Hong Kong Journal of Emergency Medicine A study of orotracheal intubation in emergency departments of five district hospitals in Hong Kong YF Choi, TW Wong, CC Lau, AYC Siu, CB Lo, MC Yuen, WK Tung, P Ng, CW Kam, TK Mui, WL Yuen, B Lim, ACH Lit Objective: To study the success rates and complications of orotracheal intubation in emergency departments of five district hospitals in Hong Kong in order to identify ways for improvement. Method: This was a prospective observational study. The emergency department doctors performing the intubation were asked to complete an intubation study form immediately after the procedure over a period of four months. Data collected included vital signs, experiences of intubators, method of intubation and complications. Results: A total of 347 cases were collected and 93% of them were non-trauma cases. Fifty-two percent (52%) of the cases were in cardiac arrest before intubation. Rapid sequence intubation (RSI) was applied in 36% of the cases. Junior doctors first intubated about 72% of the patients. Successful intubation was achieved in 1 and 2 attempts in 70% and 89% of the cases respectively. In 10 cases (3%), secondary methods such as laryngeal mask airway, Combitube, Trachlight or cricothyroidotomy were needed. The overall complication rate was 7.8% and the complication rate in the RSI group was 15.3%. The complication rate was even higher (20%) if intubation without medication was used in non-cardiac arrest patients. Significant drop in blood pressure was the most common complication and it could be attributed to the use of midazolam as induction medication. The success rate was found to correlate with the experience of the first intubator (p<0.05) and the laryngeal view (p<0.001). The complication rate increased with repeated attempts (p<0.001) and was higher among junior doctors (p<0.05). Early use of elastic gum bougie was associated with lower complication and higher success rates. Conclusion: Orotracheal intubation in the emergency department was associated with high complication rate. Many complications came from junior intubators. Hypotension was the most common complication. Potentially avoidable complications may be a result of failure to use RSI in non-cardiac arrest patients and failure to use bougie in cases of poor laryngeal view. (Hong Kong j.emerg.med. 2003;10:138-145) Keywords: Complications, rapid sequence intubation Correspondence to: Choi Yu Fai, MBBS(HK), FRCSEd Pamela Youde Nethersole Eastern Hospital, Accident and Emergency Department, 3 Lok Man Road, Chaiwan, Hong Kong Email: choiyf@netvigator.com Wong Tai Wai, FRCSEd, FHKCEM, FHKAM(Emergency Medicine) Lau Chor Chiu, MRCP(UK), FHKCEM, FHKAM(Emergency Medicine) North District Hospital, Accident and Emergency Department, 9 Po Kin Road, Sheung Shui, N.T., Hong Kong Siu Yuet Chung, Axel, FRCSEd, FHKCEM, FHKAM(Emergency Medicine) Lo Chi Biu, FRCSEd, FHKCEM, FHKAM(Emergency Medicine) Kwong Wah Hospital, Accident and Emergency Department, 25 Waterloo Road, Kowloon, Hong Kong Yuen Man Cheuk, MBBS(HK), FRCSEd, FHKAM(Emergency Medicine) Tung Wai Kit, FRCSEd, FHKCEM, FHKAM(Emergency Medicine) Tuen Mun Hospital, Accident and Emergency Department, Tsing Chung Koon Road, Tuen Mun, N.T., Hong Kong Paulin Ng, FRCSEd, FHKCEM, FHKAM(Emergency Medicine) Kam Chak Wah, MRCP(UK), FRCSEd, FHKAM(Emergency Medicine) Caritas Medical Centre, Accident and Emergency Department, Shamshuipo, Kowloon, Hong Kong Mui Tsz Kuen, FRCSEd, FHKCEM, FHKAM(Emergency Medicine) Yuen Wing Lok, FRCSEd, FHKAM(Surgery), FHKAM(Emergency Medicine) Yan Chai Hospital, Accident and Emergency Department, Tsuen Wan, N.T., Hong Kong Lim Bun, MRCP(UK), FHKCEM, FHKAM(Emergency Medicine) Lit Chau Hung, Albert, MRCP(UK), FRCSEd, FHKAM(Emergency Medicine)
Choi et al./orotracheal intubation in ED 139 Introduction Emergency airway management in the emergency department (ED) is an important life saving procedure and it is not uncommon to encounter difficulties and complications, which make this procedure challenging. An appropriate method of controlling the airway and the safety of the procedure are our main concern. The reported complication rates of emergency intubation by ED doctors in early studies were alarmingly high. 1 The application of rapid sequence intubation (RSI) has gained popularity as it has been shown to reduce the complication rate of intubation in certain groups of patients. 2-4 However, intubation related complications were still encountered quite commonly despite the availability of guidelines. In this study, we would like to examine the methods, success rates and complication rates of orotracheal intubation performed by ED doctors in five district hospitals in Hong Kong. Our aim was to identify weaknesses so that we could find some ways to improve the outcome. Methods Study design and setting This was a prospective observational study carried out in the ED of five district hospitals in Hong Kong. All the five district hospitals involved were recognized training centres for emergency medicine. None of them were tertiary referral centres or university teaching centres. The study was carried out in the five hospitals at the same time over a period of four months. Study protocol All patients who required orotracheal intubation in the ED and were intubated first by ED doctors were included. Those who had pre-hospital intubation or intubation by other specialists such as anaesthetists or paediatricians were excluded. Nasotracheal intubation was rarely performed in local EDs and was excluded. The attending ED doctors were asked to complete an intubation audit form immediately after the procedure. Measurements The intubators were required to complete an intubation study form (Appendix), recording demographic data of the patients, the initial diagnoses, the rank and training status of the intubators by sequence of attempts and their choice of intubation methods. The use of bougie, the total number of attempts and the laryngeal view by direct laryngoscopy were recorded for their initial attempts of RSI. All medications given to the patients for the purpose of intubation including pre-medications, induction agents and muscle relaxants were entered. Secondary methods of airway access e.g. laryngeal mask airway after failing endotracheal intubation were also recorded. The haemodynamic status and oxygen saturation of patients just before and five minutes after intubation were also noted. Finally, the intubators were asked to report the occurrence of any complications that were related to intubation. The data were verified by the research co-ordinator in each center. Definitions A single attempt was defined as an act of trying to view the vocal cord using the direct laryngoscope. It was a failed attempt if the tube was not negotiated through the vocal cords or if it was wrongly inserted into the oesophagus. ED doctors were defined as junior if they were pre-fellowship medical officers and defined as senior if they were either senior medical officers or post-fellowship medical officers. Significant hypotension was defined by a decrease in systolic blood pressure below 90 mmhg or a reduction by more than 20% five minutes after intubation. Desaturation was defined by falling of pulse oximetry saturation below 85% at any time within five minutes of intubation. Data processing The authors collected the intubation audit forms at the end of the four-month period and Statistical Package for Social Science (SPSS) software was used for data input and analysis. Descriptive analysis was applied to show the frequency and percentage of certain items such as demographic data, success rate and complication rate. Chi-squared test for significant difference was applied to compare the performance of various sub-groups.
140 Hong Kong j. emerg. med. Vol. 10(3) Jul 2003 Results (1) General A total of 347 cases were collected from the five hospitals during the four-month study period. The male to female ratio was 1.3:1. The mean age was 66.2 years with a range from one month to 98 years old. There were only six patients who were below seventeen and there were eight adult cases of unknown age. Non-traumatic cases accounted for 92.8% of the total, and 52.4% of the patients were in cardiac arrest before intubation. The initial ED diagnoses are shown in Table 1. RSI was performed in 35.7% (n=124) while intubation without medication represented 60.8% (n=211) as 52.4% of the patients were in cardiac arrest. The remaining 12 cases were intubated with sedation only, without using neuromuscular blocking agents. No medication would usually be given for intubation if the patient was in cardiac arrest. However, 35 out of the 165 non-cardiac arrest patients (21.2%) did not receive any medication before they were intubated. (2) Success rate Successful intubation could be achieved in the first attempt in 70.0% and cumulatively within two attempts in 88.5% of the cases. There were 35 cases (10.1%) that required three attempts and five cases (1.1%) required more than three attempts. Secondary methods of airway access were used in 10 cases (2.9%). These secondary methods were carried out after one to four unsuccessful attempts by direct laryngoscopy alone. An airway could finally be established successfully by Trachlight (4 cases), laryngeal mask airway (3 cases), Combitube (2 cases) and cricothyroidotomy (1 case). Elastic gum bougie or equivalent endotracheal-tube introducers were used in 26 cases (7.5% of all cases) and all of them were associated with poor laryngoscopic view. In nine cases bougie was used in the first attempt and in the rest bougie was resorted to after the initial attempts had failed. Bougie failed to establish an airway in one case only and that patient was the one who finally needed a cricothyroidotomy. Junior doctors performed the first attempt of intubation in 72.4% of the cases. We found that the experience of the first intubator was significantly associated with the success rates in the first attempt (p<0.05, Table 2) as well as within two attempts (p<0.01, Table 3). In the 36 cases where the first intubators were juniors and intubations could not be established within two attempts, senior supervision Table 1. Initial ED diagnoses of all patients ED diagnosis Frequency Percentage Cardiopulmonary arrest 182 52.4 Respiratory failure 50 14.4 Coma 29 8.4 Acute pulmonary oedema 24 6.9 Cerebrovascular accident 21 6.1 Head injury 8 Drug overdose 9 Convulsion 5 Airway obstruction 3 Persistent shock 3 Facial injury 2 Burn injury 1 Neck or airway trauma 1 Others 9 Total 347 Table 2. Success rates by 1st attempt according to 1st intubators' experience 1 attempt 2 or more Total attempts Junior as 1st intubator 168 83 251 Senior as 1st intubator 75 21 96 Total 243 104 347 (Chi-squared: p=0.048) Table 3. Success rates within 2 attempts according to 1st intubators' experience Within More than Total 2 attempts 2 attempts Junior 1st intubator 215 36 251 Senior 1st intubator 92 4 96 Total 307 40 347 (Chi-squared: p=0.008)
Choi et al./orotracheal intubation in ED 141 and intervention were present in 29 of them (80.6%). Besides, the overall success rate of intubation was also found to correlate with direct laryngoscopic view (p<0.001, Table 4). (3) Complications There were totally 35 complications reported in 27 patients (Table 5). The observed complication rate was 7.8%. The most common complication was significant reduction in blood pressure. Hypotension was reported in 10 cases after intubation, of which two had already been borderline hypotensive before intubation. In the remaining eight cases, six could be attributed to the use of medications (one had thiopentone and five had midazolam). Oesophageal intubation occurred in nine cases and eight of them were detected immediately after checking tube position and one was detected by chest X-ray. There were totally six cases of local trauma including dental trauma and oral bleeding. Oxygen saturation by pulse oximetry was found to be below 85% in four cases and it was suspected that these could be attributed to poor peripheral circulation as the endotracheal tube positions were verified to be correct. Right bronchial intubation occurred in two patients. One transient cardiac arrest was recorded shortly after four attempts of intubation without any medication in a 92 years old comatose patient. Vocal cord was not visualized by direct laryngoscopy and bougie was not used in all the four attempts. The systolic blood pressure was 98 mmhg before intubation and was 108 mmhg five minutes after intubation. The cardiac arrest was transient and responded to resuscitation and the patient was admitted to the intensive care unit. After investigating the complication rates of various sub-groups, it was found that the complication rate in the RSI intubation group was 14.0%. This was significantly higher than the group of patients intubated without medication (p=0.002; Table 6). Complication rates were also significantly higher if the first intubators were junior (p<0.05, Table 7) and if the total attempts were more than two (p<0.001, Table 8). The complication rate was also higher if vocal Table 4. Successful intubation within 2 attempts and laryngeal view Vocal cord Vocal cord Total seen not seen Within 2 attempts 261 32 293 More than 2 attempts 23 17 40 Total 284 49 #333 (Chi-squared: p<0.001; #Laryngeal views were missing in 14 cases) Table 5. The incidence of various complications Complications Incidence Hypotension 10 Oesophageal intubation 9 Local trauma 6 Desaturation 4 Right bronchial intubation 2 Pneumothorax 1 Tube dislodgement 1 Aspiration 1 Cardiac arrest 1 Total 35 Table 6. Complication rates in RSI and no medication groups No complication Complication Total (%) No medication 202 9 211 (4.5%) RSI/sedation 118 18 136 (15.3%) Total 320 27 347 (7.8%) (Chi-squared: p=0.002) Table 7. Complication rates and 1st intubators' experience No complication Complication Total Junior as 1st intubator 225 24 249 Senior as 1st intubator 95 3 98 Total 320 27 347 (Chi-squared: p=0.039) Table 8. Complication rates and repeated attempts No complication Complication Total Within 2 attempts 292 15 307 More than 2 attempts 28 12 40 Total 320 27 347 (Chi-squared: p<0.001)
142 Hong Kong j. emerg. med. Vol. 10(3) Jul 2003 cord was not visualized in initial direct laryngoscopy but it was not statistically significant. No complication was reported in 25 cases using bougie within two attempts. On the other hand, in 35 non-cardiac arrest patients intubated without medication, seven complications (20.0%) were reported. (4) RSI usage In the RSI group, we have investigated the choice of medications. Lignocaine with or without fentanyl were commonly used as pre-medication in cases of head injury or suspected intracerebral haemorrhage in order to prevent rise in intracranial pressure during intubation. However, defasciculation using nondepolarizing neuromuscular blockers was rarely used for this purpose. The most popular neuromuscular blocking agent was suxamethonium (110 cases). Concerning the drug of choice for induction, midazolam was the most popular agent while etomidate was the drug of choice in one of the five hospitals. Thiopentone and ketamine were occasionally used. Significant drop in blood pressure was found to occur only if thiopentone (one out of two) or midazolam (five out of 68) was used as induction agent. The dose of midazolam ranged from 2 mg to 5 mg, which was already a low dose comparing to the recommended dose of 0.2-0.3 mg per Kg. The mean systolic blood pressures before and after using midazolam in the 68 cases were 170.1 mmhg and 152.5 mmhg respectively and the difference was equivalent to a 10% drop (two tailed t-test: p=0.001). No significant drop in blood pressure was found for ketamine and etomidate. The use of neuromuscular blocking agents was not found to be associated with hypotension. (5) Patient outcomes Concerning the outcome after intubation in ED, 127 out of 182 patients (69.8%) who were in cardiac arrest on arrival were certified dead in ED. The outcomes after admission were not followed in this study. Discussion Gaining access to the airway in an emergency department is an essential skill for ED doctors. RSI has become more and more widely practised in local EDs. RSI has been shown by many recent studies to reduce certain complications such as struggle-related airway trauma and aspiration 5,6 but it has its own problems too. We reported an overall intubation complication rate of 7.8%, which was similar to other studies with similar settings. (Table 9) However, the complication rate of 15.3% in our RSI group was much higher than the overall rate. The difference could be partly explained by the fact that in the non-rsi group, 66% were in cardiac arrest and most of them never regained circulation. On the other hand, most of the reported complications in the RSI group were hypotension. If haemodynamic related complication was disregarded, the complication rate in the RSI group would be reduced to 7.4%. However, hypotension is a potentially serious complication that cannot be disregarded. It is found to be associated with a high Table 9. Complication rates in some other studies Study Complication rate Remarks Dufour 4 6.5% RSI study in ED Tayal 5 10.6% RSI study in ED Sackles 6 8.0% Both RSI and non-rsi in ED Rotondo 9 12% Trauma cases Hedges 10 21% Pre-hospital by paramedics Syverud 11 28% Pre-hospital by paramedics on helicopter
Choi et al./orotracheal intubation in ED 143 incidence of mortality and morbidity. 7,8 Our RSI group's complication rate was higher than expected when compared with other studies of ED intubation. Other intubation studies revealed that the complication rates were higher in trauma cases and in pre-hospital settings 9-11 but intubations using RSI done by ED doctors in the emergency departments had the lowest complication rate. Therefore we should find out the reason for such a high complication rate in our RSI group. Hypotension was the most common complication and it was suspected that the choice of induction agent might be the cause. No hypotension was reported if ketamine or etomidate was used as induction agent while significant decrease in blood pressure was observed while midazolam was used. Further study is necessary to compare the haemodynamic effect of various induction agents. We observed that both complication rate and success rate were related to the experience of the intubators. In the 36 cases in which the first intubators were juniors and more than two attempts were needed to secure the airway, complications were reported in 11 of them (30.6%). In Hong Kong, emergency medicine trainees do not have formal anaesthesiology attachment and they gain their experiences in airway control through on-job training. For this reason the majority of the intubations (72%) were first attempted by juniors. Ideally, senior staff should attend to critical patients who need prompt airway management in ED. In practice, junior ED staff should gain experience by performing airway intubation under close support and supervision of their seniors. Difficult airway should be anticipated and immediate senior supervision and intervention should be available. In addition, regular workshops on intubation training using manikins might be very helpful. Our results showed that in 21% of the non-cardiac arrest cases, intubation without medication was performed instead of using RSI. Intubation without sedation and paralysis in conscious or semiconscious patients are unpleasant to the patients and it was reported to be associated with an increased complication rate. We found that the complication rate in this group of patients was 20%, which was higher than average. It is probably desirable to give muscle relaxants to patients who are comatose but not in cardiac or respiratory arrest before intubation even if their jaw muscles were relaxed as unexpected struggling and aspiration could lead to complications. Oesophageal intubation was commonly caused by poor laryngeal view. Immediate detection can reduce harmful complications such as hypoxia, gastric distension and oesophageal trauma. Oesophageal intubation detectors and end-tidal CO 2 detector should be used routinely to avoid such complications as recommended in the Guidelines 2000 of the American Heart Association. 12 As oesophageal intubation is usually related to poor laryngeal view, the use of elastic gum bougie or a similarly designed stylet could be another way of reducing this complication. Some authors recommended the use of bougie in difficult cases or if vocal cord could not be directly visualized. It is probably helpful in reducing intubation time and complications if used early. 13-15 Our data showed that bougie was not always used even if vocal cord was not seen. Vocal cords were not visualized by direct laryngoscopy in 49 cases in our study but bougie was used in 26 cases only. Furthermore, in 17 of these 26 cases, bougie was probably not used in the first intubation attempt. The complication rate for intubation using bougie was 3.8% (1 in 26 cases) and no complication was reported if bougie was used within two attempts. ED doctors should be trained to use the bougie and early consideration for its use should be encouraged. Lastly, there are several limitations of this study. The final outcomes of the patients after leaving ED were not followed so some late complications of intubation such as aspiration pneumonia might be under-reported. Owing to the small number of cases, we could not compare haemodynamic effects
144 Hong Kong j. emerg. med. Vol. 10(3) Jul 2003 of all the agents used in RSI. Moreover, in many returned audit forms, the doses of medications were not entered and we could not assess whether haemodynamic disturbance of a drug was doserelated. Conclusions Orotracheal intubation performed in ED was associated with a significant complication rate. Junior intubators reported a significantly lower success rate and a higher complication rate. Hypotension was the most common complication and we suspected that the use of midazolam as induction agent might be the cause. Intubation without medication for non-arrested patient should be avoided because of the associated higher complication rate. The early use of bougie when encountering poor laryngeal view should be encouraged to avoid oesophageal intubation. Acknowledgement The authors would like to express sincere appreciation to all ED staff who had contributed to the data collection. References 1. Taryle DA, Chandler JE, Good JT Jr, Potts DE, Sahn SA. Emergency room intubations--complications and survival. Chest 1979;75(5):541-3. 2. Fortney JP, Bodner M, Lewis LM. Comparison of rapid sequence intubation with conscious sedation or awake intubation in the ED. Acad Emerg Med 1996;4:458. 3. Li J, Murphy-Lavoie H, Bugas C, Martinez J, Preston C. Complications of emergency intubation with and without paralysis. Am J Emerg Med 1999;17(2): 141-3. 4. Dufour DG, Larose DL, Clement SC. Rapid sequence intubation in the emergency department. J Emerg Med 1995;13(5):705-10. 5. Tayal VS, Riggs RW, Marx JA, Tomaszewski CA, Schneider RE. Rapid-sequence intubation at an emergency medicine residency: success rate and adverse events during a two-year period. Acad Emerg Med 1999; 6(1):31-7. 6. Sakles JC, Laurin EG, Rantapaa AA, Panacek EA. Airway management in the emergency department: a one-year study of 610 tracheal intubations. Ann Emerg Med 1998;31(3):325-32. 7. Schwartz DE, Matthay MA, Cohen NH. Death and other complications of emergency airway management in critically ill adults. A prospective investigation of 297 tracheal intubations. Anesthesiology 1995;82(2): 367-76. 8. Franklin C, Samuel J, Hu TC. Life-threatening hypotension associated with emergency intubation and the initiation of mechanical ventilation. Am J Emerg Med 1994;12(4):425-8. 9. Rotondo MF, McGonigal MD, Schwab CW, Kauder DR, Hanson CW. Urgent paralysis and intubation of trauma patients: is it safe? J Trauma 1993;34(2):242-6. 10. Hedges JR, Dronen SC, Feero S, Hawkins S, Syverud SA, Shultz B. Succinylcholine-assisted intubations in prehospital care. Ann Emerg Med 1988;17(5): 469-72. 11. Syverud SA, Borron SW, Storer DL, et al. Prehospital use of neuromuscular blocking agents in a helicopter ambulance program. Ann Emerg Med 1988;17(3): 236-42. 12. Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 6: advanced cardiovascular life support: section 3: adjuncts for oxygenation, ventilation and airway control. The American Heart Association in collaboration with the International Liaison Committee on Resuscitation. Circulation 2000;102(8 Suppl):I95-104. 13. Nolan JP, Wilson ME. Orotracheal intubation in patients with potential cervical spine injuries. An indication for the gum elastic bougie. Anaesthesia 1993; 48(7):630-3. 14. Gataure PS, Vaughan RS, Latto IP. Simulated difficult intubation. Comparison of the gum elastic bougie and the stylet. Anaesthesia 1996;51(10):935-8. 15. Moscati R, Jehle D, Christiansen G, et al. Endotracheal tube introducer for failed intubations: a variant of the gum elastic bougie. Ann Emerg Med 2000;36(1): 52-6.
Choi et al./orotracheal intubation in ED 145 Appendix