Skip Ribbon Commands
Skip to main content
SharePoint

PublicComment

 Feedback

If you have any comments or questions on this page, please email us at:
 

Recovery Position

Question Type:
Intervention
Full Question:
Among adults outside of a hospital with a non-traumatic illness who are unresponsive but breathing normally  (P), does positioning in any specific position  (I), compared with a supine or other proposed recovery position (C), change overall mortality, complications, incidence of cardiac arrest, the incidence of aspiration, the likelihood of cervical spinal injury, need for airway management (O)?
Consensus on Science:
Lateral, Side-Lying Recovery Position Compared With Supine Position For the critical outcome of the incidence of aspiration, we identified very-low-quality evidence (downgraded for imprecision) from 1 observational study with a total of 142 patients(Adnet 1999, 745) found in the left lateral decubitus or supine position demonstrating no benefit to being in the left lateral position (relative risk [RR], 0.93; 95% confidence interval [CI], 0.55–1.58). The same observational study had a total of 132 patients found in the right lateral decubitus or supine position and demonstrated no benefit to being in the right lateral position (RR, 1.15; 95% CI, 0.67–1.96). For the critical outcome of need for airway management, only studies with indirect measures of potential need for airway management were identified, including measures of total airway volume and stridor scores. Very-low-quality evidence (downgraded for risk of bias, indirectness, and imprecision) from 1 observational study with 17 patients(Litman 2005, 484) demonstrating the benefit of the lateral position by increasing total airway volume (mean difference [MD], 2.7; 95% CI, 0.88–4.52), and very-low-quality evidence (downgraded for indirectness, and imprecision) from 1 observational study with 30 patients(Arai 2004, 1638) demonstrating the benefit of the lateral position by decreasing stridor score (MD, −0.9; 95% CI, −1.21 to −0.59). HAINES Modified Recovery Position Compared With Lateral Recovery Position For the critical outcome of the likelihood of cervical spinal injury, we identified very-low-quality evidence (downgraded for indirectness and imprecision) from 1 observational study with 2 healthy volunteers(Gunn 1995, 239-244) demonstrating less overall lateral cervical spine flexion with the HAINES position (MD, −17; 95% CI, −21.39 to −12.62), no difference in lateral flexion of the upper cervical spine with the HAINES position (MD, −4.5; 95% CI, −11.7 to 2.7), and less lateral flexion of the lower cervical spine with the HAINES position (MD, −12.5; 95% CI, −21.52 to −3.47). We have also identified very-low-quality evidence (downgraded for indirectness and imprecision) from 1 observational study with 10 cadavers with surgically created cervical instability(Del Rossi 2014, 539) demonstrating no difference in linear translation between the HAINES recovery position and the 1992 ERC lateral recovery position in terms of medial/lateral movement (MD, −1.1; 95% CI, −5.17 to 2.97), compression/distraction (MD, −1.06; 95% CI, −3.7 to 1.58), or anterior/posterior movement (MD, −0.24; 95% CI, −2.96 to 2.48). Left Lateral Position Compared With Right Lateral Position For the critical outcome of the incidence of aspiration, we identified very-low-quality evidence (downgraded for imprecision) from 1 observational study with a total of 50 patients(Adnet 1999, 745) who were found in the left lateral decubitus or right lateral decubitus position, demonstrating no benefit to the left versus the right lateral position (RR, 0.82; 95% CI, 0.42–1.6). 1992 ERC Recovery Position Compared With Old Left Lateral, Semiprone Resuscitation Council (UK) Recovery Position For the critical outcome of complications, we identified very-low-quality evidence (downgraded for imprecision) from 1 observational study with 6 healthy volunteers(Fulstow 1993, 89) demonstrating no difference in either position in terms of venous occlusion (RR, 5; 95% CI, 0.29–86.44), arterial insufficiency with venous occlusion (RR, 5; 95% CI, 0.29–86.44), or left arm discomfort (RR, 7; 95% CI, 0.44–111.92). 1997 Resuscitation Council (UK) Recovery Position Compared With 1992 ERC Recovery Position For the critical outcome of complications, we identified very-low-quality evidence (downgraded for risk of bias, imprecision, and indirectness) from 1 observational study with 100 healthy volunteers(Doxey 1998, 161) demonstrating less pain/discomfort with the 1992 ERC recovery position (RR, 3.25; 95% CI, 1.81–5.83). AHA Semiprone Recovery Position Compared With 1992 ERC Recovery Position For the critical outcome of complications, we identified very-low-quality evidence (downgraded for risk of bias, imprecision, and indirectness) from 1 observational study with 40 healthy volunteers placed in 1 or both of the positions(Rathgeber 1996, 13) demonstrating less discomfort with the AHA recovery position (RR, 0.36; 95% CI, 0.14–0.95). Morrison, Mirakhur, and Craig Recovery Position Compared With Rautek Recovery Position For the critical outcome of complications, we identified very-low-quality evidence (downgraded for risk of bias, imprecision, and indirectness) from 1 observational study with 20 healthy volunteers placed in 1 or both of the positions(Rathgeber 1996, 13) demonstrating no difference in discomfort between the positions (RR, 1.25; 95% CI, 0.47–3.33). AHA Semiprone Recovery Position Compared With Morrison, Mirakhur, and Craig Recovery Position For the critical outcome of complications, we identified very-low-quality evidence (downgraded for risk of bias, imprecision, and indirectness) from 1 observational study with 30 healthy volunteers placed in 1 or both of the positions(Rathgeber 1996, 13) demonstrating no difference in discomfort between the positions (RR, 0.4; 95% CI, 0.14–1.17). AHA Semiprone Recovery Position Compared With Rautek Recovery Position For the critical outcome of complications, we identified very-low-quality evidence (downgraded for risk of bias, imprecision, and indirectness) from 1 observational study with 30 healthy volunteers placed in 1 or both of the positions(Rathgeber 1996, 13) demonstrating no difference in discomfort between the positions (RR, 0.5; 95% CI, 0.16–1.59). 1992 ERC Recovery Position Compared With Morrison, Mirakhur, and Craig Recovery Position For the critical outcome of complications, we identified very-low-quality evidence (downgraded for risk of bias, imprecision, and indirectness) from 1 observational study with 30 healthy volunteers placed in 1 or both of the positions(Rathgeber 1996, 13) demonstrating no difference in discomfort between the positions (RR, 1.1; 95% CI, 0.53–2.23). 1992 ERC Recovery Position Compared With Rautek Recovery Position For the critical outcome of complications, we identified very-low-quality evidence (downgraded for risk of bias, imprecision, and indirectness) from 1 observational study with 30 healthy volunteers placed in 1 or both of the positions(Rathgeber 1996, 13) demonstrating no difference in discomfort between the positions (RR, 1.38; 95% CI, 0.58–3.24). We did not identify any evidence to address the critical outcome of overall mortality or the important outcome of incidence of cardiac arrest.
Treatment Recommendation:
We suggest that first aid providers position individuals who are unresponsive and breathing normally into a lateral, side-lying recovery (lateral recumbent) position as opposed to leaving them supine (weak recommendation, very-low-quality evidence). There is little evidence to suggest the optimal recovery position. Values, Preferences, and Task Force Insights Due to the low-quality evidence, it was difficult to make a recommendation as to the best recovery position. In terms of the HAINES position versus the standard left lateral position, the task force chose to put more value in the outcomes of a study that included cadavers with surgically created cervical spine instability over a study involving 2 healthy volunteers. We discussed the need for guideline developers to clearly address situations in which a first aid provider should not move a person into a recovery position, such as in the presence of pelvic or spinal injury. Finally, discussions were held about the quality of breathing being used to help determine when it is appropriate to move an individual into the recovery position. The qualifying term “breathing normally” was included in the treatment recommendation so as to avoid the situation where a first aid provider recognizes that an individual is breathing and moves them into a recovery position when in fact chest compressions should be initiated.
CoSTR Attachments:
ILCOR Data Collection Form Template - Recovery Position 2014 12 12.xls    
Recovery position GRADE tables 2015 04 27.pdf    

 Contact Us

 
If you have any comments or questions on this page, please email us at: