Femur fracture nonunion with leg length discrepancy: a treatment strategy for difficult post-traumatic problem. *Updated with New Images

This month’s case features a 40-year old healthy man who, unfortunately, was struck by a vehicle 13 months earlier.  He sustained a closed femoral shaft fracture in the distal metadiaphyseal region.  He was treated at an outside facility with open reduction of the fracture followed by antegrade trochanteric nailing.  Suboptimal placement of the initial distal interlocking bolts was not appreciated until 5 weeks postoperatively, at which time the bolts were revised and placed accurately through the distal interlocking holes in the nail.  

Approximately one year from the date of his injury, he presented to Harborview Medical Center for an opinion regarding his ongoing pain and limitations.  His complaints include significant weightbearing and activity-related pain located to the distal thigh and anterior knee area.  While this pain was a predominant symptom, his most frustrating problem was the perception that his left leg was shorter than his right.  He is otherwise healthy and was previously an avid athlete, particularly involved in long distance running and bicycle riding.  He is a nonsmoker, and takes no medications on a regular schedule.  He has been unable to wean himself off of a walking aid because of continuing lower thigh discomfort and the inability to correct his “limp”.  He has been unable to participate in any of his running or bicycling activities.  The physical examination demonstrates well-healed surgical scars at the left hip and distal thigh regions.  The quadriceps musculature of the left thigh is atrophic.  Clinically, there is an approximate 2cm leg length discrepancy with the left leg shorter than the right.  There are no significant rotational abnormalities.  The passive hip and knee ranges of motion are symmetric to the contralateral side.  The pedal pulses were palpable, and the patient’s motor and sensory examination was normal, within the limitations of pain.  Initial bloodwork demonstrated a number of normal values, including the white cell blood count, ESR, C-reactive protein, and bone metabolism profile.

The presenting anteroposterior plain radiographs of the left femur demonstrate a well-aligned hypertrophic metadiaphyseal nonunion.  A trochanteric antegrade femoral nail is identified, and the single distal interlocking bolt is noted to be broken.  The long leg standing radiographs confirm an approximate 2cm leg length discrepancy with the left femur shorter than the right.  The operative records from the index procedure confirm the size, length, and manufacturer of the medullary implant, and also indicate that the femur fracture was reduced using an open technique prior to the placement of the nail.

The diagnosis is a well aligned, but clinically shortened, hypertrophic, aseptic distal metadiaphyseal femoral nonunion. 

The patient was extensively counseled regarding the options regarding the treatment of this problem.  Typically, the well vascularized hypertrophic nonunion develops because of fracture site instability.  Enhancing the stability of the fracture/nonunion site with a traditional compression plating technique typically results in predictable and rapid healing of the nonunion site.  Unfortunately, this strategy, by itself, does not correct the significant and clinically bothersome leg length discrepancy.  Alternatives include contralateral shortening osteotomy of the femur to equalize leg lengths, ipsilateral oblique osteotomies to regain length while still compressing the nonunion, or lengthening through the nonunion, either acutely or gradually.  Each of these options comes with risks related to healing and increased surgical complexity.  Given the patient’s desire to restore his anatomy as accurately as possible, we decided to treat this problem with acute lengthening through the nonunion site followed by stabilization with a locked reamed medullary nail with plate augmentation.   

The patient was medically stable and cleared for surgery.  A general anesthetic was administered and he was positioned supine on a radiolucent operating table.  The patient was supported with a soft left-sided lumbo-sacral support.  The entire left lower extremity was then included in the sterile operating field, and the preoperative patient/procedure verification was completed.  Antibiotic prophylaxis was withheld until deep cultures of the nonunion site were obtained.
The patient’s pre-existing trochanteric femoral nail was removed percutaneously.  Through the same skin incision, a piriformis fossa start point was obtained and an entry hole created into the proximal femur in the appropriate trajectory.  A curved ball-tipped guide wire was then placed down the femur, across the nonunion site and into the distal femur beyond the endpoint of the previously placed nail.  The canal was enlarged with sequential reaming to a total of 13mm.  The anticipated nail length was estimated using a two-guide rod technique taking into consideration the additional 2 cm of femoral lengthening.  Multiple specimens of fibrous tissue and reamings from the endosteal surface of the femur were sent for culture and sensitivity.  Subsequent to this, prophylactic antibiotics were administered.
The curved ball-tip guide rod was then retracted into the proximal femoral segment and the nonunion was then approached with a lateral extensile exposure to the femur.  After elevation of the vastus lateralis, the nonunion itself was exposed using an osteo-periosteal technique as described by Judet and Judet.  Two 2.4mm Steinmann pins were placed into the anterolateral surface of the femur on either side of the nonunion.  The distance between these two pins was measured and recorded.  Peripheral to these pins, two bicortical 5mm Schanz pins were placed and attached to a large Universal Distractor.

The nonunion site was mobilized then distracted 2cm and confirmed accurate by repeat measurement using the 2.4mm Steinmann pins.

The nonunion site was then stabilized with a posterolateral locking plate using screws strategically placed to be out of the way of the anticipated nail pathway.

The 2.4mm Steinmann pins, the 5mm Schanz pins, and the Universal Distractor were then removed.  A straight insertion rod was then placed through the piriformis fossa nail entry point and positioned into the distal femur, followed by the definitive statically locked medullary nail.  A portion of the anterior and anterolateral aspects of the hypertrophic nonunion were removed with an osteotome, morselized and reinserted into the nonunion gap created by the lengthening.  The wounds were then closed and final clinical and radiographic assessments in the operating room demonstrated symmetric leg lengths, angulation, and rotation.

The patient had an uneventful postoperative recovery course and the intraoperative specimens sent for microbiologic examination demonstrated no growth of organisms at the two week mark.  Six weeks postoperatively, he was progressed to from touchdown to full weightbearing.  He had no pain and full knee range of motion.  Twelve weeks postoperatively strengthening exercises  were initiated and the patient was weaned from his walking aids.  Two months later, the patient began light jogging and workouts with a stationary bicycle.  Approximately one year postoperatively, the patient has no pain and is back to his recreational activiities, including completion of the “Escape From Alcatraz” triathlon.

Authored By: David P. Barei.,M.D