Historically, bone-tendon-bone grafts were preferred for anterior cruciate ligament reconstruction as a result of successful initial fixation techniques and outcomes6,7. As concern regarding anterior knee pain, extensor mechanism dysfunction, and the postoperative range of motion increased, alternate methods, particularly use of hamstring tendon grafts, became prevalent8. With increasing acceptance of the belief that they provided decreased donor-site morbidity and favorable biomechanical properties, hamstring and soft-tissue allografts have been used increasingly for anterior cruciate ligament reconstruction9,10. There are various techniques for hamstring graft fixation on the femoral side, including use of interference screws, button techniques, and cross-pin fixation. Studies have shown cross-pinning methods to be superior to other techniques with respect to pullout stiffness, pullout strength, and slippage6,7,9-16.
The Arthrotek Bone Mulch Screw consists of a tapered screw with a smooth beam at one end. The hamstring graft is passed through the primary femoral tunnel for the graft and secured by the screw, which is directed perpendicular to the femoral tunnel from the lateral aspect of the femur, through an 8-mm-diameter accessory drilled tunnel, until the smooth beam touches the medial wall of the graft femoral tunnel. With use of a passing suture, the graft is looped over the smooth distal beam of the screw, passed back distally through the femoral tunnel, tensioned, and secured with use of a washer locking device to the tibia.
There has been increasing concern about the prevalence of supracondylar femoral fractures after anterior cruciate ligament reconstruction4,17-20. Wiener and Siliski described a distal femoral shaft fracture resulting from gradual fatigue and overuse seven months following anterior cruciate ligament reconstruction with a patellar tendon autograft2. They concluded that multiple trocar cortical holes resulted in stress risers leading to the fracture. This conclusion was supported by a finding of sclerosis at the trocar holes, which was thought to be caused by thermal necrosis2. Multiple drill-hole sites were not apparent on the imaging studies of our patient, and they were not described in the operative report on the initial surgery.
The concept of fracture facilitation by cortical weakening from stress risers is well supported by biomechanical studies21,22. Brooks et al. determined that diaphyseal drill holes with a diameter of >20% of the bone decrease the torsional energy-absorbing capacity of bone by 55%22. In a later study of rabbits, bending strength was reduced by 30% when holes on the tension side of the cortex were <30% of the bone diameter23. A 10.5 × 30-mm hole in the lateral femoral condyle can represent a substantial stress riser in the lateral aspect of the distal part of the femur if one considers that the anterior-to-posterior dimension of the femur ranges from 50 to 60 mm. Furthermore, while the manufacturer's recommendations were followed in this case, drilling an 8-mm-diameter lateral femoral tunnel and then inserting a 10.5-mm screw may have resulted in increased circumferential tensile forces around the screw that could have predisposed our patient to a fracture. One would expect such a fracture to be more likely to occur in the early postoperative period rather than eleven years after the initial surgery, when bone remodeling should have decreased the circumferential tensile stress around the screw. However, we are not aware of any data in the literature indicating how distal femoral bone remodeling over time may alter the stress-rising effects of a large osseous defect. An alternative cross-pinning technique, such as the use of two smaller-diameter cross-pins as a substitute for the larger single screw, could have been used to minimize this problem.
To our knowledge, all case reports describing fractures resulting from stress risers have noted that they originated in the femoral tunnel for the graft. This has typically yielded the characteristic fracture pattern extending superolaterally from the intraosseous tunnel created for the graft. Typically, these are oblique fractures in the coronal plane extending from the tunnel site or the site of the interference screw2,12,17,24. The weakest area of the distal part of the femur is believed to be its posterior aspect25. In a recent report, Mithoefer et al. illustrated the importance of surgical technique as it relates to optimal femoral tunnel placement during reconstruction24. They noted that optimally the tunnel should be placed as far posterior as possible without compromising the posterior cortex. The fracture pattern in our patient is unique in that, on the coronal computed tomography section, it was seen as a shearing fracture not through the graft tunnel but instead through the lateral femoral tunnel drilled for the cross-pinning fixation device. On the axial cuts, the fracture extended from medial to lateral as opposed to the posterior-to-anterior orientation of fractures that occur through the graft femoral tunnel.
An additional factor that may play a role in the development of a late fracture after anterior cruciate ligament reconstruction is enlargement of the bone tunnel, which has been reported to occur in up to 68% of cases26. The precise etiology of this process has yet to be determined. It was initially thought to be an immune-mediated phenomenon, but research has shown no difference in the prevalence of tunnel enlargement associated with allografts compared with that associated with autografts27. It is now thought that tunnel lysis is a complex imbalance of osteoclastic, osteoblastic, immune, and nonspecific biological factors. Enlargement of both the graft femoral tunnel and the lateral femoral tunnel drilled for a cross-pinning device may additionally weaken the distal part of the femur, but this was not a relevant factor in our case, judging by the computed tomography images, which did not demonstrate unusual enlargement of the tunnels.
Our case illustrates that the risk of fracture from the stress-rising effect of femoral fixation of an anterior cruciate ligament graft by cross-pinning with a large screw device may persist many years after the initial repair, and perhaps indefinitely. Patients should be informed of the risk of postoperative fractures and warned that a fracture may occur even many years after a repair done with proper technique. Also, minimization of the size of any cortical defect should be emphasized. Additional avenues of research include the relationships between drill-hole size and screw placement and the prevalence of fractures. Studies evaluating how stress distributions around a distal femoral cortical stress riser evolve over time are also needed.