European Journal of Obstetrics & Gynecology and Reproductive Biology 109 (2003) 8087

A study of pelvic ligament strength

Michel Cossona,*, Malik Boukerroua, Sophie Lacazea, Eric Lambaudiea, Jean Faselb, Henri Mesdaghc, Pierre Lobryd, Anne Egoa
a

le de Chirurgie Gyne cologique, Ho pital Jeanne de Flandre, Centre Hospitalier Re gional Universitaire de Lille, 59037 Lille, France Po b partement de Morphologie Centre Me dical Universitaire, 1, Rue Michel Servet CH-1211, Geneva 4, Switzerland De c gional de Lille, Faculte de Me decine de Lille, Lille, France Laboratoire danatomie, Centre Hospitalier Re d riaux et Technologies Nouvelles, Institut Catholique des Arts et Me tiers de Lille, 6 Rue Auber 59046 Lille, France Laboratoire Mate Received 20 June 2002; accepted 29 November 2002

Abstract Objectives: To measure the strength at tearing of pelvic ligaments used in the cure of prolapse and urinary incontinence. Material and methods: We performed our measurements on pelvis ligaments from cadaveric specimens. We dissected 29 human female pelvis cadavers of which storage conditions differed. Ten were frozen, 10 fresh and 9 were stored in formalin. In each cadaver we dissected pre-vertebral ligaments at promontory and right and left symmetrical ligaments. These were the iliopectineal, sacrospinous and arcus tendineus of pelvic fascia. A subjective clinical evaluation of the ligament properties was performed by visual observation as well as nger palpation. Ligaments were classied into three groups. Group A contained high quality ligaments, in terms of thickness and apparent strength following nger palpation. Ligaments of doubtful quality were classied in group B and low apparent quality ligaments in group C. Then the ligaments were stitched by a suture taking the entire ligament and a force was applied on the vagina axis until tearing. The device used for strength measurement during traction was a SAMSON type force gauge, model EASY, serial number SMS-R-ES 300N manufactured by Andilog that was developed for the purpose of our study. Measurements were given in Newton (N). Results: There was a great variability in the values obtained at tearing with minimal values at around 20 N and maximal values at 200 N. Individually measured, ligament strength varied between individuals, and for the same patient between the type of ligaments and the side. The pre-vertebral ligament was on average the strongest. There was no signicant difference according to the storage condition except for the pre-vertebral ligament in formalin cadavers. For bilateral ligaments, there was no difference between the left and right side. The iliopectineal ligament was statistically signicantly stronger than the sacrospinous and arcus tendineus of pelvic fascia. There was a correlation between subjective evaluation and objective strength measurements. Discussion: No papers have been published on the strength of pelvic ligaments at tearing. These are however routinely used in the cure of prolapse and urinary incontinence. Our results show that there is a great variability in strength between individuals, and for a same patient between the types of ligaments and side. These observations could explain some of the surgical intervention failures and demonstrate the importance of per-operative strength evaluation. Per-operative subjective evaluation of strength is related to objective measurements and could be used to determine the type of ligaments to be used for surgical suspension. Freezing does not damage pre-vertebral ligament strength and further studies are required to evaluate elasticity of pelvic ligaments. # 2003 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Pelvic ligaments; Strength measurement; Vaginal prolapse

1. Introduction A wide range of surgical techniques routinely used for the treatment of urinary incontinence as well as vaginal prolapse consists in the suspension of the vaginal tissue. This is performed either by using surgical suture (Burch colposuspension procedure, suspension of the vagina by Richter,

* Corresponding author. Tel.: 33-320993216. E-mail address: m-cosson@chru-lille.fr (M. Cosson).

para-vaginal suspension) or by using a prosthesis suspended by a suture (promontoxation). All these techniques require the use of four principal ligaments, three of which are bilateral and symmetrical. The right and left iliopectineal ligaments (pectineal ligament) are used in the Burch procedure for the treatment of urinary incontinence with indirect suspension of the urethrovesical junction [13]. Sacrospinal vault suspension as described by Richter [4] for the treatment of prolapsed vagina requires attachment to the right and left sacrospinous ligaments (sacrospinal ligament). For the treatment of urin-

0301-2115/03/$ see front matter # 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0301-2115(02)00487-6

M. Cosson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 109 (2003) 8087

81

` le [5], the arcus ary incontinence or in case of Cystoce tendineus (arcus tendineus fascia pelvis) reinforcing the pelvic fascia and pelvic aponevrosis is used in para-vaginal suspension. Finally the common anterior longitudinal vertebral ligament or pre-vertebral ligament (anterius longitudinal ligament) is the attachment point for articial prosthesis used in the treatment of genital and rectal prolapse with promontoxation [6,7]. For the majority of surgeon, the nature of the ligament used is recognised as being the strongest element of the assembly unlike the vaginal tissue being the weakest. Signicant failure rate are seen in these interventions recognised as being mainly due to tearing from the vaginal tissue [810]. A literature search (Pubmed) did not retrieve published papers measuring the strength of the above ligaments. We also did not found previous studies on biomechanical properties of these ligaments as well as no study on their maximal strength. We therefore propose to measure the strength of human female pelvis ligaments obtained from cadavers.

Frozen cadavers were always defrosted at room temperature during an average period of 36 h. Then they were dissected at the anatomy laboratory of Lille (Pr. Mesdagh) or in the anatomic laboratory at Geneva School of Medicine headed by Pr. Fasel (Switzerland). 2.2. Dissection and suture insertion All the dissections were always performed by the rst four authors, who all had experience in prolapse intervention, urinary incontinence signs and pelvic ligaments dissection. Only one member of a team of two carried out the ligaments dissection. However, the two operators did the subjective evaluation successively. A quality control check to evaluate the ligament grip was always performed by the non-operating member before inserting the suture material in order to minimise technical defaults. The cadaver was placed in the supine position. We started by removing the anterior wall. Then we fully removed the small pelvis. The bladder, the uterus, the annexes, the vagina and the recto-sigmoid junction were widely dissected and removed from the small pelvis ensuring that the ligaments needed for our measurements were not damaged. The use of the retro peritoneal route and removal of fatty tissues permitted to show the following ligaments needed for our measurements: right and left iliopectineal ligaments, tendineus arch of pelvic fascia, both sacrospinous ligament and the common anterior vertebral ligament. Before inserting the suture, a subjective clinical evaluation of the ligament properties was performed by visual observation as well as nger palpation. After the ligaments were classied into three groups. Group A contained high quality ligaments, in terms of thickness and apparent strength following nger palpation. Ligaments of doubtful quality were classied in group B and low apparent quality ligaments in group C. A double braided lament suture having more strength than the ligaments switched each of the dissected ligaments. Braided sutures known as being the strongest for traction and of least elasticity were chosen. MERSUTURE 2 and VICRYL 1 and 2 were selected for their characteristics. The anterior pre-vertebral and iliopectineal ligaments were widely switched without considering the adjacent blood vessels or nerves. The suspension localisation was performed similarly than during prolapse or urinary incontinence interventions. The iliopectineal ligament was switched on its median part at approximately 2 cm from the median line on the inferior and posterior side of the pubic ramication. The anterior pre-vertebral ligament was switched 12 mm deep without penetrating the intervertebral disc in order to best able to visualise the suture by transparency as during promontoxation interventions. Sacrospinous ligaments were switched 2 cm away from the ischial spine and perpendicular to the ligament surface plane as described in the ` le by Richter [1]. Arcus tendineus were treatment of Rectoce

2. Materials and methods 2.1. The criterions for selecting patients and conservations procedures We performed our measurements on pelvic ligaments obtained from cadaveric specimens. A total of 29 cadavers were dissected at the anatomy laboratory both in Lille and Geneva. Different methods of storage had been used for the cadaveric specimen, 10 were deep-frozen, 10 used fresh and 9 stored in formalin. Dissection was solely carried out in patients who consented prior to death to have their cadaver used for medical purposes and according to current regulations. Our study was performed over the period 2 October 2000 to 4 December 2001. All patients for whom the identication of all the ligaments required was not possible were excluded from our study. Thus no cadaveric specimens originating from patient with a past medical history of pelvic surgery or prolapse treatment were used. One patient with a history of pelvic tumour was excluded. Dissection was done at the anatomy laboratory of Lille (France) headed by Pr. Mesdagh for fresh and formalin cadavers. Some of the latter have been stored for a few years. Fresh cadavers were always dissected within 24 h of death before any preparation or storage procedure was started. The formalin conservation used a mixture of 3 l of formalin, 3 l of ethyl alcohol, 1 l of phenol, 3 l of glycerin, and 6 l of water. The embalmed cadavers were prepared with and intravascular injection of this mixture. A period of 1 year was needed to gather 10 cadaveric specimens. The amount was limited by regulations and by obligation of research purposes.

82

M. Cosson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 109 (2003) 8087

Fig. 1. Picture of the pelvic floor with the six ligaments ready to be measured. TAPF: tendineus arch of pelvic fascia; R: right; L: left.

switched at around 1 cm from the ischial spine as performed ` le. by Scotti et al. [5] in the cure of Cystoce 2.3. Measurement of the maximal strength of the ligaments Once the sutures were inserted (Fig. 1), measurements on the ligaments maximal strength at tearing were performed. The material used was developed in collaboration with Pr. tiers Lobry from Institut Catholique des Arts et Me (ICAM) in Lille. This measuring device enabled measures being performed already in the dissection room. It also enabled axis directed traction as close as possible to the vagina by mean of a pulley. This device was produced and calibrated at ICAM in Lille. The device used for strength measurement during traction is a SAMSON type force gauge, model EASY, serial number SMS-R-ES 300N manufactured by Andilog (http://www.andilog.com). The latest calibration took place on 28 April 2000. Results were in Newton and maximal tension values at the time of tearing were saved. Measurements were given with 0.6 N precision, i.e. 0.2%. The device was installed on a tripod (similar to those used for photographic camera (Fig. 2)). Axis directed traction was possible by using a second directional pulley, around which the suture xed at the ligament was attached. This enabled orientation in a vertical and horizontal surface plane in order to guide the way of traction. Two bars at the bottom of the device ensured the stability of the assembly. These were place in the gutter of the dissection table. A spirit level was used to constantly verify the horizontality of the assembly. The ligament suture was inserted on the pulley and connected to the strength sensor. After

verifying the horizontality of the mounting, we applied a progressive force on the ligament until tearing. At this moment the force gauge indicated the value obtained in Newton (Fig. 3). EPI INFO software was used for statistical analysis of the results (French version 6.04 cfr Ecole National de Sante Publique in Rennes). The analysis of variance was calculated using software SPSS version 9 for Windows1 and all results were given with 5% error. Correlation was the

Fig. 2. The device used for strength measurement placed under the cadaver.

M. Cosson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 109 (2003) 8087

83

Fig. 3. Schematic representation of the device used for strength measurement.

methods of analysis used when studying possible associations.

3. Results We dissected 29 female cadaveric specimens, 10 were fresh, 10 previously deep-frozen and 9 stored in formalin. The mean age of formalin-stored cadavers was 84 years (ranging from 64 to 96 years). For deep-frozen and fresh cadaver, mean age was 71.4 years (4986 years) and 81.4 years (6492 years), respectively. The average time between death and dissection was 20 months for formalin cadavers, and the average freezing time for frozen cadavers was 82.4 weeks (1288 weeks). Fresh cadavers were dissected within

24 h after death. A preliminary remark was the variability in extreme values in strength obtained for pelvic ligaments, with values ranging from 1 to 10 (Fig. 4). There was also a great variability in same type ligaments, with differences from a factor one to ve as for example seen for the prevertebral ligaments with a maximum of 203.48 N and a minimum of 47.73 N. The highest values were observed for pre-vertebral and iliopectineal ligaments with a maximum of 180200 N. For the other types of ligaments, we observed minimal values of 2030 N for the arcus tendineus pelvic fascia and sacrospinous ligaments. A wide disparity in the results obtained for a same patient regardless of the ligament types and storage method used (Figs. 57). With regards to iliopectineal ligaments, we found concordant results for the left and right side in the

Fig. 4. Ligaments strength for the whole population (n 29). ALL: anterior longitudinal ligament, pre-vertebral ligament; TAPF: tendineus arch of pelvic fascia; R: right; L: left; sd: standard deviation.

84

M. Cosson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 109 (2003) 8087

Fig. 5. Ligament strength for fresh specimens (n 10). ALL: anterior longitudinal ligament, pre-vertebral ligament; TAPF: tendineus arch of pelvic fascia; R: right; L: left.

Fig. 6. Ligament strength for frozen specimens (n 10). ALL: anterior longitudinal ligament, pre-vertebral ligament; TAPF: tendineus arch of pelvic fascia; R: right; L: left.

same range of values. There were however differences between subjects. For subjects I and V (fresh cadavers), the results obtained for the right iliopectineal ligament were 107.3 and 57.2 N, respectively, and 35.7 and 105.9 N, respectively, for the left iliopectineal ligament. For both subjects, the clinical evaluation was in agreement with measured values; ligaments of high strength were classied in group A whereas those of least strength were in group C.

The following values were obtained for fresh cadavers. For patient IX, values obtained for the right and left arcus tendineus were 110 and 54.7 N, respectively. Measurement on the right and left sacrospinous ligaments for patient III was 37.4 and 104.8 N, respectively. Unlike to values obtained for patients I and V where subjective evaluation was in harmony with measured strength, differences were seen for patients IX and III. The right and left arcus

Fig. 7. Ligaments strength of formalin specimens (n 9). ALL: anterior longitudinal ligament, pre-vertebral ligament; TAPF: tendineus arch of pelvic fascia; R: right; L: left.

M. Cosson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 109 (2003) 8087

85

tendineus for patient IX were classied in group A and sacrospinous ligaments for patient III were found in group B after clinical evaluation. Standard deviation calculated for the ligaments were 31.1 for iliopectineal ligament, 28.2 for arcus tendineus and 33.2 for sacrospinous. The mean variation was 6.6 N for iliopectineal ligament, 4.8 N for arcus tendineus and 9 N for sacrospinous ligaments. Several comparisons on the various ligaments were therefore possible regarding the mean results obtained (Fig. 4). We initially investigated whether the storage condition had any impact on the maximal strength measurements. An analysis of variance found a signicant impact of the storage condition only for pre-vertebral ligaments (P 0:012). Its strength was signicantly reduced after prolonged formalin storage. No similar results were obtained for the other types of ligaments, with no statistically signicant difference on the storage condition for symmetrical ligaments; right (P 0:467) and left (P 0:799) iliopectineal ligament, right (P 0:482) and left (P 0:831) sacrospinous ligament, and right (P 0:287) and left (P 0:472) arcus tendineus. The following statistical results are given irrespectively of the storage condition except for pre-vertebral ligaments. We found a statistical difference for the pre-vertebral ligament obtained from fresh (Fig. 5) and deep-frozen cadavers (mean value 141:3 N (Fig. 6)) when comparing the maximal strength mean value obtained for the various ligaments. The lowest mean value was found in the formalin stored cadaver with a value of 46.03 N measured for the sacrospinous ligament (Fig. 7). Then we compared the values obtained for symmetrical ligaments depending on their lateralisation. This analysis was performed by a two way analysis of variance. We did not observe a signicant statistical difference regarding the side of measurement (right or left) for the three ligaments. However, there was a statistical difference between the iliopectineal ligaments, which were found to be more resistant than sacrospinous and arcus tendineus (P < 0:001). No difference was seen between the sacrospinous ligaments and the tendineus arch of pelvic fascia. Finally by an analysis of variance, we compared the values obtained after subjective clinical evaluation of the ligaments and their classication in groups A, B and C and the maximal strength measured (Table 1). This analysis conrmed the
Table 1 Comparison of ligaments strength subjective and objective measuresa Subjective measure A B C Total
a

statistical correlation between the subjective evaluation and the maximal strength measurement (P < 0:001). This correlation was observed irrespectively of the storage conditions and ligaments studied. There are however some exceptions. For patient IX (fresh cadaver), the values obtained for the right and left iliopectineal ligament were 51.54 and 92.66 N, respectively, although both ligaments were classied in group A after subjective evaluation; and for patient V (fresh cadaver), the values obtained for the right and left arcus tendineus were 42.1 and 19.4 N although they were both classied in group B. For patient VII (formalin cadaver), we observed a difference of 60 N between the right and left iliopectineal ligament although the clinical evaluation was similar for these two ligaments. A 50 N difference was also observed in this patient for both sacrospinous ligaments with again a disparity between clinical evaluation and measurement. For patient III (formalin cadaver), we measured a 40 N difference between both arcus tendineus of pelvic fascia although the clinical evaluation was similar (group A).

4. Discussion Limitations in our study were due to numerous material constraints. It would not have been possible to study the resistance of ligaments in vivo, as it requires its traction until tearing. We therefore decided on using cadaveric specimens where dissection was easier, exempt of risk and less haemorrhagic. It also permitted to perform measurements in optimal conditions. No similar study measuring the strength of ligaments on humans or animals has previously been published [11]. It is therefore not possible to compare the differences obtained according to the storage condition. In light of the results obtained, freezing has a negative effect on the resistance and elasticity of tendineus structures [12], but no data are available on the effect of freezing or formalin storage on the small pelvis ligaments. We therefore compared ligaments strength according to three different storage conditions. Furthermore due to technical constraints and the obligation to perform measurements at the anatomy laboratory, we were not able to study elasticity plots before tearing, which would have been interesting [13]. This would have required an additional and bulky measurement device. The device we developed only permitted strength measurements at tearing but no data on elasticity could be measured. Our study was performed over a 14-month period on 29 fresh, deep-frozen or formalin stored cadaver, by objectively measuring the maximal resistance at tearing of pelvis ligaments used in the surgical treatment of urinary incontinence and in cases of prolapse. There are few data on the biomechanical properties of the pelvis namely on the mechanical aspect of biological tissue as well as on the

Mean objective measure 84.4 54.6 38.3 71.1

N 131 38 34 203

Standard deviation 39.2 27.2 14.3 38.8

Analysis of variance: statistical significance with P < 0:0001.

86

M. Cosson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 109 (2003) 8087

components used. Some papers describe the results of investigations of resistance and strain rates in the horse and rabbit [1416] as well as the evaluation of human tendons properties [12,17]. But pelvis ligaments are used in numerous surgical interventions in the treatment of urinary incontinence as well as in rectal and vaginal prolapse. In this type of intervention, they ensure the assembly is solid and represent the suspension point. However their resistance was never studied. One could imagine that healing reinforces the resistance of the assembly. But healing may last for a few weeks during which constraints are put on the suspension assembly. For pre-vertebral ligaments, we measured resistance until tearing of up to 200 N which represents 20 kg. In some cases, ligaments resistance can be weak at around 20 N as measured for some sacrospinous and arcus tendineus. This could explain the 1030% surgical intervention failure described elsewhere [110], on the outside of technical problem, specially for the sacrospinous ligament or the tendineus arch of fascia pelvis. We have shown that pre-vertebral and iliopectineal ligaments are signicatively more resistant than sacrospinous and arcus tendineus. These two ligaments are usually used in the treatment of prolapse or incontinence by the upper route although sacrospinous and arcus tendineus ligament are mainly used in surgical intervention by the lower route. This difference in resistance by the lower route may explain the higher failure rate in this type of intervention as previously described in the literature [810]. Difference in values of resistance obtained between the right and left side may explain unilateral failure of ligaments suspension. These differences were not explained by difcult ligaments grip, as dissection conditions were adequate. Cadaver ligament dissection is easier than in vivo interventions due to the absence of haemorrhagic complication and exposition to the operative eld. We have also shown that subjective clinical evaluation of ligaments was statistically correlated to objective resistance. This result can be applied during surgical intervention to evaluate the apparent quality of ligaments. A doubtful quality ligament observed during a surgical intervention would necessitate to reinforce the resistance of a suspension by using additional suspension or synthetic material as well as adhesive to enhance secondary clinical results. Thus in Richters type intervention, xation of the vagina on the sacrospinal ligament by the vaginal route is usually done unilaterally. In front of a sacrospinal ligament of doubtful quality, the operator should consider a controlateral dissection to perform a bilateral suspension [18]. However it is still required to dene the maximal strength the assembly in the cure of prolapse are subject to in particular during cough episode and pushing effort to fully interpret our ndings. After this preliminary step, we would have to consider the suture, prosthesis and vaginal tissue and the para-vagina tissue. We would then know the resistance potential and weakness of surgical assembly.

5. Conclusion This novel work on the biomechanical properties of pelvis ligaments has consequences on surgical attitude especially for static pelvis intervention. Individually measured, ligament strength varies between individuals, and for a same patient between the types of ligaments and their side. We have shown that the ligaments used for vaginal intervention (sacrospinous and arcus tendineus fascia pelvis) are signicatively less resistant than ligaments used for abdominal route (pre vertebral and iliopectineal ligament). The ligaments quality accounts for the failure rate encountered during prolapse intervention and in the treatment of urinary incontinence. Clinical evaluation enables a globally reliable evaluation of the actual ligament strength. A per-operative subjective evaluation of the strength is correlated to the objective measures and should be used a determining factor in the choice of ligaments used for suspension techniques. Freezing does not alter the resistance of pre-vertebral ligament and further study is required to study their elasticity. References
[1] Burch JC. Urethrovaginal fixation to Coopers ligament for correction of stress incontinence. Am J Obstet Gynecol 1961;81:28190. [2] Burch JC. Coopers ligament urethrovesical suspension for stress incontinence. Am J Obstet Gynecol 1968;100:76474. [3] Stanton SL, Williams JE, Ritchie D. The colposuspension operation of urinary incontinence. Br J Obstet Gynecol 1976;83:8905. [4] Richter K, Albrich W. Long-term results following fixation of the vagina on the sacrospinal ligament by the vaginal route (vaginaefixatio sacrospinalis vaginalis). Am J Obstet Gynecol 1981;141: 8116. [5] Scotti RJ, Garely AD, Greston WM, Flora RF, Olson TR. Paravaginal repair of lateral vaginal wall defects by fixation to the ischial periosteum and obturator membrane. Am J Obstet Gynecol 1998;179:143645. [6] Addison WA, Livengood CH, Sutton GP, Parker RT. Abdominal sacral colpopexy with Mersilene mesh in the retroperitoneal position in the management of post hysterectomy vaginal vault prolapse and enterocele. Am J Obstet Gynecol 1985;153:1406. [7] Snyder TE, Krantz KE. Abdominal-retroperitoneal sacral colpopexy for the correction of vaginal prolapse. Obstet Gynecol 1991;77:944 9. [8] Paraiso MR, Ballard LA, Walters MD, Lee JC, Mithinson AR. Pelvis support defects and visceral and sexual function in women treated with sacrospinous ligament suspension and pelvic reconstruction. Am J Obstet Gynecol 1996;175:142331. [9] Benson JT, Lucente V, McClellan E. Vaginal versus abdominal reconstructive surgery for the treatment of pelvic support defects: aprospective randomized study with long term outcome evaluation. Am J Obstet Gynecol 1996;175:141822. [10] Hardiman PJ, Drutz HP. Sacrospinous vault suspension and abdominal colposacropexy: success rates and complications. Am J Obstet Gynecol 1996;175:6126. [11] Duck FA. Physical properties of tissue a comprehensive reference book academic press limited. London: Harcourt Brace Jovanovich Publishers; 1990.

M. Cosson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 109 (2003) 8087 [12] Clavert P, Kempf JF, Bonnomet F, Boutemy M, Kahn JL. Efects of freezing/thawing on the biomechanical properties of human tendons. Surg Radiol Anat 2001;23:25962. [13] Fung YC. Biomechanics. Mechanical properties of living tissues, 2nd ed. New York: Springer; 1993. [14] Lochner FK, Milne DW, Mills EJ, Groom JJ. In vivo and in vitro measurement of tendon strain in the horse. Am J Vet Res 1980;41:192937. [15] Yamamoto E, Tokura S, Yamamoto N, Hayashi K. Mechanical

87

properties of collagen fascicles from in situ frozen and stress-shield rabbit patellar tendons. Clin Biomech 2000;15:28491. [16] Herrick WC, Kingsbury HB, Lou DY. A study of strain. J Biomed Mat Res 1978;12:87794. [17] Harris EH, Walker LB, Bass BR. Stress-strain studies in cadaveric human tendon and an anomaly in the Youngs modulus thereof. Med Biol Eng 1966;4:2539. [18] Pohl JF, Frattarelli JL. Bilateral transvaginal sacrospinous colpopexy: preliminary experience. Am J Obstet Gynecol 1977;177:135662.