Noninvasive Reduction of Open-Book Pelvic Fractures by ...biomechresearch.org/PDF/Noninvasive Reduction of Open...Consecutive open-book fracture model, subsequently simulating two

$Page 1: Noninvasive Reduction of Open-Book Pelvic Fractures by ...biomechresearch.org/PDF/Noninvasive Reduction of Open...Consecutive open-book fracture model, subsequently simulating two$
Noninvasive Reduction of Open-Book Pelvic Fractures byCircumferential Compression

Michael Bottlang, *Tamara Simpson, Juergen Sigg, James C. Krieg, Steven M. Madey,and William B. Long

Biomechanics Laboratory, Legacy Health System, and *Oregon Health Sciences University, Portland, Oregon, U.S.A.

Objectives: To determine the efficacy and optimal applicationparameters of circumferential compression to reduce externalrotation-type pelvic fractures.Design: Biomechanical investigation on human cadavericspecimens.Setting: Biomechanics laboratory.Intervention: Partially stable and unstable external rotationinjuries of the pelvic ring (OTA classification 61-B1 and 61-C1) were created in seven human cadaveric specimens. A pro-totype pelvic strap was applied subsequently at three distincttransverse levels around the pelvis. Circumferential pelviccompression was induced by gradual tensioning of the strap toattempt complete reduction of the symphysis diastasis.Main Outcome Measurements: Pelvic reduction was evalu-ated with respect to strap tension and the strap application site.The effect of circumferential compression on intraperitonealpressure and skin–strap interface pressure was measured.

Results: A successive increase in circumferential compressionconsistently induced a gradual decrease in symphysis diastasis.An optimal strap application site was determined, at whichcircumferential compression most effectively yielded pelvic re-duction. The minimum strap tension required to achieve com-plete reduction of symphysis diastasis was determined to be177 ± 44 Newtons and 180 ± 50 Newtons in the partially stableand unstable pelvis, respectively.Conclusions: Application of circumferential compression tothe pelvic soft tissue envelope with a pelvic strap was an effi-cient means to achieve controlled reduction of external rota-tion-type pelvic fractures. This study derived application pa-rameters with direct clinical implication for noninvasive emer-gent management of traumatic pelvic ring disruptions.Key Words: Pelvic ring disruption, Hemorrhagic shock, Emer-gent care.

Unstable pelvic fractures are associated with a highincidence of morbidity and mortality (9,16,18,25,33),with hemorrhage being the leading cause of death (7,10,22). Although the source of hemorrhage can be frommultiple sites, the degree and type of pelvic disruptiondirectly correlate with retroperitoneal blood loss (6,11,16,26). Even though a limited set of treatment modalitiesfor hemodynamically unstable patients with unstable pel-vic fractures is available, no general consensus existsregarding the optimal methods of emergent management.Several authors have proposed treatment algorithms forthe management of life-threatening hemorrhage in pelvictrauma (3,16,18,19,25,27,28,29). These managementschemes include fluid resuscitation, correction of coagu-lopathy, selective angiography, and temporary pelvic re-

duction and stabilization. Among these, pelvic reductionand stabilization in the early post-traumatic phase arereported to provide the most effective means of control-ling venous hemorrhage (3,8,13,27). Pelvic reduction re-aligns bleeding fracture surfaces, whereas pelvic stabili-zation facilitates clot formation by reducing fracture sitemotion (14). Despite the potential benefits of such inter-vention, few devices for emergent, noninvasive pelvicreduction and stabilization at the accident site are avail-able (34).

Current treatment options for immediate pelvic stabi-lization include open reduction and internal fixation(ORIF) (32,35), closed reduction and percutaneous fixa-tion (27), external fixation (25), posterior pelvic reduc-tion clamps (13), inflatable pneumatic trousers (2), andcircumferential wrapping of the pelvic region with asheet or strap (28,34). All of these approaches exhibitpotential problems in an emergent situation. ORIF andpercutaneous fixation have the potential to provide su-perior reduction and the most stable fixation, yet areinvasive, require an operating room environment, andmay delay angiography (27,32,35). In addition, ORIFmay release the evolving pelvic hematoma. Anterior ex-ternal fixation has inherent deficiencies in reducing and

Accepted July 9, 2001.Address correspondence and reprint requests to Dr. Michael

Bottlang, Legacy Clinical Research & Technology Center, 1225 NE2nd Avenue, Portland, OR 97232, U.S.A.

Supported by a grant from the Legacy Research Advisory Commit-tee. The authors have received nothing else of value.

The device that is subject of this manuscript is not FDA approvedand is not commercially available in the United States.

Journal of Orthopaedic TraumaVol. 16, No. 6, pp. 367–373© 2002 Lippincott Williams & Wilkins, Inc., Philadelphia

367

stabilizing posterior pelvic disruptions (15,20,24). Pelvicantishock clamps may be difficult to apply and havecaused severe complications (14,19). Inflatable pneu-matic trousers limit access to critical areas of interest inthe polytraumatized patient, namely, the perineum andabdomen. They can cause systemic complications (4,16,21), and their pressurization has been associated withcompartmental syndromes in the lower extremities (12,17,21). Most recently, the use of circumferential pelvicsheets and straps for reduction and stabilization of pelvicfractures has been reported (3,27,28,34). The noninva-sive nature of these devices minimizes the risk of seriouscomplication and allows application at the emergencysite. Although this technique has recently been recom-mended by the American College of Surgeons Commit-tee on Trauma (1), no data are available to describe theapplication, efficacy, and safety of circumferential pelviccompression to unstable pelvic fractures.

We conducted a cadaveric, biomechanical investiga-tion on the effectiveness of a noninvasive pelvic strap toinduce pelvic reduction by means of circumferentialcompression. We attempted to determine the most effec-tive strap application site as well as the strap tensionrequired to reduce a partially stable and unstable cadav-eric pelvic fracture model.

MATERIALS AND METHODS

Four male and three female nonembalmed whole bodycadaveric specimens were used, ranging in age from sev-enty-three to ninety-three years with an average age of79.7 years. The average specimen height was 170 centi-meters, ranging from 157 to 185 centimeters. The bodymass index ranged from 21.7 to 30.8 kilograms persquare meter with an average value of 25.7 kilograms per

square meter (normal body mass index range 19 to 25kilograms per square meter). Each specimen was instru-mented with three different sensing devices as shown inFigure 1. Two six-degrees-of-freedom motion trackingsensors (pcBird, Ascension Tech. Corp., Burlington, VT,USA) were rigidly mounted on the superior pubic rami ina parasymphyseal location, using custom-built nonferro-magnetic mounting clamps. These sensors were used toassess the spatial location and motion of each hemipel-vis. These hemipelvic kinematics data, in combinationwith coordinate digitization on the superior symphysealregion, allowed us to compute the apparent displacementvector ds at the pubis symphysis. Furthermore, a sym-physis pubis sensor was designed to detect the symphy-seal contact that occurred with complete reduction ofsymphysis diastasis. This sensor determined electricalconductivity between two metallic films that were fittedto each osteotomy surface after symphysiotomy. Finally,intraperitoneal pressure was monitored with a pressuresensor (HP 78534, Hewlett Packard, Englewood, CO,USA). To obtain reproducible and reliable recordings,this sensor measured the pressure inside a fifty-cubiccentimeter fluid-filled pressure pouch, which was surgi-cally inserted into the intraperitoneal cavity. After speci-men instrumentation the pelvic ring was weakened in adefined manner to achieve reproducible fractures. Ante-riorly, a complete symphysiotomy was performed. Pos-teriorly, the left sacroiliac ligaments and joint capsulewere exposed via an anterior approach and surgicallytransected. Subsequently, two distinct unilateral pelvicring fractures were induced by forced external rotation ofthe left hemipelvis (Fig. 2). For this purpose one-inch-wide belts were inserted through each obturator foramen.Lateral tensioning of these belts was manually appliedwith the aid of a lever arm. The motion tracking system

FIG. 1. Instrumentation of specimens and pelvic strap for assessment of intraperitoneal pressure, strap tension, skin–strap interfacepressure, symphysis contact, and spatial motion of each hemi-pelvis.

M. BOTTLANG ET AL.368

J Orthop Trauma, Vol. 16, No. 6, 2002

was used to provide continuous feedback of the apparentsymphysis diastasis. First, a partially stable external ro-tation fracture (OTA classification 61-B1) was modeledas having a symphysis diastasis of fifty millimeters withunilateral disruption of the anterior sacroiliac joint. Thisfracture was subsequently expanded to an unstable pelvicfracture (OTA classification 61-C1) characterized by100-millimeter symphysis diastasis and complete unilat-eral sacroiliac joint disruption.

An experimental prototype pelvic strap was designed,consisting of a fifty-millimeter-wide, flexible, nonelasticrubber strap, which encircled the pelvis and was guided

anteriorly over rollers contained in a buckle (Fig. 3a).Manual application of tension FT to the ends of the strapgradually induced circumferential pelvic compression.Two sensors were integrated into the strap to providereal-time assessment of the applied strap tension (ELFS-500N, Entran, Fairfield, NJ, USA) and the resulting lat-eral strap–skin interface pressure (HP 78534, HewlettPackard, Englewood, CO, USA). For each specimen thestrap was applied in three distinct transverse plane levelsaround the pelvis (Fig. 3b). At transverse plane level I thepelvic strap was located to circumscribe the symphysispubis and the greater trochanteric region. Transverse

FIG. 2. Consecutive open-book fracture model, subsequently simulating two distinct injury severity patterns, corresponding to OTAclassifications 61-B1 and 61-C1.

FIG. 3. Independent experimental variables of the pelvic strap. a: Strap tension Ft. b: Application levels I to III.

NONINVASIVE REDUCTION OF PELVIC FRACTURES 369


level II was centered midway between the symphysispubis and the iliac crest. At transverse level III the pelvicstrap was applied between the anterior–superior iliacspine and the iliac crest. Anteroposterior fluoroscopicvisualization was used to ensure proper strap placement.Circumferential pelvic compression was induced manu-ally by gradual application of laterally directed tension toeach strap end. The pelvic strap was first applied to re-duce the partially stable, and subsequently, the unstablepelvic fracture. Resulting strap tension, intraperitonealpressure, strap–skin interface pressure, and spatial mo-tion of the hemipelvis relative to each other were simul-taneously recorded (SCXI 1000, National Instruments,Austin, TX, USA) until symphysis contact was detected.Finally, fracture patterns were documented on pelvic an-teroposterior as well as inlet and outlet radiographs, ob-tained for each specimen before and after fracture reduc-tion. In addition, a computerized tomographic (CT) scanof one specimen was obtained before and after reductionof the unstable pelvis to visualize pelvic ring congruity inthe transverse plane.

Outcome parameters were statistically analyzed usinga two-tailed t test for paired samples. Statistical signifi-cance was determined at ! < 0.05, and mean and stan-dard error of the mean were documented.

RESULTS

Fluoroscopic radiographs of the pelvis demonstratedconsistent and graded pelvic ring fractures in each of theseven specimens (Fig. 4). Motion tracking recordingsyielded a symphyseal diastasis at the time of maximalpelvic disruption of 56.2 ± 6.4 millimeters for the par-tially stable pelvis and 99.0 ± 5.8 millimeters for theunstable pelvis. After release of the disruption force, astress–relaxation recoil behavior of the pelvis was ob-served accompanied by a significant decrease in sym-physis diastasis. This pelvic relaxation reached a steadystate after about sixty seconds, at which time the sym-physis diastasis of the partially stable and unstable pelvis

decreased by 50.1 ± 14.8 percent and 54.3 ± 7.2 percent,respectively (Fig. 5).

With pelvic strap application, a continuous increase ofstrap tension FT up to 200 Newtons induced a gradualdecrease in displacement ds at the pubis symphysis (Fig.6). After closure of symphysis diastasis, detected by thesymphysis contact sensor, the residual symphyseal mis-alignment in the sagittal plane was ds ! 5.8 ± 3.6 mil-limeters for reduction of the unstable pelvis by strapapplication at level I. In this scenario closure of symphy-sis diastasis was achieved at a strap tension of 177 ± 44Newtons for the partially stable pelvis and 180 ± 50Newtons for the unstable pelvis (Fig. 7). Symphysis di-astasis closure due to strap application at levels II and IIIrequired a significantly higher strap tension of 228 ± 55and 262 ± 79 Newtons, respectively.

Reduction of the unstable pelvic fracture by strap ap-plication at level I was characterized by an intraperito-neal pressure increase of 6.2 ± 5.8 millimeters of mer-cury and a strap–skin interface pressure of twenty-fourmillimeters of mercury. Pelvic reduction by strap appli-cation at levels II and III caused an intraperitoneal pres-sure increase of 19.4 ± 13.8 and 20.9 ± 13.2 millimetersof mercury, respectively.

CT images of one specimen obtained at levels of thefifth lumbar vertebra and the symphysis pubis demon-strated that strap application at level I effectively inducedcircumferential soft tissue compression, which evokednear-anatomic reduction of the sacroiliac joint and thesymphysis pubis (Fig. 8).

DISCUSSION

Unstable pelvic ring disruptions benefit from emer-gent reduction and stabilization, aimed at controlling ret-roperitoneal hemorrhage (8,12,23,26,27). While initialclinical results obtained with pelvic sheets and strapshave been encouraging, published information remainsrare and a biomechanical technique evaluation has beenabsent. Vermeulen et al. (34) reported the application of

FIG. 4. Anteroposterior fluoroscopicradiographs of the partially stable (a)and unstable (b) pelvic fracture model,obtained after recoil of the pelvic ring.



a pelvic strap to nineteen patients with pelvic injuries.Their pelvic strap was applied at the accident scene byparamedics upon suspicion of unstable pelvic lesions.Two of their patients showed no abnormalities on initialpelvic radiographs obtained with the strap in place. How-ever, a four-centimeter opening of the symphysis pubisappeared after strap removal, and associated sacrum frac-tures were present in both cases. They reported a pelvicstrap application time of approximately thirty secondsand emphasized significant time savings compared withthe use of pneumatic garments (34). Routt et al. (28)encircled the pelvic region of one hemodynamically un-stable patient with bilateral pubic ramus fractures andsacroiliac joint disruptions in a snugly applied sheet. Thesheet was applied in the angiography suite. They empha-sized the importance of early stabilization by circumfer-ential pelvic compression but pointed out the potentialdamage to nerve roots that can be caused by overcompres-

sion in patients with sacral foramina fractures. However,no detailed information on the amount of applied pelviccompression or its application site has been reported.

This research delineates for the first time the applica-tion and efficacy of this noninvasive pelvic stabilizationapproach in a biomechanical study on cadaveric speci-mens. Our study design was based on established frac-ture models. Unstable unilateral open-book fractures indenuded pelvic specimens have previously been createdto assess the rigidity provided by external fixation frames(5) and ORIF (30,31). Brown et al. (5) concluded thatsupplementary support from soft tissue is an importantfactor in successful clinical modeling of pelvic ring ri-gidity. Ghanayem et al. (14) used specimens with anintact soft tissue envelope to create unilateral open-bookpelvic fractures by direct dissection of the pubis sym-physis and the left sacroiliac joint. Grimm et al. (15) usedforced external rotation to both iliac wings to create

FIG. 6. Symphyseal displacement atmaximum pelvic disruption, at sixtyseconds after passive relaxation, andafter application of pelvic circumferen-tial compression under strap tensionup to 200 Newtons.

FIG. 5. Passive relaxation of the pelvicring, measured sixty seconds afterforced, maximum external expansion upto a symphysis diastasis of 56.2 ± 6.4millimeters and 99 ± 5.8 millimeters inthe partially stable and unstable pelvicfracture model, respectively.



open-book pelvic fractures. After unsuccessful fractureattempts in some specimens, further disruption was fa-cilitated by transsection of the pubis symphysis. Expand-ing on the Grimm et al. (15) experience, we created aconsecutive open-book fracture model, simulating twodistinct injury severity patterns in each specimen. Thismodel delivered for the first time a graded, highly repro-ducible symphysis diastasis and was essential to system-atically assess the effects of strap tension and position onpelvic reduction.

The cadaveric biomechanical study design was wellsuited to establish optimal application parameters for a

pelvic strap in terms of its most effective application site(i.e., transverse plane, including the pubis symphysis andthe greater trochanteric region) and the required straptension of 180 Newtons. For comparison this tension isequivalent to the force required for lifting an eighteen-kilogram weight. A large amount of passive recoil of thepelvic ring after fracture was consistently observed. Thissuggests that the actual damage sustained by an indi-vidual may be severely underestimated if predictedsolely by the magnitude of the diastasis apparent onpostinjury radiographs. The reported skin–strap interfacepressure of twenty-four millimeters of mercury is directly

FIG. 8. CT images of a specimen with an unstable open-book pelvic fracture (OTA, Type 61-C1), obtained in two transverse planesbefore and after application of the pelvic strap at level I.

FIG. 7. Effect of strap application levelon the strap tension required to inducereduction of partially stable and un-stable open-book fractures. Reductionwas defined by symphysis contact asindicated by the symphysis contactsensor.



comparable to the inflation pressure of antishock gar-ments, which typically ranges from twenty to thirty mil-limeters of mercury. Such garments can be left in placefor up to forty-eight hours without causing adverse ef-fects on skin viability (2).

Although this study provides clinically relevant pa-rameters for pelvic strap application, its outcome canonly theoretically infer the potential of a pelvic strap toreduce retroperitoneal hemorrhage. However, based onthe assumption that early reduction and stabilization ofan unstable pelvis are crucial to decrease the extent ofretroperitoneal hemorrhage, pelvic straps are likely toprovide a highly effective, emergently applicable inter-vention (3,13,27). Specimen size and weight have notbeen considered as experimental variables, which con-fines result interpretation to an adult population withinthe reported specimen size and weight range. Renderingof the pelvic geometry from CT scans before and afterpelvic strap application was obtained for only one speci-men. It therefore can serve only as an example to quali-tatively visualize pelvic reduction induced by circumfer-ential compression.

Finally, whereas this study provided clinically relevantparameters for pelvic strap application, the results cannotaddress potential complications associated with a pelvicstrap. Further laboratory studies are therefore required toinvestigate the potential for harm if the pelvis strap isapplied at the accident site where the pelvic fracturepattern cannot be reliably assessed. In addition, furtherinvestigations should address the effects of patient move-ment and transportation on the quality of reduction anddegree of stabilization over time.

Within these constraints and limitations, the presentresearch delineated key parameters for the provision ofemergent pelvic reduction and stabilization with a pelvisstrap. These outcome data are directly applicable to ad-vance the clinical concept of circumferential pelvic com-pression toward a more controlled and effective emer-gent care procedure.

Acknowledgment: The authors thank Markus Mohr for hisassistance.

REFERENCES1. American College of Surgeons. Advanced trauma life support for

doctors, ATLS. Instructor Course Manual 1997;24:206–209.2. Batalden DJ, Wickstrom PH, Ruiz E. Value of the G suite in patients

with severe pelvic fractures. Arch Surg 1974;109:326–329.3. Bottlang M, Sigg J, Simpson T, et al. Emergent Non-invasive

Reduction of Pelvic Ring Disruptions. Chicago: Trans 24th AmSoc Biomech, 2000:45–46.

4. Brotman S, Soderstrom CA, Oster-Granite M, et al. Managementof severe bleeding in fractures of the pelvis. Surg Gynecol Obstet1981;153:823–826.

5. Brown TD, Stone JP, Schuster JH, et al. External fixation of un-stable pelvic ring fractures: comparative rigidity of some currentframe configurations. Med Biol Eng Comput 1982;20:727–733.

6. Burgess AR, et al. Pelvic ring disruptions: effective classificationsystem and treatment protocols. J Trauma 1990;30:848–856.

7. Cryer HM, Miller FB, Evers BM, et al. Pelvic fracture classifica-tion: correlation with hemorrhage. J Trauma 1988;28:973–980.

8. Culemann U, Reilmann H. Injury of the pelvic ring. Unfallchirurg1997;100:487–496.

9. Dujardin FH, Hossenbaccus M, Duparc F, et al. Long-term func-tional prognosis of posterior injuries in high-energy pelvic disrup-tions. J Orthop Trauma 1998;12:145–151.

10. Evers M, Cryer H, Miller FB. Pelvic fracture hemorrhage. ArchSurg 1989;124:422–424.

11. Failinger MS, McGanity LJ. Current concepts review: unstable frac-tures of the pelvic ring. J Bone Joint Surg Am 1992;74:781–791.

12. Flint L, Babikian G, Anders M, et al. Definitive control of mor-tality from severe pelvic fracture. Ann Surg 1990;211:703–707.

13. Ganz R, Krushell RJ, Jakob RP, et al. The antishock pelvic clamp.Clin Orthop 1991;267:71–78.

14. Ghanayem AJ, Wilber JH, Lieberman JM, et al. The effect oflaparotomy and external fixator stabilization on pelvic volume inan unstable pelvic injury. J Trauma 1995;38:396–401.

15. Grimm MR, Vrahas MS, Thomas KA. Pressure-volume character-ization of the intact and disrupted pelvic retroperitoneum. JTrauma 1998;44:454–459.

16. Henry S, Tornetta P, Scalea T. Damage control for devastating pelvicand extremity injuries. Surg Clin North Am 1997;77:879–895.

17. Jennings TJ, Seaworth JF, Tripp LD. The effects of inflation ofantishock trousers on hemodynamics in normovolemic subjects. JTrauma 1986;26:544–548.

18. Kregor PJ, Routt MLC. Unstable pelvic ring disruptions in un-stable patients. Injury 1999;30:19–28.

19. Lazarus MD, Born CT. Advances in the management of muscu-loskeletal trauma. Curr Opin Gen Surg 1993;46–54.

20. Lindahl J, Hervensalo E, Bostman O, et al. Failure of reductionwith an external fixator in the management of injuries of the pelvicring: long term evaluation of 110 patients. J Bone Joint Surg Br1999;81:955–962.

21. Mattox KL, Bickell W, Pepe PE, et al. Prospective MAST study in911 patients. J Trauma 1989;29:1104–1112.

22. Moreno C, Moore EE, Rosenberger A, et al. Hemorrhage associ-ated with major pelvic fractures: a multispecialty challenge. JTrauma 1986;26:987–994.

23. Mucha P, Farnell MB. Analysis of pelvic fracture management. JTrauma 1984;24:379–386.

24. Palmer S, Fairbank AC, Bircher M. Surgical complications andimplications of external fixation of pelvic injuries. Injury 1997;28:649–653.

25. Riemer BL, et al. Acute mortality associated with injuries to thepelvic ring: the role of early patient mobilization and externalfixation. J Trauma 1993;35:671–677.

26. Rothenberger DA, Fischer RP, Perry JF. Major vascular injuriessecondary to pelvic fractures: an unsolved clinical problem. Am JSurg 1978;136:660–662.

27. Routt MLC, Simonian PT, Ballmer F. A rational approach to pel-vic trauma: resuscitation and early definitive stabilization. ClinOrthop 1995;318:61–74.

28. Routt MLC, Simonian PT, Swiontkowski MF. Stabilization of pel-vic ring disruptions. Orthop Clin North Am 1997;28:369–388.

29. Sigg J, Bottlang M, Simpson T, et al. Open Book Pelvic Fractures:Effect of Pelvic Reduction and Hematoma Formation on Retroper-itoneal Pressure. Chicago: Trans 24th Am Soc Biomech,2000:167–168.

30. Simonian PT, Routt MLC, Harrington RM, et al. Internal fixationof the unstable anterior pelvic ring: a biomechanical comparison ofstandard plating techniques and the retrograde medullary superiorpubic ramus screw. J Orthop Trauma 1994;8:467–482.

31. Simonian PT, Routt MLC, Harrington RM, et al. Anterior versusposterior provisional fixation in the unstable pelvis. Clin Orthop1995;310:245–251.

32. Tile M. Pelvic ring fractures: should they be fixed. J Bone JointSurg Br 1988;70:1–12.

33. Tile M, Pennal GF. Pelvic disruptions: principles of management.Clin Orthop 1980;151:56–64.

34. Vermeulen B, Peter R, Hoffmeyer P, et al. Prehospital stabilizationof pelvic dislocations: a new strap belt to provide temporary he-modynamic stabilization. Swiss Surg 1999;5:43–46.

35. Vrahas MS, Wilson SC, Cummings PD, et al. Comparison of fixa-tion methods for preventing pelvic ring expansion. Orthopedics1998;21:285–289.



Documents

Noninvasive Reduction of Open-Book Pelvic Fractures by ...biomechresearch.org/PDF/Noninvasive Reduction of Open...Consecutive open-book fracture model, subsequently simulating two