The prevalence of notch reformation varies in the literature (Fig. 25.3). In a rabbit model, gross and histologic examination found that exposed cancellous bone at the notchplasty site became covered with fibrous scar tissue, but no osteochondral reconstitution occurred in any specimen [46]. Other animal studies using canine models have found significant refilling of the notchplasty site with fibrous tissue, fibrocartilage, and bone [58, 59]. Clinical studies also have found variable results. A computed tomography (CT) study of patients who had 5-mm notchplasty found no significant differences in multiple notch measurements between 1 week and 1 year postoperatively [60]. During second-look arthroscopy after a mean of 21.2 months, Ahn et al. [61] found 1- to 3-mm notch reformation in 33 % of knees, >3-mm reformation in 7 %, and some degree of graft impingement in 26 % regardless of the graft type. On MRI 6 months after ACL reconstruction, including a 3- to 5-mm notchplasty, Bents et al. [62] found that 97 % of knees showed notchplasty site recortication and 49 % had regrowth of 2–3 mm but that clinical evidence of graft impingement was found in smaller percentages of knees on MRI (22 %) or physical examination (11 %). Another MRI study reported that 6 months following notchplasty, 94 % of patients had 0.5- to 1.5-mm recortication overlying the notchplasty, site and 64 % of patients had a second 1- to 5-mm layer with signal intensity similar to that of hyaline cartilage [63]. While the biologic response to notchplasty appears variable, studies have consistently shown that native anatomy is not restored following notchplasty.

The effect of notchplasty on the articular surfaces of the patellofemoral joint has also been investigated. Morgan et al. [64] measured patellofemoral contact areas and pressures in cadaveric knees and after 3-, 6-, and 9-mm notchplasties. There were no statistical differences between groups at 90°, 105°, and 120° of knee flexion. On the basis of these data, the authors suggested that notchplasty up to 9 mm may be performed without causing anterior knee pain or patellofemoral joint deterioration. Shino et al. [65] evaluated the patellofemoral articular surface at the time of ACL reconstruction and again during second-look arthroscopy at an average of 19 months later. They identified patellofemoral articular surface deterioration in 51 % (93/181) of knees. In the 101 patients who had arthroscopic reconstruction, the incidence of joint surface deterioration was 54 % (13/24) in patients who had notchplasty compared to 38 % (29/77) in those who did not undergo notchplasty. This difference was not statistically significant, and the authors concluded that a 4- to 6-mm notchplasty did not adversely affect the patellofemoral articular surface [65].

Animal studies have shown that an aggressive notchplasty may have an effect on patellofemoral articular cartilage. Using a canine model, LaPrade et al. [59] compared the effects of sham surgery, 4-mm notchplasty (correlating with a 6 to 8-mm notchplasty in humans), and 7 to 8-mm notchplasty (12 to 16-mm notchplasty in humans). Compared with the control group, at 6 months the notchplasty groups had macroscopic articular cartilage changes and significant loss of lateral femoral condyle and trochlear groove articular surface proteoglycans. The authors noted that these histopathologic changes of the articular cartilage were consistent with the changes seen in early degenerative osteoarthrosis. The authors recommended that notchplasty should be as limited as possible or not performed if avoidable. In a similar study using rabbits, Asahina et al. [46] compared patellar articular cartilage changes among a control group, a 1-mm notchplasty group (correlating with a 5-mm notchplasty in humans), and a 3-mm notchplasty group (15-mm notchplasty in humans). There were no microscopic differences between the control and 1-mm notchplasty groups; however, extensive articular deterioration was seen in the 3-mm notchplasty group. These findings were more common when notchplasty was performed in combination with bone–patellar tendon–bone autograft harvest. Despite these reports, there is a lack of high-quality evidence on the effect of notchplasty on patellofemoral articular cartilage in humans.

There is little published data on notchplasty-associated blood loss. In a prospective clinical study, Pape et al. [66] found that notchplasty with a motorized burr resulted in significantly increased blood loss and decreased serum hematocrit compared to no notchplasty; however, there were no clinical differences between groups at 12 months postoperatively. Another study reported significantly less blood loss when notchplasty was performed with a radiofrequency device as opposed to a motorized shaver [67]. While increased intra-articular bleeding may promote fat pad fibrosis that could compromise range of motion, additional studies would be needed to determine the effect of blood loss from the notchplasty site [1, 68].

Markolf et al. [69] conducted the first study on the biomechanical effects of notchplasty. They found that 2- and 4-mm notchplasties resulted in abnormal graft laxity patterns, greater graft excursion, greater graft forces, and higher pretension requirements to restore normal laxity. In a porcine model, Keklikci et al. [70] compared the intact ACL, an ACL-deficient knee, anatomic single-bundle ACL reconstruction, and anatomic single-bundle ACL reconstruction with a 5-mm notchplasty. They found significant differences in the notchplasty group, including greater ATT at 30° and 60° of knee flexion; lower in situ graft force with ATT at 30°, 60°, and 90° of knee flexion; and greater internal rotation tibial torque at 60° of knee flexion. Hame et al. [48] conducted a cadaveric study to determine the effects of varying femoral tunnels before and after 2-mm notchplasty. They found no difference in bone–patellar tendon–bone graft excursions prior to notchplasty but significantly greater graft tightening during 20–90° of knee flexion for all femoral tunnel positions.

Seo et al. [71] used a porcine model to investigate the effects of notchplasty on femoral tunnel diameter and orifice area following ACL reconstruction with suspensory fixation and cyclic loading. In the notchplasty group following testing, there was significantly increased mean longest tunnel diameter, area of the intra-articular orifice, and volumetric bone loss at the anterior margin of the tunnel compared with before testing. In the non-notchplasty group, there were no significant differences in tunnel morphology before and after testing. The authors hypothesize that even with anatomic femoral tunnel placement, removing harder cortical bone during notchplasty exposes softer cancellous bone that may be more susceptible to deformation with cyclic loading. Such alterations in tunnel geometry could result in graft–tunnel mismatch and affect graft position, biomechanics, and laxity.

Fu et al. [72] argue that notchplasty laterally displaces the femoral graft insertion and can result in abnormal knee kinematics. Other studies agree with this hypothesis and contend that observed biomechanical differences may be due to altered tibial–femoral kinematics or because notchplasty recesses the femoral tunnel aperture, effectively altering tunnel length and/or graft length, orientation, loading, and function. Brown et al. [73] suggest that if notchplasty is required, it should be performed after femoral tunnel drilling to avoid lateral displacement of the femoral tunnel. In general, these studies conclude that as little bone as possible should be removed from the intercondylar notch during ACL reconstruction.

Koga et al. [1] conducted the first clinical study on the effect of notchplasty following anatomic double-bundle ACL reconstruction (Tables 25.1, 25.2, 25.3, and 25.4). They found significantly greater objective and subjective loss of extension in the notchplasty group, with six of those patients requiring additional arthroscopic synovectomy for prolonged extension deficit (compared to no patients in the control group). The authors suggested that notchplasty site bleeding caused infrapatellar pad fibrosis and subsequent extension deficit. There were no differences between groups regarding muscle strength, patellofemoral findings, Lysholm or Tegner scores, Lachman or pivot-shift tests, graft failure, or return to sport. ATT as measured by KT-1000 was significantly less in the notchplasty group (0.4 versus 1.2 mm, P = 0.002); however, this was attributed to knee over-constraint in the six notchplasty patients with extension deficit compared to only one in the control group. This study concluded that anatomic double-bundle ACL reconstruction without notchplasty allowed for physiologic graft–roof impingement without extension deficit. In a retrospective review of 75 patients, Muneta et al. [42] found no statistical differences in radiographic or clinical outcomes between notchplasty and non-notchplasty groups but reported that postoperative chronic synovitis occurred only in two patients in the non-notchplasty group. A recent case series found that smaller intercondylar notch dimensions were not a risk factor for graft failure following anatomic single- or double-bundle ACL reconstruction, and the authors did not endorse the use of notchplasty in conjunction with these reconstruction techniques [74].