Original Article Received: April 22, 2014 Accepted after revision: August 6, 2014 Published online: December 20, 2014
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
Magnetic Resonance-Visible Polypropylene Mesh for Pelvic Organ Prolapse Repair Kerstin A. Brocker a Florian Lippus d Céline D. Alt b Peter Hallscheidt e Fejér Zsolt g Irina Soljanik c Florian Lenz f Michael Bock d Christof Sohn a
Departments of a Obstetrics and Gynecology and b Diagnostic and Interventional Radiology, and c Section of Neuro-Urology, Spinal Cord Center, University Hospital Heidelberg, Heidelberg, d Radiology – Medical Physics, University Medical Center Freiburg, Freiburg, e Radiological Department, Alice Hospital, Darmstadt, and f Department of Obstetrics and Gynecology, Hospital Hetzelstift, Neustadt an der Weinstrasse, Germany; g Institute of Anatomy and Cell Biology, Semmelweis University of Budapest, Budapest, Hungary
Abstract Aim: To develop a magnetic resonance (MR)-visible mesh using iron oxides and prove visibility. Methods: In a phantom study, a suitable iron oxide, Fe3O4 [iron(II,III) oxide] and FeOOH [iron(III) oxide-hydroxide], concentration was determined using relaxometric MR measurements of the transverse relaxation rates R2 and R2*. Next, a nonabsorbable mesh was designed from the MR-visible threads woven into a polypropylene mesh. The mesh was implanted into a fresh female cadaver via the transobturator route, and MR visibility was assessed with various MR pulse sequences in a clinical 3-tesla system. Results: Optimal contrast was achieved with Fe3O4 at 0.2 weight-% in all imaging sequences, and the optimal contrast was achieved in a 3D spoiled gradient-echo (fast low-angle shot) acquisition. In this concentration range
© 2014 S. Karger AG, Basel 0378–7346/14/0792–0101$39.50/0 E-Mail [email protected] www.karger.com/goi
the apparent transverse relaxation rate R2* is below 10 ms. The mesh was visible in the cadaver on T1-weighted 3D spoiled gradient-echo images and T1-weighted fast spinecho images. Conclusion: Mesh materials can be manufactured to be visible on MR with a negative contrast. Fe3O4 meshes could simplify follow-up examinations and help diagnose origins of postsurgical lesions after urogynecological procedures with mesh material. © 2014 S. Karger AG, Basel
Introduction
The number of elderly people is steadily increasing, and there has been a corresponding growth in the number of common diseases such as female pelvic organ prolapse or stress urinary incontinence. Pelvic organ prolapse is a multifactorial disorder and the vaginal wall of concerned patients is often described as weakened or altered [1, 2]. These pelvic floor disorders are treated either with noninvasive Kerstin Aneta Brocker, MD, MSc Department of Obstetrics and Gynecology University of Heidelberg Medical School INF 440, DE–69120 Heidelberg (Germany) E-Mail kerstin.brocker @ med.uni-heidelberg.de
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
Key Words Female cadaver · Ferric compounds · Magnetic resonance imaging · Pelvic organ prolapse · Surgical mesh
Color version available online
a
b
Fig. 1. a Photograph of the mesh materials.
From bottom left to top right: samples 1–4. b Photograph of the prototype mesh with
six fixation points. Both the dotted lines and the zigzag lines schematically illustrate the position of the MR-visible threads in the mesh and are not absorbable.
Materials and Methods This experimental study was designed to develop MR-visible mesh material and to assess its visibility on MRI. We therefore determined suitable concentrations of iron oxide labeling in a phantom study. Based on these results, a labeled, MR-visible thread was woven into a mesh. To assess MR visibility in a realistic setting, the mesh was then implanted in a fresh female cadaver and MR images were acquired using optimized MRI protocols. The institutional ethical board approved the cadaver study (IRB No. 53/2010).
102
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
Substance Development and Magnetic Particle Concentration Two iron oxide substances were tested to make the meshes MR visible: iron(III) oxide-hydroxide (FeOOH) and iron(II,III) oxide (Fe3O4). These substances were chosen for their biocompatibility, as both iron oxides are used as pigments (FeOOH as C.I. pigment yellow 42, and Fe3O4 as C.I. pigment black 11) and are internationally approved [13]. Initially, measurements of the transverse relaxation rates R2 and R2* were performed to determine the optimal concentration of the iron oxide to be used in the fabrication of the mesh material. For this purpose, different concentrations of FeOOH and Fe3O4 in both gel (gelatin) and palm oil were used, ranging between 0 and 10% for FeOOH and between 0 and 0.2% for Fe3O4. R2 was measured with a 32-echo spin-echo sequence, and R2* was determined from a 8-echo gradient-echo data set. Relaxation rates were calculated from a signal fit to the exponential signal decay as a function of echo time (TE). From the relaxation data, the minimum concentration was determined that provides a R2 of 100 s–1 or more, i.e. which shortens the transverse relation time T2 = 1/R2 to less than 10 ms. Since relaxation data were only accessible in gel and in palm oil, a final test was performed with the iron oxides embedded in the thread carrier material. Therefore, four threads were manufactured from polypropylene granulate: two with FeOOH at concentrations of 2.5% (sample 2) and 5% (sample 1), and two with Fe3O4 at 0.02% (sample 4) and 0.2% (sample 3). Four bands were woven from the polypropylene threads as they are used in Serasis PP® (Serag Wiessner, Naila, Germany) mesh implants (samples 1–4; fig. 1a). For MRI, the bands were placed in a water bath, and images were acquired in a 1.5-tesla clinical MR system (Tim Symphony; Siemens Healthcare, Erlangen, Germany) using three standard imaging sequences: a 3D spoiled gradient-echo sequence [fast low-angle shot (FLASH)], a fast spin-echo sequence, and a balanced steady state free precession sequence or true fast imaging with steady precession (trueFISP). Imaging parameters of these three pulse sequences used to test the MR visibility of the mesh implants were repetition time (TR) = 15 ms, TE = 6 ms, BW = 179 Hz/px for FLASH 3D; TR = 4,000 ms, TE = 110 ms, BW = 260 Hz/ px for turbo spin-echo, and TR = 3.8 ms, TE = 1.79 ms, BW = 751 Hz/px for trueFISP. A field of view of 400 mm and a matrix of 512 × 512 was applied in all sequences. To optimally visualize and compare the signal void produced by the iron oxide in the bands, minimum intensity projections were calculated from all three 3D data sets.
Brocker/Lippus/Alt/Hallscheidt/Zsolt/ Soljanik/Lenz/Bock/Sohn
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
methods such as pelvic floor exercises or pessary therapy, or with classical surgical methods such as vaginal sacrospinous fixation, abdominal sacrocolpopexy, or Burch colposuspension [3, 4]. Recently, partially absorbable vaginal meshes and tension-free vaginal slings have been introduced into the portfolio of urogynecological surgery [3, 5]. The current literature on complications after mesh surgery includes reports on postoperative pain, hematoma, nerve distractions, mesh shrinking, de novo dyspareunia, and infections [6, 7]. To assess surgical outcome and postoperative complications, the patients’ subjective rating is combined with clinical examination and ultrasound imaging. Current imaging modalities reach their limits, however, when they are used to evaluate the mesh arms and their fixation point, especially when trying to verify a correct location or diagnosing misplacement [8]. Magnetic resonance imaging (MRI) could be a promising imaging alternative to further investigate the implanted mesh [9]. Nevertheless, current meshes are hardly visible in magnetic resonance (MR) images besides a few exceptions such as the polyvinylidene fluoride meshes recently described [10–12]. The aim of this study is therefore to develop a mesh material for reconstructive surgery of female pelvic organ disorders that uses iron oxides to make it visible on MR.
Results
Substance Development and Magnetic Particle Concentration Figure 2 shows R2 as a function of iron oxide concentration in gelatin. As expected, in this concentration range R2 increases linearly with concentration, so that an increase in iron oxide concentration will improve mesh visibility. However, as higher iron oxide concentrations might have a negative effect on thread stability and could also create excessively large artifacts, two different concentrations were chosen for thread production. These concentrations were 2.5 and 5% for FeOOH and 0.02 and 0.2% for Fe3O4. For both FeOOH concentrations and the 0.2% Fe3O4 concentration, an R2 of MR-Visible Polypropylene Mesh
Color version available online
100 80 60 40 20 0
FeOOH Fe3O4 0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Concentration (%)
Fig. 2. Transverse relaxation rate R2 as a function of the concentra-
tion of Fe3O4 and FeOOH in gelatin.
Table 1. Imaging parameters of the sequences performed to test
the MR visibility of the mesh implanted in the female cadaver
T1 3D WAVE transversal T2 3D VISTA transversal T1 tse transversal T2 tse sagittal T2 tse transversal SSH coronal T1 tse fs transversal
TR, ms
TE, ms
ST, mm
4.8 1,979.1 493 1,865 2,117.2 1,918.5 590
2.4 200 10 80 70 120 10
1.9 4 4 4 4 4 4
WAVE = Spoiled gradient-echo sequence; VISTA = volumetric isotropic turbo spin-echo acquisition; SSH = ultrafast spin-echo sequence; tse = turbo spin-echo; fs = fat saturated; ST = slice thickness.
100 s–1 or more was expected (i.e. a T2 of 10 ms or less; fig. 2). The mentioned concentrations of FeOOH and Fe3O4 are weight percent. While figure 3a–c shows that all the bands woven from these threads are visible in the MR images, sample 3 (Fe3O4 at 0.2%) achieves optimal contrast in all the standard imaging sequences. In this concentration range the apparent transverse relaxation rate T2* (which determines the contrast in gradient-echo images) is well below 10 ms, so that the labeled meshes are always delineated with a negative contrast in gradient-echo image data. Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
103
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
Cadaver Experiment and Proof of MR Visibility In order to visualize the newly constructed MR-visible mesh in the pelvic floor, it was implanted into a postmenopausal female cadaver using the transobturator approach and securing the proximal mesh arms through the sacrospinous ligament on both sides. For secure and standardized implantation, the mesh was placed according to the manufacturer’s recommendation and employing reusable introducer needles (Serag Wiessner®). The surgeon (F.L.) implanting the mesh was experienced in the field of urogynecological surgery (>1,000 vaginal mesh implantations). To fill the bladder, a transurethral catheter was inserted and blocked after filling the bladder with 200 ml water. The rectum was cleaned of stool and ultrasound gel was inserted for better demarcation of the rectal walls. Afterwards the anus was closed with several stiches to keep the ultrasound gel from being expelled. After full implantation of the vaginal mesh, ultrasound gel was inserted into the vagina for better demarcation of the vaginal walls. The cadaver was then imaged in a 3-tesla MR system (Achieva, Philips) to assess the MR visibility of the mesh. Sequences were performed that are used in clinical routine: a T1-weighted 3D spoiled gradient-echo sequence, a T2-weighted sequence with volumetric isotropic turbo spin-echo acquisition, T1-weighted turbo spin-echosequences with and without fat saturation, T2-weighted turbo spin-echo sequences, and a T2-weighted ultrafast spin-echo sequence. The parameters are listed in table 1.
120
R2 (1/s)
A full mesh was produced from the band with the best image contrast. It consisted of conventional partially absorbable threads and MR-visible threads (fig. 1b). The mesh was designed based on a Seratom E PA® (Serag Wiessner) mesh, such as what is used for combined reconstruction of a descended uterus (level I defect according to DeLancey et al. [14]) with the help of an additional sacrospinous fixation and an anterior vaginal wall defect (level II defect according to DeLancey et al. [14]) [15]. The MR-visible threads were woven into all mesh arms and partially into the mesh body using a combination of straight threads and threads woven in a zigzag pattern (fig. 1b). The non-MR-visible threads are absorbable.
a
c
b
Fig. 3. Minimum intensity projection images from 3D FLASH (a), turbo spin-echo (b) and TrueFISP (c) data. In
all pulse sequences sample 3 has the highest contrast.
a
b
c
d
echo) sequence, the arrows mark the four transobturator mesh arms. c T1-weighted 3D WAVE sequence, the arrow shows the
Cadaver Experiment and Proof of MR Visibility Surgical implantation of the MR-visible mesh into the female cadaver was performed in 45 min. The woven iron oxide filaments in the mesh were best visible in the T1weighted 3D spoiled gradient-echo sequence and the T1weighted fast spin-echo sequence. Figure 4a–d presents images of the mesh in the cadaver showing the anterior and medial arms of the mesh implanted via the transobturator route or displaying the zigzag pattern of the woven threads near the cervix of the uterus. 104
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
zigzag pattern of the mesh. d T1-weighted 3D WAVE, the arrow shows the mesh visible as a line (negative echo signal) between the cervix and rectum. AH = Artificial hip; B = bladder; DC = dilated catheter in bladder.
Discussion
In this experimental study, we made MR-visible mesh material for implantation into the pelvic floor. Its purpose is to overcome the current limitations in visibility on MR or ultrasound of meshes used in pelvic reconstructive surgery. The mesh material was optimized for rapid MRI to facilitate better detection of mesh placement in the pelvic floor and potentially find the origin of complications related to mesh repair. In the current literature, meshes in Brocker/Lippus/Alt/Hallscheidt/Zsolt/ Soljanik/Lenz/Bock/Sohn
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
Fig. 4. a T2-weighted, the arrows depict both medial mesh arms. b T1-weighted 3D WAVE (T1-weighted 3D spoiled gradient-
general and misplaced ones in particular might be the cause of nerve irritations, uncertain postsurgical pelvic floor pain, or dyspareunia [6, 16, 17]. The published rates of complications such as mesh erosion, de novo dyspareunia, and pelvic organ prolapse recurrence are heterogeneous [6, 7]. Until now it has often not been possible to determine a definite reason for their occurrence because the complete location of the mesh and its anchoring mesh arms were not visible [6, 7, 17]. Iron oxides were chosen as labeling compounds due to their known biocompatibility [18]. Iron is a natural element found in the human body, and iron oxides are widely approved for human use. For example, iron oxides are licensed as food coloring agents under the number E172, have been approved by the FDA for use in contact lenses (§ 73.3125), and they form the contrast-generating compound in the MRI contrast agent Lumirem® (ferumoxsil; Guerbet Groupe, France), which is approved for oral and rectal application at concentrations up to 175 mg Fe/l [9, 19, 20]. Thus, incorporation of iron oxides into the mesh material does not seem to pose any risks of side effects, such as might be present with other MR contrast agents, such as Gd-DTPA from which the toxic element gadolinium could potentially be washed out of the mesh during the long exposure times [21–23]. Relaxometric studies show that a concentration of 0.2 weight-% Fe3O4 is a good compromise between signal decay in the proximity of the threads and clear visibility of the mesh. Therefore Fe3O4 was selected over FeOOH as the labeling compound since much higher concentrations would have been needed with FeOOH to create the negative image contrast. At all the MR settings subsequently employed in our study, the MR visibility of the labeled mesh was demonstrated, and spoiled gradientecho sequences showed the highest contrast between the mesh and surrounding tissue, which was also confirmed in the cadaver experiment. In theory, more mesh material with other iron oxide concentrations could have been manufactured to evaluate the contrast mechanism in more detail. In addition, more experimental images are needed to determine the effect on visibility at other field strengths. However, the current results already indicate that the concentration of 0.2 weight-% is sufficient to reliably create the desired contrast, and a higher concentration might only compromise visibility and mesh stability. Preliminary tests at higher field strengths indicate that the contrast behavior is very similar and that the meshes will also be suitable for higher field strengths such as those we performed in the cadaver experiment using a 3-tesla machine. To as-
sess the reproducibility of our results, more than one cadaver study should be performed. While cadaver studies can be generally applied, they often have limitations due to the presence of entrapped air, which can create artifacts similar to those of meshes [24]. Additionally, in order to learn more about the biological stability of this mesh, future experimental studies on live animal models seem useful and are initiated as described by other authors [11, 12]. The present study demonstrates a new iron oxide-labeled mesh with proven MR visibility in an experimental 3-tesla MR scenario with a fresh human female cadaver for use in the reconstructive surgery of pelvic floor disorders. Although more follow-up experiments are needed, the results of this study might be helpful to facilitate the introduction of the mesh in human studies. In addition, the introduction of an MR-visible mesh in pelvic floor reconstructive surgery may help to precisely assess the anatomical and functional relations of implanted meshes and to correlate it with patients’ symptoms. The cost-effectiveness of visible meshes should also be evaluated with regard to the economic benefit of successfully treated complications after pelvic floor reconstructive surgery. In conclusion, the MR visibility of the mesh was achieved in a 3-tesla MR setting with a fresh human female cadaver when an adequate concentration of iron oxide particles was added to the mesh material. MR-visible meshes might help to analyze and quantify objectively the surgical outcome and complications after mesh repair.
MR-Visible Polypropylene Mesh
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
Acknowledgements Special thanks go to Dr. Anna Németh (Institute of Anatomy and Cell Biology, Semmelweis University of Budapest, Budapest, Hungary) and Prof. Gabor Rudas (MR Research Center, Szentágothai Knowledge Center, Semmelweis University of Budapest, Hungary).
Disclosure Statement
105
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
K.A. Brocker received a research fellowship from the University of Heidelberg and a research scholarship from the Forum Urodynamicum e.V. to perform this work. K.A. Brocker received research funding from Serag Wiessner to perform this work. F. Lenz received speaking honoraria for lectures and speeches on international meetings from Serag Wiessner. M. Bock has received grants from Siemens Healthcare outside of the research outlined in this study. No money was used for the research presented in this manuscript. For all the remaining authors, no conflict of interest was declared.
References
106
9 Arsalani N, Fattahi H, Nazarpoor M: Synthesis and characterization of PVP-functionalized superparamagnetic Fe3O4 nanoparticles as an MRI contrast agent. Express Polym Lett 2010;4:329–338. 10 Ciritsis A, Hansen NL, Barabasch A, Kuehnert N, Otto J, Conze J, Klinge U, Kuhl CK, Kraemer NA: Time-dependent changes of magnetic resonance imaging-visible mesh implants in patients. Invest Radiol 2014; 49: 439–444. 11 Kuehnert N, Kraemer NA, Otto J, Donker HC, Slabu I, Baumann M, Kuhl CK, Klinge U: In vivo MRI visualization of mesh shrinkage using surgical implants loaded with superparamagnetic iron oxides. Surg Endosc 2011; 26:1468–1475. 12 Endo M, Feola A, Sindhwani N, Manodoro S, Vlacil J, Engels AC, Claus F, Deprest JA: Mesh contraction: in vivo documentation of changes in apparent surface area utilizing meshes visible on magnetic resonance imaging in the rabbit abdominal wall model. Int Urogynecol J 2013;25:737–743. 13 Title 21, Food and Drugs Chapter I, Food and Drug Administration, Department of Health and Human Services, Subchapter A, Part 73, Listing of Color Additives Exempt from Certification, Subpart D, Medical Devices. 73.3125 Iron Oxides. 2012 (revised April 1, 2013). http://www.accessdata.fda.gov/scripts/cdrh/ cfdocs/cfCFR/CFRSearch.cfm?fr=73.3125 (accessed January 11, 2014). 14 DeLancey JO: Anatomic aspects of vaginal eversion after hysterectomy. Am J Obstet Gynecol 1992; 166: 1717–1724; discussion 1724– 1728. 15 Huddleston HT, Dunnihoo DR, Huddleston PM 3rd, Meyers PC Sr: Magnetic resonance imaging of defects in DeLancey’s vaginal support levels I, II, and III. Am J Obstet Gynecol 1995;172:1778–1782; discussion 1782–1784.
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
16 Jacquetin B, Cosson M: Complications of vaginal mesh: our experience. Int Urogynecol J Pelvic Floor Dysfunct 2009;20:893–896. 17 Margulies RU, Lewicky-Gaupp C, Fenner DE, McGuire EJ, Clemens JQ, Delancey JO: Complications requiring reoperation following vaginal mesh kit procedures for prolapse. Am J Obstet Gynecol 2008;199:678.e1–e4. 18 Gould P: Nanomagnetism shows in vivo potential. Nanotoday 2006;1:34–39. 19 Hallgren B, Sourander P: The effect of age on the non-haemin iron in the human brain. J Neurochem 1958;3:41–51. 20 Hentze MW, Caughman SW, Rouault TA, Barriocanal JG, Dancis A, Harford JB, Klausner RD: Identification of the iron-responsive element for the translational regulation of human ferritin mRNA. Science 1987;238:1570– 1573. 21 Weinmann HJ, Brasch RC, Press WR, Wesbey GE: Characteristics of gadolinium-DTPA complex: a potential NMR contrast agent. AJR Am J Roentgenol 1984;142:619–624. 22 Kornmesser W, Laniado M, Hamm B, Clauss W, Weinmann HJ, Schulz E, Wolf KJ, Felix R: Oral contrast medium for magnetic resonance tomography of the abdomen. II. Phase I clinical testing of gadolinium-DTPA (in German). Rofo 1987;147:550–556. 23 Okada S, Katagiri K, Kumazaki T, Yokoyama H: Safety of gadolinium contrast agent in hemodialysis patients. Acta Radiol 2001; 42: 339–341. 24 Dietrich O, Reiser MF, Schoenberg SO: Artifacts in 3-T MRI: physical background and reduction strategies. Eur J Radiol 2008;65:29– 35.
Brocker/Lippus/Alt/Hallscheidt/Zsolt/ Soljanik/Lenz/Bock/Sohn
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
1 Meijerink AM, van Rijssel RH, van der Linden PJ: Tissue composition of the vaginal wall in women with pelvic organ prolapse. Gynecol Obstet Invest 2013;75:21–27. 2 Martins P, Lopes Silva-Filho A, Rodrigues Maciel da Fonseca AM, Santos A, Santos L, Mascarenhas T, Natal Jorge RM, Ferreira AJ: Biomechanical properties of vaginal tissue in women with pelvic organ prolapse. Gynecol Obstet Invest 2013;75:85–92. 3 Maher C, Baessler K: Surgical management of anterior vaginal wall prolapse: an evidencebased literature review. Int Urogynecol J Pelvic Floor Dysfunct 2006;17:195–201. 4 Jelovsek JE, Maher C, Barber MD: Pelvic organ prolapse. Lancet 2007;369:1027–1038. 5 Ulmsten U: An introduction to tension-free vaginal tape (TVT) – a new surgical procedure for treatment of female urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct 2001;12:S3–S4. 6 Baessler K, Hewson AD, Tunn R, Schuessler B, Maher CF: Severe mesh complications following intravaginal slingplasty. Obstet Gynecol 2005;106:713–716. 7 Farthmann J, Watermann D, Niesel A, Funfgeld C, Kraus A, Lenz F, Augenstein HJ, Graf E, Gabriel B: Lower exposure rates of partially absorbable mesh compared to nonabsorbable mesh for cystocele treatment: 3-year followup of a prospective randomized trial. Int Urogynecol J 2013;24:749–758. 8 Shek KL, Dietz HP, Rane A, Balakrishnan S: Transobturator mesh for cystocele repair: a short- to medium-term follow-up using 3D/4D ultrasound. Ultrasound Obstet Gynecol 2008;32:82–86.
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
Magnetic Resonance-Visible Polypropylene Mesh for Pelvic Organ Prolapse Repair Kerstin A. Brocker a Florian Lippus d Céline D. Alt b Peter Hallscheidt e Fejér Zsolt g Irina Soljanik c Florian Lenz f Michael Bock d Christof Sohn a
Departments of a Obstetrics and Gynecology and b Diagnostic and Interventional Radiology, and c Section of Neuro-Urology, Spinal Cord Center, University Hospital Heidelberg, Heidelberg, d Radiology – Medical Physics, University Medical Center Freiburg, Freiburg, e Radiological Department, Alice Hospital, Darmstadt, and f Department of Obstetrics and Gynecology, Hospital Hetzelstift, Neustadt an der Weinstrasse, Germany; g Institute of Anatomy and Cell Biology, Semmelweis University of Budapest, Budapest, Hungary
Abstract Aim: To develop a magnetic resonance (MR)-visible mesh using iron oxides and prove visibility. Methods: In a phantom study, a suitable iron oxide, Fe3O4 [iron(II,III) oxide] and FeOOH [iron(III) oxide-hydroxide], concentration was determined using relaxometric MR measurements of the transverse relaxation rates R2 and R2*. Next, a nonabsorbable mesh was designed from the MR-visible threads woven into a polypropylene mesh. The mesh was implanted into a fresh female cadaver via the transobturator route, and MR visibility was assessed with various MR pulse sequences in a clinical 3-tesla system. Results: Optimal contrast was achieved with Fe3O4 at 0.2 weight-% in all imaging sequences, and the optimal contrast was achieved in a 3D spoiled gradient-echo (fast low-angle shot) acquisition. In this concentration range
© 2014 S. Karger AG, Basel 0378–7346/14/0792–0101$39.50/0 E-Mail [email protected] www.karger.com/goi
the apparent transverse relaxation rate R2* is below 10 ms. The mesh was visible in the cadaver on T1-weighted 3D spoiled gradient-echo images and T1-weighted fast spinecho images. Conclusion: Mesh materials can be manufactured to be visible on MR with a negative contrast. Fe3O4 meshes could simplify follow-up examinations and help diagnose origins of postsurgical lesions after urogynecological procedures with mesh material. © 2014 S. Karger AG, Basel
Introduction
The number of elderly people is steadily increasing, and there has been a corresponding growth in the number of common diseases such as female pelvic organ prolapse or stress urinary incontinence. Pelvic organ prolapse is a multifactorial disorder and the vaginal wall of concerned patients is often described as weakened or altered [1, 2]. These pelvic floor disorders are treated either with noninvasive Kerstin Aneta Brocker, MD, MSc Department of Obstetrics and Gynecology University of Heidelberg Medical School INF 440, DE–69120 Heidelberg (Germany) E-Mail kerstin.brocker @ med.uni-heidelberg.de
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
Key Words Female cadaver · Ferric compounds · Magnetic resonance imaging · Pelvic organ prolapse · Surgical mesh
Color version available online
a
b
Fig. 1. a Photograph of the mesh materials.
From bottom left to top right: samples 1–4. b Photograph of the prototype mesh with
six fixation points. Both the dotted lines and the zigzag lines schematically illustrate the position of the MR-visible threads in the mesh and are not absorbable.
Materials and Methods This experimental study was designed to develop MR-visible mesh material and to assess its visibility on MRI. We therefore determined suitable concentrations of iron oxide labeling in a phantom study. Based on these results, a labeled, MR-visible thread was woven into a mesh. To assess MR visibility in a realistic setting, the mesh was then implanted in a fresh female cadaver and MR images were acquired using optimized MRI protocols. The institutional ethical board approved the cadaver study (IRB No. 53/2010).
102
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
Substance Development and Magnetic Particle Concentration Two iron oxide substances were tested to make the meshes MR visible: iron(III) oxide-hydroxide (FeOOH) and iron(II,III) oxide (Fe3O4). These substances were chosen for their biocompatibility, as both iron oxides are used as pigments (FeOOH as C.I. pigment yellow 42, and Fe3O4 as C.I. pigment black 11) and are internationally approved [13]. Initially, measurements of the transverse relaxation rates R2 and R2* were performed to determine the optimal concentration of the iron oxide to be used in the fabrication of the mesh material. For this purpose, different concentrations of FeOOH and Fe3O4 in both gel (gelatin) and palm oil were used, ranging between 0 and 10% for FeOOH and between 0 and 0.2% for Fe3O4. R2 was measured with a 32-echo spin-echo sequence, and R2* was determined from a 8-echo gradient-echo data set. Relaxation rates were calculated from a signal fit to the exponential signal decay as a function of echo time (TE). From the relaxation data, the minimum concentration was determined that provides a R2 of 100 s–1 or more, i.e. which shortens the transverse relation time T2 = 1/R2 to less than 10 ms. Since relaxation data were only accessible in gel and in palm oil, a final test was performed with the iron oxides embedded in the thread carrier material. Therefore, four threads were manufactured from polypropylene granulate: two with FeOOH at concentrations of 2.5% (sample 2) and 5% (sample 1), and two with Fe3O4 at 0.02% (sample 4) and 0.2% (sample 3). Four bands were woven from the polypropylene threads as they are used in Serasis PP® (Serag Wiessner, Naila, Germany) mesh implants (samples 1–4; fig. 1a). For MRI, the bands were placed in a water bath, and images were acquired in a 1.5-tesla clinical MR system (Tim Symphony; Siemens Healthcare, Erlangen, Germany) using three standard imaging sequences: a 3D spoiled gradient-echo sequence [fast low-angle shot (FLASH)], a fast spin-echo sequence, and a balanced steady state free precession sequence or true fast imaging with steady precession (trueFISP). Imaging parameters of these three pulse sequences used to test the MR visibility of the mesh implants were repetition time (TR) = 15 ms, TE = 6 ms, BW = 179 Hz/px for FLASH 3D; TR = 4,000 ms, TE = 110 ms, BW = 260 Hz/ px for turbo spin-echo, and TR = 3.8 ms, TE = 1.79 ms, BW = 751 Hz/px for trueFISP. A field of view of 400 mm and a matrix of 512 × 512 was applied in all sequences. To optimally visualize and compare the signal void produced by the iron oxide in the bands, minimum intensity projections were calculated from all three 3D data sets.
Brocker/Lippus/Alt/Hallscheidt/Zsolt/ Soljanik/Lenz/Bock/Sohn
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
methods such as pelvic floor exercises or pessary therapy, or with classical surgical methods such as vaginal sacrospinous fixation, abdominal sacrocolpopexy, or Burch colposuspension [3, 4]. Recently, partially absorbable vaginal meshes and tension-free vaginal slings have been introduced into the portfolio of urogynecological surgery [3, 5]. The current literature on complications after mesh surgery includes reports on postoperative pain, hematoma, nerve distractions, mesh shrinking, de novo dyspareunia, and infections [6, 7]. To assess surgical outcome and postoperative complications, the patients’ subjective rating is combined with clinical examination and ultrasound imaging. Current imaging modalities reach their limits, however, when they are used to evaluate the mesh arms and their fixation point, especially when trying to verify a correct location or diagnosing misplacement [8]. Magnetic resonance imaging (MRI) could be a promising imaging alternative to further investigate the implanted mesh [9]. Nevertheless, current meshes are hardly visible in magnetic resonance (MR) images besides a few exceptions such as the polyvinylidene fluoride meshes recently described [10–12]. The aim of this study is therefore to develop a mesh material for reconstructive surgery of female pelvic organ disorders that uses iron oxides to make it visible on MR.
Results
Substance Development and Magnetic Particle Concentration Figure 2 shows R2 as a function of iron oxide concentration in gelatin. As expected, in this concentration range R2 increases linearly with concentration, so that an increase in iron oxide concentration will improve mesh visibility. However, as higher iron oxide concentrations might have a negative effect on thread stability and could also create excessively large artifacts, two different concentrations were chosen for thread production. These concentrations were 2.5 and 5% for FeOOH and 0.02 and 0.2% for Fe3O4. For both FeOOH concentrations and the 0.2% Fe3O4 concentration, an R2 of MR-Visible Polypropylene Mesh
Color version available online
100 80 60 40 20 0
FeOOH Fe3O4 0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Concentration (%)
Fig. 2. Transverse relaxation rate R2 as a function of the concentra-
tion of Fe3O4 and FeOOH in gelatin.
Table 1. Imaging parameters of the sequences performed to test
the MR visibility of the mesh implanted in the female cadaver
T1 3D WAVE transversal T2 3D VISTA transversal T1 tse transversal T2 tse sagittal T2 tse transversal SSH coronal T1 tse fs transversal
TR, ms
TE, ms
ST, mm
4.8 1,979.1 493 1,865 2,117.2 1,918.5 590
2.4 200 10 80 70 120 10
1.9 4 4 4 4 4 4
WAVE = Spoiled gradient-echo sequence; VISTA = volumetric isotropic turbo spin-echo acquisition; SSH = ultrafast spin-echo sequence; tse = turbo spin-echo; fs = fat saturated; ST = slice thickness.
100 s–1 or more was expected (i.e. a T2 of 10 ms or less; fig. 2). The mentioned concentrations of FeOOH and Fe3O4 are weight percent. While figure 3a–c shows that all the bands woven from these threads are visible in the MR images, sample 3 (Fe3O4 at 0.2%) achieves optimal contrast in all the standard imaging sequences. In this concentration range the apparent transverse relaxation rate T2* (which determines the contrast in gradient-echo images) is well below 10 ms, so that the labeled meshes are always delineated with a negative contrast in gradient-echo image data. Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
103
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
Cadaver Experiment and Proof of MR Visibility In order to visualize the newly constructed MR-visible mesh in the pelvic floor, it was implanted into a postmenopausal female cadaver using the transobturator approach and securing the proximal mesh arms through the sacrospinous ligament on both sides. For secure and standardized implantation, the mesh was placed according to the manufacturer’s recommendation and employing reusable introducer needles (Serag Wiessner®). The surgeon (F.L.) implanting the mesh was experienced in the field of urogynecological surgery (>1,000 vaginal mesh implantations). To fill the bladder, a transurethral catheter was inserted and blocked after filling the bladder with 200 ml water. The rectum was cleaned of stool and ultrasound gel was inserted for better demarcation of the rectal walls. Afterwards the anus was closed with several stiches to keep the ultrasound gel from being expelled. After full implantation of the vaginal mesh, ultrasound gel was inserted into the vagina for better demarcation of the vaginal walls. The cadaver was then imaged in a 3-tesla MR system (Achieva, Philips) to assess the MR visibility of the mesh. Sequences were performed that are used in clinical routine: a T1-weighted 3D spoiled gradient-echo sequence, a T2-weighted sequence with volumetric isotropic turbo spin-echo acquisition, T1-weighted turbo spin-echosequences with and without fat saturation, T2-weighted turbo spin-echo sequences, and a T2-weighted ultrafast spin-echo sequence. The parameters are listed in table 1.
120
R2 (1/s)
A full mesh was produced from the band with the best image contrast. It consisted of conventional partially absorbable threads and MR-visible threads (fig. 1b). The mesh was designed based on a Seratom E PA® (Serag Wiessner) mesh, such as what is used for combined reconstruction of a descended uterus (level I defect according to DeLancey et al. [14]) with the help of an additional sacrospinous fixation and an anterior vaginal wall defect (level II defect according to DeLancey et al. [14]) [15]. The MR-visible threads were woven into all mesh arms and partially into the mesh body using a combination of straight threads and threads woven in a zigzag pattern (fig. 1b). The non-MR-visible threads are absorbable.
a
c
b
Fig. 3. Minimum intensity projection images from 3D FLASH (a), turbo spin-echo (b) and TrueFISP (c) data. In
all pulse sequences sample 3 has the highest contrast.
a
b
c
d
echo) sequence, the arrows mark the four transobturator mesh arms. c T1-weighted 3D WAVE sequence, the arrow shows the
Cadaver Experiment and Proof of MR Visibility Surgical implantation of the MR-visible mesh into the female cadaver was performed in 45 min. The woven iron oxide filaments in the mesh were best visible in the T1weighted 3D spoiled gradient-echo sequence and the T1weighted fast spin-echo sequence. Figure 4a–d presents images of the mesh in the cadaver showing the anterior and medial arms of the mesh implanted via the transobturator route or displaying the zigzag pattern of the woven threads near the cervix of the uterus. 104
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
zigzag pattern of the mesh. d T1-weighted 3D WAVE, the arrow shows the mesh visible as a line (negative echo signal) between the cervix and rectum. AH = Artificial hip; B = bladder; DC = dilated catheter in bladder.
Discussion
In this experimental study, we made MR-visible mesh material for implantation into the pelvic floor. Its purpose is to overcome the current limitations in visibility on MR or ultrasound of meshes used in pelvic reconstructive surgery. The mesh material was optimized for rapid MRI to facilitate better detection of mesh placement in the pelvic floor and potentially find the origin of complications related to mesh repair. In the current literature, meshes in Brocker/Lippus/Alt/Hallscheidt/Zsolt/ Soljanik/Lenz/Bock/Sohn
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
Fig. 4. a T2-weighted, the arrows depict both medial mesh arms. b T1-weighted 3D WAVE (T1-weighted 3D spoiled gradient-
general and misplaced ones in particular might be the cause of nerve irritations, uncertain postsurgical pelvic floor pain, or dyspareunia [6, 16, 17]. The published rates of complications such as mesh erosion, de novo dyspareunia, and pelvic organ prolapse recurrence are heterogeneous [6, 7]. Until now it has often not been possible to determine a definite reason for their occurrence because the complete location of the mesh and its anchoring mesh arms were not visible [6, 7, 17]. Iron oxides were chosen as labeling compounds due to their known biocompatibility [18]. Iron is a natural element found in the human body, and iron oxides are widely approved for human use. For example, iron oxides are licensed as food coloring agents under the number E172, have been approved by the FDA for use in contact lenses (§ 73.3125), and they form the contrast-generating compound in the MRI contrast agent Lumirem® (ferumoxsil; Guerbet Groupe, France), which is approved for oral and rectal application at concentrations up to 175 mg Fe/l [9, 19, 20]. Thus, incorporation of iron oxides into the mesh material does not seem to pose any risks of side effects, such as might be present with other MR contrast agents, such as Gd-DTPA from which the toxic element gadolinium could potentially be washed out of the mesh during the long exposure times [21–23]. Relaxometric studies show that a concentration of 0.2 weight-% Fe3O4 is a good compromise between signal decay in the proximity of the threads and clear visibility of the mesh. Therefore Fe3O4 was selected over FeOOH as the labeling compound since much higher concentrations would have been needed with FeOOH to create the negative image contrast. At all the MR settings subsequently employed in our study, the MR visibility of the labeled mesh was demonstrated, and spoiled gradientecho sequences showed the highest contrast between the mesh and surrounding tissue, which was also confirmed in the cadaver experiment. In theory, more mesh material with other iron oxide concentrations could have been manufactured to evaluate the contrast mechanism in more detail. In addition, more experimental images are needed to determine the effect on visibility at other field strengths. However, the current results already indicate that the concentration of 0.2 weight-% is sufficient to reliably create the desired contrast, and a higher concentration might only compromise visibility and mesh stability. Preliminary tests at higher field strengths indicate that the contrast behavior is very similar and that the meshes will also be suitable for higher field strengths such as those we performed in the cadaver experiment using a 3-tesla machine. To as-
sess the reproducibility of our results, more than one cadaver study should be performed. While cadaver studies can be generally applied, they often have limitations due to the presence of entrapped air, which can create artifacts similar to those of meshes [24]. Additionally, in order to learn more about the biological stability of this mesh, future experimental studies on live animal models seem useful and are initiated as described by other authors [11, 12]. The present study demonstrates a new iron oxide-labeled mesh with proven MR visibility in an experimental 3-tesla MR scenario with a fresh human female cadaver for use in the reconstructive surgery of pelvic floor disorders. Although more follow-up experiments are needed, the results of this study might be helpful to facilitate the introduction of the mesh in human studies. In addition, the introduction of an MR-visible mesh in pelvic floor reconstructive surgery may help to precisely assess the anatomical and functional relations of implanted meshes and to correlate it with patients’ symptoms. The cost-effectiveness of visible meshes should also be evaluated with regard to the economic benefit of successfully treated complications after pelvic floor reconstructive surgery. In conclusion, the MR visibility of the mesh was achieved in a 3-tesla MR setting with a fresh human female cadaver when an adequate concentration of iron oxide particles was added to the mesh material. MR-visible meshes might help to analyze and quantify objectively the surgical outcome and complications after mesh repair.
MR-Visible Polypropylene Mesh
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
Acknowledgements Special thanks go to Dr. Anna Németh (Institute of Anatomy and Cell Biology, Semmelweis University of Budapest, Budapest, Hungary) and Prof. Gabor Rudas (MR Research Center, Szentágothai Knowledge Center, Semmelweis University of Budapest, Hungary).
Disclosure Statement
105
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
K.A. Brocker received a research fellowship from the University of Heidelberg and a research scholarship from the Forum Urodynamicum e.V. to perform this work. K.A. Brocker received research funding from Serag Wiessner to perform this work. F. Lenz received speaking honoraria for lectures and speeches on international meetings from Serag Wiessner. M. Bock has received grants from Siemens Healthcare outside of the research outlined in this study. No money was used for the research presented in this manuscript. For all the remaining authors, no conflict of interest was declared.
References
106
9 Arsalani N, Fattahi H, Nazarpoor M: Synthesis and characterization of PVP-functionalized superparamagnetic Fe3O4 nanoparticles as an MRI contrast agent. Express Polym Lett 2010;4:329–338. 10 Ciritsis A, Hansen NL, Barabasch A, Kuehnert N, Otto J, Conze J, Klinge U, Kuhl CK, Kraemer NA: Time-dependent changes of magnetic resonance imaging-visible mesh implants in patients. Invest Radiol 2014; 49: 439–444. 11 Kuehnert N, Kraemer NA, Otto J, Donker HC, Slabu I, Baumann M, Kuhl CK, Klinge U: In vivo MRI visualization of mesh shrinkage using surgical implants loaded with superparamagnetic iron oxides. Surg Endosc 2011; 26:1468–1475. 12 Endo M, Feola A, Sindhwani N, Manodoro S, Vlacil J, Engels AC, Claus F, Deprest JA: Mesh contraction: in vivo documentation of changes in apparent surface area utilizing meshes visible on magnetic resonance imaging in the rabbit abdominal wall model. Int Urogynecol J 2013;25:737–743. 13 Title 21, Food and Drugs Chapter I, Food and Drug Administration, Department of Health and Human Services, Subchapter A, Part 73, Listing of Color Additives Exempt from Certification, Subpart D, Medical Devices. 73.3125 Iron Oxides. 2012 (revised April 1, 2013). http://www.accessdata.fda.gov/scripts/cdrh/ cfdocs/cfCFR/CFRSearch.cfm?fr=73.3125 (accessed January 11, 2014). 14 DeLancey JO: Anatomic aspects of vaginal eversion after hysterectomy. Am J Obstet Gynecol 1992; 166: 1717–1724; discussion 1724– 1728. 15 Huddleston HT, Dunnihoo DR, Huddleston PM 3rd, Meyers PC Sr: Magnetic resonance imaging of defects in DeLancey’s vaginal support levels I, II, and III. Am J Obstet Gynecol 1995;172:1778–1782; discussion 1782–1784.
Gynecol Obstet Invest 2015;79:101–106 DOI: 10.1159/000366442
16 Jacquetin B, Cosson M: Complications of vaginal mesh: our experience. Int Urogynecol J Pelvic Floor Dysfunct 2009;20:893–896. 17 Margulies RU, Lewicky-Gaupp C, Fenner DE, McGuire EJ, Clemens JQ, Delancey JO: Complications requiring reoperation following vaginal mesh kit procedures for prolapse. Am J Obstet Gynecol 2008;199:678.e1–e4. 18 Gould P: Nanomagnetism shows in vivo potential. Nanotoday 2006;1:34–39. 19 Hallgren B, Sourander P: The effect of age on the non-haemin iron in the human brain. J Neurochem 1958;3:41–51. 20 Hentze MW, Caughman SW, Rouault TA, Barriocanal JG, Dancis A, Harford JB, Klausner RD: Identification of the iron-responsive element for the translational regulation of human ferritin mRNA. Science 1987;238:1570– 1573. 21 Weinmann HJ, Brasch RC, Press WR, Wesbey GE: Characteristics of gadolinium-DTPA complex: a potential NMR contrast agent. AJR Am J Roentgenol 1984;142:619–624. 22 Kornmesser W, Laniado M, Hamm B, Clauss W, Weinmann HJ, Schulz E, Wolf KJ, Felix R: Oral contrast medium for magnetic resonance tomography of the abdomen. II. Phase I clinical testing of gadolinium-DTPA (in German). Rofo 1987;147:550–556. 23 Okada S, Katagiri K, Kumazaki T, Yokoyama H: Safety of gadolinium contrast agent in hemodialysis patients. Acta Radiol 2001; 42: 339–341. 24 Dietrich O, Reiser MF, Schoenberg SO: Artifacts in 3-T MRI: physical background and reduction strategies. Eur J Radiol 2008;65:29– 35.
Brocker/Lippus/Alt/Hallscheidt/Zsolt/ Soljanik/Lenz/Bock/Sohn
Downloaded by: University of Exeter 149.126.76.97 - 8/15/2015 8:38:30 AM
1 Meijerink AM, van Rijssel RH, van der Linden PJ: Tissue composition of the vaginal wall in women with pelvic organ prolapse. Gynecol Obstet Invest 2013;75:21–27. 2 Martins P, Lopes Silva-Filho A, Rodrigues Maciel da Fonseca AM, Santos A, Santos L, Mascarenhas T, Natal Jorge RM, Ferreira AJ: Biomechanical properties of vaginal tissue in women with pelvic organ prolapse. Gynecol Obstet Invest 2013;75:85–92. 3 Maher C, Baessler K: Surgical management of anterior vaginal wall prolapse: an evidencebased literature review. Int Urogynecol J Pelvic Floor Dysfunct 2006;17:195–201. 4 Jelovsek JE, Maher C, Barber MD: Pelvic organ prolapse. Lancet 2007;369:1027–1038. 5 Ulmsten U: An introduction to tension-free vaginal tape (TVT) – a new surgical procedure for treatment of female urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct 2001;12:S3–S4. 6 Baessler K, Hewson AD, Tunn R, Schuessler B, Maher CF: Severe mesh complications following intravaginal slingplasty. Obstet Gynecol 2005;106:713–716. 7 Farthmann J, Watermann D, Niesel A, Funfgeld C, Kraus A, Lenz F, Augenstein HJ, Graf E, Gabriel B: Lower exposure rates of partially absorbable mesh compared to nonabsorbable mesh for cystocele treatment: 3-year followup of a prospective randomized trial. Int Urogynecol J 2013;24:749–758. 8 Shek KL, Dietz HP, Rane A, Balakrishnan S: Transobturator mesh for cystocele repair: a short- to medium-term follow-up using 3D/4D ultrasound. Ultrasound Obstet Gynecol 2008;32:82–86.
