This is the first study to measure the distance from the tumor to the dissection margin using preoperative contrast-enhanced CT. Based on our investigation, it was considered that the distance from the tumor to the dissection margin should be at least 20 mm. However, Case No. 6 had a distance of 25 mm with a positive dissected stump [12]. Since PAS propagates and proliferates in the intima of the PA, it is difficult to determine how far it has progressed by contrast-enhanced CT and PET-CT. In terms of pathological findings, the PAS extension site is 0.01-0.02 mm for the normal PA intima, whereas the PA intima for PAS is 0.1-0.2 mm, which is not as thick (Figures 2A and B). Siordia et al. reported that primary PAS is better treated with pneumonectomy than with PE, which is better suited for palliative treatment [13]. With regard to the pathological findings, it is difficult to make a macroscopic judgment during surgery, and we agree with the opinions of Siordia et al. Extended PAS emergency surgery is often performed to save lives to remove a tumor that is originally symptomatic and has grown. However, since there are reports of patients living for more than 5 years due to extended surgery [14], one indicator would be that the distance between the tumor and the stump during surgery is 20 mm.
We also reported that cardiovascular and lung surgeries are performed on different days for PAS surgery. The advantage is that in the case of one-stage surgery, a surgical heart-lung machine tends to cause bleeding, so a two-stage helps ensure a clear surgical field of view. In addition, lung surgery is easier for general thoracic surgeons than surgery with a median sternotomy, by performing a posterior lateral incision. The disadvantage is that it is complicated because the surgery is performed in two stages. For Case No. 10, we attempted to perform the operation in two stages, but on the night of 0 POD of cardiovascular surgery, the bleeding from the trachea did not stop and the operation was accelerated. We believe that two-stage surgery is more effective, for example, when there is a high probability of adhesions in the chest cavity.
Kim et al. reported that early detection contributes significantly to prognosis [15]. For every doubling of time from symptom onset to diagnosis, the odds of death increased by 46% [16]. However, this tumor is often misdiagnosed as acute or chronic pulmonary embolism because it is characterized by luminal obstruction and intraluminal growth. Gan et al. reported that the wall eclipsing sign on PA-CT angiography is pathognomonic for PAS [17]. Endovascular catheter biopsy and endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) may be used to diagnose PAS [18–20]. However, EBUS-TBNA increases the risk of bleeding and causes massive hemoptysis. PET-CT has been previously reported to be helpful in diagnostic workup [21, 22]. Ito et al. reported that the median maximum standardized uptake value of FDG is 7.63 in PAS and 2.36 in pulmonary embolism [23]. In our case, PET-CT was performed in four cases, with three positive cases. However, Case No. 8 was suspected of having pulmonary embolism because of a negative PET-CT, and was initially treated with heparin, but chest CT after 2 weeks showed no change in the PA intravascular nodule. Therefore, we suspected PAS and performed the surgery. PAS should be suspected if there is a large intravascular filling defect and no clinical improvement on anticoagulant therapy.
Hoarseness was observed in 40% (4/10) of the patients as a minor complication after surgery. The incidence of postoperative recurrent laryngeal nerve palsy is 0.077% in general surgery [24], but the incidence after cardiovascular surgery is as high as 1.9%-6.9% [25, 26]. Itagaki et al. reported that macrovascular surgery was associated with a high risk (odds ratio 5.6) [27]. After branching from the vagus nerve, the recurrent laryngeal nerve travels posteriorly at the subclavian artery on the right and the ductus arteriosus on the left and ascends the tracheoesophageal groove. It then splits into anterior and posterior branches before reaching the upper edge of the cricoid cartilage. The causes of paralysis are thought to be indirect disorders, in addition to direct disorders of the recurrent laryngeal nerve due to surgical operations on large blood vessels. The first is due to compression of the recurrent laryngeal nerve. Traction of the subclavian artery is triggered by intraoperative repositioning, curvature of the endotracheal tube due to cervical extension, intraoperative operations such as sternum traction, displacement of the cuff position due to repositioning, gastric tube and transesophageal echocardiogram (TEE), and excessive sternum traction. The second is hypoperfusion of the recurrent laryngeal nerve feeding blood vessels due to peripheral circulatory insufficiency during use of the heart-lung machine and transient recurrent laryngeal nerve palsy due to hypothermia. In general, the risk factors for postoperative recurrent laryngeal nerve palsy include hypertension, diabetes, female sex, macrovascular surgery, heart-lung machine, TEE, long intubation, and surgery time. The median hospitalization time of patients with hoarseness complications was 18.5 days, which was longer than that in cases without complications. Although it is a major operation, it requires careful operation.
In this case, one perioperative death was due to right-sided heart failure. Kruger et al. reported an early postoperative mortality rate of 22% [7]. In other reports, perioperative mortality was 0–15%, with right heart failure and respiratory failure being the most common causes [15, 28–30]. In addition, Mussot et al. reported that two out of 31 cases required reoperation and two deaths were due to adult respiratory distress syndrome [31]. Based on this result, it should be remembered that surgery for PAS is major surgery.
Limitations
Our study had a few limitations. First, the sample size was small owing to the rarity of the original disease and was a retrospective observational analysis spanning nearly 15 years. Further research with a larger population and a longer follow-up period is necessary.