Complications related to sitting position during Pediatric Neurosurgery: An institutional experience and review of literature
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.222852
Source of Support: None, Conflict of Interest: None
Keywords: Anesthesia, children, complication, posterior fossa surgery, sitting position
Posterior cranial fossa and posterior cervical spine surgery present a distinct challenge in terms of surgical exposure. Sitting, lateral, or prone position can be used to access these surgical sites, and each of these positions has its own advantages and disadvantages. At times, neurosurgeons prefer the sitting position because it allows better anatomical orientation and physical access to the lesion, improved venous drainage and gravitational drainage of cerebrospinal fluid (CSF), and better hemostasis as accumulated blood drains away from the operative site.,,, Similarly, the sitting position is also advantageous to the anesthesiologists, as it provides better access to the endotracheal tube (ETT), chest wall and extremities, and freer diaphragmatic movements. Thus, a lower airway pressure is required to expand the chest wall; the position also enables the observation of the face for signs of surgical stimulation of cranial nerves during cranial nerve montoring procedures. Despite the above-mentioned benefits, the use of sitting position has declined worldwide because of the potential life-threatening complications it may be associated with, such as venous air embolism (VAE), paradoxical air embolism (PAE), and tension pneumocephalus.,, Although several studies have demonstrated the relative safety of sitting position for neurosurgery,, its use continues to remain controversial. Posterior fossa surgery in sitting position is still practiced, both in adults as well as in pediatric patients, in many countries. At our institution, the sitting position is practiced based upon the neurosurgeon's preference. We believe that the sitting position offers many advantages with acceptable risks even in pediatric patients. This study was carried out to analyze the perioperative complications and postoperative course of children who underwent neurosurgery in sitting position.
Available medical records of all children up to the age of 18 years, who underwent neurosurgery in sitting position, over a period of 12 years (January 2001 to December 2012) were analyzed. For each child, data were collected by a detailed review of the medical records pertaining to preanesthetic evaluation as well as the intraoperative and postoperative course. Preoperative data included physical characteristics such as age, sex, weight, American Society of Anesthesiologists (ASA) physical status, associated medical illness, hemoglobin concentration, serum electrolytes, and radiological diagnosis. No routine workup was undertaken to detect a patent foramen ovale unless it was clinically indicated. The intraoperative data included a recording of the anesthetic technique, fluids infused, blood loss that occurred and blood transfusion that was given. Intraoperative complications such as hemodynamic alterations (bradycardia and tachycardia, hypotension and hypertension; defined as 20% alteration from the baseline values), VAE [defined as sudden and sustained fall of 4 mmHg or more in end-tidal carbon dioxide (EtCO2)], and respiratory complications (bronchospasm, laryngospasm, and accidental dislodgement of the endotracheal tube [ETT]) were recorded. Data on postoperative parameters included indications for mechanical ventilation, duration of ventilation, incidence of reintubation, surgical complications, fever, seizures, duration of intensive care unit (ICU) and hospital stay, and neurological status at discharge.
A standard anesthetic protocol was followed for all the children. They received either oral atropine or intramuscular glycopyrrolate as premedication. Induction of anesthesia was done by either intravenous (IV) or inhalational technique depending upon the presence or absence of an IV cannula. The trachea was intubated with a reinforced ETT, which was facilitated by nondepolarizing muscle relaxants. A soft bite block was placed after intubation. Anesthesia was maintained with O2:N2O (40:60), and isoflurane or sevoflurane [minimum alveolar concentration (MAC) 0.8–1.2] along with intermittent boluses of vecuronium and fentanyl. Nitrous oxide was used in all children; however, it was discontinued when VAE was suspected. The children were mechanically ventilated without positive end-expiratory pressure to maintain mild hypocapnia to normocapnia. Monitoring modalities included heart rate, electrocardiogram (ECG), pulse oximetry (SpO2), noninvasive and invasive blood pressure (NIBP and IBP), central venous pressure (CVP), EtCO2, minimum alveolar concentration (MAC), temperature, and urine output. The transducer of the IBP monitoring equipment was positioned at the level of the mastoid process and the transducer of CVP was kept at level of the heart. Transesophageal echocardiography (TEE) was used, depending upon its availability and the anesthesiologist's preference. Prior to the year 2005, no leg wrapping with compression bandage was used during positioning; however, after 2005, it was done with a wrinkle-free elastic crepe bandage to avoid pooling of blood into the legs. After induction of anesthesia, the children were preloaded with 10 ml/kg of crystalloids, to avoid postural hypotension and then they were gradually moved into the sitting position over a period of 5 min with continuous monitoring of blood pressure [Figure 1]. During fixation of the head, a distance of at least two-finger width was ensured between the chin and sternum. All potential pressure points were padded with cotton. At the end of surgery, the children were made supine. The patient was extubated after reversal from neuromuscular blockade. The reinforced ETT was changed to a polyvinyl chloride tube if elective ventilation was planned.
Statistical analysis was done using STATA 11.2 (Statacorp, Texas, USA). Data are represented as mean [with standard deviation (SD)], median [with range], or number [with percentage]. Chi-square analysis was done to compare categorical variables. Nonparametric numerical data were analyzed using Mann–Whitney test. A P value <0.05 was considered as statistically significant.
The medical records of 97 children, who underwent various neurosurgical procedures in sitting position during the study period, were analyzed. The median age of the children was 12 years (3–18 years) [Table 1]. All the children belonged to the ASA physical status I and II. There was no clinical evidence of septal defect in the heart in any of these patients. VAE was observed in 21 (21.6%) children. Considering the fact that the sagittal sinus pressure remains positive in children even in sitting position up to the age of 9 years, VAE was compared in children between the age range of above 2 years and below 9 years, and in those above 9 years of age [Table 2]. However, the incidence of VAE was found to be comparable (P = 0.7) in between the two groups. There were a total of 38 separate episodes of VAE in 21 children (21.6%); the maximum number of episodes occurring in a single child were seven. Nine (42.8%) children experienced more than one episode of VAE during their surgery. Most of the episodes of VAE occurred during excision of cerebellar tumors, brainstem gliomas, and third ventricular tumors [Figure 2]. One child had VAE during decompression of foramen magnum for Chiari malformation type I. VAE was accompanied by hypotension in 5 (23.8%) and desaturation with SpO2≤95% in four (19.1%) of 21 children. The maximum drop of saturation was upto 85%, observed only in children who experienced multiple episodes of VAE. Most of the episodes of VAE occurred during the craniotomy and resection of tumors; however, in one patient, it occurred during elevation of the myofascial flap. Intraoperative TEE was used in 8 children. All VAE episodes were evident with TEE and EtCO2 when both were used in the same patient. VAE was managed by informing the neurosurgeon who covered the surgical field with saline-soaked sponges, as well as by switching the patients on 100% oxygen and by aspiration of air from the central venous catheter. Jugular compression was not required in any of these children. Intravenous fluid boluses or vasopressors were administered when VAE was associated with hypotension. Intraoperative hemodynamic instability (unrelated to VAE) occurred in 12 (12.3%) children. These instabilities were mainly bradycardia or tachycardia and occasional ventricular premature beats, at times associated with hypotension or hypertension, due to tumor manipulation near the brainstem. They resolved spontaneously after cessation of surgical stimulation. In one child, hypotension was associated with rapid blood loss during tumor removal. Forty-seven (48.4%) children required postoperative elective ventilation; among them, 10 children had suffered an intraoperative VAE. However, VAE was not the indication for elective ventilation in any of these patients. Three children remained drowsy following reversal from residual neuromuscular blockade, and extubation was not immediately performed in them. The most common indication for continuation of postoperative ventilation was brainstem handling (42.5%). The other indications were preoperative lower cranial nerve palsy, disorientation, and the presence of residual tumor. Intraoperative blood loss ranged from 2 to 80 ml/kg (median 6 ml/kg). Twenty-four (24.7%) children required an intraoperative transfusion of blood and blood products; 13 (13.4%) requiring transfusion of blood more than 10 ml/kg [Table 2]. Six (6.1%) children developed tension pneumocephalus diagnosed on computed tomographic (CT) scan in the postoperative period; three developed seizures and one among them had deteriorated consciousness. They were managed conservatively. Two other children presented with delayed recovery from anesthesia; a computed tomographic (CT) scan followed by twist drill burr-hole evacuation of the air improved their condition. One child who was extubated in the operating room became drowsy and desaturated 4 h postoperatively, following which endotracheal intubation and twist drill burr-hole evacuation of the pneumocephalus was done. Six children developed pneumoventricle in the postoperative period; two of them had a pneumoventricle in association with tension pneumocephalus, and the remaining four had clinically insignificant pneumoventricle. One child developed an acute subdural hematoma (SDH) in the immediate postoperative period that required surgical intervention. Five (5%) children required reintubation, due to various reasons such as post-extubation altered sensorium (1), pulmonary edema (1), seizure following tension pneumocephalus (2), and subdural hematoma (SDH) (1). Other complications encountered were postoperative fever, meningitis, hydrocephalus, CSF leak, and pseudomeningocele [Table 3]. One child required surgical re-exploration due to the development of postoperative SDH. Apart from tension pneumocephalus, no other morbidities associated with the sitting position, such as macroglossia, quadriplegia, and peripheral nerve injury, were encountered.
The median duration of the ICU stay was 87 (6–720) h and the mean duration of hospital stay was 12.8 ± 6.2 days. There was no difference in the incidence of postoperative complications in children with VAE or in those without it; no mortality could be attributable to VAE. Eighteen of 21 patients who experienced VAE were discharged with a mild disability [Table 4].
The sitting position for neurosurgery is as controversial today as it was when first advocated by De Martel in 1931. Although the use of sitting position is declining in neurosurgery, some centers all over the world are still practicing it owing to its distinct advantages. In our hospital, this positioning is mainly used for posterior fossa surgeries but not for upper cervical spine surgeries. The main concern in the sitting position is VAE and its sequel, the true incidence of which cannot be defined as it depends upon the sensitivity of detection method used. The reported incidence of VAE in adults, when detected by a TEE was 100%, and when detected by a precordial Doppler was 50%. Several articles have reported the incidence of VAE occurring during the sitting position in adults; however, there are only a limited number of studies reporting its incidence in children. A wide range of incidence in children has been proposed (9.3–33%), which is certainly less as compared to the adults.,,, This discrepancy in the incidence of VAE can be attributed to different monitoring methods used to detect VAE, the variable age groups in which the study was conducted and the influence of age on the cerebral sinus pressure, the effect of preloading of the patients, the use of antistatic stockings, and the use of variations in the sitting (or lounging) positions utilized. The nature of the study (that is, whether it was prospective or retrospective) may also influence the incidence of VAE detectable during surgery in sitting position. Two prospective studies suggested the incidence of VAE in children based on EtCO2 monitoring to be 22 and 26.3%, respectively., In our study, the incidence of VAE was 21.6%, a result which was similar to that seen in both of these studies. However, Harrison et al., reviewed 407 children in the age group of 1.5 months to 17 years and observed a much lower incidence of VAE (9.3%), as detected by EtCO2. The authors attributed the lower incidence of VAE detected in children to the presence of relatively higher dural sinus pressure in children as compared to the adults. On the contrary, Matjasko et al., reported an incidence of 62% of VAE in children by Doppler and capnometry as compared to 23% in adult patients. However, in this study, there was a wide variability in the patient population that was largely composed of adults (541 adults and 13 children). Had TEE been used routinely, the incidence of VAE detected would probably have been more in this study. Another significant point to be evaluated is whether or not the entire spectrum of “TEE detected” VAE would actually have been clinically significant. Iwabuchi et al., examined dural sinus pressure [confluens sinus pressure (CSP)] in 47 cases (11 were children). In sitting position, adults showed a negative CSP, whereas all children of age ≤9 years (n = 8) showed a positive CSP pressure. Possibly, a positive CSP in younger children is responsible for the lower incidence of VAE in this age group. Grady et al., studied the relationship of superior sagittal pressure (SSP) to head position in 15 children (within the age range of 1–17 years) and concluded that progressive head elevation significantly decreased the mean SSP in 5 patients; however, they did not mention the age group of these children. We did not find any significant difference in the incidence of VAE in children above or below 9 years of age, which is in accordance with the findings of our previous study.
There is a hypothesis that the incidence of VAE associated with hypotension in children might be higher than in adults. This is because an equivalent size of air bubble would be larger relative to the small cardiac volume in children, thus causing pronounced hypotension. This hypothesis was supported by Cucchiara and Bowers  as they retrospectively reviewed 96 patients undergoing suboccipital craniotomy in sitting position. They observed that the incidence of VAE associated with hypotension (fall in systolic arterial pressure more than 25 mmHg) was greater in children (69%) than in adults (36%). However, there are three prospective studies in pediatric patients reporting conflicting results. Meyer et al., observed that severe hypotension (fall in mean arterial pressure more than 25 mmHg) occurred in all 60 children with VAE, as detected by capnometry. However, Fuchs et al., did not observe any child developing hypotension associated with VAE in a series of 30 children. Bithal et al., reported that the incidence of VAE associated with hypotension was similar in both adults (37%) and children (33%). Our results were comparable to the results of the study by Harrison et al., who reported an incidence of 20.9% of VAE associated with hypotension. Any preexisting hypovolemia, even mild, predisposes patients to VAE and associated hypotension. Maintenance of adequate intravascular volume is even more important in children, as air bubbles in the relatively smaller right atrium might result in more frequent cardiovascular changes as compared to adults. Glenski et al., have shown that transcutaneous oxygen tension monitored in neurosurgical patients provided an earlier and more reliable indication of VAE than did EtCO2, but no direct study has suggested that changes in SpO2 may be regarded as an early manifestation of VAE in humans. In this study, oxygen desaturation was observed only in those patients who experienced multiple episodes of VAE. It may be hypothesized that an acute air embolism which is superimposed on an earlier incompletely resolved embolism might have caused a further increase in PaCO2. This combination of an elevated PaCO2 together with venous admixture resulted in clinically significant arterial desaturation. Nevertheless, in none of the children was the surgical position changed or the surgery was abandoned owing to the occurrence of VAE, which implies that this complication is quite manageable.
The sitting position may reduce blood loss by facilitating drainage of venous blood away from the operating site. Black et al., reported in adult patients that intraoperative blood loss and volume of blood transfused were significantly lesser in patients undergoing posterior fossa craniectomy in sitting position as compared to those operated in horizontal position. In our study, blood loss was relatively lesser (6 ml/kg); only one patient had massive blood loss (80 ml/kg), as the tumor was large and highly vascular. Orliaquet et al., compared perioperative complications in children undergoing posterior fossa surgery in sitting and prone positions and concluded that the volume of intraoperative blood transfusion was significantly larger in the prone position as compared to the sitting position.
Intraoperative hemodynamic instabilities can occur during surgery conducted in sitting position, particularly, during positioning, episodes of VAE, and brainstem or cranial nerve handling. Harrison et al., reported the occurrence of intraoperative hemodynamic instabilities (other than during VAE) in 24.1% children undergoing posterior fossa surgery in sitting position. However, this incidence was much more as compared to that found in our study (12.3%), possibly because the criteria of defining hemodynamic instability (10% vs 20% change in heart rate or systolic blood pressure) were different in the two studies.
In our study, postoperative mechanical ventilation was required in a large number (48.4%) of children. It has been suggested that children who undergo posterior fossa surgery should be extubated only after the integrity of bulbar function and lower cranial function, apart from adequate recovery from the anesthesia, has been ascertained; otherwise, the child should be kept intubated for 24h. The children presenting to our hospital usually have a large tumor at the time of admission due to our hospital being an apex referral center. The large size of the tumors encountered poses difficulty in tumor dissection with more chances of intraoperative brainstem handling.
The development of tension pneumocephalus is a serious and life-threatening emergency condition. This complication has been attributed to the diminution of brain volume secondary to mannitol administration, hyperventilation, removal of the space occupying mass, and contraction of intravascular blood volume due to the intraoperative hemorrhage encountered. Intraoperative drainage of cerebrospinal fluid, commonly via the subarachnoid drainage and the gravitational effect of sitting position, may increase the likelihood of air entry into the subdural space. Lunsford et al., have proposed the “inverted pop bottle” analogy to describe this phenomenon. The symptoms may occur 2–4h after surgery, which may present as delayed recovery, severe restlessness, deterioration of consciousness, and seizures. The incidence of symptomatic tension pneumocephalus in our study was similar to that found in the study conducted by Orliaquet et al. Most of the children in that study had a preoperative ventriculo-peritoneal (VP) shunt installed. It was concluded based upon the observations made in that study as well as ours that in children in a similar situation, the sitting position should be avoided.
Postoperative seizures are relatively rare in posterior fossa surgery. Three children presented with seizures owing to tension pneumocephalus, which was in accordance with the hypothesis postulated by Suri et al.
Macroglossia is a known but uncommon complication of neurosurgical procedures in sitting position. We did not encounter macroglossia in any of the children, possibly due to the usage of a soft (cotton) bite block, which prevented compression of the tongue by the teeth; and, due to the maintenance of an adequate distance between the chin and sternum.
Peripheral nerve injuries have been reported in association with the sitting position for neurosurgery. The incidence of peripheral nerve injury (foot drop following common peroneal nerve injury) in a series of 488 patients was <1%. No such complication was encountered in this series. It might be due to careful padding of all pressure points with cotton and also due to precautions taken to prevent hyperflexion at hip. There are occasional reports of quadriplegia/paraplegia in the postoperative period after neurosurgical procedures in sitting position. It may occur in patients with degenerative cervical spinal disease or cerebrovascular disease., These comorbidities are usually not encountered in the pediatric population, making the possibility of quadriplegia rare.
Interestingly, in this study, there was no mortality which could have been directly attributable to the sitting position. Two children died on the 15th postoperative day, following a protracted course of sepsis and related morbidities. In a series of 1792 sitting procedures, Himes et al., found the overall incidence of mortality to be as low as 1.45%, which included clinically significant VAE as one of the etiological factors.
In conclusion, this study observed a relatively lesser incidence of VAE and associated complications in children undergoing neurosurgery in sitting position. The other major complication was tension pneumocephalus with seizures, which occurred in 5% children who were managed successfully without any adverse outcome. We believe that avoidance of sitting position in children with a VP shunt in situ may reduce the incidence of tension pneumocephalus. There was reduced requirement of blood transfusion in children operated upon in sitting position. Overall, this study reported very few intraoperative complications; hence, with meticulous anesthetic and surgical planning, the sitting position can be safely practiced in children.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]