Content uploaded by John Haines
Author content
All content in this area was uploaded by John Haines on Dec 08, 2018
Content may be subject to copyright.
ARTICLE IN PRESS
UNCORRECTED PROOF
The position of the spine in the recovery position*/an experimental
comparison between the lateral recovery position and the modified
HAINES position
W.E.D. Blake
a,
*, B.C. Stillman
b
, N. Eizenberg
a
, C. Briggs
a
, J.M. McMeeken
b
a
Department of Anatomy and Cell Biology, University of Melbourne, Parkville, Victoria 3051, Australia
b
School of Physiotherapy, University of Melbourne, Parkville, Victoria 3051, Australia
Received 15 October 2001; received in revised form 16 October 2001; accepted 25 February 2002
/13 /Abstract
/14 /
/
15 /The lateral recovery position is widely used for the positioning of unconscious patients. Ideally, in the setting of trauma it is
/16 /avoided because of concerns about spinal cord injury. However, unconscious individuals with unsuspected trauma or trauma victims
/17 /attended by partially trained first-aiders may be placed in the recovery position, potentially endangering the cord. Excessive
/18 /movement of the spine in the recovery position may increase the risk of spinal cord injury in these situations. A new recovery
/19 /position, termed the modified HAINES position, is described and the position of the spine in this position is compared with the
/20 /lateral recovery position. Hypothesis: That the modified HAINES position results in less distortion of the position of the spine than
/21 /the lateral recovery position. Methods: Thirty-eight healthy volunteers were imaged in the two different positions. Measurements of
/22 /rotation, flexion and lateral flexion of the cervical and thoraco-lumbar spine were made. Two tailed paired t-tests were employed to
/23 /compare measurements of the two positions and a McNemar test was used to compare the subjects’ subjective experiences. Results:
/24 /The modified HAINES position resulted in 13.08(99% CI: 7.5 /18.5) less lateral flexion and 12.68(99% CI: 9.4/15.9) less extension
/25 /of the cervical spine while the position of the thoraco-lumbar spine was similar in both positions. Nineteen of 28 subjects found the
/26 /modified HAINES position more comfortable (not significant). Conclusion: The modified HAINES position results in a more
/27 /neutral position of the spine making it preferable to the lateral recovery position in the management of patients when trauma may
/28 /have occurred. Further research is required to ensure that the recovery positions in use today are the best possible. #2002
/29 /Published by Elsevier Science Ireland Ltd.
/30 /Keywords: Adult; Basic life support; Spinal cord; Resuscitation; Safety
/31 /1. Introduction
/32 /Resuscitation guidelines advocate placing an uncon-
/33 /scious patient who has a pulse and is breathing into a
/34 /recovery position [1]. The main benefit of this is to
/35 /spontaneously maintain a clear airway. In the setting of
/36 /trauma any movement of the victim is avoided until
/37 /rigid in line immobilisation is secured. However, in three
/38 /settings trauma patients may be placed into a recovery
/39 /position: (1) the trauma is unrecognised [2]; (2) the
/40 /rescuer is not well trained and makes a decision to use
/41/the recovery position; and (3) an adequate airway
/42/cannot be maintained with the patient supine. In these
/43/situations the position of the spine is vital. Recovery
/44/positions also have wide application within hospitals
/45/especially in the recovery room and endoscopy units.
/46/Ideally a single recovery position should be recom-
/47/mended and taught for use in, and out, of hospital.
/48/Positioning of the unconscious patient at the scene of
/49/an incident and during transportation to hospital is the
/50/subject of many recommendations but very little science.
/51/A Medline literature search for data on recovery
/52/positions using the search terms ‘recovery’, ‘position’,
/53/‘transport’, ‘transfer’, ‘trauma’, and ‘unconscious’ re-
/54/vealed only three experimental papers [3 /5]. These and a
/55/small number of case reports and interview-based
/56/studies [6,7] are discussed in detail below. Given the
G:/Elsevier Science/Shannon/RESUS/articles/resus1974/RESUS1974.3d[x] Tuesday, 12th March 2002 11:2:8
3* Corresponding author. Present address: 28a Lisson Grove,
4Hawthorn, Victoria 3122, Australia. Tel.: 613-9819-3119.
5E-mail address: wblake@bigpond.net.au (W.E.D. Blake).
6
7
8
9
10
11
12
Resuscitation 00 (2002) 1 /8
www.elsevier.com/locate/resuscitation
10300-9572/02/$ - see front matter #2002 Published by Elsevier Science Ireland Ltd.
2PII: S 0 3 0 0 - 9 5 7 2 ( 0 2 ) 0 0 0 3 7 - 0
ARTICLE IN PRESS
UNCORRECTED PROOF
/57 /lack of evidence supporting any one position, many
/58 /organisations haveavoided endorsing a single recovery
/59 /position. Instead, priorities for recovery positions have
/60 /been published by ILCOR (the International Liaison
/61 /Committee on Resuscitation) and are summarised in six
/62 /points. These are: ‘‘(1) the victim should be in as near to
/63 /a true lateral position as possible, with the head
/64 /dependant to allow free drainage of fluid; (2) the
/65 /position should be stable; (3) any pressure on the chest
/66 /that impairs breathing should be avoided; (4) it should
/67 /be possible to turn the victim onto his or her side and
/68 /return to the back easily and safely, having particular
/69 /regard for the possibility of cervical spine injury; (5)
/70 /good observation of and access to the airway should be
/71 /possible; and (6) the position itself should not give rise
/72 /to any injury to the victim’’ [8]. Other expert bodies,
/73 /including the American Heart Association and the
/74 /European and Australian Resuscitation Councils have
/75 /endorsed nearly identical guidelines [9].
/76 /Several recovery positions havebeenadvocated by
/77 /various organisations at various times and some discus-
/78 /sion of their pros and cons appears below. The lateral
/79 /recovery position (Fig. 1) and its close variant the ‘How’
/80 /or 1992 European Resuscitation Council (ERC) posi-
/81 /tion, are the most widely illustrated, taught and used
/82 /positions at this time. In the lateral recovery position the
/83 /cervical spine is observed to be laterally flexed and
/84 /possibly rotated while the thoraco-lumbar spine looks
/85 /rotated also. We propose that a refined position, the
/86 /modified HAINES position (Fig. 2) may achieve a more
/87 /neutral position of the spine. In the modified HAINES
/88 /position the head rests on the fully abducted dependant
/89 /arm and both legs are drawn up, flexed at the hip and
/90 /the knee. These changes aim to reduce rotation and
/91 /flexion of the spine.
/92 /2. Aim
/93 /We aimed to test the hypothesis that the modified
/94 /HAINES position results in less distortion of the
/95 /position of the spine than the lateral recovery position
/96 /(Figs. 1 and 2). Measurements were made of flexion,
/97 /rotation and lateral flexion of the cervical and thoraco-
/98/lumbar regions of the spine to allow comparison of the
/99/two positions.
/100/3. Materials and methods
/101/Thirty-eight healthy volunteer subjects were recruited
/102/to participate in the study. Key bony landmarks were
/103/identified by palpation and marked using a surgical
/104/skin-marking pen. These were: the spinous processes of
/105/C7, T2, L2 and L4, the suprasternal notch and the
/106/xiphoid process. Each subject lay on a firm examination
/107/couch and was positioned in the lateral recovery
/108/position and later in the modified HAINES position.
/109/A qualified paramedic or his assistant undertook the
/110/positioning. In each position four digital photographs
/111/were taken: one from in front, one from behind, one
/112/from above and one along the longitudinal axis of the
/113/body from the head end. Care was taken to standardise
/114/the four positions of the camera.
/115/A total of 304 digital images were analysed using
/116/ImageJ software published by the National Institutes of
/117/Health, USA [10]. Assessments were made of six angles
/118/allowing measurement of thoraco-lumbar flexion, lateral
/119/flexion and rotation and cervical flexion, lateral flexion
/120/and rotation.
/121/Lateral flexion was assessed using images taken from
/122/in front and behind the subject. The midline of the
/123/sternum, marked at the xiphoid and suprasternal notch,
/124/was compared to the midline of the face, bridge of nose
/125/and midpoint of chin, to measure cervical side flexion
/126/(Fig. 3). Marks at C7 and T2 were joined to make a line
/127/and similar marks on L2 and L4 were joined to form a
/128/line. The angle between these lines was thoraco-lumbar
/129/lateral flexion (Fig. 4). An arbitrary sign convention was
/130/adopted so that flexion to the right was positive and to
/131/the left negative.
/132/Flexion was assessed using an image taken from
/133/overhead using a newly developed but simple technique.
/134/To allow measurement of the position of the bony
/135/vertebral column from the photograph, a 15 cm ply-
/136/wood square was applied to the lumbar spine with its
/137/lower border at the L4 mark and a 10 cm square was
/138/applied to the cervical area with its lower border at the
/139/T2 mark. Without such equipment the required bony
/140/landmarks are obscured from the camera by the
/141/subject’s body. The protruding edges of these squares
G:/Elsevier Science/Shannon/RESUS/articles/resus1974/RESUS1974.3d[x] Tuesday, 12th March 2002 11:2:10
Fig. 1. The lateral recovery position.
Fig. 2. The modified HAINES position.
W.E.D. Blake et al. / Resuscitation 00 (2002) 1/82
ARTICLE IN PRESS
UNCORRECTED PROOF
/142 /were parallel to the section of the bony spine over which
/143 /the square rested. It is important to note that this
/144 /methodology was not designed to measure the absolute
/145 /position of the vertebral column but instead, to allow
/146 /comparison between the two positions under investiga-
/147 /tion. Thoraco-lumbar flexion was determined by mea-
/148/suring the angle between the exposed edges of the
/149/wooden squares and cervical flexion was determined
/150/by measuring the angle between the upper square and
/151/the plane of the face (Fig. 5). The plane of the face was
/152/arbitrarily defined as a line joining the tip of the
/153/supraorbital ridge to the tip of the chin. Again, it is
G:/Elsevier Science/Shannon/RESUS/articles/resus1974/RESUS1974.3d[x] Tuesday, 12th March 2002 11:2:11
Fig. 3. Cervical lateral flexion was measured by comparing the midline of the face to the midline of the sternum. Here it is 68.
Fig. 4. Thoraco-lumbar lateral flexion was measured by marking the spinous processes of C7, T1, L2 and L4. Here it is 68.
W.E.D. Blake et al. / Resuscitation 00 (2002) 1/83
ARTICLE IN PRESS
UNCORRECTED PROOF
/154 /important to note that the absolute value of cervical
/155 /flexion is not assessed; this methodology was designed to
/156 /allow a comparison to be made between the two
/157 /positions.
/158 /Rotation was measured using an axial photograph
/159 /taken from the head end of the couch. To enable the
/160 /planes of the skeleton to be measured wooden squares
/161 /were employed again. A 15 cm square was placed
/162 /against the bony pelvis resting on the posterior superior
/163 /iliac spines and a 10 cm square was placed transversely
/164 /at the level of T2. A further marker was then applied to
/165 /the external occipital protuberance. Thus thoraco-lum-
/166 /bar rotation could be measured from the two exposed
/167 /sides of the wooden squares and cervical rotation was
/168 /measured by comparing a line drawn from nose to
/169 /external occipital protuberance with the free edge of the
/170 /upper square (Fig. 6).
/171 /The weight and height of each subject were recorded
/172 /and each was asked for any subjective impressions of the
/173 /two positions. These were recorded on a standardised
/174 /recording sheet.
/175 /Two investigators undertook image analysis sepa-
/176 /rately. The two sets of data were compared and
/177 /whenever discrepancies of 48or more existed between
/178 /the results for any given image the analysis was
/179 /repeated. The two investigators produced highly corre-
/180 /lated data sets (correlation coefficients varied from 0.94
/181 /to 0.99). The final data set was derived by taking the
/182 /mean of the two sets of data produced by the two
/183 /analysts.
/184/A paired two-tailed Student’s t-test has been em-
/185/ployed to compare the mean difference between the two
/186/positions for each of the six measured angles. Data
/187/showed little skew and few outliers making the Student’s
/188/test appropriate. Pvalues less than 0.01 were considered
/189/to indicate statistical significance.
/190/Ethics approval for this study was obtained from the
/191/University of Melbourne and written informed consent,
/192/including consent for the publication of images, was
/193/obtained from all subjects.
/194/4. Results
/195/Characteristics of the 38 volunteer subjects are
/196/detailed in Table 1. The subjects were young adults
/197/with a preponderance of women over men.
/198/Data from the analysis of the 304 digital images is
/199/summarised in Table 2. Statistically significant results,
/200/indicated by a 99% confidence interval which excludes
/201/zero, are recorded for four of the six angles measured.
/202/There is more cervical lateral flexion in the lateral
/203/recovery position. The neck is also more extended (the
/204/value for flexion is more negative) in this position. In the
/205/thoraco-lumbar region there is more lateral flexion in
/206/the modified HAINES position but the difference is
/207/small. Values for thoraco-lumbar rotation are interest-
/208/ing in that both positions have a similar magnitude to
/209/the rotation measured but the directions are opposite.
/210/The subjective impressions of the 38 volunteers were
/211/recorded on a standardised recording sheet at the end of
G:/Elsevier Science/Shannon/RESUS/articles/resus1974/RESUS1974.3d[x] Tuesday, 12th March 2002 11:2:12
Fig. 5. Flexion of the cervical and thoraco-lumbar spine was assessed from above. Here the angles are 25 and 228respectively. See text for details.
W.E.D. Blake et al. / Resuscitation 00 (2002) 1/84
ARTICLE IN PRESS
UNCORRECTED PROOF
/212 /each imaging episode. Thirty-four of the 38 subjects
/213 /offered a comment with six subjects saying both
/214 /positions were equally comfortable. Of the remaining
/215 /28, 19 preferred the modified HAINES position and
/216 /nine the lateral recovery position. This difference is not
/217 /significant (McNemar test). Several subjects commented
/218 /that the neck was more comfortable when the head
/219 /rested on an abducted arm. The other frequent comment
/220/was that the modified HAINES position felt less stable
/221/than the lateral recovery position. Two subjects reported
/222/a ‘pins and needles’ feeling in the abducted arm in the
/223/modified HAINES position.
/224/5. Discussion
/225/Injuries to the spinal cord havedevastating conse-
/226/quences for the individuals afflicted and every effort
/227/must be taken to reduce the number of such injuries. It is
/228/possible that first aid and paramedic interventions might
/229/cause or exacerbate spinal cord injuries. The use of in
/230/line immobilisation, spine boards and hard cervical
/231/collars is routine practise when there is any suspicion
/232/of a vertebral fracture. However, in some situations
/233/patients will be moved without the benefit of rigid in line
G:/Elsevier Science/Shannon/RESUS/articles/resus1974/RESUS1974.3d[x] Tuesday, 12th March 2002 11:2:12
Fig. 6. Rotation was measured using an axial image. Here there is 198of cervical rotation and 108of thoraco-lumbar rotation.
Table 1
Demographic data
Number 38
Age
a
19 (17 /33)
Sex 27 female, 11 male
Height
a
171 cm (160 /193)
Weight
a
62 kg (50 /93)
a
Expressed as mean and range.
Table 2
Summary of data
Lateral recovery position Modified HAINES position Difference 99% CI of difference Pvalue
Cervical region
Flexion 36.89 23.91 12.98 7.47 /18.49 0.001
Rotation 1.85 2.05 0.20 4.28 /4.68 0.91 NS
Lateral flexion 22.35 9.74 12.61 9.37 /15.85 0.001
Thoraco-lumbar region
Flexion 23.90 23.44 0.46 2.17 /3.09 0.64 NS
Rotation 4.64 5.22 9.86 6.41 /13.31 0.001
Lateral flexion 7.01 10.72 3.71 0.28 /7.14 B0.01
Expressed as mean of 38 observations, all angles in degrees, see text for sign convention.
W.E.D. Blake et al. / Resuscitation 00 (2002) 1/85
ARTICLE IN PRESS
UNCORRECTED PROOF
/234 /immobilisation. Rarely, this may include patients with
/235 /vertebral injuries. In this situation it is imperative that a
/236 /recovery position is employed which provides a reliable
/237 /airway while minimising the probability of injury to the
/238 /spinal cord or, indeed, any other structure in the body.
/239 /To reduce the chance of converting a patient with an
/240 /unstable vertebral fracture but intact cord into a patient
/241 /with a cord injury, the degree of distortion of the
/242 /vertebral column must be minimised. Extension and
/243 /lateral flexion have been associated with reduction in the
/244 /cross-sectional area of the injured cervical spinal canal
/245 /[11]. In this study extension was the most detrimental
/246 /movement with 13.79/6.18of extension reducing the
/247 /area of the canal by 429/20% (P/0.02). Therefore, a
/248 /recovery position which produces less distortion of the
/249 /position of the vertebral column is preferred over one
/250 /that produces more distortion provided both adequately
/251 /address airway issues.
/252 /Attempts to find an ideal recovery position have been
/253 /ongoing for many years. The early ‘coma’ position,
/254 /advocated extensively in the 1980s (Fig. 7) and illu-
/255 /strated in the American Heart Association guidelines of
/256 /1992 [12], was an attempt to find a safe and stable
/257 /position for prehospital care of the unconscious patient.
/258 /In the prehospital and transportation phases of patient
/259 /management it has some strengths. The airway is
/260 /maintained by allowing the tongue and jaw to hang
/261 /forward and secretions or vomitus to drain from the
/262 /mouth. This occurs passively as the head rotates and
/263 /flexes under the effect of gravity. Amongst its disadvan-
/264 /tages is poor protection of the cervical spine during the
/265 /transition from supine to the ‘coma’ position. The final
/266 /position of the head is difficult to control and some
/267 /individuals tend to become almost prone with conse-
/268 /quent loss of access to the airway. The thoraco-lumbar
/269 /spine is variably rotated in the ‘coma’ position, again
/270 /potentially threatening the spinal cord. There have been
/271 /some reports that the position is excessively uncomfor-
/272 /table [13,14]. Lastly, the patient in the ‘coma’ position
/273 /cannot be returned easily to the supine position for
/274 /cardiopulmonary resuscitation (CPR) or other proce-
/275 /dures because the posteriorly positioned dependant
/276 /upper limb will tend to interfere with rolling.
/277/Addressing some of the concerns about the ‘coma’
/278/position an improved, and now nearly standard, lateral
/279/recovery position was developed (Fig. 1). There are two
/280/important variants of this position, namely the lateral
/281/recovery position as illustrated and examined in this
/282/study and the 1992 ERC or ‘How’ position. One or the
/283/other is the standard position for care of the uncon-
/284/scious patient adopted by many organisations managing
/285/trauma and first aid [1,15 /17], and it is illustrated in the
/286/International Resuscitation Guidelines [1]. In these
/287/positions the dependant upper limb is flexed to 908at
/288/the shoulder rather than being behind the trunk as in the
/289/coma position. The physical process of rolling a patient
/290/into the lateral recovery position is more controlled and
/291/precise than for the ‘coma’ position [6]. The position of
/292/the head is predictable and allows access to the airway.
/293/The patient is stable but can be easily returned to the
/294/supine position for CPR if required. Direct, but
/295/subjective, comparisons of these positions and the
/296/‘coma’ position have indicated that the lateral recovery
/297/position is easier to use and provides better stability and
/298/airway access than the ‘coma’ position [7]. Although
/299/widely used, the lateral recovery position has been the
/300/subject of very little research.
/301/The 1992 ERC position differs from the lateral
/302/recovery position as illustrated here in that the hand
/303/of the victim’s upper arm is tucked under the head to
/304/reduce lateral flexion and rotation of the cervical spine.
/305/This has been demonstrated to reduce perfusion and
/306/cause pain in the dependant arm [3]. The position is also
/307/associated with neuropraxias [18] and venous congestion
/308/of the dependant arm [5]. To avoid this the lateral
/309/recovery position tested in this study rests the upper arm
/310/on the ground where it is not loaded with the weight of
/311/the head and is unlikely to cause compression. The
/312/lateral recovery position has been criticised because of
/313/its tendency to produce significant distortion of the
/314/vertebral column with the fear that this could be
/315/associated with damage to the spinal cord in the neck
/316/or thorax. When vertebral fractures have been unsus-
/317/pected, basic first aid procedures have been associated
/318/with spinal cord damage [2].
/319/To minimise the risk to the cervical spine Gunn et al.
/320/[4] proposed a modification of the lateral recovery
/321/position in 1995. Termed the HAINES position (an
/322/acronym for High Arm IN Endangered Spine and
/323/recognising the contribution of its inventor, John
/324/Haines), it has two significant differences when com-
/325/pared with the lateral recovery position. The dependant
/326/upper limb is fully abducted and lies under the head
/327/where it reduces lateral flexion and, secondly, both
/328/lower limbs are flexed at the hip and the knee resulting
/329/in one lying on top of the other possibly reducing torque
/330/on the thoraco-lumbar spine. The position is easily
/331/achieved by a single rescuer. In his paper Gunn uses a
/332/combination of digital image analysis and fluoroscopy
G:/Elsevier Science/Shannon/RESUS/articles/resus1974/RESUS1974.3d[x] Tuesday, 12th March 2002 11:2:13
Fig. 7. The coma position.
W.E.D. Blake et al. / Resuscitation 00 (2002) 1/86
ARTICLE IN PRESS
UNCORRECTED PROOF
/333 /to demonstrate that the supporting dependant upper
/334 /limb reduces lateral flexion of the cervical spine (from 28
/335 /to 118) whilst not compromising the airway. Unfortu-
/336 /nately, Gunn did not go on to examine other movements
/337 /of the cervical spine or the thoraco-lumbar spine and his
/338 /study was restricted to two subjects. The two subjects
/339 /who were studied displayed quite different cervical
/340 /motion suggesting quite marked variation within the
/341 /population. A further modification has been proposed
/342 /since Gunn’s original article [19]. In the modified
/343 /HAINES position, which we investigated, the upper-
/344 /most upper limb is placed longitudinally down the trunk
/345 /rather than in front of the patient in an effort to further
/346 /reduce the rotation of the thoraco-lumbar spine (Fig. 2).
/347 /Gunn imaged the airway using fluoroscopy in the
/348 /lateral recovery position and in the HAINES position
/349 /and showed that the dimensions of the airway are
/350 /equivalent. The modified HAINES position offers a
/351 /large improvement in the position of the cervical spine.
/352 /It is 138less extended and 138less laterally flexed in the
/353 /modified HAINES position. In the injured spine,
/354 /differences of this magnitude make a huge difference
/355 /to the dimensions of the vertebral canal [11]. Changes in
/356 /the thoraco-lumbar spine are less impressivewitha
/357 /statistically significant but probably unimportant 38
/358 /increase in lateral flexion and an equivalent amount of
/359 /rotation when compared with the lateral recovery
/360 /position.
/361 /A separate issue in choosing a recovery position is the
/362 /stability of the unconscious patient in that position. The
/363 /‘coma’ position was rejected partly because patients
/364 /tended to roll forward into a prone position. Several of
/365 /the subjects in the current study commented on the fact
/366 /that the modified HAINES position felt less stable than
/367 /the standard lateral recovery position. It should be
/368 /recognised that the modified HAINES position is most
/369 /suitable for an attended patient. Because the upper-
/370 /most arm is not acting as a stabilising prop to
/371 /prevent rotation there is a tendency for the patient
/372 /to roll forwards. The attendant is also required to
/373 /monitor the position of the head and ensure that it
/374 /does not slip off the abducted arm posteriorly. Having
/375 /demonstrated that the degree of rotation of the thoraco-
/376 /lumbar spine in not an important consideration in
/377 /choosing between the two positions we have compared,
/378 /the stability of the HAINES position could be greatly
/379 /improved by bringing the upper arm forward into 908of
/380 /flexion.
/381 /Only two subjects described any discomfort during
/382 /the study. Both described a mild and transient ‘pins and
/383 /needles’ feeling in the dependant upper limb in the
/384 /modified HAINES position. Attention must be focused
/385 /on the brachial plexus and ulnar nerve in this position,
/386 /as in all recovery positions. Significant secondary
/387/injuries would greatly detract from the usefulness of the
/388/position.
/389/The methodology employed in this study is deliber-
/390/ately simple and non-invasive. While the angles mea-
/391/sured are based on easily identifiable bony surface
/392/landmarks, no imaging of the skeleton itself has been
/393/undertaken. This has allowed us to recruit a significant
/394/number of volunteers none of whom has been exposed
/395/to radiation particularly X-rays. The authors recognise
/396/that it has not been possible to exclude all possible
/397/sources of subjective bias in positioning subjects and
/398/analysing images.
/399/Several directions for further investigation immedi-
/400/ately present themselves. The ideal study design, a
/401/randomised double blinded controlled trial, does not
/402/lend itself to investigation of the relationship between
/403/recovery positions and spinal cord injuries. The end
/404/point of cord injury is too rare, there is no opportunity
/405/for randomisation and blinding participants is impos-
/406/sible. However, this should not prevent further attempts
/407/to investigate the under-researched area of recovery
/408/positions. There are no published studies using uncon-
/409/scious or anaesthetised individuals where the elimination
/410/of protective reflexes and resting muscle tone would be
/411/expected to make results more reliable. There is a need
/412/for research in populations of older and less healthy
/413/people than our study cohort. Radiological studies,
/414/while necessitating the exposure of subjects to ionising
/415/radiation, may enable more precise delineation of
/416/exactly where in the spine flexion, rotation and lateral
/417/flexion are occurring.
/418/6. Conclusion
/419/The modified HAINES recovery position results in a
/420/better position of the head than the standard lateral
/421/recovery position. By being less extended and less
/422/laterally flexed the dimensions of the spinal canal are
/423/better preserved. For an unconscious person with an
/424/unsuspected, unstable vertebral fracture its use may
/425/reduce the chance of causing or exacerbating a spinal
/426/cord injury. In the case of an unconscious patient with a
/427/suspected vertebral fracture where in-line immobilisa-
/428/tion is not available, the position of the head is crucial
/429/and the modified HAINES position is recommended in
/430/preference to the lateral recovery position. Standard
/431/management still requires supine in line immobilisation
/432/but where this is impossible the modified HAINES
/433/position is the better alternative. The modified HAINES
/434/position, like any recovery position, requires rescuers to
/435/be adequately trained and also requires that the
/436/attendant stay to observe the patient. Further empirical
/437/research, that builds on the findings of this study, is
G:/Elsevier Science/Shannon/RESUS/articles/resus1974/RESUS1974.3d[x] Tuesday, 12th March 2002 11:2:15
W.E.D. Blake et al. / Resuscitation 00 (2002) 1/87
ARTICLE IN PRESS
UNCORRECTED PROOF
/438 /desperately required to ensure that the recovery posi-
/439 /tions in use worldwide are the best possible.
/440 /Acknowledgements
/441 /This work has been funded by the Department of
/442 /Anatomy and Cell Biology, University of Melbourne,
/443 /Parkville, Victoria, Australia. The authors are not aware
/444 /of any conflict of interest.
/445 /References
/446 /[1] The American Heart Association in collaboration with the
/447 /International Committee on Resuscitation (ILCOR). Interna-
/448 /tional Resuscitation Guidelines*/a consensus on science Part 3:
/449 /adult basic life support. Resuscitation 2000;46:29/71.
/450 /[2] Powell RM, Heath KJ. Quadriplegia in a patient with an
/451 /undiagnosed odontoid peg fracture. The importance of cervical
/452 /spine immobilisation in patients with head injuries. J Army
/453 /Medical Corps 1996;142:79/81.
/454 /[3] Rathgeber J, Panzer W, Gu
¨nther U, Scholz M, Hoeft A, Bahr J,
/455 /Kettler D. Influence of different types of recovery positions on
/456 /perfusion indices of the forearm. Resuscitation 1996;32:13/7.
/457 /[4] Gunn BD, Eizenberg N, Silberstein M, et al. How should an
/458 /unconscious person with a suspected neck injury be positioned?
/459 /Prehospital Disaster Med 1995;10:239/44.
/460 /[5] Fulstow R, Smith GB. The new recovery position, a cautionary
/461 /tale. Resuscitation 1993;26:89/91.
/462 /[6] Doxey J. Comparing 1997 resuscitation council (UK) recovery
/463 /position with recovery position of 1992 European Resuscitation
/464 /Council guidelines: a user’s perspective. Resuscitation
/465 /1998;39:161 /9.
/466/[7] Turner SA, Turner I, Chapman D, et al. A comparative study of
/467/the 1992 and 1997 recovery positions for use in the UK.
/468/Resuscitation 39;1998:153/60.
/469/[8] Handley AJ, Becker LB, Allen M, et al. Single-rescuer adult basic
/470/life support: an advisory statement from the Basic Life Support
/471/Working Group of the International Liaison Committee on
/472/Resuscitation. Circulation 1997;95:2174/9.
/473/[9] Pepe P, Gay M, Cobb L, et al. Action sequence for lay-
/474/person cardiopulmonary resuscitation. Ann Emerg Med
/475/2001;37:S17/25.
/476/[10] Available free of charge from http://rsb.info.nih.gov/ij.
/477/[11] Ching RP, Watson NA, Carter JW, Tencer AF. The effect of
/478/post-injury spinal position on canal occlusion in a cervical spine
/479/burst fracture model. Spine 1997;22:1710 /5.
/480/[12] Emergency cardiac care committee and subcommittees, American
/481/Heart Association. Guidelines for cardiopulmonary resuscitation
/482/and emergency cardiac care, Part II adult basic life support. J Am
/483/Med Assoc 1992;268:2184/98.
/484/[13] Handley A. Resucitation Council (UK) wants everyone who uses
/485/new recovery position to report experiences. Br Med J
/486/1997;315:1308.
/487/[14] Leaves S, Donnelley P, Lester C, Assar D. Trainees’ adverse
/488/experiences of the new recovery position. Br Med J
/489/1998;316:1748 /9.
/490/[15] Resuscitation Council (UK), Basic Life Support /Resuscitation
/491/Guidelines 2000, at http://www.resus.org.uk/.
/492/[16] St. John Ambulance, Australian First Aid: an authorised manual
/493/of St. John Ambulance Australia, 4 ed., Canberra, St. John
/494/Ambulance Australia, 1998, p. 30.
/495/[17] Australian Red Cross, Australian Red Cross first aid, Artarmon,
/496/NSW, Mosby Lifeline, 1995, p. 70.
/497/[18] Kumar P, Touquet R. Perils of the recovery position: neuropraxia
/498/of the radial and common peroneal nerve. Emerg Med J
/499/1996;13:69/70.
/500/[19] Personal correspondence between the authors, Mr J Haines and
/501/Dr Bror Gardelof.
/502 /
G:/Elsevier Science/Shannon/RESUS/articles/resus1974/RESUS1974.3d[x] Tuesday, 12th March 2002 11:2:28
W.E.D. Blake et al. / Resuscitation 00 (2002) 1/88
