Epidural fibrosis results from the surgical intervention in the epidural space. The scar tissue envelops the nerve roots and covers the exposed portion of the dura within the spinal canal. Although this process is normal and an expected result of healing of the surgical site or wound, actual formation of scar tissue and adhesions may lead to a less than optimal clinical outcome. Specifically, epidural fibrosis has been associated with the persistence or recurrence of back pain after discectomy [
A variety of methods have been used in an attempt to prevent and reduce epidural fibrosis. The most notable method may be the ADCON-L (Gliatech Inc., Cleveland, OH), a carbohydrate polymer gel designed to provide a mechanical barrier to inhibit fibroblast migration and adhesion. While the early results of the ADCON-L showed reduced scar tissue formation and improved outcomes [
Another antiadhesion gel (Oxiplex or Medishield, Fziomed Inc., San Luis Obispo, CA), composed of carboxymethylcellulose and polyethylene oxide, has been shown to improve some outcomes and reduce the rate of reoperation [
The use of human amniotic membrane based products has also been considered for the prevention of epidural fibrosis. Amniotic membrane, comprised of both amnion and chorion membranes, is metabolically active tissue which continually remodels the extracellular matrix through processes controlled by paracrine growth factors [
Dehydrated human amnion/chorion membrane (dHACM) is a dehydrated human allograft comprised of laminated amnion and chorion membranes derived from donated human placentas according to the American Association of Tissue Banks (AATB) standards and is considered a tissue product under Section 361 of the Public Health Service Act. PURION processed dHACM has been shown to retain the growth factors in natural amniotic membrane including PDGF-AA, PDGF-BB, bFGF, TGF-
Our primary objective was to evaluate the use of dehydrated human amnion/chorion membrane (dHACM) allograft (AmnioFix, MiMedx, Marietta, GA) and the formation of soft tissue scarring in the epidural space, in patients having transforaminal lumbar interbody lumbar fusion (TLIF) with posterior instrumentation and subsequent reexploration with instrumentation removal. Secondary outcomes included intraoperative and postoperative complications related to dHACM and patient outcomes relative to disability, pain, and functional health.
The study design, protocol, and consent forms (clinical trial number:
Patients underwent standard TLIF with posterior pedicle screw fixation. After the insertion of the interbody fusion device, bone graft material, and posterior stabilization devices, the dHACM (AmnioFix, MiMedx, Marietta, GA) was oriented with the appropriate side up, cut to fit the dura exposed by the decompression, and placed in the epidural space. Meticulous hemostasis was achieved and a hemovac drain placed prior to routine closure.
After radiological studies had been performed to document adequate fusion, as evidenced by bridging bone within the interbody space and adjacent intertransverse region, the patient was scheduled for instrumentation removal with exploration of the epidural space. After exposure, the hardware was removed and the pedicles reconstructed with crushed cancellous allograft. The ease of dissection/extent of adhesions in the epidural space was determined by the surgeon and a piece of tissue adjacent to the epidural space was sent for histopathologic analysis.
During the instrumentation removal surgery, the spine surgeon assessed the degree of scar formation adjacent to the epidural space using a 4-level scoring system ranging from “no adhesion” to “sharp dissection required” (Table
Scoring scheme for the presence of adhesions and quality of the dissection plane.
Score | Description |
---|---|
1 |
|
|
|
2 |
|
|
|
3 |
|
|
|
4 |
|
During the instrumentation removal surgery, a small sample of the scar tissue adjacent to the epidural space was collected for histological analysis. A 1-cm2 section of the lumbar tissue was removed with a small scalpel blade, clearly marking the surface adjacent to the dura. The tissue sample was treated with H & E staining. The slides were evaluated for the presence and extent of scar tissue (mm of fibrosis and percent of fat infiltration).
Intraoperative and postoperative complications were recorded. Patients were evaluated at three weeks, three months, six months, and twelve months after the TLIF surgery and again after the instrumentation removal surgery. Patients’ outcomes were collected using standard validated questionnaires including Oswestry Disability Index (ODI), Numerical Rating Pain Scales, and SF-36. The average change in patient outcomes was assessed with repeated-measures analysis (SPSS Inc, Chicago, IL) with alpha = 0.05.
During the study period a total of 5 patients (males, average age 61 years) without history of prior spinal surgery underwent TLIF with dHACM and subsequent reexploration. Patient diagnosis, levels of TLIF, adhesion score assessed at reexploration, and histological analysis of fibrosis and fat infiltration are reported in Table
Adhesion score, fibrosis, fat infiltration, and length of surgery.
Patient | Diagnosis | TLIF | Adhesion score | Fibrosis (mm) | % fat infiltration | Surgery time (min) |
---|---|---|---|---|---|---|
1 | Spondylolisthesis | L4-L5 | 4 | 2.5 | 10 |
|
2 | Stenosis | L4-L5 | 2 | 5 | 0 |
|
3 | Spondylolisthesis | L3-L5 | 2 | 2.5 | 0 |
|
4 | Stenosis | L3-L5 | 2 | 6 | 5 |
|
5 | Spondylosis | L4-S1 | 2 | 2.5 | 5 |
|
Adhesion score: 4 = sharp dissection required; 2 = easily detachable.
Histologic overview (H & E, original magnification ×100 or ×40).
Patient 1: Fibroadipose tissue, 10% fat (×100)
Patient 2: Fibrous tissue (×40)
Patient 3: Fibrous tissue (×40)
Patient 4: Fibroadipose tissue, 5% fat (×40)
Patient 5: Fibroadipose tissue, 5% fat (×40)
There were no intraoperative complications, specifically no durotomy/cerebrospinal fluid leaks.
Similarly, there were no postoperative complications, specifically no wound infections/dehiscence.
Individual patient outcomes are reported in Table
Patient outcomes.
Patient | Before TLIF | Before instrumentation removal | After instrumentation removal | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
ODI | Back pain | Leg pain | PCS | MCS | ODI | Back pain | Leg pain | PCS | MCS | ODI | Back pain | Leg pain | PCS | MCS | |
1 | 62 | 7 | 4 | 42.0 | 22.0 | 54 | 3 | 1 | 40.4 | 35.2 | 33 | 4 | 4 | 30.6 | 30.4 |
2 | 32 | 8 | 0 | 41.6 | 36.9 | 20 | 6 | 0 | 44.3 | 61.7 | 24 | 5 | 0 | 37.0 | 63.2 |
3 | 32 | 7 | 5 | 29.5 | 52.6 | 6 | 2 | 2 | 47.7 | 60.2 | 2 | 2 | 3 | 47.8 | 61.1 |
4 | 22 | 6 | 3 | 32.2 | 50.9 | 8 | 1 | 0 | 52.7 | 61.1 | 0 | 3 | 0 | 45.5 | 61.1 |
5 | 40 | 8 | 8 | 42.5 | 46.1 | 0 | 1 | 1 | 46.9 | 61.5 | 0 | 0 | 0 | 55.8 | 58.9 |
|
|||||||||||||||
Average | 37.6 | 7.2 | 4.0 | 37.6 | 41.7 | 17.6 | 2.6 | 0.8 | 46.4 | 55.9 | 14.4 | 2.8 | 1.4 | 45.2 | 54.9 |
Sample averages for the Oswestry Disability Index (ODI), the Physical Component Summary (PCS), and the Mental Component Summary (MCS) of the Medical Outcomes Study Questionnaire Short Form 36. ODI:
Sample averages for back pain and leg pain. Back pain:
While there is also an improvement in physical well-being (PCS), it is not statistically significant (Figure
Both back and leg pain decreased on average. The decrease was significant for back pain but not leg pain (Figure
Patients exhibited individual variation in their responses to treatment: half the patients reported minimal to no disability and pain while half the patients improved but experienced persistent pain and disability. Patient 2 underwent a sacroiliac joint fusion to further help with back pain. Patient 4 underwent an unrelated cervical fusion for cervical myelopathy.
Supplemental instrumentation is widely used in spinal fusion surgery. Biomechanical studies have demonstrated that supplemental instrumentation enhances the stability of the target motion segment [
This pilot study investigated the use of a dHACM barrier in reducing epidural fibrosis and facilitating dissection in revision spinal surgery. Although the small sample size of the study limits the ability to generalize our results, we believe the use of a dHACM barrier was clearly useful in limiting epidural fibrosis and promoting dissection in revision spinal surgery. Furthermore, the dHACM did not lead to any adverse events such as infection or spinal fluid leak.
Compared to other barriers that we have utilized, the dHACM performed better in revision surgery. In the past, we have used ADCON-L gel (Gliatech, Cleveland, OH, USA) but no longer utilize the product given the reports in the literature of associated radiculitis and spinal fluid leak. We have also used DuraSeal Xact (Covidien, Mansfield, MA) but contrary to Fransen study [
Although this study is too small to properly evaluate the effectiveness of dHACM in preventing epidural fibrosis and facilitating dissection in revision spinal surgery, it clearly shows promise in achieving these goals. These results and the impact of the dHACM barrier on patient outcomes need to be confirmed with larger studies with longer term follow-up, before more definitive conclusions may be drawn.
The authors declare that there is no conflict of interests regarding the publication of this paper.