Ischemic stroke is one of the leading causes of death worldwide, and stroke recurrence worsens the prognosis of patients and increases both the patient's and society’s costs. Numerous studies have been conducted on the recurrence rate of ischemic stroke and treatment for the prevention of recurrence.33–37 In particular, the incidence of recurrent ischemic stroke at 5 years in patients classified as LAA based on the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification was 8.2–16.8%.34 However, a clear treatment for preventing recurrence of ischemic stroke due to LAA has yet to be established. As previously mentioned, several attempts to improve the disease through intracranial stent insertion in patients with intracranial stenosis have not been as successful as expected.3–5 Conversely, there are some reports that nonemergent EIB, whose effectiveness had previously been questioned, is helpful in limited indications.25–28
Nonemergent EIB for patients with symptomatic and hemodynamic compromised LAA
Two representative RCTs studied the effectiveness of nonemergent bypass for hemodynamically compromised LAA.38,39 The EIBT analyzed the efficacy and safety of EIB in patients with TIA or stroke that had occurred within 3 months before entry into the study.38 Perioperative ischemic events occurred in 12.2%, and major infarction occurred in 4.5%.38 Furthermore, 18% and 20% of single strokes occurred in the medical treatment group and the EIB group, respectively, and two or more strokes occurred in 10% and 11% of each group over an average of 55.8 months, which demonstrated that EIB had no benefit in preventing stroke.38 Another study was the COSS, which also compared a medical treatment group and EIB group, which consisted of ICA-occluded patients with TIA or ischemic stroke that had occurred within 120 days.39 In that study, the 30-day rates for ipsilateral ischemic stroke were 14.3% in the EIB group and 2.0% in the medical treatment group, and the 2-year rates for the primary endpoint were 21.0% and 22.7%, respectively, indicating that EIB had no preventive effectiveness.39 However, the problem with these two studies was that the perioperative morbidity was high, 12.2% and 14.3%, respectively.39 Recently, as bypass technology has been developed and become widespread, several studies on EIB have reported good results with low perioperative morbidity.16–23, 25–28, 40–46 Nevertheless, since most studies are on emergent EIB in AIS, they are of limited use in demonstrating the effectiveness and safety of nonemergent bypass for symptomatic and hemodynamically compromised LAA. We performed elective EIB for highly selected patients with symptomatic and hemodynamically compromised LAA, and cerebral infarction occurred in only 4.7% of patients over a mean period of 3 years without major complications associated with surgery, which is remarkably lower than the natural course of hemodynamically compromised LAA reported previously.33–37
Following quantitative analysis with RAPID software, we showed that cerebral perfusion could be improved and hemodynamic instability could be alleviated through EIB. Perfusion parameters before, immediately after surgery, and long-term indexes up to 6 months after surgery were compared sequentially, and the results demonstrated that most of the parameters improved after EIB. In particular, using imaging evidence, we established that EIB can prevent cerebral infarction, confirming that the volume of Tmax > 6 s, an index of penumbra,47 is significantly improved. The recurrent TIA symptoms in all patients resolved within 1 month, and in patients with previous infarction, based only on if the operation was successful, symptoms disappeared in 97.6%; the other 2 of the 84 (2.4%) patients developed recurrent infarction. When nonemergent EC-IC bypass was performed on patients with symptomatic hemodynamically compromised chronic LAA under our indications, the surgical success rate was 97.7% (84/86), and the probability of symptom improvement was 97.6% (82/84).
Comparison of perfusion parameters with or without previous infarction
When we compared patients with and without previous infarction, there were no significant differences in other characteristics, including preoperative perfusion parameters and intraoperative STA flow, with the exception of the preoperative mRS score. However, postoperative perfusion parameters were significantly better in the noninfarction group than in the previous infarction group. Although there is no known evidence for this, in the previous infarction group, it was assumed that the area where the infarction had already occurred would not have effectively increased perfusion. Additionally, since the previous-infarction group was more likely to have more severe arterial atherosclerosis, we also hypothesized that the intracranial arterial network, such as the leptomeningeal collateral flow, would be relatively poor, so less improvement in perfusion status may occur. Based on this, it can be hypothesized that in patients with symptomatic LAA, EIB before infarction may be more helpful in improving perfusion parameters.
Relationship between bypass flow and perfusion change
According to the results of this study, STA bypass flow as measured with a flowmeter was positively correlated with the degree of perfusion improvement. Similarly, Inoue et al. demonstrated that the STA mean flow was significantly associated with regional cerebral blood flow when cerebral hemodynamics were evaluated by STA duplex ultrasonography.48 Therefore, for perfusion status improvement, it is advantageous to use a method that increases bypass flow, and it may be helpful to select and design the STA and anastomosis site based on Poiseulle's law, which states that flow is proportional to the fourth power of diameter and inversely proportional to length. Accordingly, it is advantageous to select an STA or recipient artery with a larger diameter, although caution is required because there is a risk of hemorrhagic cerebral hyperperfusion syndrome if the STA is significantly thicker than the recipient artery.48–51 Nakamizo et al. reported that the STA diameter and mean STA flow were significantly associated with cerebrovascular reserve capacity, with an STA diameter cutoff value of 1.8 mm.49 In addition, long-slit arteriotomy and fish-mouth cutting of the donor artery to enlarge the anastomosis site may be helpful.50,51 Furthermore, according to Poiseulle's law, to reduce resistance, the length of the STA should be minimized to within the distance from the recipient artery as much as possible, which will contribute to increasing bypass flow.
Limitations
There are several limitations in this study. First, this study was conducted retrospectively. Additionally, it is a single-arm study without a control group. To overcome these limitations, the effectiveness of EIB was evaluated quantitatively using CTP, and we compared the results indirectly with the results of other studies that demonstrated the natural course of symptomatic LAA. Additionally, since this study was conducted in a single tertiary general hospital that has much experiences with bypass surgeries and postoperative care, there may be limitations in applying this result to all hospitals. However, as the bypass technique has recently improved and become more common, this treatment can be considered sufficient for other institutions.