Morphometric study of the femoral intercondylar notch of knees with and without injuries of anterior cruciate ligament (A.C.L.), by the use of software in digitalized radiographic images

Abstracts

Os autores sugerem a aplicação de um software sobre imagens radiográficas digitalizadas para análise morfométrica da fossa intercondilar e dos côndilos femorais. O programa permite o tracejamento de linhas guias que facilitam a mensuração da extremidade distal do fêmur. Foram analisadas 39 radiografias simples da fossa intercondilar femoral dos joelhos direito e esquerdo, obtidas de indivíduos do sexo masculino reunidos em grupo normal (n=23) e grupo lesionado (n=16). A média de idade da amostra foi de 26,56 anos. As variáveis analisadas foram largura bicondilar femoral, larguras da fossa intercondilar ao nível do sulco poplíteo e da base da fossa, a altura da fossa intercondilar e a largura do côndilo femoral lateral. A fossa intercondilar foi classificada quanto ao formato em cônica, circular e retangular. Os resultados encontrados sugerem que a largura da base da fossa e a largura do côndilo femoral lateral seriam fatores de risco importantes na lesão do L.C.A. Os valores médios das variáveis analisadas aproximam-se dos descritos na literatura especializada em mensurações diretas em peças cadavéricas e ressonância nuclear magnética e demonstram que a aplicação de um software sobre as imagens radiográficas digitalizadas proporciona uma mensuração confiável, mesmo utilizando-se de imagens radiográficas simples e de baixo custo.

Fossa intercondilar femoral; Ligamento cruzado anterior; Morfologia; Joelho

ORIGINAL ARTICLE

Morphometric study of the femoral intercondylar notch of knees with and without injuries of anterior cruciate ligament (A.C.L.), by the use of software in digitalized radiographic images

Rita di Cássia de Oliveira Angelo^I; Sílvia Regina Arruda de Moraes^II; Luciano Carvalho Suruagy ^III; TetsuoTashiro^IV; Helena Medeiros Costa^V

^IMaster Graduate in Morphology at Anatomy Department of UFPE

^IIAssociated Professor in Morphology of Anatomy Department of UFPE

^IIIDoctor of Arthroscopy Service and Knee Surgery of CMH/PMPE

^IVAssistant Professor of Physical Education Department of UFPE

^VPhysiotherapist

^{Correspondence}

SUMMARY

The authors suggest the use of software in digitalized radiographic images to morphometric analysis of the intercondylar notch and the femoral condyles. The software allows the draw of guide lines which facilitate the measurement of the distal extremity of femur. Thirty-nine radiographic simple has been analyzed of femoral intercondylar notch of right and left knees, of male sex individuals collected into normal (n=23) and injured (n=16) groups. The age average was 26-56 years old.The analyzed variable had been femoral bicondylar width, widths of intercondylar notch to the level of the popliteal groove and the notch base, intercondylar notch height and the width of lateral femoral condyle. The intercondylar notch was classified according to its shape into conical , circular and rectangular. The results suggest that the base width of the notch and the width of the femoral condyle would be important risk factors of the injury of A.C.L. The average range of analyzed variables are near to the ones described in the specialized literature in direct measurements in corpse pieces and magnetic resonance imaging and demonstrate that the use of a software in digitalized radiographic images provides a realiable measurement, even if simple and low cost radiographic images are used.

Key words: Femoral intercondylar notch; Anterior cruciat ligament; Morphology; Knee.

INTRODUCTION

The acute tear of anterior cruciate ligament (A.C.L.) is common in sports requiring frequent that rotational movements, rough stops or jumps. It is a belief that the morphology of the femoral intercondylar notch can predispose the injuries of A.C.L., and its morphometry analysis could provide important data to be used in the prevention and prognostic of these injuries⁽¹³⁾.

The intercondylar notch has already been analyzed by computed tomography (C.T.)⁽⁶⁾, by magnetic resonance imaging (M.R.I.)^(1,7), by film radiography ^{(2,9,7,12,13,14)} and, also by the direct measurement in corpses^{(3,4,10,11,15)}.

In 1937, Holmblad⁽⁸⁾ considered a radiographic view for better visualization of femoral intercondylar notch where notch image was gotten with 75º knees flexed, the tibia approximately parallel on examination table and the X-rays blister centered in popliteal cavum, perpendicularly to the tibial axis. Other authors^(9,13,14) ,equally using Holmblad method (1937)^{(8) ,} had carried out intercondylar notch radiographic images studies, however, the measurement of the variables was manual and directly on the radiographic film.

This study proposes using a software for analysis of the morphometric and morphologic aspects in radiographic images of femoral intercondylar notch and the lateral femoral condyle, and suggest a draw method of guide lines that facilitate the measurement of variables as width and height of femoral intercondylar notch, intercondylar notch base width and lateral femoral condyle, femoral bicondylar width and the Intercondyle Width Index, providing more accurate data in order to trace an anatomic profile of femoral intercondylar notch and femoral condyles of young adults of both sex, in active age group with and without injured of A.C.L.

MATERIALS AND METHODS

Thirty-nine radiographic studies have been analyzed of the femoral intercondylar notch of left and right knees, of 39 male sex individuals, in 14 — 45 age range, collected into two groups: 46 normal knees of 23 individuals with no prior pathologies of inferior members (group 1) and 32 knees of 16 individuals with a diagnostic of unilateral acute injury of A.C.L. (group 2). The data have been sellected and collected at Centro Médico Hospitalar da Polícia Militar de Pernambuco. The volunteers signed a commitment term to give the radiographic images for research. The present study have been aproved in according to the resolution nº 196/96 of Comitê de Ética em Pesquisa involving human beings of Centro de Ciências da Saúde da Universidade Federal de Pernambuco (CEP/CCS/UFPE).

To obtain the radiographic images of femoral intercondylar notch, has been adopted the method of Holmblad (1937)⁽⁸⁾. The individuals were positioned on the examination table in genuflected with the hands and knees over the table (Figure 1). The articular amplitude of knees was fixed in 75º ⁽⁸⁾ and determined by a clinical goniometer, any variation of legs, ankles and feet position were corrected by the researcher who was present during the collection of all images.

The radiographic images were digitalized by the scanner TCÊ MK^® 600u and the software MGI Photo Suite SE^®, that allowed the draw of reference lines with bone sharp contours. For a better visualization of notch limits, various clicks of the mouse were made to mark the area with 10 to 20 points. Afterwards, the variables were measured by the software KEYCAD complete version 1.0^© (1993).

The variables analyzed were: femoral bicondylar width (fbw), femoral intercondylar notch width (inw), Intercondyle Width Index (I.W.I.)⁽¹⁴⁾, femoral intercondylar notch base width (B); lateral femoral condyle width (lcw) and the notch height (nh). To measure the variables, reference lines were drawn (Figure 2): Line A, horizontally traced through the popliteal groove, to measure bicondylar distance, intercondylar notch width and lateral femoral condyle; Line B, tangent to the articular ventral surface of femoral condyles; reference to the measurement of intercondylar notch height; Line C, join the more distal vertex of intercondylar notch faces, determines the value of intercondylar notch base; Line D, perpendicular to line B, towards the intercondylar notch roof, determines the notch height (distance between the roof and line C). The I.W.I. was calculated by the division between the femoral intercondyle notch width and femoral bicondylar width⁽¹⁴⁾ to line A level.

The obtained values were compared between groups 1 (normal) and 2 (injured) as for the laterality and, afterwards, submitted to statistical analysis. The Student t test was used in non coupled parametrical samples comparision to correlate the data as laterality and knee clinical condition, with and without A.C.L. injury. In all the carried out tests, 5% significance level was adopted (pd"0,05). The femoral intercondylar notch shape was classified according to criteria adopted by Ellera Gomes; Scarton (1998)⁽⁸⁾ in conical, circular and rectangular. The Sign Test was used in the final phase of the study to observe the likeness frequency of femoral intercondylar notch shape between right and left knees of group 1 and between right and left knees of group 2.

RESULTS

The correlation of average values obtained by the measurement of variables of both knees of group 1 (without injury of A.C.L.) and group 2 (injured A.C.L.) are shown on tables 1, 2, 3, 4, 5 and 6. The comparative analysis of obtained data did not show significant statistically differences (p £ 0,05). Concerning about the femoral intercondylar notch shape, 78,26% of group 1 knees were considered conical (Figure 3), 17,4% circular (Figure 4) and 4,34% rectangular (Figure 5). In group 2, 56,25% showed the conical femoral intercondylar notch, 31,25% circular and 12,5% rectangular (Figures 6,7).

DISCUSSION

The most common and accessible method to measure the femoral intercondylar notch have been radiography⁽⁹⁾. Additional studies using radiographic images, has been linking femoral intercondylar notch morphologic and morphometry to A.C.L. injuries^{(2,9,12,13,14)} The radiographies can offer useful information in abnormalities detection, admiting that the error limit will be bigger and that measurements will not be as precise as those obtained in multiple planes section M.R.I.⁽⁷⁾ studies.

The use of a software for the measurement, aims a more reliable result, decreasing likely distortions due to make the measurement directly on the radiographic film with a millimeter rule, used method by some authors^(9,13,14). The drawing of reference lines, at this study, was based on the methods considered by Herzog et al. (1994)⁽⁷⁾ and Ellera Gomes and Scarton (1998)⁽²⁾ and adapted for computerized measurement through KEYCAD^® software. Tillman et al. 2002⁽¹⁵⁾ have been analyzed digital photos of femoral intercondylar notch, by the software MATLAB 5,2 (MATHWORKS, INC., NATICK, MILIAMPÈRE, EUA) in corpse study; the drawing of reference lines with a mouse and the femoral condyles contours outlines with a series of points (least of 10) is similar to our method.

However, we agree to Ellera Gomes and Scarton⁽²⁾ when they affirm that there is not an appropriate reference, uniformly accepted in literature for the measurement of femoral intercondylar notch.

In this study, during the radiographic examination, a 75º knee flexion was established, considering a 0º extension movement amplitude. The 70º to 75º knee flexion provides a bigger relaxation of joint structures, increasing the joint space between tibia and femur, supplying a better intercondylar notch visualization⁽⁸⁾ .

Comparative analysis of femoral bicondylar width referring data showed similarity between normal and injured knees. Our results differ from some authors, who describe superior values for bicondylar width in knees without A.C.L. injury⁽⁵⁾. Our group 1 findings are similar to the "maximum femoral condylar width" avarage value found by Herzog et al. (1994)⁽⁷⁾ (82,00mm) in radiographic measurements of corpse femoral intercondylar notches and with inferior to Mall et al.⁽¹¹⁾ (84,00mm) data in direct measurement of corpses. In revised literature, the majority of the authors, however, do not correlate absolute values of this variable to the ligament injury probability, the frequent use of bicondylar width is observed in the Intercondyle Width Index calculus (I.W.I.)^(1,13,14,15) and in sex determination^{(3,11) .} Studies suggest that the femoral bicondylar width would increase disproportionatily more than the width notch, according with the individual height variation^{(1,12) .} and corporal weight^{(12) .} The discrepancy between data, therefore, is probably due to our constitutional morphologic characteristics sample, since our anthropometric standards considerably vary between different populations^(11).

Our results have not shown difference statistically significant among the femoral intercondylar notch width of analyzed groups.The avarage values presented by Herzog et al.⁽⁷⁾ (23,9mm) in radiographic study and by Anderson et al.⁽¹⁾ (23,7mm) in M.R.I.study, are equivalent to our result in the left side of group 1. The values of left side of both groups are kindred, in accordance with some authors which describes similar values in normal knees and knees with acute injuries of A.C.L.^(6,7). Such similarity can be probably explained by the limited sample of this study that did not allow to recognize small differences in the width notch, or by osteophites non formation as result of joint instability in injured knees, since group 2 was constituted by individuals with acute ligament injuries, or still because there is no morphometrical differences between normal knees and acute ligament knees injuries. Studies had compared individuals without injury and with A.C.L. uni and bilateral injury and had found estenosis of femoral intercondylar notch in bilateral ligament knee injury. According to authors, patients with bilateral injuries are young, present relative estenosis of femoral intercondylar notch in the radiographic images and the initial injury generally occurs without contact⁽¹³⁾. It is important to point out that injuried group in this study was constituted by individuals with unilateral injuries.

According to some authors the estenosis of anterior part of femoral intercondylar notch in A.C.L. knees injury occurs precociously due to formation of ostheophites, probably result of the secondary instability to injury; however, the precise beginning of measurable estenosis could not be determined⁽⁴⁾. However, studies state that the prevalence of lesional mechanism without contact suggests that the A.C.L. injury risk is intrinsic for nature and that congenital estenosis of femoral intercondylar notch can be considered as one of the intrinsic risk factors in the AC.L. injury ^(5,13,14)

In accordance with the literature^(6,7,14), we have observed similar values in I.W.I. of both groups, in particular on left side. The average value found in right knees of normal group is equivalent to values found in M.R.I. studies^(1,7). The avarage value observed in the injuried group is similar to the one described in disabled A .C.L. knees (0,28) by M.R.I. study^(7). Our values from normal group are similar to those found in african men in corpse direct measurement (0,26) and are not in accordance with the data observed by some authors for European descendant (0,24)⁽¹⁵⁾; that can be explained by the Brazilian miscegenation. The reliability of I.W.I. as determinant of the notch width, however, is controvertible. This measure intends to eliminate the variation of magnification when comparing radiographic measures and standardize patients with different morphologic biotypes⁽¹⁴⁾. Some authors, however, disagree with this index^(1,7,12). In order to be a valid mathematical measure of notch size, both variable of this index must vary proportionally to the individual's heights. However, some studies suggest that the notch width is not directly proportional to the height, though the femoral bicondylar width increases with the addition of stature⁽¹²⁾.

The comparative analysis of the base width of intercondylar notch between both sides of groups 1 and 2 showed higher values for the normal group, in agreement with literature findings⁽²⁾ which relates a significant distal narrowness in unstable knees occuring in the order of 0,0446mm/month after injury. In corpse study, authors demonstrated that the intercondylar notch is more distally strait and more proximally wide⁽¹⁰⁾, leading to question if the base of stenotic notch would not be a more important risk factor of A.C.L. injury than its width in the level of popliteal groove.

Palmer apud Shelbourn and Kerr⁽¹²⁾ was the first to correlate the femoral intercondylar notch proportions and A.C.L. injury prediposition. Observed that A.C.L. gets constitucionally a vulnerable position, being retaining over the lateral femoral condyle internal face during knees rotational movements, facilitating its rupture. Our results at this study, in spite of statistically non significant, show clearly a lateral femoral condyle wider in group with A.C.L. injury Our values are consistent with the ones found by some authors in direct measurement of corpse pieces (29,6mm)^{(4) .}In tomography study, authors observed that patients with bilateral injuries of A.C.L. had the lateral femoral condyle wider than individuals without injury and concluded that the width of lateral femoral condyle can interfere in the notch width and constitute an intrinsic risk factor to the injury of that ligament⁽⁶⁾. However, the present study do not prove such hypothesis, since the femoral intercondylar notch width did not show decresead value in injured knees originated from the wider lateral femoral condyle. It is important to say that the sample was filled with unilateral A.C.L. injuries individuals.

In spite of statistically non significant, our data demonstrate that the intercondylar notch height of normal group was higher than the injured knees group. This argument is consistent with the results of some authors⁽⁷⁾, however, our values are numerically higher. The numerical disparity can be explained by the ethnic heterogeneity of the population studied sample, characterized by the intense Brazilian miscegenation. In corpse study, the notch height was measured separating the samples by racial groups, the authors have concluded that black people have the intercondylar notch height higher than white people; the reason of this morphological variation is an anterior curvature of femoral diaphysis lower in black population. Theoretically, the lower femoral curvature causes an upward rotation of femoral condyles, producing a higher notch⁽³⁾.

In this study the conical shape was predominant in both groups, with higher percentages in normal group in both sides. The higher percentage of conical shape in injured knees agrees to the literature findings for anterolateral instability knees (54,4%)⁽²⁾. We observed a discrete high percentage of circular shape in the injured knees group, independent of laterality, in relation to knees of normal group as to circular and rectangular shapes. In circular shape the femoral intercondylar notch base width is lesser than its apex; it is pobable that a narrower base acts an intrinsic risk factor for A.C.L.injury. The difference of the femoral intercondylar notch shape can partially explain the probability of circular A.C.L. injury⁽¹⁵⁾. This would explain a bigger percentage of circular notch in the injured group in relation to the rectangular shape and the percentages of circular and rectangular shapes found in right and left knees of the normal group.

The percentage values of shape equality of femoral intercondylar notch, determined by the Sign Test, show higher percentages of different shapes of intercondylar notch of injured individuals group. It is probable that such finding is related to presence of factors as height, corporal weight, member dominance in relation to the other one, that can influence in predominance of laterality in ligament injuries, however we do not find data in revised literature to oppose such findings.

CONCLUSION

The results presented in this study did not presented significant statistical differences between knees without ligament injury and knees with unilateral acute A.C.L. injury; such finding can be related to a small population sample, that did not allow the observation of peculiar differences between the analyzed group. However we can not discard the hypothesis not to have difference between normal knees and knees with recent ligamentar injuries. Estenosis of femoral intercondylar notch is praised as risk factor for A.C.L. injury, however we believe there is necessity further investigation in order to establish the real influence of intercondylar notch estenosis on this ligament, if it is determined congenitally or if it occurs after the ligament injury.

It seems to us that theintercondylar base width and lateral femoral condyle width would be more important risk factors in A.C.L. injury than the intercondylar width to popliteal groove level, and that such variable can influence in notch format, what would explain the differentiated behavior of circular shapes in injuried knees. It is necessary higher verification of such finding in order to correlate the notch shape to A.C.L. probability of rupture.

As to the used measurement method, the results found in this study, very near to the ones found in the specialized literature, demonstrate that the use of a software in digitalized radiographic images can provide a practical and accurate measurement, even though, the used radiographic images were simple and low cost, in accordance with the reality of Brazilian health service.

REFERÊNCIAS BIBLIOGRÁFICAS

Trabalho recebido em 29/04/2003.

Aprovado em 20/06/2004.

Work performed at the Anatomy Department of the Federal University of Pernambuco (UFPE)

• 1. Anderson AF, Dome DC, Gautam S, Awh MH, Rennirt GW. Correlation of anthropometric measurements, strength, anterior cruciate ligament size, and intercondylar notch characteristics to sex differences in anterior cruciate ligament tear rates. Am J Sports Med 29:58-66, 2001.
• 2. Ellera Gomes JL, Scarton A. Planificação radiológica da abertura intercondiliana na reconstrução do ligamento anterior. Rev Bras Ortop 33:651-654, 1998.
• 3. Gill GW. Racial variation in the proximal and distal femur: heritability and forensic utility. J Forensic Sci 46:791-799, 2001.
• 4. Good L, Odensten M, Gillquist J. Intercondylar notch measurements with special reference to anterior cruciate ligament surgery. Clin Orthop 263:184-189, 1991.
• 5. Harner CD, Paulos LE, Greenwald AE, Rosenberg TD, Cooley VC. Detailed analysis of patients with bilateral anterior cruciate ligament injuries. Am J Sports Med 22:37-43, 1994.
• 6. Herzog RJ, Siliman JF, Hutton K, Rodkey WG, Steadman JR. Measurements of the intercondylar notch by plain film radiography and magnetic resonance imaging. Am J Sports Med 22:204-210, 1994.
• 7. Holmblad EC. Postero-anterior x-ray view of knee in flexion. JAMA 9:1196-1197, 1937.
• 8. Ireland ML, Ballantyne BT, Little K, McClay IS. A radiographic analysis of the relationship between the size and shape of the intercondylar notch and anterior cruciate ligament injury. Knee Surg Sports Traumatol Arthrosc 9:200-205, 2001.
• 9. Koukoubis TD, Glisson RR, Bolognesi M, Vail TP. Dimensions of the intercondylar notch of the knee. Am J Knee Surg 10:83-87, 1997.
• 10. Mall G, Graw M, Gehring KD, Hubig M. Determination of sex from femora. Forensic Sci Int 113:315-321, 2000.
• 11. Palmer I. On the injuries to the ligaments of the knee joint. A clinical study. Acta Chir Scand 81: 1-282, 1938.
• 12. Shelbourne KD, Kerr B. The relationship of femoral intercondylar notch width to height, weight, and sex in patients with intact anterior cruciate ligaments. Am J Knee Surg 14:92-96, 2001.
• 13. Souryal TO, Freeman TR. Intercondylar notch size and anterior cruciate ligament injuries in athletes: a prospective study. Am J Sports Med 21:535-539, 1993.
• 14. Souryal TO, Moore HA, Evans J.P. Bilaterality in anterior cruciate ligament injuries: associated intercondylar notch stenosis. Am J Sports Med 16:449-454, 1988.
• 15. Tillman MD, Smith KR, Bauer JA, Cauraugh JH, Falsetti AB. Differences in three intercondylar notch geometry indices between males and females: a cadaver study. Knee 9:41- 46, 2002.

Publication Dates

• Publication in this collection
  16 Nov 2004
• Date of issue
  Sept 2004

History

• Accepted
  20 June 2004
• Received
  29 Apr 2003