ABSTRACT:
OBJECTIVE To evaluate possible connections between the weight and height of patients submitted
to total knee arthroplasty (TKA), with the length, width and area of the anterior
cruciate ligament (ACL) fovea, as verified during surgery.
METHODS A total of 33 proximal tibial joint surfaces, obtained from TKA tibial sections of
33 patients, were used in the present study. The ACL was resected with a delicate
scalpel to expose the ACL tibial fovea. Then the periphery of this fovea was delimited
with a marker pen by means of small dots. Each piece was photographed, and the ACL
tibial fovea length, width, and area were measured with the ImageJ (National Institutes
of Health, Bethesda, MD, USA) software. Statistical analysis studied the correlation
between anthropometrics data of the patients and the measurements of the ACL tibial
fovea.
RESULTS The ACL tibial fovea length, width, and area were, respectively, 11.7 ± 2.0 mm, 7.1
± 1.4 mm and 151.3 ± 22.2 mm2. There was a statistically significant relationship between the height of the patients
and the width of the ACL tibial fovea. The width of the ACL fovea could be predicted
by the formula:
width
=
107
.
294
−
133
.
179
×
height
+
44
.
009
×
squared
height
.
CONCLUSION The height of the patients may predict the width of the ACL tibial fovea, and therefore,
may allow surgeons to choose the more adequate graft for each patient in ACL reconstruction.
Keywords: tibia; anterior cruciate ligament; anatomy.
RESUMO:
OBJETIVO Avaliar possíveis relações entre o peso e altura de pacientes submetidos à artroplastia
total do joelho com o comprimento, largura e área da fóvea do ligamento cruzado anterior,
verificados na cirurgia.
MÉTODOS Um total de 33 superfícies articulares tibias proximais, obtidas nos cortes tibiais
de artroplastia total do joelho de 33 pacientes, foram utilizadas no presente estudo.
O ligamento cruzado anterior foi dissecado cuidadosamente e ressecado com bisturi
delicado, para expor sua fóvea. Depois, a periferia dessa fóvea foi demarcada por
pequenos pontos, com um marcador. Cada peça foi fotografada e as medições do comprimento,
largura e área da fóvea tibial do ligamento cruzado anterior foram feitas com o programa
ImageJ (National Institutes of Health, Bethesda, MD, EUA). A análise estatística avaliou
a correlação entre os dados antropométricos dos pacientes com as medidas da fóvea
tibial do ligamento cruzado anterior.
RESULTADOS O comprimento, a largura e a área médios da fóvea tibial do ligamento cruzado anterior
foram, respectivamente 11,7 ± 2,0 mm, 7,1 ± 1,4 mm e 151,3 ± 22,2 mm2. Houve relação estatisticamente significativa entre a altura dos pacientes e a largura
da fóvea tibial do ligamento cruzado anterior. A largura da fóvea do ligamento cruzado
anterior pôde ser predita pela fórmula:
largura
=
107
,
294
−
133
,
179
×
altura
+
44
,
009
×
altura
ao
quadrado
.
CONCLUSÃO A altura dos pacientes pôde a prever a largura da fóvea tibial do ligamento cruzado
anterior e, assim, pode ajudar os cirurgiões escolher o enxerto mais adequado para
cada paciente, nas reconstruções do ligamento cruzado anterior.
Palavras-chave: tíbia; ligamento cruzado anterior; anatomia.
FIGURES
| Citation: Gali JC, Del Giglio DB, Patriarcha LF, Cruz BAP, Gali Filho JC. Correlation bettween Individual Anthropometric Characteristics and Anterior Cruciate Ligament Tibial Fovea Measurements*. 55(1):88. doi:10.1055/s-0039-1700816 |
| Note: * The present study was conducted at the Department of Orthopedics, Faculty of Medical and Health Sciences of the Pontifícia Universidade Católica de São Paulo, Sorocaba, SP, Brazil. |
| Conflict of Interests: The authors have no conflict of interests to declare. |
| Received: August 17 2018; Accepted: October 16 2018 |
INTRODUCTION
The number of anterior cruciate ligament (ACL) reconstructions in the period between 1994 and 2006 increased by 50%, especially among women and people < 20 and > 40 years old.1 The outcome of ACL reconstruction depends on the biological response of the patient, the surgical technique, and rehabilitation. The current trend is to perform anatomical reconstructions, defined as those with the intention of seeking the functional restoration of the ACL to its native dimensions, collagen orientation and insertion sites in order to try to reproduce normal anatomy, restore kinematics and provide long-term joint health.2
A more recent conception is that reconstructions need to be individualized. Since no two ACLs are the same, the graft must be shaped for each individual.3 Since ACL reconstruction can be considered a graft transplant, its size should be especially evaluated. However, in most reconstructive surgeries using autologous grafts, the size of the ACL is mainly determined by the size of the graft obtained and not by the measurement of the ACL insertion. If the surgeon could predict the size of the ACL tibial insertion before surgery, he could also choose the most appropriate autograft size.4
Therefore, anticipating the dimensions of the ACL insertion can be valuable for better surgical planning. This can be done by CT scan5,6 or magnetic resonance imaging (MRI),7-11 but it can also be predicted by the individual physical characteristics.3,4
In this decade, some authors sought to correlate the weight and height of patients with the size of their tibial ACL insertion. These studies were conducted in the United States3 and in South Korea.11-13 Therefore, a study is important to evaluate if there is any correlation between the weight and height of the patients and the ACL tibial foveal dimensions in the Brazilian population, composed of a mixture of ethnicities, particularly African descendant, Portuguese, Italian, South-American Native, Spanish, German, Japanese, Polish, Arab or Semite.14
The objective of this study was to evaluate possible correlations between the weight and height of patients undergoing total knee arthroplasty (TKA) and the length, width and area of the ACL tibial fovea found in TKA. Our hypothesis was that the preoperative evaluation of the anthropometric characteristics of our patients could anticipate the ACL tibial fovea dimensions.
METHODS
We have obtained the approval of the Ethics Committee of our institution for the present study (CAAE: 21220013.6.0000.5373, opinion: 509.126). Information on the weight and height of the patients was obtained from the medical records of the Hospital of the Institution where the study was performed.
From September 2014 to June 2017, 74 TKAs were performed in our service. Only knees whose ACLs had normal macroscopic appearance were included in our study. Exclusion factors were history of previous joint trauma, infectious or inflammatory diseases that could alter the ACL tibial insertion, macroscopic morphological alterations of the ACL and presence of osteophytes around the tibial insertion of the ACL at the time of evaluation. In cases of bilateral surgery, only one knee (right or left) was chosen by lot (head or tails). Thus, 41 knees were excluded from our investigation, 7 (9.4%) because they were from patients who had bilateral arthroplasty, and 34 due to macroscopic changes in the normal anatomy of the ACL or the presence of osteophytes that altered ACL insertion (45.9%).
Thus, our final sample consisted of 33 knees from 33 patients. Nineteen were women (57.6%) and 14 were men (42.4%). Eighteen knees were right (54.5%) and 15 were left (45.5%). The average age of the subjects was 69.5 years old (± 7.3), the median weight was 78.4 kg (± 12.3) and the average height was 1.58 m (± 0.08).
Each piece, corresponding to the proximal tibial portion, taken at the tibial section of the TKA, was evaluated by the same surgeon, senior of the team, after surgery, with a four-times magnifying lens. After thorough ACL dissection, it was resected with a delicate scalpel to expose its tibial fovea, a depression between the intercondylar eminences corresponding to the tibial insertion of the ACL.5 The fovea was delimited, in its periphery, by means of small dots, with a marker pen. A millimeter ruler was placed next to the pieces to serve as a parameter for measurements, and each of them was photographed with a Nikon D3100 (Nikon, Tokyo, Japan) digital camera, which was positioned parallel to the plane of the examination table of the pieces.
All of the measurements were made subsequently, using the ImageJ (National Institutes of Health, Bethesda, MD, USA) software, by another member of the staff. The length of the ACL tibial fovea was defined as its largest axis in the anteroposterior direction, and its width as its largest axis in the lateromedial direction, perpendicular to the anteroposterior axis. To measure the area, the previously demarcated dots were joined in the computer (Figure 1). We also obtained the ratio between the length and width of the ACL tibial fovea. The measurements were made twice, and the average of the values obtained was used.

Statistical Analysis
We used multivariate analysis of variance (MANOVA) to interpret the results, since the sample can be considered small, and has three interrelated variables (width, length and area). Thus, we adopted as a strategy the application of a one-factor MANOVA (gender) and two covariates (height and weight) and, simultaneously with the MANOVA, we performed the confidence interval calculations and multiple comparisons adjusted by bootstrap (n = 1.000; with bias correction) to pinpoint where the significant differences were.
To examine the normality of each of the variables, the Shapiro-Wilk test was used, with a 5% significance level, and we rejected univariate normality in all dependent variables except length. The Levene test was used for genders and sides and, in all variables, we do not reject the hypothesis of homogeneity of variances (Table 1).
| Levene* | Shapiro-Wilk | |||||
|---|---|---|---|---|---|---|
| Statistic | DF | Sig. | Statistic | DF | Sig. | |
| Age | 0.686 | 31 | 0.414 | 0.957 | 33 | 0.208 |
| Weight | 0.173 | 31 | 0.68 | 0.973 | 33 | 0.554 |
| Height | 2.337 | 31 | 0.136 | 0.928 | 33 | 0.030 |
| Width | 0.929 | 31 | 0.343 | 0.850 | 33 | 0.000 |
| Length | 0.437 | 31 | 0.514 | 0.977 | 33 | 0.680 |
| Area | 1.967 | 31 | 0.171 | 0.831 | 33 | 0.000 |
Abbreviation: DF, degrees of freedom (sample size - the number of estimated parameters).
* Grouping variable: Gender.
RESULTS
The mean length, width and area of the ACL tibial fovea were, respectively, 11.7 ± 2.0 mm; 7.1 ± 1.4 mm and 151.3 ± 22.2 mm2. On average, the ACL fovea length was 1.64 times the width. In 27 knees (81.8%) the length of the ACL tibial fovea was < 14 mm and in 28 patients (84.8%) the fovea width was < 8 mm. The full results are described in Table 2.
| PATIENT | AGE | GENDER | WEIGHT | HEIGHT | SIDE | WIDTH | LENGTH | AREA |
|---|---|---|---|---|---|---|---|---|
| Kg | m | mm | mm | mm2 | ||||
| 1 | 69 | F | 51 | 1.48 | R | 5.96 | 10.27 | 50.4 |
| 2 | 72 | M | 73 | 1.62 | L | 6.29 | 10.10 | 52.8 |
| 3 | 60 | M | 79 | 1.65 | L | 7.55 | 14.27 | 78.9 |
| 4 | 67 | F | 80 | 1.58 | R | 7.03 | 12.16 | 61.8 |
| 5 | 66 | F | 77 | 1.53 | L | 7.92 | 11.92 | 67.1 |
| 6 | 72 | M | 108 | 1.64 | L | 6.48 | 9.70 | 49.4 |
| 7 | 75 | F | 80 | 1.50 | R | 5.09 | 9.56 | 35.8 |
| 8 | 68 | F | 85 | 1.57 | R | 6.76 | 9.52 | 49.2 |
| 9 | 81 | M | 70 | 1.70 | R | 7.22 | 10.97 | 52.3 |
| 10 | 79 | F | 69 | 1.50 | L | 6.99 | 11.38 | 51.2 |
| 11 | 81 | M | 70 | 1.70 | L | 6.11 | 9.44 | 51.3 |
| 12 | 62 | F | 77 | 1.56 | L | 7.87 | 14.62 | 88.2 |
| 13 | 62 | M | 76 | 1.78 | R | 8.75 | 15.86 | 90.8 |
| 14 | 58 | F | 72 | 1.50 | R | 6.96 | 10.13 | 47.7 |
| 15 | 74 | M | 81 | 1.60 | R | 8.33 | 12.98 | 76.6 |
| 16 | 64 | M | 71 | 1.62 | R | 7.52 | 11.68 | 68.0 |
| 17 | 70 | F | 94 | 1.58 | L | 7.60 | 12.65 | 70.4 |
| 18 | 83 | F | 89 | 1.55 | L | 7.13 | 10.85 | 51.0 |
| 19 | 62 | F | 84 | 1.60 | R | 7.00 | 10.75 | 52.9 |
| 20 | 58 | M | 88 | 1.64 | L | 8.46 | 12.20 | 66.5 |
| 21 | 72 | F | 63 | 1.50 | L | 5.13 | 8.62 | 33.4 |
| 22 | 62 | F | 62 | 1.49 | R | 6.96 | 13.26 | 68.9 |
| 23 | 60 | M | 92 | 1.64 | R | 7.32 | 11.89 | 61.9 |
| 24 | 68 | M | 66 | 1.58 | R | 8.16 | 14.10 | 96.6 |
| 25 | 73 | F | 60 | 1.50 | L | 4.87 | 10.69 | 40.0 |
| 26 | 71 | M | 84 | 1.58 | R | 6.56 | 11.54 | 56.2 |
| 27 | 65 | F | 90 | 1.55 | R | 4.98 | 13.11 | 54.0 |
| 28 | 66 | F | 73 | 1.45 | L | 7.13 | 11.84 | 66.0 |
| 29 | 77 | M | 76 | 1.64 | R | 6.23 | 13.13 | 60.1 |
| 30 | 64 | F | 108 | 1.60 | L | 7.34 | 10.00 | 53.7 |
| 31 | 73 | F | 77 | 1.55 | R | 5.90 | 8.06 | 37.3 |
| 32 | 83 | F | 75 | 1.53 | R | 7.74 | 14.51 | 84.1 |
| 33 | 79 | M | 88 | 1.85 | L | 12.80 | 16.69 | 151.3 |
| MINIMUM | 58.00 | H 14 | 51.00 | 1.45 | R 18 | 4.87 | 8.06 | 33.40 |
| MAXIMUM | 83.00 | M 19 | 108.00 | 1.85 | L 15 | 12.80 | 16.69 | 151.30 |
| MEAN | 69.58 | 78.42 | 1.59 | 7.10 | 11.77 | 62.90 | ||
| STANDARD DEVIATION | 7.391001 | 12.36434 | 0.086446 | 1.430719 | 2.052747789 | 22.29646111 |
All of the individual correlations were significant, but none of the coefficients exceeded 0.90, the limit of multicollinearity. We found a statistically significant relationship only between patient height and ACL tibial foveal width. The relationship between these two variables is displayed in Figure 2. In this figure, there is a value that stands out, but which did not seem to us to be an extreme value (outlier), given that a height of 1.85 m is perfectly plausible. There was a quadratic relationship between height and width, which fit much better than a linear relationship (Figure 3).


The adjustment of this chart straight line by Ordinary Least Squares (OLS) returned an adjusted R2 of 0.369, while the quadratic curve adjustment returned an adjusted R2 value of 0.471. The R2 value of this model was 0.71, indicating that 71% of the ACL tibial foveal width variations can be explained by the height variation of the patients. The coefficient values of the quadratic model, all significant at the 5% confidence level, were: Width = 107.294-(133.179 × height) + (44.009 × height squared).
DISCUSSION
The main finding of our study was that the width of the ACL tibial fovea found in knees of individuals undergoing TKA was related to the height of the patient. In 45.9% of our initial series, there were macroscopic changes in the normal ACL anatomy or osteophytes around the ligament insertion. Douglas et al,15 in 95 patients undergoing TKA, found no ACL in 12% of the patients and macroscopic abnormalities in 45.2% of the subjects. Lee et al16 evaluated the integrity of the ACL in 107 knees submitted to TKA. In 41 patients (39%) the ligament was deficient at the time of surgery, and 31% of these could remember an episode consistent with an ACL injury or a history of previous instability. To Mullaji et al,17 40% of arthritic knees at the time of arthroplasty showed severely degenerated ACLs.
In 27 knees (81.8%) of our sample, the length of the ACL fovea was < 14 mm, and in 28 patients (84.8%) its width was < 8 mm. In our methodology, we resected the tibial portion of the ACL and measured its fovea. This assessment method differed from most studies found in the literature, in which the tibial insertion area of the ACL was measured outlining its periphery, which also includes expansions of its fixation. That is why our measurements are very similar to those obtained by Purnell et al,5 and smaller than those of the other authors. Another possible explanation for our measurements being smaller than those of other authors was that the height of our patients was relatively low (1.59 ± 0.08m) and, as the ACL fovea width was related to the height of the patients, its measures were also smaller. The comparison between our results and those of other authors, by the various modes of evaluation, is found in Table 3.3-13,18-24
| Reference | Length (mm) | Width (mm) | Ratio L/W | Area (mm2) |
|---|---|---|---|---|
| Morgan et al.18 (1995)a | 18 | 10 | 1.8 | NE |
| Park et al.12 (2015) a | 13.8 | 9.8 | 1.4 | NE |
| Park et al.13 (2017) a | 14.3 | 10.2 | 1.4 | NE |
| Tállay et aI.19 (2008) a | 19.5 ± 2.6 | 10.3 ± 1.9 | 1.89 | NE |
| Kim et aI.11 (2018) a | 13.8 | 7.2 | 1.91 | NE |
| Kopf et aI.3 (2011) b | 17,0 ± 2,0 | AM 9,2 ± 1,1 | - | NE |
| PL 7,0 ± 1,0 | - | |||
| Guenther et al.8 (2017) b | 16.3 ± 1.6 | 9.7 ± 1.4 | 1.65 | 123.8 ± 21.5 |
| Guenther et al.20 (2017) b | 16.1 ± 2.6 | 9.6 ± 1.5 | 1.67 | NE |
| Edwards et al.21 (2007) c | 18 ± 2 | 9 ± 2 | 2.0 | NE |
| Purnell et al.5 (2008) c | 10.7 ± 1.3 | 7.4 ± 1.2 | 1.44 | NE |
| Ferretti et al.22 (2012) c | 18.1 ± 2.8 | 9 ± 2 | 2.0 | NE |
| Siebold et al.23 (2015) c | 12.6 ± 2.3 | NE | - | 110.9 ± 14.7 |
| Fujimaki et al.24 (2016) c | NE | NE | - | 175.8 ± 64.3 |
| Tampere et al.6 (2017) c | NE | NE | - | 159.2 |
| Iriuchishima et aI.4 (2015) c | NE | NE | - | 133.8 ± 31.3 |
| Guenther et al.8 (2017) d* | 16.6 ± 1.3 | 10.2 ± 1.0 | 1.62 | 132.8 ± 15.7 |
| Guenther et al.8 (2017) d** | 16.7 ± 1.3 | 10.4 ± 0.9 | 1.60 | 136.7 ± 15.4 |
| Tashiro et al.9 (2017) d | 15.6 ± 1,8 | 14.5 ± 2.2 | 1.07 | 182.7 ± 41.1 |
| Ichiba et al.7 (2014) d | 15.2 ± 1.9 | NE | - | NA |
| Araki et al.10 (2017) d | NE | NE | - | 134.7 ± 22.9 |
| Kim et aI.11 (2018) a | 12.4 | 8.8 | 1.40 | NA |
| Araki et al.10 (2017) e | NE | NE | - | 135.2 ± 15.1 |
| Our results a | 11.7 ± 2.0 | 7.1 ± 1.4 | 1.64 | 151.3 ± 22.29 |

In fact, Lord et al25 published a study of the function of the ACL tibial insertion fibers in the resistance of anterior and rotational displacements, with robotics, in six degrees of freedom of movement. They concluded that the central, posterior and lateral fibers of the ACL tibial insertion play a secondary role in the restriction of tibial displacements and that the most important fibers are the anterior and the anteromedial peripheral portion. However, we cannot be sure that this happens in vivo, given the array of interactions of static and dynamic restrictors responsible for knee stabilization in the various simultaneous three-dimensional movements that happen during the daily activities of the human being.
In our view, the graft is different from the original ligament, so it does not necessarily have to reproduce the same measurements of this ligament, but rather have similar failure force and stiffness to the primary ligament. In our study, the ACL insertion length was 1.64 times the width size, so the latter is the limiting factor in determining the graft diameter to be used.
Hamada et al26 reported that the graft should not cover the entire ligament insertion area, as they have greater mechanical properties than the native ACL and become hypertrophic after implantation. In contrast, LaPrade et al27 believe that the prediction of the width of the ACL tibial insertion has been neglected, and large grafts can damage the roots of the anterior or posterior lateral meniscus.
Our measurements were made on knees of patients during TKA. There was a positive correlation between patient height and ACL tibial foveal width. Kopf et al3 quantified ACL insertion measurements in 137 patients during ACL reconstructions. The mean length of the tibial insertion of the ACL was between 16 and 18 mm in 66.4% of the subjects. There was a significant positive correlation between the weight and height of the patients and the length of the ACL tibial insertion (p < 0.05). However, the delimitation of the tibial insertion, in arthroscopy, was made with electrocautery, and measurements were made with a ruler, which can generate a subjectivity bias.
Ichiba et al7 evaluated ACL tibial insertion measurements in 100 MRIs of 100 patients, 50 men and 50 women. For these authors, the average length of this insertion in men was 16.1 mm and, in women, 14.3 mm (mean 15.2 ± 1.9 mm). There was a significant correlation between insertion length and weight and height of the patients (p < 0.05). The width and tibial insertion area of the ACL were not evaluated by the authors.
Park et al12 published the results of ACL tibial insertion measurements made in 127 patients during TKA. Measurements were made at the peripheral margin of the ligament insertion. The average length of the ACL tibial insertion was 13.8 mm (10.0-18.0 mm), and the average width was 9.8 mm (6.3-13.5 mm). Tibial length < 14 mm was found in 53.5% of the patients. There was a significant correlation between ACL insertion length and weight and height of the patients. Width was significantly related to weight, but correlation was poor.
In a most recent article, Park et al13 reported prospective data on ACL tibial insertion measurements performed in 209 patients, during TKA, in which the peripheral margins of the ACL insertion were also marked. In this study, the average length of the ACL tibial insertion was 14.3 mm (10.0-19.7 mm), and the average width was 10.2 mm (6.3-13.7 mm). The insertion length was < 14mm in 46.5% of the studied population. The width of the ACL insertion in the tibia was associated with weight by multivariable linear regression.
Our study has some limitations. Although we excluded the knees with macroscopic morphological ligament changes and those with osteophytes around the tibial insertion of the ACL, the mean age of our population (69.5 years old) is higher than that of individuals who undergo ACL surgical reconstruction. Our sample may be considered small, but it is larger than several previously published studies.5,6,22-24 In addition, MANOVA was specifically chosen to interpret the results because we considered the sample small. Finally, all of the measurements were made by a single member of our group, which may be a bias generator, although they were performed twice, and the mean results were used.
CONCLUSIONS
The width of the ACL tibial fovea could be predicted by the height of the patients, providing a means to help the surgeon in the choice of the most suitable graft for each patient in ACL reconstructions.




