The timing of tracheal extubation in patients undergoing large intraoperative procedures is controversial. The use of mechanical postoperative ventilation in these patients is increasingly being questioned . Cost containment, better use of personnel, and a reduction in some postoperative complications are cited as advantages in support of early extubation .
Advances in selection and management of liver transplant recipients have improved patient survival and quality of life . However, perioperative expenses can be a significant economic burden. While early extubation has reduced perioperative costs associated with other large operative procedures , tracheal extubation of patients immediately after liver transplantation has not been evaluated. Provided this is a safe and cost-effective intervention, clinical guidelines for candidate selection need to be established. In this preliminary study we compare the experience at two centers using different clinical strategies to safely extubate the tracheas of patients after liver transplantation.
The two study centers were the University of Colorado Health Sciences Center (UC) and the University of California at San Francisco (UCSF). Both centers use risk stratification defined by the United Network of Organ Sharing (UNOS) which assigns each patient a status code indicating the degree of medical urgency for transplantation (1): status 1, life expectancy less than 7 days due to acute and chronic liver failure; status 2, patients require continuous hospitalization in an acute care bed for at least 5 days; status 3, patients require ongoing interactions with health care system without continuous hospitalization; and status 4, patients at home and functioning normally.
(1) United Network for Organ Sharing. Liver allocation: degree of medical urgency. Policy 3.6, section 3.6.4; June 1996.
Before extubation studies were initiated, the anesthesiology team at UC established clinical guidelines based on previous experience. We examined the records of all liver transplant patients for 1994 who were postoperatively ventilated less than 8 h and discharged from the intensive care unit (ICU) in less than 48 h, reasoning that this was a group representing uncomplicated postoperative outcome. Twelve of 44 patients who received transplants in 1994 met these criteria.
All 12 patients were less than 50 yr of age, UNOS Status 3, and free of coexistent disease and prior encephalopathy. During surgery, all received less than 10 U of packed red cells and had good donor graft function. A stable base deficit, normal electrolytes, and normalization of coagulation were considered indicative of adequate graft function . At the conclusion of surgery, the alveolar-arterial oxygen gradient was less than 150 mm Hg and none of these 12 patients required vasoactive support.
After investigational review board approval, the 1994 clinical observations summarized in Table 1 served as guidelines to determine whether a patient was eligible for immediate postoperative tracheal extubation in 1995. The 12 patients with uncomplicated postoperative outcome in 1994 served as the cost comparison group for the prospective study at UC. The number and cost of ICU services were compared between the two groups using an unpaired Student's t-test.
In a separate study at UCSF, a trial of immediate extubation of liver transplant recipients was conducted during the same calendar year. Criteria used at UC were unknown to investigators at UCSF. Patient candidates at UCSF were selected based on clinical judgment by the attending anesthesiologist at the conclusion of surgery. The records of those patients were retrospectively evaluated and are compared here with prospectively studied patients at UC.
At both centers anesthetic technique during liver transplantation was determined by the attending physician. We noted that less benzodiazepines and narcotics were administered to patients at UC whose tracheas were extubated less than 8 h after surgery in 1994 than to those patients who were ventilated for a longer period. Therefore, to ensure comparability at UC between the 1994 control group and the 1995 study group, we used the following anesthetic regimen: After rapid sequence induction of anesthesia with thiopental (5 mg/kg) and succinylcholine (1 mg/kg), isoflurane or desflurane were used for maintenance anesthesia. No more than 25 micro g/kg of fentanyl and 2 mg of lorazepam were administered as intravenous adjuvants. No other sedative-hypnotic drugs were administered. Neuromuscular blockade with doxacurium was intermittently assessed by peripheral neuromuscular stimulator to maintain one twitch of a train-of-four response after reperfusion of the donor graft.
Patients who met all predetermined intraoperative criteria (Table 1), were prepared for emergence and tracheal extubation. Patients were given neostigmine (0.05 mg/kg) and glycopyrrolate (0.01-0.02 mg/kg) for neuromuscular reversal and the inhaled anesthetic was discontinued. Patients who did not rapidly awaken were given incremental doses of naloxone to a maximum of 0.2 mg or flumazenil to a maximum of 5 mg at the discretion of the anesthesiologist. When patients were awake and able to follow commands, the usual criteria for extubation included a positive gag response, tidal volume more than 8 mL/kg, respiratory rate less than 20 breaths/min, normocarbia as judged by end-tidal carbon dioxide analysis, and clinical evidence of neuromuscular reversal. Patients were tracheally extubated with the head elevated and transferred to the ICU with oxygen given by nasal cannula at 2 L/min.
At UCSF, the usual intraoperative requirements for immediate postoperative extubation included good donor graft function and hemodynamic stability at the conclusion of surgery. Hepatic graft function was assessed as described above. No standardized anesthetic regimen was used for patients who appeared to be candidates for early tracheal extubation.
During 1995, 16 of 67 liver transplant patients at UC and 25 of 106 patients at UCSF were tracheally extubated immediately postoperatively. Surgical time averaged 6-8 h at both institutions. There were no reintubations at UC. Two of 25 patients at UCSF required reintubation. Neither of these patients met all the UC criteria: one was reintubated within 30 min due to hypoventilation and decreased level of consciousness. This patient was UNOS status 1 with a history of recent encephalopathy (Table 2).
The second patient was reintubated within 10 min due to clinical impression of respiratory failure. Postreperfusion coagulopathy suggested inadequate donor graft function. After graft reperfusion, administration of blood products resulted in a central venous pressure of 18 cm H2 O. During this time, the alveolar-arterial gradient widened as the partial pressure of arterial oxygen decreased from 287 mm Hg to 94 mm Hg while breathing 50% oxygen. Both patients were ventilated for 12-14 h and subsequently extubated without problem.
Of the remaining 23 UCSF patients, 17 did not meet all of the UC criteria. Seven patients were over 50 yr old, UNOS Status 2 was present in 11 patients, 2 patients required more than 10 U of packed red cells, and 4 had coexistent disease. However, none of these 17 patients required reintubation (Table 2).
The 1994 historical control patients at UC were ventilated for an average of 6 h with a range of 1-8 h (Table 3). A comparison of hospital services showed a statistically significant reduction in the length of stay and use of arterial blood gases for patients extubated in 1995 compared to the control group. Acuity levels of nursing were reduced for extubated patients, averaging 2 h compared to 8 h for control patients. An average cost saving of $2709 was associated with reduced ICU services in patients tracheally extubated immediately postoperatively.
The use of conservative criteria based on our prior experience at UC resulted in successful immediate postoperative extubation of the trachea in approximately one quarter of our liver transplantation patients. The experience at UCSF using less structured criteria was similar. As isolated criteria, age greater than 50 yr and UNOS status 2 did not preclude successful immediate postoperative extubation in this preliminary study.
Liver transplant recipients are usually ventilated for an average of 18 to 36 hours postoperatively [3-5]. Siliciano  suggested that postoperative positive pressure ventilation with sedation may decrease surgical stress response and improve hemodynamic stability, facilitating early recovery in patients after large operative procedures. However, accentuated pulmonary impedance due to lung inflation increases right ventricular workload . Backward flow into the inferior vena cava and hepatic veins from associated tricuspid regurgitation produces venous congestion . The reduction in splanchnic blood flow may pose special problems for the transplant recipient by impairing graft function [8,9]. Therefore, mechanical ventilation may not be of benefit in all liver transplant patients.
In spontaneously breathing patients, reduced intrapleural pressure improves venous return, increasing cardiac end-diastolic volume, cardiac output, and hepatic blood flow [1,10]. Thus, spontaneous ventilation may be beneficial in the hemodynamically stable transplant patient, promoting hepatic venous drainage and donor graft circulation .
Early extubation protocols have been used successfully in other large surgical procedures [11-14]. Revascularization studies in particular have demonstrated the importance of identifying patient characteristics associated with uncomplicated outcome [1,11]. Other early extubation studies using selected low-risk patient populations have been completed for esophagectomy , transsternal thymectomy , and abdominal aortic reconstruction patients . These studies have shown that carefully selected patients channeled into institutional protocols designed to facilitate early extubation have favorable outcomes [1,11,12,13,15].
Preoperative criteria for early extubation in most protocols have concentrated on age, coexistent illness, urgency of surgery, and severity of illness at the time of operation [1,11-13,15]. The prospective study at UC used similar characteristics. Encephalopathy was added as an exclusion criterion since it has been associated with delayed awakening and prolonged postoperative ventilation after liver transplantation . UNOS status, reflecting both severity of illness and urgency of surgery, is a well accepted prognostic tool . Although well defined, UNOS status is based on subjective evaluation and is not always uniformly scored. Different approaches to determining UNOS status assessment may have affected patient selection at the two institutions. Accepting this limitation, it appears that UNOS status 2 and age greater than 50 yr may not independently preclude immediate postoperative tracheal extubation in liver transplant recipients.
Intraoperative criteria, which have been used to predict outcome, were added to the protocol to further identify a potential population of patients at low risk of respiratory failure after extubation. Donor graft function is one of the most important criteria independently determining perioperative morbidity and mortality . The importance of this factor in predicting immediate postoperative extubation outcome is suggested after reintubation of the second patient who had intraoperative evidence of poor graft function. Blood loss  is both a dependent and independent predictor of outcome during liver transplantation while hemodynamic stability, an expression of multiple physiological events, reflects overall patient and graft well being. An increased alveolar-arterial oxygen gradient has been used to predict postextubation respiratory failure [13,19,20] and was therefore added to the UC intraoperative criteria.
Immediately extubated patients experienced a shorter stay in the ICU. Consequently, the need for ancillary services compared to ventilated patients was reduced, resulting in an economic advantage similar to that in other studies [1,11,12,21]. Cost reduction was apparent when extubated patients were compared to those ventilated for an average of only six hours. The cost benefit would have been much greater if compared to longer elective ventilation times. Total length of stay was not studied because the postoperative course of transplant patients can be complicated by episodes of rejection, increasing the length of hospitalization, and there is probably no relationship between extubation and rejection.
The results at both institutions suggest that immediate extubation of selected liver transplant patients is safe and cost effective. It appears that acceptable guidelines for immediate postoperative extubation lies between the experience of the two centers. None of the patients who met the criteria used at UC required reintubation. The UCSF data suggests age greater than 50 yr and UNOS status 2 may not preclude successful immediate postoperative extubation in all liver transplant patients. In contrast, a history of encephalopathy and poor graft function are probably contraindications to extubation.
Because immediate postoperative tracheal extubation of selected liver transplant recipients appears to be safe and cost effective, more extensive studies are warranted to help identify patient and surgical factors that predict successful extubation. The prognostic value of criteria used at UC and the effects of a standardized anesthetic technique on the success of immediate postoperative extubation will require further testing. Presumably smaller doses of anesthetics may facilitate immediate postoperative extubation in selected patients, but this is not clear from the present study. Optimally, all patient or surgical factors that may influence the selection of extubation candidates will require large-sample statistical analysis.
1. Higgins JE. Pro: early endotracheal extubation is preferable to late extubation in patients following coronary artery surgery. J Cardiothorac Vasc Anesth 1992;6:488-93.
2. Michel BC, Van Hout BA, Bonsel GJ. Assessing the benefits of transplant services. Baillieres Clin Gastroenterol 1993;8:411-23.
3. Carton EG, Plevak DJ, Kranner PW, et al. Perioperative care of the liver transplant patients. Part 2. Anesth Analg 1994;78:382-99.
4. Plevak DJ, Sourthorn PA, Narr BJ. Intensive care unit experience in the Mayo liver transplant program: the first 100 cases. Mayo Clin Proc 1989;64:433-45.
5. Muralidahar V, Jayalaxmi TS. Anaesthesia for liver transplantation: perioperative problems and management. Trop Gastroenterol 1994;4:191-203.
6. Siliciano D. Con: early extubation is not preferable to late extubation in patients undergoing coronary artery surgery. J Cardiothorac Vasc Anesth 1992;6:494-8.
7. Dorinsky PM, Hamlin RL, Gadek JE. Alterations in regional blood flow during positive end-expiratory pressure ventilation. Crit Care Med 1987;15:106-12.
8. Jullien T, Valtier B, Hongnat JM, et al. Incidence of tricuspid regurgitation and vena caval backward flow in mechanically ventilated patients. A color Doppler and contrast echocardiographic study. Chest 1995;107:488-93.
9. Rossaint R, Slama K, Jaeger M, et al. Fluid restriction and early extubation for successful liver transplantation. Transplant Proc 1990;22:1533-4.
10. Ben-Haim SA, Amar R, Shofty R, Dinnar U. Low positive end expiratory pressures improve the left ventricular workload versus coronary blood flow relationship. J Cardiovasc Surg 1991;32:239-45.
11. Gross SB. Early extubation: preliminary experience in the cardiothoracic patient population. Am J Crit Care 1995;4:262-6.
12. Caldwell MT, Murphy PG, Page R, et al. Timing of extubation after oesophagectomy. Br J Surg 1993;80:1537-9.
13. Gorback MS, Moon RE, Massey JM. Extubation after transsternal thymectomy for myasthenia gravis: a prospective analysis. South Med J 1991;84:701-6.
14. Shackford SR, Virgilio RW, Peter RM. Early extubation versus prophylactic ventilation in the high-risk patient: a comparison of postoperative management in the prevention of respiratory complications. Anesth Analg 1981;60:76-80.
15. Quasha AL, Loeber N, Feeley TW, et al. Postoperative respiratory care: a controlled trial of early and late extubation following coronary artery bypass. Anesthesiology 1980;52:135-41.
16. Spanier TB, Klein RD, Nasarway SA, et al. Multiple organ failure after liver transplantation. Crit Care Med 1995;23:466-573.
17. Kamath GS, Plevak DJ, Weissner RH, et al. Primary nonfunction of the liver graft. When should we retransplant? Transplant Proc 1991;23:1954.
18. Shaw BW Jr, Wood RP, Gordon RD, et al. Influence of selected patient variables and operative blood loss on six-month survival following liver transplantation. Semin Liver Dis 1985;5:384-93.
19. Rosenthal MH. Evaluation and management of postoperative respiratory insufficiency and hypoxemia. Annual refresher course lectures. Atlanta: American Society of Anesthesiologists, 1995;532:1-4.
20. Whelan J, Simpson SQ, Levy H. Unplanned extubation. Predictors of successful termination of mechanical ventilatory support. Chest 1994;105:1808-12.
21. Laussen PC, Reid RW, Stene RA, et al. Tracheal extubation of children in the operating room after atrial septal defect repair as part of a clinical practice guideline. Anesth Analg 1996;82:988-93.