Cardiovascular Anesthesia: Research Report
The safety of liver resection depends mainly on the control of bleeding during parenchymal resection. The Cavitron Ultrasonic Surgical Aspirator® (CUSA®; Valleylab Inc, Boulder, CO) has become a standard surgical tool for liver resection. Although controversy exists, the advantage of CUSA is reported to be the markedly reduced bleeding from the cut surface of the liver during surgery through aspiration of parenchymal tissue and ligation of the remaining Glisson’s structure (1).
Venous air embolism (VAE), a potential complication of surgery in the sitting position (2) or laparoscopy,(3,4) is unlikely to occur during laparotomy in a horizontal position. VAE was reported during several types of hepatic interventions, such as electrocauterization (5), argon-enhanced coagulation (6,7), water jet dissection (8), ultrasonic dissection (9), microwave coagulation therapy, and radiofrequency ablation.(10) However, the incidence of VAE using the CUSA® has not been determined.
This study was designed to compare the incidence and severity of air embolism between the clamp-crushing (CC) method and CUSA® during liver resection and to document whether these emboli were detected using transesophageal echocardiography (TEE). The changes in hemodynamic values, end-tidal CO2 partial pressure (Petco2) and arterial CO2 partial pressure (Paco2) were also compared.
The study protocol was approved by our IRB, and written, informed consent was obtained from all patients. The subjects of our study were 50 adults scheduled for elective hepatectomy from April 2003 to November 2003. None of them had known cardiopulmonary diseases. Patients with dysphagia, hiatal hernia, or esophageal disease were excluded because they had relative contraindications for TEE (SONOS 4500; Philips, Boeblingen, Germany). The patients were randomly assigned to either CC group or CUSA® groups. Randomization was performed by opening a sealed envelope before induction of anesthesia. In the CUSA® group (n = 25), hepatic resection was performed using CUSA®, and the parenchymal crushing technique with a Péan clamp (11) was performed in the CC group (n = 25). Inflow occlusion was not performed in either group. In the CUSA® group, suction pressure was 200–250 mm Hg and amplitude was 80 W. The disease and the type of operation in each group are shown in Table 1. The same surgeon performed all 50 of the operations. The patients received 2 mg of IV midazolam and 0.2 mg of IV glycopyrrolate preoperatively. After insertion of an epidural catheter for postoperative analgesia, anesthesia was induced with thiopental 5 mg/kg and fentanyl 100 μg. Neuromuscular relaxation was achieved with 0.6 mg/kg IV rocuronium followed by a continuous IV infusion of 2–5 μg·kg−1·min−1. Patients received 50% O2/50% air, with end-tidal sevoflurane concentration adjusted according to hemodynamic indications. The end-tidal CO2 partial pressure (Petco2) was maintained at 30–35 mm Hg. One liter of crystalloid solution was given preoperatively, and intraoperative fluid administration was adjusted to maintain a urine output ≥1 mL·kg−1·h−1. Stomach contents were aspirated via a nasogastric tube after tracheal intubation to enhance the TEE visualization. A 5.0-MHz multiplane TEE probe was then inserted. The TEE view was continuously monitored for cardiac chamber size, wall motion, gas entry, and valvular regurgitation with the patients in the supine position, except during the periods when a complete TEE examination was performed.
The complete TEE examination included the 4-chamber view to examine valvular function and intracardiac air. If air had entered the heart, a longitudinal view of the superior vena cava and inferior vena cava was obtained to document its pathway. Cardiovascular instability during the period of air entry was defined as a sudden decrease in mean arterial blood pressure >20 mm Hg or an acute episode of pulse oximetric saturation (Spo2 <90%). TEE images were videotaped for further analysis. To avoid interobserver variability, an independent cardiac anesthesiologist certified for echocardiography in our institution reviewed the tapes for air embolism and TEE interpretation. Air emboli were staged (Table 2) (12). Arterial blood gas analysis was performed whenever more than stage II VAE was observed with TEE. Otherwise, arterial blood gas analysis was obtained 30 min after the beginning of hepatic transection. If a larger amount of VAE occurred after that, arterial blood gas analysis was repeated. At that time, the average stage of VAE, arterial blood gas values, and hemodynamics were compared between the two groups.
Systolic and diastolic blood pressure, central venous pressure, Petco2, and Spo2 were monitored throughout the surgery. Blood loss was measured both during the transection and at the end of the surgery. Fresh packed red blood cells were transfused only when hematocrit decreased to <25%.
Data were expressed as mean ± sd. The Mann-Whitney U-test, Student’s t-test, Fisher’s exact test, and one-way repeated-measures analysis of variance were used to determine statistical differences between the two groups. A P value <0.05 was considered significant.
Differences in the demographic data and duration of surgery between groups were not significant (Table 3). There were no significant differences in hemodynamic, blood gas, or Petco2 values between the two groups. All these values were stable throughout the intervention (Table 4). There were no episodes of cardiac arrhythmia in either group and differences between groups for blood loss and transfusion were not statistically significant (Table 3).
The presence of foramen ovale was examined before the hepatic resection through TEE and no patient had patent foramen ovale.
During hepatectomy with the CC method, no air embolism was detected by two-dimensional TEE in 32.0% of the patients (Fig 1). For those patients who had any VAE, all the emboli were smaller than half the diameter of the right atrium (RA) or right ventricle (RV) (Fig. 2).
During liver resection using CUSA®, air emboli were detected in all the patients (Fig. 1). The detected emboli were classified into 4 different stages, and 20.0%, 36.0%, 36.0%, and 8% of the patients in the CUSA® group were in stage I, II, III and IV respectively (Fig. 1). In the CUSA® group, 44% of patients showed air emboli that filled more than half the diameter of the RA or RV. The stage of VAE was significantly more advanced in the CUSA® group than in the CC group (Mann-Whitney U-test, U = 214.00, z = −4.25, P < 0.0001).
The right heart cavities were cleared of all air emboli within seconds; no air emboli were noted in the left heart cavities in any cases. There was no correlation between episodes of VAE and blood gas variables or Petco2. There was no TEE evidence of patent foramen ovale (PFO). No postoperative neurologic deficits were observed.
This randomized trial using two-dimensional TEE, which is the most sensitive method for the detection of VAE (12,13), revealed venous air embolism in all patients during hepatectomy using CUSA®. Higher stage VAE were observed in the CUSA® compared with the CC group. There was no direct correlation observed between the VAE and cardiorespiratory events.
CUSA®, combined with bipolar cautery and a saline irrigation system, allows hepatic parenchyma resection from the anterior surface of the liver in a bloodless field without retracting the hepatic lobe or occluding inflow vessels at the hepatic hilum (the anterior approach) (14). Aspiration of parenchymal tissue and ligation of the remaining Glisson’s structures have markedly reduced bleeding from the resection surfaces. This technique has also decreased morbidity in patients who have a smaller hepatic functional reserve (1). Although CUSA® has made hepatic resection much easier, successive ligation of the fragile peripheral Glisson’s tissue and the extremely thin hepatic veins is time consuming and can be unsuccessful. Insufficient ligation is still a frequent cause of late bleeding and bile leakage. Tearing the small vessels causes oozing from the cut surfaces (1). Our results were consistent with those of Takayama et al. (15), who showed that ultrasonic dissection offered no reduction in blood loss compared with the CC method for transection of the liver.
VAE have been reported during several types of hepatic interventions (5–10). During hepatic resection, vena cava manipulation or compression may narrow lumen diameter at its junction with the hepatic veins. In such a situation, the venous distending pressure of the constricted portion of the inferior vena cava may be less than in the nonconstricted part, and it may even become subatmospheric when blood passes through the narrowed portion with a rapid flow rate. Air could be sucked into the inferior vena cava via the large number of small hepatic veins exposed to the atmosphere (5). Evaporation of gases, including nitrous oxide dissolved in the blood and tissue, could be a cause of gas bubbles produced during microwave coagulation therapy or radiofrequency ablation (10).
Our results showed that there were no significant changes in arterial blood pressure, central venous pressure, Petco2, or blood gas variables even when a large amount of VAE occurred. The finding that all of the embolic episodes in this study were not clinically significant was consistent with results of other studies that used TEE to detect gas emboli during laparoscopic surgery (16,17). Farges et al. (18) have reported that none of 21 patients who underwent laparoscopic liver resection experienced gas embolism, but they monitored Petco2, oxygen saturation, and transesophageal Doppler in only four patients. Therefore, their study cannot persuade us that VAE does not occur during liver resection.
To produce a significant hemodynamic effect detectable by means other than TEE, a large amount of air must enter the venous circulation (19). Such severe VAE are rare. The current study does reveal, however, that many episodes of VAE occur during open hepatic resection using CUSA®.
Many of the patients who undergo hepatectomy have liver cirrhosis. VAE is particularly dangerous in such patients because 15%–45% of them have pulmonary abnormalities, including intrapulmonary shunting caused by pulmonary vascular dilation and arteriovenous communication (20,21). In these patients, paradoxical emboli can occur during air embolism even if intracardiac abnormalities are not present (9,10).
Paradoxical air emboli are more likely to occur in those patients having a probe-PFO, especially when the normal transatrial (left > right) pressure gradient is reversed. Reversal of this gradient is favored by hypovolemia and, perhaps, by positive end-expiratory pressure. Two autopsy studies have determined incidences of PFO. The first study (n = 1100) revealed a 29% incidence of a “probe” PFO (0.2 cm to 0.5 cm maximum dimension) and a 6% incidence for “pencil” PFO (0.6 cm to 1.0 cm) (22). The second study (n = 965) revealed a PFO incidence of 27.3%, with the PFOs varying in size from 1 mm to 19 mm.(23) Fortunately, PFO was not found in our patients.
In conclusion, all the patients undergoing resection of the liver using the CUSA® had VAE, and the amount of the VAE was larger in the CUSA® group than in the CC group. Therefore, we suggest the following recommendations for hepatectomies using the CUSA®: the minimal required duration for the application of the CUSA® should be used; precaution should be taken for harmful VAE, particularly for those patients with the potential for intracardiac right to left shunt, such as patients having PFO or liver cirrhosis; nitrous oxide should not be used for these procedures; central venous access, which may allow the aspiration of entrained air, should be considered; and appropriate monitors, including TEE, should be used in patients having extensive resection.
1. Yamamoto Y, Ikai I, Kume M, et al. New simple technique for hepatic parenchymal resection using a Cavitron Ultrasonic Surgical Aspirator® and bipolar cautery equipped with a channel for water dripping. World J Surg 1999;23:1032–7.
2. Papadopoulos G, Kuhly P, Brock M, et al. Venous and paradoxical air embolism in the sitting position: a prospective study with transesophageal echocardiography. Acta Neurochir 1994;126:140–3.
3. Diakun TA. Carbon dioxide embolism: successful resuscitation with cardiopulmonary bypass. Anesthesiology 1991;74:1151–3.
4. Hashizume M, Takenaka K, Yanaga K, et al. Laparoscopic hepatic resection for hepatic cellular carcinoma. Surg Endosc 1995;9:1289–91.
5. Hatano Y, Murakawa M, Segawa H, et al. Venous air embolism during hepatic resection. Anesthesiology 1990;73:1282–5.
6. Veyckemans F, Michel I. Venous air embolism from an argon coagulator. Anesthesiology 1996;85:443–4.
7. Palmer M, Miller CW, van Way III, CW Orton EC. Venous gas embolism associated with argon-enhanced coagulation of the liver. J Invest Surg 1993;6:391–9.
8. Smith JA. Possible venous air embolism with a new water jet dissector. Br J Anaesth 1993;70:446–7.
9. Lee SY, Choi BIW, Kim JS, Park KS. Paradoxical air embolism during hepatic resection. Br J Anaesth 2002;88:136–8.
10. Nakayama R, Yano T, Mizutamari E, et al. Possible pulmonary gas embolism associated with localized thermal therapy of the liver. Anesthesiology 2003;99:227–8.
11. Cunningham JD, Fong Y, Shriver C, et al. One hundred consecutive hepatic resections: blood loss, transfusion and operative technique. Arch Surg 1994;129:1050–6.
12. Schmandra TC, Mierdl S, Bauer H, et al. Transoesophageal echocardiography shows high risk of gas embolism during laparoscopic hepatic resection under carbon dioxide pneumoperitoneum. Br J Surg 2002;89:870–6.
13. Glenski JA, Cucchiara RF, Michenfelder JD. Transesophageal echocardiography and transcutaneous O2 and CO2 monitoring for detection of venous air embolism. Anesthesiology 1986;64:541–5.
14. Hodgson WJB, Morgan J, Byrne D, DelGuercio LRM. Hepatic resection for primary and metastatic tumors using the ultrasonic surgical dissector. Am J Surg 1992;163:246–50.
15. Takayama T, Makuuchi M, Kubota K, et al. Randomized comparison of ultrasonic versus clamp transection of the liver. Arch Surg 2001;136:922–8.
16. Derouin M, Boudreault D, Couture P, et al. Detection of CO2 venous embolism during laparoscopic surgery. Anesthesiology 1994;81:A560.
17. O’Sullivan DC, Micali S, Averch TD, et al. Factors involved in gas embolism after laparoscopic injury to inferior vena cava. J Endourol 1998;12:149–54.
18. Farges O, Jagot P, Kirstetter P, et al. Prospective assessment of the safety and benefit of laparoscopic liver resections. J Hepatobiliary Pancreat Surg 2002;9:242–8.
19. Dion YM, Levesque C, Doillon CJ. Experimental carbon dioxide pulmonary embolization after vena cava laceration under pneumoperitoneum. Surg Endosc 1995;9:1065–9.
20. Krowka MJ, Tajik AJ, Dikson ER, et al. Intrapulmonary vascular dilatations (IPVD) in liver transplant candidates: screening by two-dimensional contrast-enhanced echocardiography. Chest 1990;97:1165–70.
21. Hopkins WE, Waggoner AD, Barzilai B. Frequency and significance of intrapulmonary right-to-left shunting in end-stage hepatic disease. Am J Cardiol 1992;70:516–9.
22. Thompson T, Evans W. Paradoxical embolism. Q J Med 1930;23:129–35.
© 2005 International Anesthesia Research Society
23. Hagen PT, Scholz DG, Edwards WD. Incidence and size patent foramen ovale during first 10 decades of life: an autopsy study of 965 normal hearts. Mayo Clin Proc 1984;59:17–20.