Isoflurane

Study objectives: To evaluate the effect of perioperative thoracic epidural analgesia followed by postoperative epidural analgesia compared with conventional IV anesthesia on the occurrence of late postoperative hypoxemia in patients undergoing elective coronary bypass graft (CABG) surgery.

Design: Randomized controlled trial.

Setting: Cardiac surgery unit at a university hospital.

Patients: A total of 50 patients undergoing elective CABG surgery.

Intervention: Patients were randomly assigned to receive either conventional IV anesthesia (CON) or general anesthesia combined with thoracic epidural anesthesia followed by postoperative epidural analgesia (TEA) with bupivacaine. Postoperatively, the patients were monitored in the surgical ward with a pulse oximeter for a total of two postoperative nights (the second and third postoperative nights).

Measurements and results: The overall incidence of episodic hypoxemia was 56% (28 of 50 patients) on the second postoperative night and 89% (41 of 46 patients) on the third postoperative night. More than 30 episodes of hypoxemia developed on the second night in 22% of patients (11 of 50 patients), and on the third night in 30% of patients (14 of 46 patients). Despite oxygen therapy, 7% of patients (3 of 46 patients) experienced constant hypoxemia on the third night. In general, hypoxemia seemed to be slightly worse on the third postoperative night compared with the second postoperative night. Significantly more patients in the TEA group (25 of 25 patients) experienced episodic hypoxemia on the third postoperative night compared with the CON group (16 of 21 patients; p < 0.05). Otherwise, there were no significant differences between the two regimens.

Conclusions: Both episodic and constant hypoxemia were common in the late postoperative period in patients on the ward after CABG surgery with no clinically significant intergroup differences. Thus, perioperative epidural anesthesia/analgesia combined with postoperative epidural anesthesia/analgesia was not protective against hypoxemia, and therapy with opioids did not seem to be of importance for the occurrence of late postoperative hypoxemia on nights 2 and 3 after CABG surgery.

Key words: bupivacaine: coronary artery bypass grafting; postoperative complication; postoperative hypoxemia; prevention; pulse oximeter; thoracic epidural analgesia

Abbreviations: CABG = coronary artery bypass graft; CON = conventional IV anesthesia; REM = rapid eye movement; TEA = thoracic epidural anesthesia/analgesia

**********

Increased cardiac morbidity after various surgical procedures has been found in relation to late postoperative hypoxemia, (1-4) and several studies (1-4) have described a time-related connection between late postoperative hypoxemia and myocardial ischemia after noncardiac surgery. Further, it has been shown in animal experiments that arterial hypoxemia can worsen ischemia in an already ischemic myocardium. (5) In addition to ischemia, there may be a connection between hypoxemia and cardiac arrhythmias like atrioventricular blockade, ventricular ectopy, and periodic tachycardia. (1) Thus, the data suggest that postoperative hypoxemia may be a risk factor for cardiac complications after major abdominal and thoracic (noncardiac) surgery. (6) No data, however, are available on the occurrence of episodic hypoxemia on the second and third nights after coronary artery bypass graft (CABG) surgery. The effect of different postoperative analgesic techniques on late postoperative hypoxemia has been investigated in only a limited number of studies, and the incidence of hypoxemia after cardiac surgery is unknown. Regional anesthetic techniques, like thoracic epidural anesthesia/analgesia (TEA), may reduce the surgical stress response and has been suggested to induce less frequent sleep disturbances postoperatively. (7,8) We hypothesize that TEA may reduce the incidence of postoperative hypoxemia. Thus, the aim of this study was to investigate the effect of thoracic epidural analgesia on the occurrence of late postoperative hypoxemia in patients undergoing elective CABG.

MATERIALS AND METHODS

Approvals

The study was approved by the local ethics committee. Patients were included in the study after giving written informed consent.

Study Protocol

Patients were randomly assigned to receive either conventional IV anesthesia (CON) or general anesthesia combined with TEA. The randomization list was generated from a table of random numbers.

Patients

Sixty-six patients who were scheduled for elective CABG surgery were screened. The criteria for inclusion in the study were age > 18 years, sinus rhythm on the ECG preoperatively, and written and oral informed consent. The criteria for exclusion from the study were oral anticoagulation and coagulopathy. Patients were included consecutively in the study from January 3, 2000, to December 12, 2000. Patients who could not be extubated and moved to a stationary ward within the first 24 h after surgery were excluded from the study. Of the 66 possible patients, 55 patients were included in the study as 11 patients did not give informed consent. Five patients (one from the CON group and four from the TEA group) were excluded from the study because they had received prolonged mechanical ventilation. The study therefore included 50 patients for data analysis (TEA group, 26 patients; CON group, 24 patients). Four patients withdrew consent after the second postoperative night. Thus, 25 patients in the TEA group and 21 patients in the CON group remained in the study on the third postoperative night.

For patients undergoing CABG surgery, the incidence of episodic hypoxemia was not known before designing the study. Based on studies dealing with noncardiac patients, in which the incidence of patients with episodic hypoxemia was approximately 60%, the number of patients needed for investigation was 84 ([alpha] = 0.05; [beta] = 0.80; 50% reduction in incidence). Because of a time limit and a lack of success including patients in the study, this number was not reached. A total of 50 patients were included in the study, a number that was accepted because of the uncertain preconditions for the estimation of the level of patients needed for investigation.

Primary End Points

Episodic hypoxemia was defined as a drop in arterial oxygen saturation below 90% or a drop of > 4% from baseline in which the decline occurred within a maximum period of 2 min. (1,8) Constant hypoxemia was defined as a mean oxygen saturation of < 90% during the period of observation. (9)

Oximetry Monitoring

The patients were monitored in the surgical ward with a pulse oximeter on the second and the third postoperative nights. The period of monitoring was from 11:00 PM to 7:00 AM the following morning. We used a pulse oximeter (N-3000; Nellcor Inc; Pleasanton, CA) with adhesive finger probes on the second or third finger. All of the previous data in the oximeter were reset before a new measuring period began. The display of the pulse oximeter was hidden from both the patient and the department staff, and all acoustic alarms were turned off. The patients received oxygen (2 to 3 L/rain) via a nasal catheter during the periods of monitoring according to the routine of the department. Data from the pulse oximeter were downloaded to a personal computer, and the data analyses were blinded in relation to any clinical information.

TEA Group

The day before surgery, an epidural catheter was inserted at the Th1-Th3 level by the median approach using the hanging drop technique. Correct placement was tested with 2 mL of 2% lidocaine. On arrival in the operating room, epidural analgesia was induced with 8 to 10 mL of bupivacaine, 5 mg/mL. The spread of blockade was evaluated with ice after 15 min. An extension from at least Th1 to Th8 was accepted. Anesthesia was induced with IV midazolam, 3 to 5 mg, fentanyl, 0.3 mg, and a dose of pancuronium, 0.1 mg/kg. Anesthesia was maintained with isoflurane and a continuous epidural infusion of bupivacaine, 1.25 mg/mL, with morphine, 25 [micro]g/mL and 5 mL/h. Top-up bolus doses of 4 mL of bupivacaine, 5 mg/mL, were administered hourly during the operation. The administration of fentanyl was restricted to the dose given at the induction of general anesthesia. Postoperatively, the epidural infusion continued, and bolus doses of 4 mL of 0.25% bupivacaine were administered as needed. The efficacy of the epidural blockade was tested on a daily basis and recorded. All patients in the TEA group had a detectable blockade, and top-up bolus doses were administered to achieve a spread of blockade from at least Th1 to Th8. The catheter was removed on postoperative day 4 or 5. The coagulation status was not controlled at this point. (10)

CON Group

Anesthesia was induced with midazolam, 3 to 5 rag, fentanyl, 15 to 30 [micro]g/kg, and pancuronium, 0.1 mg/kg, which were administered at induction only and not repeated. Anesthesia was maintained with isoflurane. Analgesia in the CON group was provided with intermittent IV morphine boluses (2.5 to 5 mg) until the morning of the first postoperative day when morphine, 5 to 10 rag, orally were given as needed. Both groups received their usual medication preoperatively on the day of surgery, followed by diazepam, 10 mg orally, for premeditation. Postoperatively, all patients received oral paracetamol, 1 g every 6 h.

Statistical Analysis

The Wilcoxon signed rank test, Fisher exact test, [chi square] test, and McNemar test were used to test the difference between paired observations in the different patient groups, while the Mann-Whitney U test and the Fisher exact test were used for comparison between the groups. The analyses were performed using a statistical software package (SPSS, version10.0; SPSS; Chicago, IL). A p value of < 0.05 was regarded as being significant.

RESULTS

Patient characteristics and the duration of the period of study admittance are shown in Table 1. There were no statistically significant differences between the two groups.

Oxygenation results are shown in Tables 2 to 4. For all patients, the mean and the lowest oxygen saturation levels were significantly lower during the third postoperative night than during the second postoperative night. In the CON group, the number of hypoxemic episodes was significantly higher on the third postoperative night compared with the second night. There was no significant difference between the two nights, either for the number of patients with constant or episodic hypoxemia or for the number of patients who experienced > 30 episodes per night, except among patients in the TEA group, who experienced more hypoxemic episodes on the third postoperative night compared with the second night.

When comparing the TEA and the CON groups, there were no statistically significant differences with respect to the mean and lowest arterial oxygen saturation levels or the number of hypoxemic episodes on the second postoperative night. Likewise, there was no significant difference in the number of patients who experienced > 30 episodes of hypoxemia per night. On the third postoperative night, 16 of the 21 patients in the CON group had episodic hypoxemia compared to 25 of 25 patients in the TEA group who experienced at least one hypoxemic episode (p < 0.05).

The overall incidence of episodic hypoxemia was 89% (41 of 46 patients) on the third postoperative night and 56% (28 of 50 patients) on the second postoperative night. More than 30 episodes of hypoxemia occurred in 30% of the patients (14 of 46 patients) on the third postoperative night and in 22% of the patients (11 of 50 patients) on the second postoperative night. There was no constant hypoxemia the second postoperative night, whereas 7% of patients (3 of 46 patients) experienced constant hypoxemia on the third postoperative night.

DISCUSSION

We have demonstrated a high incidence of episodic hypoxemia and also some degree of constant hypoxemia despite oxygen therapy on the second and third postoperative nights after uncomplicated CABG surgery. For some of the parameters, hypoxemia seemed to be slightly worse on the third postoperative night compared with the second postoperative night.

Only three patients experienced constant hypoxemia. This is not surprising, as all patients were treated with postoperative nasal oxygen therapy according to the routine of the department. A previous study in which the use of oxygen therapy was compared with the breathing of atmospheric air using a Hudson mask in patients who had undergone hip surgery showed that oxygen therapy reduced constant hypoxemia on the second postoperative night, but also that oxygen therapy had no effect on the number of hypoxemic episodes. (11) Correspondingly, in the present study, we found many hypoxemic episodes despite the application of oxygen therapy.

Late postoperative hypoxemia has been a well-known phenomenon for many years. In general, two kinds of hypoxemia are in focus; namely, episodic hypoxemia and constant hypoxemia. Episodic hypoxemia is most likely to occur during the second and third postoperative nights, and the mean oxygen saturation (constant hypoxemia) has the lowest level during the second postoperative night. This condition has been shown to occur in patients both after major abdominal surgery and after noncardiac thoracotomy. (1,2,8,9) Thus, among 78 patients who underwent major abdominal surgery without postoperative oxygen therapy, 41% had a mean oxygen saturation level during the second postoperative night that was < 90%, 38% experienced episodes with sudden desaturation to < 80%, and 23% experienced > 30 episodes of sudden desaturation. (11)

Until now, no studies have evaluated the incidence and severity of episodic hypoxemia on the second and third nights after CABG surgery. A previous study (12) using blood gas analysis (thus evaluating constant hypoxemia) showed severe hypoxemia (Pa[O.sub.2], < 60 mm Hg) in 31% of patients on the first postoperative day, in 50% on the second postoperative day, and in 40% on the third postoperative day.

Our study did not indicate a reduction in hypoxemia when using intraoperative anesthesia/analgesia combined with postoperative TEA with bupivacaine compared with traditional general anesthesia with postoperative pain relief based on therapy with opioids, although significantly more patients in the TEA group experienced episodic hypoxemia on the third postoperative night compared with those in the CON group. On the third postoperative night, six patients in the TEA group experienced only a single episode of hypoxemia (Table 3). These six patients therefore weigh heavily when the two patient groups are compared with respect to the number of patients with episodic hypoxemia that is given in Table 2. The data therefore suggest no clinically significant difference between the patient groups regarding late postoperative hypoxemia.

The perioperative use of TEA may provide both intraoperative and postoperative benefits. There is evidence to indicate improved myocardial oxygen supply and hemodynamic stability due to perioperative sympatholysis, [13] and a recent metaanalysis by Liu et al (14) assessed the effects of perioperative TEA on outcome after CABG surgery. The incidence of mortality or myocardial infarction were not affected, and TEA produced reduced risks of pulmonary complications and atrial fibrillation, reduced time to tracheal extubation, and reduced pain scores. (15,16) Still, instrumentation of the epidural space in patients undergoing anticoagulation therapy remains controversial because it may lead to unrecognized bleeding and catastrophic hematoma formation. (17) The first case report of an epidural hematoma associated with TEA and cardiac surgery with cardiopulmonary bypass was published only recently. (18)

However, the epidural catheter was inserted following the induction of anesthesia, and the patient was ambulating without difficulty on the first postoperative day after an uneventful immediate postoperative course. Contributing factors in this case included the initiation of IV heparin prosthetic valve thromboprophylaxis, the use of alteplase (a thrombolytic drug) to flush a dysfunctional IV catheter, thrombocytopenia, and the removal of the epidural catheter while the activated partial thromboplastin time was prolonged. However, the neurologic risk remains inadequately defined. To date, the scientific literature has reported the preoperative placement of > 6,000 epidural catheters in patients undergoing cardiopulmonary bypass. In most of these reports, the catheter was placed on the day preceding surgery. This probably confers a maximum risk of 1 in 1,500 eases (minimum risk, 1 in 150,000). (19) The use of TEA in cardiac surgery patients remains controversial, although the risk is small compared with that of the surgery and when proper care is taken. (10)

Episodic hypoxemia is primarily caused by episodes of apnea and hypoventilation. (20) Constant hypoxemia depends primarily on lung volume, and a decrease in lung volume in the late postoperative period may therefore worsen episodic desaturation during a given apnea. Further, the maximum decrease in functional residual capacity occurs later than 16 h after surgery, which partly explains why episodic hypoxemia is more common during the second and third postoperative nights compared with the first postoperative night (data from abdominal surgery). (8)

Despite our results, other studies have shown that TEA has a protective effect against a decrease in pulmonary function. In one study, (21) the effects of high-dose fentanyl anesthesia vs TEA and low-dose fentanyl anesthesia on pulmonary function were compared. The results demonstrated a considerable reduction in pulmonary function after CABG surgery, while patients receiving TEA experienced improved FE[V.sub.1] on day 2 and peak expiratory flow rate on day 2 and 3 after surgery compared with the control group. (21) In an another study, (16) pulmonary function measured by mean maximal expiratory lung volume immediately before extubation was preserved in patients undergoing elective CABG surgery when receiving TEA compared to general anesthesia.

Apart from the changes in lung volume, surgical patients also exhibit the suppression of rapid eye movement (REM) sleep primarily during the first postoperative night, followed by rebound with a longer duration and a higher intensity of REM sleep during the second and the third postoperative nights. (7) REM sleep is associated with an increase in apnea and episodic oxygen desaturation. Furthermore, episodic hypoxemia occurs more often during periods of rebound REM sleep than during other sleep stages in the postoperative period. (22) This sleep disturbance may therefore be a key pathogenic factor in the development of episodic hypoxemia after operation.

In a number of studies in patients undergoing noncardiac surgery, the effect of therapy with opioids on late postoperative hypoxemia has been investigated. The focus has been on both the direct effect on respiratory pattern and the suppression of REM sleep with subsequent REM rebound with episodic hypoxemia. (7) A study (20) involving the first 16 h after surgery has suggested a connection between the use of IV opioid analgesia and periods of apnea and episodic hypoxemia. Several other studies (8) focusing on the second postoperative night showed either no correlation or a weak correlation between opioid dose and the number of hypoxemic episodes, which therefore probably makes it clinically irrelevant.

It is a limitation of the study that epidural morphine might have influenced the occurrence of oxygen desaturation in the TEA group. Opioids produce a dose-related depression of the ventilatory response to C[O.sub.2] by a direct effect on respiratory centers in the medulla. In addition, morphine has been shown to blunt the response to hypoxia. (23) Although the dose of epidurally administered morphine in this study was small (25 [micro]g/mL, 5 mL/h), a TEA approach with local anesthetics could only have altered the results.

Despite this, our results indicate, as have those in previous studies in noncardiac surgery patients, that therapy with opioids has no major influence on hypoxemia on the second and third postoperative nights. When it comes to anesthetic techniques, a high degree of TEA did not have a protective effect against late postoperative hypoxemia. However, our sample size was small, so the results should be verified in larger studies. In the light of both our results and the results in other studies, other mechanisms for late postoperative hypoxemia than the use of opioids and methods of anesthesia should be studied. Studies (8) in noncardiac surgery patients have shown that one of the most important reasons for episodic hypoxemia may be the suppression of REM sleep and the resulting rebound, and this field should therefore be explored in cardiac surgery patients as well. Factors like pain and therapy with opioids may be involved, but the most important reasons for the suppression of REM sleep are properly related to the surgical stress response, (7) whereas isoflurane anesthesia did not have a REM suppressant effect on volunteers in a previous study. (7,24)

Another limitation of the study might be the number of patients included. Before designing the study, the incidence of late postoperative hypoxemia for patients undergoing CABG was unknown. A power analysis based on studies dealing with noncardiac patients suggested that 84 patients should be used in the investigation. Because we were dealing with a quite different population of patients, this estimate was roughly used to indicate the number of patients needed for the study. Therefore, 50 patients were accepted for investigation.

In conclusion, the use of opioids did not seem to be of importance for the occurrence of late postoperative hypoxemia after coronary bypass surgery, and epidural anesthesia was not protective against hypoxemia on nights 2 and 3 after surgery. Further studies should investigate the relation between cardiac arrhythmias/ischemia and hypoxemia, and should explore the possible relationship between sleep disturbances and hypoxemia after CABG surgery.

REFERENCES

(1) Rosenberg J, Rasmussen V, von Jessen F, et al. Late postoperative episodic and constant hypoxaemia and associated ECG abnormalities. Br J Anaesth 1990; 65:684-691

(2) Rosenberg J, Dirkes WE, Kehlet H. Episodic arterial oxygen desaturation and heart rate variations following major abdominal surgery. Br J Anaesth 1989; 63:651-654

(3) Gill NP, Wright B, Reilly CS. Relationship between hypoxaemic and cardiac ischaemic events in the perioperative period. Br J Anaesth 1992; 68:471-473

(4) Reeder MK, Muir AD, Foex P, et al. Postoperative myocardial ischaemia: temporal association with nocturnal hypoxaemia. Br J Anaesth 1991; 67:626-631

(5) Coetzee A, Foex P, Holland D, et al. Effect of hypoxia on the normal and ischemic myocardium. Crit Care Med 1984; 12:1027-1031

(6) Rosenberg J. Hypoxaemia in the general surgical ward: a potential risk factor? Eur J Surg 1994; 160:657-661

(7) Rosenberg-Adamsen S, Kehlet H, Dodds C, et al. Postoperative sleep disturbances: mechanisms and clinical implications. Br J Anaesth 1996; 76:552-559

(8) Rosenberg J. Late postoperative hypoxaemia: mechanisms and clinical implications. Dan Med Bull 1995; 42:40-46

(9) Entwistle MD, Roe PG, Sapsford DJ, et al. Patterns of oxygenation after thoracotomy. Br J Anaesth 1991; 67:704-711

(10) Sanchez R, Nygard E. Epidural anesthesia in cardiac surgery: is there an increased risk? J Cardiothorac Vase Anesth 1998; 12:170-173

(11) Rosenberg J, Pedersen MH, Gebuhr P, et al. Effect of oxygen therapy on late postoperative episodic and constant hypoxaemia. Br J Anaesth 1992; 68:18-22

(12) Rapati D, Morandi A, Paje A, et al. Hypoxemia after heart surgery: an exception or a normal development? Minerva Anestesiol 1992; 58:355-360

(13) Liem TH, Booij LH, Hasenbos MA, et al. Coronary artery bypass grafting using two different anesthetic techniques: part I. Hemodynamic results. J Cardiothorac Vasc Anesth 1992; 6:148-155

(14) Liu SS, Block BM, Wu CL. Effects of perioperative central neuraxial analgesia on outcome after coronary artery bypass surgery: a recta-analysis. Anesthesiology 2004; 101:153-161

(15) Stenseth R, Bjella L, Berg EM, et al. Thoracic epidural analgesia in aortoeoronary bypass surgery: I. Haemodynamic effects. Acta Anaesthesiol Scand 1994; 38:826-833

(16) Scott NB, Turfrey DJ, Ray DA, et al. A prospective randomized study of the potential benefits of thoracic epidural anesthesia and analgesia in patients undergoing coronary artery bypass grafting. Anesth Analg 2001; 93:528-535

(17) Gravlee GP. Epidural analgesia and coronary artery bypass grafting: the controversy continues. J Cardiothorac Vasc Anesth 2003; 17:151-153

(18) Rosen DA, Hawkinberry DW, Rosen KR, et al. An epidural hematoma in an adolescent patient after cardiac surgery. Anesth Analg 2004; 98:966-969

(19) Ho AM, Chung DC, Joynt GM. Neuraxial blockade and hematoma in cardiac surgery: estimating the risk of a rare adverse event that has not (yet) occurred. Chest 2900; 117:551-555

(20) Catley DM, Thornton C, Jordan C, et al. Pronounced, episodic oxygen desaturation in the postoperative period: its association with ventilatory pattern and analgesic regimen. Anesthesiology 1985; 63:20-28

(21) Stenseth R, Bjella L, Berg EM, et al. Effects of thoracic epidural analgesia on pulmonary function after coronary artery bypass surgery. Eur J Cardiothorac Surg 1996; 10:859-865

(22) Rosenberg J, Wildschiodtz G, Pedersen MH, et al. Late postoperative nocturnal episodic hypoxaemia and associated sleep pattern. Br J Anaesth 1994; 72:145-150

(23) Chaney MA. Side effects of intrathecal and epidural opioids. Can J Anaesth 1995; 42:891-903

(24) Moote CA, Knill RL. Isoflurane anesthesia causes a transient alteration in nocturnal sleep. Anesthesiology 1988; 69:327-331

Lars Hyldborg Lundstrom, MD; Eigil Nygard, MD; Lamia Bada Hviid, MD; Finn Moller Pedersen, MD; Jesper Ravn, MD; Jan Aldershvile, MD, DSc; ([dagger]) and Jacob Rosenberg, MD, Dsc

* From the Departments of Thoracic Anesthesiology (Drs. Lundstrom, Nygard, and Pedersen), Thoracic Surgery (Drs. Hviid and Ravn), and Cardiology (Dr. Aldershvile), The Heart Centre Rigshospitalet, Copenhagen University, Copenhagen, Denmark; and the Department of Surgical Gastroenterology (Dr. Rosenberg), Gentofte University Hospital, Gentofte, Denmark. ([dagger]) Jan Aldershvile died on August 12, 2003.

This research was supported by The Danish Heart Foundation, Copenhagen, Denmark, by research grants No. 99-2-3-79-22764 and 99-1-5-92-22709.

Manuscript received April 8, 2003; revision accepted March 20, 2005.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml).

Correspondence to: Lars Hyldborg Lundstrom, Skovbrynet 22, Kagerup, 3200 Helsinge, Denmark; e-mail: lars.hyldborg@ dadlnet.dk

COPYRIGHT 2005 American College of Chest Physicians
COPYRIGHT 2005 Gale Group