A prospective study including 2,138 hip and 540 knee operations in 2,678 patients undergoing TJA was conducted at our institute between September 2009 and December 2013. The study group included 1,697 females (63.3 %) and 981 males (36.7 %). The THA study group included 153 revisions (7 %), while the TKA group included 17 (3 %). All patients gave informed consent.

Patients with a previous history of DVT, chronic venous insufficiency, varicose veins, renal insufficiency, heart failure, or who were taking oral steroidal/hormonal/anticoagulant drugs for any medical condition, were excluded from the study.

Preoperative assessment for DVT was done in all patients on both lower limbs by color Doppler ultrasonography using an ATL 5000 HDI machine. Assessment included examination of bilateral common femoral, superficial femoral, popliteal, anterior tibial, and posterior tibial veins. Veins were assessed for flow, visualized thrombus, compressibility, and augmentation. A diagnosis of DVT was made in cases of visualization of thrombosis, absence of flow, lack of compressibility, or lack of augmentation. The venous segments examined were the same as for venography. Iliac veins were not visualized, and the deep femoral vein was visualized only at the junction site. Thigh veins were examined with the patient in a supine position, whereas popliteal and calf veins were examined with the patient in a seated position. Pulsed-wave Doppler and color Doppler modalities have been used for anatomic orientation and venous examination, but not for documentation of the venous findings. Documentation consisted of five video/digital sequences of each leg (thigh veins, popliteal veins, peroneal veins, posterior tibial veins, and anterior tibial veins) for approximately 60 s.

The postero-lateral approach was used on all patients undergoing THA. All patients undergoing TKA received a standard paramedical approach: a tourniquet was routinely used. Patients were assessed daily for any signs of DVT. All patients were given prophylaxis for DVT for 35 days postoperatively: a daily single dose of LMWH medication (Nadroparin, 0.4 mL) was given to 32 % of the patients, while a daily single dose of Fondaparinux was given to 78 % of the patients. Two hundred and thirty-four patients (9 %) were preoperatively taking warfarin because of a collateral pathology, so warfarin was suspended for 5 days preoperatively and replaced by a personalized dose of LMWH (“bridging anticoagulation”) (Lohr et al. 1995). Warfarin was resumed in the early postoperative period (48–72 h).

Nadroparin is a porcine-derived LMWH which accelerates the inactivation of factor II and factor Xa when bound to antithrombin III (ATIII). Nadroparin halts the coagulation pathway by inhibiting the activation of thrombin (factor IIa) by factor Xa. The amplification of the fibrin clotting cascade is stopped once factors Xa and IIa are inactivated.

Fondaparinux (Arixtra) is a synthetic pentasaccharide anticoagulant. Apart from the O-methyl group at the reducing end of the molecule, the identity and sequence of the five monomeric sugar units contained in Fondaparinux are identical to a sequence of five monomeric sugar units which can be isolated after either chemical or enzymatic cleavage of the polymeric glycosaminoglycan heparin and heparan sulfate (HS). This monomeric sequence in heparin and HS is thought to form the high-affinity binding site for the natural anticoagulant factor, antithrombin III (ATIII). Binding of heparin/HS to ATIII has been shown to increase the anticoagulant activity of antithrombin III 1,000-fold. Fondaparinux potentiates the neutralizing action of ATIII on activated Factor ×300-fold.

All patients used an intermittent pneumatic boot compression device in the first 24 h postoperatively, while being monitored in a sub-intensive care unit and then during the hospital stay in all non-deambulatory situations. As in the preoperative period, assessment for postoperative DVT was carried out by color Doppler ultrasonography on postoperative day 4. The same angiologist, experienced in color Doppler ultrasonography, repeated the study in all cases. All patients wore below-the-knee elastic compression stockings for 35 days postoperatively. Statistical analysis was performed using Student’s t-test.

The presence of DVT in the lower extremities following TJA is considered a potentially life-threatening situation (Lotke et al. 1994). Prevention of DVT using duplex ultrasonography to avoid the complication of pulmonary embolism has shown its efficacy (Grady-Benson et al. 1994). The fundamental aid from preoperative color Doppler evaluation is to identify the location of a silent DVT. In fact, proximal DVTs are well known to be closely associated with increased risk of pulmonary embolism and are conventionally treated more aggressively with closer monitoring.

On the other hand, the role of distal DVT is less obvious. It was once thought to be quite benign (Doouss 1976), but it has also been reported to be associated with pulmonary embolism, especially when there is propagation of a distal DVT to a more proximal location (Masuda et al. 1998). Recently, Grady-Benson et al. (1994) used serial venous Doppler flow measurements to document a propagation rate of 24 % in the calf. DVT formation during TJA could be related to the flexion position of the lower limb during the procedure, or to the use of a tourniquet on the thigh causing stasis or surgical trauma to the surrounding vasculature during the release.

Unfortunately, Duplex ultrasonography as a non-invasive screening tool has not yet received universal acceptance (Davidson et al. 1992), despite promising reports (Ko et al. 2003). This study shows that preoperative Doppler ultrasonography prevents any false positive cases and forewarns the surgeon of an increased risk of DVT and need for prophylaxis. The preoperative ultrasound screening demonstrated a DVT in 4.5 % of our patients: without screening they were at risk of developing a perioperative pulmonary embolism. Therefore, in our hands, it is beneficial to carry out preoperative Doppler ultrasonography to detect any pre-existing DVT, especially in patients who are considered at high risk as shown in Table 8.2. In fact, preoperative and postoperative clinical findings alone are generally considered poor predictors of the presence of DVT (Lieberman and Pellegrini 1999). Once a DVT episode has been detected, surveillance with duplex scanning is also mandatory in determining efficacy and duration of therapeutic anticoagulation for DVT (Grady-Benson et al. 1994).

Our study intentionally does not address the issues of the indication and timing of initiating anticoagulation therapy for patients undergoing TJA. Although the value of routine pharmacologic thromboprophylaxis in TJA has been questioned by a metanalysis (Murray et al. 1996), the majority of surgeons would still recommend some form of prophylaxis against DVT (Brookenthal et al. 2001) in consideration of the morbidity and mortality associated with postoperative DVT in THA and TKA. The authors of the current study recognize that venous thromboembolic events following primary hip and knee arthroplasty have decreased significantly over the past two decades, mainly because of new multidisciplinary approaches. Rapid postoperative mobilization, optimization of surgical techniques, and improved perioperative pain management (which includes the use of regional anesthesia) have all contributed to decreasing the DVT risk. At the author’s institution we use pharmacologic and mechanical approaches for thromboprophylaxis after TJA as suggested by the 2012 American College of Chest Physicians (ACCP) Evidence-Based Clinical Practice Guidelines (Gordon et al. 2012) and the 2009 Tuscany Region Protocol for Thromboprophylaxis in Orthopaedic Surgery (www.​regione.​toscana.​it). All patients undergoing a TJA procedure receive a form of pharmacologic thromboprophylaxis (LMWH or Fondaparinux) for 35 days postoperatively, use an intermittent pneumatic compression device during the hospital stay, and wear below-the-knee elastic compression stockings for 35 days postoperatively. The main difference from 2012 ACCP guidelines is that all our patients underwent Doppler ultrasonography screening preoperatively and postoperatively before hospital discharge. Doppler ultrasonography is historically a non-invasive procedure and provides good sensitivity (89 %) and specificity (100 %) for detecting DVT (Cronan et al. 1987). There is a lack of published data on the appropriateness of preoperative Doppler ultrasonography in patients undergoing TJA: Sisodia et al. (2013) recently reported 22.4 % preoperative asymptomatic DVTs in a group of patients awaiting TJA.

The current study supports the benefit of Doppler ultrasonography as a preoperative and postoperative investigation method in patients undergoing TJA.

Deep venous thrombosis (DVT), with thromboembolic complications which may lead to pulmonary embolism, is a serious disease and potentially fatal. This often complicates the clinical course of patients suffering with another disease, already hospitalized or no, but it also affects individuals in apparently good health.

The most important clinical objectives of an early and correct diagnosis and treatment are:

Venous thrombosis in the majority of cases involves the veins of the legs and, depending on the locations, is deep venous thrombosis if it involve the venous system subfascial and a superficial venous thrombosis if it involves overfascial veins (Fig. 8.1).

The incidence of pulmonary thromboembolism increases with age, aggressiveness of surgical procedures, and the length of hospital stay, and is highest in patients treated in intensive care because of the overlap of these risk factors and prolonged immobility.

It is estimated that only in approximately 30 % of ambulatory patients in whom a clinical suspicion of DVT has been made is actually confirmed by objective investigations.

In the absence of prophylaxis, the incidence of DVT in the hospital population varies between 10 % in medical patients and 40–60 % in patients undergoing orthopedic major surgery (Turpie et al. 2002; Geerts et al. 2008; Cohen et al. 2007; Heit et al. 2005; Blann and Lip 2006).

The clinical manifestations of DVT of the lower limbs are multiple (spontaneous pain, flushing, cyanosis, increased skin temperature, cramps, increase in the size of the limb, edema, development of collateral circulation, phlegmasia alba dolens, etc.). However, the clinical diagnosis of DVT is not accurate because it is based on symptoms and signs that, taken separately or together, are neither sensitive nor specific. Lack of a pathognomonic element requires that the diagnosis should be definitively confirmed by instrumental examination.