Hypoxia during upper gastrointestinal endoscopy with and without sedation and the effect of pre-oxygenation on oxygen saturation

Transcription

1 Hypoxia during upper gastrointestinal endoscopy with and without sedation and the effect of pre-oxygenation on oxygen saturation C. Y. Wang, 1 L. C. Ling, 2 M. S. Cardosa, 2 A. K. H. Wong 2 and N. W. Wong 3 1 Department of Anaesthesiology & Intensive Care, and 3 Department of Medicine, University of Malaya Medical Centre, Lembah Pantai, Kuala Lumpur, Malaysia 2 Department of Anaesthesiology & Intensive Care, Kuala Lumpur Hospital, Kuala Lumpur, Malaysia Summary In Study A, the incidence of arterial oxygen desaturation was studied using pulse oximetry (S a o 2 )in 100 sedated and 100 nonsedated patients breathing room air who underwent diagnostic upper gastrointestinal endoscopy. Hypoxia (S a o 2 92% or less of at least 15 s duration) occurred in 17% and 6% of sedated patients and nonsedated patients, respectively (p, 0.03). Mild desaturation (S a o 2 94% or less and less than 15 s duration) occurred in 47% of sedated patients compared with 12% of nonsedated patients (p, 0.001). In Study B, the effects of supplementary oxygen therapy and the effects of different pre-oxygenation times on arterial oxygen saturation (S a o 2 ) in sedated patients were studied using pulse oximetry. One hundred and twenty patients who underwent diagnostic upper gastrointestinal endoscopy with intravenous sedation were studied. Patients were randomly allocated to one of four groups: Group A received no supplementary oxygen while Groups B±D received supplementary oxygen at 4 l.min 21 via nasal cannulae. The pre-oxygenation time in Group B was zero minutes, Group C was 2 min and Group D was 5 min before sedation and introduction of the endoscope. Hypoxia occurred in seven of the 30 patients in Group A and none in groups B, C and D (p, 0.001). We conclude that desaturation and hypoxia is common in patients undergoing upper gastrointestinal endoscopy with and without sedation. Sedation significantly increases the incidence of desaturation and hypoxia. Supplementary nasal oxygen at 4 l.min 21 in sedated patients abolishes desaturation and hypoxia. Pre-oxygenation confers no additional benefit. Keywords Surgery: gastrointestinal endoscopy. Complications: hypoxia. Monitoring: pulse oximetry.... Correspondence to: Dr C. Y. Wang Accepted: 30 January 2000 Upper gastrointestinal endoscopy has been a common procedure for more than 20 years, but only recently has attention been focused on the associated cardiorespiratory complications. These account for some 50% of the morbidity and 60% of the mortality from endoscopy [1± 3]. Hypoxia during upper gastrointestinal endoscopy is a well-recognised complication. It is unclear if the hypoxia results primarily from the respiratory depressant effects of sedation or the presence of the endoscope obstructing the upper airway. Upper gastrointestinal endoscopy can be performed with or without sedation. Although it is often considered to be unpleasant, a number of endoscopic centres perform the procedure without sedation [4, 5]. However, it is uncertain whether endoscopy performed without sedation reduces the incidence of complications. The first study (Study A) was a prospective joint study between the University of Malaya Medical Centre (UMMC) and Kuala Lumpur Hospital (KLH). At the time of the study, patients undergoing upper gastrointestinal endoscopy in KLH did not receive sedation routinely, while in UMMC all patients undergoing upper gastrointestinal endoscopy were given sedation unless they refused it. The aims of Study A were to determine the 654 q 2000 Blackwell Science Ltd

2 C. Y. Wang et al. Hypoxia during upper gastrointestinal endoscopy incidence of hypoxia during upper gastrointestinal endoscopy, with or without sedation, and to survey patient attitudes to sedation. Since Study A showed that there was a high incidence of hypoxia in sedated patients, we proceeded to investigate prospectively whether supplementary oxygen by nasal cannulae prevented hypoxia in patients who received sedation while undergoing upper gastrointestinal endoscopy at the UMMC (Study B). As there have been few studies that have examined the effects of preoxygenation before sedation in the prevention of hypoxia during upper gastrointestinal endoscopy, we also investigated the effect of different pre-oxygenation times in these patients. Patients and methods The study was approved by the Hospital Ethics Committee and written informed consent was obtained from all patients. Both studies were performed on unpremedicated adult ASA I and II patients who underwent upper gastrointestinal endoscopy. Patients with severe cardiopulmonary disease and all patients with baseline pulse oximetry arterial oxygen saturation of 94% or below were not studied. Study A Two hundred patients who underwent upper gastrointestinal endoscopy without oxygen supplementation were studied. Group I consisted of 100 consecutive patients from Kuala Lumpur Hospital (KLH) where no sedation was given. Group II consisted of 100 consecutive patients from the University of Malaya Medical Centre (UMMC) who underwent the same procedure and received intravenous midazolam (2.5± 5.0 mg) for sedation. Past medical history for chronic obstructive pulmonary disease, bronchial asthma and smoking was taken. The patients' height and weight were measured and the body mass index (BMI) was calculated according to the formula: BMI ˆ weight/ height 2 (kg.m 22 ). In both groups, the endoscopy was performed by one of two endoscopists, each with more than 3 years of experience. The endoscopists were blinded to the monitoring by an opaque screen placed between the patient and the endoscopist. All patients received 10% lidocaine spray to anaesthetise the oropharynx before undergoing endoscopy in the left lateral position. Arterial oxygen saturation (S a o 2 ) and pulse rate were monitored continuously with a Nellcor (N200E) pulse oximeter using a finger probe. In both groups, prior to endoscopy and sedation, baseline saturation and pulse rate were obtained after 30 s following finger probe placement. Patients with a baseline saturation of S a o 2 94% or below were not studied. Sedation was given by intravenous bolus of midazolam, 2.5 mg for patients 65 years or older and 5 mg for patients less than 65 years. All episodes of desaturation to 94% or below were recorded: X desaturation to 92% or below for at least 15 s was defined as hypoxia ± 92% was taken to represent hypoxia because it marked the mid-point of the shoulder of the oxyhaemoglobin dissociation curve; X all other episodes of desaturation to 94% or below of any duration were considered as mild desaturation. Oxygen supplementation was given when necessary. After the endoscopy, each patient was asked if they could remember the procedure. If they did, they were asked if the procedure was unpleasant and if they would wish to have sedation in future. Study B In the second part of the study (Study B), 120 patients from the UMMC who underwent upper gastrointestinal endoscopy with intravenous sedation were studied. The patients were randomly allocated before the start of the study to one of four groups by opening sealed envelopes in the endoscopy room. Each group consisted of 30 patients. Group A received no supplementary oxygen while all other groups were given supplementary oxygen 4 l.min 21 via nasal cannulae. In addition to receiving continuous intranasal oxygen during the procedure, patients in Group B were pre-oxygenated for zero minutes, Group C for 2 min and Group D for 5 min prior to sedation. The endoscopist was blinded to the randomisation process. He was only allowed to come into the endoscopy theatre after the pre-oxygenation and sedation were given to the patient. An opaque screen was also placed between the patient and the pulse oximeter and oxygen flowmeter so that the endoscopists could not see the pulse oximeter or witness the administration of oxygen. Audible tones and alarms of the pulse oximeter were silenced. Sedation was given by intravenous bolus midazolam 2.5 mg for patients older than 65 years and 5 mg for patients less than 65 years. The procedure then proceeded as in Study A. Data analysis The data were analysed using the Chi-squared test. A two-tailed probability of less than 0.05 was the criterion for statistical significance. Results Study A The two groups were similar with respect to age, sex distribution and mean BMI. The percentage of patients q 2000 Blackwell Science Ltd 655

3 C. Y. Wang et al. Hypoxia during upper gastrointestinal endoscopy Anaesthesia, 2000, 55, pages 654±658 Table 1 Physical characteristics and baseline readings of mean body mass index, baseline oxygen saturation and endoscopy time, and the percentage of patients who were smokers, chronic obstructive airway disease and asthma. Results are shown as mean (SD). 'Not unpleasant' 'Slightly unpleasant' * 'Very unpleasant' Group I Group II % Age; years (17.3) (17.01) Sex ratio; male: female 67 : : 44 Body mass index; kg.m (4.26) (3.12) Smokers; % Chronic obstructive airway disease/asthma; % 5 5 Previous scope; % Mean endoscopy time; min 6.11 (3.49) 6.27 (3.5) Mean baseline S a o 2 ; % (1.37) (1.38) % of patients % 61% who were smokers, and who had chronic obstructive pulmonary disease or asthma were similar between groups. The number of patients with previous experience of upper gastrointestinal endoscopy, the mean endoscopy time and also the mean baseline oxygen saturation were also similar between the two groups (Table 1). Table 2 shows the incidence of arterial oxygen desaturation. Mild desaturation occurred in 47% of the sedated group compared with 12% in the unsedated group (p, 0.001). Significantly more sedated patients (17%) suffered hypoxia compared with 6% in the unsedated patients (p, 0.03). The mean lowest S a o 2 was significantly lower in the sedated group (94.1%) compared with the unsedated group (97.1%) (p, 0.001). Thirty-one per cent of the unsedated patients thought that the procedure was `very unpleasant' compared with 12% in the sedated group (p, 0.003) (Fig. 1). Seventythree per cent in the sedated group reported a preference for sedation for future repeat procedures compared with 38% in the unsedated group. Seventy per cent of the sedated patients reported amnesia (Fig. 2). Table 2 Incidence of desaturation (%) and the mean lowest oxygen saturation during upper gastrointestinal endoscopy. Values are given as mean (SD). Group I Group II No. of patients with mild desaturation; % 12 (12) 47 (47)* No. of patients with hypoxia; % 6 (6) 17 (17)** Mean lowest S a o 2 ; % (2.6) 94.1 (3.9)* (Group I ˆ no sedation, Group II ˆ with sedation.) *p, 0.001, **p, 0.03 vs. no sedation % 27% 12% No sedation Sedated Figure 1 Opinions of patients regarding the procedure. The values shown are percentages (%). *p, vs. sedated group. Study B One hundred and twenty patients were studied. The four groups were similar with respect to age, sex distribution, mean BMI, number of patients with previous endoscopy done and the percentage of patients who were smokers or had chronic obstructive pulmonary disease or asthma. % of patients % 46% 16% Unsedated 'Yes' 'No' Indifferent 73% 3% 24% Sedated Figure 2 Preference for sedation for a repeat procedure. The values shown are percentages (%). 656 q 2000 Blackwell Science Ltd

4 C. Y. Wang et al. Hypoxia during upper gastrointestinal endoscopy Table 3 Physical characteristics and baseline readings of mean body mass index, baseline oxygen saturation and endoscopy time, and the percentage of patients who were smokers, chronic obstructive airway disease and asthma. The two groups values are given as mean (SD). Group A Group B Group C Group D Mean age; years 46.6 (17.2) 51.0 (19.4) 51.0 (14.4) 51.1 (17.0) Sex; M/F 15/15 17/13 17/12 19/10 BMI; kg.m (4.58) (4.68) (3.73) (4.38) Smokers; % Chronic obstructive airway disease/asthma; % Previous scope; % Endoscopy time; min 7.8 (4.9) 7.5 (4.0) 7.4 (3.4) 9.5 (7.3) Baseline oxygen saturation; % 98.2 (1.4) (1.5) (1.4) (1.0) The mean endoscopy time and the mean baseline oxygen saturation were also similar (Table 3). Sixteen patients in Group A (53.3%) had mild desaturation compared with none in Groups B, C and D(p, 0.001). Seven patients in Group A (23.3%) had hypoxia compared with none in Groups B, C and D (p, 0.001) (Table 4). Discussion Our study showed that hypoxia is a common problem during upper gastrointestinal endoscopy with or without sedation, and that sedation significantly increases the incidence of hypoxia. These findings are similar to those of Reed and colleagues, who reported that oxygen desaturation during upper gastrointestinal endoscopy occurs in both sedated and nonsedated patients, and that the incidence of hypoxia increased with sedation [6]. Several studies have looked at the mode of delivery of supplementary oxygen for patients undergoing upper gastrointestinal endoscopy: oxygen adminstration via nasal catheters [7] or oxygenating mouthguards [8], the effect of different oxygen flow rates [9] on maintenance of normoxaemia, the effect of pre-oxygenation versus only giving oxygen during endoscopy [10] and the effect of supplemental oxygen during the procedure itself [11]. However, no other study has attempted to compare the effect of varying the time of pre-oxygenation, as conducted in our study. Rigg et al. [10] and Reed et al. [6] both recommended oxygen supplementation at 4 l.min 21 intranasally, to be given at 5 min and 2 min, respectively, before the start of endoscopy. The reasons for these recommendations were not clear. Neither study compared different pre-oxygenation times. In Study B, we found that in Group A (no supplementary oxygen), 53.3% of patients had mild desaturation and 23.3% had hypoxia compared with none in Groups B, C and D (all patients were given supplementary oxygen). Although patients in Groups B, C and D were `pre-oxygenated' for different times, ranging from zero minutes to 5 min, we did not find any difference in the incidence of desaturation or hypoxia among the three groups. Pre-oxygenation is designed to achieve an increase in the body oxygen stores, sufficient to avoid hypoxia during the procedure. For effective pre-oxygenation it is necessary to breathe 100% oxygen for 3 min using a tightly fitting face mask. Any leaks in the breathing system will allow ingress of room air, preventing effective pre-oxygenation. Berthaoud et al. showed that allowing inspiration of even a small amount of room air during pre-oxygenation had a marked effect on denitrogenation [12, 13]. When using nasal cannulae, there is little control over the inspired concentration, which can vary from 25 to 40% with flow rates of 2±4 l.min 21, depending upon the pattern and volume of Table 4 Incidence of desaturation (%) and the mean lowest oxygen saturation during upper gastrointestinal endoscopy. Values are given as mean (SD). Group A Group B Group C Group D No. of patients with mild desaturation; % 16 (53.3%) 0 (0%)** 0 (0%)** 0 (0%)** No. of patients with hypoxia; % 7 (23.3%) 0 (0%)** 0 (0%)** 0 (0%)** Mean lowest S p o 2 ; % (3.7) (1.0)* (1.4)* (1.4)* (Group A ˆ no oxygen supplementation, Group B ˆ oxygen supplementation with no pre-oxygenation, Group C ˆ oxygen supplementation with 2 min of pre-oxygenation, Group D ˆ oxygen supplementation with 5 min of pre-oxygenation.) *p, 0.01, **p, vs. no oxygen supplementation. q 2000 Blackwell Science Ltd 657

5 C. Y. Wang et al. Hypoxia during upper gastrointestinal endoscopy Anaesthesia, 2000, 55, pages 654±658 ventilation [14, 15]. This may be one possible reason for the lack of additional benefit of pre-oxygenation with nasal cannulae. Our findings indicate that pre-oxygenation is unnecessary and nasal oxygen supplementation at 4 l.min 21 given during the procedure is sufficient to abolish episodes of desaturation and hypoxia. Although upper gastrointestinal endoscopy can be performed without sedation [4], many patients find this method unpleasant despite the recent use of smalldiameter endoscopes [5]. These findings were confirmed in our study, where an unacceptably large number of unsedated patients (31%) found the procedure very unpleasant. Thirty-eight per cent of this same (unsedated) group also reported a preference for sedation for future repeat procedures. Amnesia, which occurred in 70% of sedated patients, is an additional benefit of sedation, particularly in procedures such as upper gastrointestinal endoscopy which often have to be repeated. In conclusion, endoscopy performed without sedation is uncomfortable, and we recommend that sedation should be given routinely during such procedures. Withholding sedation does not necessarily prevent desaturation and hypoxia in patients undergoing endoscopy, and although sedation increases the incidence of desaturation and hypoxia, this can be completely abolished by supplementary nasal oxygen at 4 l.min 21. Since our study has shown that pre-oxygenation confers no additional benefit, we propose that the `pre-oxygenation' time recommended by other authors [6±10] should be omitted. This could be cost and time saving, especially in a busy endoscopy unit. Acknowledgments We thank the nursing staff of the endoscopy unit, University of Malaya Medical Centre and Kuala Lumpur Hospital for assistance with this study. References 1 Hart R, Classen M. Complications of diagnostic gastrointestinal endoscopy. Endoscopy 1990; 22: 229±33. 2 Bell GD, McCloy RF, Charlton JE et al. Recommendations for standards of sedation and patient monitoring during gastrointestinal endoscopy. Gut 1991; 32: 823±7. 3 Lieberman DA, Wuerker CK, Katon RM. Cardiopulmonary risk of esophagogastroduodenoscopy. Role of endoscope diameter and systemic sedation. Gastroenterology 1985; 88: 468±72. 4 Fisher NC, Bailey S, Gibson JA. A prospective, randomized controlled trial of sedation vs. no sedation in out-patient diagnostic upper gastrointestinal endoscopy. Endoscopy 1998; 30: 21±4. 5 Thompson DG, Lennard-Jones JE. Evans SJ, Cowan RE, Murray RS, Wright JT. Patients appreciate premedication for endoscopy. Lancet 1980; 30: 469±70. 6 Reed MWR, O'Leary DP, Duncan JL, Majeed AW, Wright B, Reilly CS. Effects of sedation and supplemental oxygen during upper alimentary tract endoscopy. Scandinavian Journal of Gastroenterology 1993; 28: 319±22. 7 Bell GD, Brown S, Morden A, Coady T, Logan RFA. Prevention of hypoxaemia during upper gastrointestinal endoscopy by means of oxygen via nasal cannulae. Lancet 1987; i: 1022±4. 8 Brandl S, Borody TJ, Andrews P, Morgan A, Hayland L, Devine M. Oxygenating mouthguard alleviates hypoxia during gastroscopy. Gastrointestinal Endoscopy 1992; 38: 415± Block R, Jankowski J, Johnston D, Colvin JR, Wormsley KG. The administration of supplementary oxygen to prevent hypoxia during upper alimentary endoscopy. Endoscopy 1993; 25: 269± Rigg JD, Watt TC, Tweedle DEF, Martin DF. Oxygen saturation during endoscopic retrograde cholangiopancreatography: a comparison of two protocols of oxygen administration. Gut 1994; 35: 408± Bell GD, Quine A, Antrobus JHL, Morden A, Burridge SM, Lee J, Coady TJ. Upper gastrointestinal endoscopy: a prospective randomised study comparing continuous supplemental oxygen given via the nasal or oral route. Gastrointestinal Endoscopy. 1992; 38: 319± Berthoud M, Read DH, Norman J. Pre-oxygenation ± how long? Anaesthesia 1983; 38: 96± Drummond GB, Park GR. Arterial oxygen saturation before intubation of the trachea. British Journal of Anaesthesia 1984; 56: 987± Sellers WFS, Higgs CMB. Comparison of tracheal oxygen concentrations using Hudson mask, nasal cannula and nasal catheter. Anesthesia and Analgesia 1987; 66: S Fairfield JE, Goroszeniuk T, Tully AM, Adams AP. Oxygen delivery systems ± a comparison of two devices. Anaesthesia 1991; 46: 135± q 2000 Blackwell Science Ltd