Abstract
Background: Given the increasing use of allografts in anterior cruciate ligament reconstruction, selection of appropriate-sized grafts may help individual surgeons as well as the efficiency of the overall system for graft distribution.
Hypothesis: Recipient patient height can predict the desired length for the tendinous portion of a patellar bone-tendon-bone allograft in anterior cruciate ligament reconstruction.
Study Design: Cohort study (Prognosis); Level of evidence, 2.
Methods: A series of 414 knees in 392 consecutive patients undergoing magnetic resonance imaging evaluation of knee pain were enrolled in the study. Data collected from magnetic resonance imaging included patella and patellar tendon length and intra-articular length of the anterior cruciate ligament. Patient age, height, weight, and gender were recorded. Linear regression analysis assessed the correlation between patient height and intra-articular length of the anterior cruciate ligament as well as patellar tendon length. The effect of variance in age, weight, and gender on anterior cruciate ligament intra-articular length was also measured.
Results: A strong positive correlation was found between intra-articular length of the anterior cruciate ligament and patient height (Pearson r = 0.73; P < .001). Anterior cruciate ligament length (y, in millimeters) as a function of height (x, in inches) can be expressed as y = 1.17x — 41.29. As a function of height (x, in centimeters), anterior cruciate ligament length (y, in millimeters) can be expressed as y = 0.4606x — 41.29. Age, gender, and weight did not significantly influence this relationship. A weak positive association was found between patient height and patellar tendon length.
Conclusion: Patient height can predict the desired length of the tendinous portion of a patellar bone-tendon-bone allograft. An addition of 10 mm is made to the predicted anterior cruciate ligament length to allow for aperture tibial and femoral fixation. Patellar bone-tendon-bone allografts can be requested based on recipient patient height as follows: 5 ft, 0 in to 5 ft, 6 in: tendinous length/total length, 45 mm/95 mm; 5 ft, 7 in to 6 ft, 1 in: 50 mm/100 mm; >6 ft, 1 in: 55 mm/105 mm.
Although endoscopic ACL reconstruction using patellar bone-tendon-bone (BTB) autograft has developed into a successful, predictable procedure to restore stability to the ACL-deficient knee, the use of patellar BTB allograft has been advocated to reduce postoperative pain and maximize postoperative rehabilitation.3,7,11,13,14 One potential pitfall when performing endoscopic ACL reconstruction with patellar BTB is graft size mismatch,1,2,5,12,15 a problem that may be more commonly encountered in allograft reconstruction than in autograft reconstruction. In graft size mismatch, the tendon distance between the tibial bone plug and the femoral bone plug is either too short or, more often, too long. In the latter case, the bone plug in the tibia protrudes out of the tunnel, requiring removal of excess bone and compromising the total amount of bone fixation on the tibial side, resulting in less than optimal graft fixation. Although early studies demonstrated that the rate of graft mismatch when using patellar BTB could be as high as 26%,12,15 a recent study by Verma et al 16 reported an overall rate of 13%, with an even higher rate of 20% when using allograft patellar BTB.
Although graft size mismatch can be managed using a variety of techniques,12,15,16 avoiding the mismatch in the first place would obviously be a preferable approach. Several techniques and formulas have evolved in an effort to ensure adequate tibial tunnel length to optimize bone plug fixation with the interference screw.2,8,12,15 Obviously, knowing the length of the tunnel helps avoid graft tunnel mismatch. 6 A number of studies have looked at the relationship between the length of the tibial tunnel (millimeters) and angle of the tibial guide (degrees). Miller and Hinkin 9 suggested the “N + 7” rule, in which the length of the tendinous portion of the graft (millimeters) plus 7° is the optimal angle of the tibial guide. Olszewski et al 10 suggested using a combination of the “N + 2 mm” rule, in which the optimal length of the tibial tunnel is patellar tendon length (millimeters) + 2 mm, and the “N + 7” rule. More recently, Verma et al 16 suggested the use of the “N + 10” rule. As a group, these techniques focus on length of the tibial tunnel, not on the graft itself, and therefore do not effectively address the primary issue at hand, that being inappropriate length of the tendinous portion of the graft.
Another problem exists in requesting the graft itself. Before the development of our method, a graft was simply requested as a patellar BTB graft with no specification of length. Obviously, there is a significant variability in donor graft sizes that may or may not be appropriate for a particular patient. For example, a surgeon may request a patellar BTB graft and receive a graft with a 60-mm tendon length for a patient who is optimally treated with a 45-mm graft tendon length.
We have developed a novel technique that uses the height of the patient to accurately predict the correct length of allograft tendon, the x-factor, to achieve optimal bone plug fixation (Figure 1). The purpose of this study was to validate this method, testing the hypothesis that the x-factor is correlated to patient height. We also tested the hypothesis that patellar tendon length is correlated to patient height.
Model of patellar bone-tendon-bone graft.
Materials and Methods
The study protocol was approved by the institutional review board of the study site. Patients undergoing MRI for evaluation of nonspecific knee pain with no history of inflammatory or degenerative arthritis were considered for inclusion in the study. Patients with a history of previous knee surgery and pregnant patients were excluded from the study.
If they agreed to participate in the study, the patients’ age, gender, height, and weight were recorded. An MRI was obtained from each knee using an Oni OrthOne 1.0-T dedicated MRI (ONI Medical Systems Inc, Wilmington, Mass). MediSurf software (Algotec Systems Ltd, Raanana, Israel), an on-screen direct image measuring program calibrated for the program and images was used to review the coronal and sagittal images of the knee simultaneously. The longest dimension of the patella was measured and recorded. The patellar tendon was measured at the midline of the patella on the sagittal view based on the coronal image. Finally, the intra-articular length of the ACL was measured from the image in which the tibial and femoral ACL attachments were visualized.
Statistical and linear regression analysis was used to determine the correlation between patellar tendon length and patient height as well as intra-articular length of the ACL and patient height. The effect of variance in age, weight, and gender on intra-articular length of the ACL was also measured.
Results
A series of 414 knees in 392 consecutive patients undergoing MRI evaluation of knee pain were enrolled in the study. Summary demographic data are presented in Table 1. The most common findings on MRI included medial meniscal tears (155 knees), lateral meniscal tears (85 knees), chondromalacia (52 knees), significant bone edema (35 knees), and articular cartilage lesions (35 knees). No findings (normal knee) were reported in 48 knees.
Subject Characteristics a
There were 221 right legs and 193 left legs imaged; there were 217 male and 197 female participants.
A strong positive correlation was found between the intra-articular length of the ACL and patient height (Pearson r = 0.73, P < .001). The linear regression equation for ACL length (y, in millimeters) as a function of height (x, in inches) can be expressed as y = 1.17x – 41.29 (see Figure 2). Based on height in centimeters, ACL length (y, in millimeters) as a function of height (x, in centimeters) can be expressed as y = 0.4606x – 41.29. Demographic variables such as age, gender, and weight did not significantly influence this relationship. No significant association was found between patient height and patellar tendon length.
Intra-articular length of ACL related to patient height.
Discussion
For ACL reconstruction using patellar BTB allograft, patient height can be used as an accurate predictor of the desired length of the tendinous portion of the graft. An addition of 10 mm is made to the predicted length of the tendinous portion of the graft to achieve aperture tibial and femoral fixation while allowing some flexibility for tibial tunnel length. The appropriate x-factor roughly correlates with height as follows: 5 ft to 5 ft, 6 in (152-168 cm), 39 to 46 mm; 5 ft, 6 in to 6 ft, 1 in (168-185 cm), 46 to 54 mm; >6 ft, 1 in (185 cm), 54 mm. For total graft length (x-factor plus bone plugs), an additional 25 mm is added for each bone plug (Table 2). The study site currently requests patellar BTB allografts from the tissue bank based on the height of the patient scheduled for ACL reconstruction as shown in Table 2. As a rough guide to the accuracy of this formula, if it were used on all of the patients in this study, the estimate would be off by more than 5 mm in 6.5% of patients and off by more than 10 mm in only 1.7% of patients.
Selection Criteria for Allograft Patella Bone-Tendon-Bone
Patellar tendon length did not correlate with patient height or intra-articular ACL length. This suggests that preoperative planning for ACL reconstruction using patellar BTB autograft should recognize the potential for mismatch between patellar tendon length and intra-articular ACL length. In most cases, the preoperative radiograph can be used to measure the patellar tendon length. This value can be compared with the predicted intra-articular ACL length based on the patient's height. There may be cases in which an MRI measurement of the intra-articular length of the ACL on the contralateral limb may be warranted to help predict whether patellar BTB autograft will be appropriately sized. For example, patients with patella alta are likely to generate autografts with a long tendinous portion of the graft. Because the incidence of graft tunnel mismatch has been shown to be higher when the tendinous portion of the graft is >50 mm, 12 additional preoperative information such as an MRI of the contralateral knee may be helpful.
The authors are not aware of any other published algorithm to predict appropriate allograft length for ACL reconstruction, although a couple of studies have assessed the relationship between patient height and ACL length. Denti et al 5 looked at graft-tunnel mismatch in 50 reconstructed and 9 cadaveric knees and found no significant relationship between length of intra-articular ACL graft or patellar tendon length and patient height and weight. Another study reviewing 30 cadaveric knees also found no relation between height and patellar tendon length, 4 although this study did find a correlation of 0.40 between height and ACL length. These studies were based on relatively small cohorts in comparison with the large, diverse population used as a basis for the x-factor.
The authors recognize that the literature has shown that the mean tendon and intra-articular ACL graft lengths are 45 to 50 mm and 20 to 25 mm, respectively.5,12 However, both of these studies demonstrated considerable variability in the length of the patellar tendon. In the study by Denti et al, 5 the mean patellar tendon length was 45.5 ± 4.7 mm. One quarter of the tendons were shorter than 42 mm, and one quarter were longer than 48 mm. Shaffer et al 12 reported a mean patellar tendon length of 48.4 ± 6.0 mm with a range of 40 to 63 mm. Fifteen percent of the tendons were shorter than 42 mm, and 20% were longer than 52 mm. There is clearly some variation in patellar tendon length across patients, which probably relates to the rates of graft tunnel mismatch reported in the literature. The intra-articular graft length showed similar variability, with a range of 21 to 33 mm in the Shaffer et al 12 study and 15 to 26 mm in the Denti et al 5 study.
Using the x-factor as a guide to ordering patellar BTB allograft for ACL reconstruction protects against graft size mismatch. This minimizes the chance of receiving a graft that is too short or too long, which may put graft fixation at risk. Use of the x-factor at the study site has virtually eliminated graft size mismatch in this clinical setting. Patient height can be used to predict appropriate length of allograft to be used in ACL reconstruction.