Acute penetrating injury of the spinal cord by a wooden spike with delayed surgery: a case report

Abstract

Rarely, penetrating injuries to the spinal cord result from wooden objects, creating unique challenges to mitigate neurological injury and high rates of infection and foreign body reactions. We report a man who sustained a penetrating cervical spinal cord injury from a sharpened stick. While initially tetraparetic, he rapidly recovered function. The risks of neurological deterioration during surgical removal made the patient reluctant to consent to surgery despite the impalement of the spinal cord. A repeat MRI on day 3 showed an extension of edema indicating progressive inflammation. On the 7^th day after injury, fever and paresthesias occurred with a large increase in serum inflammatory indicators, and the patient agreed to undergo surgical removal of the wooden object. We discuss the management nuances related to wood, the longitudinal evolution of MRI findings, infection risk, surgical risk mitigation and technique, an inflammatory marker profile, long-term recovery, and the surprisingly minimal neurological deficits associated with low-velocity midline spinal cord injuries. The patient had an excellent clinical outcome. The main lessons are that a wooden penetrating central nervous system injury has a high risk for infection, and that surgical removal from the spinal cord should be performed soon after injury and under direct visualization.

Introduction

Most acute spinal cord injuries (SCI) are caused by high-energy trauma to the spine, usually contusive forces from displaced bone (Ahuja et al., 2017), but SCI also results from metallic bullets (Avila et al., 2021) and shrapnel (Lawless et al., 2018). Rarely, acute SCI is due to non-metallic penetrating objects such as wood (Pal et al., 2006) that may be retained (Imokawa et al., 2003). In these rare patients, there are substantial specific challenges around imaging (Gul et al., 2010), mitigation of the substantial infection risks (Zentner et al., 1991), and surgical removal of the foreign body. The main challenges, in this case, were to prevent further cord injury, avoid central nervous system and wound infection, and accomplish surgical removal safely. This case report illustrates challenges in imaging of penetrating wooden objects, the time course of inflammation and infection, surgical principles, and review the relevant literature.

Case Presentation

A 23-year-old man was fighting with friends using sharpened sticks late at night. A stick entered his posterior lower neck, snapping off, and he was immediately paralyzed in all limbs and fell to the ground. He was taken to Kunming Tongren hospital, where he underwent an emergency computed tomography (CT) scan and was noted to be rapidly recovering neurologically but complained of neck pain. The size of the foreign body was not immediately apparent on physical exam, as only a small penetration site was observed. The initial interpretation of the CT scan was subcutaneous air with no definite object (Figure 1). However, subsequent magnetic resonance imaging (MRI) showed a definite foreign body, presumptively a wooden spike (Figure 2, days 1 and 3) that impaled the spinal cord at C5/6 based on correlation to the history of injury.

Figure 1: A cervical CT scan obtained soon after injury.(A, B) A faint air attenuation signal is seen within the spinal canal at C5/6 extending into the anterior disc space (white arrows). (C) A small axial air attenuation signal is seen in the midline between the C5 and C6 spinous processes (white arrow). (D) A CT 3-D reconstruction appears normal and the wooden object is not visible. These images show that other than air signal, wood may have insufficient attenuation to be visible on a CT scan. CT: Computed tomography.

Figure 2: MRI panel displaying the initial injury, surgical removal, and follow-up.Sequential sagittal midline images. On Day 1, the wooden spike is clearly defined between the C5 and C6 spinous processes (white arrow) penetrating the spinal cord with a small region of associated edema. There is no evidence of intra- or extradural hemorrhage. By day 3, the extent of edema has expanded longitudinally (white arrows), and there is swelling of the spinal cord. Day 12 is 5 days post-surgery, and considerable dorsal spinal cord swelling is present. This may be appreciated by comparing the size of the smaller white triangle opposite C3 as compared to the larger triangle spanning the spinal cord diameter at C5. There is a small fluid pocket posterior to the laminectomy site (white asterisk). By 25 days post-injury, the spinal cord swelling has partially remitted and there is a region of probable necrosis (white arrow). By 3 months post-injury, a focal dorsal region of tissue loss is present at C5–C6 (black arrow) and the previous fluid collection is minimal. Marker = 5 cm. MRI: Magnetic resonance imaging.

The patient had blood drawn for lab assessments, including serial serum inflammatory markers, was started on prophylactic antibiotics, and was administered tetanus toxoid. Surgery was recommended, and the risks were explained in detail, including potential spinal damage and the possibility of spinal cord abscess and meningitis, especially if removal of the foreign body was not performed promptly. The patient and his family were reluctant to have him undergo surgery, given that he had no apparent neurological deficit. The patient’s case, suitably de-identified, was promptly discussed with members of the Kunming working group, an advisory panel of SCI experts, including clinicians such as neurosurgeons. The discussion focused on how to remove the object that was impaling the spinal cord in a manner to minimize the risk of infection and additional spinal cord trauma.

It was apparent that the following goals were of immediate importance: 1) Delicate removal of the spike without tearing spinal cord tissue and causing additional injury, 2) Complete removal of all the wooden material to prevent chronic granulation and infection.

The patient and his family continued to be reluctant to consent to surgery, however, on the 6^th day after injury, the patient developed a fever and began to feel dysesthesia and numbness in his legs. Being additionally concerned by these new symptoms, he provided informed consent for the surgery, including the use of his data and images for possible education and publication. Throughout this period, great care was taken with the movement of the patient’s neck and head to avoid spinal cord tension or distortion since the wooden spike had impaled the spinal cord, and there was a residual spike in the neck soft tissues.

Formal ethics approval for the project was waived after consulting the local research ethics board at Kunming Tongren Hospital before undertaking this case report; however, written informed consent was obtained from the patient presented in the case report.

History and physical examination

The patient had no significant prior medical or surgical history. Vital signs, including temperature, were serially documented. The stick entry site was carefully assessed for evidence of retained debris or the development of infection.

Neurological examinations

To establish the baseline neurological status, the patient was examined using the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) (Rupp et al., 2021) as well as specifically directed examinations related to cervical dorsal midline injury such as heel-shin tracking and perception of joint movements for proprioception. The ISNSCI exam was documented on days 1, 5, and 21 after injury, and his functional abilities were assessed with periodic remote video follow-up after discharge.

Imaging

As described, the initial radiological study was with CT scan followed by serial MRIs. These images defined the location and extent of injury in the spinal cord as well the orientation of the wooden spike with respect to vertebral structures.

Laboratory tests

The patient’s blood and cerebrospinal fluid (CSF) were serially assessed for evidence of an acute inflammatory response, and appropriate cultures were obtained.

Antibiotics

To reduce the risk of a serious infection, the patient received the following antibiotics, cefmenoxime IV, during the 6 days prior to surgery. On the day of surgery, ceftriaxone, vancomycin, and ornidazole, then ceftriaxone and ornidazole for an additional days, and then ceftriaxone until the 18^th day after injury.

Surgery

The primary goals of surgery were the removal of all foreign body material and minimal manipulation and injury to the spinal cord. Caution regarding neck movement continued during preparation for surgery, induction of anesthesia and intubation, and prone positioning. During the subcutaneous and muscle dissection, small wood fragments were encountered and removed. The wooden stick had penetrated between the spinous processes of C5 and C6, and there were small fractures of the C5 and C6 spinous processes. The wooden stick was isolated from surrounding bone contact by bone removal. Prior to dural opening, the wound cavity was irrigated with diluted hydrogen peroxide and flushed with normal saline Vancomycin solution. Care was taken during the dural exposure to minimize the movement of the wood fragment (Figure 3A and B). The subdural portion of the wood was found to be 2.5 cm in length, with a width of 0.3 cm at the distal tip and 0.7 cm at the dural entry. The total length of the removed portion was 6 cm (Figure 3D).

Figure 3: Intraoperative and gross images of the wooden spike.(A) The posterior tissues of the neck have been dissected away from the wooden spike, with the dura open, showing the midline position of the spike. (B) After the removal of the spike some maceration in the dorsal spinal cord midline is evident (black arrow). (C) Pre-operative skin level position of the spike (black arrow) after hair removal prior to initial positioning for surgery. (D) The full wooden extracted spike is shown next to a black malleable rod with one-centimeter markings for reference (1 cm = space between black arrows).

Drilling of the wooden spike as well as morcelization was avoided to minimize creating foreign body wood dust and fragments that could constitute residual foreign bodies. This is different from the approach to e.g., diastematomyelia, a congenital bone spike that bisects the spinal cord, and is an important surgical issue in this case.

The anterior spinal cord could not be visualized at surgery, and thus it was not possible to determine if any residue of the stick remained, thus the patient was continued on antibiotics for a total of 18 days.

Rehabilitation

Following surgery, the patient was mobilized and commenced rehabilitation, including the Kunming locomotion training program (Liu et al., 2021).

Results

On initial physical exam, the patient had a small entry wound at the lower hairline and slightly to the left side, where a 3–4 mm area of wood could be seen (Figure 3C).

Neurological examinations

Although initially tetraparetic, the patient recovered quickly and was able to control his bladder by the third post-injury day. The only abnormality detected on the sequential ISNCSCI exams was 4/5 weakness of ankle plantar flexors on day 1. This had resolved by the exam on day 5 post-injury.

Imaging studies

The initial CT and the MRI exams in sequence are shown in Figures 1 and 2. MRI was superior to CT in defining the wooden spike penetrating through to the anterior spinal cord. There was no evidence of penetration of the intervertebral discs. At no point was signal change detected in the disc spaces.

Laboratory values

In the perioperative period 6–8 days post-injury, there was a notable spike in white blood cells, C-reactive protein (CRP), and interleukin-6 (IL-6) and a very high transient neutrophil-to-lymphocyte ratio (NLR; Figure 4). Values had decreased, approaching the normal range on days 10–11 post-injury, when another spike occurred.

Figure 4: Selected laboratory values by day-post-injury.The time frame is shown by the shaded boxes, lightest- pre-op, intermediate-peri-op, and darkest-post-op. The pre- and post-operative boxes show the normal range for the values. The patient developed a low-grade fever and paresthesias in his legs on day 6 after injury and then consented to surgery. Note that an elevation of interleukin-6 (IL-6) precedes the elevations in C-reactive protein (CRP) that are observed to have 3 spikes. The onset of fever corresponds to a very high neutrophil-to-lymphocyte ratio (NLR) ratio. WBC: White blood cell.

CSF cultures

CSF was gathered by a lumbar puncture on four occasions. The CSF culture on day 9, 2 days after surgery, was positive for Staphylococcus haemolyticus. On days 20 and 26, the report was negative for growth. On day 31, a culture was positive for Staphylococcus epidermidis considered a skin contaminant.

Outcome and follow-up

The patient’s neurological recovery and ability to return to work and to life with his family were monitored during follow-up through systematic in-person and remote exams. At 1.5 years post-injury, the patient’s function was examined by video recording, including his ability to stand still with eyes closed, heel-toe step on a balance beam, and to climb and descend stairs. No clinically detectable deficits were observed.

Discussion

Penetrating injuries caused by wood

Non-metallic penetrating injuries are a relatively uncommon cause of SCI. Because they occur at a low velocity relative to bullets and shrapnel, they transmit less force into the spinal cord. Wooden object penetration has been reported in several ocular and orbital injuries, but SCI is rare. This is because the posterior tissues that protect the spinal cord include large muscles, the bones of the lamina and spinous processes, and the ligamentum flavum. Thus, a rare constellation of events is necessary for a wooden object to penetrate the spinal cord.

Literature review

A thorough literature search on spinal cord impalement by wood was conducted. A report by Santangelo et al. (2018) stated there were only 6 prior reports of wooden objects causing SCI in the literature. The Web of Science, National Library of Medicine, and Scopus were reviewed using the abstract search terms (wood AND spinal), which yielded 30 returns, of which four were specifically relevant (Rayward, 2002; Gul et al., 2010; Fattahi et al., 2018; Santangelo et al., 2018). For the Scopus search, the string TITLE-ABS-KEY(“wood”) AND TITLE-ABS-KEY(“spinal”) yielded 107 document returns of which an additional five were relevant (Lunawat and Taneja, 2000; Liliang et al., 2001; Pal et al., 2006; Yamaguchi et al., 2007; Yamasaki et al., 2013). A review of the article bibliographies yielded an additional five reports (Peacock et al., 1977; Criado et al., 1990; Pyhtinen et al., 1995; Tomaras et al., 1997; Arunkumar et al., 1999). Thus, a total of 13 articles were identified in the accessible medical literature that specifically report SCI by wooden objects.

Other injuries have relevance, such as intraorbital penetrations (Zentner et al., 1991; Yamasaki et al., 2013), and there is a larger literature on knife and other metallic object spinal cord penetrations. The greatest reported experience of CNS penetration by wooden objects is from the eye (Pyhtinen et al., 1995; Arunkumar et al., 1999).

There are several case reports of minimal to no apparent clinical SCI after knife wounds of the spinal cord. Castillo and Kahn reported that a broken knife blade was retained, transfixing the thoracic spinal cord for 23 years before the onset of progressive symptoms (Castillo and Kahn, 1950). In separate reports, Jones and Wolf reported similar cases with brief transient paresthesia following by a long asymptomatic period before onset of increasing paralysis (Jones, 1943; Wolf, 1973). Sarkar et al. (2018) reported a thoracic midline nail gun injury with no significant neurological deficits and suggested the term –”Midline Cord Syndrome”. Similar to our case they reported no apparent deficits on the ISNCSCI exam despite a large midline nail penetrating the spinal cord visible on both X-Ray and CT. The length of the nail was almost identical to the wood spike reported here. Cord bisection was confirmed on post-operative MRI. Thus, there is a precedent that some penetrating injuries may initially have minimal symptoms.

Biology of wood as a foreign body

An understanding of the properties of wood as a material is useful in considering this case and others. Wood is a porous, fibrous polysaccharide polymer of cellulose and hemicellulose fibers embedded in a lignin matrix and contains several ion species and “extractives” such as terpenes, phenols, and aliphatic compounds. Extractives confer resistance to fungi and insects (Kirker et al., 2013). Normally, wood contains some water and has evolved mechanisms to avoid environmental degradation.

The body’s reaction to wood is apparent to most people that have had difficulty removing a splinter. As a retained foreign body, chronic abscesses, and draining fistulas have been reported. The lignin components of wood have free-radical scavenging activity (Dizhbite et al., 2004), and polyphenols and tannins have antibacterial properties (Aviat et al., 2016). Although wood may have some antibacterial properties when dry, macerated wood is a porous foreign body. Contact with soil before penetration by wood creates a risk for clostridial and fungus infection (Meyer and Hood, 1977). Cellulose polymer exhibits slow biodegradation and becomes thinly encapsulated. Macrophages and foreign body giant cells are typical. Both pro-inflammatory (IL-6, tumor necrosis factor-alpha) and ani-inflammatory cytokine IL-10 secretion have been reported (Rashad et al., 2019) near retained wood.

A role for antibiotic prophylaxis

Wood objects are likely to be contaminated and lead to infection. On the day of surgery, the patient had a fever, and CSF obtained from surgery was positive for Staphylococcus haemolyticus, a coagulase-negative pathogen that can form biofilms (Czekaj et al., 2015) and infect foreign bodies (Hitzenbichler et al., 2017). Retained wood objects may present with delayed infection, and discs are at higher infection risk due to their limited blood supply.

Imaging of wooden objects

CT scans may fail to detect some spinal injuries (Kolli et al., 2010). Several reports indicate difficulties with the visualization of wooden foreign bodies using imaging techniques (Sawar et al., 1993). On CT scans, wooden objects show low attenuation and may be initially confused with imaging artifacts or not discerned (Imokawa et al., 2003). X-rays may fail to detect wooden foreign bodies (Zentner et al., 1991), although dry wood usually contains air, as seen in this case. Sonography is generally effective (Crawford and Matheson, 1989). Thus, after accidents, retention of penetrating wooden foreign bodies may escape detection.

Surgical approach

Whilst one could argue for a less invasive surgical approach, such as simply pulling the stick out from a subcutaneous exposure that approach would leave the surgeon blind to possible damage to spinal cord tissue, related bleeding, an unclosed dural opening, and residual wood fragments. Thus, such as approach would risk serious added spinal cord injury. In surgery for diastematomyelia, one can carefully reduce the bone spur by drilling with a fine-grained bit (Vissarionov et al., 2018); however, in this case, drilling down the wooden spike would have created distributed material that could cause a foreign body response. Spinal fixation was performed prophylactically using lateral mass screws between C4–C6.

Laboratory values

Notable changes included a high perioperative NLR, with an associated peak in CRP and IL-6. Elevated NLR has been reported in bacterial meningitis (Naess et al., 2017) and perioperative stress (Josse et al., 2016). Post-surgically inflammatory biomarkers returned to near normal levels except for a transient increase two weeks post-injury, the cause of which is unclear (Figure 4). IL-6 is known to induce CRP production by the liver in the presence of infection (Buck et al., 1994).

Sensitivity of the ISNCSCI exam in midline injuries

Despite an evident SCI, the patient’s ISNCSCI exams were reported as normal. Thus, a substantial injury was present with an apparently normal exam. For contusive injuries, recovery to AIS E status has been reported for injuries with edema only and not hemorrhage (Miyanji et al., 2007). Martineau et al. (2019) reported that patients with traumatic SCI that converted to AIS E (“normal”) had up to 25 mm of initial signal change. Thus, the literature confirms that some degree of SCI is compatible with recovery that appears normal when the ISNCSCI exam is performed. Examination of the dorsal column function by ISNCSCI is based on an ordinal light touch score without a test of proprioception. Thus, the exam may not be especially sensitive to midline injury.

Learning points

• Midline penetrating spinal cord injuries may present with minimal neurological symptoms.
• The severity of these injuries may be underestimated by standard neurological exams.
• Surgery removal of retained wooden objects is necessary to reduce infection risk.

Conclusion

Penetrating injuries of the spinal cord with wooden objects are rare. The associated risk of infection is high. Meticulous surgery, debridement, and infection protection are necessary.

Author contributions:Conceptualization, writing original draft, final draft, editorial revisions: JDG. Investigation, data curation, manuscript preparation, clinical supervision: HZ. Investigation, clinical guidance: ZL. Clinical reports, supervision: YL. Writing original draft, data curation: XMX. All authors approved the final version of the manuscript.

Conflicts of interest:None declared.

Data availability statement:Not applicable.

C-Editors: Zhao M, Liu WJ; S-Editor: Li CH; L-Editors: Li CH, Song LP; T-Editor: Jia Y