Deep vein thrombosis (DVT) and pulmonary embolism (PE), both clinical expressions of the same entity, venous thromboembolism (VTE), are major surgery-related complications.

The incidence of DVT exceeds 20% in patients undergoing major surgery, and may even reach 40% in cases of major orthopedic surgery.1 The risk of post-operative PE may reach 5% in patients with a high thrombotic risk, and is associated with 1%–10% of hospital deaths.2 It is the most common cause of preventable mortality in hospital.

Numerous studies have described the prevalence of VTE in patients undergoing major surgery. However, little is known to date about its prevalence in the thoracic surgery post-operative period, and few authors have studied this problem.3 The aim of this study was to define the prevalence (overall and by risk groups) of thromboembolic complications in patients undergoing elective thoracic surgery, to quantify the risk of PE in pneumonectomy compared to patients undergoing lobectomy, and to characterize the risk of suffering a VTE.

Between January 1994 and November 2011, 6004 patients underwent elective thoracic surgical procedures under general anesthesia. The demographic and preoperative data for the study group are documented in Table 2. Lobectomy was performed in 1378 cases and pneumonectomy in 305 patients. Among the remaining surgeries, 118 bilobectomies, 59 typical and 1238 atypical segmentectomies, 289 lung biopsies, 137 chest wall resections, 74 thymectomies and 168 thyroidectomies were performed, among others. Among the diagnoses, lung cancer was the most common (2245 cases); 522 cases of lung metastases, 442 cases of spontaneous pneumothorax, 321 cases of pulmonary nodule, 211 cases of pleural effusion, 178 cases of interstitial lung disease, 161 cases of goiter, 149 cases of tracheal stenosis, 115 cases of sarcoidosis, 108 cases of chest wall tumors and 92 cases of thymomas were recorded, among others. The most common risk factors presented by patients in our series were age over 60 years (54%), obesity (18.1%) and history of cancer (68.7%).

All surgical procedures generate a state of venous stasis and hypercoagulability, as well as a certain level of endothelial damage, the three factors that bring together Virchow's triad and which favor the development of VTE in these patients. VTE is a major complication that affects hospitalized patients and which is closely related with surgery.

Various studies have described its prevalence in patients undergoing specific chest procedures. Thus, in a study by Ziomek et al.5 in which 77 patients who underwent lung resection surgery for lung cancer were analyzed, 20 patients (26%) had a thromboembolic episode in the postoperative period; there were 15 cases of DVT and 5 cases of PE. According to other authors, the mortality derived from these thromboembolic complications varies between 15% and 20%.6 Kalweit et al.7 analyzed 1735 patients who underwent lung resection surgery. In their series, PE was the second most common cause of death during the postoperative period, accounting for 15.2% of deaths post-resection. The high incidence of thromboembolic complications observed in these series could be due to a lack of standard protocols for heparin thromboprophylaxis, as happened previously in the 1990s. In a more recent study, Dentali et al.8 described, in a series of 693 patients who underwent thoracotomy for lung cancer under an antithrombotic prophylaxis protocol, an incidence of VTE and PE of 1.7% and 1.3%, respectively, with 0.6% mortality attributed to PE. These figures are more similar to those obtained in our series, in which eight patients of the 1749 who underwent lung resection surgery for lung cancer had thromboembolic complications (0.005%), with a mortality due to this cause of 0.002%.

The incidence of VTE in surgical patients without prophylaxis may reach 80% in patients with a very high thrombotic risk.4 However, although most patients undergoing elective thoracic surgery procedures have a high or very high thrombotic risk due to their characteristics (80.8%), in our study the prevalence of VTE was 0.18%. This low prevalence may be attributable to the use of thromboprophylaxis protocols, which include the use of anticoagulant drugs and mechanical measures (intermittent compression systems and elastic stockings) and early ambulation, following international recommendation guidelines.9 Taking these data into account, it is vitally important to perform an individualized assessment of each patient, determining their thrombotic risk in order to apply the best thromboprophylaxis possible based on the latest recommendations in the guidelines.

The postoperative period in thoracic surgery represents an increased risk of VTE, both due to the advanced age of the patients and the high frequency of neoplastic diseases as a reason for the intervention. Although the precise prevalence of VTE is unknown, we do know that the incidence of DVT in people aged 65–69 years is 1.8 cases/1000 persons/year and increases to 3.1 cases/1000 persons/year between 85 and 89 years.10 In our series, the mean age of the patients who presented VTE was 65.95 years, considerably older than the group of patients who did not have this complication (57.53 years).

The presence of neoplasm increased the risk of VTE between six- and sevenfold, and the development of PE threefold.11,12 According to a recent systematic review in which 26 studies were assessed, the prevalence of PE in patients with lung cancer was 3.6%.13 In our series, 90.9% of patients who had VTE were diagnosed with neoplasm, either primary or metastatic, which was the reason for the surgery.

Among patients who undergo surgical resection due to oncological processes, VTE is considered a major cause of mortality and may serve as an important predictor of survival. According to De Martino et al.,14 the variation in the risk of VTE among patients undergoing surgical resection not only depends on the type of cancer, but is also associated with the magnitude of the surgery necessary for the resection. In the case of surgery due to lung cancer, they reported an incidence postoperatively or within the first 30 days after surgery of DVT, PE and VTE of 1.3%, 1.3%, and 2.3%, respectively. Interestingly, the hospital stay (>1 week) is established as one of the major risk factors associated with VTE. Lyman et al.15 established that the risk of having a fatal PE after surgery for lung cancer is three times higher than in patients undergoing the same surgery for benign causes.

Pneumonectomy is associated with a higher rate of complications and mortality than any other thoracic surgery procedure. Patients undergoing pneumonectomy have a higher risk of VTE than patients undergoing lobectomy for stage I and II lung cancer, due to greater activation of coagulation, which appears from the seventh postoperative day.16 These data are consistent with those observed in our series, where patients who had VTE had undergone pneumonectomy (5 cases), the most common surgical procedure, three of whom died during the postoperative period as a result of a PE (all had undergone right pneumonectomy). There are several studies that have determined the prevalence and mortality associated with thromboembolic episodes in patients undergoing pneumonectomy due to lung cancer. Mason et al.17 analyzed 336 patients, of whom 25 presented a thromboembolic episode after the pneumonectomy (17 cases of DVT, 5 cases of PE and 3 cases of DVT+PE), with a maximum peak incidence on the seventh day postoperatively. With respect to mortality, Weder et al.,18 in their series of 176 patients who underwent pneumonectomy due to lung cancer after neoadjuvant chemotherapy, described a mortality of 3%; half were attributed to pulmonary thromboembolic complications. However, in our series, the prevalence of thromboembolic complications in patients who underwent pneumonectomy was lower than that described by other authors: 0.32% for DVT and 1.3% for PE. This low prevalence may be explained by the use of anti-thrombotic prophylactic protocols with anticoagulant drugs, and by the use of mechanical measures employed comprehensively and routinely in our hospital.

Seven of the patients who presented VTE in our series developed the complication while in hospital. Therefore, it is important to carry out close clinical follow-up with a high level of suspicion for early diagnosis of these complications. In four of the cases with VTE, the complication appeared after discharge. Thus, in patients with a high thrombotic risk, such as neoplastic patients in whom oncological surgery is performed, prophylaxis is recommended for 2–3 weeks after discharge.4 According to the results of two randomized studies, prolonging the duration of anticoagulation treatment beyond four weeks after surgery reduces the risk of VTE.19,20

There are more than 20 international guidelines that recommend VTE prophylaxis in high risk patients. The recent Antithrombotic Therapy and Prevention of Thrombosis guidelines from the American College of Chest Physicians (9th edition), recommends the use of LMWH, unfractionated heparin (UFH) or intermittent pneumatic pressure stockings in patients with moderate thrombotic risk. For patients with high thrombotic risk and low bleeding risk, it recommends the use of LMWH, UFH and mechanical measures, with either elastic stockings or intermittent pneumatic pressure stockings. In patients with a high bleeding risk, the use of mechanical measures is recommended, preferably intermittent compression systems, until the risk of bleeding decreases and pharmacological prophylaxis can be started. In our series of patients who underwent elective thoracic surgery, routine follow-up of these anti-thrombotic prophylactic protocols based on a thrombotic risk stratification has been shown to be highly efficient, with a low prevalence of VTE (0.18%) and excellent results.

This study has some significant limitations. Firstly, all the techniques for establishing the diagnosis of VTE were not used in all patients with radiological changes and postoperative dyspnea, so the real prevalence of VTE may be different to that described in our results. Furthermore, this is a retrospective study, so it is impossible to establish causality between the different thoracic surgery procedures and thromboembolic complications. Although data collection was prospective, there could be bias in the collection of risk factors in cases that presented the complication compared to those that did not. An active search for thromboembolic complications was carried out in all patients in the series, although autopsies were not performed on all patients, so PE could be the unsuspected cause of death in any patient who had another type of complication. Nevertheless, the results obtained in our series could be used to conduct prospective cohort studies designed to validate the risk of VTE in patients undergoing thoracic surgery procedures.

Patients undergoing elective thoracic surgery have a low prevalence of VTE (0.18%), attributable to the use of anti-thrombotic prophylaxis protocols based on proper stratification of thrombotic risk. However, in patients undergoing lung resection, especially in pneumonectomies, the risk is not inconsiderable and this complication should be taken into account as a priority suspected diagnosis in patients with acute respiratory episodes. The figures may be used as reference values, as they have been extracted from a large series with prospective records and uniform diagnostic criteria.

The authors declare that they have not had any funding to conduct this study.

The authors declare that there are no conflicts of interests of any type in relation to the publication of this document.