1 Department of Anaesthesia, University of Basel/Kantonsspital, CH-4031 Basel, Switzerland. 2 Department of Anaesthesia, Spital Zollikerberg, CH-8125 Zollikerberg, Switzerland
* Corresponding author. E-mail: bvonungern{at}uhbs.ch
Accepted for publication August 19, 2004.
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. Eighty-four patients having midline laparotomy for gynaecological procedures successfully completed the study. Premedication, anaesthesia and analgesia were standardized. The patients were given a free choice between epidural analgesia (EDA) (n=42) or opioids (n=42) for postoperative analgesia. We performed spirometry to measure vital capacity (VC), forced vital capacity, peak expiratory flow, mid-expiratory flow and forced expiratory volume in 1 s at preoperative assessment, 3060 min after premedication and 20 min, 1 h, 3 h and 6 h after extubation.
Results. Baseline values were all within the normal range. All perioperative spirometric values decreased significantly with increasing body mass index (BMI). The greatest reduction in VC occurred directly after extubation, but was less in the EDA group than in the opioid group: mean of 23(SD 8)% versus 30(12)% (P<0.001). In obese patients (BMI>30) the difference in VC was significantly more pronounced than in patients of normal weight (BMI<25): 45(10)% versus 33(4)% (P<0.001). Recovery of spirometric values was significantly quicker in patients receiving EDA, particularly in obese patients.
Conclusion. We conclude that EDA should be considered in obese patients undergoing midline laparotomy to improve postoperative spirometry.
Keywords: anaesthetic techniques, epidural ; complications, obesity ; lung, respiratory function ; monitoring, spirometry ; surgery, laparotomy
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
General anaesthesia
In both groups, premedication consisted of oral midazolam 7.5 mg given 3060 min before surgery. In the EDA group, EDA was initiated after local infiltration according to our routine using an 18 gauge Tuohy needle and a 20 gauge multiport epidural catheter inserted at the T7T8 or T8T9 interspace. After a negative test dose consisting of lidocaine 2% (3 ml) with 1:200 000 epinephrine, an epidural bolus injection of bupivacaine 15 mg and fentanyl 100 µg in sodium chloride 0.9% (10 ml) were given. Further bolus injections of bupivacaine 0.5% followed according to clinical needs. In both groups, general anaesthesia was induced with propofol 2 mg kg1 and fentanyl 2 µg kg1 i.v. Tracheal intubation was facilitated by atracurium 0.5 mg kg1 i.v. Anaesthesia was maintained with nitrous oxide 66% in oxygen and propofol by infusion using the Bristol formula (10 mg kg1 h1 for the first 10 min, 8 mg kg1 h1 for a further 10 min and thereafter 6 mg kg1 h1 or adjusted to individual needs).7 Ventilation was controlled using an ADU Ventilator (Datex Ohmeda, S/5 ADU Helsinki, Finland) with a circle system. Repeated doses of fentanyl were given during surgery as necessary based on clinical signs (heart rate, arterial pressure, pupil size and sweating), but not within 60 min of the estimated end of surgery. To have the patient fully alert and cooperative for spirometry, we substituted sevoflurane for propofol 3060 min before the estimated end of surgery as this was considered, on the basis of clinical observations, to allow for a more rapid recovery.4 Increments of atracurium 5 mg i.v. were given to maintain muscle relaxation which was monitored by train-of-four stimulation. Neostigmine 2.5 mg and glycopyrrolate 0.5 mg i.v. were given as needed to antagonize residual neuromuscular block. Before extubation of the trachea, four equal twitches in the train-of-four without tetanic fade (50 Hz over 5 s) were required as well as recovery of consciousness (eye opening on demand), protective airway reflexes and adequate spontaneous ventilation.
Postoperative pain management
In both groups, postoperative basic analgesia consisted of paracetamol 1000 mg rectally or orally every 6 h starting directly after the operation. In the EDA group, a continuous infusion of epidural bupivacaine 0.125% with fentanyl 2 µg ml1 was administered. The infusion rate was adjusted to obtain a sensory level of T5 (range 510 ml h1) and adequate analgesia. Adequate analgesia was defined as a pain score 20 mm while coughing, which was assessed on the 100 mm visual analogue scale (VAS), where 0 mm represented no pain or no dyspnoea while 100 mm indicated the worst possible pain or dyspnoea). If pain persisted in the EDA group despite a sufficient sensory level, as a first measure an additional epidural dose of fentanyl 100 µg in sodium chloride 0.9% (10 ml) was given. In both groups, according to our standards, increments of methadone 2 mg i.v. were given to the patients in order to achieve adequate analgesia. The total dose of methadone given to each patient was neither limited nor weight adjusted.
Spirometry
Spirometry was standardized with each patient in a 30° head-up position using a Vitalograph 2120 (Vitalograph, Hamburg, Germany). At the pre-anaesthetic visit, a baseline spirometry measurement was taken (T0) after a thorough demonstration of the correct usage. Vital capacity (VC), forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1) mid-expiratory flow (MEF2575) and peak expiratory flow (PEF) were measured and the FEV1/FVC ratio was calculated. At each assessment time, spirometry was performed at least three times to be able to meet the criteria of the European Respiratory Society (ERS)8 and the best measurement was recorded. When the patient arrived in the operating theatre (about 3060 min after premedication), we repeated spirometry (T1) after initiation of effective EDA (where applicable) and before induction of anaesthesia in order to compare the effect of premedication alone with the effect of premedication plus effective EDA on spirometric tests. After extubation, as soon as the patient was alert and fully cooperative, pain and dyspnoea were assessed during coughing using the VAS before and, if necessary, after analgesic therapy. Pain was not assessed during the VC manoeuvre itself. As soon as a patient had a VAS pain score 20 mm during coughing (all patients met this criterion within 20 min of extubation), we performed spirometry for the third time (T2). Spirometric assessments were repeated in the postanaesthetic care unit at 1 h (T3), 3 h (T4) and 6 h (T5) after extubation. Prior to each assessment, as soon as the patients were free from pain during coughing, methadone requirements were documented and sensory levels of EDA were evaluated. All postoperative measurements, including spirometry, were performed by postanaesthetic care unit nurses trained to use the spirometer but unaware of the study hypothesis and otherwise not involved in this study.
Statistical analysis
We measured the weight and height of each patient to obtain the exact BMI. In order to quantify the effect of obesity, we allocated the patients according to their BMI as follows: normal weight (BMI<25), mildly obese (BMI 2530) and obese (BMI>30). To allow for comparisons between the patients and the groups, pulmonary function values were calculated as the percentage deviation from baseline (T0). To compare data within the groups, repeated-measure analysis of variance (ANOVA) was applied. To compare data between the groups, a Wilcoxon rank-sum test was performed. For post hoc comparisons, a Bonferroni test was applied and probability values were calculated. The Spearman rank correlation test was used to calculate the correlation coefficients between spirometric measurements and BMI as the BMI data were skewed. A P-value <0.05 was considered significant. The Statview for Windows software package (SAS Institute Inc., Cary, NC, USA, Version 5.0.1) was used for statistical calculations.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
Intraoperative opioid requirement, postoperative pain scores and pain relief
There were no differences in pain scores between the groups when spirometry was performed; the maximum VAS value was 20 mm. There was no correlation between the VAS scores and the reduction in spirometric measurements in either group of patients. During the whole study period, no patient in either group complained of dyspnoea.
There were significant differences between the groups regarding intraoperative and postoperative opioid requirements. Mean intraoperative fentanyl doses were 0.30 (0.10) mg in the EDA group versus 0.62 (0.17) mg in the opioid group.
Postoperatively, seven patients in the EDA group received a single dose of epidural fentanyl 100 µg, while four patients received a single dose of methadone 2 mg i.v. Postoperative analgesic requirements were higher in the opioid group, as indicated by mean methadone doses of 0.9 (1.1) mg immediately after extubation, 4.5 (2.2) mg between T2 and T3, 4.1 (2.1) mg between T3 and T4, and 4.4 (1.9) mg between T4 and T5. This resulted in a total methadone dose within the first six postoperative hours of 14 (3.6) mg for the opioid group compared with 0.2 (0.7) mg for the EDA group. There was no correlation between the opioid consumption and the reduction of VC within either the EDA or the opioid groups.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Effect of premedication and initiation of effective EDA
The effect of premedication was similar to that shown in a previous study,4 although EDA was introduced as a new variable. Premedication resulted in a BMI-related reduction of VC with no significant difference between the groups and no further change by effective EDA. However, there was a comparatively wide range of individual responses to the effect of premedication. VC is known to be a good index of respiratory muscle strength in patients with neuromuscular disorders.9 Benzodiazepines have a spinally mediated relaxant action on respiratory muscles9 and could have been expected to affect respiration, with obese patients being more affected because of their greater work of breathing.10 Another explanation for the impairment is sedation induced by midazolam that could have interfered with spirometric performance. However, sedation or lack of volition would have affected both obese and non-obese patients equally, although we did not include an alternative measure of the patient's ability to cooperate and perform maximally. Moreover, changes in MEF2575 values, which depend less on patient cooperation than PEF values, were in parallel to the changes in PEF values during the whole study period.11 The observation that premedication with benzodiazepines resulted in a marked reduction of VC in obese patients indicates that this class of patient should be given either no or a reduced dose of premedication. Thus, if premedication is given to obese patients, they should be closely observed in the preoperative period for signs of respiratory impairment.
Surprisingly, EDA did not influence spirometric measurements (except for a reduction of PEF and MEF2572 values), even though initiation of EDA may have accounted for some degree of muscle relaxation as shown by changes in dynamic rather than static spirometric measurements of respiratory function.12 13
Anaesthesia and immediate postoperative respiratory function
As previously described,4 the lowest spirometric values are observed during the first assessment after extubation. The decrease in VC, FVC, FEV1, MEF2575 and PEF followed the same pattern (Table 2), and the FEV1/FVC ratio did not change. This suggests a restrictive pattern of respiratory compromise in the immediate postoperative period, as previously described.4 14
Postoperative impairment of spirometric measurements was probably not related to insufficient cooperation since all patients were alert and fully compliant within 20 min of extubation. Additionally, any lack of cooperation would have affected the whole study population to a comparable degree. The reduction of spirometric volumes observed in our study may have been caused by impaired respiratory mechanics, obesity and atelectasis formation promoted by general anaesthesia in the supine position, as well as by abdominal surgery.1517 A reduction in both inspiratory and expiratory reserve volumes would not only have an impact on VC,14 but might interfere directly with the ability to cough effectively as a result of decreased inspiratory capacity and thus predispose to respiratory complications.2 14
Body mass index and immediate postoperative respiratory function
As previously reported,4 the compromise of spirometric measurements correlated significantly with increasing BMI, persisted over the entire study period and was more severe in obese patients. Six hours after an operation, the mean VC reduction in the opioid group was 42% for obese but only 21% for normal-weight patients. These data differed significantly from those obtained within the EDA group, in which there was a mean reduction in VC of 24% for obese and 13% for normal-weight patients.
Thoracic epidural analgesia and respiratory function
The effect of EDA on spirometric measurements is controversial. High-thoracic EDA was shown to decrease spirometric measurements by blocking intercostal muscle innervation.18 19 A recent study showed a 2530% decrease in FVC and FEV1 after initiation of EDA in patients undergoing cardiac surgery. This decrease was mainly attributable to change of position, since baseline measurements performed in the sitting position were compared with subsequent measurements in the supine position.20 In contrast with the latter results, and in line with our study, others have not found that EDA has any influence on spirometry or lung dynamics.2124
During forced expiration (e.g. spirometry), the principal expiratory muscles are those of the abdominal wall and, to a lesser extent, the internal intercostal muscles. EDA with sensory levels extending from approximately T4 to L1 is likely to be accompanied by some degree of muscle paralysis, even if low concentrations of local anaesthetics are used,13 and is more likely to block the muscles of the abdominal wall (innervation T6L1) than the diaphragm (C3C5) or the intercostal muscles (T1T11). This blockade of abdominal muscles because of low-thoracic EDA is reflected by a reduction of the dynamic parameters PEFR and MEF2575, which depend more on active exhalation, but is without significant changes in comparatively static spirometric measurements (e.g. VC).19 2527 Even a subtle decrease in abdominal muscle tone because of EDA will affect dynamic parameters before impairing static parameters.
Some older studies show a reverse relationship between reduced spirometric measurements and increased postoperative complication rates.28 Despite the lack of evidence that EDA reduces in-hospital mortality, a recent large randomized trial showed a significant reduction in postoperative respiratory failure rates (23% versus 30%).5 However, since lung volumes were not measured, this trial did not answer the question as to whether a reduction in spirometric values was predictive of postoperative complications.
Overall, the positive effects of EDA on spirometric tests, which became even more important in obese patients, might add to other benefits of EDA shown in previous studies, such as earlier mobilization, more rapid recovery of bowel function, thus allowing oral nutrition, and less disturbance of mental status in the elderly.29 30
Postoperative pain and respiratory function
Spirometric measurements have been used to quantify postoperative pain.31 32 Therefore it is crucial for a patient to be free of pain during spirometry and thus to be as close to preoperative baseline conditions as possible to avoid factors that affect test performance. Pain probably influenced the results of earlier studies in which insufficient postoperative pain relief might have contributed to a greater decrease of VC.28 33
Although all our patients were free of pain during coughing (VAS20 mm), there might also have been some degree of abdominal tension because of volume shifts into the third space.14 The pain score during the VC manoeuvre itself was not measured. Theoretically, reduction of abdominal wall tension induced by EDA might result in a decrease of diaphragmatic strain and ease displacement of the abdominal contents during breathing, and thus might have contributed to the measured differences between the two techniques. Therefore inspiratory volumes would be increased in the EDA group, improving all spirometric measurements provided that active expiration is intact.12
Sedation induced by the larger doses of opioids required during surgery and for postoperative analgesia might have interfered with spirometry of subjects not receiving the benefits of EDA, although there was no correlation between the opioid requirements and spirometric performance within the groups.
Limitations
An observational rather than a randomized study design was used in our study. With randomization, this study would not have been finished within a reasonable time span in our hospital. Many patients refuse thoracic EDA for fear of neurological complications after being informed about its risks. In consequence, the patients opted for systemic opioid analgesia technique.
The potential for a selection bias was minimized by the support of anaesthetists not involved in this study who were responsible for giving patients preoperative information. Patients were only asked for informed consent once they had decided on their perioperative pain regimen. Additionally, postoperative spirometry was performed by trained nurses who were unaware of the study hypothesis and were not involved in this study.
Our findings do not allow us to draw conclusions regarding the mechanism of VC loss or to distinguish between the loss of inspiratory and expiratory power. Nevertheless, the primary aim of our study was to examine the potential of different perioperative anaesthetic regimens for modifying spirometrically measured lung volumes, and to assess whether there were clinically relevant differences in postoperative respiratory impairment during the immediate postoperative period when the impact of surgical trauma and anaesthesia are likely to peak and trigger postoperative pulmonary morbidity.4
We conclude that obesity is an important risk factor for perioperative impairment of spirometric measurements in patients undergoing laparotomy. The moderate reduction of spirometric tests induced by midazolam as premedication was not enhanced further by EDA. The reduction postoperatively was significantly greater in obese than in normal-weight patients. In all patients, the severity of postoperative lung volume reduction measured by spirometry was reduced by the presence of EDA and postoperative restoration of lung volumes was significantly quicker. The use of EDA should be considered for obese patients undergoing midline laparotomy.
![]() |
Acknowledgments |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Hansen G, Drablos PA, Steinert R. Pulmonary complications. Ventilation and blood gases after upper abdominal surgery. Acta Anaesthesiol Scand 1977; 21: 21115[ISI][Medline]
3 Blouw EL, Rudolph AD, Narr BJ, Sarr MG. The frequency of respiratory failure in patients with morbid obesity undergoing gastric bypass. AANA J 2003; 71: 4550[Medline]
4 von Ungern-Sternberg BS, Regli A, Schneider MC, et al. The effect of obesity and site of surgery on perioperative lung volumes. Br J Anaesth 2004; 92: 2027
5 Rigg JRA, Myles PS, Silbert BS, et al. Epidural anaesthesia and analgesia and outcome of major surgery: a randomised trial. Lancet 2002; 359: 127682[CrossRef][ISI][Medline]
6 Kehlet H, Holte K. Effect of postoperative analgesia on surgical outcome. Br J Anaesth 2001; 87: 6272
7 Roberts FL, Dixon J, Lewis GT, et al. Induction and maintenance of propofol anaesthesia. A manual infusion scheme. Anaesthesia 1988; 43 (Suppl): 147[Medline]
8 Standardized lung function testing. Official statement of the European Respiratory Society. Eur Respir J 1993; 16 (Suppl): 1100[CrossRef]
9 Jolly E, Aguirre L, Jorge E, Luna C. Efecto agudo del lorazepam sobre los musculos respiratorios en los pacientes con epoc estable. Medicina (B Aires) 1996; 56: 4728[ISI][Medline]
10 Ray C, Sue D. Bray G, et al. Effects of obesity on respiratory function. Am Rev Respir Dis 1983; 128: 5016[ISI][Medline]
11 West JB. Respiratory Physiology. The Essentials, 6th edn. Philadelphia, PA: LippincottWilliams & Wilkins, 1999
12 von Ungern-Sternberg BS, Regli A, Bucher E, et al. The effect of epidural analgesia on the respiratory function of pregnant women during labour. Anaesthesia 2004; 59: 3503[CrossRef][ISI][Medline]
13 Zaric D, Nydahl PA, Philipson L, et al. The effect of continuous lumbar epidural infusion of ropivacaine (0.1%, 0.2%, and 0.3%) and 0.25% bupivacaine on sensory and motor block in volunteers: a double-blind study. Reg Anesth 1996; 21: 1425[ISI][Medline]
14 Craig D. Postoperative recovery of pulmonary function. Anesth Analg 1981; 60: 4652[ISI][Medline]
15 Pelosi P, Croci M, Ravagnan I, et al. The effects of body mass on lung volumes, respiratory mechanics, and gas exchange during general anaesthesia. Anesth Analg 1998; 87: 65460[Abstract]
16 Rigg JR. Pulmonary atelectasis after anaesthesia: pathophysiology and management. Can Anaesth Soc J 1981; 28: 30513[ISI][Medline]
17 Eichenberger A, Proietti S, Wicky S, et al. Morbid obesity and postoperative pulmonary atelectasis: an underestimated problem. Anesth Analg 2002; 95: 178892
18 Takasaki M, Takahashi T. Respiratory function during cervical and thoracic extradural analgesia in patients with normal lungs. Br J Anaesth 1980; 52: 12715[Abstract]
19 Sjögren S, Wright B. Respiratory changes during continuous epidural blockade. Acta Anaesthesiol Scand 1972; 16: 2746[ISI][Medline]
20 Tenling A, Joachimsson P-O, Tyden H, Hedenstierna G. Thoracic epidural analgesia as an adjunct to general anaesthesia for cardiac surgery. Effects on pulmonary mechanics. Acta Anaesthesiol Scand 2000; 44: 10716[CrossRef][ISI][Medline]
21 McCarthy GS. The effect of thoracic extradural analgesia on pulmonary gas distribution, functional residual capacity and airway closure. Br J Anaesth 1976; 48: 2438[Abstract]
22 Wahba WM, Craig DB, Don HF, Becklake MR. The cardiorespiratory effects of thoracic epidural anaesthesia. Can Anaesth Soc J 1972; 19: 819[ISI][Medline]
23 Sundberg A, Wattwil M, Arvill A. Respiratory effects of high thoracic epidural anaesthesia. Acta Anaesthesiol Scand 1986; 30: 21517[ISI][Medline]
24 Moir DD. Ventilatory function during epidural analgesia. Br J Anaesth 1963; 35: 37[ISI]
25 Freund FG, Bonica JJ, Ward RJ, et al. Ventilatory reserve and level of motor block during high spinal and epidural anesthesia. Anesthesiology 1967; 28: 8347[ISI][Medline]
26 Egbert LD, Tamersoy K, Deas TC. Pulmonary function during spinal anaesthesia: the mechanism of cough depression. Anesthesiology 1961; 27: 8825
27 Freund FG. Respiratory effects of subarachnoid and epidural block. Clin Anesth 1969; 2: 97107[Medline]
28 Parbrook G, Stell D, Dalrymple D. Factors predisposing to postoperative pain and pulmonary complications. Br J Anaesth 1973; 45: 2132.[ISI][Medline]
29 Carli F, Mayo N, Klubien K, et al. Epidural analgesia enhances functional exercise capacity and health-related quality of life after colonic surgery: results of a randomised trial. Anaesthesiology 2002; 97: 5409[ISI][Medline]
30 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: 52835
31 Bromage P. Spirometry in assessment of analgesia after abdominal surgery: a method of comparing analgesic drugs. Br Med J 1955; 2: 58993.[ISI]
32 Macias A, Monedero P, Adame M, et al. A randomized, double-blinded comparison of thoracic epidural ropivacaine, ropivacaine/fentanyl, or bupivacaine/fentanyl for postthoracotomy analgesia. Anesth Analg 2002; 95: 134450.
33 Simpson BR, Parkhouse J, Marshall R, Lambrechts W. Extradural analgesia and the prevention of postoperative respiratory complications. Br J Anaesth 1961; 33: 62841[ISI]
|