- Email: [email protected]
Innate healing in the fetal sheep model of myelomeningocele: A standardized defect grading system
Journal of Pediatric Surgery xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Innate healing in the fetal sheep model of myelomeningocele: A standardized defect grading system Erin G. Brown ⁎, Benjamin A. Keller, Christopher D. Pivetti, Diana L. Farmer University of California, Davis Health System, Sacramento, CA, USA
a r t i c l e
i n f o
Article history: Received 21 July 2014 Received in revised form 1 October 2014 Accepted 1 October 2014 Available online xxxx Key words: Myelomeningocele Fetal healing Grading system Spina bifida
a b s t r a c t Background/Purpose: The fetal sheep model of myelomeningocele (MMC) is well-established. While the variability of innate fetal healing of the defect at the time of the repair operation has been acknowledged, it remains poorly described. We characterized the healing within the fetal sheep MMC model and present a standardized defect grading system. Methods: Forty-three fetuses underwent surgical MMC creation at a gestational age of 75 days and repair at 100 days. Defects were graded based on percentage of exposed spinal cord and degree of scarring. A scale was developed and validated by interrater reliability testing. Results: The phenotypic appearance of the defect at the time of the repair operation was highly variable. The defect length ranged from 3.0 to 5.5 cm. Three spinal cords remained fully exposed across their length; 18 were completely covered in fibrinous exudate. Twenty-two fetuses demonstrated varying degrees of open spinal cord. Interrater reliability revealed consistent grades 91% of the time. Conclusion: Despite consistent defect creation, the fetal MMC defect has a wide spectrum of appearance at the time of the second surgery. This study describes the innate fetal healing within the model and presents a feasible and valid grading scale to ensure scientific rigor within the model. © 2014 Elsevier Inc. All rights reserved.
The fetal sheep model of myelomeningocele (MMC), as described by Meuli et al, [1], has been used for over two decades and is the most commonly utilized large animal model of MMC. The relative tolerance of the sheep uterus to instrumentation, long gestation period, and large size of the fetal lamb have facilitated the use of this large animal model for MMC. The surgical technique of lumbar laminectomy and dural excision described in this study has been adopted by surgeon-researchers worldwide with general consensus that the best results are obtained when defect creation is performed at a gestational age of 75 days with subsequent repair occurring at a gestational age of 100 days. While the operation to create the MMC defect is standardized and the resultant histologic and neurologic outcomes are well characterized, the varying degree of innate fetal defect closure at the time of the repair operation is not well described. Many researchers proficient with the fetal sheep model of MMC have alluded to the inconsistent appearance of the defect at the time of the second operation [2–7]. However, the degree of wound closure has never been described in detail, and no formal evaluation of this variability has been conducted. Although the clinical significance of the amount of healing and variability in healing seen at the repair procedure is not yet known, it is not unreasonable to believe that the degree of healing may have an impact on resultant neurologic function. We ⁎ Corresponding author at: University of California, Davis Department of Surgery, 2315 Stockton Blvd, OP512, Sacramento, CA 95817, USA. Tel.: + 1 916 734 2724; fax: + 1 916 734 5633. E-mail address: [email protected] (E.G. Brown).
propose that standardized assessment of this healing should be incorporated into all studies using the fetal sheep MMC model in order to improve our understanding and to ensure adequate scientific rigor within the model. 1. Methods Forty-three fetuses were used to develop and validate the defect grading scale. An MMC defect was surgically created in all fetal lambs as previously described [2,8]. In brief, each ewe underwent a survival laparotomy and hysterotomy at a gestational age of approximately 75 days. The defect was created by removing a large area of overlying skin, soft tissue, and the paraspinal muscles, followed by removal of six lumbar vertebrae to expose the spinal cord. Lastly, the dura over the exposed cord was excised. Defects, at the time of creation, were measured (length and width of skin defect and length of spinal cord defect) (Fig. 1). These data were analyzed for consistency. The uterus was then filled with normal saline and closed in an imbricated running fashion with absorbable suture. Approximately twenty-five days later, a second survival laparotomy and hysterotomy was performed. At the time of the second surgery, intraoperative photographs of the resultant defect were taken, and notes on defect length, length of exposed spinal cord segments, and degree of scarring were recorded. Operative records, including notes and intraoperative photographs, were reviewed to develop the defect grading system. Defects were graded based on the percentage of exposed spinal cord and the degree/thickness of scarring over the remaining cord based on
http://dx.doi.org/10.1016/j.jpedsurg.2014.10.021 0022-3468/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Brown EG, et al, Innate healing in the fetal sheep model of myelomeningocele: A standardized defect grading system, J Pediatr Surg (2014), http://dx.doi.org/10.1016/j.jpedsurg.2014.10.021
2
E.G. Brown et al. / Journal of Pediatric Surgery xxx (2014) xxx–xxx
operation (Fig. 1). On average, the MMC defect was surgically created on gestational age of 77 days (range: 73 to 81). The spinal cord defect was consistently 3.0 ± 0.4 cm in length (mean: 3.0 cm, median 3.0 cm). The skin defect was also consistent at 4.0 ± 0.5 cm by 4.0 ± 0.5 cm (mean length: 4.2 cm, median 4.0 cm; mean width: 4.0 cm, median 4.0 cm). Six lumbar vertebrae were removed in each fetus, and dura was removed from the entire length of the exposed spinal cord. The appearance of the defect at the time of the second surgery was highly variable among the fetuses studied (Table 2). While the average length of the spinal cord defect was 3.7 cm, the range in length was substantial from 3.0 cm to 5.5 cm. Three spinal cords remained fully exposed across their length, while 18 defects were covered in varying degrees of fibrinous exudate over the entire length of the lesion. The remaining 22 fetuses demonstrated varying degrees of open spinal cord/ defect healing. Defects were assigned a numerical grade of 0 through 3 based on the percentage of exposed spinal cord, as described above. Results of our interrater reliability testing demonstrated the feasibility of applying this scale both intraoperatively and retrospectively via photographic documentation. Grades were consistent among all three reviewers 91% of the time. In the cases of discordance, grades were within 1 point 100% of the time. Scores assigned intraoperatively matched photographically assessed scores 100% of the time. Fleiss’ kappa demonstrated excellent agreement among reviewers (κ = 0.92). Fig. 1. Operative creation of the defect: typical appearance of defect. Average length of spinal cord defect is 3.0 cm. Skin defect is approximately 4.0 cm by 4.0 cm.
intraoperative photographs and notes from the second surgery (Table 1). Cords covered by scar tissue in their entirety received a grade of 0. Up to 33% of exposed cord were Grade 1, 34%–66% were Grade 2, and 67%–100% were Grade 3 (Fig. 2). The degree of exudate over the covered cord was noted to be either minimal or moderate in quantity (Fig. 3). To provide an objective measure of the amount of scarring present, the ability to visualize the cord through the exudate was used. Defects with transparent scar tissue were denoted with an “a,” while opaque scar was noted as “b” (Fig. 2). All individuals involved in surgical defect creation were trained in the grading scale. Following defect grading, the scar tissue overlying the spinal cord was removed in order to reexpose the spinal cord. After experimental treatment was performed, normal saline was used to restore amniotic fluid volume and standard uterine closure was performed, as described above. Consistency of the grading scale was tested by assessing interrater reliability for all forty-three fetuses. Three individuals, blinded to the operating surgeon and date of procedure, independently reviewed operative photographs from the repair surgery to determine a grade, and all scores were compared. Additionally, to evaluate reliability of photographic grades compared to live intraoperative grades, scores assigned intraoperatively by the operating surgeon were compared to grades assigned by blinded reviewers based on intraoperative photographs in five fetuses. These scores were compared for accuracy. Statistical analysis was performed with percentage of concordance and Fleiss’ Kappa statistic. 2. Results Overall fetal mortality was 66%. Review of operative notes and photographs demonstrated consistency in resulting defects from the first Table 1 Grading system. Grade
% Exposed spinal cord
Class
Degree of fibrinous exudate
0 1 2 3
0% 1%–33% 34%–66% 67%–100%
a b
Minimal (transparent) Moderate (opaque)
Defects were assigned a grade of 0, 1, 2, or 3 based on the percentage of exposed spinal cord. They were additionally characterized by the degree of fibrinous exudate present; class “a” was considered minimal (or transparent), while “b” was moderate (or opaque).
3. Discussion The fetal sheep model is a well-characterized model of MMC. Although the defect is surgically created and not developmentally acquired, the resultant spinal cord defect is remarkably similar to that which naturally occurs in humans [8]. Furthermore, previous work has led to current clinical application [9]. However, despite consistent creation of a uniform defect at the initial surgery, the phenotypic appearance of the resulting defect is widely variable at the time of the second surgery. The grading scale presented here represents a standardized assessment of the degree of naturally occurring fetal healing. The current study demonstrates that in the face of uniform defect creation at the initial surgery and minimal variation in size, the phenotypic appearance of the resulting defect remains highly variable. There was no correlation between larger defects at the first surgery and increased percentage of exposed spinal cord or less scar at the second surgery. The reasons for this variability are unknown, and while precision in defect creation is a critical component of this model, it may not be possible to control the degree of innate fetal healing within this model. The phases of adult wound healing (inflammation, proliferation, and remodeling) have been well described in the literature [10,11]. The result is a wound with less tensile strength than the original tissue and a visible scar. However, in contrast to the adult, the fetus has the remarkable ability to heal wounds without scar formation [12]. This unique property has been thoroughly investigated and fetal wound healing has been shown to be accelerated, completely regenerate the dermis, and lack fibrosis associated with adult scars [13–15]. Further understanding distinctions and their impact on healing within the fetal sheep model of MMC may be critical to the successful of fetal surgery for MMC. The impact of the degree of fetal healing on neurologic outcome is unknown at this time, but it may contribute important prognostic information for neurologic function. For instance, animals with spinal cords completely covered by granulation tissue or other signs of healing may be exposed to less injury from intrauterine trauma or amniotic fluid exposure. Alternatively, fetuses with a greater degree of healing may have had a more significant injury at the time of defect creation leading to a more pronounced inflammatory response. Future studies could perform histologic assessment of the defect at 100 days gestation in order to delineate the relationship between the defect phenotype and the degree of spinal cord injury. Given the time and financial investment involved with the fetal sheep model, we propose that consistent
Please cite this article as: Brown EG, et al, Innate healing in the fetal sheep model of myelomeningocele: A standardized defect grading system, J Pediatr Surg (2014), http://dx.doi.org/10.1016/j.jpedsurg.2014.10.021
E.G. Brown et al. / Journal of Pediatric Surgery xxx (2014) xxx–xxx
3
Fig. 2. MMC numeric grade. (0) Entire length of defect is covered in fibrinous exudate. (1) 0–33% of exposed cord is open. (2) 34%–66% of exposed cord is open. (3) 67%–100% of exposed cord is open.
needed to investigate possible correlations. Consistency in the surgical creation of the defect is undeniably important in this model. We argue that the same consistency should be applied to documentation and reporting of the defect appearance at the time of repair in order to ensure adequate scientific rigor within the model. The grading scale presented in this study is a feasible, accurate method of standardized assessment for use in the fetal sheep model of MMC. References
Fig. 3. Degree of Scar. (A) Minimal amount of fibrinous exudate (transparent). (B) Moderate amount of fibrinous exudate (opaque).
reporting of defect appearance may help elucidate the relationship between phenotype and degree of injury, add scientific rigor to the model, and aid research for all who utilize the model. We acknowledge that the significance of the phenotypic variability remains largely unknown, but we believe that this defect grading system will enhance the scientific rigor of the model. While the clinical importance of such variability is unknown, the potential significance should not be overlooked, and future studies are
[1] Meuli M, Mueli-Simmen C, Hutchins GM, et al. In utero surgery rescues neurologic function at birth in sheep with spina bifida. Nature Med 1995;1(4):342–7. [2] Mueli M, Mueli-Simmen C, Yingling CD, et al. In utero repair of experimental myelomeningocele saves neurologic function at birth. J Pediatr Surg 1996;31(3):397–402. [3] Eggink AJ, Roelofs LA, Feitz WF, et al. Delayed intrauterine repair of an experimental spina bifida with a collagen biomatrix. Pediatr Neurosurg 2008;44(1):29–35. [4] Eggink AJ, Roelofs LA, Feitz WF, et al. In utero repair of an experimental neural tube defect in a chronic sheep model using biomatrices. Fetal Diagn Ther 2005;20(5): 335–40. [5] von Koch CS, Compagnone A, Hirose S, et al. Myelomeningocele: characterization of a surgically induced sheep model and its central nervous system similarities and differences to the human disease. Am J Obstet Gynecol 2005;193(4):1456–62. [6] Encinas JL, Garcia-Cabezas MA, Barkovich J, et al. Maldevelopment of the cerebral cortex in the surgically induced model of myelomeningocele: implications for fetal neurosurgery. J Pediatr Surg 2011;46(4):713–22. [7] Fontecha CG, Peiro JL, Aguirre M, et al. Inert patch with bioadhesive for gentle fetal surgery of myelomeningocele in a sheep model. Eur J Obstet Gynecol Reprod Biol 2009;146(2):174–9. [8] Mueili M, Meuli-Simmen C, Yingling CD, et al. Creation of myelomeningocle in utero: a model of functional damage from spinal cord exposure in fetal sheep. J Pediatr Surg 1995;30(7):1028–33. [9] Adzick NS TE, Spong CY, Brock JW, et al. A randomized trial of prenatal versus postnatal repair of myelomeningocele. N Engl J Med 2011;364(11):993–1004. [10] Longaker M, Gurtner GC. Introduction: wound repair. Semin Cell Dev Biol 2012;23:945. [11] Hu M, Maan ZN, Wu JC, et al. Tissue engineering and regenerative repair in wound healing. Ann Biomed Eng 2014;42:1494–507. [12] Rowlatt U. Intrauterine wound healing in a 20 week human fetus. Virchows Arch A Pathol Anat Histol 1979;381:353–61. [13] Adzick N, Longaker MT. Animal models for the study of fetal tissue repair. J Surg Res 1991;51:216–22. [14] Adzick N, Longaker MT. Scarless fetal healing. Therapeutic implications. Ann Surg 1992;215:3–7. [15] Lorenz H, Longaker MT, Perkocha LA, et al. Scarless wound repair: a human fetal skin model. Development 1992;114:253–9.
Table 2 Distribution of grades: number of lambs (and percentage) demonstrating each grade at the time of the second surgery. Grade
n (%)
0a 0b 1a 1b 2a 2b 3a 3b
8 (19) 10 (23) 6 (14) 4 (9) 11 (26) 1 (2) 3(7) 0 (0)
Please cite this article as: Brown EG, et al, Innate healing in the fetal sheep model of myelomeningocele: A standardized defect grading system, J Pediatr Surg (2014), http://dx.doi.org/10.1016/j.jpedsurg.2014.10.021
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Innate healing in the fetal sheep model of myelomeningocele: A standardized defect grading system Erin G. Brown ⁎, Benjamin A. Keller, Christopher D. Pivetti, Diana L. Farmer University of California, Davis Health System, Sacramento, CA, USA
a r t i c l e
i n f o
Article history: Received 21 July 2014 Received in revised form 1 October 2014 Accepted 1 October 2014 Available online xxxx Key words: Myelomeningocele Fetal healing Grading system Spina bifida
a b s t r a c t Background/Purpose: The fetal sheep model of myelomeningocele (MMC) is well-established. While the variability of innate fetal healing of the defect at the time of the repair operation has been acknowledged, it remains poorly described. We characterized the healing within the fetal sheep MMC model and present a standardized defect grading system. Methods: Forty-three fetuses underwent surgical MMC creation at a gestational age of 75 days and repair at 100 days. Defects were graded based on percentage of exposed spinal cord and degree of scarring. A scale was developed and validated by interrater reliability testing. Results: The phenotypic appearance of the defect at the time of the repair operation was highly variable. The defect length ranged from 3.0 to 5.5 cm. Three spinal cords remained fully exposed across their length; 18 were completely covered in fibrinous exudate. Twenty-two fetuses demonstrated varying degrees of open spinal cord. Interrater reliability revealed consistent grades 91% of the time. Conclusion: Despite consistent defect creation, the fetal MMC defect has a wide spectrum of appearance at the time of the second surgery. This study describes the innate fetal healing within the model and presents a feasible and valid grading scale to ensure scientific rigor within the model. © 2014 Elsevier Inc. All rights reserved.
The fetal sheep model of myelomeningocele (MMC), as described by Meuli et al, [1], has been used for over two decades and is the most commonly utilized large animal model of MMC. The relative tolerance of the sheep uterus to instrumentation, long gestation period, and large size of the fetal lamb have facilitated the use of this large animal model for MMC. The surgical technique of lumbar laminectomy and dural excision described in this study has been adopted by surgeon-researchers worldwide with general consensus that the best results are obtained when defect creation is performed at a gestational age of 75 days with subsequent repair occurring at a gestational age of 100 days. While the operation to create the MMC defect is standardized and the resultant histologic and neurologic outcomes are well characterized, the varying degree of innate fetal defect closure at the time of the repair operation is not well described. Many researchers proficient with the fetal sheep model of MMC have alluded to the inconsistent appearance of the defect at the time of the second operation [2–7]. However, the degree of wound closure has never been described in detail, and no formal evaluation of this variability has been conducted. Although the clinical significance of the amount of healing and variability in healing seen at the repair procedure is not yet known, it is not unreasonable to believe that the degree of healing may have an impact on resultant neurologic function. We ⁎ Corresponding author at: University of California, Davis Department of Surgery, 2315 Stockton Blvd, OP512, Sacramento, CA 95817, USA. Tel.: + 1 916 734 2724; fax: + 1 916 734 5633. E-mail address: [email protected] (E.G. Brown).
propose that standardized assessment of this healing should be incorporated into all studies using the fetal sheep MMC model in order to improve our understanding and to ensure adequate scientific rigor within the model. 1. Methods Forty-three fetuses were used to develop and validate the defect grading scale. An MMC defect was surgically created in all fetal lambs as previously described [2,8]. In brief, each ewe underwent a survival laparotomy and hysterotomy at a gestational age of approximately 75 days. The defect was created by removing a large area of overlying skin, soft tissue, and the paraspinal muscles, followed by removal of six lumbar vertebrae to expose the spinal cord. Lastly, the dura over the exposed cord was excised. Defects, at the time of creation, were measured (length and width of skin defect and length of spinal cord defect) (Fig. 1). These data were analyzed for consistency. The uterus was then filled with normal saline and closed in an imbricated running fashion with absorbable suture. Approximately twenty-five days later, a second survival laparotomy and hysterotomy was performed. At the time of the second surgery, intraoperative photographs of the resultant defect were taken, and notes on defect length, length of exposed spinal cord segments, and degree of scarring were recorded. Operative records, including notes and intraoperative photographs, were reviewed to develop the defect grading system. Defects were graded based on the percentage of exposed spinal cord and the degree/thickness of scarring over the remaining cord based on
http://dx.doi.org/10.1016/j.jpedsurg.2014.10.021 0022-3468/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Brown EG, et al, Innate healing in the fetal sheep model of myelomeningocele: A standardized defect grading system, J Pediatr Surg (2014), http://dx.doi.org/10.1016/j.jpedsurg.2014.10.021
2
E.G. Brown et al. / Journal of Pediatric Surgery xxx (2014) xxx–xxx
operation (Fig. 1). On average, the MMC defect was surgically created on gestational age of 77 days (range: 73 to 81). The spinal cord defect was consistently 3.0 ± 0.4 cm in length (mean: 3.0 cm, median 3.0 cm). The skin defect was also consistent at 4.0 ± 0.5 cm by 4.0 ± 0.5 cm (mean length: 4.2 cm, median 4.0 cm; mean width: 4.0 cm, median 4.0 cm). Six lumbar vertebrae were removed in each fetus, and dura was removed from the entire length of the exposed spinal cord. The appearance of the defect at the time of the second surgery was highly variable among the fetuses studied (Table 2). While the average length of the spinal cord defect was 3.7 cm, the range in length was substantial from 3.0 cm to 5.5 cm. Three spinal cords remained fully exposed across their length, while 18 defects were covered in varying degrees of fibrinous exudate over the entire length of the lesion. The remaining 22 fetuses demonstrated varying degrees of open spinal cord/ defect healing. Defects were assigned a numerical grade of 0 through 3 based on the percentage of exposed spinal cord, as described above. Results of our interrater reliability testing demonstrated the feasibility of applying this scale both intraoperatively and retrospectively via photographic documentation. Grades were consistent among all three reviewers 91% of the time. In the cases of discordance, grades were within 1 point 100% of the time. Scores assigned intraoperatively matched photographically assessed scores 100% of the time. Fleiss’ kappa demonstrated excellent agreement among reviewers (κ = 0.92). Fig. 1. Operative creation of the defect: typical appearance of defect. Average length of spinal cord defect is 3.0 cm. Skin defect is approximately 4.0 cm by 4.0 cm.
intraoperative photographs and notes from the second surgery (Table 1). Cords covered by scar tissue in their entirety received a grade of 0. Up to 33% of exposed cord were Grade 1, 34%–66% were Grade 2, and 67%–100% were Grade 3 (Fig. 2). The degree of exudate over the covered cord was noted to be either minimal or moderate in quantity (Fig. 3). To provide an objective measure of the amount of scarring present, the ability to visualize the cord through the exudate was used. Defects with transparent scar tissue were denoted with an “a,” while opaque scar was noted as “b” (Fig. 2). All individuals involved in surgical defect creation were trained in the grading scale. Following defect grading, the scar tissue overlying the spinal cord was removed in order to reexpose the spinal cord. After experimental treatment was performed, normal saline was used to restore amniotic fluid volume and standard uterine closure was performed, as described above. Consistency of the grading scale was tested by assessing interrater reliability for all forty-three fetuses. Three individuals, blinded to the operating surgeon and date of procedure, independently reviewed operative photographs from the repair surgery to determine a grade, and all scores were compared. Additionally, to evaluate reliability of photographic grades compared to live intraoperative grades, scores assigned intraoperatively by the operating surgeon were compared to grades assigned by blinded reviewers based on intraoperative photographs in five fetuses. These scores were compared for accuracy. Statistical analysis was performed with percentage of concordance and Fleiss’ Kappa statistic. 2. Results Overall fetal mortality was 66%. Review of operative notes and photographs demonstrated consistency in resulting defects from the first Table 1 Grading system. Grade
% Exposed spinal cord
Class
Degree of fibrinous exudate
0 1 2 3
0% 1%–33% 34%–66% 67%–100%
a b
Minimal (transparent) Moderate (opaque)
Defects were assigned a grade of 0, 1, 2, or 3 based on the percentage of exposed spinal cord. They were additionally characterized by the degree of fibrinous exudate present; class “a” was considered minimal (or transparent), while “b” was moderate (or opaque).
3. Discussion The fetal sheep model is a well-characterized model of MMC. Although the defect is surgically created and not developmentally acquired, the resultant spinal cord defect is remarkably similar to that which naturally occurs in humans [8]. Furthermore, previous work has led to current clinical application [9]. However, despite consistent creation of a uniform defect at the initial surgery, the phenotypic appearance of the resulting defect is widely variable at the time of the second surgery. The grading scale presented here represents a standardized assessment of the degree of naturally occurring fetal healing. The current study demonstrates that in the face of uniform defect creation at the initial surgery and minimal variation in size, the phenotypic appearance of the resulting defect remains highly variable. There was no correlation between larger defects at the first surgery and increased percentage of exposed spinal cord or less scar at the second surgery. The reasons for this variability are unknown, and while precision in defect creation is a critical component of this model, it may not be possible to control the degree of innate fetal healing within this model. The phases of adult wound healing (inflammation, proliferation, and remodeling) have been well described in the literature [10,11]. The result is a wound with less tensile strength than the original tissue and a visible scar. However, in contrast to the adult, the fetus has the remarkable ability to heal wounds without scar formation [12]. This unique property has been thoroughly investigated and fetal wound healing has been shown to be accelerated, completely regenerate the dermis, and lack fibrosis associated with adult scars [13–15]. Further understanding distinctions and their impact on healing within the fetal sheep model of MMC may be critical to the successful of fetal surgery for MMC. The impact of the degree of fetal healing on neurologic outcome is unknown at this time, but it may contribute important prognostic information for neurologic function. For instance, animals with spinal cords completely covered by granulation tissue or other signs of healing may be exposed to less injury from intrauterine trauma or amniotic fluid exposure. Alternatively, fetuses with a greater degree of healing may have had a more significant injury at the time of defect creation leading to a more pronounced inflammatory response. Future studies could perform histologic assessment of the defect at 100 days gestation in order to delineate the relationship between the defect phenotype and the degree of spinal cord injury. Given the time and financial investment involved with the fetal sheep model, we propose that consistent
Please cite this article as: Brown EG, et al, Innate healing in the fetal sheep model of myelomeningocele: A standardized defect grading system, J Pediatr Surg (2014), http://dx.doi.org/10.1016/j.jpedsurg.2014.10.021
E.G. Brown et al. / Journal of Pediatric Surgery xxx (2014) xxx–xxx
3
Fig. 2. MMC numeric grade. (0) Entire length of defect is covered in fibrinous exudate. (1) 0–33% of exposed cord is open. (2) 34%–66% of exposed cord is open. (3) 67%–100% of exposed cord is open.
needed to investigate possible correlations. Consistency in the surgical creation of the defect is undeniably important in this model. We argue that the same consistency should be applied to documentation and reporting of the defect appearance at the time of repair in order to ensure adequate scientific rigor within the model. The grading scale presented in this study is a feasible, accurate method of standardized assessment for use in the fetal sheep model of MMC. References
Fig. 3. Degree of Scar. (A) Minimal amount of fibrinous exudate (transparent). (B) Moderate amount of fibrinous exudate (opaque).
reporting of defect appearance may help elucidate the relationship between phenotype and degree of injury, add scientific rigor to the model, and aid research for all who utilize the model. We acknowledge that the significance of the phenotypic variability remains largely unknown, but we believe that this defect grading system will enhance the scientific rigor of the model. While the clinical importance of such variability is unknown, the potential significance should not be overlooked, and future studies are
[1] Meuli M, Mueli-Simmen C, Hutchins GM, et al. In utero surgery rescues neurologic function at birth in sheep with spina bifida. Nature Med 1995;1(4):342–7. [2] Mueli M, Mueli-Simmen C, Yingling CD, et al. In utero repair of experimental myelomeningocele saves neurologic function at birth. J Pediatr Surg 1996;31(3):397–402. [3] Eggink AJ, Roelofs LA, Feitz WF, et al. Delayed intrauterine repair of an experimental spina bifida with a collagen biomatrix. Pediatr Neurosurg 2008;44(1):29–35. [4] Eggink AJ, Roelofs LA, Feitz WF, et al. In utero repair of an experimental neural tube defect in a chronic sheep model using biomatrices. Fetal Diagn Ther 2005;20(5): 335–40. [5] von Koch CS, Compagnone A, Hirose S, et al. Myelomeningocele: characterization of a surgically induced sheep model and its central nervous system similarities and differences to the human disease. Am J Obstet Gynecol 2005;193(4):1456–62. [6] Encinas JL, Garcia-Cabezas MA, Barkovich J, et al. Maldevelopment of the cerebral cortex in the surgically induced model of myelomeningocele: implications for fetal neurosurgery. J Pediatr Surg 2011;46(4):713–22. [7] Fontecha CG, Peiro JL, Aguirre M, et al. Inert patch with bioadhesive for gentle fetal surgery of myelomeningocele in a sheep model. Eur J Obstet Gynecol Reprod Biol 2009;146(2):174–9. [8] Mueili M, Meuli-Simmen C, Yingling CD, et al. Creation of myelomeningocle in utero: a model of functional damage from spinal cord exposure in fetal sheep. J Pediatr Surg 1995;30(7):1028–33. [9] Adzick NS TE, Spong CY, Brock JW, et al. A randomized trial of prenatal versus postnatal repair of myelomeningocele. N Engl J Med 2011;364(11):993–1004. [10] Longaker M, Gurtner GC. Introduction: wound repair. Semin Cell Dev Biol 2012;23:945. [11] Hu M, Maan ZN, Wu JC, et al. Tissue engineering and regenerative repair in wound healing. Ann Biomed Eng 2014;42:1494–507. [12] Rowlatt U. Intrauterine wound healing in a 20 week human fetus. Virchows Arch A Pathol Anat Histol 1979;381:353–61. [13] Adzick N, Longaker MT. Animal models for the study of fetal tissue repair. J Surg Res 1991;51:216–22. [14] Adzick N, Longaker MT. Scarless fetal healing. Therapeutic implications. Ann Surg 1992;215:3–7. [15] Lorenz H, Longaker MT, Perkocha LA, et al. Scarless wound repair: a human fetal skin model. Development 1992;114:253–9.
Table 2 Distribution of grades: number of lambs (and percentage) demonstrating each grade at the time of the second surgery. Grade
n (%)
0a 0b 1a 1b 2a 2b 3a 3b
8 (19) 10 (23) 6 (14) 4 (9) 11 (26) 1 (2) 3(7) 0 (0)
Please cite this article as: Brown EG, et al, Innate healing in the fetal sheep model of myelomeningocele: A standardized defect grading system, J Pediatr Surg (2014), http://dx.doi.org/10.1016/j.jpedsurg.2014.10.021