- Email: [email protected]
Anisotropy of human linea alba: A biomechanical study
Journal of Surgical Research 124, 118 –125 (2005) doi:10.1016/j.jss.2004.10.010
Anisotropy of Human Linea Alba: A Biomechanical Study David Gräel,*,† Andreas Prescher, M.D.,† Sabine Fitzek, M.D.,‡ Diedrich Graf v. Keyserlingk, M.D.,† and Hubertus Axer, M.D.‡,1 *Institute of Diagnostic and Interventional Radiology, Friedrich-Schiller-University Jena, Erlanger Allee 101, D-07747 Jena, Germany; †Department of Anatomy I, RWTH Aachen, Pauwelsstr. 30, D-52057 Aachen, Germany; and ‡Department of Neurology, Friedrich-Schiller-University Jena, Erlanger Allee 101, D-07747 Jena, Germany Submitted for publication July 18, 2004
Background. Recently, a new model of fiber architecture of the linea alba has been described consisting of an oblique fiber layer of intermingling oblique fibers, a transverse fiber layer containing mainly transverse fibril bundles, and a variable, small irregular fiber layer. In this study the morphological model was proven using direction-specific biomechanical measurements of the linea alba. Material and methods. Thirty-one human abdominal walls were analyzed (16 male and 15 female). Six strips of collagen tissue with a width of 1 cm were exsected from each linea alba transversely, obliquely, and longitudinally according to the main fiber directions. An increasing force from 2 to 24 N was applied to these strips, and the corresponding strain represented by the relative elongation was measured, which allows the calculation of a direction-specific compliance of the tissue. Results. The compliance is highest in longitudinal and smallest in transverse direction. In the infraumbilical part of the female linea alba the compliance was significantly smaller in the transverse direction than in the oblique direction. Moreover, the compliance in the transverse direction was significantly smaller in women than in men. Conclusions. A distinct anisotropy of morphological and biomechanical properties was demonstrated as well as sex-dependent differences. The compliance correlates with the distribution of fiber orientation in the linea alba. These biomechanical results constitute the functional correlation with the fiber morphology
1
To whom correspondence and reprint requests should be addressed at Department of Neurology, Friedrich-Schiller-University Jena, Erlanger Allee 101, D-07747 Jena, Germany. E-mail: [email protected].
0022-4804/05 $30.00 © 2005 Elsevier Inc. All rights reserved.
of the linea alba and correspond well to our earlier proposed model of fiber architecture. © 2005 Elsevier Inc. All rights reserved.
Key Words: linea alba; anisotropy; collagen fiber architecture; biomechanics; gender differences. INTRODUCTION
The anterior abdominal wall plays a significant role in surgery since it is a common way to open the peritoneal cavity [1]. It consists of a muscular part and a collagenous part. The oblique external, the oblique internal, and the transverse abdominal muscles constitute the ventrolateral abdominal muscles, while the rectus abdominis and the pyramidal muscles form the ventroanterior muscular part. The collagenous parts in the abdominal midline contain the linea alba and the rectus sheaths. A disturbance of these stabilizing parts of the abdominal wall leads to hernia. The collagenous parts of the ventral abdominal wall are the linea alba and the rectus sheaths. The linea alba is composed of the tendon fibers of the abdominal muscles, thus playing a most significant role in stabilization of the abdominal wall. The architecture of collagen fibers in linea alba has been investigated in a previous study [2, 3]. The result was a new anatomical model of fiber architecture of the linea alba considering the anisotropic order of collagen fibers. Anisotropy means that the tissue exhibits properties with different values when measured in different directions. In short, using confocal laser scanning microscopy the following architectural pattern of the collagen fibers was described [2]. The collagen fibers in linea alba show the same orientation as the muscle fibers of the ventrolateral abdominal wall: transverse, oblique I (directed from right upward to left downward), and oblique II (directed from left upward to right down-
118
¨ EL ET AL.: BIOMECHANICS OF LINEA ALBA GRA
119
oblique and crossed the midline. Thus the linea alba was considered as a line of decussation of fibers. Two patterns of decussation were described: the single and triple pattern of decussation. The close relationship between morphology and function of collagenous tissue [6, 7] inspired us to reinvestigate the linea alba from a biomechanical point of view, this being most important in hernial sac formation [8 –10] and for the development of implantable meshes for abdominal wall repair [11, 12]. The intention of the present study was (1) to determine directional compliances of the linea alba and correlate the biomechanic anisotropy, and (2) to define any sex differences in biomechanic anisotropy of the linea alba. MATERIALS AND METHODS Preparation of the Specimen
FIG. 1. Anatomical model of collagen fiber architecture in the linea alba. The architectural zones are described from ventral to dorsal: (1) oblique fiber layer, (2) transverse fiber layer, and (3) irregular fiber layer.
ward). In linea alba the following three different zones of fiber orientation follow each other from ventral to dorsal (Fig. 1):
Thirty-one fresh abdominal walls were derived from persons (15 female, 16 male, age: 63 to 95 years) who donated their bodies for anatomical study. The abdominal walls were free of scars or any other pathologies. Skin and subcutaneous tissue were sharply dissected. The rectus abdominis muscles were carefully removed from the rectus sheaths and linea alba without injuring the collagen structure of the tissue. The remaining tissue, consisting of the rectus sheaths and the linea alba, was cut transversely at the level of the umbilicus, to distinguish the supraumbilical from the infraumbilical part of the specimen. Three strips were cut out of each part. The three single strips were oriented in longitudinal, transverse, and oblique direction with respect to linea alba (Fig. 2) in randomly changing order. The lines of intersection were angled by means of the eye. Concerning the oblique ones, we tried to follow the trend of the fibers if visible. Otherwise we approximated angles between 40 and 50 degrees with respect to linea alba. Specimens were not obtained in all directions from all donors. Parts of the linea alba had adhesions to the bowels which could not be properly dissected, and others had holes. Those parts were deter-
1. The oblique fiber layer consists of intermingling oblique fibers (in average four to six layers of fibers). 2. The transverse fiber layer contains mainly transverse fibril bundles (in average four to six layers of fibers). 3. An inconstant, small irregular fiber layer is composed of one to two layers of oblique fibers. Moreover, gender differences were demonstrated in the relation of oblique and transverse fibers [3]. A larger amount of transverse fibers relative to the oblique fibers was found in females than in males (60% versus 37.5%). Different regions can be distinguished in the craniocaudal course of the linea alba: supraumbilical part, umbilical part, transition zone, and infraarcuate part, but the general structure of fibers is the same in all four regions. This model of fiber architecture is in contrast with findings proposed by Askar [4] and Rizk [5] describing six aponeurotic layers in linea alba, which all were
FIG. 2. Orientation of the tissue strips in relation to the linea alba.
120
JOURNAL OF SURGICAL RESEARCH: VOL. 124, NO. 1, MARCH 2005 to the loading force per width of the tissue strip. Using the Law of Laplace, Klinge et al. [11] demonstrated the physiological stress of the abdominal wall not to exceed 16 N/cm. (The Law of Laplace states the proportionality of pressure and compliance of a flexible sphere for constant radii of curvature.) At each step, the positions of the needles were recorded using a digital steadycam mounted on the frame. The strain rate was variable during the experiment. Elongation alternated with the taking of pictures. One step over 2 N took about 10 s for both actions. We performed frequent tests of viscous behavior of the tissue. At maximum stress (25 N/cm) we recorded some extra values of absolute elongation with a time delay of 1 min per measurement. These measurements resulted in a negligible elongation even after 2 min. This time corresponds to the distance of the whole process of elongation during the actual measurement of one strip (12 steps times 10 s).
Data Acquisition and Analysis
FIG. 3.
(A) Test device. (B) Specimen fixed to the clamps.
mined to be useless for further examination. A few strips slipped out of the clamps of the test device and were excluded from analysis. Thus, we analyzed 58 different samples in the transverse direction (females: 14 supraumbilical and 13 infraumbilical samples; males: 16 supraumbilical and 15 infraumbilical samples), 55 in oblique direction (females: 11 supraumbilical and 14 infraumbilical samples; males: 15 supraumbilical and 15 infraumbilical samples), and 52 in longitudinal direction (females: 12 supraumbilical and 13 infraumbilical samples; males: 14 supraumbilical and 13 infraumbilical samples). The width of the strips was strictly kept to 1 cm on the relaxed condition. The specimens were kept moist in a buffered Ringer solution as used earlier in other biomechanical studies [13]. Biomechanical measurements were performed at constant room temperature (21°C).
Test Device and Procedure of Measurement The biomechanical test device consists of an octagonal frame with eight concentric arms (Fig. 3A). Each arm is equipped with a crank carrying a spring scale with a clamp fixed on its end. The clamps were used to link the strips to the frame via a spring scale. To avoid slipping during the test, the strips were covered with emery paper before they were jammed in the clamps (Fig. 3B). Afterwards the strips were marked with two needles at the lateral edges of the linea alba, where the double layer of rectus sheaths emerges. The effective length of the tested tissue strip is the distance between the two needles and we supposed to have constant mechanical properties of the tissue between them. The part of the rectus sheath beyond the needles, where the tissue was clamped, was of no interest. The distance between the needles varied depending on the width of linea alba and the direction of the cut. During the test the loading force was increased from 2 to 24 N in steps of 2 N, implying stresses from 2 to 24 N/cm. In this context the term stress is related
To acquire data about the dependence of the relative elongation of the strips on the loading force, the distance between the two needles was measured. The digitally recorded pictures were evaluated by taking the coordinates at the center of the needles’ heads and calculating their distance in units of pixels. The correlation between absolute distance and loading force reveals an interval of approximate linearity between 6 and 16 N. The strong increase of distance in the interval between 2 and 6 N is artificial since it is caused by the elevation of the sagging strip. By means of linear regression, the slope of the absolute distance versus loading force was calculated in this interval. This way we obtained the original distance in the relaxed condition by extrapolating the line of regression to zero Newton. The strain is represented by the relative elongation which results from the absolute distance in relation to the originating distance at zero Newton. The usage of relative elongation as measurement of strain instead of the absolute elongation allows a comparison of the specimens despite different needle distances. The specimens were divided into the supraumbilical and infraumbilical part of the linea alba and were grouped according to the different orientations (transverse, oblique, and longitudinal, Fig. 2). The strain was plotted versus the loading force in each direction, separately for male and female specimens. The slope of the curve within the interval between 6 and 16 N is described by the parameter ␣, which connects the strain with the loading force via the following formula, derived from the law of Hooke:
⫽
ᐉ ⫺ ᐉ0 ᐉ0
⫽␣
F b
where F is the loading force, b is the width of the tissue strip (strictly kept to 1 cm on relaxed condition), l is the distance between the needles’ heads, l 0 is the extrapolated originating distance at zero Newton, and is the relative elongation respectively strain of the tissue strip. The parameter ␣ is not exactly the reciprocal of Young’s modulus respectively modulus of elasticity, because it is not independent of the thickness of the clamped tissue strip. The parameter ␣ represents the compliance per unit thickness and its value is quantified in units of cm/N. Due to its biological spread, the thickness is a main source for variations in the value of ␣. For statistical testing the unpaired Student’s t test was used with two-sided significance levels of P ⫽ 0.05.
RESULTS
The relative elongation was plotted versus the stress in N/cm in each direction, separately for male and female specimens (Fig. 4). The values of the compliance per unit thickness were calculated from these graphs,
¨ EL ET AL.: BIOMECHANICS OF LINEA ALBA GRA
FIG. 4.
121
Strain plotted versus stress, shown separately for the different directions of force, region of linea alba, and gender.
and Fig. 5 shows the compliance per unit thickness derived from the single measurements. The standard deviations of the longitudinal compliance were much larger than the standard deviations of the compliance in transverse and oblique directions. For both sexes the compliance is highest in longitudinal direction. These values differ significantly from the measurements in transverse and oblique direction in most cases (Table 1). The compliance in transverse direction was smallest in both sexes but the transverse compliance was significantly smaller than the oblique one in the infraumbilical part of the female linea alba (Table 1, Fig. 6). Table 2 shows the sex-dependent differences of compliance per unit thickness in the different directions of tensile force. In the supra- and infraumbilical region the compliance per unit thickness in transverse direction was significantly smaller in women than in men. In an earlier study [2, 3] the directions of the collagen fiber bundles in the linea alba were determined with confocal laser scanning microscopy. In the male, it was determined that 44.9% of the fibril bundles were transverse in the supraumbilical part, while 55.1% of the fibers were oblique. In the male infraumbilical part 37.5% of the fibers were transverse and 62.5% were oblique. In contrast, in females 47.6% of the fibers were transverse in the supraumbilical part, while 52.4%
were oblique. In the female infraumbilical part 60.4% of the fibers were found to be transverse, while 39.6% were oblique. There was a significant sex difference between the infraumbilical transverse fiber directions. Figure 7 shows the compliance per unit thickness in comparison with the distribution of fiber direction in the linea alba. Both parameters are correlated significantly to each other (Pearson correlation coefficient: ⫺0.919, P ⬍ 0.001). DISCUSSION
This study focuses on the biomechanical properties of the linea alba in relation to its collagen fiber architecture. It was clearly demonstrated that the linea alba shows a distinct anisotropy in its biomechanical properties with highest compliance in longitudinal direction and lowest in transverse direction. These results conform to the proposed properties of the fiber model described earlier [2]. The strain was plotted versus the corresponding stress (loading force) in each direction. Above 16 N the slopes become flatter and the graphs seem to lead into saturation (Fig. 4). This behavior seems to be due to the mechanical properties of the scissors-like lattice structure of the collagen fibers. Table 2 shows the compliance per unit thickness in
122
JOURNAL OF SURGICAL RESEARCH: VOL. 124, NO. 1, MARCH 2005
demonstrates that the more fibers that are located in a special direction, the lower is the compliance. If the amounts of fibers in two directions differ too much, the differences in compliance become significant. The low value of the compliance in transverse direction clearly contradicts the fiber model described by Askar [4] as this model did not at all describe fibers running transversely to the linea alba. In this case the value of compliance in the transverse direction would have been larger than in the oblique direction. The high amount of transverse fibers enhances the stiffness of linea alba in the transverse direction. Interestingly the well-known Langer lines, which describe the direction of skin incision producing the least tension of the skin in a specific anatomical region, are also oriented transversely to the linea alba [14]. However, biomechanical measurements in this study were performed only on collagen tissue of linea alba without the skin, which was removed from the specimens before. Moreover, sex-dependent differences were demonstrated. The compliance per unit thickness in the transverse direction was significantly smaller in the investigated women than in men. This again corresponds well to the sex differences found in the morphological fiber architecture of the linea alba [3]. Here, it was demonstrated that the female linea alba has more transverse fibers in relation to the oblique fibers than the male one. Moreover, this finding was most distinct in the infraumbilical part. In this study the compliance in the transverse direction is significantly different from that in the oblique direction in the infraumbilical part of the female linea alba (Table 1). It was proposed that these morphological differences could be due to the elevated intraabdominal TABLE 1
FIG. 5. Values of the measured compliance per unit thickness in men and women. Weighted mean values and standard errors of the mean are shown at the top of the diagrams and visualized as gray boxes.
comparison with the distribution of fiber direction in the linea alba. Moreover, both parameters are correlated significantly with each other (Pearson correlation coefficient: ⫺0.919, P ⬍ 0.001). The information about the distribution of fiber direction was derived from an earlier morphological study analyzing the fiber architecture of linea alba [3]. Using confocal laser scanning microscopy, the diameter of each layer of fibril bundles was measured in linea alba of 12 human cadavers, and each fibril bundle was classified according to its orientation (oblique I and II, transverse). Correlating the biomechanical results of this study with the morphological properties of the linea alba
Comparison of Compliance per Unit Thickness Related to the Direction of Force Gender Male
Region
Comparison
P
df
Supraumbilical
Transverse–oblique Longitudinal–transverse Longitudinal–oblique Transverse–oblique Longitudinal–transverse Longitudinal–oblique Transverse–oblique Longitudinal–transverse Longitudinal–oblique Transverse–oblique Longitudinal–transverse Longitudinal–oblique
n.s. 0.002 0.001 n.s. n.s. 0.022 n.s. 0.035 0.006 0.018 n.s. 0.033
29 16 15 28 21 15 13 16 12 20 17 13
Infraumbilical
Female
Supraumbilical
Infraumbilical
Note. The table shows the significance of the difference between the values of mean compliance per unit thickness obtained from samples oriented in different directions. Specimens were not obtained in all directions from all donors; therefore, the degrees of freedom differ. n.s. ⫽ not significant; df ⫽ degrees of freedom.
¨ EL ET AL.: BIOMECHANICS OF LINEA ALBA GRA
FIG. 6. Values of the mean elasticity coefficients in women. Weighted mean values and standard errors of the mean are shown at the top of the diagrams and visualized as gray boxes. Histograms of the values are plotted at the bottom of each diagram.
pressure during pregnancy as the lattice of the oblique fibers would be stressed and deformed so that most of the fibers would run in the transverse direction [3]. Indeed, most of the transverse fibers were found in the infraumbilical linea alba and in this study the compliance per unit thickness in transverse direction becomes significantly different from that in the oblique direction in the infraumbilical part. The compliance of the maternal abdominal wall (which is reciprocal to its stiffness) was found to be inversely related to gestational age [15], a finding which could also be related to such a mechanism. However, this hypothesis has yet to be proven in detail since no information about pregnancies in the history of the body donors of this study is available. Rath et al. [16] performed biomechanical measurements on human abdominal walls. They did not find differences in age, sex, or morphotype. However, they only investigated the compliance of the tissue in the
123
transverse direction of the linea alba, and they were interested in the biomechanical properties of the whole abdominal wall. It must be kept in mind that the specimens of this study are relatively old, as they represent a normal distribution of age of body donors in an anatomical department. Thus, care has to be taken to generalize the presented results to younger populations. However, the relationship between function and morphology of the connective tissue has been clearly demonstrated in this study. Overall, a distinct anistropy of morphological and biomechanical properties was demonstrated as well as gender-dependent differences. These biomechanical results are the functional correlates of the fiber morphology of the linea alba. The abdominal wall is an important structure serving many different functions. The two major functions are movements of the trunk [17] and regulation of intraabdominal pressure [18]. Moreover, it supports respiration [19] and plays a role in stabilization of the spine [20, 21]. All these functions are facilitated by the coordinated and task-specific activation pattern of the abdominal muscles [22]. However, electromyographic observations clearly demonstrated the transversus muscle to be most significant in intraabdominal pressure production [18, 22] and to be recruited preferentially in breathing [19]. In contrast, the internal and external oblique muscles are more effective in contributing to trunk movements [23, 24] and stabilization of the spine [25]. The linea alba has to be seen as the decussation of the tendon fibers of the abdominal wall muscles. The three-dimensional fiber architecture of the linea alba shows the functional anatomical entity of this structure as it acts as the midline insertion of the abdominal wall muscles. Thus, it has to perform the stabilization of the abdominal wall while the muscles resemble the dynamic elements. The abdominal wall (and the linea alba as the collagenous part of it) is a complex dynamic system which performs multiple interactions within this framework of bones, muscles, and collagen fibers. From a surgical point of view the anterior abdominal wall is the object of investigations regarding wound healing [26] and wound closure [27, 28]. Midline ventral abdominal wall hernias include epigastric, umbilical, paraumbilical, and also incisional hernias [9, 10, 29]. The structure of the collagen fibers may be of significance in the choice of direction of laparotomy incision. It was demonstrated that transverse laparotomy incisions are more resistant to rupture than longitudinal incisions [26, 30, 31]. In fact, in transverse laparotomy the transverse fibers in linea alba and rectus sheaths are not cut and can serve their function, while in lon-
124
JOURNAL OF SURGICAL RESEARCH: VOL. 124, NO. 1, MARCH 2005
TABLE 2 Comparison of Compliance per Unit Thickness (␣, in cm/N) with Relative Distribution of Fiber Orientation in Linea Alba Related to Gender Region Supraumbilical
Infraumbilical
Male ␣
Female ␣
P
df
Male distribution
Female distribution
Transverse Oblique
0.67 0.70
0.50 0.73
0.014 n.s.
28 13
Longitudinal Transverse Oblique
1.28 0.73 0.78
1.42 0.55 0.86
n.s. 0.017 n.s.
17 26 27
Longitudinal
1.07
1.14
n.s.
18
44.9% I: 25.5% II: 29.6% 0% 37.5% I: 31.6% II: 30.9% 0%
47.6% I: 22.9% II: 29.5% 0% 60.4% I: 16.7% II: 22.9% 0%
Direction
Note. n.s. ⫽ not significant, df ⫽ degree of freedom, I ⫽ oblique I (directed from right upward to left downward), II ⫽ oblique II (directed from left upward to right downward).
gitudinal incisions the sutures are not fixed in between the cut transverse fibers [32, 33]. Mesh implantation to repair incisional hernias is a common approach. Because the meshes work as a mechanical closure of the defect and induce scar tissue, the biomechanical properties of the implantable meshes should reproduce the physiological biomechanics of the linea alba and the abdominal wall [12, 34]. If the demonstrated sex differences in biomechanical properties of the linea alba imply different sex-oriented surgical procedures cannot be answered from these data, then this topic could be an interesting viewpoint in the future.
ACKNOWLEDGMENTS We thank Anita Agbedor, Andre Doering, and Hans Dietz for technical assistance and Wolfgang Graulich, who drew Fig. 1.
REFERENCES 1.
2.
3.
4.
5. 6. 7.
8. 9. 10. 11.
12.
FIG. 7. Comparison of unweighted mean elasticity coefficients with relative distribution of fiber orientation in linea alba. The morphological data were derived in an earlier study [3]. I ⫽ oblique I (directed from right upward to left downward), II ⫽ oblique II (directed from left upward to right downward).
13.
14.
Prescher, A. Surgical anatomy. In V. Schumpelick, A. N. Kingsnorth (Eds.), Incisional Hernia. Berlin: Springer, 1999. Pp. 45– 60. Axer, H., Keyserlingk, D. G. V., and Prescher, A. Collagen fibers in linea alba and rectus sheaths. I. General scheme and morphological aspects. J. Surg. Res. 96: 127, 2001, doi:10.1006/ jsre.2000.6070. Axer, H., Keyserlingk, D. G. V., and Prescher, A. Collagen fibers in linea alba and rectus sheaths. II. Variability and biomechanical aspects. J. Surg. Res. 96: 239, 2001, doi:10.1006/ jsre.2000.6071. Askar, O. M. Surgical anatomy of the aponeurotic expansions of the anterior abdominal wall. Ann. R. Coll. Surg. Engl. 59: 313, 1977. Rizk, N. N. A new description of the anterior abdominal wall in man and mammals. J. Anat. 131: 373, 1980. Viidik, A. On the correlation between structure and mechanical function of soft connective tissues. Verh. Anat. Ges. 72: 75, 1978. Nilsson, T. Biomechanical studies of rabbit abdominal wall. Part II –The mechanical properties of specimens in relation to length, width, and fibre orientation. J. Biomech. 15: 131, 1982. Engin, A. E., and Akkas, N. Etiology and biomechanics of hernial sac formation. J. Biomed. Eng. 5: 329, 1983. Schumpelick, V. Hernien, 4th ed. Stuttgart: Enke, 2000. Schumpelick, V., and Kingsnorth, A. N. Incisional Hernia. Berlin: Springer, 1999. Klinge, U., Klosterhalfen, B., Muller, M., Anurov, M., Ottinger, A., and Schumpelick, V. Influence of polyglactin-coating on functional and morphological parameters of polypropylene-mesh modifications for abdominal wall repair. Biomaterials 20: 613, 1999. Klinge, U., Klosterhalfen, B., Conze, J., Limberg, W., Obolenski, B., Ottinger, A. P., and Schumpelick, V. Modified mesh for hernia repair that is adapted to the physiology of the abdominal wall. Eur. J. Surg. 164: 951, 1998. Nilsson, T. Biomechanical studies of rabbit abdominal wall. Part I. The mechanical properties of specimens from different anatomical positions. J. Biomech. 15: 123, 1982. Maurel, W. Biomechanical Models for Soft Tissue Simulation. Berlin: Springer, 1998.
¨ EL ET AL.: BIOMECHANICS OF LINEA ALBA GRA 15.
Cohen, D., Timbs, A. E., and Dalton, K. J. Measurement of compliance of the maternal abdominal wall in pregnancy. Eur. J. Obst. Gyn. Repr. Biol. 23: 267, 1986. 16. Rath, A. M., Attali, P., Dumas, J. L., Goldlust, D., Zhang, J., and Chevrel, J. P. The abdominal linea alba: an anatomoradiologic and biomechanical study. Surg. Radiol. Anat. 18: 281, 1996. 17. De Troyer, A. Mechanical role of the abdominal muscles in relation to posture. Respir. Physiol. 53: 341, 1983. 18. Cresswell, A. G. Responses of intra-abdominal pressure and abdominal muscle activity during dynamic trunk loading in man. Eur. J. Appl. Physiol. 66: 315, 1993.
25.
19.
29.
De Troyer, A., Estenne, M., Ninane, V., Van Gansbeke, D., and Gorini, M. Transversus abdominis muscle function in humans. J. Appl. Physiol. 68: 1010, 1990.
20.
Gracovetsky, S., Farfan, H., and Helleur, C. The abdominal mechanism. Spine 10: 317, 1985.
21.
Tesh, K. M., Shaw Dunn, J., and Evans, J. H. The abdominal muscles and vertebral stability. Spine 12: 501, 1987.
22.
Cresswell, A. G., Grundström, H., and Thorstensson, A. Observations on intra-abdominal pressure and patterns of abdominal intra-muscular activity in man. Acta Physiol. Scand. 144: 409, 1992.
23.
McGill, S. A revised anatomical model of the abdominal musculature for torso flexion efforts. J. Biomech. 29: 973, 1996.
24.
Torén, A., and Öberg, K. Maximum isometric trunk muscle strength and activity at trunk axial rotation during sitting. Appl. Ergonomics 30: 515, 1999.
26.
27.
28.
30. 31.
32.
33.
34.
125
Gardner-Morse, M. G., and Stokes, I. A. F. The effects of abdominal muscle coactivation on lumbar spine stability. Spine 23: 68, 1998. Nilsson, T. The relative rate of wound healing in longitudinal and transverse laparotomy incisions. Acta Chir. Scand. 148: 251, 1982. Campbell, J. A., Temple, W. J., Frank, C. B., and Huchcroft, S. A. A biomechanical study of suture pullout in linea alba. Surgery 106: 888, 1989. Descôteaux, J. G., Temple, W. J., Huchcroft, S. A., Frank, C. B., and Shrive, N. G. Linea alba closure: Determination of ideal distance between sutures. Invest. Surg. 6: 201, 1993. Naraynsingh, V., and Ariyanayagam, D. Rectus repair for midline ventral abdominal wall hernia. Br. J. Surg. 80: 614, 1993. Halasz, N. A. Dehiscence of laparotomy wounds. Am. J. Surg. 116: 210, 1968. Greenall, M. J., Evans, M., and Pollock, A. V. Midline or transverse laparotomy? A random controlled clinical trial. Part I: Influence on healing. Br. J. Surg. 67: 188, 1980. Blomstedt, B., and Welin-Berger, T. Incisional hernias: A comparison between midline, oblique and transrectal incisions. Acta Chir. Scand. 138: 275, 1972. ˚ berg, C. Tissue strength of structures involved Tera, H., and A in musculo-aponeurotic layer sutures in laparotomy incisions. Acta Chir. Scand. 142: 349, 1976. Junge, K., Klinge, U., Prescher, A., Giboni, P., Niewiera, M., and Schumpelick, V. Elasticity of the anterior abdominal wall and impact for reparation of incisional hernias using mesh implants. Hernia 5: 113, 2001.
Anisotropy of Human Linea Alba: A Biomechanical Study David Gräel,*,† Andreas Prescher, M.D.,† Sabine Fitzek, M.D.,‡ Diedrich Graf v. Keyserlingk, M.D.,† and Hubertus Axer, M.D.‡,1 *Institute of Diagnostic and Interventional Radiology, Friedrich-Schiller-University Jena, Erlanger Allee 101, D-07747 Jena, Germany; †Department of Anatomy I, RWTH Aachen, Pauwelsstr. 30, D-52057 Aachen, Germany; and ‡Department of Neurology, Friedrich-Schiller-University Jena, Erlanger Allee 101, D-07747 Jena, Germany Submitted for publication July 18, 2004
Background. Recently, a new model of fiber architecture of the linea alba has been described consisting of an oblique fiber layer of intermingling oblique fibers, a transverse fiber layer containing mainly transverse fibril bundles, and a variable, small irregular fiber layer. In this study the morphological model was proven using direction-specific biomechanical measurements of the linea alba. Material and methods. Thirty-one human abdominal walls were analyzed (16 male and 15 female). Six strips of collagen tissue with a width of 1 cm were exsected from each linea alba transversely, obliquely, and longitudinally according to the main fiber directions. An increasing force from 2 to 24 N was applied to these strips, and the corresponding strain represented by the relative elongation was measured, which allows the calculation of a direction-specific compliance of the tissue. Results. The compliance is highest in longitudinal and smallest in transverse direction. In the infraumbilical part of the female linea alba the compliance was significantly smaller in the transverse direction than in the oblique direction. Moreover, the compliance in the transverse direction was significantly smaller in women than in men. Conclusions. A distinct anisotropy of morphological and biomechanical properties was demonstrated as well as sex-dependent differences. The compliance correlates with the distribution of fiber orientation in the linea alba. These biomechanical results constitute the functional correlation with the fiber morphology
1
To whom correspondence and reprint requests should be addressed at Department of Neurology, Friedrich-Schiller-University Jena, Erlanger Allee 101, D-07747 Jena, Germany. E-mail: [email protected].
0022-4804/05 $30.00 © 2005 Elsevier Inc. All rights reserved.
of the linea alba and correspond well to our earlier proposed model of fiber architecture. © 2005 Elsevier Inc. All rights reserved.
Key Words: linea alba; anisotropy; collagen fiber architecture; biomechanics; gender differences. INTRODUCTION
The anterior abdominal wall plays a significant role in surgery since it is a common way to open the peritoneal cavity [1]. It consists of a muscular part and a collagenous part. The oblique external, the oblique internal, and the transverse abdominal muscles constitute the ventrolateral abdominal muscles, while the rectus abdominis and the pyramidal muscles form the ventroanterior muscular part. The collagenous parts in the abdominal midline contain the linea alba and the rectus sheaths. A disturbance of these stabilizing parts of the abdominal wall leads to hernia. The collagenous parts of the ventral abdominal wall are the linea alba and the rectus sheaths. The linea alba is composed of the tendon fibers of the abdominal muscles, thus playing a most significant role in stabilization of the abdominal wall. The architecture of collagen fibers in linea alba has been investigated in a previous study [2, 3]. The result was a new anatomical model of fiber architecture of the linea alba considering the anisotropic order of collagen fibers. Anisotropy means that the tissue exhibits properties with different values when measured in different directions. In short, using confocal laser scanning microscopy the following architectural pattern of the collagen fibers was described [2]. The collagen fibers in linea alba show the same orientation as the muscle fibers of the ventrolateral abdominal wall: transverse, oblique I (directed from right upward to left downward), and oblique II (directed from left upward to right down-
118
¨ EL ET AL.: BIOMECHANICS OF LINEA ALBA GRA
119
oblique and crossed the midline. Thus the linea alba was considered as a line of decussation of fibers. Two patterns of decussation were described: the single and triple pattern of decussation. The close relationship between morphology and function of collagenous tissue [6, 7] inspired us to reinvestigate the linea alba from a biomechanical point of view, this being most important in hernial sac formation [8 –10] and for the development of implantable meshes for abdominal wall repair [11, 12]. The intention of the present study was (1) to determine directional compliances of the linea alba and correlate the biomechanic anisotropy, and (2) to define any sex differences in biomechanic anisotropy of the linea alba. MATERIALS AND METHODS Preparation of the Specimen
FIG. 1. Anatomical model of collagen fiber architecture in the linea alba. The architectural zones are described from ventral to dorsal: (1) oblique fiber layer, (2) transverse fiber layer, and (3) irregular fiber layer.
ward). In linea alba the following three different zones of fiber orientation follow each other from ventral to dorsal (Fig. 1):
Thirty-one fresh abdominal walls were derived from persons (15 female, 16 male, age: 63 to 95 years) who donated their bodies for anatomical study. The abdominal walls were free of scars or any other pathologies. Skin and subcutaneous tissue were sharply dissected. The rectus abdominis muscles were carefully removed from the rectus sheaths and linea alba without injuring the collagen structure of the tissue. The remaining tissue, consisting of the rectus sheaths and the linea alba, was cut transversely at the level of the umbilicus, to distinguish the supraumbilical from the infraumbilical part of the specimen. Three strips were cut out of each part. The three single strips were oriented in longitudinal, transverse, and oblique direction with respect to linea alba (Fig. 2) in randomly changing order. The lines of intersection were angled by means of the eye. Concerning the oblique ones, we tried to follow the trend of the fibers if visible. Otherwise we approximated angles between 40 and 50 degrees with respect to linea alba. Specimens were not obtained in all directions from all donors. Parts of the linea alba had adhesions to the bowels which could not be properly dissected, and others had holes. Those parts were deter-
1. The oblique fiber layer consists of intermingling oblique fibers (in average four to six layers of fibers). 2. The transverse fiber layer contains mainly transverse fibril bundles (in average four to six layers of fibers). 3. An inconstant, small irregular fiber layer is composed of one to two layers of oblique fibers. Moreover, gender differences were demonstrated in the relation of oblique and transverse fibers [3]. A larger amount of transverse fibers relative to the oblique fibers was found in females than in males (60% versus 37.5%). Different regions can be distinguished in the craniocaudal course of the linea alba: supraumbilical part, umbilical part, transition zone, and infraarcuate part, but the general structure of fibers is the same in all four regions. This model of fiber architecture is in contrast with findings proposed by Askar [4] and Rizk [5] describing six aponeurotic layers in linea alba, which all were
FIG. 2. Orientation of the tissue strips in relation to the linea alba.
120
JOURNAL OF SURGICAL RESEARCH: VOL. 124, NO. 1, MARCH 2005 to the loading force per width of the tissue strip. Using the Law of Laplace, Klinge et al. [11] demonstrated the physiological stress of the abdominal wall not to exceed 16 N/cm. (The Law of Laplace states the proportionality of pressure and compliance of a flexible sphere for constant radii of curvature.) At each step, the positions of the needles were recorded using a digital steadycam mounted on the frame. The strain rate was variable during the experiment. Elongation alternated with the taking of pictures. One step over 2 N took about 10 s for both actions. We performed frequent tests of viscous behavior of the tissue. At maximum stress (25 N/cm) we recorded some extra values of absolute elongation with a time delay of 1 min per measurement. These measurements resulted in a negligible elongation even after 2 min. This time corresponds to the distance of the whole process of elongation during the actual measurement of one strip (12 steps times 10 s).
Data Acquisition and Analysis
FIG. 3.
(A) Test device. (B) Specimen fixed to the clamps.
mined to be useless for further examination. A few strips slipped out of the clamps of the test device and were excluded from analysis. Thus, we analyzed 58 different samples in the transverse direction (females: 14 supraumbilical and 13 infraumbilical samples; males: 16 supraumbilical and 15 infraumbilical samples), 55 in oblique direction (females: 11 supraumbilical and 14 infraumbilical samples; males: 15 supraumbilical and 15 infraumbilical samples), and 52 in longitudinal direction (females: 12 supraumbilical and 13 infraumbilical samples; males: 14 supraumbilical and 13 infraumbilical samples). The width of the strips was strictly kept to 1 cm on the relaxed condition. The specimens were kept moist in a buffered Ringer solution as used earlier in other biomechanical studies [13]. Biomechanical measurements were performed at constant room temperature (21°C).
Test Device and Procedure of Measurement The biomechanical test device consists of an octagonal frame with eight concentric arms (Fig. 3A). Each arm is equipped with a crank carrying a spring scale with a clamp fixed on its end. The clamps were used to link the strips to the frame via a spring scale. To avoid slipping during the test, the strips were covered with emery paper before they were jammed in the clamps (Fig. 3B). Afterwards the strips were marked with two needles at the lateral edges of the linea alba, where the double layer of rectus sheaths emerges. The effective length of the tested tissue strip is the distance between the two needles and we supposed to have constant mechanical properties of the tissue between them. The part of the rectus sheath beyond the needles, where the tissue was clamped, was of no interest. The distance between the needles varied depending on the width of linea alba and the direction of the cut. During the test the loading force was increased from 2 to 24 N in steps of 2 N, implying stresses from 2 to 24 N/cm. In this context the term stress is related
To acquire data about the dependence of the relative elongation of the strips on the loading force, the distance between the two needles was measured. The digitally recorded pictures were evaluated by taking the coordinates at the center of the needles’ heads and calculating their distance in units of pixels. The correlation between absolute distance and loading force reveals an interval of approximate linearity between 6 and 16 N. The strong increase of distance in the interval between 2 and 6 N is artificial since it is caused by the elevation of the sagging strip. By means of linear regression, the slope of the absolute distance versus loading force was calculated in this interval. This way we obtained the original distance in the relaxed condition by extrapolating the line of regression to zero Newton. The strain is represented by the relative elongation which results from the absolute distance in relation to the originating distance at zero Newton. The usage of relative elongation as measurement of strain instead of the absolute elongation allows a comparison of the specimens despite different needle distances. The specimens were divided into the supraumbilical and infraumbilical part of the linea alba and were grouped according to the different orientations (transverse, oblique, and longitudinal, Fig. 2). The strain was plotted versus the loading force in each direction, separately for male and female specimens. The slope of the curve within the interval between 6 and 16 N is described by the parameter ␣, which connects the strain with the loading force via the following formula, derived from the law of Hooke:
⫽
ᐉ ⫺ ᐉ0 ᐉ0
⫽␣
F b
where F is the loading force, b is the width of the tissue strip (strictly kept to 1 cm on relaxed condition), l is the distance between the needles’ heads, l 0 is the extrapolated originating distance at zero Newton, and is the relative elongation respectively strain of the tissue strip. The parameter ␣ is not exactly the reciprocal of Young’s modulus respectively modulus of elasticity, because it is not independent of the thickness of the clamped tissue strip. The parameter ␣ represents the compliance per unit thickness and its value is quantified in units of cm/N. Due to its biological spread, the thickness is a main source for variations in the value of ␣. For statistical testing the unpaired Student’s t test was used with two-sided significance levels of P ⫽ 0.05.
RESULTS
The relative elongation was plotted versus the stress in N/cm in each direction, separately for male and female specimens (Fig. 4). The values of the compliance per unit thickness were calculated from these graphs,
¨ EL ET AL.: BIOMECHANICS OF LINEA ALBA GRA
FIG. 4.
121
Strain plotted versus stress, shown separately for the different directions of force, region of linea alba, and gender.
and Fig. 5 shows the compliance per unit thickness derived from the single measurements. The standard deviations of the longitudinal compliance were much larger than the standard deviations of the compliance in transverse and oblique directions. For both sexes the compliance is highest in longitudinal direction. These values differ significantly from the measurements in transverse and oblique direction in most cases (Table 1). The compliance in transverse direction was smallest in both sexes but the transverse compliance was significantly smaller than the oblique one in the infraumbilical part of the female linea alba (Table 1, Fig. 6). Table 2 shows the sex-dependent differences of compliance per unit thickness in the different directions of tensile force. In the supra- and infraumbilical region the compliance per unit thickness in transverse direction was significantly smaller in women than in men. In an earlier study [2, 3] the directions of the collagen fiber bundles in the linea alba were determined with confocal laser scanning microscopy. In the male, it was determined that 44.9% of the fibril bundles were transverse in the supraumbilical part, while 55.1% of the fibers were oblique. In the male infraumbilical part 37.5% of the fibers were transverse and 62.5% were oblique. In contrast, in females 47.6% of the fibers were transverse in the supraumbilical part, while 52.4%
were oblique. In the female infraumbilical part 60.4% of the fibers were found to be transverse, while 39.6% were oblique. There was a significant sex difference between the infraumbilical transverse fiber directions. Figure 7 shows the compliance per unit thickness in comparison with the distribution of fiber direction in the linea alba. Both parameters are correlated significantly to each other (Pearson correlation coefficient: ⫺0.919, P ⬍ 0.001). DISCUSSION
This study focuses on the biomechanical properties of the linea alba in relation to its collagen fiber architecture. It was clearly demonstrated that the linea alba shows a distinct anisotropy in its biomechanical properties with highest compliance in longitudinal direction and lowest in transverse direction. These results conform to the proposed properties of the fiber model described earlier [2]. The strain was plotted versus the corresponding stress (loading force) in each direction. Above 16 N the slopes become flatter and the graphs seem to lead into saturation (Fig. 4). This behavior seems to be due to the mechanical properties of the scissors-like lattice structure of the collagen fibers. Table 2 shows the compliance per unit thickness in
122
JOURNAL OF SURGICAL RESEARCH: VOL. 124, NO. 1, MARCH 2005
demonstrates that the more fibers that are located in a special direction, the lower is the compliance. If the amounts of fibers in two directions differ too much, the differences in compliance become significant. The low value of the compliance in transverse direction clearly contradicts the fiber model described by Askar [4] as this model did not at all describe fibers running transversely to the linea alba. In this case the value of compliance in the transverse direction would have been larger than in the oblique direction. The high amount of transverse fibers enhances the stiffness of linea alba in the transverse direction. Interestingly the well-known Langer lines, which describe the direction of skin incision producing the least tension of the skin in a specific anatomical region, are also oriented transversely to the linea alba [14]. However, biomechanical measurements in this study were performed only on collagen tissue of linea alba without the skin, which was removed from the specimens before. Moreover, sex-dependent differences were demonstrated. The compliance per unit thickness in the transverse direction was significantly smaller in the investigated women than in men. This again corresponds well to the sex differences found in the morphological fiber architecture of the linea alba [3]. Here, it was demonstrated that the female linea alba has more transverse fibers in relation to the oblique fibers than the male one. Moreover, this finding was most distinct in the infraumbilical part. In this study the compliance in the transverse direction is significantly different from that in the oblique direction in the infraumbilical part of the female linea alba (Table 1). It was proposed that these morphological differences could be due to the elevated intraabdominal TABLE 1
FIG. 5. Values of the measured compliance per unit thickness in men and women. Weighted mean values and standard errors of the mean are shown at the top of the diagrams and visualized as gray boxes.
comparison with the distribution of fiber direction in the linea alba. Moreover, both parameters are correlated significantly with each other (Pearson correlation coefficient: ⫺0.919, P ⬍ 0.001). The information about the distribution of fiber direction was derived from an earlier morphological study analyzing the fiber architecture of linea alba [3]. Using confocal laser scanning microscopy, the diameter of each layer of fibril bundles was measured in linea alba of 12 human cadavers, and each fibril bundle was classified according to its orientation (oblique I and II, transverse). Correlating the biomechanical results of this study with the morphological properties of the linea alba
Comparison of Compliance per Unit Thickness Related to the Direction of Force Gender Male
Region
Comparison
P
df
Supraumbilical
Transverse–oblique Longitudinal–transverse Longitudinal–oblique Transverse–oblique Longitudinal–transverse Longitudinal–oblique Transverse–oblique Longitudinal–transverse Longitudinal–oblique Transverse–oblique Longitudinal–transverse Longitudinal–oblique
n.s. 0.002 0.001 n.s. n.s. 0.022 n.s. 0.035 0.006 0.018 n.s. 0.033
29 16 15 28 21 15 13 16 12 20 17 13
Infraumbilical
Female
Supraumbilical
Infraumbilical
Note. The table shows the significance of the difference between the values of mean compliance per unit thickness obtained from samples oriented in different directions. Specimens were not obtained in all directions from all donors; therefore, the degrees of freedom differ. n.s. ⫽ not significant; df ⫽ degrees of freedom.
¨ EL ET AL.: BIOMECHANICS OF LINEA ALBA GRA
FIG. 6. Values of the mean elasticity coefficients in women. Weighted mean values and standard errors of the mean are shown at the top of the diagrams and visualized as gray boxes. Histograms of the values are plotted at the bottom of each diagram.
pressure during pregnancy as the lattice of the oblique fibers would be stressed and deformed so that most of the fibers would run in the transverse direction [3]. Indeed, most of the transverse fibers were found in the infraumbilical linea alba and in this study the compliance per unit thickness in transverse direction becomes significantly different from that in the oblique direction in the infraumbilical part. The compliance of the maternal abdominal wall (which is reciprocal to its stiffness) was found to be inversely related to gestational age [15], a finding which could also be related to such a mechanism. However, this hypothesis has yet to be proven in detail since no information about pregnancies in the history of the body donors of this study is available. Rath et al. [16] performed biomechanical measurements on human abdominal walls. They did not find differences in age, sex, or morphotype. However, they only investigated the compliance of the tissue in the
123
transverse direction of the linea alba, and they were interested in the biomechanical properties of the whole abdominal wall. It must be kept in mind that the specimens of this study are relatively old, as they represent a normal distribution of age of body donors in an anatomical department. Thus, care has to be taken to generalize the presented results to younger populations. However, the relationship between function and morphology of the connective tissue has been clearly demonstrated in this study. Overall, a distinct anistropy of morphological and biomechanical properties was demonstrated as well as gender-dependent differences. These biomechanical results are the functional correlates of the fiber morphology of the linea alba. The abdominal wall is an important structure serving many different functions. The two major functions are movements of the trunk [17] and regulation of intraabdominal pressure [18]. Moreover, it supports respiration [19] and plays a role in stabilization of the spine [20, 21]. All these functions are facilitated by the coordinated and task-specific activation pattern of the abdominal muscles [22]. However, electromyographic observations clearly demonstrated the transversus muscle to be most significant in intraabdominal pressure production [18, 22] and to be recruited preferentially in breathing [19]. In contrast, the internal and external oblique muscles are more effective in contributing to trunk movements [23, 24] and stabilization of the spine [25]. The linea alba has to be seen as the decussation of the tendon fibers of the abdominal wall muscles. The three-dimensional fiber architecture of the linea alba shows the functional anatomical entity of this structure as it acts as the midline insertion of the abdominal wall muscles. Thus, it has to perform the stabilization of the abdominal wall while the muscles resemble the dynamic elements. The abdominal wall (and the linea alba as the collagenous part of it) is a complex dynamic system which performs multiple interactions within this framework of bones, muscles, and collagen fibers. From a surgical point of view the anterior abdominal wall is the object of investigations regarding wound healing [26] and wound closure [27, 28]. Midline ventral abdominal wall hernias include epigastric, umbilical, paraumbilical, and also incisional hernias [9, 10, 29]. The structure of the collagen fibers may be of significance in the choice of direction of laparotomy incision. It was demonstrated that transverse laparotomy incisions are more resistant to rupture than longitudinal incisions [26, 30, 31]. In fact, in transverse laparotomy the transverse fibers in linea alba and rectus sheaths are not cut and can serve their function, while in lon-
124
JOURNAL OF SURGICAL RESEARCH: VOL. 124, NO. 1, MARCH 2005
TABLE 2 Comparison of Compliance per Unit Thickness (␣, in cm/N) with Relative Distribution of Fiber Orientation in Linea Alba Related to Gender Region Supraumbilical
Infraumbilical
Male ␣
Female ␣
P
df
Male distribution
Female distribution
Transverse Oblique
0.67 0.70
0.50 0.73
0.014 n.s.
28 13
Longitudinal Transverse Oblique
1.28 0.73 0.78
1.42 0.55 0.86
n.s. 0.017 n.s.
17 26 27
Longitudinal
1.07
1.14
n.s.
18
44.9% I: 25.5% II: 29.6% 0% 37.5% I: 31.6% II: 30.9% 0%
47.6% I: 22.9% II: 29.5% 0% 60.4% I: 16.7% II: 22.9% 0%
Direction
Note. n.s. ⫽ not significant, df ⫽ degree of freedom, I ⫽ oblique I (directed from right upward to left downward), II ⫽ oblique II (directed from left upward to right downward).
gitudinal incisions the sutures are not fixed in between the cut transverse fibers [32, 33]. Mesh implantation to repair incisional hernias is a common approach. Because the meshes work as a mechanical closure of the defect and induce scar tissue, the biomechanical properties of the implantable meshes should reproduce the physiological biomechanics of the linea alba and the abdominal wall [12, 34]. If the demonstrated sex differences in biomechanical properties of the linea alba imply different sex-oriented surgical procedures cannot be answered from these data, then this topic could be an interesting viewpoint in the future.
ACKNOWLEDGMENTS We thank Anita Agbedor, Andre Doering, and Hans Dietz for technical assistance and Wolfgang Graulich, who drew Fig. 1.
REFERENCES 1.
2.
3.
4.
5. 6. 7.
8. 9. 10. 11.
12.
FIG. 7. Comparison of unweighted mean elasticity coefficients with relative distribution of fiber orientation in linea alba. The morphological data were derived in an earlier study [3]. I ⫽ oblique I (directed from right upward to left downward), II ⫽ oblique II (directed from left upward to right downward).
13.
14.
Prescher, A. Surgical anatomy. In V. Schumpelick, A. N. Kingsnorth (Eds.), Incisional Hernia. Berlin: Springer, 1999. Pp. 45– 60. Axer, H., Keyserlingk, D. G. V., and Prescher, A. Collagen fibers in linea alba and rectus sheaths. I. General scheme and morphological aspects. J. Surg. Res. 96: 127, 2001, doi:10.1006/ jsre.2000.6070. Axer, H., Keyserlingk, D. G. V., and Prescher, A. Collagen fibers in linea alba and rectus sheaths. II. Variability and biomechanical aspects. J. Surg. Res. 96: 239, 2001, doi:10.1006/ jsre.2000.6071. Askar, O. M. Surgical anatomy of the aponeurotic expansions of the anterior abdominal wall. Ann. R. Coll. Surg. Engl. 59: 313, 1977. Rizk, N. N. A new description of the anterior abdominal wall in man and mammals. J. Anat. 131: 373, 1980. Viidik, A. On the correlation between structure and mechanical function of soft connective tissues. Verh. Anat. Ges. 72: 75, 1978. Nilsson, T. Biomechanical studies of rabbit abdominal wall. Part II –The mechanical properties of specimens in relation to length, width, and fibre orientation. J. Biomech. 15: 131, 1982. Engin, A. E., and Akkas, N. Etiology and biomechanics of hernial sac formation. J. Biomed. Eng. 5: 329, 1983. Schumpelick, V. Hernien, 4th ed. Stuttgart: Enke, 2000. Schumpelick, V., and Kingsnorth, A. N. Incisional Hernia. Berlin: Springer, 1999. Klinge, U., Klosterhalfen, B., Muller, M., Anurov, M., Ottinger, A., and Schumpelick, V. Influence of polyglactin-coating on functional and morphological parameters of polypropylene-mesh modifications for abdominal wall repair. Biomaterials 20: 613, 1999. Klinge, U., Klosterhalfen, B., Conze, J., Limberg, W., Obolenski, B., Ottinger, A. P., and Schumpelick, V. Modified mesh for hernia repair that is adapted to the physiology of the abdominal wall. Eur. J. Surg. 164: 951, 1998. Nilsson, T. Biomechanical studies of rabbit abdominal wall. Part I. The mechanical properties of specimens from different anatomical positions. J. Biomech. 15: 123, 1982. Maurel, W. Biomechanical Models for Soft Tissue Simulation. Berlin: Springer, 1998.
¨ EL ET AL.: BIOMECHANICS OF LINEA ALBA GRA 15.
Cohen, D., Timbs, A. E., and Dalton, K. J. Measurement of compliance of the maternal abdominal wall in pregnancy. Eur. J. Obst. Gyn. Repr. Biol. 23: 267, 1986. 16. Rath, A. M., Attali, P., Dumas, J. L., Goldlust, D., Zhang, J., and Chevrel, J. P. The abdominal linea alba: an anatomoradiologic and biomechanical study. Surg. Radiol. Anat. 18: 281, 1996. 17. De Troyer, A. Mechanical role of the abdominal muscles in relation to posture. Respir. Physiol. 53: 341, 1983. 18. Cresswell, A. G. Responses of intra-abdominal pressure and abdominal muscle activity during dynamic trunk loading in man. Eur. J. Appl. Physiol. 66: 315, 1993.
25.
19.
29.
De Troyer, A., Estenne, M., Ninane, V., Van Gansbeke, D., and Gorini, M. Transversus abdominis muscle function in humans. J. Appl. Physiol. 68: 1010, 1990.
20.
Gracovetsky, S., Farfan, H., and Helleur, C. The abdominal mechanism. Spine 10: 317, 1985.
21.
Tesh, K. M., Shaw Dunn, J., and Evans, J. H. The abdominal muscles and vertebral stability. Spine 12: 501, 1987.
22.
Cresswell, A. G., Grundström, H., and Thorstensson, A. Observations on intra-abdominal pressure and patterns of abdominal intra-muscular activity in man. Acta Physiol. Scand. 144: 409, 1992.
23.
McGill, S. A revised anatomical model of the abdominal musculature for torso flexion efforts. J. Biomech. 29: 973, 1996.
24.
Torén, A., and Öberg, K. Maximum isometric trunk muscle strength and activity at trunk axial rotation during sitting. Appl. Ergonomics 30: 515, 1999.
26.
27.
28.
30. 31.
32.
33.
34.
125
Gardner-Morse, M. G., and Stokes, I. A. F. The effects of abdominal muscle coactivation on lumbar spine stability. Spine 23: 68, 1998. Nilsson, T. The relative rate of wound healing in longitudinal and transverse laparotomy incisions. Acta Chir. Scand. 148: 251, 1982. Campbell, J. A., Temple, W. J., Frank, C. B., and Huchcroft, S. A. A biomechanical study of suture pullout in linea alba. Surgery 106: 888, 1989. Descôteaux, J. G., Temple, W. J., Huchcroft, S. A., Frank, C. B., and Shrive, N. G. Linea alba closure: Determination of ideal distance between sutures. Invest. Surg. 6: 201, 1993. Naraynsingh, V., and Ariyanayagam, D. Rectus repair for midline ventral abdominal wall hernia. Br. J. Surg. 80: 614, 1993. Halasz, N. A. Dehiscence of laparotomy wounds. Am. J. Surg. 116: 210, 1968. Greenall, M. J., Evans, M., and Pollock, A. V. Midline or transverse laparotomy? A random controlled clinical trial. Part I: Influence on healing. Br. J. Surg. 67: 188, 1980. Blomstedt, B., and Welin-Berger, T. Incisional hernias: A comparison between midline, oblique and transrectal incisions. Acta Chir. Scand. 138: 275, 1972. ˚ berg, C. Tissue strength of structures involved Tera, H., and A in musculo-aponeurotic layer sutures in laparotomy incisions. Acta Chir. Scand. 142: 349, 1976. Junge, K., Klinge, U., Prescher, A., Giboni, P., Niewiera, M., and Schumpelick, V. Elasticity of the anterior abdominal wall and impact for reparation of incisional hernias using mesh implants. Hernia 5: 113, 2001.