Supplementary material, on-line appendix 1
Subgroup analyses of difficult tracheal intubation
Subgroup analysis: depolarizing versus non-depolarizing NMBA. In the subgroup of trials where depolarizing NMBA was used there was an increased risk of DTI with no use of NMBA, RR 5.79 (2.64 to 12.72, 95% CI, P < 0.0001, NNTH = 5.6, I2 = 69%). Likewise, the subgroup using non-depolarizing NMBA a DTI was associated with no use of NMBA with a RR 4.72 (3.17 to 7.02, 95% CI, P < 0.0001, NNTH = 5.8, I2 = 74%). There was no significant subgroup differences, P = 0.65.
Subgroup analysis: remifentanil versus other opioids. In 11 trials 26, 29, 32-34, 38, 44, 46, 50, 54, 59remifentanil was used as opioid. However, in four studies 26, 34, 44, 59 remifentanil was only used in the intervention groups while no opioids were used in the control group. In one 26 of the three trials remifentanil and alfentanil were used as opioids in two different intervention groups. In another three trials 32, 33, 38 the amount of remifentanil used in the intervention group exceeded the amount used in the control group. Among trials using equal amount of remifentanil in the intervention- and control groups 29, 46, 50, 54 there was a significant increased risk of DTI when avoiding NMBA for tracheal intubation, RR 15.86 (4.43 to 56.71, 95% CI, P < 0.0001, NNTH = 6.4, I2 = 0%). In trials where other opioids than remifentanil were used, avoiding NMBA was significantly associated with DTI with RR 5.20 (3.53 to 7.64, 95% CI, P < 0.0001, NNTH = 6.5, I2 = 74%). There was no significant subgroup differences, P = 0.10. One trial did not use any opioids 37.
Subgroup analysis: alfentanil versus other opioids. In 13 trials 26-28, 30, 31, 35, 41, 49, 52, 53, 55, 56, 58alfentanil was used as opioid. Five trials 26, 28, 53, 56, 58only used alfentanil in the intervention group while no opioids were used in the control group. In one of the five trials remifentanil and alfentanilwere use as opioids in two different intervention groups 26. In two trials the amount of alfentanil used in the intervention group exceeded the amount used in the control group 30, 55. Among the six trials with equal amount of alfentanil in the intervention- and control groups there was a significant increased risk of DTI when avoiding NMBA for tracheal intubation, RR 4.46 (1.66 to 11.98, 95% CI, P = 0.0002, I2 = 79%, NNTH = 5.2) 27, 31, 35, 41, 49, 52. In trials using other opioids than alfentanil the risk of DTI increased when avoiding NMBA, RR 5.10 (3.34 to 7.79, 95% CI, P < 0.0001, I2 = 72%, NNTH = 4.8). There were no significant subgroup differences, P = 0.81. One trial did not use any opioids 37.
Subgroup analysis: local anaesthetic versus no use of local anaesthetic. A local anaesthetic was used in eight trials. However, three trials 37, 52, 56used local anaesthetic only in the intervention group and not in the control group.Thus, we included five trials 27, 31, 34, 50, 57all using local anaesthesia for intubation in both control and intervention groups. Comparing the trials using local anaesthesia with the trials not using local anaesthesia the risk of DTI was increased with no use of NMBA. When using local anaesthesia, RR 1.90 (1.14 to 3.18, 95% CI, P = 0.01, NNTH = 9.4, I2 = 35%) and with no use of local anaesthesia, RR 6.26 (4.15 to 9.44, 95% CI, P < 0.0001, NNTH = 5.7, I2 = 69%), respectively. There was a significant subgroup differences, P < 0.0001.
Subgroup analysis: exclusion of patients with expected DTI versus no exclusion of patients with expected DTI.In the subgroup where patients with expected DTI were excluded from the individual trials, there was a significantly increased risk of DTI when avoiding NMBA, RR 5.32 (3.54 to 8.00, 95% CI, P < 0.00001, NNTH = 6.3, I2 = 72%).In the subgroup where there was no exclusion of patients with expected DTI there was a significantly increased risk of DTI when avoiding NMBA, RR 4.40 (1.71 to 11.29, 95% CI, P = 0.0002, NNTH = 5.7, I2 = 74%). There were no significant subgroup differences, P = 0.72.
Subgroup analyses of upper airway discomfort or injury
Subgroup analysis: depolarising versus non-depolarizing NMBA. In the two trials where depolarising NMBA was used there was no significant association between avoidance of NMBA and upper airway discomfort or injury, RR 1.48 (0.83 to 2.65, 95% CI, P = 0.19, I2 = 0%) 33, 55. Likewise in the subgroup of five trials using non-depolarizing NMBA there was no significant association, RR 1.37 (0.97 to 1.94, 95% CI, P = 0.07, I2 = 15%) 29, 30, 32, 45, 46. There was no significant subgroup difference, P = 0.83.
Subgroup analysis: remifentanil versus other opioids. In the two trials 29, 46 where remifentanil was used there was no significant association between avoidance of NMBA and upper airway discomfort or injury, RR 1.12 (0.61 to 2.08, 95% CI, P = 0.14, I2 = 55%). In the five trials using other opioids than remifentanil there was a significant association between avoiding NMBA and upper airway discomfort or injury?, RR 1.42 (1.16 to 1.75, 95% CI, P = 0.0009, NNTH = 7.5, I2 = 0%) 30, 32, 33, 45, 55. There were no significant subgroup differences, P = 0.47.
Subgroup analysis: alfentanil versus other opioids. In two trials the amount of alfentanil used in the intervention group exceeded the amount used in the control group, thus this trial was excluded from the meta-analysis 30, 55. In the remaining trials other opioids than alfentanil were used 29, 32, 33, 45, 46. Here there was no significant association between no use of NMBA and upper airway discomfort or injury, RR 1.47 (0.85 to 2.53, 95% CI, P = 0.17, I2 = 49%).
Subgroup analysis: local anaesthetic versus no use of local anaesthetic. None of the trials reporting upper airway discomfort or injury used local anaesthetic.
Subgroup analysis: exclusion of patients with expected DTI versus no exclusion of patients with expected DTI. Patients with anticipated difficult airway management were excluded in six of the trials reporting upper airway discomfort or injury 29, 30, 32, 45, 46, 55.There was a significantly increased risk of upper airway discomfort or injury when not relaxing the patient with a RR 1.37 (1.05 to 1.79, 95% CI, P = 0.02, NNTH = 8.9, I2 = 29%). One trial included participants with expected difficult airway management 33. Here there was no significant association between avoiding NMBA and upper airway discomfort or injury, RR 2.00 (0.39 to 10.31, 95% CI, P = 0.41. There was no significant subgroup difference.