Imaging of the postoperative spine

Abstract

Background

There has been a significant increase in the number of spinal surgical procedures performed over the last few decades, resulting in a proportionate increase in the number of postoperative imaging studies.

Methods

An exhaustive literature search was performed along with consideration of various guidelines and expert opinions regarding postoperative spine imaging. Complications are divided into early (in the first few weeks) and delayed, depending on the time of onset. Some complications occurring in the early postoperative period are common for both the instrumented and non-instrumented spine. Delayed complications are specific to the type of surgery performed and have been described as such. This review discusses the normal postoperative appearance and the various early and delayed complications.

Conclusion

An understanding of the normal postoperative appearance is pertinent to distinguish normal from abnormal. A plain radiograph is the primary imaging modality for immediate postoperative assessment and long-term follow-up after spinal instrumentation. MRI with or without contrast is the imaging modality of choice for the evaluation of the postoperative spine. CT is the best modality for the assessment of the instrumented spine and status of bony fusion. Imaging assessment of the postoperative spine is complex and requires knowledge of key factors for interpretation like indications for the initial surgical procedure, type and approach of the surgical procedure, instrumentation used, time elapsed since surgery, and clinical complaints.

Key Points

• For proper interpretation of the postoperative spine, it is very important to understand the indication and type of spinal procedure involved

• Baseline postoperative radiographs are important to detect any change in the position of metalwork and implant integration.

• Computed tomography is the modality of choice to evaluate bony fusion and metalwork-specific complications

• Postoperative fluid collection is the most common early complication and MRI is the imaging modality of choice for the identification of the same.

• Intravenous gadolinium is helpful to differentiate between scar/vascularized granulation tissue and recurrent or residual disc.

Citation Format

• Kaur S, Lalam R, Trivedi R. Imaging of the postoperative spine. Rofo 2025; 197: 1017–1032

 

MissingLinkText

Zusammenfassung

Hintergrund

In den letzten Jahrzehnten hat die Zahl der Wirbelsäulenoperationen deutlich zugenommen, was zu einem entsprechenden Anstieg der Zahl der postoperativen Bildgebungsstudien geführt hat.

Methoden

Es wurde eine umfassende Literaturrecherche durchgeführt und verschiedene Richtlinien und Expertenmeinungen zur postoperativen Wirbelsäulenbildgebung berücksichtigt. Komplikationen werden je nach Zeitpunkt des Auftretens in frühe (in den ersten Wochen) und verzögerte Komplikationen unterteilt. Einige Komplikationen, die in der frühen postoperativen Phase auftreten, sind sowohl bei instrumentierten als auch bei nicht instrumentierten Wirbelsäulen üblich. Verzögerte Komplikationen sind spezifisch für die Art der durchgeführten Operation und wurden als solche beschrieben. In dieser Übersicht werden das normale postoperative Erscheinungsbild und die verschiedenen frühen und verzögerten Komplikationen erörtert.

Fazit

Ein Verständnis des normalen postoperativen Erscheinungsbilds ist wichtig, um Normales von Abnormalem zu unterscheiden. Eine einfache Röntgenaufnahme ist das primäre Bildgebungsverfahren für die unmittelbare postoperative Beurteilung und die langfristige Nachuntersuchung nach Wirbelsäuleninstrumentierung. MRT mit oder ohne Kontrastmittel ist das Bildgebungsverfahren der Wahl zur Beurteilung der postoperativen Wirbelsäule. Die CT ist die beste Methode zur Beurteilung der instrumentierten Wirbelsäule und des Status der Knochenfusion. Die bildgebende Beurteilung der postoperativen Wirbelsäule ist komplex und erfordert Kenntnisse über Schlüsselfaktoren zur Interpretation wie Indikationen für den ersten chirurgischen Eingriff, Art und Vorgehensweise des chirurgischen Eingriffs, verwendete Instrumente, seit der Operation vergangene Zeit und klinische Beschwerden.

Kernaussagen

• Für eine korrekte Interpretation der postoperativen Wirbelsäule ist es sehr wichtig, die Indikation und Art des Wirbelsäuleneingriffs zu verstehen.

• Postoperative Röntgenaufnahmen sind wichtig, um jegliche Positionsänderungen von Metallimplantaten und die Implantatintegration zu erkennen.

• Die Computertomografie ist die Methode der Wahl zur Beurteilung der Knochenfusion und implantatspezifischer Komplikationen.

• Die postoperative Flüssigkeitsansammlung ist die häufigste frühe Komplikation und MRT ist die bildgebende Methode zur Identifizierung derselben.

• Intravenöses Gadolinium ist hilfreich, um zwischen Narben-/vaskularisiertem Granulationsgewebe und rezidivierenden oder Restbandscheiben zu unterscheiden

 

MissingLinkText

Keywords

spine - metalwork - decompression - discectomy - fusion - postoperative

Introduction

Owing to the advancements in surgical techniques and equipment, there has been a manifold increase in the number of spinal surgical procedures performed in the past few decades. Between the years of 2004 and 2015, the number of spine surgeries increased by over 60 percent in the United States of America [1]. With the increase in the number of spine surgeries, there has been an increase in the number of postoperative imaging studies. As a result, it is pertinent that radiologists are well versed in the normal postoperative appearance, possible postoperative complications, and their appearance on imaging which will aid in early detection and guide further management.

 

MissingLinkText

Imaging modalities

Radiography

Plain radiography is a primary imaging modality for immediate postoperative assessment and long-term follow-up after spinal instrumentation procedures, given its low cost, ease of availability, and low radiation dose [2]. It allows for assessment of hardware, alignment correction, implant failure or migration, progression of osseous fusion, and detection of fractures of adjacent segment degeneration [3]. Plain radiography also allows for dynamic assessment by obtaining weight-bearing images along with extension and lateral bending to detect implant instability [4]. Baseline postoperative radiographs are very important to detect any change in device position and implant integration [2]. However, it is of no use in the diagnostic workup of an un-instrumented spine [5].

 

MissingLinkText

Ultrasound

Ultrasound has a role in detecting and characterizing superficial collections. However, depth of involvement is difficult to assess, and deeper collections cannot be evaluated with this modality.

 

MissingLinkText

Radionuclide imaging

A triple phase technetium 99m Methyl Diphosphonate (MDP) scan is very sensitive to bone metabolism alterations before they become apparent on conventional anatomic imaging modalities [6]. It is often combined with Gallium 67 scintigraphy to improve the specificity. Labelled leukocyte scintigraphy and PET imaging are useful for detecting and diagnosing spinal infection. Single-Photon Emission Computed Tomography (SPECT) often helps as a problem-solving tool in patients with recurrent or persistent pain after spinal surgery. It can act as an adjunct to detect pseudoarthrosis, adjacent segment degeneration, and hardware failure [6].

Computed Tomography (CT)

CT demonstrates the osseous structures with great detail and has a high spatial resolution which enables three-dimensional multiplanar reformats, allowing the evaluation of implant position, alignment, and bony fusion [2]. CT is routinely used to evaluate the location and integrity of the hardware, hardware failure, and loosening and to assess osseous fusion after arthrodesis [7]. In the presence of metallic hardware, there can be significant artifacts because of the Xray beam traversing the high attenuation implants resulting in photon starvation, beam hardening, and beam scattering. These are manifested as alternate dark and bright bandlike areas or steak artefacts interfering with the evaluation of spine and surrounding structures [8]. There are fixed and variable factors that influence the severity of the artefacts. The fixed factors or non-modifiable factors are based on the inherent characteristic of the implant, which includes composition and geometric features. Titanium is less dense than stainless steel, and as a result produces fewer artifacts [9]. The interbody cage devices are made from polyetheretherketone, which produces minimal artifacts on CT [3]. The modifiable factors include altering of the various imaging parameters and protocols such as using high peak voltage, high tube current, thin sections, and narrow collimation [10]. Other advanced techniques for artifact reduction include metal artifact reduction reconstruction algorithms and dual-energy data acquisition with virtual monoenergetic extrapolation postprocessing [11].

CT myelogram is helpful in patients who are claustrophobic or in whom susceptibility artifacts from the metalwork make it impossible to interpret the MRI images. It is also particularly helpful in diagnosing a dural tear or pseudo meningocele in patients with negative MRI findings but a strong clinical suspicion [7].

 

MissingLinkText

Magnetic Resonance Imaging (MRI)

MRI is the primary imaging modality for the evaluation of the postoperative spine. 3 Tesla MRI provides a higher signal-to-noise ratio and faster acquisition time, which is one of the most important factors when imaging postoperative patients with pain and discomfort. However, because of the increase in susceptibility artifacts from the metal implants and increased motion artifacts in the cervical spine due to breathing and swallowing, a 1.5 Tesla MRI scanner is the scanner of choice for imaging the postoperative spine [12].

The imaging protocols are similar to the imaging protocols for the native spine with an emphasis on minimizing susceptibility artifacts. The protocol includes axial and sagittal T1- and T2-weighted turbo/spin echo sequences with an additional coronal T2-weighted sequence. The coronal T2-weighted sequence improves the detection of lumbosacral transitional anatomy and extraforaminal far lateral disc herniations. At least one fat-suppressed sequence in one plane is essential to look for postoperative collections and seromas. Short tau inversion recovery (STIR) is the most reliable and robust fat suppression technique in the presence of metallic hardware. It does not depend on the precession frequency and is not sensitive to magnetic field inhomogeneities. In the presence of metal resulting in an inhomogeneous magnetic field, there is a shift in the precession frequency of fat surrounding the metal, thereby making the chemical shift fat suppression techniques unreliable. Gradient echo sequences are also avoided because of their magnetic susceptibility. Patients in pain and with an inability to stay still resulting in excessive motion artifacts can be imaged using faster scanning techniques like BLADE (Siemens) and PROPELLER (periodically rotated overlapping parallel lines with enhanced reconstruction) (General Electric).

One of the strategies for metal artifact reduction includes altering the parameters of conventional MRI like increasing the bandwidth and matrix size and using small voxels. But these techniques come at the price of a reduction in the signal-to-noise ratio, and the number of excitations must be increased to compensate for the loss of signal which in turn results in longer acquisition times. Multispectral acquisition techniques like slice-encoding for metal artifact correction (SEMAC), multi-acquisition variable-resonance image combination (MAVRIC), and a combination of SEMAC and MAVRIC known as MAVRIS-SL [13]. Multi Spectral Imaging (MSI) techniques reduce metal artifact distortion along with three-dimensional imaging to increase the signal-to-noise ratio.

 

MissingLinkText

 

MissingLinkText

 

MissingLinkText

Intravenous Gadolinium

There are several indications for using intravenous gadolinium in postoperative spine imaging which includes differentiating between scar tissue/vascularized granulation tissue and a recurrent or residual disc fragment. The granulation tissue or the scar tissue enhances as there is a lot of extracellular space in developing fibrosis and granulation tissue which allows for the ease of permeation/extravasation of the intravascular contrast. On the other hand, a residual or recurrent disc fragment does not demonstrate enhancement [14]. The scar tissue enhances consistently, and this feature makes it possible to distinguish between scar and disc with an accuracy of 96 percent [15]. IV contrast also helps to determine the extent, characterize the collections, and differentiate them from phlegmon.

 

MissingLinkText

Reporting a postoperative scan

Imaging of the postoperative spine is often done for one of the following indications:

1. To confirm that the surgery is satisfactory, i.e., correct level has been operated on with adequate decompression and correct positioning of the metalwork and other adjuncts ([Fig. 1]).

2. Failed back surgery for patients having continuing symptoms.

3. To assess complications (immediate and delayed).

Imaging assessment of the postoperative scan is often complex and requires thorough knowledge of the following:

• Initial spinal pathological condition for which the procedure was performed

• Normal anatomy of the patient

• Preoperative imaging

• Type of surgery performed

• Clinical presentation and time of development of symptoms since surgery

• Time interval between surgery and imaging

• Any symptom-free interval

• Whether the symptoms are recurrent or new

• Biochemistry

Fig. 1 Sagittal T2-weighted MRI a shows canal stenosis at L4/L5 with buckling of the proximal cauda equina nerve roots. Axial T2-weighted MRI b at the level of L3/L4 depicts a normal central canal with no canal or lateral recess compromise. Axial T2-weighted MRI c at L4/L5 demonstrates lumbar canal stenosis with complete CSF effacement. Axial T2-weighted MRI d demonstrates bilateral lateral recess stenosis at L5/S1. The surgical plan was to decompress the L4/L5 and L5/S1 levels. Postoperative sagittal T2-weighted MRI e demonstrates an ill-defined postoperative hematoma at the L3/L4 and L4/L5 levels. Axial T2-weighted MRI at the L3/L4 f, L4/L5 g, and L5/S1 h levels depicts satisfactory decompressive surgery at L3/L4 and L4/L5 and not at the L5/S1 level as originally planned. It is common that an adequately decompressed thecal sac does not return to its normal caliber immediately after surgery and the CSF can look effaced, despite satisfactory decompression. This was a case of decompression at the wrong level.

 

MissingLinkText

Spinal decompression and instrumentation techniques

To interpret postoperative imaging, it is important to understand the indications for the procedure and the performed procedure. The main surgical interventions include discectomy, spinal decompression, spinal decompression combined with arthrodesis, spinal fusion and stabilization, and deformity correction. The aim of spinal decompression is to relieve the compression on the spinal cord, canal, or the nerve roots. This is secondary to disc degeneration and prolapse, ligamentum flavum hypertrophy, or facet arthropathy, which may be accompanied by spondylolisthesis [16]. Decompression procedures are frequently combined with arthrodesis to fuse the spinal segments that may otherwise be subjected to excessive stress secondary to altered biomechanics [16]. Fusion is also indicated when there is instability due to degenerative disc and facet joint disease, spondylolysis and spondylolisthesis, trauma, infection, or malignancy. The use of spinal implants and instrumentation improves the rate of fusion by providing early stability until mature osseous fusion is achieved.

The various fusion and instrumentation procedures are named based on the surgical approach used to place the cages/grafts in the disc space [16]. In the lumbar spine, for example, they include transforaminal lumbar interbody fusion (TLIF), posterior lumbar interbody fusion (PLIF), anterior lumbar interbody fusion (ALIF), lateral or extreme lateral lumbar interbody fusion (LLIF/XLIF), and oblique lumbar interbody fusion (OLIF).

Many anterior and posterior approaches are used in the cervical spine depending on the underlying abnormality and its level of involvement [2]. The classic approach is the Smith-Robinson anterior approach used to gain access to C2 to T1 vertebral bodies for procedures such as anterior cervical discectomy and fusion (ACDF), disc replacement, anterior corpectomy, and fusion. A posterior approach is used for laminectomy, laminoplasty, and posterior instrumentation in the form of screws and interconnecting rods. The most common technique in the cervical spine is the ACDF which involves discectomy and placement of an interbody cage or bone graft to restore the normal foraminal height and lordosis for interbody fusion to occur across the disc space.

Disc replacements are done in younger patients with a view to preserve physiological segmental motion [3]. Scoliosis correction procedures involve the use of various rods along with transpedicular screws. Growing rods are adjustable fixation constructs used in pediatric/adolescent scoliosis correction. These allow for incremental lengthening noninvasively or by means of minimally invasive procedures [17].

 

MissingLinkText

Normal postoperative MRI appearance

In the initial postoperative period there is disruption of the dorsal paraspinal musculature and fascia, along with normal edema and postoperative collection or seroma. There may be a small epidural seroma or hematoma at the site of discectomy which resolves within 2–3 months and is replaced by an epidural scar [18]. There might be patchy edema in the paraspinal muscles if a posterior approach has been used. This is a normal finding. If the medial branch of the dorsal ramus of the spinal nerve has been damaged during the procedure, a denervation edema pattern is seen within the multifidus muscle [19]. This is replaced by atrophy and fatty infiltration as time progresses, because of the chronic denervation. The intervertebral discs can sometimes appear hyperintense on T2-weighted images with disruption of the annulus fibrosis and they demonstrate contrast enhancement in 80 percent of cases [10]. This is known as chemical or mechanical discitis which usually disappears in 4–5 weeks and is not associated with elevated inflammatory markers [20].

 

MissingLinkText

Normal postoperative CT appearance

In the immediate post-surgical period, expected CT findings include pockets of gas and fluid in the surgical bed and along the surgical access site [2]. The transpedicular screws should traverse the center of the pedicle, parallel to the end plates and remain within the confines of the bone without any anterior or medial breach. The anterior cervical plate in ACDF should have a lordotic curvature and should sit flush against the anterior margin of the vertebral bodies ([Fig. 2]). There should be at least 5 mm between the margin of the plate and the adjacent vertebral body endplates to avoid peri-plate ossification [21]. The interbody cages may subside by up to 3mm into the adjacent vertebral body endplates before osseous fusion occurs [22]. For optimum positioning of the interbody cage, there should be a minimum distance of 2mm between the posterior margin of the endplate and the radio-opaque marker at the posterior margin of the graft [2]. There is usually trabecular osseous bridging across or around the interbody graft 6 months after interbody graft placement and postero-laterally after posterolateral graft placement with transpedicular screw fixation. This trabecular bony bridging should be replaced by mature trabeculation and solid cortical bone by 1 year after surgery [2].

Fig. 2 Axial T2-weighted MRI a depicts an adequately positioned interbody cage device that is centrally located. Axial T2-weighted MRI b in another patient demonstrates the displacement of the eccentrically placed interbody cage device as depicted by the arrow, into the spinal canal resulting in left lateral recess stenosis and impingement on the transiting nerve roots. Lateral radiograph c in a patient with transpedicular screw fixation from L4-S1 and interbody cage device at L4/L5 shows satisfactory metalwork. Lateral radiograph d in another patient depicts the dorsal extrusion of the interbody graft with the radiopaque marker projecting beyond the posterior vertebral body margin (white arrow). Plain radiograph e in another patient with ACDF from C4-C7 demonstrates the adequate metalwork positioning which has a lordotic curvature and sits flush against the anterior margin of vertebral bodies.

 

MissingLinkText

Complications

The decision-making process to determine what surgery and how much surgery each patient receives is performed on an individual basis and is often complex. The surgeon often tries to strike a balance between how much resection or fusion can be performed without compromising spinal stability immediately and in the long term. This often determines whether further instrumentation is performed to protect spinal stability. The aim is to perform as little as necessary, because of the increased risk of complications secondary to long and excessive surgical dissection and subsequent procedure. In this process, too little or too much surgery is sometimes performed, resulting in postoperative complications that should always be kept in mind when reporting postoperative imaging.

Complications can be divided into early (in the first few weeks) and delayed, based on the time of onset. Early complications include hematoma, infection, abscess, and pseudo-meningocele and can occur in both the non-instrumented and the instrumented spine. Complications like hardware fractures or malposition and neurovascular injury secondary to the metalwork are specific to an instrumented spine. There can be some rare catastrophic early postoperative complications after spinal surgery like cord infarction and bowel or peritoneal injury ([Fig. 3]).

Fig. 3 Sagittal T1-weighted MRI a in a patient with multi-level decompression depicts a large compressive epidural hematoma which extends cranially to the level of the T10/T11 disc space and caudally to the S1 sacral segments (arrows). Sagittal T2-weighted MRI b and axial T2-weighted MRI c in a patient with complaints of headache post-surgery demonstrates a large ill-defined fluid signal intensity collection in the epidural space with extension into the subcutaneous plane suggestive of CSF leak secondary to a dural tear. Axial CT d and coronal reformatted CT e demonstrate abnormal medial positioning of the left-sided L3 screw with breach of the medial wall and impingement on the lateral recess. Axial post-contrast T1-weighted MRI f demonstrates enhancing inflammatory tissue in the pre- and paravertebral spaces with the presence of small peripherally enhancing epidural (bold arrow) and paravertebral abscesses (dotted arrow). Sagittal T2-weighted MRI g in a young 14-year-old patient immediately after idiopathic thoracic scoliosis repair depicts diffuse swelling of the cord with an abnormal increased signal throughout suggestive of cord infarction. Axial CT h, i in another patient with XLIF procedure demonstrates air within the bowel wall in the ascending colon suggestive of bowel injury (arrow) and a large left-sided retroperitoneal hematoma (arrow) secondary to extreme lateral access during the procedure.

Delayed complications are specific to the surgery that is performed and should be interpreted and evaluated carefully. After discectomy, the most common delayed complications include recurrent disc herniations and epidural fibrosis. Complications after decompression can occur because of either inadequate decompression resulting in continued nerve compression, or too much decompression resulting in instability. Complications specific to spinal fusion and stabilization include failure of fusion or pseudoarthrosis, adjacent segment degeneration, hardware failure, and implant migration. The complexity of spinal surgery is increasing. Deformity correction and long segment fusion can result in hardware-related complications as mentioned above along with proximal junctional kyphosis. Complications like arachnoiditis ossificans are not specific and can be seen after any spinal intervention ([Table 1]).

Table 1 Early complications can be divided into complications common to both the instrumented and non-instrumented spine and those specific to the instrumented spine as depicted in the table below. Delayed complications are usually specific to the surgical procedure and, therefore, knowledge of the surgical procedure, approach, and metalwork used is very important for identification and characterization of these complications.

EARLY COMPLICATIONS
1	Common to both the instrumented and non-instrumented spine	Postoperative seroma, hematoma, CSF leak, infection
2	Instrumented spine	Hardware/metalwork malposition, neurovascular injury (most vulnerable – common iliac vein)
3	Catastrophic complications	Cord injury/ischemia, bowel or retroperitoneal injury (specific to XLIF)
DELAYED COMPLICATIONS
1	Specific to surgical procedure
a.	Discectomy	Recurrent/residual disc, epidural fibrosis
b.	Decompression	Inadequate decompression, instability
c.	Fusion	Failure of fusion/pseudoarthrosis, adjacent segment degeneration
d.	Scoliosis/deformity correction	Proximal junctional kyphosis
2	Metalwork-related	Loosening, screw fractures, displacement of metalwork, implant failure and mechanical engagement of the components
3	Non-specific	Arachnoiditis/arachnoiditis ossificans, meningomyelocele, donor site-related complications

 

MissingLinkText

Early complications

Postoperative collection

Early complications within the first 3 months are often iatrogenic and MRI is the main imaging modality to help identify these complications. The most common complication is a postoperative fluid collection which can be a seroma, hematoma, abscess, or pseudo-meningocele.

Seromas are the most common postoperative fluid collections and have a low T1 and bright T2 signal on MRI. A small fluid collection in the postoperative surgical bed is a common and expected finding. However, a compressive fluid collection resulting in neurological compromise is a time-sensitive complication. The greatest challenge is determining the degree of thecal sac compression that warrants repeat surgery. This should be solely based on a patients’ symptoms and neurological examination. It is often seen that a satisfactorily decompressed thecal sac does not return to its normal caliber immediately after surgery and the CSF can look effaced, despite satisfactory decompression. This can be due to a multitude of factors, including the time needed for the compressed sac to expand and normal postoperative edema and inflammation secondary to surgical trauma. The mass effect and CSF effacement in the immediate postoperative period are significant only if the patient is symptomatic. This stresses the importance of collaborating closely with the spinal surgical team and the need for effective communication.

Acute hematomas have an iso-hypointense signal on T1-weighted images and a hypo-hyperintense signal on T2-weighted images. In the subacute phase, hematomas are visualized with an increase in both the T1 and T2 signal with moderate peripheral enhancement. Hematomas may have a bright T1 signal and layering because of the fluid hematocrit level. Hemosiderin in hematomas shows blooming on gradient echo images, which have an extremely limited value in the presence of metallic hardware because of excessive susceptibility artifacts. It is sometimes difficult to characterize the location of the hematoma as epidural or subdural and some signs on MRI may facilitate this differentiation. Centrally clumped nerve roots with laterally exiting nerve roots give rise to a Mercedes Benz sign suggestive of a subdural hematoma [23]. The cap sign, because of the presence of epidural fat, can also be seen as the subdural blood gravitates and settles dorsally [23].

Surgical site infection is the most common cause of morbidity in patients undergoing spinal surgery. The most common causative organisms are Staphylococcus aureus, Staphylococcus epidermidis, and Enterococcus faecalis [25]. Osseous and soft-tissue edema is the earliest and most sensitive imaging finding of an infection, but it is non-specific in the early postoperative period, due to the presence of edema caused by the surgical procedure itself. The absence of edema, on the other hand, has a very strong negative predictive value for infection. MRI is the imaging modality of choice to detect spondylodiscitis or abscess formation. However, abscesses can have an identical imaging appearance to a postoperative collection or hematoma. Diffusion-weighted imaging may be of help as abscesses have a lower apparent diffusion coefficient compared to other collections like seroma or pseudo-meningocele. The persistence of locules of gas or air within and beyond the margins of a surgical collection are secondary signs of an underlying infection or abscess. The development of collections in tissues beyond the surgical bed, like the psoas muscles in lumbar spine surgery and prevertebral tissues in cervical spine operations, should raise the suspicion of an underlying abscess [26]. MRI may depict an erosive change in the endplate and resorption around the metal hardware. There might be a high signal within the intervertebral disc in spondylodiscitis with an associated epidural or pre/paravertebral collection. The imaging findings, especially in the postoperative setting, should always be correlated with the clinical signs and symptoms and laboratory parameters like CRP and ESR.

Dural tears develop due to accidental durotomy at the time of surgery and are seen in approximately 5% of lumbar spine surgeries [27]. If the tear is not identified intraoperatively or if the surgical repair fails, it results in the formation of pseudo-meningoceles or extradural CSF collections. The clinical presentation includes headache, backache, radicular symptoms, and localized subcutaneous swelling. On MRI, these collections follow the CSF signal, i.e. low T1 and fluid bright T2 signal, and can exhibit peripheral enhancement. The communicating tract between the dura and the collection can manifest as a “jet phenomenon” due to the dephasing of the proton spins because of fast CSF flow [28]. Sometimes the dural tears are too small to be identified on MRI. A CT myelogram may be helpful in demonstrating communication between the extradural collection and the thecal sac.

 

MissingLinkText

Hardware malposition

Hardware malpositioning can be a cause of immediate postoperative back or radicular pain. Normally the screws should be contained within the pedicels and positioned parallel to the vertebral body endplates. Abnormal medial positioning can result in impingement on the nerve roots in the lateral recess resulting in radicular symptoms. The screws can also erroneously enter the neural foramina resulting in nerve impingement and neuritis. Not all screw breaches are symptomatic and, therefore, do not need corrective surgery. In fact, removing an asymptomatic screw lying next to a nerve root or vessel can cause further trauma during the unscrewing process. There is a reported incidence of approximately 5.1% for pedicel screw breaches. However, the frequency of neurological symptoms is only 0.2% [29]. A nerve that is being impinged secondary to the screw may demonstrate swelling, abnormal increased T2 hyperintensity, or contrast enhancement suggestive of neuritis.

 

MissingLinkText

Neurovascular injury

Screws that pass beyond the anterior vertebral cortex are generally not a cause for concern. However, they can rarely injure and penetrate iliac veins. This can result in vascular injury, including thrombosis, pseudoaneurysm, and fistula formation. The common iliac vein is the most vulnerable to direct injury because of the ventral positioning with respect to the vertebral body predominantly at the level of L5 [30].

 

MissingLinkText

 

MissingLinkText

Late complications

Delayed complications generally occur 12 weeks after surgery and are specific to the surgical procedure that was performed. Delayed complications are often predictable depending on the original surgery and the time of onset of postoperative symptoms. Complications are usually either due to too little or too much surgery or due to adjacent segment problems secondary to altered dynamics. Knowing the nature of the original surgery and the clinical indications narrows the potential delayed complications to two or three possibilities. Knowing this will help the radiologist to seek these abnormalities.

When complex spinal surgery is performed and a reporting radiologist identifies one problem, they must understand the mechanical consequences and cause of the identified radiological finding. Subsequently, this should lead them to seek out and identify other potential radiological findings that would confirm the mechanical complication. For example, if one sees a loose screw, they should then seek out loosening of the opposite sided screw either at that level or the level immediately above or below. Radiologists should also seek out evidence of motion in the disc and facet joints, edema related to the excessive movement, the absence of fusion bone mass, and fracture/migration of metalwork, which are all features compatible with non-fusion and excessive movement ([Fig. 4]). These additional features may be seen on the index radiological examination or are likely to develop on subsequent imaging.

Fig. 4 Sagittal T1-weighted a and T2-weighted b MRI images demonstrate an interbody fusion device at L4/L5 with good positioning and fusion. There are fatty endplate changes but no endplate edema. However, at the adjacent segment L5/S1, gas is present within the intervertebral disc, which is suggestive of motion across the segment. Sagittal c and coronal d reconstructed CT confirms trabecular bone bridging across the interbody cage at L4/L5 with unremarkable appearance of the screws. Once again there is gas within the disc at L5/S1 as seen on MRI and loosening of the S1 screws further confirming motion at L5/S1.

Post-discectomy

Recurrent disc herniation is the most common complication after discectomy with a reported incidence of 3% to 18% [31]. It is often a challenge to differentiate between recurrent and residual disc prolapse and careful comparison with preoperative imaging is helpful. Epidural fibrosis is a common occurrence after discectomy. However, it is difficult to differentiate from recurrent disc material as both have a similar signal intensity on T1- and T2-weighted images and both can cause a significant mass effect. IV gadolinium is helpful as the recurrent disc does not demonstrate any contrast enhancement as compared to epidural fibrosis which shows varying degrees of enhancement ([Fig. 5]). Epidural fibrosis is ill-defined with irregular margins and may demonstrate thecal sac retraction secondary to the fibrosis. On the other hand, a herniated disc has smooth margins with no significant retraction. It is very important for radiologists to differentiate between recurrent disc and epidural fibrosis as a recurrent disc can be managed with repeat surgery, whereas surgery has no role in epidural fibrosis and can make it worse resulting in more scarring and adhesions.

Fig. 5 Sagittal T2-weighted a and axial T2-weighted b MRI in a patient with recurrent radicular symptoms after discectomy. A recurrent right paracentral disc protrusion impinging on the transiting nerve root is seen. Axial T1 post-contrast MRI c depicts peripheral enhancement (arrow) around the prolapsed disc. Axial T2-weighted MRI d in another patient with similar complaints has a similar appearance to that of the patient in a, b. However, axial T1 post-contrast MRI e demonstrates diffuse ill-defined enhancing tissue around the nerve root suggestive of epidural fibrosis rather than recurrent disc prolapse.

 

MissingLinkText

Spinal decompression

Spinal decompression procedures include laminotomy and laminectomy along with variable degrees of facetectomy. Complications can arise due to inadequate decompression resulting in recurrent compression and symptoms. It is important to compare the postoperative images with preoperative scans to comment on the adequacy of the decompression. An adequately decompressed central canal has restitution of the CSF within the thecal sac with clear lateral recesses. However, complications can arise because of vigorous/too much decompression resulting in instability. If too much of the facet joint is removed during the procedure, the remnant spinal segment can become unstable. Any segment that is unstable is characterized by excessive motion at that segment which in turn predisposes rapid disc and facet joint degeneration. The main clue for identifying instability is to look for motion. Motion is characterized by the presence of fluid or gas within the facet joints and exaggerated facet joint degeneration resulting in spondylolisthesis along with rapid disc degeneration. This can result in foraminal or canal stenosis and patients usually present with radiculopathy, claudication, or cauda equina syndrome.

 

MissingLinkText

Spinal fusion and decompression with arthrodesis

Spinal fusion is usually achieved by interbody fusion devices which can be supported with posterolateral instrumentation and bone grafts. There can be complications secondary to the failure of bone fusion or pseudoarthrosis, implant failure and fracture, implant migration, and adjacent segment degeneration.

Pseudoarthrosis or failure of bony fusion can occur in up to 5% to 35% of patients [32] and may be asymptomatic. Fusion spanning more than 3 segments, revision surgery, infection, smoking, and previous radiation are some of the risk factors associated with pseudoarthrosis [16]. Spinal implant loosening can occur as a result of micromotion, infection, and reaction to implant or metal debris. CT is the imaging modality of choice for evaluating non-fusion/pseudoarthrosis. The presence of circumferential peri-implant osteolysis of more than 2mm is suggestive of implant loosening [33]. The most definitive sign of loosening or osteolysis is the demonstration of a change in position of the implant on serial imaging ([Fig. 6]). Bone resorption around the screws results in loosening which in turn results in increased motion, further bone resorption and more loosening [34]. One of the important principles when evaluating the postoperative spine is to look for additional findings if one abnormality is seen. Failure of fusion results in a cascade of other findings which act as important clues for the diagnosis. Once lucency is noted around the screws, the adjacent disc and facet joints should be evaluated, as often they will show evidence of movement and degeneration. Loosening can also be suggested by graft subsidence, implant migration, screw backout, and implant fracture ([Fig. 7]). Failure of mature osseous bridging across the interbody graft or posterolateral grafts 24 months after surgery is suggestive of non-union [2]. Repetitive or excessive stress can result in implant failure or fracture. Subsequently, this instability can interfere with osseous fusion resulting in pseudoarthrosis. The various components of the fixation construct namely screws, rods, bolts, and plates should sit flush against each other. In the presence of loosening or failure, there can be mechanical disengagement between the various components ([Table 2]).

Fig. 6 Plain radiograph a in the immediate postoperative period demonstrates long segment deformity correction for scoliosis with transpedicular screws and interconnecting rods. Baseline postoperative radiographs are important to detect any change in the implant position or implant migration. Plain radiograph b 3 years after surgery demonstrates lucency around L5 transpedicular screws bilaterally (arrows). Coronal T2-weighted MRI c demonstrates bone marrow edema within the L5 vertebral body which is abnormal 3 years after surgery. Persistent or new bone marrow edema is an indirect sign of loosening which is confirmed on axial CT d.

Fig. 7 Anteroposterior a and lateral b radiographs in a patient with T11 to sacral fusion with transpedicular screws and interconnecting rods and bilateral iliac screws depict fracture of the left-sided rod just below the level of the L5 screw. This results in alteration in the biomechanics and there is abnormal increased stress on the rest of the fixation construct resulting in fracture of the right-sided iliac screw and the left L5 screw subsequently (circles) depicted on anteroposterior c and lateral d radiographs 14 months later. Coronal reformatted CT e depicts failure of facet joint fusion (dotted arrow) and fracture of the posterolateral fusion mass (arrow) secondary to the implant fracture.

Table 2 Fusion can be assessed by direct and indirect signs (suggestive of motion) as depicted in the table below.

HOW TO ASSESS FUSION?
DIRECT SIGNS	Solid bridging bone across the disc space
	Posterolateral bone mass
	Fusion of the facet joints
INDIRECT SIGNS SUGGESTIVE OF MOTION – SIGNIFYING NON-UNION	Fluid/gas in disc space or facet joints
	New/increased disc prolapse
	New/increased Type I modic end plate changes
	Reactive bone formation at site of movement
	New vertebral compression
CHECK METALWORK	Screw loosening
	Fractures
	Interbody device migration
	Mechanical disengagement of fixation construct
	Screw backouts

Although CT is the imaging modality of choice for diagnosing pseudoarthrosis, various findings on MRI can be suggestive of a failure of fusion. T1-weighted MRI images are the most helpful images for evaluating fusion and can depict mature osseous bridging across the disc space or posterior elements. Persistence of end plate edema or increasing end plate edema as compared to previous scans is highly suggestive of movement/failure of fusion. Other findings on MRI include the presence of gas in the disc space, increasing disc prolapse, facet joint degeneration with effusion or gas within the joints, and absence of solid posterolateral fusion mass.

It is important to differentiate aseptic mechanical loosening from infection-related peri-implant osteolysis ([Fig. 8]). In pseudoarthrosis, the lucency around the screw is more pronounced around the distal tip of the screw because the tip forms the pivot around which the motion occurs. However, lucency in osteolysis secondary to infection is more diffuse [2]. Differentiating between the two based on imaging alone is very difficult and needs correlation with the clinical symptoms and inflammatory markers. Radionuclide imaging with labelled WBCs has a role in diagnosing osteolysis due to underlying infection.

Fig. 8 Sagittal T1-weighted MRI a, b and sagittal post-contrast T1-weighted MRI c, d in a patient with a 3-month history of L4-S1 interbody and posterior fusion complaining of pain on the left side of the lumbosacral junction demonstrates transpedicular screw fixation with interbody cage devices at L4/L5 and L5/S1. The cage at L4/L5 is adequately positioned. Post-contrast MRI depicts the enhancement in the interbody graft (white arrow), which represents normal vascularized granulation tissue and should not be confused with discitis. However, the interbody cage at L5/S1 (dotted arrow) is anteriorly displaced. There is a low T1 signal in the S1 vertebral body suggestive of edema, which enhances on post-contrast images and is suspicious for infection. Sagittal CT e and axial CT f demonstrate fusion and adequate positioning of the cage at L4/L5 with anterior displacement of the L5/S1 cage and loosening around the left S1 screw.

Anything that results in alteration in the biomechanics across a spinal segment like loosening, infection, trauma, or implant fracture can result in implant migration or graft extrusion. There is an increased risk of dorsal extrusion of interbody grafts if the distance between the posterior margin of the graft and the vertebral body is less than 2mm resulting in spinal canal or lateral recess compromise. Lateral graft extrusion can result in injury or damage to the lumbosacral plexus. A subsidence of an interbody graft in the adjacent vertebral body endplates of more than 3mm is a sign of non-fusion and often results in the reduction of disc height with foraminal narrowing. If there is an abnormal lateralized position of the interbody cage device, this can result in uneven transmission of forces and abnormal axial loading across the segment, resulting in accelerated asymmetrical degeneration.

All of the above complications result from loosening or culminate in loosening through a vicious cycle. However, if there is solid fusion at the operated segment, there is a change in the biomechanics. This change increases the stress on the levels above or below the fixed immobile segment, resulting in adjacent segment degeneration ([Fig. 9]). It is more commonly seen in multiple level fusion as compared to single level fusion. The instrumented immobilized spine acts as a lever arm, resulting in the summation of forces and increased transmission at the adjacent motion segments both caudal and cranial to the fused spine [2].

Fig. 9 Sagittal T2-weighted MRI a in a patient with cervical spondylosis depicting canal stenosis and cord impingement with myelomalacia at C5/C6. The patient underwent ACDF and postoperative sagittal T2-weighted MRI b depicts good decompression with restitution of the CSF around the cord. Small focus of residual myelomalacia is seen at the level of the operated segment. However, 4 years after surgery, the patient became symptomatic again and sagittal T2-weighted MRI c depicts an adjacent segment degenerative change at C4/C5 with canal narrowing and cord impingement. Sagittal T2-weighted MRI d in another patient with lumbar canal stenosis demonstrates multilevel stenosis from L2 to L5. The patient underwent multilevel decompression without any instrumentation or fixation. However, he presented 6 years later with similar symptoms and sagittal T2-weighted MRI e at the time of presentation demonstrates instability at L3/L4 with severe reduction of disc height and anterolisthesis with Type I modic endplate changes resulting in recurrent stenosis with buckling of the caudal equina nerve roots. SPECT-CT f in a patient with transpedicular screw fixation at L3/L4 demonstrates increased uptake at the adjacent L4/L5 segment with solid and unremarkable fusion across L3/L4. SPECT-CT can act as a problem-solving tool in patients with satisfactory metalwork and solid fusion but persistent back pain.

 

MissingLinkText

Deformity correction

There has been an increased incidence of complex long segment surgical procedures for deformity correction. It is, therefore, very important to evaluate the proximal and distal junctions in long segment fusion, as the failure often starts at these junctions ([Fig. 10]). When metal failure/fracture occurs, it is usually at the midpoint of the long construct, where the stresses are at a maximum. All the above-mentioned complications associated with spinal instrumentation and arthrodesis can also be seen with deformity correction surgeries. Proximal junctional failure, however, is specific to deformity correction surgery.

Fig. 10 Sagittal T2-weighted a and sagittal T1-weighted b MRI in a patient with long segment un-instrumented fusion in the cervico-thoracic spine. There is solid fusion with formation of a solid posterolateral bone mass along with fusion of the facet joints (dotted arrow). The intervertebral discs in the thoracic spine are pristine with no evidence of degeneration which is also an indirect sign of fusion and lack of motion across the spinal segments. There is alteration in the biomechanics which increases the stress on the levels proximal and distal to the fixed immobilized segment resulting in disc and facet joint degeneration (arrows). Sagittal CT reformat c confirms the solid posterolateral bone mass formation. Vacuum phenomenon and disc degeneration secondary to increased transmission of forces is seen at the adjacent motion segments both cranial and caudal to the fused spine.

Proximal junctional failure is a major complication when a long segment (at least four vertebral levels) has been fused [16]. The fixation construct is very stiff and there is increased mechanical stress proximally at the junction of the instrumented-immobilized segment and the immediately proximal mobile segment. This results in instability at this level, accelerated disc degeneration with loss of disc height, and proximal junctional kyphosis ([Fig. 11]). The increased mechanical stress can also result in wedge compression fractures at the most proximal instrumented level or one segment proximal to it. Once again, a vicious cycle starts, resulting in implant failure and disruption of the posterior ligamentous structure with screw backouts and mechanical disengagement of the fixation construct.

Fig. 11 Plain radiograph a in the immediate postoperative period in a patient with scoliosis repair depicts satisfactory positioning of metalwork with the transpedicular screws being parallel to the vertebral end plates and the posterior rods are flush.10 months later, plain radiograph b and sagittal CT c, d demonstrate a loss of vertebral body height at the proximal level with wedge compression of the T10 vertebral body (arrow) resulting in proximal junctional kyphosis. The screws in T10 are no longer parallel to the vertebral body end plates and are encroaching on the intervertebral disc space. The interconnecting rod is also not sitting flush with the posterior elements (dotted arrow) best depicted on CT.

 

MissingLinkText

 

MissingLinkText

Miscellaneous

Arachnoiditis

Arachnoiditis is inflammation of the cauda equina nerve roots and can occur secondary to surgery, trauma, infection, hemorrhage, and intrathecal injection of compounds. The nerve roots are abnormally thickened and fibrosed and can clump together, giving rise to an intradural soft-tissue mass. These can also clump and adhere to the theca resulting in an empty thecal sac sign. The nerve roots may demonstrate abnormal thickening and fanning on sagittal sequences. In some cases, there is osteoblastic metaplasia of arachnoid cells resulting in the deposition of osseous material known as arachnoiditis ossificans ([Fig. 12]). Domenicucci et al. proposed a classification system for arachnoiditis ossificans [35]:

Fig. 12 Sagittal T1-weighted MRI a in a patient demonstrates hyperintensity (arrow) in relation to the cauda equina nerve roots which appear clumped together in the anterior part of the thecal sac. This T1 hyperintensity can act as a clue for the diagnosis of arachnoiditis ossificans which is confirmed on the corresponding axial CT b. Coronal reformatted CT c and axial CT d in another patient with spinal decompression and fusion demonstrates ossification of the clumped nerve roots in the thecal sac.

• Type I: presence of peripheral semi-circular ossification

• Type II: the entire circumference is involved

• Type III: the internal contents of the thecal sac are also involved giving rise to a “honeycomb” pattern [35].

 

MissingLinkText

Complications in relation to graft donor site

The posterior iliac bone serves as the most common donor site for bone grafts used in spinal fusion. There can be complications like donor site infection, hemorrhage, abscess, and sinus formation. MRI is the modality of choice for diagnosis and intravenous gadolinium can be helpful in cases of abscess and sinus formation.

 

MissingLinkText

 

MissingLinkText

Conclusion

Understanding the normal radiological postoperative appearance of the spine and its associated complications requires knowledge of the following: the indication for surgery, the appearance of preoperative imaging, the surgical procedure performed, the instrumentation used, clinical complaints, and the time interval between surgery and symptoms. MRI is the mainstay imaging modality in the non-instrumented spine because of its high soft-tissue contrast resolution, allowing for accurate assessment of the spinal canal, nerve roots, and foramina. CT is the gold-standard imaging modality for evaluating the success of spinal instrumentation surgery and the detection of implant-related complications. Modified MRI protocols augmented with metal artifact reduction techniques in conjunction with CT can help detect most surgical complications.

 

MissingLinkText

 

MissingLinkText

Conflict of Interest

The authors declare that they have no conflict of interest.

• References

• 1 Martin BI, Mirza SK, Spina N. et al. Trends in Lumbar Fusion Procedure Rates and Associated Hospital Costs for Degenerative Spinal Diseases in the United States, 2004 to 2015. Spine (Phila Pa 1976) 2019; 44: 369-376
• 2 Ghodasara N, Yi PH, Clark K. et al. Postoperative Spinal CT: What the Radiologist Needs to Know. Radiographics 2019; 39: 1840-1861
• 3 Petscavage-Thomas JM, Ha AS. Imaging current spine hardware: part 1, cervical spine and fracture fixation. AJR Am J Roentgenol 2014; 203: 394-405
• 4 Venu V, Vertinsky AT, Malfair D. et al. Plain radiograph assessment of spinal hardware. Semin Musculoskelet Radiol 2011; 15: 151-162
• 5 Salgado R, Van Goethem JWM, van den Hauwe L. et al. Imaging of the postoperative spine. Semin Roentgenol 2006; 41: 312-326
• 6 Al-Riyami K, Gnanasegaran G, Van den Wyngaert T. et al. Bone SPECT/CT in the postoperative spine: a focus on spinal fusion. Eur J Nucl Med Mol Imaging 2017; 44: 2094-2104
• 7 Krishnan K, Queler SC, Sneag DB. Principles of Postoperative Spine MRI. MRI of the Spine 2020; 237-251
• 8 McLellan AM, Daniel S, Corcuera-Solano I. et al. Optimized imaging of the postoperative spine. Neuroimaging Clin N Am 2014; 24: 349-364
• 9 Rutherford EE, Tarplett LJ, Davies EM. et al. Lumbar spine fusion and stabilization: hardware, techniques, and imaging appearances. Radiographics 2007; 27: 1737-1749
• 10 Splendiani A, D’Orazio F, Patriarca L. et al. Imaging of post-operative spine in intervertebral disc pathology. Musculoskelet Surg 2017; 101: 75-84
• 11 Khodarahmi I, Fishman EK, Fritz J. Dedicated CT and MRI Techniques for the Evaluation of the Postoperative Knee. Semin Musculoskelet Radiol 2018; 22: 444-456
• 12 Hargreaves BA, Worters PW, Pauly KB. et al. Metal-induced artifacts in MRI. AJR Am J Roentgenol 2011; 197: 547-555
• 13 Koch KM, Brau AC, Chen W. et al. Imaging near metal with a MAVRIC-SEMAC hybrid. Magn Reson Med 2011; 65: 71-82
• 14 Hayashi D, Roemer FW, Mian A. et al. Imaging features of postoperative complications after spinal surgery and instrumentation. AJR Am J Roentgenol 2012; 199
• 15 Ross JS, Masaryk TJ, Schrader M. et al. MR imaging of the postoperative lumbar spine: assessment with gadopentetate dimeglumine. AJR Am J Roentgenol 1990; 155: 867-872
• 16 Abel F, Tan ET, Chazen JL. et al. MRI after Lumbar Spine Decompression and Fusion Surgery: Technical Considerations, Expected Findings, and Complications. Radiology 2023; 308
• 17 Yazici M, Olgun ZD. Growing rod concepts: state of the art. Eur Spine J 2013; 22: 2
• 18 Wang Y, Luo G, Wang J. et al. Early Postoperative Magnetic Resonance Imaging Findings After Percutaneous Endoscopic Lumbar Discectomy and Their Correlations with Clinical Outcomes. World Neurosurg 2018; 111: e241-e249
• 19 Stevens KJ, Spenciner DB, Griffiths KL. et al. Comparison of minimally invasive and conventional open posterolateral lumbar fusion using magnetic resonance imaging and retraction pressure studies. J Spinal Disord Tech 2006; 19: 77-86
• 20 Grane P, Lindqvist M. Evaluation of the post-operative lumbar spine with MR imaging: The role of contrast enhancement and thickening in nerve roots. Acta radiol 1997; 38: 1035-1042
• 21 Park JB, Cho YS, Riew KD. Development of adjacent-level ossification in patients with an anterior cervical plate. J Bone Joint Surg Am 2005; 87: 558-563
• 22 Thakkar RS, Malloy JP, Thakkar SC. et al. Imaging the postoperative spine. Radiol Clin North Am 2012; 50: 731-747
• 23 Krishnan P, Banerjee TK. Classical imaging findings in spinal subdural hematoma – „Mercedes-Benz“ and „Cap“ signs. Br J Neurosurg 2016; 30: 99-100
• 24 Pierce JL, Donahue JH, Nacey NC. et al. Spinal Hematomas: What a Radiologist Needs to Know. https://doi. org/101148/rg2018180099 2018; 38: 1516-1535
• 25 Meredith DS, Kepler CK, Huang RC. et al. Postoperative infections of the lumbar spine: presentation and management. Int Orthop 2012; 36: 439-444
• 26 Acharya J, Gibbs WN. Imaging spinal infection. Radiology of Infectious Diseases 2016;
• 27 Strömqvist F, Sigmundsson FG, Strömqvist B. et al. Incidental durotomy in degenerative lumbar spine surgery – a register study of 64,431 operations. Spine J 2019; 19: 624-630
• 28 Radcliff K, Morrison WB, Kepler C. et al. Distinguishing Pseudomeningocele, Epidural Hematoma, and Postoperative Infection on Postoperative MRI. Clin Spine Surg 2016; 29: E471-E474
• 29 Salzmann SN, Plais N, Shue J. et al. Lumbar disc replacement surgery-successes and obstacles to widespread adoption. Curr Rev Musculoskelet Med 2017; 10: 153-159
• 30 Groot OQ, Hundersmarck D, Lans A. et al. Postoperative adverse events secondary to iatrogenic vascular injury during anterior lumbar spinal surgery. The Spine Journal 2021; 21: 795-802
• 31 McGirt MJ, Garcés Ambrossi GL, Datoo G. et al. Recurrent disc herniation and long-term back pain after primary lumbar discectomy: review of outcomes reported for limited versus aggressive disc removal. Neurosurgery 2009; 64: 338-344
• 32 Chun DS, Baker KC, Hsu WK. Lumbar pseudarthrosis: a review of current diagnosis and treatment. Neurosurg Focus 2015; 39
• 33 Malhotra A, Kalra VB, Wu X. et al. Imaging of lumbar spinal surgery complications. Insights Imaging 2015; 6: 579-590
• 34 Corona-Cedillo R, Saavedra-Navarrete MT, Espinoza-Garcia JJ. et al. Imaging Assessment of the Postoperative Spine: An Updated Pictorial Review of Selected Complications. Biomed Res Int 2021; 2021
• 35 Domenicucci M, Ramieri A, Passacantilli E. et al. Spinal arachnoiditis ossificans: report of three cases. Neurosurgery 2004; 55: 985

Publikationsverlauf

Eingereicht: 01. Oktober 2024

Angenommen nach Revision: 04. Dezember 2024

Artikel online veröffentlicht:
07. Februar 2025