Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Laryngeal mask airway versus endotracheal intubation or bag-mask ventilation for cardiac arrest in adults

Three Part Question

In [adults with cardiac arrest requiring airway management] is the [classic laryngeal mask airway (LMA) optimal to endotracheal intubation (ETI) or bag-mask ventilation] in terms of [improved outcomes]?

Clinical Scenario

A 30-year-old woman suffers a witnessed out-of-hospital cardiac arrest. You lack experience with ETI and wonder what evidence supports routine use of LMA compared to ETI or bag mask ventilation for airway management in cardiac arrests.

Search Strategy

MEDLINE (OVID, 1950 to October Week 4 2009), EMBASE (1980 to 2009 Week 44), CINAHL (1982 to October 2009), Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 4, 2009), and Google Scholar. Bibliographies of retrieved articles were reviewed for additional references.

MEDLINE, EMBASE, CINAHL and Cochrane Central Register of Controlled Trials search: [laryngeal mask$.mp OR exp Laryngeal Masks (MeSH) OR] AND [cardiac or exp Heart Arrest (MeSH) OR OR exp Cardiopulmonary Resuscitation (MeSH) OR], limited to humans (EMBASE and MEDLINE), adults and English language (CINAHL, EMBASE and MEDLINE).

Google Scholar search: “laryngeal mask airway OR LMA” and “arrest” in the title of the article.

Search Outcome

Altogether 83 papers were found in MEDLINE, 65 in EMBASE, 12 in CINAHL, 34 in The Cochrane Central Register of Controlled Trials and 8 in Google Scholar, of which 7 were relevant or of sufficient quality. No further papers were found by scanning the references of relevant papers. All relevant papers are summarised in the table.

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Yoshida et al,
66 witnessed out-of-hospital cardiac arrest patients (mean age 61.6 years), including 31 resuscitated with LMA and 13 with tracheal tube (TT). No statistical significance in patient characteristics, sex, complication, 24 hr-survival rate or the causes of cardiac arrest. Prospective, non-randomised controlled trialArterial blood gas (ABG) pHLMA: 7.090 vs TT: 7.050Small study, not randomised. Significantly larger LMA group compared to TT. ABG collected after hospital admission. No data on HCO3, which could affect pH. Full details of methods not reported.
ABG PaC02LMA: 56.86 mmHg vs TT: 45.56 mmHg
ABG Pa02LMA: 183.431 mmHg vs TT: 190.544 mmHg
Samarkandi et al,
Saudi Arabia
20 adults (mean age 54 ± 24 years) including 5 adults resuscitated with endotracheal tube (ETT) vs 7 adults given LMA, following in-hospital cardiac arrest with either asystole or severe bradycardia. Prospective, non-randomised controlled trialOxygen saturation (mean ± SD)LMA: 50-98% (74±24%) vs ETT: 77-89% (83±6%)Very small numbers, not randomised or standardised. Small number of successful survivors to compare functional outcomes. Data poorly presented preventing some outcomes being interpreted.
RegurgitationLMA: 0 patients vs ETT: 0 patients
SOS-KANTO study group,
322 patients (mean age 64) with bystander witnessed cardiac-verified out-of-hospital ventricular fibrillation or pulseless ventricular tachycardia. 173 LMA vs 200 bag-valve-mask (BVM).Prospective multicenter, non-randomized control trial ABG pH (median)LMA: 7.117 vs BVM: 7.075 (p=0.02)Non-randomised study. Inability of the authors to check that blood gas samples were taken from arteries. Timing of measuring blood gases was after the transfer of patients to the hospital.
ABG PaCO2 (median)LMA: 52.9 mmHg BMV: 55.3 mmHg (p=0.06)
ABG PaO2 (median) LMA: 64.6 mmHg vs BVM: 71.9mmHg (p=0.56)
Return of spontaneous circulationLMA: 38.7% vs BVM: 40% (p=0.89)
Admitted to hospitalLMA: 27.3% vs BVM: 25.1% (p=0.74)
Survival at 24hrsLMA: 21.5% vs BVM: 17.3% (p=0.48)
Survival at 7 daysLMA: 15.7% vs BVM: 10.7% (p=0.20)
Survival to hospital dischargeLMA: 13.4% vs BVM: 6.1% (p=0.03)
Stone et al,
In-hospital cardiac arrest patients. 466 BVM or BVM followed by ETI vs 86 LMA or LMA followed by ETI.Prospective, non-randomised, comparative studyRegurgitation during CPRLMA: 3.5% vs BVM: 12.4% (p<0.05)Study not randomised or standardised. No patient demographics or details of cardiac arrest provided. Significantly more BVM than LMA patients studied. Not just BVM or LMA alone, but also followed by ETI in some cases. Number of secondary ETI cases not provided.
Regurgitation after CPRLMA: 0% vs BVM: 1.5%
Ocker et al,
An experimental bench model simulating an unintubated patient with cardiac arrest. 20 paramedics performed ventilation, including with BVM and LMA.Prospective comparative studyTime to first adequate tidal lung volume (median)LMA: 35 seconds vs BVM: 9 seconds (p<0.001)Study uses a single bench model, not actual patients. Does not account for respiratory system compliance such as that seen during CPR.
Lung tidal volume (mean ± SEM)LMA: 743 ± 70ml vs BVM: 353 ± 26ml (p<0.001)
Peak airway pressure (mean ± SEM)LMA: 21 ± 2 cm H20 vs BVM: 17 ± 2 cm H20 (p<0.05)
Peak oesophageal pressure (mean ± SEM)LMA: 2 ± 1 cm H20 vs BVM: 15 ± 1 cm H20 (p<0.001)
Gastric tidal volume (mean ± SEM)LMA: 25 ± 13ml vs BVM: 313 ± 30ml (p<0.001)
Dorges et al,
An experimental bench model simulating an unintubated patient with cardiac arrest. 31 non-anaesthesia house officers performed ventilation, including with BVM and LMA (with adult self-inflating bag).Prospective, experimental, comparative studyTime to first adequate tidal lung volume (median)LMA: 29 seconds vs BVM 14 seconds (p<0.01)Study only uses a single bench model, not actual patients. Does not account for respiratory system compliance such as that seen during CPR.
Lung tidal volume (mean ± SEM)LMA: 727 ± 53ml vs BVM 271 ± 33ml (p<0.01)
Peak airway pressure (mean ± SEM)LMA: 20 ± 1 cm H20 vs BVM: 14 ± 1 cm H20 (p<0.01)
Peak oesophageal pressure (mean ± SEM)LMA: 3 ± 0.5 cm H20 vs BVM: 12 ± 1 cm H20 (p<0.01)
Oesophageal tidal volume (mean ± SEM)LMA: 8 ± 3ml vs BVM: 272 ± 24ml (p<0.01)
Ventilation failuresLMA: 10% vs BVM: 6.5%
Stomach inflationLMA: 7/28 volunteers vs BVM: 29/29 volunteers (p<0.01)
Dorges et al,
An experimental bench model simulating an unintubated patient with cardiac arrest. 21 student nurses performed ventilation with BVM and LMA.Prospective, experimental, comparative studyTime to first adequate tidal lung volume (median)LMA: 37 seconds vs BVM: 22 seconds (p<0.01)Study only uses a single bench model, not actual patients. Does not account for respiratory system compliance such as that seen during CPR.
Lung tidal volume (mean ± SD)LMA: 715 ± 416ml vs BVM: 243 ± 131ml (p<0.01)
Peak airway pressure (mean ± SD)LMA: 26 ± 10 cm H20 vs BVM: 14 ± 5 cm H20 (p<0.01)
Peak oesophageal pressure (mean ± SD)LMA: 4 ± 5 cm H20 vs BVM: 15 ± 6 cm H20 (p<0.01)
Gastric volume (mean ± SD)LMA: 0.6 ± 0.8 l/min vs BVM: 3.0 ± 2.1 l/min (p<0.01)
Lung volume (mean ± SD)LMA: 15.0 ± 6.6 l/min vs BVM: 4.8 ± 2.7 l/min (p<0.01)
Ventilation failuresLMA: 0% vs BVM: 20%
Stomach inflationLMA: 8/21 volunteers vs BVM: 17/17 volunteers


Airway management is critical to successful outcomes during cardiopulmonary resuscitation (CPR). Ensuring adequate oxygenation and ventilation is of primary importance in cardiac arrest patients, whilst also protecting airway patency from risk of regurgitation and pulmonary aspiration. ETI remains the optimal method to provide and maintain a clear and secure airway, however, it requires much skill and experience with a high risk of unrecognised misplaced tracheal tube. Bag-mask ventilation also requires reasonable experience with risk of ineffective tidal volumes and gastric inflation. LMA has been shown to be a reliable and effective alternative airway device, providing rapid attainment of a clear airway in cardiac and non-cardiac studies. However, there is a need to know how well it compares to other devices to support or refute routine use in adults with cardiac arrest. Only two heterogeneous studies directly compare ETI with LMA. One study showed higher oxygen saturation with in-hospital cardiac arrest patients using ETI, however this was not significant. This may be because of the very small size of the study group [2]. The other study showed insignificant differences with ABG pH, PaC02, and Pa02 between ETI and LMA with out-of-hospital cardiac arrest patients [1]. Methodological problems in both studies limit interpretation and applicability of results. There is no comparison of tidal volumes to compare effectiveness of ventilation of LMA with ETI. From the limited available evidence, LMA does not show improved outcomes compared to ETI, although ETI has not demonstrated significant advantage to LMA. Five studies compare LMA with bag-mask ventilation. A series of three studies in a cardiac arrest model show bag-mask ventilation is significantly faster in achieving the first adequate tidal volume [5-7]. However, LMA has significantly higher tidal volume and peak airway pressure, whilst lower peak oesophageal pressure, stomach inflation and gastric tidal volume [5-7]. This is seen even when LMA is delivered with more inexperienced personnel, such as student nurses [7]. This suggests LMA being more effective in terms of ventilation, whilst also safer by reducing gastric inflation. However, the applicability of these results to actual patients may be limited. A relatively large study comparing out-of-hospital cardiac arrest patients found ABG pH was significantly superior with LMA, although not clinically different, whilst Pa02 and PaC02 was insignificant [3]. Thus, LMA may not be greatly beneficial to respiratory status. The group also found survival to hospital discharge significantly higher with LMA than bag-mask ventilation, but this may be attributed to numerous confounders. A study with in-hospital cardiac arrest patients found LMA significantly less likely to cause regurgitation during CPR [4]. Together the evidence on LMA compared with bag-mask ventilation is stronger than ETI, but still lacks homogeneity of patients and outcomes measured, making conclusions still quite weak. However, LMA does seem to offer greater advantages compared to bag-mask ventilation. Randomised controlled trials comparing LMA with ETI and bag-mask ventilation are warranted.

Clinical Bottom Line

LMA appears to offer better outcomes compared to bag-mask ventilation, but not ETI in experienced individuals.


  1. Yoshida R, Hasegawa N, Kobayashi Y. Is airway management with tracheal tube on out-of-hospital cardiac arrest more beneficial than laryngeal mask airway or esophageal-tracheal combitube? Resuscitation. 2006; 70(2):301.
  2. Samarkandi AH, Seraj MA, el Dawlatly A, Mastan M, Bakhamees HB. The role of laryngeal mask airway in cardiopulmonary resuscitation. Resuscitation. 1994; 28(2):103-6.
  3. SOS-KANTO study group. Comparison of arterial blood gases of laryngeal mask airway and bag-valve-mask ventilation in out-of-hospital cardiac arrests. Circulation Journal. 2009; 73(3):490-6.
  4. Stone BJ, Chantler PJ, Baskett PJF. The incidence of regurgitation during cardiopulmonary resuscitation: a comparison between the bag valve mask and laryngeal mask airway. Resuscitation. 1998; 38(1):3-6.
  5. Ocker H, Wenzel V, Schmucker P, Dorges V. Effectiveness of various airway management techniques in a bench model simulating a cardiac arrest patient. Journal of Emergency Medicine . 2001; 20(1):7-12.
  6. Dorges V, Ocker H, Wenzel V, Sauer C, Schmucker P. Emergency airway management by non-anaesthesia house officers – a comparison of three strategies. Emergency Medicine Journal. 2001; 18(2):90-94.
  7. Dorges V, Sauer C, Ocker H, Wenzel V, Schmucker P. Airway management during cardiopulmonary resuscitation—a comparative study of bag–valve–mask, laryngeal mask airway and combitube in a bench model. Resuscitation. 1999; 41(1):63-69.