Venous thromboembolic complications with and without intermittent intraoperative and postoperative pneumatic compression in patients with glioblastoma multiforme using intraoperative magnetic resonance imaging. A retrospective study

Objective. - To evaluate the effectiveness of intraoperative and postoperative intermittent pneumatic compression (IPC) as a method used to decrease the incidence of deep venous thrombosis (DVT), in comparison to the standard use of graduated compression stockings, low-molecular weight heparin (LMWH) and physiotherapy during the hospital stay. All patients in this study underwent intracranial surgery for glioblastoma multiforme (GBM) using intraoperative magnetic resonance imaging (MRI) guidance. Patients and methods. - We performed a single center retrospective study of a cohort of 153 patients who underwent surgery for GBM aided by intraoperative MRI from October of 2009 to January of 2015 at the International Neuroscience Institute (INI), Hannover, Germany. Out of all patients, 75 in comparison to 78 were operated with and without the additional use of IPC, respectively. Both groups received graduated compression stockings, LMWH and physiotherapy postoperatively as a basic thromboprophylaxis. Postoperatively the patients were screened for DVT by Doppler ultrasonography of the limbs and pulmonary embolism (PE) by CT-scan of the chest. Results. - DVTs were found in 6 patients with IPC and in 3 patients without IPC. The incidence of developing DVTs was therefore not significantly increased with the application of IPC from 3.9% to 8% (P-value: 0.33). No statistically significant differences were found in the probability of occurrence of pulmonary embolism (PE) with a reduction from 2.6% to 1.3% (P-value: 0.59). Conclusion. - Our results demonstrate, that the surgical intervention and the subsequent patient immobilization, as well as the thromboprophylactic techniques used have a relatively low influence on the occurrence of thromboembolic complications than we expected. Our findings might be attributed to the overall low number of these complications in a glioblastoma multiforme patient population expected to be at a high risk for coagulopathy. In other words, in order to produce statistically significant results, we would need to increase the patient cohort. By doing so we may better detect a positive therapeutic effect. Alternatively, because of the multitude of possible complex risk-factors leading to coagulopathy in a glioblastoma patient population it might be the case that IPC has little or no effect and that there is a different underlying mechanism responsible for the observed coagulopathy. (C) 2018 Published by Elsevier Masson SAS.