Outcomes of patients with pelvic-ring fractures managed by open reduction internal fixation.

There is an absence of literature to describe the course of injury and outcome of patients following multiple traumas. Clinicians dealing with these patients do not have a basis for comparison to determine how a patient is performing in relation to others with similar injuries. The patients described in this case report had multiple traumas, including unstable pelvic-ring fractures that were managed by open reduction internal fixation (ORIF) as well as associated injuries. This report describes a group of patients with injuries so severe that the patients were not allowed to bear weight on either lower extremity for 3 months.

A fracture of the pelvic ring represents severe trauma. Most pelvic fractures are the result of motor vehicle accidents or falls from great heights. The prevalence of pelvic fractures is 3% of all skeletal injuries. The associated morbidity, however, is disproportionately high, which presents a challenge for the medical management of thcse patients.[1])

Unstable pelvic-ring fractures have been managed by bed rest, skeletal traction, or use of pelvic slings or hip spica casts.[2] All of these methods of immobilization require prolonged periods of recumbency, which result in increased risks of pulmonary, urologic, and cutaneous complications. In addition, there is a considerable risk of a nonunion or a malunion of the pelvic fracture.[1] Because of these problems, methods of external and internal fixation of the pelvis have been developed to stabilize unstable pelvic-ring fractures.[1]

Percutaneous pins with one end placed internally into the anterior iliac crests are attached with the other end externally to a rigid frame for external fixation. Most of the problems associated with external fixation are related to a failure to achieve adequate stabilization of the posterior portion of the pelvic ring. As a result, there have been reports of posterior pelvic pain,[1,2] sitting imbalance attributed to a malaligned pelvis,[1] and decreased activity levels.[1,2]

Internal fixation requires surgery to explore the pelvis, followed by reduction of the displaced fracture segments. Plate fixation of the bone then holds the fragments together. The major advantage of this approach is that the anterior and posterior portions of the pelvic ring are rigidly stabilized. The disadvantages include the potential for hemorrhage or iatrogenic nerve injury. With the anatomical reduction provided by this method, however, the likelihood of nonunion and malunion is reduced. In addition, the rigid stabilization of the posterior pelvis allows the patient to move around sooner, including an earlier advancement to full weight bearing.[1,2]

To provide a framework for describing patients in the report, the International Classification of Impairments, Disabilities, and Handicaps[3] (ICIDH) developed by the World Health Organization will be used. The first category of the ICIDH, "disease," represents any pathologic process associated with a characteristic and identifiable set of symptoms and signs. The second component is "impairment," which is described as any loss or abnormality of psychologic, physiologic, or anatomic structure within a specific organ or system of the body. The next element, "disability," is defined as any restriction or lack of ability to perform routine activities of daily living. The final component, "handicap," refers to a disadvantage that limits or prevents an individual's fulfillment of a role that is considered normal based on the individual's age, gender, and cultural background, including occupational responsibilities and desired social interactions.[4,5]

The fracture of the pelvic ring represents an anatomical "impairment" caused by the traumatic "disease." During the initial assessment of these patients by a physician, clinical and radiologic examinations are performed. An unstable pelvic fracture exists if the plane radiographs and the computerized axial tomography scan show rotational or vertical displacement of the pelvic ring.[1] A classification system developed by Tile[2] defines the type, location, and extent of the pelvic fractures (Tab. 1). Secondary impairments may include limited range of motion (ROM) or decreased strength of the body segments adjacent to the pelvis.

Table 1.

Tile Classification of Pelvic Fracture(a)

Type A--stable
  A1--fractures of the pelvis not involving the pelvic ring
  A2--stable, minimally displaced fractures of the pelvic ring
Type B--rotationolly unstable, vertically stable
  B1--open book
  B2--lateral compression: ipsilateral
  B3--lateral compression: contralateral (bucket handle)
Type C--rotationally and vertically unstable
  C1--unilateral
  C2--bilateral
  C3--associated with an acetabular fracture

(a) Adapted from Tile.[2]

The ICIDH dimensions that reflect on the daily life of patients with multiple traumas and pelvic fractures, specifically the dimensions of "disability" and "handicap," in these patients have not been reported. These patients undergo lengthy surgeries, prolonged periods of non-weight bearing, and limited activity, as well as time away from household and occupational responsibilities. Studies investigating whether these patients are able to return to their previous occupational and physical activities do not exist.

The purpose of this report is to describe the impairments, disabilities, and handicaps of patients with multiple traumas following ORIF of unstable pelvic fractures. The physical performance measures consisted of lift capacity, forward bending, and ambulation. Limitations of activities of daily living and occupational responsibilities were examined through self-report.

Subjects

All subjects were recruited by the second author, an orthopedic surgeon at the University of Pittsburgh Medical Center (UPMC), as part of his follow-up of patients with trauma of the pelvis managed by ORIF. Of 230 consecutive patients admitted to UPMC with pelvic fractures and associated multiple traumas, 54 patients received surgical repair by ORIF. Three of these patients expired (2 with multisystem organ failure, I with pulmonary embolus), and 3 patients could not be located. The remaining 48 patients were invited to be examined as part of their follow-up. Thirty-one patients agreed to participate in this examination. The 17 patients who did not participate in the examination received physician follow-up with an office visit and completed the Sickness Impact Profile (SIP)[6] by phone. Nineteen men and women, with a mean age of 38 yea (SD = 16, range = 18 -71), were examined on a single day by the physician and three physical therapists. Each physical therapist performed the same assessment throughout the day. This examination took place at least 1 year (X=14.5 months) after gustaining multiple traumas, which included an unstable pelvic-ring fracture as defined by Tile.[2] All subjects were observed by the physician to have had their fractures reduced to within 2 mm, as determined by follow-up pelvic radiographs.

All subjects participated on a volunteer basis. The mechanisms of injury included motor vehicle crashes (52%), pedestrians struck by automobiles (17%), motorcycle crashes (11%), falls (7%), crush injuries (7%), and other (7%). Table 2 shows the types and distribution of multiple injuries sustained by the subjects. A large number of subjects sustained multiple fractures, with 57% of all fractures occurring in the upper and lower extremities. Many subjects also experienced gastrointestinal/abdominal (32%), genitourinary/bladder (19%), and cardiopulmonary (29%) systemic injuries. In addition, 7 subjects had closed head injuries (CHIs). The average length of hospital stay following injury was 21.2 [+ or -]16.5 days (X [+ or -] SD).

Table 2.

Associated Traumatic Injuries of the Subjects

                                           Percentage
Injury Type                  Prevalence    of Subjects

Gastrointestinal/abdominal   10            32
Genitourinary/bladder         6            19
Closed head injury            7            23
Peripheral neurologic         1             3
Deep venous thrombosis        2             6
Cardiopulmonary               9            29
Fractures
Lower extremity               12           34
  Acetabular                   4           11
  Upper extremity              8           23
  Axial                       11           32

Patient Management

Following ORIF of the pelvis, subjects were maintained on bed rest for 1 to 2 days. Physical therapy was then initiated to begin transfer and exercise training with the subjects. The short-term goals included independence with transfers and wheelchair mobility. Subjects also were expected to be able to perform a home exercise program prior to discharge from the hospital. The time frame for meeting these goals varied from 2 to 6 weeks, depending on the subjects' medical status.

Transfer training was accomplished with or without a sliding board, depending on the comfort and ease of the transfer. The subject was instructed to push with the arms on the supporting surface in order to raise the hips and begin moving to the transfer surface. This transfer was done slowly with small "scooting" movements. Most often, the subject managed to move the upper body, but the lower extremities may have been too weak or painful to be lifted as the transfer progressed. When this was the case, the wheelchair leg rests were fully elevated to maintain a long sitting position with the knees fully extended. The legs were then pulled along with the hips as the subject moved. If this method was too uncomfortable, someone manually guided the subject's legs as the subject moved the upper body. If there were also upperextremity or trunk injuries, some assistance may have been needed to achieve sitting balance, arm extension, or "scooting." The subject also practiced wheelchair mobility to gain access with independence.

The home exercise program included basic ROM and strengthening exercises intended to prevent contracture and reduce atrophy while the subjects were nonambulatory. Each subject performed isometric exercises (eg, gluteal or quadriceps femoris muscle sets), ROM exercises, and upper-extremity resistive exercises (eg, shoulder and elbow flexion and extension) until fatigued. The number of repetitions varied with each subject. Resistive exercises were not performed with the lower extremities because of the subjects' non-weight-bearing status. Many subjects had difficulty with full active ROM because of increased pain or weakness. Gentle active-assisted exercises, therefore, were often performed, as tolerated. The therapist assisted the subject by putting one hand under the subject's knee and the other hand under the subject's ankle for support, while slowly guiding the lower extremity through tolerable ranges (ie, as pain or discomfort would allow). Each subject was different, with progression based on individual responses. Some subjects tolerated only 5 to 10 repetitions of lower-extremity ROM exercise with assistance initially, whereas others could complete 20 to 30 repetitions with supervision. This then became the starting point, from which the subject further progressed. Ideally, the subjects could perform the entire home exercise program without assistance before they were discharged from the hospital. When assistance was still needed, however, a family member was shown how to assist the subject.

In many cases, family training was an important component of discharge planning. As mentioned, a family member may have been instructed to perform activeassisted exercises or to assist transfers by safely guiding the lower extremities while the subject moved in or out of the wheelchair. The disposition of the subject was in part based on his or her ability and resources at home. In some cases, home services or even skilled nursing facility placement was needed until the subject's weight-bearing status progressed to full weight bearing at 3 months postsurgery.

Once weight bearing was resumed, physical therapy consisted of gait training and resistive exercises for the trunk and extremities, along with cardiovascular training (eg, treadmill or exercise bicycle training). Aquatic therapy was used when available. Most subjects attended therapy on an outpatient basis; however, inpatient rehabilitation was also an option. Because of individual differences, the course of therapy was difficult to predict. During the acute stage following the injury, the subjects had similar restrictions and were wheelchair dependent. This was no longer true, however, during later phases of rehabilitation when the subjects could bear weight. The subjects' progression was no longer bound by any special limitations that were previously maintained by all subjects. The physical therapist then evaluated and treated according to the specific needs of the individual. Each therapy program varied accordingly.

All of the subjects participated in physical therapy while in the hospital and again as outpatients or in a rehabilitation center following the resumption of weight bearing, as previously described.

Outcome Assessment

All data were collected on the same day, during which all 31 subjects rotated through a series of testing stations. Physical testing consisted of forward bending, lifting, and ambulation. Self-report consisted of the SIP[6] and the Oswestry Low Back Pain Scale.[7]

Forward bending was selected as a routine movement that involves the pelvis and surrounding body segments. This measure was useful because individual joint measures were tedious and impractical due to time constraints. Lumbar ROM was assessed during forward bending using the inclinometer method, which is reputed to yield reliable measurements of flexion.[8] The spine was palpated to locate the spinous processes of T-12 and L-1, which were marked on the skin surface. The two-point contact base of the inclinometer was held in place to span the T-12 and L-1 spinous processes. The second goniometer was placed over the sacrum. Angular readings were taken from the inclinometers with the subject in a neutral standing position. The subject was then instructed to bend forward slowly and smoothly as far as possible. Upon reaching the end range, the inclinometer measurements were again recorded. The upper inclinometer reading represents the gross motion, and the sacral inclinometer measures the pelvic or hip motion. A measurement of lumbar motion is then derived by subtracting the pelvic motion from the gross motion and is expressed in degrees of flexion. This process was performed three times, and the mean was used to represent the measurement.[9]

Lifting was selected as a measure of gross strength because it is part of everyday activities and it involves muscle contractions across and adjacent to the pelvis. Lifting was evaluated using a static lift station to measure maximal lift capacity.[10] The station consisted of a dynamometer connected to a platform with a chain and handle. The subject was instructed to stand on the platform with toes level to the chain attachment. The handle was held while comfortably bending both knees (20[degrees]-40[degrees]) and keeping the trunk erect. The subject was then instructed to slowly build up tension, and pull up as hard as possible on the handle. The maximal effort from three trials was used for data analysis. Reliability of these measurements was not assessed.

Ambulation was chosen as a functional task required for many activities of daily living. In addition, the pelvis links the lower extremities to the upper body during this reciprocal forward translation. Ambulation was evaluated with a heel-strike analysis, during which data were collected on temporal-distance variables.[11] These variables were selected based on clinical observation that patients typically ambulate slowly with small, guarded steps following prolonged bed rest. This method of gait analysis provides a clinically feasible measurement that is reputed to be reliable when replicated with the same instrumentation 1 week apart.[12]

Markers were attached to each subject's heels. The subject then ambulated along a 6.1-mm (20-ft) paper walkway that was taped securely to the floor. A central portion of the walk was timed with a stopwatch for three consecutive unilateral heel-strikes. Data were collected for three trials, with approximately 2 minutes of rest between trials.

Distance measures were collected from the paper within the timed segments using a yardstick. Stride lengths represent the distance between ipsilateral heel-strikes. Gait speed was calculated as the total timed walking distance divided by the elapsed time of the segment. Mean stride lengths of each lower extremity and mean gait speed were calculated from the three trials of ambulation.

Self-reports were used to measure levels of disability and handicap. Survey instruments were selected to allow subjects to describe their functional outcomes. The Oswestry Low Back Pain Questionnaire[7] was selected to determine whether low back pain, often seen with pelvic fracture following bed rest or external fixation, was present. This questionnaire contains 10 sections designed to assess limitations of various activities of daily living (eg, walking, traveling, personal care). Each of the 10 sections contains six statements, each of which describes a greater degree of activity than the preceding statement.[7] The survey responses are scored with a percentage from 0% to 100%, with the percentage increasing as perceived disability increases. The Oswestry questionnaire provides reliable measurements for assessing disability as it relates to low back pain.[13]

The subjects also completed the SIP.[6] The SIP is a behaviorally based measure of health status developed to provide a measure of perceived well-being. The SIP consists of 136 statements about health-related dysfunction in 12 areas of activity: sleep and rest, body care and movement, eating, ambulation, mobility, emotional behavior, alertness behavior, social interaction, communication, home management, recreation, and work. The scores range from 0 to 100, increasing as the perceived disability increases. The SIP has been widely tested for reliability and validity across a variety of patient populations.[6]

Outcomes

Subjects in this multiple-subject case report were assigned to one of three separate groups: (1) subjects who were -50 years of age, (2) subjects who were >50 years of age, and (3) subjects with CHI (not included in either of the other two groups). The mean value and standard deviation for each variable as well as the minimum and maximum scores are presented. Results are then compared with published normal values[10,14-16] for all physical performance measures. The normal data are provided simply as a reference in this multiple-subject case report (Tab. 3).

[TABULAR DATA 3 OMITTED]

The SIP and the Oswestry questionnaire were scored as described by their accompanying literature.[6,7] These scores were then compared with the scales of disability provided with each survey instrument. The SIP data also are provided in groups defined by Tile's pelvic-fracture classification[2] (Tab. 4).

[TABULAR DATA 4 OMITTED]

The results of the SIP phone survey that was conducted with the 17 patients who could not attend the examination were compared with the SIP scores of the subjects who participated in the examination. Student's t test for continuous variables and Fisher's Exact chi-square test for dichotomous variables were used to compare the groups' scores.

All three subject groups scored below the means for lumbar ROM, maximum lift, stride length, and walking speed obtained in previous studies of subjects without known impairments or disabilities (Tab. 3). The mean value for lumbar ROM of the subjects most closely approximated the normal values. Data for gait speed appeared to be most limited when compared with data obtained from subjects without injuries. Subjects had a tendency to walk at a much slower rate, approximately 40 to 60 cm/s (30%-45%) less, than average. The stride lengths, however, were less limited, at approximately 5 to 40 cm (5%-25%) shorter than average. Lifting scores varied widely for all three groups, with the mean value falling between the normal values for women and men.

The [less than or equal to]50-year-old group had the highest mean scores for all measures of physical performance except the lifting test, in which the CHI group scored the highest. The >50-year-old group typically had the lowest mean scores.

Overall, the subjects' scores were lower than those of subjects without injury. Some individual data from each group, however, fell within normal ranges for each measure, with the exception of gait speed. Only four subjects, who were in the [less than or equal to]50-year-old age group, had gait speed scores that were within normal ranges. No subjects from the other two groups had gait speed scores that were within normal ranges (Tab. 5). The self-report survey data for individual subjects are presented in Table 6.

[TABULAR DATA 5 & 6 OMITTED]

The average SIP scores for the subjects were as follows: total SIP score = 9.34[+ or -]7.47, physical dimension=7.79[+ or -]6.93, and psychosocial dimension= 8.24[+ or -]9.61. By group, the SIP scores were as follows: (1) for the [less than or equal to]50-year-old group, total SIP score= 10.6[+ or -]11.6, physical dimension =7.61[+ or -]9.6, and psychosocial dimension=10.2[+ or -]16.3, (2) for the >50-year-old group, total SIP score=11.2[+ or -]9.6, physical dimension =10.9[+ or -]9.6, and psychosocial dimension=7.7[+ or -]9.6, and (3) for the CHI group, total SIP score=4.4[+ or -]6.5, physical dimension =4.7[+ or -]7.5, and psychosocial dimension= 3.0[+ or -]5.1. The average scores in all of these areas represent minimal levels of disability according to the literature.6 The CHI group's mean scores were approximately 40% lower than the mean scores of the other two groups, with smaller standard deviations.

The SIP scores were also examined in relation to the Tile classification of pelvic fractures[2] (Tab. 4). The subjects with B1-class ("open-book") pelvic fractures had a tendency to score higher across individual SIP categories. The B1 and B2 groups typically had the lowest scores, with both groups having similar associated injuries.

At the time of the examinations, 76% of those subjects who were employed prior to the injury had returned to their previous employment. Of this group, 62% returned to full-time employment and 14% returned with job modifications. In addition, 7 subjects were attending school prior to injury. Of this group, 6 subjects returned to school full-time and I subject returned to school part-time.

The average Oswestry score for all subjects was 13.26%[+ or -]15.41%. The scores by group were as follows: (1) [less than or equal to]50-year-old group = 17.3%[+ or -]16.4%, (2) >50-year=old group = 23.1%[+ or -]17.3%, and (3) CHI group= 14.7% [+ or -] 20.7%. All mean scores reflect minimal disability according to the literature[6,7] (Tab. 7). The SIP comparisons revealed no differences (P<.05) between the subjects who were examined and the 17 subjects who did not attend the examination.

Table 7.

Interpretation of Oswestry Disability Scores(a)

Disability Score                 Interpretation
0%-20%--minimal                  This group can cope with most
living
disability                       activities. Usually no treatment
is
                                 indicated, apart from advice on
                                 lifting, sitting posture, physical
                                 fitness, and diet. In this group,
                                 some patients have particular
                                 difficulty with sitting, and this
may
                                 be important if their occupation
is
                                 sedentary (eg, a typist or lorry
                                 driver).
20%-40%--moderate                This group experiences more pain
disability                       and problems with sitting,
lifting,
                                 and standing. Travel and social
                                 life are more difficult, and the
                                 patients may well be off work.
                                 Travel, sexual activity, and
                                 sleeping are not grossly affected,
                                 and the back condition can usually
                                 be managed by conservative
                                 means.
40%-60%--severe                  Pain remains the main problem in
disability                       this group of patients, but
travel,
                                 personal care, social life, sexual
                                 activity, and sleep are also
                                 affected. These patients require
                                 detailed investigation.
60%-80%--crippled                Back pain impinges on all aspects
of
                                 these patients' lives, both at
home
                                 and at work, and positive
                                 intervention is required.
80%-100%                         These patients are either bed
bound
                                 or exaggerating their symptoms.
                                 This can be evaluated by careful,
                                 observation of the patient during
                                 the medical examination.

Discussion

There is an absence of literature that documents the outcomes of patients with multiple traumas following ORIF of unstable pelvic-ring fractures. This case report represents the only published work of this nature on this patient population. The record of this group's performance can be used as a basis of comparison for treatment and research involving patients with. multiple traumas and pelvic-ring fractures.

The physical testing stations that we used were selected as a practical means of examining routine activities. Decreased gait speed appeared to reflect the most involved physical disability. This finding may be related to lingering impairments in ROM or strength, or it may be present due to fear of falling, which was not measured in our subjects.

Although physical performance is important, perhaps even more crucial is the subject's perception of his or her functioning. The >50-year-old group had the greatest mean score for the physical dimension of the SIP, whereas the [less than or equal to]50-year-old group had the highest mean psychosocial dimension score. Perhaps the older group had more difficulty with physical recovery because of age, whereas the younger group may have had more difficulty with coping and socialization following the trauma. The CHI group had the lowest mean scores across all areas; however, two of the seven subjects in this group scored within the upper third. Nonetheless, the average group scores in all of these areas represent levels of mild disability. The scores provide examples of how individual patients with multiple traumas and pelvic fractures interpret disability and handicap I year postoperatively.

With regard to fracture classification, it may be useful to consider recovery in terms of initial severity of pelvic injury. The tendency for patients with B1-class fractures to score higher on SIP categories may reflect greater difficulty with recovery at approximately 1 year following this type of pelvic displacement. As a result, the physical therapist may anticipate greater involvement with patients having this type of fracture.

The prevalence of back pain was also of interest in these patients because of the way pelvic fractures are usually managed. Tile[2] examined 218 patients following pelvic-ring fractures managed by bed rest or external fixation. The report indicated that stable injuries resulted in few major long-term problems. In contrast, patients with vertically unstable fractures had many problems, with 60% of the patients continuing to report considerable pain, typically along the posterior sacroiliac region or lower lumbar spine.[2] Unlike the patients in Tile's study, our subjects were managed by ORIF, which provided anatomical reduction to within 2 mm. The average Oswestry score for all subjects in our report was 13.26%[+ or -]15.41%, which represents minimal disability. Complaints of back pain do not appear to occur often in these patients.

We believe that early initiation of physical therapy with these patients following surgery is an important factor in determining placement following discharge. The transfer training is very challenging because of increased pain and associated injuries. If patients and therapists can overcome obstacles, discharge to the home can be accomplished. We contend that the importance of continuing with a home exercise program should be emphasized because these patients will be nonambulatory for 3 months. Although our subjects reported following through with the home exercise program, we cannot be certain how closely they adhered to our recommendations.

Additional information about the subjects in our case report would have been useful. A comparison of exactly what type and amount of follow-up was received by each subject would have been helpful. Additionally, a record of each subject's prior level of function was not available because of the unexpected nature of the injury. In response to this need, further data collection has been initiated on a new group of patients with traumatic injuries, which will include completion of the SIP by the patient or a family member (if the patient is cognitively impaired) to determine function prior to the trauma. Other factors such as age and motivation also may affect recovery.

The subjects in our case report displayed varying degrees of impairment, disability, and handicap following ORIF of pelvic fractures. The limitations in most cases were mild or minimal. As a result, the subjects' recovery appeared remarkable, considering the severity and extent of the initial injuries. Although individuals sustaining multiple traumas may still face functional limitations, after I year the subjects in this report have regained much of the function that was lost immediately following injury.

Acknowledgments

We express our gratitude to Charles Olsen, PT, Linda Palmer, PT, Ellyce Johnson, Rebecca Gruen, RN, JD, and John Hakim, JD, for their time and effort toward helping with the data collection. The normal data for maximal lift capacity provided by Thomas Rudy, PhD, is also appreciated.

References

[1] Gruen GS, Mears CM. Surgery of the Musculoskeletal System. 2nd ed. New York, NY: Churchill Livingstone Inc; 1990:2493-2515.

[2] Tile M. Pelvic ring fractures: Should they be fixed? J Bone Joint Surg

[Br]. 1988;70:1-12.

[3 ] International Classification of Impairments, Disabilities, and Handicaps. Geneva, Switzerland: World Health Organization, 1980.

[4] Jette AM. Diagnosis and classification by physical therapists: a special communication. Phys Ther. 1989;69:967-969.

[5] Guccione A. Physical therapy diagnosis and the relationship between impairments and function. Phys Ther. 1991;71:499-504.

[6] Bergner M, Bobbitt RA, Carter WB, Gilson BS. The sickness impact profile: development and final revision of a health status measure. Med Care. 1981;19:787-805.

[7] Fairbank JC, Couper C, Davies JB, O'Brien JP. The Oswestry low back pain questionnaire. Psychotherapy. 1980;66:271-273.

[8] MacKenzie EJ, Burgess AR, McAndrew MC, et al. Patient-oriented functional outcome after unilateral lower extremity fracture. J Orthop Trauma. 1993;7:393-401.

[9] Mayer TG, Tencer AF, Kristoferson S, Mooney V. Use of noninvasive techniques for quantification of spinal range-of-motion in normal subjects and chronic low-back dysfunction patients. Spine. 1984;9:588-595.

[10] Rudy TE, Liber S, Turk DC. Development of a functional capacity protocol for chronic back pain subjects. Clin J Pain. 1991;7:62.

[11] Shores M. Footprint analysis in gait documentation. Phys Ther. 1980;60:1163-1167.

[12] Boenig D. Evaluation of a clinical method of gait analysis. Phys Ther. 1977;57:795-798.

[13] Delitto A, Cibulka CT, Erhard R.E, et al. Evidence for use of an extension-mobilization category in acute low back syndrome: a prescriptive validation pilot study. Phys Ther. 1993;73:216-228.

[14] Murray MP. Gait as a total pattern of movement. Am J Phys Med. 1967;46:290-333.

[15] Murray MP, Kory R, Sepic SB. Walking patterns of normal women. Arch Phys Med Rehabil. 1970;51:637-650.

[16] Dillard J, Trafimow J, Andersson J, Cronin K. Motion of the lumbar spine: reliability of two measurement techniques. Spine. 1991;16:321-324.

RM Hakim, PT, is Student Program Coordinator for Northeast Area 18 and Rehabilitation Team Leader, Nova Care Inc, 321 Spruce St, Scranton, PA 18503 (USA) (RMH [e-mail]). Address all correspondence to Ms Hakim.

GS Gruen, MD, is Assistant Professor, Department of Orthopaedic Surgery, and Chief, Division of Orthopaedic Traumatology, University of Pittsburgh Medical Center, Musculoskeletal Institute, Kaufmann Bldg, Suite 1010, 3471 Fifth Ave, Pittsburgh, PA 15261.

A Delitto, PhD, PT, is Chairman, Department of Physical Therapy, School of Health and Rehabilitation Sciences, University of Pittsburgh, 101 Pennsylvania Hall, Pittsburgh, PA, 15261.

This work was completed by Ms Hakim as an independent study in partial fulfillment of her, Master of Science degree in neurologic physical therapy, Department of Physical Therapy, University of Pittsburgh.