Annals of Biomedical Engineering, Vol. 28, pp. 619628, 2000 Printed in the USA. All rights reserved.

0090-6964/2000/286/619/10/$15.00 Copyright 2000 Biomedical Engineering Society

Optimization of Design and Surgical Positioning of Inatable Penile Prostheses

A. GEFEN,1 J. CHEN,2 and D. ELAD1
Department of Biomedical Engineering, Faculty of Engineering, Tel Aviv University, Tel Aviv 69978, Israel and 2Sexual Dysfunction Clinic, Department of Urology, Tel Aviv Sourasky Medical Center, Tel Aviv 64239, Israel
(Received 10 December 1999; accepted 23 May 2000)
1

AbstractThe interaction between the cylinders of an inatable penile prosthesis IPP and the surrounding tissues during IPP-aided erection may result in local elevated stresses. These stresses may reach values that can obstruct penile blood vessels and cause ischemia and/or stimulate nerves around the operation site, thereby inducing sensations of pain. A new numerical model was used to analyze penile stresses postimplantation of different IPP types, in order to optimize prosthesis design and surgical positioning by enabling minimal stress transfer to dorsal blood vessels and nerves. The results suggest that intraluminal pressures should be maintained at low levels about 80 kPa while cylinder thickness and stiffness should be kept just high enough approximately 15% of the radius and 1000 MPa, respectively to eliminate deleterious cylinder-tissue contact stresses. Smaller prosthetic cylinders, i.e., occupying about 45% of the cavernosal space, may be advantageous in terms of reducing dorsal stresses, but lower penile rigidity should be expected. A signicant decrease of dorsal stresses can also be achieved by encouraging the surgeon to position the cylinders toward the lower part of the corpora. The numerical simulations indicate that circular cylinders may allow greater biomechanical compatibility of the IPP with the penis structure than elliptic ones, and this should be a subject for clinical investigations. 2000 Biomedical Engineering Society. S0090-69640001106-1 KeywordsErectile dysfunction, model, Finite element method. Impotence, Numerical

INTRODUCTION The inatable penile prosthesis IPP was rst introduced in the early 1970s as an interventional treatment of impotence.14 The IPP is composed of three components: a reservoir that is located in the perivesical space, two inatable cylinders that are surgically inserted along the erectile bodies of the penis i.e., corpus cavernosa, and a pump that is implanted within the scrotum. When erection is desired, the penis can be distended to a near normal erectile condition by repeatedly squeezing the pump, thereby transferring uid from the reservoir to
Address correspondence to Professor David Elad, Department of Biomedical Engineering, Faculty of Engineering, Tel Aviv University, Tel Aviv 69978, Israel. Electronic mail: elad@eng.tau.ac.il

inate the cylinders. Manual release of the valve at the lower portion of the pump will cause the uid to ow back into the reservoir, returning the penis to a accid state. Since this mechanism of operation mimics the normal erectile process, it is generally accepted that the IPP provides the best physiological results and, accordingly, numerous successful IPP implantation procedures were performed during the last two decades.13,18 However, a variety of postoperative complications have been encountered in up to 20% of the patients, including infections, mechanical failures of the prosthesis, severe pain associated with its operation, prosthesis protrusion through the glans penis, and aneurysms.4,12,15 The mechanical interaction between the articially inated prosthetic cylinders and the surrounding penile tissues induces mechanical stresses within the penis, which may include sites of localized, highly elevated stresses. Destruction of some penile tissues during the surgical procedure exposes nerve endings on the internal surfaces surrounding the cylinders. Many researchers now agree that under these conditions, these nerve endings develop hypersensitivity alodynia and, therefore, their being stimulated by excessive stretching and/or compression of adjacent tissue are perceived by the patient as painful sensations.2,10,16 These sensations could also be augmented by obstruction of some dorsal penile blood vessels due to intensive compression, i.e., ischemic pain. These effects of local intensied stresses could be the underlying causes for penile pain during ination of the prosthetic cylinders, which was clinically observed to be the most common cause for dissatisfaction with the device.6 Therefore, the biomechanical compatibility of an IPP can be characterized by the stresses developing around the primary nerve roots and blood vessels due to ination of the cylinders. Gefen et al.3 recently introduced a computational model of the penis/IPP complex that allows for quantitative analysis of the distribution of stresses within the different penile tissues during IPPaided erections. In the present study, we utilized this computational model to optimize the stresses within the 619

620

GEFEN, CHEN, and ELAD

FIGURE 1. Biomechanical model of the human penisIPP complex and its adaptation to different prosthesis designs, orientations and positioning: a d circular cylinder cross sections and e h elliptic cylinder cross sections.

penile tissues during IPP-aided erections that had been obtained by using different IPP structural designs with different surgical positions of the cylinders within the cavernosal spaces. Our aim was to minimize stress transfer to primary nerves and blood vessels and thereby alleviate some of the complications sometimes produced by these devices. METHODS The Penis/IPP Structural Model The methodology used to build a two-dimensional 2D computational model for analysis of structural stresses in the penile tissues during IPP-aided erection is described in detail in Gefen et al.;3 its essential components relevant to the present report are given later. The symmetrical geometry of a typical 2D transverse cross section of the human penis was extracted from an anatomical scheme. The model includes the following elements: IPP cylinders, compressed corpus cavernosa, tunica albuginea, skin, dorsal blood vessels, and the ure-

thra Fig. 1a. The corpus spongiosum, whose crosssectional area is usually signicantly smaller than that of the corpus cavernosa, was excluded. The prosthesis bulk and penile soft tissues were assumed to be made of homogenous, isotropic, and linear elastic materials, whose mechanical properties are detailed in Gefen et al.3 The stress distribution was determined for each conguration by employing a commercial nite element analysis software package ANSYS which was used to solve the general equilibrium equations for plane stress. Automatic meshing was used to generate optimally converging meshes of 25003000 quadrilateral and triangular elements that described the cross-sectional geometry Fig. 2. The meshes were determined by a converging process in which the mesh density was gradually increased until the deviation in the produced stress values did not exceed 5%. During the meshing process, special attention was paid to avoid extreme transitions in sizes of adjacent elements, which may induce local inaccuracies in the numerical solution. The boundary conditions included an intralumenal device pressure ( P i ) applied to the interior

Optimization of Design and Positioning of Penile Prostheses

621

FIGURE 2. Examples of meshes of the penisIPP model geometry: a and b circular cylinder cross sections see Figs. 1b and 1c for respective model congurations and c and d elliptic cylinder cross sections see Figs. 1f and 1h for respective model congurations.

proles of the cylinders, as well as four constraints on the lateral and dorsal-plantar aspects of the penis which allowed its expansion but not rotation in response to ination of the cylinders. The nite element analysis provided the structural stress distribution in terms of von Mises ( v . M . ) equivalent stress
2 1/2 v.M . 2 1 2 1 2 ,

IPP Design Optimization The model was implemented to simulate the interaction of soft tissues of a typical penis with a series of inatable prostheses that differ by their cross-sectional geometry, cylinder material properties, and intraluminal operational pressures. In addition, we tested different positioning of the cylinders within the cavernosal spaces which is also likely to play an important role in determining the distribution of stresses. Values for these parameters were selected to conform to the design parameters of commercial IPPs. Cylinder Geometry Two types of cylinder cross-sectional shapes, circular and elliptic, are presently commercially available at different lengths 1028 cm accid, diameters 1018 mm accid, and wall thickness. When elliptic cylinders are considered, alignment of their principal axes with respect to the corporal anatomy during surgery is also important and was taken into consideration in this study. The following penis/IPP congurations were selected for the present optimization analysis Fig. 1: a large and dorsally positioned thick circular cylinders, which are dened as a reference case for the comparative analysis; b large and dorsally positioned thin circular cylinders; c large and ventrally positioned thick circular cylinders; d small and ventrally positioned thick circular cylinders; e large and dorsally positioned thick el-

which weighs both principal tension ( 1 ) and principal compression ( 2 ) effects. In order to characterize the biomechanical performances of a given IPP, we examined the averaged values of the stresses transferred to the dorsal nerve roots and blood vessels. Since the deformation of the penis during IPP-aided erection varies among different cases, we dened the average stress at this region as

1 S

S 0

v.M .d ,

where the linear course of length S originates at the center of the dorsal face of the penis cross section above the dorsal vein and artery, crosses the dorsal nerve roots and the nerves of the tunica albuginea, and terminates at the apex of the corpus cavernosum Fig. 1b.

622

GEFEN, CHEN, and ELAD

liptic cylinders with an aspect ratio of 0.9 and the long axes directed toward the dorsal-lateral and ventral-medial aspects of the penis; f small and dorsally positioned thick elliptic cylinders with an aspect ratio of 0.9 and the long axes directed toward the dorsal-lateral aspect of the penis; g small and centrally positioned thick elliptic cylinders with an aspect ratio of 0.7 and the long axes aligned with the medial-lateral direction of the penis; and h small and centrally positioned thick elliptic cylinders with an aspect ratio of 0.7 and the long axes aligned with the dorsal-ventral direction of the penis. Based on measurements of commercial cylinder walls, thickness of a thick cylinder wall was taken as being 15% that of the internal radius, whereas for the thin cylinder, it was taken as being 5% of the radius. Large cylinders were built to occupy 60% of the cavernosal space, while small cylinders occupied 45% of it. The earlier set of eight geometrical congurations enabled analysis of various IPP designs by altering the material properties of the cylinder walls and the intraluminal cylinder pressures, as detailed in the following paragraphs. Cylinder Material Properties Inatable prostheses are generally made of thermoplastic elastomers. Silicone rubber and Bioex polyurethane a polymer which is more durable and less elastic than silicone are the most common materials in current use.7 Since this group of materials demonstrates a variety of highly nonlinear stress-strain relationships, it is difcult to establish any generalized method for determining characteristic material stiffness. In the present study, the generally accepted characterization of the Young modulus as an initial slope of the stress-strain curve was used, yielding elastic moduli in the range of 101000 MPa.11 Consequently, for the present analysis, we dened soft ( E 10 MPa, medium ( E 100 MPa, and hard ( E 1000 MPa material types. The Poisson ratio of the cylinders was set as 0.4 in all cases. Cylinder Intraluminal Operational Pressures Pescatory and Goldstein12 conducted in vitro experiments with several commercial IPPs and measured mean intraluminal cylinder pressures between 80 and 173 kPa at maximum inated volumes 1 kPa equals 7.5 mm Hg. These pressure values are signicantly higher than the physiological cavernosal pressure during normal erection, which is approximately 13.3 kPa.17 This difference appears to be mostly due to the fundamental differences between the behaviors of the natural anatomy of the penis and its structure post-IPP-implantation. In the natural penis, internal stresses during erection are borne not only by the main structural framework of the penis, the tunica albuginea,1,5 but also by the intracavernous structure of bromuscular columns which penetrate the cor-

pora and attach to the intracavernosal vasculature as well.12 During IPP implantation, most of this tissue is removed to provide space for the cylinders which become the main structural support. In order to examine the role of the cylinder intraluminal pressure, we simulated conditions of low ( P i 80 kPa and high ( P i 173 kPa ination pressures. RESULTS The model was utilized to study penile stress distributions following implantation of an IPP in each of the cases shown in Fig. 1. Results are presented during full IPP-aided erection and in terms of von Mises equivalent stress Eq. 1. Figure 3 demonstrates the resulting stress distributions during IPP-aided erections for dorsally positioned thin and thick circular cylinders subjected to low and high intraluminal pressures. The results for each case are shown for only one half of the symmetric transverse cross-section: the cylinders made of the hard material are shown on the left side and those of the soft one appear on the right. In all cases, the tunica albuginea, which is considered the structural framework of the penis,5 is shown to be the most stress-loaded biological component. With the decrease in stiffness of the prosthetic cylinders, i.e., from hard to soft material characteristics, the load carried by the tunica albuginea dramatically increases by a factor of approximately 8 Fig. 3a. Obviously, some of the elevated tunical stresses are also transferred to the dense network of nerves that is interlaced within this structure. The combination of soft cylinders and high ination pressures not only causes further loading of the ventral aspect of the tunica around the urethra, but also induces a region of concentrated compression stresses at the central part of the dorsal aspect of the penis Fig. 3b in the vicinity of the dorsal artery. When thin and soft cylinders are used, these phenomena become even more pronounced, as elevated compression stresses spread at the dorsal aspect of the penis, reaching up to the skin surface if intraluminal pressures are kept high Figs. 3c and 3d/. The thin cylinders also tend to lose their original circular cross section during ination, eventually adopting an egg-shaped cross-sectional area whose local small-radii edge is compressed against the dorsal nerve roots and blood vessels. Table 1 details the results of a quantitative performance analysis for the circular cylinders when implanted at the dorsal aspects of the corpus , IPP cavernosa. According to the criterion of min cylinders ideally should be thick, made of a hard elastomer, and inated to low intraluminal pressures. Reduction of the cylinder thickness by 65% increased the mean stress values by a factor of about 2.1. Reduction of the Young modulus of the cylinders by one order of magnitude increased the average stress values by a factor of

Optimization of Design and Positioning of Penile Prostheses

623

FIGURE 3. Distribution of von Mises stresses during IPPaided erection for dorsally positioned circular cylinders with a thick walls subjected to low intraluminal pressure; b thick walls subjected to high intraluminal pressure; c thin walls subjected to low intraluminal pressure; and d thin walls subjected to high intraluminal pressure. For each of the earlier cases, three diagrams are shown: the left one is the stress distribution for the hard cylinders, the middle one is for the soft cylinders, and the right one presents curves of stresses along the line S logarithmic scale for both the hard solid line and soft dashed line cylinders.

about 2.3. Alteration of the intraluminal pressure from low to high values also raised the stresses by a factor of about 1.4. In order to examine if the biomechanical compatibility of circular cylinders could be enhanced by positioning them as far as possible from the dorsal nerve roots and blood vessels, we analyzed cases of large and small cylinders positioned lower within the corpora cavernosa, toward the ventral aspect of the penis Figs. 1c and 1d. Analysis of the stress distributions that had resulted due to ination of the large cylinders showed that positioning of the cylinders close to the ventral aspect of the corpora only about 4 mm lower than in Figs. 1a ) reduced stresses at the vicinity of the dorsal nerves (

by as much as 20%40%, compared with the reference case Figs. 4a and 4b. Such positioning, however, also increased the stresses around the urethra. Use of cylinders with smaller diameter Figs. 4c and 4d not only effectively eliminated this problem, but also minimized the dorsal stresses to negligible levels Table 1. It should be borne in mind that small-diameter cylinders will also have relatively less penile rigidity during IPPaided erection. In the following stage, interaction of the penis with elliptic cylinders was analyzed and compared to performances of circular cylinders. Since the combination of thick walls, hard material, and low ination pressure was shown to provide the best results for circular cylinders,

624

GEFEN, CHEN, and ELAD

TABLE 1. Average von Mises stresses through a representative section AA in the region of primary penile nerves and blood vessels for dorsally and ventrally positioned circular cylinders. Average stress transferred to nerves (kPa) and blood vessels Dorsally positioned cylinders Thin cylinder wall Cylinder material E (MPa) 10 100 1000 Low pressure 176 71 58 High pressure Thick cylinder wall Low pressure High pressure

246 128 160 103 46 65 74 16 23 Ventrally positioned thick cylinders Small cylinders Low pressure 2 High pressure 4

large, dorsally located thick circular cylinders with identical material properties. Use of small, laterally medially orientated elliptical cylinders limited the inversion of the ellipse poles Fig. 5c. This conguration, however, induced highly elevated stress sites at the medial, ventral, and dorsal aspects of the penis, thereby providing the poorest results among the set of elliptic cylinders in terms of Table 2. Rotation of the cylinders so that their long axis aligned with the dorsal and ventral directions moderately reduced stresses at the dorsal aspect of the penis, but inversion of the poles during ination caused stress concentrations to appear at both lateral aspects. DISCUSSION It is commonly accepted that adequate mechanical interaction between the articial implant and the surrounding biological tissues in vivo plays an important role in the success of insertion of a penile prosthesis. It has been shown that mechanical factors such as cylinder collapse or tissue stiffening are signicant causes for severe penile pain due to stress concentrations and excessive local deformations during IPP-aided erection.3,8 Elevated local stresses and deformations may also accelerate cylinder wear and malfunction in the form of aneurysms or tubing uid leaks that were reported to occur in 1%4% of the cases.9,15 In view of the earlier circumstances which eventually require a traumatic and often expensive revision procedure to replace the faulty implant, it is highly recommended that both the design of the IPP and the surgical positioning of its cylinders within the penis be aimed towards minimal stress transfer to nerves and blood vessels during IPP-aided erections. In order to optimize the in vivo performances of IPPs by reducing the stresses transferred to the penile soft tissues as much as possible, we utilized a recently developed structural model of the penis/IPP complex. The model was adapted to analyze not only the effect of the IPP design on its biomechanical compatibility, but also the outcomes of surgical decisions of positioning and alignment of the cylinders within the corpora. The simulation of stress distributions at various postimplantation conditions indicated that most of the load during IPPaided erection is carried by the dorsal and lateral aspects of the tunica albuginea, which are also the most vulnerable sites since they contain the penile nerve roots and major blood vessels. Our objective was to minimize stresses at these regions, and the results of the present analysis suggest that an ideal IPP could be designed to more closely mimic the structural behavior of the human penis by achieving functional rigidity at lower cylinder pressures and preferably at pressures that approach the physiological value of about 13.3 kPa. The results also

Large cylinders Low pressure 1000 10 High pressure 19

all elliptic cylinders were constructed using the same set of design parameters. Hence, the analysis of elliptic cylinders was focused on the effect of prosthesis dimensions, position, and orientation, aiming to further improve the performances of circular cylinders. The resulting stress distributions and prosthesis deformations during IPP-aided erections that were obtained by using the elliptic cylinders are depicted in Fig. 5. The results of the performance analysis for these cylinders based on ) are detailed in Table 2. These the criterion of min results clearly demonstrated that all the elliptic cylinders tested in this study generated signicantly higher stresses at the vicinity of the dorsal nerve roots and blood vessels, compared with circular cylinders of identical thickness, material properties and intraluminal pressure. In an attempt to minimize stresses in the vicinity of dorsal nerve roots and blood vessels, large and small elliptic cylinders were prepositioned with the large-radii curve of the ellipse facing the dorsal-central surface of the penis Figs. 1e and 1f. This approach was shown to be unsuccessful since the poles of the elliptic cross section are inverted due to the nonhomogeneous resistance of the surrounding tissues during ination of the cylinders. Thus, at the end of the ination process, as articial erection is achieved, the small-radii boundaries of the cylinders are directly facing toward the dorsalcentral surface of the penis, loading the dorsal nerve roots and blood vessels Figs. 5a and 5b. This phe obtained nomenon is well reected by the values of for the for these two cases Table 2. For instance, large inclined elliptic cylinders is as much as three times higher than the corresponding value obtained for the

Optimization of Design and Positioning of Penile Prostheses

625

FIGURE 4. Distribution of von Mises stresses kPa during IPP-aided erection for ventrally positioned and thick-walled circular cylinders: a large cylinders subjected to low intraluminal pressure; b large cylinders subjected to high intraluminal pressure; c small cylinders subjected to low intraluminal pressure; and d small cylinders subjected to high intraluminal pressure. For each of the earlier cases, three diagrams are shown: the left one is the stress distribution for the hard cylinders, the middle one is for the soft cylinders, and the right one presents curves of stresses along the line S logarithmic scale for both the hard solid line and soft dashed line cylinders.

indicate that both thickness and stiffness of the cylinders should be kept sufciently high 15% of the radius and E 1000 MPa, respectively in order to eliminate large cylinder deformations during ination, the result of which may be substantial contact stresses between the cylinders and the cavernosal tissue. Smaller prosthetic cylinders, i.e., occupying about 45% of the cavernosal space, may be advantageous in terms of reducing dorsal stresses, but relatively less penile rigidity should be expected. Insofar as this trade-off relation may have little negative effect on sexual satisfaction while it may be highly efcacious in avoiding the development of penile pains during prosthetic-aided erection, further investigation is warranted.

Signicant decreases of up to 40% relative to the reference case in the mean stress values at the dorsal sensitive part was shown to result from the selection of a 4 mm lower position for the cylinders. Current surgical techniques do not allow for very highly accurate manipulation of cylinder positioning, mainly because the dilatators used to clear intracavernosal space for the cylinders during the operation are not designed to provide precise control of the dilatation positioning. However, based on the simulation results, we believe that the implanter should aim toward the ventral aspect, considering that even minimal lowering of the cylinder position would be highly advantageous in terms of decreasing the stresses within the dorsal penile tissues. This and other possible

626

GEFEN, CHEN, and ELAD

TABLE 2. Average von Mises stresses through a representative section AA in the region of primary penile nerves and blood vessels for thick-walled and hard elliptic cylinders positioned at different alignments and subjected to low intraluminal pressure P i 80 kPa. Cylinder type and alignment Large and inclined cylinders [Fig. 1(e)] Small and inclined cylinders [Fig. 1(f)] Small and laterally medially aligned cylinders [Fig. 1(g)] Small and dorsally ventrally aligned cylinders [Fig. 1(h)]

(kPa)
48 38 72 35

FIGURE 5. Distribution of von Mises stresses during IPPaided erection for elliptic and thick-walled hard cylinders subjected to low intraluminal pressure: a large and inclined cylinders Fig. 1e; b small and inclined cylinders Fig. 1f; c small and laterally medially aligned cylinders Fig. 1g; and d small and dorsally ventrally aligned cylinders Fig. 1h. The dashed, white line marks the neutral cylinder geometry for each case accid mode.

implications of the present ndings on the surgical technique and tools for IPP implantation should now be experimentally evaluated in animal models in order to test their validity in vivo. Based upon the set of congurations tested in the framework of the present study, it was clearly demonstrated that circular cylinder cross sections allow far greater biomechanical compatibility of the IPP with the penis structure compared with elliptic cylinders. Inversion of the elliptic poles was observed in three out of four congurations containing elliptic cylinders due to nonhomogenous resistance of the surrounding tissues to ination. This phenomenon, which not only causes difculties in predicting the nal steady state of deformation of the cylinders in vivo, but also induces undesired

stress concentrations which compress nerves and blood vessels in the regions of the inverted poles, suggests that an elliptic cross section is less suitable for IPP cylinders. Nevertheless, clinical studies comparing performances of commercially available prostheses of both types in adequate numbers of patients are still required before specic manufacturing recommendations could be established. Undesirable alteration of the cross-sectional shape of the prosthetic cylinders toward a more elliptic prole increases the risk for localized collapse under compression conditions, mainly due to the effect of buckling. Buckling collapse commonly induces sharp geometry transitions within the deformed prosthesis, as demonstrated by magnetic resonance imaging MRI by Moncada et al.8 These sharp geometry transitions are clearly capable of inducing sites of concentrated mechanical stresses within the surrounding penile tissues, in which mechanical failure could appear in the form of microtears. If some contamination of the cylinders occurs during the implantation procedure, infection will develop within these sites, leading to pain. Since the present study deals with a sterile model, the issue of infection is not applicable, and the discussion in the following paragraph is limited to the structural behavior of the penis/ prosthesis complex. Buckling of an inatable penile prosthesis during coitus may be the result of axially applied forces associated with initial vaginal penetration or axial compressive and lateral constraining forces induced by contact with the vaginal walls during continued intercourse. The Euler formula for buckling provides the critical force which induces buckling of a column, F b EI / L 2 , where E is the Young modulus, L is the column length, and I is the second moment of inertia of the column cross section. The buckling force is, therefore, linearly proportional to the moment of inertia of the column cross section. In the case of a circular column, the moment of inertia is I c D 4 /64 where D is the column diameter, while in the case of an elliptic column, the moments of inertia are I x a 3 b /64 and I y ab 3 /64 where a and b are the

Optimization of Design and Positioning of Penile Prostheses

627

dimensions of the ellipse on its long and short axes, respectively. It can be shown that for any given pair of circular and elliptic columns, I c is always greater than min( I x , I y , provided that the cross-sectional area of both columns is identical. Accordingly, elliptic cylinders will normally buckle in response to smaller loads compared with circular cylinders of the same cross-sectional area. Since the present study showed that cylinders of circular cross-sectional shape could adopt a more elliptic prole due to ination, the risk of buckling and localized collapse in these cases would need to be considered. In order to deal with the structural complexity of the penis and, especially, with the lack of experimental data on the nonlinear behavior of its tissues, assumptions were made for the purposes of simplication, and this should be kept in mind while interpreting the results. Penile tissues were assumed to be isotropic and linearly elastic. This assumption is highly likely to be adequate for analysis of the tunica albuginea, which is a dense parallel-bered collagenous tissue5 and is, therefore, expected to produce a stress-strain curve with a considerably long linear part. Since the tunica albuginea was shown to be the main load-bearing structure in the penis, predictions of the present analysis could be considered realistic. Nevertheless, in order to estimate the effects of some possible nonlinear elasticity of the tunica on the simulation results, we replaced the tunical elastic modulus with a nonlinear constitutive law of a ligamentous tissue,11 which is of similar biological structure. The resulting stresses were greater by 10%30%, depending on the prosthesis type and location. In view of the earlier, experimental data that reveal the nonlinear characteristics of the penile tissues are needed. After these data have become available, a quasilinear viscoelastic approach could be useful to obtain a more accurate representation of the structural behavior of the penis. The computational methodology of this study presents powerful biomechanical tools for optimal design, development, and examination of penile implants to improve postimplantation outcomes. Implementation of this or similar approaches as an integral part of the engineering design process of an IPP may not only enhance its performances, but could also yield important surgical guidelines in terms of the preferred position and alignment of the prosthesis within the penis. In the future, the present approach could be further expanded into a pre-operative routine for evaluation of surgical procedures for individual patients. In such cases, the real cross-sectional anatomy of a specic patient acquired by ultrasound or MRI will be directly incorporated in a nite element model. The surgical procedure can then be individually planned and its biomechanical consequences can be simulated before undertaking the actual intervention.

CONCLUSIONS The present numerical simulations demonstrate that optimal selection of engineering designed parameters for the IPP e.g., cylinder geometry and stiffness can eliminate substantial cylinder-tissue contact stresses, thereby reducing the likelihood for postoperative complications. By allowing control of anatomical structural parameters, as well as those of the prosthesis, the methodology of this study could also be applied to test the biomechanical implications of different surgical positions of the prosthetic cylinders. Hence, the use of the present modeling approach in the analysis of penile prosthesis implantation procedures signicantly increases clinical and research opportunities by delineating what are the experimental studies required for the establishment of specic manufacturing and medical recommendations.

REFERENCES
Bitsch, M., B. Kromann-Andersen, J. Schou, and E. Sjontoft. The elasticity and tensile strength of the tunica albuginea. J. Urol. (Baltimore) 143:642645, 1990. 2 Campbell, J. N., S. N. Raja, R. A. Meyer, and S. E. MacKinnon. Myelinated afferents signal the hyperalgesia associated with nerve injury. Pain 32:8994, 1988. 3 Gefen, A., J. Chen, and D. Elad. Stresses in the normal and diabetic human penis following implantation of an inatable prosthesis. Med. Biol. Eng. Comput. 37:625631, 1999. 4 Govier, F. E., R. P. Gibbons, R. J. Correa, T. R. Pritchett, and L. D. Kramer. Mechanical reliability, surgical complications, and patient and partner satisfaction of the modern three-piece inatable penile prosthesis. Urology 52:282286, 1998. 5 Hsu, G. L., G. Brock, L. Martinez-Pineiro, B. von Heyden, T. Lue, and E. A. Tanagho. Anatomy and strength of the tunica albuginea: Its relevance to penile prosthesis extrusion. J. Urol. (Baltimore) 151:12051208, 1994. 6 McLaren, R. H., and D. M. Barrett. Patient and partner satisfaction with the AMS 700 penile prosthesis. J. Urol. (Baltimore) 147:6265, 1992. 7 Merrill, D. C. Mentor inatable penile prosthesis. Urology 22:504505, 1983. 8 Moncada, I., J. J. Hernandez, L. Hernandez, J. Lafuente, E. Lledo, E. Rodriguez, and I. Saenz De Tejada. Buckling of cylinders may cause prolonged penile pain after prosthesis implantation: A case control study using magnetic resonance imaging of the penis. J. Urol. (Baltimore) 160:6771, 1998. 9 Nickas, M. E., R. Kessler, and J. N. Kabalin. Long-term experience with controlled expansion cylinders in the AMS 700CX inatable penile prosthesis and comparison with earlier versions of the Scott inatable penile prosthesis. Urology 44:400403, 1994. 10 Niv, D., and M. Devor. Does the blockade of surgical pain preempt postoperative pain and prevent its transition to chronicity? JASP Newsletter 6:27, 1993. 11 Park, J. B. Biomaterials Science and Engineering. New York, Plenum, 1984. 12 Pescatory, E. S., and I. Goldstein. Intraluminal device pressures in 3-piece inatable penile prostheses: The patho1

628

GEFEN, CHEN, and ELAD

16

physiology of mechanical malfunction. J. Urol. (Baltimore) 149:295300, 1993. 13 Quesada, E. T., and J. K. Light. The AMS 700 inatable prosthesis: Long-term experience with the controlled expansion cylinders. J. Urol. (Baltimore) 149:4648, 1998. 14 Scott, F. B., W. E. Bradley, and G. W. Timm. Management of erectile impotence. Use of implantable inatable prosthesis. Urology 2:8082, 1973. 15 Steinkohl, W. B., and G. E. Leach. Mechanical complications associated with Mentor inatable penile prosthesis. Urology 38:3234, 1991.

Torebjork, H. E., L. E. R. Lundberg, and R. H. LaMotte. Central changes in processing of mechanoreceptive input in capsaicin-induced secondary hyperalgesia in humans. J. Physiol. 448: 765780, 1992. 17 Venegas, J. G., M. P. Sullivan, S. B. Yalla, and M. A. Vickers. Assessment and modeling of the physical components of human corporovenous function. Am. J. Physiol. 269:21092123, 1995. 18 Whalen, R. K., and D. C. Merrill. Patient satisfaction with Mentor inatable penile prosthesis. Urology 37:531539, 1991.