (LMWH) may not be most applicable for the Asian population.²⁶ There may, however, still be role for aspirin as described in Kim et al.'s study,¹⁶ when weighing potential risks and benefits. Important preventive measures in the form of mechanical approaches may also prove relevant. These includes the use of graduated compression stockings and intermittent pneumatic compression devices.

4.1. Limitation

One limitation in our study is that it did not take into account patients exhibiting asymptomatic venous thromboembolism. However, this was not feasible for our study, as our goal was to maintain a cost-effective approach to treatment and manage only clinically symptomatic cases. Another limitation of this study was the fact that an assessment of the incidence of DVT after DAA was conducted in a predominantly Asian population. Recommendations may not be the most applicable to a general population.

5. Conclusion

In conclusion, the incidence of venous thromboembolic events in patients undergoing direct anterior approach total hip replacement is low, presumably because of mechanical favorability (reduced kinking of the femoral vein and consequently, occlusion of blood flow) and faster recovery times. Further large-scale studies in other study populations are thus required to validate the findings from our study.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.