Background: The purpose of this study was to compare manual traction and fracture-table traction for the reduction and nailing of femoral shaft fractures. We evaluated the quality of the reduction, operative time, complications, and functional status of the patient.
Methods: Eighty-seven consecutive adult patients with a unilateral fracture of the femoral diaphysis that did not extend into the knee joint or proximal to the lesser trochanter were enrolled in the study. Patients who were transferred to our institution more than forty-eight hours after injury; those with multiple-system injuries, injury to the ipsilateral lower extremity, or pathological fracture; and those who were unable or unwilling to provide consent or to return for follow-up were excluded. Forty-five patients were randomized to manual traction and forty-two, to fracture-table traction; all were treated in the supine position. The number of surgical assistants, operative and fluoroscopy time, complications, functional scores, and other outcomes were recorded.
Results: There were no significant differences between the groups with respect to age, gender, Glasgow Coma Score, Injury Severity Score, side or mechanism of injury, fracture type, or time from injury to treatment. Internal malrotation was significantly more common when the fracture table had been used: twelve (29%) of the forty-two femora were internally rotated by >10° compared with three (7%) of the forty-five treated with manual traction (p = 0.007). Total operative time, from the beginning of the patient positioning to the completion of the skin closure, was decreased from a mean of 139 minutes (range, 100 to 212 minutes) when the fracture table was used to a mean of 119 minutes (range, sixty-five to 180 minutes) when manual traction was used (p = 0.033). There was no significant difference between the two treatment groups with regard to the number of assistants per case (mean two; range, zero to three), fluoroscopy time, other complications including femoral shortening or lengthening, or functional status of the patient at one year.
Conclusions: Compared with fracture-table traction with the patient in a supine position, manual traction for intramedullary nailing of isolated fractures of the femoral shaft is an effective technique that decreases operative time and improves the quality of the reduction.
David J.G. Stephen, MD, BSc, FRCS(C); Hans J. Kreder, MD, MPH, FRCS(C); Lisa B. Conlan; Division of Orthopaedic Surgery (D.J.G.S., H.J.K., and L.B.C.) and Department of Health Policy Management and Evaluation (H.J.K.), University of Toronto, Sunnybrook and Women's College Health Sciences Centre, MG 365, 2075 Bayview Avenue, Toronto, ON M4N 3M5, Canada. E-mail address for H.J. Kreder: email@example.com
Emil H. Schemitsch, MD, FRCS(C); Michael D. McKee, MD, FRCS(C); Lisa Wild, RN; Department of Orthopaedic Surgery, St. Michael's Hospital, 30 Bond Street Toronto, ON M5B 1W8, Canada
Intramedullary nailing with reaming of the bone and static locking of the nail is the treatment of choice for fractures of the femoral shaft in adults 1-5. Excellent results have been reported after femoral nailing with the patient on a fracture table 2,6. The proponents of this technique mention the quality of the reduction and the physician's ability to perform the procedure either alone or with only one assistant. Conversely, the use of a fracture table can limit access to the patient for the treatment of other injuries, especially when the patient has multiple injuries. Additional time is necessary to transfer such patients from one table to another and for repeat preparation and draping. Traction with use of a fracture table has also been associated with pudendal nerve palsies, well-leg compartment syndromes, and skin sloughs of the perineum 2,7.
Several authors have reported on the use of a femoral distractor with the patient on a radiolucent table, as an alternative to a fracture table, for intramedullary nailing of the femoral shaft 8,9. In a prospective, randomized study comparing use of a femoral distractor with that of a fracture table, McFerran and Johnson found that, although use of the femoral distractor was safe and provided comparable results, it was technically demanding and time-consuming so the operative time was actually longer than it was when the fracture table was used 9. Karpos et al. performed a retrospective study comparing manual traction, without a femoral distractor, with fracture-table traction for intramedullary nailing of femoral shaft fractures 10. They noted a significant decrease in operating time (p < 0.05), with no difference in the quality of the reduction, when manual traction had been used. There is still controversy regarding the accuracy of reduction, the exact time-savings, and the need for additional scrubbed assistants when manual traction is used for intramedullary nailing of femoral shaft fractures 11.
The purpose of this study was to compare traction with use of a fracture table with manual traction for intramedullary nailing of femoral shaft fractures. We evaluated the quality of the reduction (particularly rotation and length), operative time, complications, and functional status of the patient.
Materials and Methods
Two hundred and eighty adult patients with a femoral fracture were screened for possible entry into the study at Sunnybrook and Women's College Health Sciences Centre and St. Michael's Hospital, Toronto, Ontario, between March 1997 and September 2000. One hundred and ninety-three patients were excluded, for a variety of reasons: forty-five had a proximal or distal metaphyseal femoral fracture (AO 12 type 31 or 33), forty-five had an ipsilateral or contralateral tibial or femoral fracture, fifty-eight had multiple-system injuries (including two with vascular injuries, six with unstable pelvic fractures, and two with unstable spinal fractures), thirteen had transferred from an outside institution after forty-eight hours, eighteen refused or were unable to consent to participation in the study, six had a pathological fracture, five were eligible but were missed during the randomization process, and three had traveled to the hospital from outside of the country, where follow-up was thought to be impossible. The remaining eighty-seven patients, all with an AO type-32 fracture, consented to randomization for this study. All had complete follow-up at six months. Eight patients (five treated with manual traction and three, with table traction) could not be located or refused to return for the one-year follow-up evaluation.
Our internal review board approved the trial before it was begun. A sample size calculation suggested that ninety patients would be required to provide 80% power to detect a 10° difference in malrotation (calculated by subtracting the femoral anteversion on the injured side from that on the normal side), as the primary outcome, between the two treatment groups. This sample size also provided 80% power to detect a twenty-minute difference in operative time, as a main secondary outcome. Sequentially numbered sealed envelopes were opened just prior to surgery in order to randomly assign patients to receive either manual traction or table traction. The randomization sequence was computer generated with use of a random-number generator.
Forty-five patients were randomized to manual traction; twenty-six (58%) were treated with a Russell-Taylor nail (Smith and Nephew, Memphis, Tennessee), and the remaining nineteen (42%) were treated with an ASIF nail (Synthes, Paoli, Pennsylvania). Forty-two patients were randomized to traction on the fracture table; twenty-six (62%) were treated with a Russell-Taylor nail, and the remaining sixteen (38%) were treated with an ASIF nail. All nails were inserted antegrade, and all were statically locked; two distal screws and one proximal oblique screw were used for the Russell-Taylor nail, and two transverse proximal screws were used for the ASIF nail. Preoperative skeletal traction was not used for any patient.
Manual traction: In the manual-traction group, patients were positioned supine on a radiolucent extension of the operating table, as described by Karpos et al. 10, to allow visualization of the entire affected limb with an image intensifier brought in from the opposite side. A rolled sheet was placed under the sacrum and the lumbar spine as well as under the ipsilateral shoulder to elevate the affected hip by approximately 30°. The entire limb was then draped free and included in the sterile field. During passage of the reamer guide-wire through the entry portal in the piriformis fossa, an assistant reduced the fracture using manipulation and longitudinal traction, with the patient fully paralyzed under anesthesia. In approximately one-quarter of the patients, it was necessary to control the proximal fragment with insertion of a small-diameter nail for use as an intramedullary joystick to assist in the reduction. Open reduction was not necessary in this series. An assistant maintained fracture reduction throughout the reaming and nail placement. Prior to insertion of locking screws, the length and rotation of the femur were assessed clinically and radiographically. Radiographically, the lengths of the normal and injured femora were measured with a ruler (from the tip of the greater trochanter to the margin of the knee joint), with care taken to position the two lower limbs similarly, and were compared. Rotation was compared between the injured and normal sides by reproducing the relationship between the lesser trochanter and the patellar projection over the femoral condyles on the two sides, as described by Krettek et al. 13.
Fracture-table traction: The patients were positioned supine on the fracture table with the contralateral lower limb flexed and externally rotated on a well-leg holder to facilitate fluoroscopic imaging. The foot of the affected limb was securely fastened into the traction boot on the fracture table and externally rotated approximately 10°. The image intensifier was then brought in from the side opposite the fracture, with the position checked to ensure that two orthogonal views of the fractured femur could be obtained. The reduction was obtained before preparation and draping. Length was determined by aligning the cortices of all intercalary fracture fragments. Rotation was adjusted by aligning the cortical diameters of adjacent fracture fragments in AO type-A and B fracture patterns and by evaluating the amount of anteversion of the femoral neck seen radiographically in relationship to the femoral condyles in type-C fractures 13. The amount and duration of the traction was left to the discretion of the operating surgeon. The entire thigh, including the hip and knee regions, was prepared and draped to allow full access to the affected limb.
After surgery and before reversal of the anesthesia, the rotation and length were assessed clinically in both study groups. No patient underwent revision because of malalignment identified as a result of this assessment.
The circulating nurse documented, on the basis of direct observation, the time that it took to position the patient, insert the nail, lock the nail, and close the wound and recorded the information on the patient's study form, which was then transmitted to the research coordinator. Estimated blood loss was transcribed from the operative record. Prior to discharge from the hospital, a computed tomography scan of both femora was made along with plain radiographs in order to measure the quality of the reduction, including rotation and length, in comparison with that on the contralateral side. Positioning for the computed tomography and measurement of the scans were performed according to the method described by Tornetta et al. 14. Interobserver and intraobserver consistency of computed tomography measurement of femoral anteversion is high (Spearman rho > 0.96) 15. An independent bone radiologist made all measurements without knowledge of the treatment group. An independent research assistant evaluated patients at six weeks, three months, six months, and one year. Evaluation at these times included measurement of range of motion and strength, assessment of bone-healing, determination of functional status with the Short Form-36 16 and Musculoskeletal Function Assessment Instrument 17-19, and documentation of ongoing treatment and complications.
Patient demographics, injury characteristics, and details of treatment were compared between the two study groups. The chi-square test was used to compare categorical outcomes between the treatment groups, and the Student t test was used to compare continuous outcomes.
There were no significant differences between the demographic characteristics of the two treatment groups ( Table I ).
A deep infection developed at the site of one grade-III 20 open fracture that had been treated with manual traction; the infection resolved after serial débridement. Three exchange nailing procedures (two in the manual-traction group and one in the table-traction group) were performed because of nonunion after six months. At the time of the exchange nailing in the patient in the table-traction group, derotation from a position of 35° of internal malrotation was performed to match the alignment of the contralateral femur. There were no pudendal nerve palsies or compartment syndromes.
Large deviations from normal rotation, as measured with computed tomography scanning of the fractured and intact femora, were associated with table traction ( Fig. 1 ). The mean amount of malrotation in the manual-traction group did not differ significantly from zero (p = 0.547), whereas the table-traction group had excessive internal malrotation that differed significantly from zero (p = 0.022). The percentage of patients with internal malrotation of >10° was significantly higher (p = 0.007) in the fracture-table group (29%; twelve of forty-two) than it was in the manual-traction group (7%; three of forty-five) ( Table II ). Although estimated blood loss was significantly higher in the fracture-table group (p = 0.004), the observed mean difference of 159 mL is not clinically important. Angular deformity was within 5° of normal in all patients.
Manual traction resulted in a significant savings of operative time. Positioning of the patient (from the beginning of the positioning to the incision) and locking of the nail (from the time that the nail was fully inserted to the completion of screw insertion) were both significantly faster, by an average of seven minutes, with manual traction (p = 0.047) ( Table II ). Total operative time, defined as the time from the beginning of patient positioning to skin closure, was also shorter, by an average of twenty minutes (119 minutes compared with 139 minutes), with manual traction (p = 0.033). Fluoroscopic time was similar between the two groups ( Table II ).
More than 80% of all cases were performed with no more than two scrubbed assistants. There was no difference in the number or experience of the scrubbed assistants between the two groups.
There was no significant difference in functional status between the two groups at six months or one year. A secondary analysis demonstrated a trend for worse physical function (Short Form-36 physical component summary score) with external rotation of >10° (p = 0.03); however, the result did not remain significant after correction for multiple comparisons (p > 0.05).
After stratifying by AO type, we noted trends similar to those observed in the overall analysis. Because of the small numbers of patients in the stratified groups, only some of these trends reached significance. Operative times were significantly shorter (p = 0.008) and restoration of limb length was more likely to be correct (p = 0.012) for AO type-A and B fractures treated with manual traction. Internal malrotation was less common for AO type-C fractures treated with manual traction (p = 0.036).
It has been well documented that intramedullary nailing with reaming and static locking of the nail for treatment of femoral shaft fractures results in a high rate of union and a low complication rate 1-5. Seligson argued that the role of manual traction in femoral nailing would be clarified only by a prospective, controlled study with careful documentation of alignment 11. To our knowledge, our study is the first prospective, randomized clinical trial comparing the use of manual traction with that of fracture-table traction for intramedullary nailing of isolated femoral shaft fractures. Using computed tomography scanning to provide accurate measurement of femoral rotation and length, we noted that manual traction resulted in equal or superior fracture reduction, operative time, fluoroscopy time, and functional outcome.
Femoral fractures in patients with polytrauma or multiple extremity injuries were excluded from this study to minimize bias due to the potentially unequal distribution of variables that might confound the primary or secondary outcomes. Consequently, the estimated twenty-minute time-savings with manual traction is probably an underestimation of the potential time that could be saved in a polytrauma setting, in which multiple injuries can be managed with use of the same table without repositioning or redraping. In a prospective, randomized trial, McKee et al. noted a time-savings of more than thirty minutes for polytrauma patients when the tibial fractures were nailed with use of manual instead of table traction 21. In retrospective studies, Karpos et al. reported a savings of thirty-three minutes 10 and Wolinsky et al. reported a savings of seventeen minutes when manual traction was used instead of table traction for nailing of femoral fractures 22.
Wolinsky et al. reported that there was a learning curve for femoral nailing, with or without a fracture table 22. We think that surgeons who are proficient in the technique of intramedullary nailing with the patient on a fracture table can readily master the use of manual traction. Both of our study centers are level-one trauma centers, and a large volume of femoral shaft fractures were treated by the surgeons involved in this study. At one of the study centers the technique of manual traction had been in routine use for approximately five years, while at the second site the technique was introduced for the purpose of this study. There was no significant difference between the two centers with regard to the operative time in either the manual-traction or the fracture-table group.
Despite randomization, there were more AO type-C fractures in the table-traction group ( Table I ). Comparison of operative time and alignment between the two treatment groups after stratification by AO type produced statistical trends that were similar to those demonstrated by the overall comparison.
The internal malrotation seen after use of the fracture table may have resulted from the practice of internally rotating the limb to facilitate access to the starting point and to improve visualization of the hip in the lateral plane. Furthermore, once the limb is positioned on the fracture table and the procedure is started, it can be difficult to assess rotation clinically because the limb cannot be moved readily. With manual traction, the entire injured limb is available and can be assessed clinically. Since both limbs can be imaged with fluoroscopy on the radiolucent table, the injured femur can be rotated to match the alignment of the normal limb by comparing the lesser trochanteric and patellar positions of the two limbs 13. With the technique of fracture-table traction used in this study (with the normal hip flexed and abducted), it was not possible to compare the rotatory alignments of the normal and injured femora intraoperatively. It is possible that the observed rotational malalignment of the femora treated on the fracture table might have been reduced by adjusting rotation on the injured side to match that of the normal limb before securing both limbs to the fracture table. However, this process would probably have resulted in even longer operative time for the table-traction group. Femoral anteversion can be assessed intraoperatively by evaluating the angle of neck anteversion relative to the posterior femoral condyles 13 with both manual and table traction.
One of the theoretical advantages of using a fracture table for femoral nailing is that the procedure can be done with one or even without a scrubbed assistant. In this study, there was no difference in the average number of scrubbed assistants (two) between the two groups. This may be a reflection of the fact that both study centers are teaching institutions with fellows and residents available for assistance. Manual traction on a radiolucent table may have limitations for the community orthopaedic surgeon with fewer available assistants.
Our prospective study suggests that most isolated fractures of the femoral shaft can be treated successfully with closed intramedullary nailing with use of either intraoperative manual traction or traction with a fracture table. Potential advantages of the manual-traction technique include decreased operative time and less rotational deformity. Care must be taken intraoperatively to ensure restoration of femoral length and rotation regardless of which technique is used.
Investigation performed at the University of Toronto Orthopaedic Trauma Research Group, Toronto, Ontario, Canada
In support of their research or preparation of this manuscript, one or more of the authors received grants or outside funding from the Canadian Orthopaedic Foundation and Smith and Nephew Richards. None of the authors received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.
1. Bone LB, Johnson KD, Weigelt J, Scheinberg R. Early versus delayed stabilization of femoral fractures. A prospective randomized study. J Bone Joint Surg Am
, 1989;71: 336-40.
2. Browner BD. The science and practice of intramedullary nailing. 2nd ed. Baltimore: Williams and Wilkins; 1996.
3. Winquist RA, Hansen ST Jr. Comminuted fractures of the femoral shaft treated by intramedullary nailing. Orthop Clin North Am
, 1980;11: 633-48.
4. O'Brien PJ, Meek RN, Powell JN, Blachut PA. Primary intramedullary nailing of open femoral shaft fractures. J Trauma, 1991;31: 113-6.
5. Riska EB, von Bonsdorff H, Hakkinen S, Jaroma H, Kiviluoto O, Paavilainen T. Prevention of fat embolism by early internal fixation of fractures in patients with multiple injuries. Injury
, 1976;8: 110-6.
6. Küntscher G. Practice of intramedullary nailing. Rinne HH, translator. Springfield, IL: CC Thomas; 1967.
7. Brumback RJ, Ellison TS, Molligan H, Molligan DJ, Mahaffey S, Schmidhauser C. Pudendal nerve palsy complicating intramedullary nailing of the femur. J Bone Joint Surg Am
, 1992;74: 1450-5.
8. Baumgaertel F, Dahlen C, Stiletto R, Gotzen L. Technique of using the AO-femoral distractor for femoral intramedullary nailing. J Orthop Trauma
, 1994;8: 315-21.
9. McFerran MA, Johnson KD. Intramedullary nailing of acute femoral shaft fractures without a fracture table: technique of using a femoral distractor. J Orthop Trauma
, 1992;6: 271-8.
10. Karpos PA, McFerran MA, Johnson KD. Intramedullary nailing of acute femoral shaft fractures using manual traction without a fracture table. J Orthop Trauma
, 1995;9: 57-62.
11. Seligson D. Invited commentary. Length of operative procedures: reamed femoral intramedullary nailing performed with and without a fracture table. J Orthop Trauma
, 1998;12: 495.
12. Müller ME. The principle of classification. In: Müller ME, Allgöwer M, Schneider R, Willenegger H, editors. Manual of internal fixation: techniques recommended by the AO-ASIF Group. 3rd ed. New York: Springer; 1991. p 118.
13. Krettek C, Miclau T, Grun O, Schandelmaier P, Tscherne H. Intraoperative control of axes, rotation and length in femoral and tibial fractures. Technical note. Injury
, 1998;29 Suppl 3: 29-39.
14. Tornetta P 3rd, Ritz G, Kantor A. Femoral torsion after interlocked nailing of unstable femoral fractures. J Trauma
, 1995;38: 213-9.
15. Tomczak RJ, Guenther KP, Rieber A, Mergo P, Ros PR, Brambs H-J. MR imaging measurements of the femoral antetorsion angle as a new technique: comparison with CT in children and adults. AJR Am J Roentgenol
, 1997;168: 791-4.
16. Hopman WM, Towheed T, Anastassiades T, Tenenhouse A, Poliquin S, Berger C, Joseph L, Brown JP, Murray TM, Adachi JD, Hanley DA, Papadimitropoulos E. Canadian normative data for the SF-36 health survey. Canadian Multicentre Osteoporosis Study Research Group. CMAJ
, 2000;163: 265-71.
17. Engelberg R, Martin DP, Agel J, Obremsky W, Coronado G, Swiontkowski MF. Musculoskeletal Function Assessment instrument: criterion and construct validity. J Orthop Res
, 1996;14: 182-92.
18. Martin DP, Engelberg R, Agel J, Swiontkowski MF. Comparison of the Musculoskeletal Function Assessment questionnaire with the Short Form-36, the Western Ontario and McMaster Universities Osteoarthritis Index, and the Sickness Impact Profile health-status measures. J Bone Joint Surg Am, 1997;79: 1323-35.
19. Martin DP, Engelberg R, Agel J, Snapp D, Swiontkowski MF. Develop-ment of a musculoskeletal extremity health status instrument: the Musculoskeletal Function Assessment instrument. J Orthop Res
, 1976;58: 453-58.
20. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones, Retrospective and prospective analyses.. J Bone Joint Surg Am
, 1997;79: 1323-35.
21. McKee MD, Schemitsch EH, Waddell JP, Yoo D. A prospective, randomized clinical trial comparing tibial nailing using fracture table traction versus manual traction. J Orthop Trauma
, 1999;13: 463-9.
22. Wolinsky PR, McCarty EC, Shyr Y, Johnson KD. Length of operative procedures: reamed femoral intramedullary nailing performed with and without a fracture table. J Orthop Trauma
, 1974;2: 81-4.
23. Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical scale. Lancet, 1997;79: 1323-35.
24. Baker SP, O'Neil B. The injury severity score: an update. J Trauma, 1976;16: 882-5.Copyright 2002 by The Journal of Bone and Joint Surgery, Incorporated