Perioperative cerebral infarction is a critical complication, and its incidence is dependent on the specific type of surgery performed [4, 5]; incidence of ischemic stroke and the in-hospital mortality rate associated with perioperative stroke after lobectomy or segmentectomy has been reported to be 0.6% and 32.6%, respectively . Since the first case was reported in 1989, several cases of cerebral infarction after VATS-LUL have been reported [6,7,8]. Recent studies have reported the incidence of PVT and the resulting cerebral infarction after VATS-LUL as 6.7–17.9% [3, 9, 10] and 1.9–4.6% [1, 3], respectively, and revealed that LUL was an independent risk factor of perioperative stroke . Its etiology of PVT has not been completely solved; several mechanisms of thrombus formation, such as the PV stump after VATS-LUL was longer than the other PV stumps and lower bloodstream velocity and more turbulent flow in the dissected PV stump than that in other PV stumps, were suggested [3, 9, 11]. Endothelial injury or surgical staples also can play an important role in thrombogenesis in the PV stump after lung resection . A recent study showed that central vascular ligation of PV stump may reduce intravascular exposure of the staple and contribute to a short PV stump and then contribute to reduce incidence of PVT .
Optimal antithrombotic therapy and preventive approaches for PVT remains unestablished, it should be noted that early recognition of PVT and anticoagulant therapy is important to prevent massive cerebral embolism . The induction time and the duration of anti-coagulation therapy are difficult to determine because PVT can occur even after the next day of discontinuation of anticoagulant therapy [13, 14] and can be a possible cause of cerebral embolism anytime after lung resection. In some institutions, they had implemented routine postoperative anti-coagulation therapy in patients following VATS-LUL. Along with those protocols, they changed the perioperative pain management strategy from epidural anesthesia to intercostal nerve block to prevent epidural hematoma in those patients .
The incidence of cerebral infarction after VATS-LUL (1.9–4.6%) is similar to the incidence of stroke caused by atrial fibrillation (approximately 5% 2-year age-adjusted incidence) . Lifelong anti-coagulation therapy has been shown to reduce the risk of atrial fibrillation ; therefore, it may be a promising option for preventing cerebral infarction after VATS-LUL as long as the risk of bleeding is acceptable.
We report three patients who underwent VATS for left upper lobe and developed thromboembolic complications. The patients in case 1 and case 2 developed life-threatening massive cerebral infarction without obvious cues of PVT. They had no clinical manifestation of perioperative atrial fibrillation, which is one of the major causes of stroke. Clinicians should consider PVT when patients present with cryptogenic stroke or systemic emboli after VATS-LUL [1, 2]. An early definitive diagnosis is critical to rescue the patient and to prevent severe complications. The patient in case 3 had asymptomatic PVT confirmed by CECT. We experienced case 1 and case 2 within a month, which made us change the follow-up strategy to evaluate PVT after VATS-LUL. This helped to detect PVT and prevent the stroke in case 3. Most cases of PVT are detected unexpectedly at postoperative follow-ups  while searching for the cause of nonspecific symptoms using CECT  or due to a high level of plasma D-dimer . Thus, PVT is overlooked and its incidence is underestimated . Our hospital performed 215 VATS pneumonectomy/lobectomy/segmental lung resection from 2015 to 2019; the number of VATS including left upper lobe was 30 LUL, 13 segmentectomy, and 1 left pneumonectomy. The 5-year incidence rate of cerebral infarction was 4.5% and that of clinically confirmed PVT was 2.3%, respectively. These data revealed that incidence rate of thromboembolic complications after VATS for the left upper lobe in our hospital was equivalent to those of previously reported. Although clinical evidence that shows its efficacy is limited, we currently perform CECT on POD 4 after VATS-LUL to evaluate the presence of PVT because thromboembolic complications would be critical to patients’ prognosis. According to the trends in perioperative major adverse cerebrovascular event associated with non-cardiac surgery, the incidence of ischemic stroke has increased over recent years . LUL is a new risk factor of perioperative stroke, and it is important to diagnose and treat PVT after VATS-LUL.
In conclusion, we reported three cases of postoperative PVT and cerebral infarction after VATS for left upper lobe. The diagnosis and treatment of PVT after VATS-LUL remain challenging. An evidence-based guideline for the detection and management of PVT after VATS-LUL is urgently needed.