Vertebral osteolytic defect due to cellulose particles derived from gauze fibers after posterior lumbar interbody fusion

Case report

Full access

Vertebral cystic lesions may be observed in pseudarthroses after lumbar fusion surgery. The authors report a rare case of pseudarthrosis after spinal fusion, accompanied by an expanding vertebral osteolytic defect induced by cellulose particles. A male patient originally presented at the age of 69 years with leg and low-back pain caused by a lumbar isthmic spondylolisthesis. He underwent a posterior lumbar interbody fusion, and his neurological symptoms and pain resolved within a year but recurred 14 months after surgery. Radiological imaging demonstrated a cystic lesion on the inferior endplate of L-5 and the superior endplate of S-1, which rapidly enlarged into a vertebral osteolytic defect. The patient underwent revision surgery, and his low-back pain resolved. A histopathological examination demonstrated foreign body–type multinucleated giant cells, containing 10-μm particles, in the sample collected just below the defect. Micro–Fourier transform infrared spectroscopy revealed that the foreign particles were cellulosic, presumably originating from cotton gauze fibers that had contaminated the interbody cages used during the initial surgery. Vertebral osteolytic defects that occur after interbody fusion are generally presumed to be the result of infection. This case suggests that some instances of vertebral osteolytic defects may be aseptically induced by foreign particles. Hence, this possibility should be carefully considered in such cases, to help prevent contamination of the morselized bone used for autologous grafts by foreign materials, such as gauze fibers.

Abbreviations used in this paper:CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; FTIR = Fourier transform infrared; JOA = Japanese Orthopaedic Association; PEEK = polyetheretherketone; PLDLLA = poly-l-lactide-co-d, l-lactide; PLIF = posterior lumbar interbody fusion; VAS = visual analog scale; VAS-BP = VAS for back pain; VAS-LP = VAS for leg pain.

Vertebral cystic lesions may be observed in pseudarthroses after lumbar fusion surgery. The authors report a rare case of pseudarthrosis after spinal fusion, accompanied by an expanding vertebral osteolytic defect induced by cellulose particles. A male patient originally presented at the age of 69 years with leg and low-back pain caused by a lumbar isthmic spondylolisthesis. He underwent a posterior lumbar interbody fusion, and his neurological symptoms and pain resolved within a year but recurred 14 months after surgery. Radiological imaging demonstrated a cystic lesion on the inferior endplate of L-5 and the superior endplate of S-1, which rapidly enlarged into a vertebral osteolytic defect. The patient underwent revision surgery, and his low-back pain resolved. A histopathological examination demonstrated foreign body–type multinucleated giant cells, containing 10-μm particles, in the sample collected just below the defect. Micro–Fourier transform infrared spectroscopy revealed that the foreign particles were cellulosic, presumably originating from cotton gauze fibers that had contaminated the interbody cages used during the initial surgery. Vertebral osteolytic defects that occur after interbody fusion are generally presumed to be the result of infection. This case suggests that some instances of vertebral osteolytic defects may be aseptically induced by foreign particles. Hence, this possibility should be carefully considered in such cases, to help prevent contamination of the morselized bone used for autologous grafts by foreign materials, such as gauze fibers.

Posterior lumbar interbody fusion (PLIF), involving the use of cages and pedicle screw instrumentation, is an established procedure for treating degenerative or isthmic lumbar spondylolisthesis.1 However, postoperative pseudarthrosis is a potential side effect of PLIF.3 Fujibayashi et al. indicated that vertebral cyst formation, during the early postoperative course of interbody fusion, can be a predictor of pseudarthrosis.5 These authors assumed that the cysts resulted from local mechanical stress. The cyst presented in the current case was dissimilar to a stress-induced cyst. The rapid ballooning of this vertebral cyst can be described as a vertebral osteolytic defect, with an unusual etiology. In the present report, we describe a case of pseudarthrosis after spinal fusion, accompanied by an expanding vertebral osteolytic defect induced by cellulose particles.

Case Report

Presentation and Examination

In 2009, at the age of 69 years, this male patient began to experience leg and low-back pain while walking. He described his symptoms as pain radiating down his left lower leg, worse in the foot, along with associated numbness and tingling but no weakness. He visited a local doctor who diagnosed a Grade 1 isthmic spondylolisthesis at L-5, based on radiographic findings, and referred him to our hospital in October 2011. His pain, neurological function, and functional abilities were assessed using visual analog scales (0, no pain; 100, worst pain) for back pain (VAS-BP) and leg pain (VASLP) and the Japanese Orthopaedic Association (JOA) scoring system (maximum score 29 points).8 His preoperative VAS-BP, VAS-LP, and JOA scores were 80, 50, and 16, respectively.

Initial Operation

In November 2011, the patient underwent a PLIF, using a pedicle screw placement system, at L5–S1, with lumbarization of the S-1 vertebra. Decompression was achieved by removing the loose posterior element and the fibrocartilaginous callus at the foramen. After meticulous removal of the cartilage and soft tissue from the excised posterior element, morselized bone and a piece of block bone were prepared for autologous bone grafting. Two Capstone, polyetheretherketone (PEEK) cages (Medtronic Sofamor Danek) were packed with the morselized bone. After complete removal of the intervertebral disc material and cartilaginous endplates, residual morselized bone was inserted into the anterior portions of the interbody space. The cages were then inserted into the intervertebral spaces, with a strut bone block placed between the cages (Fig. 1).

Fig. 1.
Fig. 1.

Plain anteroposterior and lateral radiographs obtained immediately after the initial surgery demonstrating good positioning of the interbody cages and pedicle screws.

Postoperative Course

The procedure and postoperative recovery were uneventful. The patient demonstrated postoperative improvement and was able to walk, without remarkable low-back or leg pain, within the first postoperative year; his VAS-BP, VAS-LP, and JOA scores at this point were 30, 20, and 27, respectively. However, fusion was not achieved, as noted on CT images (Fig. 2). Fourteen months after surgery, the patient began to complain of recurrent, severe, low-back pain while walking. His condition gradually worsened, and analgesics had no effect on the symptoms; his VAS-BP, VAS-LP, and JOA scores at this point were 90, 90, and 16, respectively. Magnetic resonance imaging of the lumbar spine revealed cystic lesions at the inferior endplate of L-5 and at the superior endplate of S-1. We suspected either bone cyst formation, caused by mechanical stress due to nonunion,5 or the presence of a low-grade infection. However, his white-cell, eosinophil, and basophil counts; C-reactive protein (CRP) level; erythrocyte sedimentation rate (ESR); and body temperature were all within the normal limits. A biopsy was impossible because of the inaccessibility of the cystic lesions.

Fig. 2.
Fig. 2.

Coronal (A–F) and sagittal (G–L) CT reconstructions and sagittal MR images (M and N) showing sequential changes in the vertebral endplate and the cystic lesions. At both 12 and 16 months after the operation, cystic lesions are evident on the inferior endplate of L-5 and the superior endplate of S-1. The clear zone around the pedicle screws is obvious, and fusion was not achieved. At 19 months postoperatively, the lesion is dramatically larger, without sclerotic changes around the vertebral osteolytic defect (arrows, D). Six months after the revision surgery, the defects are decreasing in size and bone regeneration is occurring. d = days; mo = months; p.o. = postoperative.

Sixteen months after surgery, additional surgery to treat the pseudarthrosis seemed unnecessary due to the low probability of infection. However, between postoperative months 16 and 19, follow-up CT scans demonstrated rapid ballooning of the cystic lesions, without marginal sclerosis. This change was found to be a vertebral osteolytic defect that rapidly increased in size from 8 × 7 × 10 mm to 20 × 14 × 17 mm (Fig. 2) and almost reached the superior endplate of L-5. Although a firm explanation of this rapid expansion of the cystic lesions, mimicking osteolysis, was unavailable, we believed that surgical intervention was necessary to prevent additional bone loss.

Second and Third Operations

A revision of the L5–S1 interbody fusion and surgical exploration of the vertebral osteolytic defect was performed 21 months after the initial surgery (September 2013). The pedicle screws were loose and the pedicle screw–rod constructs were easily removed. We could not retrieve the interbody PEEK cages because of severe adhesions around the thecal sac; however, there were no gross signs of infection.

One week after the second operation, the patient underwent an anterior lumbar interbody fusion with an iliac crest autograft. The PEEK cages were successfully removed through an anterior transperitoneal approach; gross purulence was not noted during the third surgery.

Final Postoperative Course

Postoperatively, the patient's recovery was uneventful, and the size of the vertebral osteolytic defect was reduced at his 6-month postoperative follow-up (Fig. 2). He was also able to walk without marked low-back or leg pain at that time. His VAS-BP, VAS-LP, and JOA scores at this point were 20, 20, and 27, respectively.

Histopathological Examination

During the third surgery, we obtained 3 histological samples—one from the packed bone in the right cage, another from the left cage, and the third from the strut bone block between the 2 cages. These samples contained collagenous connective tissue as well as bone fragments, consistent with a diagnosis of noninfectious pseudarthrosis. Foreign body–type, multinucleated giant cells containing approximately 10-μm particles (approximately 6 particles/mm2) were observed in the sample of packed bone from the right cage. This sample was obtained just below the vertebral osteolytic defect (Fig. 3); the other samples did not contain such particles. The samples were all microbiologically sterile.

Fig. 3.
Fig. 3.

Photomicrograph of a section of the sample obtained from just below the vertebral osteolytic defect. Note the presence of giant cells and phagocytized particles (arrows). H & E.

Identification of the Foreign Particles

The foreign particles in the multinucleated giant cells were examined with a micro–Fourier transform-infrared (FTIR) spectrophotometer (IRμs Molecular Microanalysis System, Thermo Fisher Scientific).7 The system consists of a rapid scan FTIR spectrometer and infrared microscope integrated into one instrument. FTIR spectroscopy revealed the foreign particles had a similar spectral pattern to that of natural cellulose. The spectrum of the particles was obviously different from those of the PEEK cages and human hair proteins (Fig. 4). An examination of a sample of cotton gauze revealed an almost identical spectrum as observed in the recovered sample (Fig. 4). Based on the results of these tests, the cellulose particles were presumed to have originated from the cotton gauze fibers used during surgery.

Fig. 4.
Fig. 4.

Infrared spectra of the foreign particles and other samples. The instrument resolution was set at 8 cm−1, and the spectra displayed were obtained using 256–2048 co-added scans. A: Sample obtained just below the vertebral osteolytic defect. B: Gauze routinely used in our institution. C: Polyetheretherketone cage. D: Protein obtained from human hair.

Discussion

Vertebral osteolytic lesions that occur after PLIF of-ten result from surgical-site infections. However, in our patient, blood samples failed to reveal any infectious agent; the white-cell counts, CRP levels, and ESR were within normal limits; gross purulence was not observed during the reoperations; and the histological and microbiological assessments did not indicate signs of infection. Thus, an aseptic mechanism appears plausible. Moreover, the possibility of an allergic reaction was unlikely, considering that this patient's eosinophil and basophil counts were within normal limits.

Fujibayashi et al. showed that vertebral cysts, formed in cases where a titanium cage is present, could predict pseudarthrosis after lumbar fusion surgery.5 They assumed that mechanical stress caused these vertebral cysts, analogous to the cysts observed in knee or hip osteoarthritis, where stress-induced microfractures and subsequent bone resorption are believed to create cysts.2 However, the absence of sclerotic changes around the defect and the dramatic speed of cyst expansion observed in the present patient did not support the hypothesis that mechanical stress was the cause of the vertebral lesions.

Aseptic vertebral osteolysis has been previously reported. Frost et al. reported that 4 out of 9 patients (44%) developed particle-induced osteolysis around the bioabsorbable poly-l-lactide-co-d, l-lactide (PLDLLA) interbody cage at 11–16 months after PLIF (Table 1).4 The authors suggested that the PLDLLA fragments induced osteolysis. Jiya et al. also found that 2 (17%) of 12 patients undergoing PLIF with PLDLLA cages developed particleinduced osteolysis approximately 1 year postoperatively (Table 1).9 We believe that these cases were similar to our case in terms of the absence of sclerotic changes around the cystic lesion, the rapid growth of osteolytic lesions approximately 1 year after surgery, and the 10-μm size of particles phagocytized by the multinucleated giant cells.13 Particles, ≤ 10 μm in size, can cause particle-induced osteolysis, which is commonly noted in cases of periprosthetic osteolysis associated with total joint arthroplasty.11 Kubo et al. showed that 11-μm polyethylene particles were more biologically active than larger particles. Their results also indicated that various other types of particles, in addition to polyethylene, could induce osteolysis;10 wear particles can be phagocytized and activate the macrophages to produce proinflammatory cytokines, chemokines, reactive oxygen species, and other mediators.6 Punt et al. reported that a wide range of particle numbers (from 1 to 1002 in 10 total hip arthroplasty patients) would elicit macrophage activation in patients requiring revision surgery, primarily for aseptic loosening;12 we observed 6 cellulose particles/mm2 in our patient with induced osteolysis.

TABLE 1:

Previous publications reporting on vertebral osteolysis associated with use of a PLDLLA interbody cage*

Authors & YearNo. of PtsNo. of Pts w/ OsteolysisOnset (mos)Possible Etiology
Frost et al., 201294 (44%)11, 15, 16, & 16PLDLLA fragments
Jiya et al., 2009122 (17%)12 & 12PLDLLA fragments, low-grade infection

The inserted interbody cages were Telamon PLDLLA Hydrosorb (Medtronic Sofamor Danek). Pts = patients.

We believe that the development of the vertebral osteolytic defect in our patient transpired in the following manner. First, postsurgical pedicle screw loosening and pseudarthrosis allowed micromotion between the vertebrae. The resultant intervertebral “grinding” process produced small particles from the cotton gauze fibers contaminating the morselized bone. When the particles were 10 μm or smaller, they were easily phagocytized by the macrophages that were thereby activated, resulting in the vertebral osteolytic defect. This may explain not only the delayed formation, but also the rapid expansion of the vertebral osteolytic defect. If the fusion had been successfully achieved within the first year after PLIF, the vertebral osteolytic defect might not have occurred.

Thus, although vertebral osteolytic defects are generally considered the results of infection, they may also develop aseptically. When preparing the bone graft, using bone cutting tools, we used cotton gauze to avoid slipping of the graft bone. We believe the contaminating gauze fibers were introduced during this step. Based on the present case, preventing the introduction of contaminating foreign materials, such as gauze fibers, into the morselized bone used for autologous grafts is important.

Disclosure

Dr. Fuji reports a consultant relationship with Daiichi-Sankyo and receipt of royalties from Showa-Ika-Kogyo and Century Medical.

Author contributions to the study and manuscript preparation include the following. Conception and design: Takenaka. Acquisition of data: Takenaka, Tateishi. Analysis and interpretation of data: Takenaka. Drafting the article: Takenaka, Critically revising the article: Mukai, Hosono. Reviewed submitted version of manuscript: Mukai, Hosono. Study supervision: Fuji.

This article contains some figures that are displayed in color online but in black-and-white in the print edition.

References

  • 1

    Brantigan JWSteffee ADLewis MLQuinn LMPersenaire JM: Lumbar interbody fusion using the Brantigan I/F cage for posterior lumbar interbody fusion and the variable pedicle screw placement system: two-year results from a Food and Drug Administration investigational device exemption clinical trial. Spine (Phila Pa 1976) 25:143714462000

    • Search Google Scholar
    • Export Citation
  • 2

    Dürr HDMartin HPellengahr CSchlemmer MMaier MJansson V: The cause of subchondral bone cysts in osteoarthrosis: a finite element analysis. Acta Orthop Scand 75:5545582004

    • Search Google Scholar
    • Export Citation
  • 3

    Etminan MGirardi FPKhan SNCammisa FP Jr: Revision strategies for lumbar pseudarthrosis. Orthop Clin North Am 33:3813922002

  • 4

    Frost ABagouri EBrown MJasani V: Osteolysis following resorbable poly-L-lactide-co-D, L-lactide PLIF cage use: a review of cases. Eur Spine J 21:4494542012

    • Search Google Scholar
    • Export Citation
  • 5

    Fujibayashi STakemoto MIzeki MTakahashi YNakayama TNeo M: Does the formation of vertebral endplate cysts predict nonunion after lumbar interbody fusion?. Spine (Phila Pa 1976) 37:E1197E12022012

    • Search Google Scholar
    • Export Citation
  • 6

    Goodman SBGibon EYao Z: The basic science of periprosthetic osteolysis. Instr Course Lect 62:2012062013

  • 7

    Griffiths PRDe Haseth JA: Introduction to vibrational spectroscopy. Fourier Transform Infrared Spectrometry ed 2Hoboken, NJJohn Wiley & Sons2007. 118

    • Search Google Scholar
    • Export Citation
  • 8

    Izumida SInoue S: [Assessment of treatment for low back pain]. J Jpn Orthop Assoc 60:3913941986. (Jpn)

  • 9

    Jiya TSmit TDeddens JMullender M: Posterior lumbar interbody fusion using nonresorbable poly-ether-ether-ketone versus resorbable poly-L-lactide-co-D,L-lactide fusion devices: a prospective, randomized study to assess fusion and clinical outcome. Spine (Phila Pa 1976) 34:2332372009

    • Search Google Scholar
    • Export Citation
  • 10

    Kubo TSawada KHirakawa KShimizu CTakamatsu THirasawa Y: Histiocyte reaction in rabbit femurs to UHMWPE, metal, and ceramic particles in different sizes. J Biomed Mater Res 45:3633691999

    • Search Google Scholar
    • Export Citation
  • 11

    Ollivere BWimhurst JAClark IMDonell ST: Current concepts in osteolysis. J Bone Joint Surg Br 94:10152012

  • 12

    Punt IMAusten SCleutjens JPKurtz SMten Broeke RHvan Rhijn LW: Are periprosthetic tissue reactions observed after revision of total disc replacement comparable to the reactions observed after total hip or knee revision surgery?. Spine (Phila Pa 1976) 37:1501592012

    • Search Google Scholar
    • Export Citation
  • 13

    Shanbhag ASJacobs JJBlack JGalante JOGlant TT: Macrophage/particle interactions: effect of size, composition and surface area. J Biomed Mater Res 28:81901994

    • Search Google Scholar
    • Export Citation

If the inline PDF is not rendering correctly, you can download the PDF file here.

Article Information

Address correspondence to: Shota Takenaka, M.D., Orthopaedic Surgery, Japan Community Health Care Organization Osaka Hospital, 4-2-78 Fukushima, Osaka 553-0003, Japan. email: show@yb3.so-net.ne.jp.

Please include this information when citing this paper: published online September 26, 2014; DOI: 10.3171/2014.8.SPINE14196.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Plain anteroposterior and lateral radiographs obtained immediately after the initial surgery demonstrating good positioning of the interbody cages and pedicle screws.

  • View in gallery

    Coronal (A–F) and sagittal (G–L) CT reconstructions and sagittal MR images (M and N) showing sequential changes in the vertebral endplate and the cystic lesions. At both 12 and 16 months after the operation, cystic lesions are evident on the inferior endplate of L-5 and the superior endplate of S-1. The clear zone around the pedicle screws is obvious, and fusion was not achieved. At 19 months postoperatively, the lesion is dramatically larger, without sclerotic changes around the vertebral osteolytic defect (arrows, D). Six months after the revision surgery, the defects are decreasing in size and bone regeneration is occurring. d = days; mo = months; p.o. = postoperative.

  • View in gallery

    Photomicrograph of a section of the sample obtained from just below the vertebral osteolytic defect. Note the presence of giant cells and phagocytized particles (arrows). H & E.

  • View in gallery

    Infrared spectra of the foreign particles and other samples. The instrument resolution was set at 8 cm−1, and the spectra displayed were obtained using 256–2048 co-added scans. A: Sample obtained just below the vertebral osteolytic defect. B: Gauze routinely used in our institution. C: Polyetheretherketone cage. D: Protein obtained from human hair.

References

  • 1

    Brantigan JWSteffee ADLewis MLQuinn LMPersenaire JM: Lumbar interbody fusion using the Brantigan I/F cage for posterior lumbar interbody fusion and the variable pedicle screw placement system: two-year results from a Food and Drug Administration investigational device exemption clinical trial. Spine (Phila Pa 1976) 25:143714462000

    • Search Google Scholar
    • Export Citation
  • 2

    Dürr HDMartin HPellengahr CSchlemmer MMaier MJansson V: The cause of subchondral bone cysts in osteoarthrosis: a finite element analysis. Acta Orthop Scand 75:5545582004

    • Search Google Scholar
    • Export Citation
  • 3

    Etminan MGirardi FPKhan SNCammisa FP Jr: Revision strategies for lumbar pseudarthrosis. Orthop Clin North Am 33:3813922002

  • 4

    Frost ABagouri EBrown MJasani V: Osteolysis following resorbable poly-L-lactide-co-D, L-lactide PLIF cage use: a review of cases. Eur Spine J 21:4494542012

    • Search Google Scholar
    • Export Citation
  • 5

    Fujibayashi STakemoto MIzeki MTakahashi YNakayama TNeo M: Does the formation of vertebral endplate cysts predict nonunion after lumbar interbody fusion?. Spine (Phila Pa 1976) 37:E1197E12022012

    • Search Google Scholar
    • Export Citation
  • 6

    Goodman SBGibon EYao Z: The basic science of periprosthetic osteolysis. Instr Course Lect 62:2012062013

  • 7

    Griffiths PRDe Haseth JA: Introduction to vibrational spectroscopy. Fourier Transform Infrared Spectrometry ed 2Hoboken, NJJohn Wiley & Sons2007. 118

    • Search Google Scholar
    • Export Citation
  • 8

    Izumida SInoue S: [Assessment of treatment for low back pain]. J Jpn Orthop Assoc 60:3913941986. (Jpn)

  • 9

    Jiya TSmit TDeddens JMullender M: Posterior lumbar interbody fusion using nonresorbable poly-ether-ether-ketone versus resorbable poly-L-lactide-co-D,L-lactide fusion devices: a prospective, randomized study to assess fusion and clinical outcome. Spine (Phila Pa 1976) 34:2332372009

    • Search Google Scholar
    • Export Citation
  • 10

    Kubo TSawada KHirakawa KShimizu CTakamatsu THirasawa Y: Histiocyte reaction in rabbit femurs to UHMWPE, metal, and ceramic particles in different sizes. J Biomed Mater Res 45:3633691999

    • Search Google Scholar
    • Export Citation
  • 11

    Ollivere BWimhurst JAClark IMDonell ST: Current concepts in osteolysis. J Bone Joint Surg Br 94:10152012

  • 12

    Punt IMAusten SCleutjens JPKurtz SMten Broeke RHvan Rhijn LW: Are periprosthetic tissue reactions observed after revision of total disc replacement comparable to the reactions observed after total hip or knee revision surgery?. Spine (Phila Pa 1976) 37:1501592012

    • Search Google Scholar
    • Export Citation
  • 13

    Shanbhag ASJacobs JJBlack JGalante JOGlant TT: Macrophage/particle interactions: effect of size, composition and surface area. J Biomed Mater Res 28:81901994

    • Search Google Scholar
    • Export Citation

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 207 207 16
PDF Downloads 163 163 7
EPUB Downloads 0 0 0

PubMed

Google Scholar