Filippi, Ronald M.D.; Derdilopoulos, Athanassios; Heimann, Axel D.V.M.; Krummenauer, Frank Ph.D.; Perneczky, Axel M.D.; Kempski, Oliver M.D.
OBJECTIVE: The purpose of this study was to test, in rabbits, the tightness of seven dural substitution materials commonly used in neurosurgery, i.e., Lyodura (B. Braun Melsungen AG, Melsungen, Germany), Tutoplast dura (Tutogen Medical, Inc., Parsippany, NJ), Tutoplast fascia lata (Tutogen Medical, Inc.), autologous periosteum, Neuropatch (B. Braun Melsungen AG), Dacron (E.I. du Pont de Nemours and Co., Wilmington, DE), and Ethisorb (Ethicon, Inc., Somerville, NJ).
METHODS: Duraplasties were performed with sutures alone or were additionally fixed with fibrin glue. Leakage pressures were assessed by infusion of artificial cerebrospinal fluid, containing sodium fluorescein, into the cisterna magna and detection of fluorescence using a charge-coupled display camera with background substraction, 3 days, 3 weeks, or 3 months after surgery.
RESULTS: Three days after implantation, the mean tightness values of duraplasties with Lyodura or Neuropatch were significantly higher (P = 0.007) than the values for the other substitutes. A significant improvement of tightness with increasing implantation time could be demonstrated for autologous periosteum (P = 0.0063). Improvement of tightness with the use of fibrin glue could be proven only for the heterologous grafts (P = 0.0071). The tightness values for Neuropatch fixed only with sutures were similar to those for the best heterologous substitutes implanted with additional fibrin glue. Lyodura, Tutoplast dura, and Neuropatch demonstrated favorable implantation characteristics; they were thin, flexible, and easily suturable. Neither adhesions to the brain nor space-occupying scars were noted.
CONCLUSION: These results confirm the excellent suitability of Lyodura and Neuropatch for dural substitution.
The search for appropriate materials that can replace the dura mater has been a major focus in neurosurgery for many decades. Different substances have been used, but most of them caused serious side effects. Many allografts, such as metals (1, 4, 12, 20) or synthetic materials (22, 23, 49), were implanted, but they led to the formation of scars and adhesions to the brain surface. Heterografts (8, 16, 19, 25, 28) often elicited rejection responses. Since 1893 (41), different autologous grafts (3, 38, 51, 53) have been tested, but extensive adhesions to the brain surface resulted from the use of these autotransplants (44).
In 1958, Sharkey et al. (45) reported their experiences using cadaverous freeze-dried (lyophilized) dura mater as a dural substitute. They observed no adhesions or disturbances of patch integration. In the following decades, homologous dura mater became the material of choice for dural substitution. Solvent-drying was subsequently developed as a new technique for the preservation of cadaverous dura mater (36, 48). This so-called “Tutoplast dura” has been used frequently. Recently, slow virus infections caused by transplantation of cadaverous human dura mater (31, 50) led to the development of new dural substitution materials (5, 35, 52).
The aim of this study was to compare the tightness of seven different dural substitution materials commonly used in neurosurgery if primary closure of the dura is impossible. The materials were implanted into rabbits with the same technique as used for human subjects, to allow clinical conclusions to be drawn. Three days, 3 weeks, and 3 months after implantation, the leakage pressures of the dural patches were measured and compared. In addition, the mechanical characteristics of the materials and the histological changes in the substitutes and the surrounding tissues were documented.
Lyodura (B. Braun Melsungen AG, Melsungen, Germany) consists of cadaverous human dura mater preserved by freeze-drying (lyophilization). Tutoplast dura (Tutogen Medical, Inc., Parsippany, NJ) also consists of cadaverous human dura mater but is preserved by solvent-drying. Tutoplast fascia lata (Tutogen Medical, Inc.) consists of solvent-dried human fascia lata. Autologous periosteum is acquired from the contralateral side of the cranium during implantation surgery. Neuropatch (B. Braun Melsungen AG) consists of a microporous fleece composed of cleaned aliphatic polyesterurethane. Dacron (E.I. du Pont de Nemours and Co., Wilmington, DE) is a woven polyester patch with low porosity (50 cc/min/cm2). Ethisorb (Ethicon, Inc., Somerville, NJ) is composed of Vicryl filaments and polydioxanone meltages.
Duraplasty was performed in rabbits weighing between 3.0 and 5.0 kg. The different dural substitutes were investigated in groups of six animals. The tightness of duraplasties without fibrin glue was measured 3 days, 3 weeks, and 3 months after implantation (7 × 6 × 3 = 126 animals). The effects of supplemental fibrin glue were tested in groups of six animals for every material but only with a survival time of 3 days. Therefore, a total of 168 animals (126 + 7 × 6 = 168) were used in the study.
The different dural substitution materials and the survival times of the animals were randomly assigned. Therefore, the same implantation quality for all of the substitutes and for all survival times could be guaranteed, despite increasing experience with the surgical procedures.
The rabbits were anesthetized with ketamine and xylazine. Initially, a mixture of 4.5 ml of ketamine and 2.5 ml of 2% xylazine (Rompun; Bayer Vital, Leverkusen, Germany) was administered intravenously at 1.5 ml/kg. These anesthetics were then administered continuously at an infusion rate of 20 ml/h. Atropine (1 mg) was injected intramuscularly. A single injection of 40 mg of gentamycin was used as antibiotic therapy.
After partial shaving, a scalp incision was made in the midline of the cranium. In the left parietal region, a 1.5- × 1-cm craniectomy was performed with an electric drill and an upcutting rongeur. The dura was opened with a dural hook and a small knife, under microscopic observation, and a 4- × 4-mm section of the dura was removed (Fig. 1 A). A 5- × 5-mm patch of dural substitute was cut, positioned, and fixed with four 10-0 polyamide sutures (Ethilon; Ethicon, Inc.) (Fig. 1B). Round needles with noncutting microtips were used, to prevent a lack of tightness of the duraplasty caused by the suture holes. The dural patch was fixed by single sutures, to yield a firm connection between the substitute and the surrounding dura at all fixation points. In addition, fibrin glue was applied for designated groups. The cranial defect was closed with polymethylacrylate (Palacos; Merck, Darmstadt, Germany) and fixed with two 4-0 polyester sutures (Ethibond; Ethicon, Inc.). Finally, the scalp was closed with 3-0 polyester sutures (Ethibond).
Three days, 3 weeks, or 3 months after surgery, the animals were again anesthetized, to test the leakage of the dural patches. The Palacos flap was removed and a solution of artifical cerebrospinal fluid (CSF) containing 0.05% sodium fluorescein was infused, under intracranial pressure control, into the cisterna magna, through a canula placed suboccipitally. Simultaneously, the dural patch was observed with a fluorescence macroscope equipped with a charge-coupled device camera (US 450; Stemmer Imaging, Puchheim, Germany), which was attached to an image processor (Argus 10; Hamamatsu Photonics KK, Hamamatsu City, Japan) with background subtraction. The leakage pressure was defined as the intracranial pressure at which fluorescence was first detected on the surface of the substitutes or the surrounding dura (Fig. 2).
During surgery, the mechanical characteristics of the implants were noted, to provide additional information regarding the suitability of the substitution materials.
After the tightness investigation, the animals were killed. The rabbits were perfused through the left ventricle with 500 ml of physiological saline, followed by 1000 ml of paraformaldehyde (4%). The dura (including the substitute) and the brain were removed, immersed in formaldehyde, embedded in paraffin, and later cut into coronal sections. The tissue sections were stained with hematoxylin/eosin and van Gieson stain and then examined for changes in the substitute, the surrounding dura, and the underlying brain tissue.
The overall hypotheses of differences among materials, with time, and with the use of fibrin glue were investigated using the design of a one-way classification, in which 42 animals were randomized to each of the seven dural substitutes. The main hypothesis of an overall treatment effect was tested using analysis of variance (14, 24). Results of the analysis of variance calculations are presented as P values; values of less than 0.05 were considered locally significant. Finally, we performed overall Kruskal-Wallis and two-sample Wilcoxon tests (treatment effects), as well as overall Friedman and signed-rank tests (time or fibrinogen effects), to provide nonparametric alternatives to the analysis of variance approach. All calculations were performed and graphical presentations were prepared by using the software packages SAS (42) and SPSS (47) (both Windows releases).
Eight animals died suddenly during anesthesia, three rabbits experienced acute cardiac arrest in their cages (without signs of infection), and one rabbit needed to be killed because of pyometritis (with no relation to the implantation surgery). The lost rabbits were replaced, to replenish the study. All other rabbits recovered from surgery without complications.
The tightness of the duraplasties was tested 3 days, 3 weeks, or 3 months after surgery. During volume loading of the subarachnoid space, the leakage pressure was recorded. Leakage pressures were observed to vary between 10 and 88 mm Hg (Table 1).
Three days after implantation, the leakage pressures for implanted Lyodura (41.2 ± 4.71 mm Hg) and Neuropatch (48.3 ± 11.99 mm Hg) were significantly higher (P = 0.007) than those for the other substitutes (Tutoplast dura, 32 ± 5.66 mm Hg; Tutoplast fascia lata, 26.8 ± 3.76 mm Hg; autologous periosteum, 27.7 ± 9.81 mm Hg; Dacron, 31 ± 15.41 mm Hg; Ethisorb, 28.8 ± 4.62 mm Hg). The analysis of all three subgroups (nonsynthetic heterologous substitutes, autologous periosteum, and synthetic materials) confirmed the quality of these materials after this implantation period; Lyodura demonstrated better tightness (P = 0.03) than the other heterologous substitutes, and Neuropatch exhibited significantly better results (P = 0.0094) than the other synthetic materials.
Three weeks after implantation, Lyodura (50.2 ± 21.16 mm Hg) and Neuropatch (41.2 ± 9.79 mm Hg) demonstrated distinctly better mean tightness values, compared with the other substitutes (Tutoplast dura, 36.3 ± 9.68 mm Hg; Tutoplast fascia lata, 36.3 ± 9.55 mm Hg; autologous periosteum, 36.3 ± 5.54 mm Hg; Dacron, 37.3 ± 7.09 mm Hg; Ethisorb, 36.0 ± 11.62 mm Hg). Three months after implantation, Lyodura and Tutoplast dura exhibited the highest leakage pressures (50.3 ± 12.53 and 42.8 ± 6.86 mm Hg, respectively), compared with Tutoplast fascia lata (37.8 ± 8.59 mm Hg), autologous periosteum (36.3 ± 8.21 mm Hg), Neuropatch (35.8 ± 12.11 mm Hg), Dacron (37.7 ± 13.68 mm Hg), and Ethisorb (39.8 ± 14.75 mm Hg). A significant improvement of tightness (P = 0.0063) with longer implantation times could be noted only for autologous periosteum.
Three days after implantation, a significant improvement in tightness (P = 0.0071) as a result of the additional application of fibrin glue was noted only for nonsynthetic heterologous grafts (Lyodura, 42.8 ± 8.33 mm Hg; Tutoplast dura, 41.5 ± 12.01 mm Hg; Tutoplast fascia lata, 38.8 ± 9.91 mm Hg; autologous periosteum, 38 ± 6.06 mm Hg; Neuropatch, 39.3 ± 5.46 mm Hg; Ethisorb, 39.8 ± 8.84 mm Hg; Dacron, 28.7 ± 9.39 mm Hg). In conclusion, very good dural closure was possible with the use of preserved cadaverous human dura mater (especially Lyodura) and synthetic Neuropatch, without the need to apply fibrin glue (Fig. 3).
The mechanical characteristics of the substitutes were rather different. Lyodura, Tutoplast dura, Tutoplast fascia lata, Neuropatch, and autologous periosteum were flexible and easily suturable. Dacron was also flexible but was very tearable. Ethisorb was very thick, rigid, and also tearable.
Rather good acceptance of the implants was observed with the biological substitutes and Neuropatch. Three months after surgery, macroscopic localization of the implant was often possible only with the use of persistent sutures. Dacron and Ethisorb produced excessive adhesions between the substitute and the underlying brain tissue.
Three days after implantation, examinations of Lyodura, Tutoplast dura, Tutoplast fascia lata, Neuropatch, and autologous periosteum revealed early fibroblast accumulation in the dura surrounding the patches. Red blood cells from the recent surgery were still scattered over the substitutes.
After 3 weeks, the bleeding caused by surgery had resolved. Early fibroblast proliferation from the edges of the dura had penetrated the collagen sponge of Lyodura, Tutoplast dura, Tutoplast fascia lata, and autologous periosteum, using its fibers as a scaffold for the deposition of new collagen. Within the group of synthetic substitutes, sufficient fibroblast infiltration occurred only with Neuropatch. This was facilitated by the very porous structure of that material. Formation of new capillaries within the microporous fleece was noted. This fibroblast proliferation into the patches observed 3 weeks after surgery possibly indicates the end of the critical postimplantation period, with risks of CSF leaks and subsequent infections.
Three months after surgery, fibroblast activity and neovascularization were well established with Lyodura, Tutoplast dura, Tutoplast fascia lata, Neuropatch, and autologous periosteum. Incorporation of the collagen sponge into the dura had occurred, and the sponge was partially collagenized. All of these materials were covered by thin neomembranes on both surfaces. Dacron was encapsulated from both sides by connective tissue, but there was neither real incorporation by ingrowing tissue nor sufficient collagen formation in the patch. Ethisorb was fragmented and had mostly dissolved. Only small amounts of the Vicryl filaments were observed; they were embedded in firm connective tissue, which had replaced the substitute. Severe adhesions to the brain surface were noted with Dacron and Ethisorb.
The ideal dural substitute is made of a material that effectively protects the brain from infections and is transformed to normal dura without any disturbances or foreign-body reactions from the meninges and the brain, so that no epileptic foci result.
Watertight closure of the dura mater is the basic requirement for prevention of CSF fistulae and central nervous system infections, with subsequent adhesions between the dura mater and the underlying brain tissue. For in vivo assessment of the tightness of duraplasties, Cargill et al. (9) reported a test for dural leaks in dogs during orotracheal intubation anesthesia. Those authors increased the intracranial pressure by laterally compressing the thorax, maintaining pulmonary inflation at 20 cm H2O. We did not perform intubation but induced anesthesia by intravenous administration of anesthetics only. Therefore, in our study, the leakage pressure of the substitute was ascertained by volume loading of the subarachnoid space with artificial CSF containing sodium fluorescein, with continuous intracranial pressure control.
The search for a suitable dural substitution material began with allografts such as rubber (1) and metals such as gold (4), silver (20), and more recently tantalum (12). These materials had to be abandoned because of the extensive formation of granulation tissue.
Early heterologous grafts (8, 16, 19, 25) also failed, because of rejection responses. The use of preserved bovine pericardium has recently been tested. Analogous to findings for other collagen sponges (33, 34), the first results after implantation indicate good ingrowth of pericardium (27, 35), without foreign-body reactions.
Several studies on autografts were conducted. Walton and Krizek (53) recommended scalp flaps only as a method for treating large dural defects. Barrow et al. (3) successfully reconstructed large dural defects with greater omentum flaps. To date, autologous periosteum and autologous fascia lata (13, 37, 51) have been successfully used for dural substitution. We can confirm that autologous periosteum is a very flexible material for the replacement of dura mater, yielding sufficient ingrowth of surrounding dura into the periosteum patches without adhesions and granulations. Only the resulting tightness of the duraplasty during the first weeks after implantation was unsatisfactory.
After the introduction of preserved cadaverous human dura mater, this material seemed to be the final solution for duraplasty, almost without complications (2, 7, 29, 40). In accordance with those reports, we obtained very good results with freeze-dried and solution-dried dura mater in our study. Implantation of the thin and flexible patches with sutures was very easy. High tightness values could be demonstrated. Fibroblastic infiltration and formation of new collagen, as well as neovascularization inside the patches of Lyodura and Tutoplast dura, were observed. No adhesions to the brain or disturbances of wound healing were noted.
Similar results have frequently been observed with the use of collagen sponges obtained from other tissues, such as the placenta or flexor tendons (10, 26, 34). In analyzing our own results, we cannot confirm these findings for solvent-dried Tutoplast fascia lata, because of the unsatisfactory tightness of those patches. Fascia lata prepared by freeze-drying was not tested, because we wanted to focus our study on the materials we had implanted earlier as dural substitutes in human subjects.
Recently, commonly used cadaverous dura mater has been associated with the transmission of slow viral infections, such as Creutzfeldt-Jakob disease (31, 50). This has led to a renewed search for appropriate synthetic dural substitutes in the past decade. Polyethylene (23), Orlon (22), Vinyon N (49), and Silastic and silicone-coated Dacron (15, 18, 28, 39), however, produced severe side effects, such as inflammation, adhering neomembranes, and meningocerebral scars. Analogous to those findings, we observed hypertrophic encapsulation and severe adhesions of the tested Dacron patches to the underlying brain tissue. Despite the extensive growth of connective tissue, the duraplasty was not very tight. Synthetic bioabsorbable polymers, such as chitosan (32) and other materials (30, 54), often induce extensive formation of adhesive connective tissue, which replaces the substitute. We noted this phenomenon after implantation of Ethisorb. In addition, the leakage pressure of Ethisorb was low, in comparison with that of most of the other substitutes. Expanded polytetrafluoroethylene sheets (Gore-Tex surgical membranes) (17, 43) and hydroxyethylmethacrylate hydrogels (5) demonstrated sufficient integration of patches without adhesions. However, the clinical value of these materials is still uncertain because they have not yet been used long enough.
We tested a microporous fleece composed of polyesterurethane (Neuropatch), with very good results. The patch was thin and flexible. Three days after implantation, the watertightness of Neuropatch was distinctly superior to that of all other substitutes. This phenomenon is possibly produced by exceptionally firm adhesions between the polyesterurethane fleece and the surrounding dura, which loosen during the first 3 weeks. Further studies are necessary to investigate the nature of these firm connections and their weakening during the first weeks after surgery. Currently, we can only speculate on the possible causes of the superior tightness achieved with Neuropatch. Also after 3 weeks and 3 months, Neuropatch, as well as Lyodura, demonstrated significantly better tightness values than did the other substitutes. Similar to cadaverous human dura mater, the resulting strong dural closeness corresponded to the ingrowth of fibroblasts and the formation of collagen inside the fleece. Our results confirm previous successful experience with the use of polyesterurethane as a dural substitute (13, 21, 52). We have also gained experience with the clinical use of Neuropatch in the past several years. This polyesterurethane fleece could be implanted in human subjects very comfortably and with high probability of sufficient tightness. It is available in eight different sizes (from 1.5 × 3 cm to 12 × 14 cm), which effectively reduces the costs of duraplasty. Our results of Neuropatch implantation in human subjects will be reported in the near future. Further studies are necessary to compare the characteristics of Neuropatch and Lyodura with those of recently developed substitution materials, such as bovine pericardium, and to compare those results with findings for primarily closed dura without substitutes.
The use of fibrin glue for duraplasty has been controversial for several years. Cain et al. (6), Dehesdin et al. (11), and Wiegand et al. (54) recommended the use of fibrin glue in addition to dural closure with sutures. In contrast, Sierra et al. (46) reported that sufficient dural tightness can usually be achieved with muscle and fat plugs and homograft dura or fascia with sutures only.
In our hands, improvements in tightness with the use of fibrin glue could be noted only after implantation of heterologous grafts, such as cadaverous human dura mater. No significant changes in tightness were observed for autologous periosteum or synthetic substitutes. This phenomenon is possibly produced by better adhesion of fibrin to collagen sponges than to synthetic materials. Nevertheless, despite the additional application of fibrin glue, the tightness of human dura mater patches was not better than the tightness of Neuropatch implanted without fibrin glue.
Our study of the tightness of the different dural substitution materials supports the following conclusions: 1) 3 days after implantation, Lyodura and Neuropatch achieve significantly better tightness values than do the other substitutes; 2) only the rather leaky autologous periosteum demonstrates significant improvement of tightness with implantation time; 3) the additional use of fibrin glue improves the tightness only of preserved cadaverous human dura mater and fascia lata; and 4) the tightness of Neuropatch fixed with sutures alone is similar to that of the best heterologous substitutes, such as Lyodura, implanted with additional fibrin glue. Both Lyodura and Neuropatch also demonstrate excellent implantation characteristics and therefore can be recommended for dural substitution.
We express sincere gratitude to Andrea Schollmayer for histological specimen processing and to Michael Malzahn and Laszlo Kopacz for dedicated technical support during the animal experiments.
The reported results are part of the doctoral thesis of Athanassios Derdilopoulos.
We have no proprietary interest in the aforementioned dural substitutes or the companies that manufacture them.
1. Abbe R: Rubber tissue for meningeal adhesions. Trans Am Surg Assoc 13:490–491, 1895. [Context Link]
2. Abdank R: Covering of dura defects in skull-brain injuries with preserved dura [in German]. Zentralbl Chir 39:1367–1369, 1968. Library Holdings [Context Link]
3. Barrow DL, Nahai F, Tindall GT: The use of greater omentum vascularized free flaps for neurosurgical disorders requiring reconstruction. J Neurosurg 60:305–311, 1984. Library Holdings [Context Link]
4. Beach HHA: Gold foil in cerebral surgery. Boston Med Surg J 136:281–282, 1897. [Context Link]
5. Bhatia S, Bergethon PR, Blease S: A synthetic dural prosthesis constructed from hydroxyethylmethacrylate hydrogels. J Neurosurg 83:897–902, 1995. Library Holdings [Context Link]
6. Cain J Jr, Dryer RF, Barton BR: Evaluation of dural closure techniques: Suture methods, fibrin adhesive sealant, and cyanoacrylate polymer. Spine 13:720–725, 1988. [Context Link]
7. Cantore G, Guidetti B, Delfini R: Neurosurgical use of human dura mater sterilized by gamma rays and stored in alcohol: Long-term results. J Neurosurg 66:93–95, 1987. Library Holdings [Context Link]
8. Caporale L, de Bernardis M: Sur les héteroplasties durales avec le catgut laminé: Contribution experimentelle. Rev Chir (Paris) 74:10–27, 1936. [Context Link]
9. Cargill HA Jr, Cawley CM, Barrow DL, Poff BC, Powell MD, Sawhney AS, Dillehay DL: Efficacy and biocompatibility of a photopolymerized, synthetic, absorbable hydrogel as a dural sealant in a canine craniotomy model. J Neurosurg 88:308–313, 1998. Library Holdings [Context Link]
10. Collins RL, Christiansen D, Zazanis GA, Silver FH: Use of collagen film as a dural substitute: Preliminary animal studies. J Biomed Mater Res 25:267–276, 1991. Library Holdings [Context Link]
11. Dehesdin D, Tadie M, Hemet J, Yun J, Bagat-D’Arc M: Dura mater healing after repair with aponeurotic plasty: Comparison of results between classical sutures and fibrin adhesive glue—Fibrin adhesive in frontobasal injuries: An experimental study [in French]. Neurochirurgie 34:311–314, 1988. [Context Link]
12. Delarue NC, Linell EA, McKenzie KG: An experimental study on the use of tantalum in the sub-dural space. J Neurosurg 1:239–242, 1944. [Context Link]
13. Ernestus RI, Ketter G, Klug N: Dura-plasty in intracranial operations [in German]. Zentralbl Neurochir 56:106–110, 1995. Library Holdings [Context Link]
14. Fleiss JL: The Design and Analysis of Clinical Experiments. New York, Wiley, 1986. [Context Link]
15. Fontana R, Talamonti G, D’Angelo V, Arena O, Monte V, Collice M: Spontaneous haematoma as unusual complication of Silastic dural substitute: Report of 2 cases. Acta Neurochir (Wien) 115:64–66, 1992. Library Holdings [Context Link]
16. Freeman L: The use of egg membrane in trephining operations upon the skull. Ann Surg 28:455–457, 1898. [Context Link]
17. Görtler M, Braun M, Becker I, Roggendorf W, Heiss E, Grote E: Animal experiments with a new dura graft (polytetrafluoroethylene): Results. Neurochirurgia (Stuttg) 34:103–106, 1991. [Context Link]
18. Gudmundsson G, Sogaard I: Complications to the use of Vicryl-collagen dural substitute. Acta Neurochir (Wien) 132:145–147, 1995. Library Holdings [Context Link]
19. Hanel D: Über alloplastischen Duraersatz. Arch Klin Chir 90:823–828, 1909. [Context Link]
20. Harris ML: The use of silver foil to prevent adhesions in brain surgery. JAMA 19:763–765, 1904. [Context Link]
21. Heiss E, Sievert T: Clinical experiences with synthetic dura substitution. Presented at the Symposium “Modern Concepts of Dural Substitution” at the 10th European Congress of Neurosurgery, Berlin, Germany, 1995, Congress Volume. [Context Link]
22. Huertas J: The use of Orlon for dural replacement. J Neurosurg 12:550–554, 1955. [Context Link]
23. Ingraham FD, Alexander E, Matson D: Polyethylene, a new synthetic plastic for use in surgery: Experimental applications in neurosurgery. JAMA 135:62–87, 1947. [Context Link]
24. Khattree R, Naik DN: Applied Multivariate Statistics with SAS Software. Carey, SAS Institute, 1995. [Context Link]
25. Kostling W: Duraplastik durch Bruchsacküberpflanzung. Med Wochenschr 58:1042, 1911. [Context Link]
26. Laquerriere A, Yun J, Tiollier J, Hemet J, Tadie M: Experimental evaluation of bilayered human collagen as a dural substitute. J Neurosurg 78:487–491, 1993. Library Holdings [Context Link]
27. Laun A, Tonn JC, Jerusalem C: Comparative study of lyophilized human dura mater and lyophilized bovine pericardium as dural substitutes in neurosurgery. Acta Neurochir (Wien) 107:16–21, 1990. Library Holdings [Context Link]
28. Lee JF, Odom GL, Tindall GT: Experimental evaluation of silicone-coated Dacron and collagen fabric-film laminate as dural substitutes. J Neurosurg 27:558–564, 1967. Library Holdings [Context Link]
29. Macfarlane MR, Symon L: Lyophilised dura mater: Experimental implantation and extended clinical neurosurgical use. J Neurol Neurosurg Psychiatry 42:854–858, 1979. Library Holdings [Context Link]
30. Mello LR, Feltrin LT, Fontes-Neto PT, Ferraz FA: Duraplasty with biosynthetic cellulose: An experimental study. J Neurosurg 86:143–150, 1997. Library Holdings [Context Link]
31. Miyashita K, Inuzuka T, Kondo H: Creutzfeldt-Jakob disease in a patient with a cadaveric dural graft. Neurology 41:940–941, 1991. Library Holdings [Context Link]
32. Muzzarelli R, Baldassarre V, Conti F, Ferrara P, Biagini G, Gazzanelli G, Vasi V: Biological activity of chitosan: Ultrastructural study. Biomaterials 9:247–252, 1988. Library Holdings [Context Link]
33. Narotam PK, van Dellen JR, Bhoola KD: A clinicopathological study of collagen sponge as a dural graft in neurosurgery. J Neurosurg 82:406–412, 1995. Library Holdings [Context Link]
34. Narotam PK, van Dellen JR, Bhoola KD, Raidoo D: Experimental evaluation of collagen sponge as a dural graft. Br J Neurosurg 7:635–641, 1993. Library Holdings [Context Link]
35. Parizek J, Husek Z, Mericka P, Tera J, Nemecek S, Spacek J, Nemeckova J, Suba P: Ovine pericardium: A new material for duraplasty. J Neurosurg 84:508–513, 1996. Library Holdings [Context Link]
36. Pesch HJ, Stöss H: The solvent-dried dura mater: A new dura transplant in the animal experiment [in German]. Chirurg 48:732–736, 1977. Library Holdings [Context Link]
37. Rauhut F, Grothe W: Die offene Schädel-Hirn-Verletzung. Chir Praxis 30:191–201, 1982. [Context Link]
38. Rehn E: Die Verwendung der autoplastischen Fett-Transplantationen bei Dura- und Hirndefekten. Langenbecks Arch Chir 101:962–996, 1913. [Context Link]
39. Robertson SC, Menezes AH: Hemorrhagic complications in association with Silastic dural substitute. Neurosurgery 40:201–206, 1997. Ovid Full Text Library Holdings [Context Link]
40. Rosomoff HL: Ethylene oxide sterilized freeze-dried dura mater for the repair of pachymeningeal defects. J Neurosurg 16:197–208, 1959. [Context Link]
41. Sacchi E: Del modo di riparare le perdite di sostanza del cranio: Plastiche della dura madre. Riforma Med 337–341, 1893. [Context Link]
42. SAS Institute:SAS Procedures Guide: Version 6.11. Cary, SAS Institute, 1990. [Context Link]
43. Schadel A: An animal experiment study using the Gore-Tex soft tissue patch as a replacement for lyophilized dura mater. HNO 38:95–98, 1990. Library Holdings [Context Link]
44. Selvaggi G: Su talune autoplastiche nelle brecce della pachimenge cerebrale. Policlinico (Chir) 43:76–96, 1936. [Context Link]
45. Sharkey PC, Usher FC, Robertson RCL: Lyophilized human dura mater as a dural substitute. J Neurosurg 15:192–198, 1958. [Context Link]
46. Sierra DH, Nissen AJ, Welch J: The use of fibrin glue in intracranial procedures: Preliminary results. Laryngoscope 100:360–363, 1990. Library Holdings [Context Link]
47. SPSS Inc.:SPSS Software, Release 6.1.3. Chicago, SPSS Inc. [Context Link]
48. Stöss H, Pesch HJ, Wildenauer HD, Tulusan AH: Light and electron microscopy studies on dura transplants preserved with solvents in animal experiments [in German]. Verh Dtsch Ges Pathol 59:569, 1975. [Context Link]
49. Teng P, Feigin I: Vinyon N as a dural substitute: An experimental study in the monkey. J Neurosurg 12:591–600, 1955. [Context Link]
50. Thadani V, Penar PL, Partington J: Creutzfeldt-Jakob disease probably acquired from a cadaveric dura mater graft. J Neurosurg 69:766–769, 1988. Library Holdings [Context Link]
51. Thammavaram KV, Benzel EC, Kesterson L: Fascia lata graft as a dural substitute in neurosurgery. South Med J 83:634–636, 1990. Library Holdings [Context Link]
52. van Loon J, Plets C, Goffin J, van Calenberg F: A prospective study of polyesterurethane (Neuro-Patch) as dura-substitute. Presented at the Symposium “Modern Concepts of Dural Substitution” at the 10th European Congress of Neurosurgery, Berlin, Germany, 1995, Congress Volume. [Context Link]
53. Walton RL, Krizek TJ: The scalp flap only onlay: A method for managing large dural defects. Plast Reconstr Surg 66:684–689, 1980. Library Holdings [Context Link]
54. Wiegand DA, Hartel MI, Quander T, Latz B, Dankle JA: Assessment of cryoprecipitate-thrombin solution for dural repair. Head Neck 16:569–573, 1994. Library Holdings [Context Link]
55. Yamada K, Miyamoto S, Nagata I: Development of a dural substitute from synthetic bioabsorbable polymers. J Neurosurg 86:1012–1017, 1997. Library Holdings
Filippi et al. provide us with a very impressive investigation of dural patch grafts. One hundred sixty-eight individual grafts were carefully analyzed under various conditions, to assess the relative merits of seven different dural substitutes. In general, I think this is a very well-conceived and well-executed study that provides us with some interesting observations with direct clinical significance. However, two minor points do come to mind as I review this article. First, I wonder about the removal of the 4- × 4-mm patch of dura. Theoretically, the dura could be carefully elevated and opened while the underlying arachnoid membrane is left intact. I do not know how easy it is to perform this procedure in rabbits, although I know it certainly is possible in larger animals. I am also not certain how obvious it is when the arachnoid membrane is violated in this particular preparation, although it does seem clear that the integrity of the underlying arachnoid membrane could influence the results of the leakage tests, as described in the experimental protocol. Second, I wonder whether the 3-day results without fibrin glue for Neuropatch are spurious. It could be assumed that the addition of fibrin glue would either have no effect or improve the tightness of the closure. The same can be said for the passage of time. It could be intuitively expected that the 3-month closure results would be as good or better than the 3-day closure results. These intuitive predictions hold true for all groups other than the Neuropatch group. Not only do the mean values trend in the opposite direction for Neuropatch, but visual inspection of the 95% confidence intervals leads me to think that they are statistically significant in the opposite direction. This, coupled with the fact that the 3-day value for Neuropatch without fibrin glue was dramatically higher than that for any other dural substitute, leads me to think that the value of 48.3 mm Hg is, in fact, spurious. I do not fault the statistical analysis performed by the authors; I merely point out that the presence of a significant P value does not guarantee the truth. It only indicates that the a priori probability of obtaining a spurious result would be relatively low if the same experiment was repeated and similar degrees of variance in the data were encountered. In this particular case, I would predict that the results for Neuropatch would be quite different if this experiment was repeated several times. Notwithstanding these minor observations, the authors have performed a very careful and detailed study.
Robert E. Breeze
Denver, Colorado
Many investigations of dural substitutes, in both human subjects and animals, have been reported throughout the years. Human cadaveric dura has been widely accepted as a suitable material for dural reconstruction (2, 6). However, its limited availability, coupled with the risk of slow virus infection, has lead to investigations of allograft pericardium (5, 9) and various biosynthetic polymers (1, 4, 7, 8). Several alternatives now exist that are surgically practical, provide satisfactory barriers to cerebrospinal fluid leakage and infection, and do not produce brain adhesions or inflammation. Very few studies have performed comparisons of the different dural substitutes (3, 10), however, and the present study provides a systematic evaluation of various types of grafts.
Reporting on seven different dural substitutes, Filippi et al. focus on the ability of these materials to provide watertight dural closures in rabbits. At 3 days, 3 weeks, and 3 months after implantation, saline containing fluorescein was infused, at varying pressures, into the cisterna magna of anesthetized subjects; cerebrospinal fluid leakage (fluorescence) detected at the duraplasty site provided a quantitative measure of tightness. The authors also examined the histological characteristics of these implanted materials, with respect to neomembrane formation and adhesion to the underlying brain tissue.
On the basis of this study, Neuropatch seems to be comparable to human cadaveric dura in its ability to provide a watertight seal in a rabbit model, without the use of fibrin glue. Furthermore, Neuropatch is apparently easy to use and causes no underlying brain adhesion. Given the stated disadvantages of human cadaveric dura, Neuropatch can be considered an appropriate dural substitute.
Dural substitutes are often used in clinical situations in which increased intracranial pressure (e.g., open head injuries) or hydrocephalus (e.g., tumors or infections affecting cerebrospinal fluid resorption) contributes to successful reconstitution without leakage. This series of experiments does not address the possible contribution of these factors to the dural seal. Furthermore, the use of sham-operated animals (in which the dura was opened and then primarily closed) would have placed these results in a broader neurosurgical context. Lastly, bovine pericardium, which is used in many centers, was not included in the comparison.
Arthur L. Day
Christopher G. Gaposchkin
Gainesville, Florida
1. Bhatia S, Bergethon PR, Blease S: A synthetic dural prothesis constructed from hydroxyethylmethacrylate hydrogels. J Neurosurg 83:897–902, 1995. Library Holdings [Context Link]
2. Cantore G, Guidetti B, Delfini R: Neurosurgical use of human dura mater sterilized by gamma rays and stored in alcohol: Long-term results. J Neurosurg 66:93–95, 1987. Library Holdings [Context Link]
3. Cargill HA Jr, Cawley CM, Barrow DL, Poff BC, Powell MD, Sawhney AS, Dillehay DL: Efficacy and biocompatibility of a photopolymerized, synthetic, absorbable hydrogel as a dural sealant in a canine craniotomy model. J Neurosurg 88:308–313, 1998. Library Holdings [Context Link]
4. Laquerriere A, Yun J, Tiollier J, Hemet J, Tadie M: Experimental evaluation of bilayered human collagen as a dural substitute. J Neurosurg 78:487–491, 1993. Library Holdings [Context Link]
5. Laun A, Tonn JC, Jerusalem C: Comparative study of lyophilized human dura mater and lyophilized bovine pericardium as dural substitutes in neurosurgery. Acta Neurochir (Wien) 107:16–21, 1990. Library Holdings [Context Link]
6. Macfarlane MR, Symon L: Lyophilised dura mater: Experimental implantation and extended clinical neurosurgical use. J Neurol Neurosurg Psychiatry 42:854–858, 1979. Library Holdings [Context Link]
7. Mello LR, Feltrin LT, Fontes-Neto PT, Ferraz FA: Duraplasty with biosynthetic cellulose: An experimental study. J Neurosurg 86:143–150, 1997. Library Holdings [Context Link]
8. Narotam PK, Van Dellen JR, Bhoola KD: A clinicopathological study of collagen sponge as a dural graft in neurosurgery. J Neurosurg 82:406–412, 1995. Library Holdings [Context Link]
9. Parizek J, Husek Z, Mericka P, Tera J, Nemecek S, Spacek J, Nemeckova J, Suba P: Ovine pericardium: A new material for duraplasty. J Neurosurg 84:508–513, 1996. Library Holdings [Context Link]
10. Yamada K, Miyamoto S, Nagata I: Development of a dural substitute from synthetic bioabsorbable polymers. J Neurosurg 86:1012–1017, 1997. Library Holdings [Context Link]
This comparative study of materials used for duraplasty was well performed, and the model is sufficient for the study. Duraplasty suturing was performed and the use of fibrin glue was assessed appropriately. The conclusion that Lyodura and Neuropatch seem to be the best materials, in comparison with the other materials tested, is justified. For surgeons who seek to avoid cadaver-derived dura, it seems that Neuropatch (a microporous fleece composed of cleaned aliphatic polyesterurethane) is quite satisfactory with respect to tightness, as well as clinical use, and is not associated with adhesions. The long-term human tolerance of Neuropatch is not discussed, but these results are promising.
James T. Robertson
Memphis, Tennessee
Filippi et al. have studied an important issue that is routinely faced by neurosurgeons in their surgical practices, i.e., the tightness of duraplasty closure. The authors meticulously studied several techniques and concluded that Lyodura and Neuropatch are appropriate patch alternatives for dural reconstruction. This type of objective information is very useful for practicing surgeons, and the authors have provided a significant service.
Edward C. Benzel
Cleveland, Ohio
This interesting study compares the tightness of various homologous and synthetic materials used for duraplasty. The experiments were conducted in rabbits. Certain conclusions were drawn regarding the greater ability of some materials, compared with others, to achieve hermetic closures.
However, the study may have a potential flaw, because observations from experiments performed on rabbits may not translate into similar results for human subjects. I suspect that rabbit tissues and human tissues do not necessarily react in the same ways (because of their different antigenicities, for example). Consequently, the conclusions reached by the authors may not be valid for human subjects, and I am not at all sure that the conclusions presented by the authors should be taken as guidelines for human use. Finally, if the conclusions are incorrect with respect to human subjects, they may be unfair to the manufacturers of duraplasty materials that were proven inferior in these experiments.
Ivan Ciric
Evanston, Illinois
Key words: Dural substitution materials; Duraplasty; Fibrin glue; Leakage pressure; Rabbits