Anesthetic factors and outcome in children undergoing indirect revascularization procedure for moyamoya disease: An Indian perspective
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.166575
Source of Support: None, Conflict of Interest: None
Background: Moya Moya disease (MMD) is one of the most common cerebro-vascular diseases in children resulting in stroke. Surgical revascularization aims at improving the perfusion to the 'at-risk' ischemic brain. Several factors including peri-operative anesthetic related ones, affect the outcome in these children. This study was performed with the aim to explore the role of anesthetic techniques, pharmacological agents and perioperative management strategies on the neurological outcome following an indirect revascularization procedure for the treatment of MMD.
Keywords: Anesthesia; hospital stay; moya moya disease; pediatrics; postoperative neurological deficits; revascularization
Moya Moya disease (MMD) is a chronic idiopathic cerebrovascular disease characterized usually by progressive bilateral stenosis of the terminal portions of the internal carotid artery (ICA) and its branches. As a compensatory mechanism, collateral circulation develops between the ICA and the external carotid arteries (ECA) giving the appearance of a “puff of smoke” on angiography. This appearance was first termed “moya moya” by Suzuki and Takaku in 1969. The incidence of MMD is highest in Japan with an incidence of 0.54 and a crude prevalence of 6.03 per 100,000 population and it is the most common cerebrovascular disease in the pediatric population. Up to 80% of children with MMD present with recurrent ischemic episodes in the form of a transient ischemic attack (TIA) or stroke mostly involving the anterior circulation of the brain. Both direct and indirect methods of revascularization are described in the literature, and are aimed at perfusing the ischemic hemisphere through collaterals from the ECA. Though direct revascularization is considered to be superior and avoids performing a repeat revascularization at a later stage, it may not always be possible because of the smaller calibre of the donor and recipient vessels in young children. A variety of indirect procedures have, therefore, been described, including encephalo-duro-arterio-myo-synangiosis (EDAMS) or a different combination of individual procedures. Based on the previous case reports and studies, maintenance of normotension and normocapnia during the perioperative period is recommended. Despite these intraoperative measures, postoperative new neurological deficits have been documented, suggesting the role of other perioperative variables like fluid balance, intraoperative blood loss, anesthetic agents, duration of procedure etc., in their causation. Currently, very little information is available in literature regarding the influence of these components of anesthetic management, on the neurological outcome in children undergoing indirect revascularization for MMD. The current study is aimed at evaluating the influence of other possible details of anesthetic management on the post-operative outcome in children undergoing EDAMS for revascularization in MMD.
All children with a diagnosis of MMD, undergoing an indirect revascularization procedure for preventing future ischemic episodes during a three year period (1st April 2011 to 31st March 2014), were included in this retrospective study. This retrospective study was approved by the NIMHANS ethics committee. Children undergoing direct revascularization and in whom complete data was unavailable were excluded from the study. The data collected in this study included demographic details (age, weight, gender), disease characteristics (pre-existing deficits and Suzuki grading on angiography) and variables related to anesthetic management (anesthetic induction and maintenance, average minimum alveolar concentration [MAC] of volatile anesthetics, use of nitrous-oxide/air, average analgesic dose, duration of anesthesia and surgery, baseline, intraoperative and postoperative blood pressure (BP), mean end-tidal carbon dioxide (ETCO2) levels, pre- and post-operative hemoglobin concentration, intraoperative fluid intake and urine output, and intraoperative blood loss). The primary outcome evaluated was the occurrence of new postoperative neurological deficit (POND) or death. The secondary outcome measure assessed was the length of post-operative hospital stay. A ten day cut off was used to describe prolonged postoperative hospital stay based on the average stay of these patients in our hospital.
Statistical analysis was performed using statistical package for the social sciences version 16.0 (SPSS 16.0 for Windows SPSS Inc., Chicago IL). Factors related to anesthetic management and presumed to be associated with new POND were analysed by using Mann-Whitney U-test for continuous variables and chi square test or Fischer exact test for categorical variables as appropriate. Factors significant on univariate analysis (P < 0.1) were entered into a regression model to identify predictive factors for new POND. Similar statistical tests were performed for analysing the postoperative length of hospital stay. Data are expressed as mean and standard deviation, or numbers and percentage as appropriate. A P value of < 0.05 was considered statistically significant.
Twenty-two children underwent thirty indirect revascularization procedures (two patients EDAS, one EMS and the rest EDAMS) during the study period. Eight children underwent bilateral procedures. Four patients had new postoperative neurological complications. The mean age and weight of children who underwent an indirect revascularization procedure for MMD was 7.3 ± 3.5 years and 24.3 ± 13.0 kg, respectively. The demographic details, neurological characteristics and anesthetic-related details in patients who developed POND and those who did not, are presented in [Table 1]. None of the factors analyzed were associated with occurrence of new POND. No clinical or statistical difference was noted in the baseline (P = 0.14), mean intraoperative (P = 0.25) and postoperative (P = 0.91) blood pressures in children who developed POND and those who did not. Factors associated with prolonged post-procedure hospital stay are shown in [Table 2]. There was no difference in the baseline (P = 0.25), mean intraoperative (P = 0.12) and postoperative (P = 0.74) blood pressures between children who had a postoperative hospital stay of less than 10 days or more. Younger age, new POND and prolonged duration of anesthesia were associated with a longer hospital stay after surgery on univariate analysis. However, on multivariate logistic regression analysis, none of these factors were predictive of prolonged post-operative hospital stay in this study.
The anesthesia related factors and their association with clinical outcome in children undergoing indirect revascularization for MMD have been inadequately studied. This study explored the possible peri-operative factors that are likely to be associated with new POND and prolonged hospital stay in twenty-two children with MMD undergoing thirty indirect revascularization procedures.
In this series, there were four children who developed new postoperative neurological complications. The first patient had a single episode of a new onset seizure. The remaining three children developed motor deficits. The first child who underwent EDAS procedure on the right side developed ipsilateral limb weakness immediately after surgery, which improved by the 3rd day. This child had a left middle cerebral artery territory hypodensity on computed tomography (CT) scan. She underwent uneventful surgery on the opposite side after six months. The second child, who developed immediate postoperative deficits, underwent bilateral revascularization on the same day and died on the 29th day due to bilateral cerebral infarcts. This patient had significant postoperative anemia (8.4 g %). The third child developed new onset hemiparesis contralateral to the operative side on the 3rd postoperative day which improved before discharge. This child had lower blood pressures during the perioperative period as compared to baseline, and also lower intraoperative ETCO2 value (29 mmHg). The same child underwent surgery on the opposite side three months later. She developed transient hemiparesis on the 3rd postoperative day without any CT evidence of infarct.
The primary goal of anesthetic management during revascularization procedures for MMD is maintaining the balance between cerebral oxygen demand and supply. Towards this end, the measures that the anesthesiologist needs to pay attention to include, maintaining normocapnia, normoxia, normothermia, normotension and normovolemia as abnormalities of these physiological parameters are shown to be associated with ischemic complications.,
In the current series, except for the two children who received sevoflurane induction, rest of the 28 children had intravenous anesthetic induction and maintenance with volatile anesthetic agents. Sato et al., demonstrated a decrease in the cortical blood flow and reduction in the frontal regional cerebral oxygen saturation with an inhalational technique as compared to total intravenous anesthesia, suggesting the occurrence of cerebral steal phenomenon. Similarly, Kikuta et al., noted that intracranial pressure was lower and frontal cerebral blood flow was higher with propofol anesthesia as compared to sevoflurane anesthesia. In contrast, another study comparing inhalational with intravenous anesthetic technique, found similar incidence of post-operative complications (44% vs 57%) in 39 pediatric patients with MMD undergoing a revascularization procedure. In our study, we did not find any difference with regard to the occurrence of POND between the two inhalational anesthetics used (isoflurane and sevoflurane). Also, use of nitrous oxide instead of air did not affect the outcome. These findings reconfirm the lack of superiority of any particular anesthetic technique on the postoperative neurological outcomes.
Perioperative hypocapnia has been extensively reported and identified as the main factor for ischemic complications in MMD., Moyamoya vessels are believed to overreact to hypocapnia, precipitating TIAs in children during hyperventilation. However, in a case series involving five children, short periods of modest hypocarbia were not associated with neurological deterioration in children with MMD anesthetized with potent inhalational agents. Though the average ETCO2 values in our study in those who developed POND was lower than those who did not, this was not statistically significant. While hypocapnia (either from crying or from pain) results in ischemic complications, excessive sedation for parental separation or the use of opioids for post-operative pain relief can also cause ischemic complications from resultant hypercarbia. Significant decreases in perfusion have been observed both during hypocapnia and hypercapnia in patients with MMD using laser doppler flowmetry. An abrupt normalization of a hypocapnic state can cause intraoperative brain swelling during a revascularization procedure for MMD. Hypercarbia, which mostly occurs from overzealous attempts to prevent hypocapnia, causes reactive dilatation of unimpaired vessels resulting in steal phenomenon and ischemia in MMD. Kurehara et al., have demonstrated significant decreases in the cortical blood flow following induction of hypercarbia (ETCO2 of 45-50 mmHg) from a normocarbic state (38.7 ± 14.4 vs 42.8 ± 13.4 ml. 100 g-l min-1, respectively) using laser doppler method in fourteen patients undergoing a revascularization procedure. The collateral pathways through the abnormal vascular network are often fully dilated while the stenosed cerebral arteries exhibit a high resistance and, hence, may not respond during hypercapnia. Furthermore, in MMD, stenosis of the anterior circulation arteries precedes that of the posterior circulation arteries, diverting blood away to regions with a relatively preserved CO2 reactivity. None of the children in this study had hypercarbia (ETCO2 >40mmHg) during the intraoperative period and hence POND cannot be attributed to this factor.
Maintenance of normotension is recommended in MMD to prevent ischemic insults. However, perioperative transient hypotension was not associated with ischemic complications in a study by Iwama et al., probably because the duration of hypotension was controlled. In our study, the mean peri-operative systolic BP in patients who developed POND was lower compared to those who did not, but the difference was not statistically significant (104.2 ± 21.3 vs 108.7 ± 13.1; P = 0.53). Postoperative hypertension is also observed in MMD following an indirect revascularization procedure. Yum et al., used preoperative heart rate variability tests to predict the occurrence of postoperative hypertension following EDAS surgery for MMD. Recently, autoregulation monitoring with near infra-red spectroscopy derived indices was used to identify the optimal blood pressure range in children with MMD. The optimal mean arterial blood pressure with best autoregulation was found to be around 60-80 mmHg.
Despite a smooth anesthetic and an uneventful intraoperative course, four children developed POND in a series described by Sakamoto and co-investigators. These postoperative deficits were delayed in nature and were attributed to crying and dehydration. In contrast, the patients who developed POND in our series had a more positive fluid balance during the peri-operative period, making dehydration a less likely cause for POND. Hypocapnia-induced seizure and neurological worsening has been reported in a child with MMD during intensive care management. It is difficult to prove the association between hypocapnia and seizure in our patient as ETCO2 was not monitored and blood gas analysis was also not performed during this postoperative episode. In the only large series involving both adult and pediatric population, in patients undergoing both direct and indirect revascularization procedures, the authors observed that POND developed in patients who suffered frequent TIAs, had precipitating factors for TIA, and underwent indirect revascularization. This study, like our study, also failed to demonstrate an association between variables relating to anesthetic management and the occurrence of POND. While the earlier study did not analyze the length of hospital stay as an outcome measure, our study demonstrated that younger age, occurrence of POND and prolonged anesthetic duration were factors associated with a longer post-operative hospital stay. Both the earlier and current study demonstrate that as long as the physiological variables are handled meticulously within normal range throughout the perioperative period to maintain perfusion to the vulnerable areas of the brain, anesthetic drugs or techniques do not seem to be associated with new POND in children undergoing an indirect revascularization procedure.
Inadequate pain management has been implicated in the development of postoperative ischemia in MMD., All children in this study had received either diclofenac or paracetamol for postoperative analgesia on a fixed schedule and dose basis. This was in addition to the transitional analgesia with opioids (morphine, pethidine or fentanyl) for immediate post-operative pain relief. This retrospective study, however, could not ascertain the relationship between pain scores in the postoperative period and the development of POND. A previous study has shown that skull block is associated with both reduced postoperative pain and analgesic requirement, and also with the reduced occurrence of ischemia in patients with MMD. Hence, all measures to prevent and treat postoperative pain have to be instituted to prevent this avoidable complication. Crying from inadequate pain relief, hunger, parental separation, fever etc., are other important predisposing factors for ischemia in patients with MMD. Two children, who developed ischemia in this series, had crying episodes in the postoperative period. Recently, in the only other Indian study examining anesthetic details, Sharma et al., have described their experience on the anesthetic management of patients (both children and adult) undergoing both types of surgical intervention (direct anastomosis and indirect revascularisation procedures). However, their study did not specifically look at the possible risk factors arising from anesthetic management in the occurrence of new POND, which our study has investigated. Further, our study differed from this study in that ours was a homogeneous population with only pediatric patients undergoing the indirect revascularization.
To conclude, in this study, anesthetic techniques or drugs did not influence the occurrence of POND after an indirect revascularization procedure in children with MMD. Factors associated with a longer postoperative hospital stay were younger age, new POND and a prolonged anesthetic duration. A larger prospective study across various centres may establish the significant predictors contributing to the occurrence of POND.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2]