The success of the ACL reconstruction is influenced by different factors. One of them is the choice of the graft. It is clear that an ideal graft for ACL reconstruction does not exist. For many surgeons, bone–patellar tendon– bone autograft (BPTB) remains still the graft of choice. Although, the graft is criticized for resulting in significant harvest-site morbidity, over the years it has proved to be a stable graft that has longlasting biomechanical properties
[3, 14, 15, 35, 36]
The primary stability of the graft is one of the most important issues in ACL reconstruction. It depends not only on the strength of the graft, but also on its fixation. Regarding BPTB, the graft can be fixed using different methods of fixation e.g.: interference screw, button, staple post and others. It can also be fixed with the use of press-fit technique without use of any hardware
[4, 20, 22, 25, 37].
The femoral press-fit fixation in this study utilizes the native femoral ACL insertion as the point of fixation. The graft is protected additionally by the angulation of the bone block against the ligament graft thereby also minimizing the risk of tunnel misplacement or bone tunnel enlargement
. The most important advantage of press-fit technique is a stable bony fixation on the femoral site without interference screw. It enables direct bone to bone healing and reduces the number of disadvantages associated with hardware fixation, like the inadvertent graft advancement, graft laceration, bone resorption or chronic synovitis, allergic reactions
[21, 23, 24]. Another important feature of this fixation technique is easier revision of ACL reconstruction
[20, 37]. The results of any kind of ACL reconstruction technique should also be evaluated from an economic point of view. As it was documented by Forssblad et al.
, patellar tendon graft is superior to hamstring graft in terms of the cost of the ACL reconstruction. It seems reasonable then, if the clinical outcome does not differ significantly between the grafts, to considered the most inexpensive graft for ACL reconstructions due to restricted economy. Taking this into account, the femoral press-fit fixation ACL reconstruction is a very low costs consuming method comparing to other techniques, especially utilizing different fixation devices.
Similar to other techniques it has also disadvantages and limitations. Comparing for example to hamstring techniques it is not a definitely cosmetic method, what may be an issue for some patients. The poor bone quality, especially in older women might be one of the possible limitations. The two incision operation method, may be technically demanding for beginners. However in our study we did not observe any complications related neither to bone quality nor to technical site.
There is not a big number of studies presenting long-term results of ACL reconstruction
[3, 14, 20, 39]
In our study we present the 15-year results of femoral press-fit ACL reconstruction with patellar tendon autograft in a series of 52 patients. Good clinical results were documented in subjective and objective evaluation. The mean Lysholm and Tegner score were 86.3 and 6.8, respectively. These results are comparable to observations reported by others
[3, 4, 14, 35]; it is reassuring to note that these results are maintained to 15 years. Subjectively, only two patients were found to have a unstable knee and reported giving-way postoperatively.
According to the IKDC standard evaluation form, the overall improvement was documented. 66.5% of 52 patients had normal or nearly normal knee (grade A or B) at the 15-year follow-up. At the time of follow-up 42 patients (81%) had 0 or 1+ on the Lachman test. There was also a marked reduction in the degree of pivot shift after the reconstruction; 43 patients (83%) had a value of 0 or 1+. Similar results have been noted in the literature after ACL reconstruction
[3, 4, 14, 20, 37]. The mean range of motion was 0° to 131°. None of our patients had a loss of extension of more than 7°. Similar results were reported by Al-Husseiny (in 3 cases loss of extension between 3° and 5°, in 2 cases loss more than 5°)
 and Buss (in 4 cases loss of extension between 1° and 5°, in 1 case loss of 8°)
The donor site morbidity seems to be the major concern of all techniques utilizing BPTB grafts. It includes complications such as anterior knee pain, pain when kneeling, patellar fracture, patellofemoral crepitation, numbness caused by damage of the infrapatellar branch of the saphenous nerve and possible loss of quadriceps strength
[14, 16, 36, 41]. In different studies symptoms related to patellar tendon are observed in 40-60% of patients
[3, 4, 16, 20, 37]; in our study it was 42%. It seems possible, like Shelbourne and Trumper
 suggested, to decrease the incidence of anterior knee pain with modern accelerated rehabilitation programs. Fracture of the patella is a rare complication and occurs in 0.1–3% of the cases
[42, 43]. In this study, no patella fracture was reported. We noted two cases of arthrofibrosis defined as 10° loss of motion in the involved knee postoperatively, that needed arthroscopic arthrolysis.
The development of early osteoarthritis might be the main consequence of ACL rupture. We already know that ACL reconstruction does not prevent osteoarthritis, and the development of osteoarthritis post-ACL reconstruction has been reported for all types of reconstructions
[1, 3, 36, 44]. In our study, radiographic examination was performed on all of 52 patients at 15 years after surgery. There was a significant increase of incidence and severity of osteoarthritis between preoperative assessment and 15 years (p = 0.001). Others have reported similar incidences of degenerative changes seen on radiographs
[3, 4, 14, 20, 39, 44]. At the same time there was no correlation between arthritic changes and stability of the knee and subjective evaluation observed.
In conclusion, most of our patients had a satisfactory outcome after ACL reconstruction using femoral press-fit technique, and the results are comparing well to the literature. This ACL reconstruction technique, after switching to arthroscopically assisted, is still used in our centre.
We are aware of limitations associated with the present study. Limitations are related with restricted subject numbers and retrospective analysis. The small number of patients is the result of very strict criteria of inclusion. We aimed to analyze a very definite group of patients. In addition certain number of patients was lost to follow-up. Despite the limitations, in our opinion the present study allows to draw rational conclusions.