SPINE Volume 29, Number 7, pp E134–E138©2004, Lippincott Williams & Wilkins, Inc.

Spine-Board Transfer Techniques and the UnstableCervical Spine

Gianluca Del Rossi, PhD,* MaryBeth Horodyski, EdD,† Timothy P. Heffernan, BS,�Michael E. Powers, PhD,‡ Ronald Siders, PhD,‡ Denis Brunt, EdD,§ andGlenn R. Rechtine, MD†

Study Design. A repeated-measures design using acadaveric model was used in this preliminary investiga-tion on the effectiveness of spine-board transfertechniques.

Objectives. To compare the amount of angulation(flexion–extension) motion that results at the cervicalspine during the execution of the log-roll maneuver andthe lift-and-slide technique; and to examine how changesto the integrity of the cervical spine impacts the amountof motion generated during the transfer process.

Summary of Background Data. Very little research hasbeen performed to establish the efficacy of spine-boardtransfer techniques. Early studies have indicated that thelog-roll maneuver may not be appropriate for transferringvictims with thoracolumbar injuries. Also, there has notbeen a single study that has reported the impact of trans-fer techniques on the unstable cervical spine. This lack ofdata necessitated the present study.

Methods. Four groups (with six participants each)were asked to execute the log-roll maneuver and thelift-and-slide technique on five cadavers. An electromag-netic motion analysis device was used to assess theamount of angulation motion generated at the C5–C6segment during the execution of these transfer techniques.To examine how changes to the integrity of the cervicalspine impacts the amount of motion that is produced duringthe transfer process, flexion–extension motion was as-sessed under various conditions: across a stable C5–C6 seg-ment, after the creation of a posterior ligamentous injury,and after a complete segmental injury.

Results. No significant differences in angulation mo-tion were noted between transfer techniques. However,significant differences were noted between all three injuryconditions. That is, as the severity of the injury increased,the average amount of angulation motion produced at thesite of the lesion also increased, regardless of technique.

Conclusion. The participants of this study were able torestrict flexion–extension motion equally well with the-log-roll maneuver as with the lift-and-slide technique.However, more research is needed to fully ascertain theeffectiveness of spine-board transfer techniques. [Keywords: cervical spine, spinal column injury, spine board,prehospital care, log-roll, lift-and-slide] Spine 2004;29:

E134–E138

Before transporting a victim of cervical spine trauma to amedical facility, full spinal immobilization must beachieved by transferring and securing the injured victim toa full-length spine board. Naturally, there is considerablerisk involved with transferring a victim onto a spine board.To facilitate this process, medical personnel use spine-board transfer techniques, such as the log-roll (LR) maneu-ver or the lift-and-slide (LS) technique. As with other med-ical procedures used in the prehospital care of the spine-injured victim, spine-board transfer techniques have beendesigned to minimize the chances of producing secondaryneurologic injuries. Namely, these techniques enable rescu-ers to maintain in-line stabilization of the spine as theysimultaneously complete the transfer process.

Sadly, it has been estimated that between 3% and25% of spinal cord injuries actually occur during theinitial stages of management.1 As a result, legitimate con-cerns have emerged regarding the effectiveness of prehos-pital medical procedures. It is, perhaps, for this reasonthat researchers have undertaken studies to verify thesafety or efficacy of airway management and intubationtechniques,2–5 stabilization maneuvers,4–6 and helmetand shoulder pad removal techniques.7–9 To contributeto this content of information, we set out to evaluate theamount of angulation motion (flexion–extension) nor-mally produced during the execution of the LR and theLS. In addition, because a cadaveric model was used inthis experiment, it was possible to create experimentallesions within the cervical spine and observe how theseverity of an injury impacts the magnitude of the move-ment generated at the injured spinal segment during thetransfer process.

Materials and Methods

After obtaining approval from the university Institutional Re-view Board, 24 qualified individuals (certified athletic trainers,athletic training students, and emergency medical technicians)were recruited for this investigation. Each participant was ran-domly assigned to one of four transfer groups such that eachgroup consisted of six individuals. Five participants from each

From the *Department of Exercise and Sport Sciences, University ofMiami, Coral Gables, FL; †Department of Orthopaedics, University ofFlorida, Gainesville, FL; ‡Department of Exercise and Sport Sciences,University of Florida, Gainesville, FL; §Department of Physical Ther-apy, East Carolina University, Greenville, NC and the �Computer andInformation Science and Engineering Department, University of Flor-ida, Gainesville, FL.Acknowledgment date: February 26, 2003. First revision date: July 22,2003. Acceptance date: September 8, 2003.Supported by the Laerdal Foundation for Acute Medicine (Stavanger,Norway).The manuscript submitted does not contain information about medicaldevice(s)/drug(s).Foundation funds were received in support of this work. No benefits inany form have been or will be received from a commercial party relateddirectly or indirectly to the subject of this manuscript.Address correspondence to Gianluca Del Rossi, PhD, University ofMiami, Department of Exercise and Sport Sciences, School of Educa-tion, 312E Merrick Building, P.O. Box 248065, Coral Gables, FL33124-2040; E-mail: delrossi@miami.edu

E134

group were needed to execute the LR maneuver and LS tech-nique, and a sixth individual was required to position the spineboard during the end stages of the procedure.10

The LR maneuver required one individual to provide man-ual in-line stabilization during the transfer process, two indi-viduals to assist in rolling the torso and upper extremities, andtwo others to assist in rolling the lower extremities (Figure 1).10

All five team members were required to roll the cadaver 90° tothe side-lying position. Once in the lateral recumbent position,the sixth and final member of the group was required to wedgethe spine board beneath the cadaver (at a 45° angle to theground). To complete this procedure, the cadaver was rolledback to the supine position onto the spine board. In most cases,after an individual has been positioned on a spine board usingthis technique, it is necessary to adjust the individual so that heor she is centered on the spine board. This component of the LRmaneuver was not performed in this investigation.

With the LS, one individual was required to maintain man-ual in-line stabilization of the head, two were needed to lift theupper torso, and two others to assist in lifting the lower ex-tremities (Figure 2).10 A sixth individual was again responsible

for placement of the spine board. Those participants responsiblefor lifting the upper torso were required to kneel by the cadaver’sshoulder, placing one hand beneath the lateral aspect of the shoul-der and the other hand beneath the torso just below the level of theaxilla. The individuals responsible for lifting the lower extremitieswere required to straddle the cadaver at the level of the thighs andlegs. With all the participants ready, the individual providing in-line stabilization of the head directed all others to raise the cadaveroff the ground. This allowed the sixth member of the team to slidethe spine board beneath the cadaver. When the board was appro-priately positioned, the cadaver was gently placed onto the boardand the procedure was completed.

In this investigation, each rescue team was required to at-tend a total of four sessions. During the first session, teamswere shown by way of a video presentation how to execute thespecific versions of the LR and the LS that were chosen for thisstudy. All groups were also required to complete a familiariza-tion session at this time. As part of the familiarization process,transfer groups completed five practice trials of each technique.For each of the three test sessions that followed this instruc-tional and familiarization session, groups were required tocomplete two trials of the LR and two of the LS on five cadav-ers. Note that for all groups the order of testing for techniqueand for cadavers was randomized.

A Fastrak motion analysis device (Polhemus Inc., Colches-ter, VT) was used to quantify the amount of motion generatedbetween the C5 and C6 spinal segments during the execution oftransfer techniques. The Fastrak device is a tracking instrumentthat uses electromagnetic fields to determine the position andorientation of its sensors in three-dimensional space. Whenused within its optimal operating range of 10–70 cm, trackingdevices such as the Fastrak have been found to be sensitiveenough to read rotational changes of 0.3° to 1.0°.

Before data collection began, we assessed the validity of theFastrak instrument by comparing angular displacement valuesobtained from this unit with those captured by an analog go-niometer. To complete this validation study, sensors were af-fixed to the arms of a goniometer as it was moved to randomlychosen angles so that angular displacements could be recordedat each angle using the Fastrak. Pearson product moment cor-relation coefficients were then calculated to assess the accuracyof Fastrak angular measurements. Tests revealed the Fastrakdevice to be very accurate with strong correlation coefficientsthat ranged between 0.991 and 0.999.

Additionally, the angles measured above were repeated in

Figure 3. A posterior ligamentous injury.

Figure 1. The log-roll maneuver.

Figure 2. The lift-and-slide technique. (From Del Rossi, et al. Man-agement of cervical-spine injuries. Athletic Therapy Today,7:46 –51).

E135Spine Board Transfer Techniques • Del Rossi et al

random order so that the reliability of angular displacementscould be determined. This was evaluated by calculating intra-class correlation coefficients using repeated measurements ob-tained by the Fastrak device. Tests of reliability confirmed thatthe Fastrak device was able to consistently measure angulardisplacement values (intraclass correlation coefficient valuesranged from 0.99 to 1.0).

Along with a control condition (stable spine), two injuryconditions were tested in this study: a posterior ligamentousinjury and a complete segmental injury. The effect of transfer-ring patients with these injury conditions was investigated bycreating the aforementioned injuries at the C5–C6 level of allcadaver spines. To standardize the injury conditions producedwith each of the cadaver spines, a spine surgeon was responsi-ble for creating all experimental lesions. On creating the exper-imental lesion, the spine surgeon was also responsible for test-ing the mobility of the spinal column (particularly at the site ofthe lesion) and for confirming that the extent of the instabilityresulted in at least an angular displacement greater than 11°.11

In order to be able to access the lower cervical vertebrae of the

cadavers, and to assess the magnitude of the movements cre-ated at the level of interest, it was necessary to displace and/orremove the structures overlying the cervical spine (both anteri-orly and posteriorly). In particular, to access the anterior aspectof the vertebral bodies of interest, it was necessary to removesections of the larynx, the esophagus, and the trachea.

Consistent with what has been reported in the literature, aposterior ligamentous injury was created by excising the supraspi-nous and interspinous ligaments, the ligamentum flavum, the facetcapsules, the posterior longitudinal ligament, and finally, the pos-terior half of the intervertebral disc (Figure 3). To create a com-plete segmental injury, the anterior longitudinal ligament and theanterior half of the intervertebral disc were also excised (Figure 4).Clearly, the order of testing for injury condition was not random-ized. Instead, the order of testing for this study followed a logicalsequence: control (stable), posterior ligamentous injury, and com-plete segmental injury.

It must be noted that to maximize the response variable ofthis study (i.e., flexion–extension motion), a cervical immobi-lization collar was not used during the execution of transfertechniques. In the case of a true emergency, it is very unlikelythat a victim would be transferred to a spine board without acervical collar in place.

To collect motion data, one sensor was positioned on the an-terior surface of the body of C5 and another on the anterior sur-face of the body of C6 (Figure 5). The position of both sensors waskept the same throughout the entire study. Each sensor recordedflexion–extension motion data from its respective location. Tosecure the sensors from the Fastrak device onto the chosen land-marks, sensors were first fitted onto small mountings constructedwith machined polyethylene. These mountings were in turn se-cured to the vertebral bodies using carbon-fiber rods (Figure 5). Itmust be noted that, while the sensors from the Fastrak device weretethered, they did not restrict motion in any way. Throughout allstages of the experiment, there was enough slack in the wires toallow the sensors to move freely with the vertebral segment towhich they were attached.

A single dependent variable was considered in this study:flexion–extension of the C5–C6 segment. This measure was

Figure 4. A complete segmental injury.

Figure 5. Location of sensors (C5, C6).

E136 Spine • Volume 29 • Number 7 • 2004

dependent on two sets of experimental conditions or two inde-pendent variables: technique and injury condition. To analyzethe collected data, a 2 � 3 (technique � injury condition) anal-ysis of variance with repeated measures was used. Simple inter-action effects and post hoc tests (using Tukey’s HSD method)were calculated when appropriate. All statistical analyses wereperformed using an SPSS statistical software package (SPSS,Inc., version 10.1, Chicago, IL) with the level of significance forall statistical tests set, a priori, at � � 0.05.

Results

The mean for age, height, and weight of the enrolledparticipants was 23.8 � 3.3 years, 169.4 � 17.8 cm, and76.1 � 21.7 kg, respectively. The mean age and weight ofthe two male and three female cadavers used in this studywas 78.8 � 14.64 years and 66.68 � 15.10 kg,respectively.

Data collected in this study are presented in Tables 1and 2. Statistical tests performed on the means of thetotal range of segmental flexion–extension motion datarevealed a significant main effect for injury conditiononly (F2,38 � 51.45, P � 0.0001). Post hoc tests revealedsignificant differences between all three injury conditions(W � 0.88°) (Table 2).

Discussion

With prehospital management of spinal injuries, the pri-mary goal is to protect neurologic tissues from furtherinjury. Preserving the integrity of these tissues is bestaccomplished by immobilizing the injured spine. As partof a protocol that is fairly implicit, a victim of spinaltrauma is shifted onto a spine board by way of a transfertechnique to enable medical personnel to secure, andthus, fully immobilize the injured spinal column. Earlyresearch, however, has indicated that the LR maneuvermay not be appropriate for transferring victims and pa-tients with thoracolumbar injuries. McGuire et al12 re-ported that in a cadaver with a surgically created, com-plete segmental instability of the L1–L2 unit, theexecution of the LR maneuver produced 21 mm of an-

teroposterior displacement, 5 mm of lateral displace-ment, and 30° of rotational displacement through thesite of the experimental lesion. Radiographic imagestaken at the completion of the LR maneuver revealed anadditional 4° of sagittal plane angulation at the unstablesegment. The results of a study by Suter et al13 addedfurther support to the notion that the LR maneuver maynot be suitable for maintaining the alignment of the tho-racolumbar spine. Suter et al13 reported that placinghealthy individuals in the lateral recumbent position (aswith the LR) produced an average of 15 mm of thoraco-lumbar deviation. Regrettably, there had not been asingle study that reported the impact that transfer tech-niques have on the (unstable) cervical spine. Further-more, it would have been inappropriate to draw conclu-sions regarding the effect of the LR maneuver on thecervical spine from results of studies that only consideredthe thoracolumbar spine.

In order to establish if the LR maneuver and the LStechnique can be used safely with victims of cervicalspine trauma, we carried out this preliminary evaluationin which we compared the amount of flexion–extensionmotion generated across a single segment of the lowercervical spine. The results of this initial study revealedthat rescuers were able to satisfactorily restrict motionusing both transfer techniques. In other words, the re-sults of this study demonstrated that the LR and the LSwere equally effective transfer techniques.

Some might argue that while this study has providedsome details regarding the effectiveness of spine-boardtransfer techniques, the true significance of our data can-not be established because the magnitude of angulationmotion that the cervical spine can tolerate before a neu-rologic injury is either precipitated or exacerbated hasnot yet been determined. Defining such a critical value,however, is no simple task. Along with structural (ana-tomic) differences among individuals, there are obviousmethodologic and ethical considerations involved withidentifying this critical value. At any rate, the issue of acritical value becomes irrelevant in the case of this studybecause the LR and the LS produced similar amounts offlexion–extension motion. It is clear that in the case of aspinal injury, the less motion there is, the better the out-come. Since our findings did not indicate one techniqueto be superior to the other, the motion produced witheither the LR or LS is, in effect, the least amount possible.

According to our data, an increase in spinal segmentmotion may not be completely avoidable when there issignificant damage to movement restraints such as liga-ments and capsules. We noticed that as the level or de-gree of instability increased, so also did the amount ofsegmental motion, regardless of technique. Naturally,this finding suggests, not surprisingly, that individualsthat suffer severe structural damage of the vertebral col-umn are at potentially greater risk of suffering neurologicinjuries that are secondary to those produced by the in-citing trauma.

Table 1. Total Flexion-Extension Motion Generated WithTransfer Techniques

Technique n Min (°) Max (°)Mean

(°) SD (°)

LR 60 1.16 10.54 3.92 2.44LS 60 .64 12.47 3.69 2.80

Table 2. Impact of Injury Severity on Performance ofTransfer Techniques

Condition n Mean (°) SD (°)

Stable 40 2.32* 0.87Posterior instability 40 3.23* 3.35Complete instability 40 5.87* 3.14

* P � 0.0001.

E137Spine Board Transfer Techniques • Del Rossi et al

As with most experimental research, our study wasnot without limitations. We recognize that spinal columnmovements generated in cadavers do not mimic thosethat may be produced in vivo. We also realize that alongwith the injury conditions we tested in this study, thereare many other types of injuries that yield unstable spinalsegments. Additionally, these diverse injuries may be af-fected differently depending on the location (spinal level)of the injury. Finally, we are aware that our sample sizewas small (chosen because of external constraints) andthat increasing the number of cadaver specimens couldhave perhaps improved the likelihood that more of ourfindings (such as differences between techniques) wouldhave been statistically significant.

Notwithstanding the shortcomings of this study, wefeel we have made some progress in learning about thesafety of spine-board transfer techniques. However,more research is warranted in this area. Above all else,we believe that to further our understanding of howtransfer techniques affect an unstable cervical spine, fu-ture investigations must include larger sample sizes andmust also quantify the amount of axial distraction andanteroposterior displacement that is produced during thetransfer process.

Conclusion

Given the results of this study, neither the LR nor the LSemerged as the clearly better technique for restrictingmotion of the lower segments of the cervical spine. Basedon the results of this study, we recommend that medicalpersonnel familiarize themselves with both techniquesand make decisions regarding which technique to use onthe circumstances of their situation (e.g., the preferenceof the rescue team, the position of the victim).

Key Points

● Neither the log-roll maneuver nor the lift-and-slide technique emerged as the clearly better spine-board transfer technique. Trained and practiced

rescue teams were able to execute the log-roll ma-neuver equally as well as the lift-and-slidetechnique.● It appears that the severity of a cervical spineinjury may impact the amount of motion that isproduced during the execution of spine-boardtransfer techniques.● Along with rotational movement (such as flex-ion–extension), translation movements (such asanterior–posterior displacement) must also be con-sidered when attempting to confirm the efficacy ofspine-board transfer techniques.

References

1. Podolsky S, Baraff LJ, Simon RB, et al. Efficacy of cervical spine immobili-zation methods. J Trauma. 1983;23:461–465.

2. Bivins HG, Ford S, Bexmalinovic Z, et al. The effect of axial traction duringorotracheal intubation of the trauma victim with an unstable cervical spine.Ann Emerg Med. 1988;17:53–57.

3. Gerling MC, Davis DP, Hamilton RS, et al. Effects of cervical spineimmobilization technique and laryngoscope blade selection on an unsta-ble cervical spine in a cadaver model of intubation. Ann Emerg Med.2000;36:293–300.

4. Lennarson PJ, Smith DW, Sawin PD, et al. Cervical spinal motion duringintubation: efficacy of stabilization maneuvers in the setting of completesegmental instability. J Neurosurg. 2001;94:265–270.

5. Lennarson PJ, Smith DW, Todd MM, et al. Segmental cervical spine motionduring orotracheal intubation of the intact and injured spine with and with-out external stabilization. J Neurosurg. 2000;92:201–206.

6. DeLorenzo RA, Olson JE, Boska M, et al. Optimal positioning for cervicalimmobilization. Ann Emerg Med. 1996;28:301–308.

7. Donaldson WF, Lauerman WC, Heil B, et al. Helmet and shoulder padremoval from a player with suspected cervical spine injury. Spine. 1998;23:1729–1733.

8. Gastel JA, Palumbo MA, Hulstyn MJ, et al. Emergency removal of footballequipment: a cadaveric cervical spine injury model. Ann Emerg Med. 1998;32:411–417.

9. Palumbo MA, Hulstyn MJ, Fadale PD, et al. The effect of protective footballequipment on alignment of the injured cervical spine. Am J Sports Med.1996;24:446–453.

10. Del Rossi G, Horodyski MB, Kaminski TW. Management of cervical-spineinjuries. Athletic Ther Today. 2002;7:46–51.

11. White AA, Panjabi MM. Clinical Biomechanics of the Spine. Philadelphia:Lippincott, 1990:284–316.

12. McGuire RA, Neville S, Green BA, et al. Spinal instability and the log-rollingmaneuver. J Trauma. 1987;27:525–531.

13. Suter RE, Tighe TV, Sartori J, et al. Thoraco-lumbar spinal instability duringvariations of the log-roll maneuver. Prehosp Dis Med. 1992;7:133–138.

E138 Spine • Volume 29 • Number 7 • 2004