Venous Air Emboli Occur During Release of Positive End-Expiratory Pressure and Repositioning After Sitting Position Surgery

Abstract

We studied the effect of positive end-expiratory pressure (PEEP) release and positioning on the occurrence of venous air embolism (VAE). Eighteen consecutive patients (8 women, 10 men; ASA grade I–III) undergoing neurosurgery in the sitting position were studied. After induction of anesthesia ventilation was controlled with a PEEP of 5 cm H₂O in an oxygen-air gas mixture. A transesophageal echocardiographic (TEE) probe was inserted. Preoperatively, a patent foramen ovale was excluded in all patients. TEE monitoring was performed during surgery, during PEEP release at the end of surgery with the patient still in the sitting position, and during change of the patient position into the supine position. The severity of VAE was differentiated as follows: grade 1 = only microbubbles; grade 2 = microbubbles and decrease of end-tidal carbon dioxide partial pressure (Petco₂) by more than 1.5 mm Hg; grade 3 = microbubbles combined with a decrease of Petco₂ by more than 1.5 mm Hg, and a decrease of mean arterial blood pressure by at least 20 mm Hg. During surgery, VAE with a grade of 1, 2 or 3 occurred in 7, 4, and 2 patients, respectively. After PEEP release, VAE of grades 1, 2, and 3 were observed in 7, 2, and 1 patients, respectively. During repositioning from sitting to supine position, VAE of grades 1, 2, and 3 was observed in 6, 1, and 1 patients, respectively. The patient with VAE grade 3 needed inotropic support until 2 h after surgery to maintain sufficient blood pressure. No patient showed any sign of paradoxical arterial embolism or cardiac dysfunction. We conclude that VAE occurs not only during surgery in the sitting position, but also with release of PEEP and during repositioning to the supine position.

Venous air embolism (VAE) is a potential complication of surgery in the sitting position and its incidence has been reported as often as 76%⁽¹⁾ when transesophageal echocardiography (TEE) was used for monitoring. Despite this, use of the sitting position is common in German neurosurgical centers ⁽²⁾. TEE is a sensitive monitor to detect VAE ^(3,4). Its use has significantly increased during the last decade as an intraoperative monitor for VAE. TEE is used in 40%–50% of the neurosurgical centers during surgery in the sitting position ⁽⁵⁾.

The use of positive end-expiratory pressure (PEEP) has been advocated to prevent and treat VAE ⁽⁶⁾, but it may increase the probability of paradoxical air embolism ⁽⁷⁾ and cardiovascular depression ⁽⁸⁾. In our setting, neurosurgery in the sitting position is only performed when preoperative TEE (performed by cardiologists) excludes a patent foramen ovale and possible right-to-left interatrial shunting. It is our routine, however, to apply a moderate PEEP of 5 cm H₂O in all patients.

Nothing is known about the effect of PEEP release at the end of surgery on the occurrence or recurrence of VAE. There are no studies on possible recurrence of VAE during the change of position at the end of surgery (sitting to supine), probably because it is common practice to remove the TEE probe before repositioning the patient into the supine position (for ease of maneuvering).

We undertook this study to prospectively evaluate the effect of PEEP release and change of patient position (sitting to supine) at the end of surgery on possible recurrence of VAE using TEE monitoring.

Patients and Methods

After obtaining local ethics committee approval and written informed consent, we prospectively studied 18 consecutive patients (11 men, 7 women) scheduled for elective neurosurgical procedures in the sitting position (17 posterior fossa procedures, 1 cervical laminectomy).

Anesthesia was induced using a bolus of propofol 2 mg/kg, and 10 μg/kg alfentanil IV. Orotracheal intubation was facilitated by atracurium 0.5 mg/kg IV. After intubation, the lungs were ventilated with an oxygen-air-mixture (Fio₂ = 0.5) and a PEEP of 5 cm H₂O to maintain an end-tidal carbon dioxide pressure (Petco₂) between 30 and 40 mm Hg. Anesthesia was maintained using a continuous infusion of propofol 100–160 μg · kg⁻¹ · min⁻¹ and alfentanil 0.1–0.2 μg · kg⁻¹ · min⁻¹.

Standard monitoring included electrocardiography, pulse oximetry, invasive blood pressure, central venous pressure via a catheter inserted into the right jugular vein with the tip placed into the right atrium (controlled by P-wave evaluation) and precordial Doppler.

TEE was performed with a multiplane TEE probe (Hewlett-Packard Sonos 2000 3-MHz probe; Hewlett-Packard, Andover, MA). After positioning the patient in the sitting position, a TEE examination was performed to detect any abnormalities in the heart structures in 2-D view mode. PEEP was increased to 15 cm H₂O and 10 mL of agitated hydroxyethyl starch was injected into the right atrium just as PEEP was released to detect a possible patent foramen ovale ⁽⁹⁾. Only patients with a negative test were operated upon in the sitting position.

Afterwards the TEE probe was positioned to obtain a routine basal short axis view at aortic valvular level (Fig. 1) to detect VAE. The TEE image was continuously monitored on a video screen during the whole procedure and recorded on a video tape. At the end of surgery PEEP was released in the sitting position and, after release of the skull head holder, the patient was repositioned to the supine position (30° head tilt up). Patients were then transferred to the intensive care unit with mechanical ventilation of the lungs continued.

Figure 1: Routine basal short axis view at the level of aortic valve enabling the anesthesiologist to focus on the right ventricular outflow track and possible paradoxical air embolism into the left atrium (LA). Right ventricular outflow track left to right: right atrium (RA), right ventricle, main pulmonary artery.

The severity of VAE was graded as follows: grade 1= only microbubbles; grade 2 = microbubbles and sudden decrease of Petco₂ by more than 2 mm Hg; grade 3 = microbubbles combined with a decrease of Petco₂ by more than 2 mm Hg, and a decrease of mean arterial blood pressure by at least 20 mm Hg. In case of intraoperative VAE, the surgeon was informed, the surgical site flushed with normal saline and jugular venous compression applied. For each patient only the highest grade of VAE was noted. Results are presented as mean (range).

Results

Mean age, weight, and height of 18 patients (ASA grade I, n = 4; ASA grade II, n = 8; ASA grade III, n = 6 patients) were 43 (20–73) yrs, 74 (55–95) kg, and 173 (162–202) cm, respectively.

In 13 patients (72%), at least one occurrence of VAE was recorded intraoperatively by TEE (Table 1). There were no electrocardiographic abnormalities such as ST-depression or arrhythmias. Except in one patient, mean arterial pressure did not decrease more than 30% below preoperative values. There was no recorded peripheral oxygen saturation less than 93% in any patient. After careful hemostasis and no further sign of air entry, surgery continued in all patients.

Table 1: Incidence of Venous Air Embolism (VAE) During Surgery, After Positive End-Expiratory Pressure (PEEP) Release, and After Changing from the Sitting Position to a Supine Position with 30 Degree Head Tilt Up

PEEP release to 0 mm Hg was achieved in all patients immediately after skin closure while still in the sitting position. PEEP release lead to VAE of different grades in 10 patients with intraoperative VAE (Table 1). Repositioning the patient to supine occurred in all patients between 5 and 10 min after PEEP release and caused VAE of various degrees in 8 patients (Fig. 1). These patients also had VAE at PEEP release. There was no patient who showed VAE during PEEP release or change of position who did not have an episode of VAE during surgery. VAE after skin closure did not exceed any previous VAE grade in any patient.

In 2 patients (11%), VAE of grade 3 occurred during surgery, with one patient showing VAE of grade 3 also during PEEP release and change to supine position. This was the only patient who needed inotropic support (dopamine infusion) to restore hemodynamic stability during surgery, The dopamine infusion was reduced gradually in the intensive care unit and was discontinued 2 h after surgery. Within 2 min after the final positioning of the patient, air entry could no longer be detected by TEE. In this patient, there was neither wall motion abnormality detected in TEE 2 D-mode nor any other sign of coronary artery air embolism. His subsequent postoperative course was uneventful. There was no patient where right-to-left shunting of air could be detected at any period of monitoring.

Discussion

This study shows that VAE occurring during a neurosurgical procedure in the sitting position can reoccur during PEEP release and during repositioning of the patient into the supine position after skin closure at the end of surgery. Patients with no episodes of VAE during surgery did not experience VAE during PEEP release or repositioning.

The use of PEEP during procedures in the sitting position is controversial and few controlled studies have been done ⁽¹⁰⁾. Earlier studies advocated the use of PEEP to decrease the risk of VAE ^(6,11) by increasing right atrial pressure and, in turn, increasing cerebral venous pressure above atmospheric pressure ⁽¹²⁾. More recent studies, however, did not confirm these findings in animals ^(13,14) or in humans ⁽⁸⁾. Before completion of this study, the application of a moderate PEEP of 5 cm H₂O was the compromise established in our setting between the advantages and disadvantages of PEEP application. The incidence of intraoperative VAE in our study was comparable to other reports where TEE was used for air detection ⁽¹⁾, and confirms TEE to be a sensitive monitor for detection of air embolism. In contrast to other reports ^(1,15), none of our patients showed any sign of right-to-left air entry during the procedure with a PEEP of 5 cm H₂O. This might be because there were no patients with patent foramen ovale (after two negative preoperative screenings in our study) or because of the fact that moderate PEEP of 5 cm H₂O does not pose a substantial risk of paradoxical air embolism.

The current study was designed the to investigate the effect of PEEP release in patients in the sitting position. There is only one study that investigated the occurrence of VAE in pigs in the supine position undergoing continuous venous air infusions ⁽¹⁶⁾. That study showed that PEEP release (decrease from 10 cm H₂O to zero PEEP) caused increased paradoxical air embolism (PAE) in 5 of 15 pigs, new PAE in 4 of 15 pigs, and a similar amount of PAE in one pig. Although it is difficult to draw any conclusions from that study for the conditions and physiological mechanisms of VAE during PEEP release in the present study, it demonstrates that PEEP release can affect air distribution and interatrial pressure gradients. This effect can be used to detect right-to-left shunt by PEEP release during preoperative contrast echocardiography ⁽⁹⁾. We hypothesize that the sudden decrease of moderate PEEP might have decreased the right atrial pressure and subsequently increased venous return from cerebral and intrathoracic veins. Because of the decrease of pressure below atmospheric levels, air, trapped inside the closed cranium, might have entered the venous blood stream and the right atrium. The fact that this occurred only in patients with previous intraoperative VAE might be explained by continuous open venous sites in the wound area, veins that are partly collapsed while in the sitting position ⁽¹⁴⁾. It also seems likely that previously entrained air could have been trapped within intrathoracic veins—undetected by the TEE focused on the heart—and the decrease of PEEP might have caused a redistribution of these air bubbles into the heart.

It has been common practice to change the position of the patient after or during severe VAE from sitting to left lateral recumbent position, mainly based on the findings by Durant et al. ⁽¹⁷⁾. More recent studies, however, have not found a significant effect of body positioning in hemodynamic performance or response to VAE ^(18,19). Even the pathophysiology of right heart failure as a result of right outflow tract obstruction by air embolus has been questioned and myocardial ischemia attributable to severe arterial hypotension and impaired coronary blood flow has been advocated as a possible explanation for the detrimental cardiovascular effect of VAE ⁽¹⁹⁾. In the latter studies, however, all animals were initially studied in the supine position and not in the sitting position. Geissler et al. ⁽¹⁹⁾ showed, that supine dogs survived a bolus injection of air only when maintained in the initial supine position, as opposed to being repositioned into either a right or left lateral decubitus position. We believe that the recurrence of VAE after change from the sitting position to supine position in our study indicates that previously trapped air was released by changing patient position. The observation that microbubbles disappeared in all patients within 10 min of PEEP release or change of position is consistent with previous findings ⁽¹⁹⁾.

The fact that none of our findings have been reported previously may be attributable to the common practice of removing the TEE probe during the sitting position to facilitate repositioning of the patient. VAE after skin closure might be the reason for hemodynamic deterioration immediately after surgery. Although this did not occur in any of our patients, we cannot exclude that patients with no sign of intraoperative VAE are not prone to VAE during PEEP release or change of position alone.

Our findings have changed our anesthetic practice. We believe that the VAE recurrence during release of even a moderate PEEP is another argument against the use of PEEP during surgery at risk of VAE. The fact that the repositioning of the patient from the sitting position to the supine position was accompanied with VAE should lead to the extension of the TEE monitoring period until the patient is placed into the supine position.