Lumbar Fusion Techniques

Lumbar fusion creates a bony bridge between vertebral segments to eliminate motion, and is used to treat instability and degenerative conditions. Interbody techniques place bone graft or a cage in the disc space for anterior column support, typically combined with posterior pedicle screw instrumentation, which provides immediate stability while the fusion mass consolidates.

Where the approaches came from. The first lumbar fusions were posterolateral, with bone graft laid between the transverse processes. Interbody techniques followed to improve fusion rates through anterior column support: PLIF was introduced in the 1950s-60s, ALIF in the 1980s, TLIF in the 1990s and the lateral approaches (XLIF/LLIF) in the 2000s.

Where practice is now. TLIF has become the most commonly performed interbody fusion because of its balance of visualisation, fusion rates and complication profile. Minimally invasive techniques are increasingly used, while multi-level constructs and adult deformity correction remain specialised applications.

View larger

The five routes at a glance. Each approach trades access for a particular risk, and the level being fused decides which trade is worth making.

Why the anterior column matters. About 80% of the axial load passes through the anterior column and 20% through the posterior elements, the facets and pedicles. An interbody cage restores disc height and anterior column support, so the construct shares that load, and the disc space itself becomes the site of fusion.

What makes a fusion take. Compressive forces promote fusion, decorticated endplates expose vascular bone, bone graft or a substitute fills the disc space, and instrumentation eliminates motion.

The posterior route (PLIF/TLIF) passes through the paraspinal muscles, lamina and facet joints to reach the dura, the cauda equina and the exiting and traversing nerve roots. Epidural veins lie in the way and can bleed significantly.

The anterior route (ALIF) crosses the retroperitoneal space. The structures in the way are the great vessels (aorta and IVC, and the iliac vessels), the sympathetic plexus (the hypogastric plexus at L5-S1), the ureter laterally and the psoas laterally.

The lateral route (LLIF/XLIF) traverses the psoas muscle directly. The lumbar plexus lies within or behind the psoas, the genitofemoral nerve runs on its anterior surface, and the segmental vessels are the other structures met on the way to the disc. The plexus does not sit in the same part of the psoas at every level, which is why the nerve at risk changes with the level being fused:

View larger

By approach. The five routes in the Overview table are the working classification. PLIF is posterior and bilateral: both sides are retracted and two cages are placed. TLIF is posterior and unilateral: a single cage passes through the foramen. ALIF is a direct anterior, retroperitoneal exposure that takes a large cage. LLIF/XLIF is a direct lateral route through the psoas, and OLIF an oblique route anterior to the psoas that avoids both the psoas and the lumbar plexus. Each has specific indications based on the target level and the patient's anatomy.

By cage. Cage selection depends on the approach, the level and the lordosis correction wanted.

• PEEK (polyetheretherketone): radiolucent, elastic modulus similar to bone
• Titanium: strong, radiopaque, which makes fusion assessment difficult
• 3D-printed titanium: porous surface for bone ingrowth
• Carbon fibre: radiolucent, historical
• Shape follows approach: banana or curved for TLIF, straight rectangular for PLIF, large lordotic for ALIF and lateral, expandable for various approaches

By extent. A posterolateral fusion has no interbody cage; an interbody fusion alone has a cage without posterior instrumentation; a circumferential (360°) fusion combines an interbody cage with posterior instrumentation; and a staged anterior/posterior fusion does the two as separate procedures. Circumferential fusion offers the highest fusion rates for complex cases.

Who needs a fusion. The indications share a theme: instability or deformity.

• Fracture with instability: unstable burst or Chance fractures
• Unstable spondylolisthesis: Grade II or higher, or dynamic instability
• Stenosis with instability, where the decompression itself destabilises the segment
• Infection requiring stabilisation, after debridement
• Deformity: scoliosis or kyphosis correction
• Neoplasm requiring stabilisation, after resection

The differential. Before committing a patient to fusion, exclude the conditions that present with overlapping back or leg pain but are managed very differently. Misattributing pain to a degenerative segment is the commonest reason for a poor fusion outcome.

Choosing an approach. Once fusion is indicated, the pathology and the level point to the route:

• TLIF for a single- or two-level fusion, radiculopathy that needs decompression, moderate loss of disc height, and revision discectomy with instability
• ALIF when L5-S1 is the primary level, when significant disc collapse needs height restoration, when the posterior structures are to be preserved, and for a failed posterior fusion (pseudarthrosis)
• Lateral (LLIF) for L1-L4 (avoiding L5-S1), multi-level degenerative disease, when indirect decompression is wanted, and when coronal or sagittal correction is needed

Contraindications. Active infection, uncontrolled medical comorbidity and the absence of a clear structural indication are absolute. The relative ones are approach-specific: previous anterior surgery or retroperitoneal scarring for ALIF, the L5-S1 level or previous psoas surgery for the lateral approach, and, for every approach, severe osteoporosis and unrealistic expectations.

View larger

Standing radiographs. Standing AP and lateral films with flexion-extension views give what a supine MRI cannot: sagittal balance, pelvic incidence and lumbar lordosis, dynamic instability and the grade of any spondylolisthesis. They are essential because alignment and instability may not be visible on the supine study.

View larger

MRI is the primary modality for soft-tissue pathology and neural compression, and is essential for planning. It grades disc degeneration (Pfirrmann), shows the neural compression and any Modic endplate changes, assesses the adjacent levels, and maps the psoas and vascular anatomy for a lateral approach. In planning terms it identifies the pathology that needs decompression, gives the disc height for cage sizing, and shows the potential approach corridors.

CT gives the bone detail: pedicle anatomy for screw placement, the state of the facet joints and any bony stenosis. It is also the study for assessing fusion postoperatively. CT angiography maps the position of the great vessels and any aberrant vessels for ALIF planning.

Bone density. DEXA is essential in elderly patients. A T-score below -2.5 is osteoporosis, and it changes the fixation strategy towards cement augmentation.

Vascular surgery consultation. Consider it for ALIF at L4 and above, after previous vascular surgery, and where the vessel anatomy is aberrant.

Conservative treatment first. Most degenerative conditions are treated non-operatively to begin with, typically for 3-6 months: physiotherapy and activity modification, with epidural or facet injections. Most patients should complete this before fusion is considered.

When to operate. Surgery is considered when there is a clear structural indication, conservative treatment has failed, there is a progressive neurological deficit, or the functional limitation is significant.

Matching the approach to the level. Approach selection depends on the target level, the pathology and the surgeon's experience:

• L5-S1: ALIF is excellent for height restoration and lordosis; TLIF is good when decompression is needed; PLIF is the alternative to TLIF
• L4-5: TLIF is the most versatile; ALIF is possible but the vascular concern is greater; the lateral approach is technically possible but L4-5 is the challenging level for it
• L1-L4: lateral (LLIF/OLIF) is good for multi-level disease; TLIF is the standard posterior approach; ALIF is limited by the vascular anatomy

Patient factors that change the plan. Previous surgery favours a different approach; ALIF planning needs the vascular anatomy on a CT angiogram; osteoporosis means planning augmentation; and a smoker must stop before elective fusion. Put together, the common decisions are these:

• Single level with radiculopathy: TLIF
• L5-S1 with significant collapse: ALIF plus posterior fixation
• Multi-level deformity: lateral plus posterior
• Revision for pseudarthrosis: consider circumferential fusion

Circumferential (360°) fusion combines ALIF with posterior instrumentation and gives the highest fusion rate. It is indicated for pseudarthrosis, high-grade spondylolisthesis and deformity, and is reserved for the complex cases that need maximum stability. Doing both stages on the same day means longer anaesthesia and more blood loss; staging means two recoveries but is possibly safer.

View larger

The lumbosacral junction. L5-S1, and any long fusion ending on the sacrum, is a special case. The junction is a high-stress cantilever with a notoriously high rate of S1-screw failure and L5-S1 pseudarthrosis, and failure to support it is a common cause of distal construct failure and pseudarthrosis. Tricortical or bicortical S1 screw purchase and good sacral fixation are the foundation, and two additions protect it:

• Anterior column support at L5-S1 (an interbody cage, by ALIF or TLIF) whenever fusing across it, and especially in a long construct: "in front and behind" loads the graft in compression and dramatically improves the fusion rate at this junction
• Sacropelvic fixation (S2-alar-iliac or iliac screws) for long constructs to the sacrum, because S1 pedicle screws alone are insufficient: it protects the S1 screws from cantilever pull-out and supports the lumbosacral fusion until it unites. The longer the construct above, the more the bottom must be anchored

View larger

TLIFALIFLLIF/XLIF

Transforaminal lumbar interbody fusion. The patient is prone on a Wilson frame or Jackson table and the exposure is midline, taken further on the symptomatic side. A complete facetectomy on the approach side is what opens the transforaminal window to the disc, and the exiting and traversing roots are identified and protected before the discectomy begins.

1. Midline incision, exposing the symptomatic side more extensively
2. Pedicle screws at the planned levels
3. Complete facetectomy on the approach side
4. Identify and protect the exiting and traversing roots
5. Discectomy through the transforaminal window
6. Prepare the endplates with curettes and shavers, and trial for size
7. Pack bone graft into the disc space and the cage
8. Insert the cage obliquely across the midline
9. Compress the pedicle screws to engage the cage

Why it has become the workhorse. The unilateral approach reduces nerve retraction compared with PLIF, and with it the nerve complication rate; the cage angled across the midline gives central support; the contralateral facet is preserved where possible; and neuromonitoring is recommended. Interbody cage plus pedicle screws makes a circumferential construct through a single posterior exposure, which is the versatility that has put it ahead of PLIF.

View larger

Anterior lumbar interbody fusion. The patient is supine, on a radiolucent table if wanted. The retroperitoneal approach through an incision left of the midline is the most common; the transperitoneal route gives more exposure but is rarely used.

1. Left paramedian or Pfannenstiel incision
2. Retroperitoneal dissection
3. Identify and mobilise the great vessels
4. Identify and protect the ureters
5. Divide the segmental vessels, ligating carefully
6. Expose the anterior disc
7. Release the anterior longitudinal ligament carefully
8. Complete discectomy and prepare the endplates
9. Size and insert a large lordotic cage, packing bone graft around it
10. Consider an anterior plate or posterior instrumentation

What it buys and what it costs. ALIF allows excellent disc space preparation and a large cage for optimal lordosis correction, and it leaves the posterior structures intact because no laminectomy is needed. L5-S1 is the most accessible level and a vascular surgeon should be involved for L4 and above. Sympathetic plexus injury causes retrograde ejaculation in 0-4%, and posterior instrumentation may be needed for stability.

View larger

Lateral lumbar interbody fusion. The patient is in the lateral decubitus position with the table broken. A lateral incision at the target level is placed under fluoroscopy, the three abdominal wall muscles are split, and the psoas is reached by blunt dissection and traversed with sequential dilators.

1. Lateral incision at the target level, fluoroscopy guided
2. Split external oblique, internal oblique and transversus
3. Blunt dissection to the psoas
4. Direct lateral transpsoas approach to the disc with sequential dilators
5. Release the annulus and complete the discectomy
6. Prepare the endplates
7. Insert a large lateral cage
8. Supplement with posterior fixation or lateral plating

Neuromonitoring is critical. Continuous EMG and a directional retractor with EMG probes identify the safe zone in the anterior third of the disc and keep the surgeon out of the posterior third, where the lumbar plexus lies within or behind the psoas. L5-S1 is avoided because the iliac crest blocks access.

What it buys. Restoring disc height decompresses indirectly, through ligamentotaxis, without posterior surgery, and the large cage gives excellent correction potential.

View larger

Minimally invasive fusion. The minimally invasive versions of these operations cause less muscle damage and less blood loss, and give a shorter hospital stay and a faster recovery. The costs are a steep learning curve, more radiation exposure, limited visualisation, and cases for which they are not suitable.

View larger

View larger

Graft. Fusion is a race between healing and hardware failure, so know the graft hierarchy. A successful graft needs osteogenic cells, an osteoinductive signal and an osteoconductive scaffold, with mechanical stability and a viable host (the "diamond").

• Autograft is the biological gold standard. Iliac crest has all three properties but carries donor-site morbidity and is rarely used now; local bone from the laminectomy and facetectomy is the workhorse for posterolateral fusion and TLIF
• Allograft is structural (femoral ring and similar) or particulate (chips, DBM). Allograft, DBM and ceramics are osteoconductive extenders with, at most, weak osteoinduction
• Synthetics: BMP-2, tricalcium phosphate and hydroxyapatite
• Cage filling is usually a combination of local autograft with allograft or DBM; BMP may be used, but it is controversial and off-label in some applications

Each route has a signature complication, and the prevention column is the operative discipline that route demands.

In theatre. The intraoperative risks are wrong-level surgery, which is why fluoroscopic confirmation is essential; bleeding from the epidural veins and the segmental vessels; and cage malposition.

Afterwards. Infection in 1-3%, pseudarthrosis in 5-15% depending on risk factors, hardware failure (loosening or breakage) and cage subsidence.

View larger

Adjacent segment disease (2-3% per year). The principal long-term cost of fusion, and the counterweight to the techniques above. A fused segment is rigid, so the load and motion it used to absorb are transferred to the neighbouring mobile segments, which then degenerate faster; the longer the fusion, the greater the effect. About 2-3% per year is the quoted rate, but the number that matters is cumulative: over a decade a meaningful proportion of fused patients develop symptomatic degeneration at the next level, and some come back for an extension of the fusion.

The motion-preservation answer. This stress transfer is the rationale for total disc replacement (lumbar arthroplasty) at single levels: by preserving motion rather than abolishing it, the implant aims to avoid the transfer that drives adjacent-segment degeneration. The trade-off is that disc replacement has its own failure modes and is unsuitable where there is facet arthropathy, instability or deformity. Fusion is the right operation for instability and deformity, but every fusion is also a bet against the next level, and choosing the shortest construct that achieves stability, and reserving fusion for patients who genuinely need it, is the main lever the surgeon controls.

Pseudarthrosis. Smoking is the most significant modifiable risk factor; diabetes, obesity, multi-level fusion, osteoporosis and poor nutrition add to it. The diagnosis is persistent pain with no bridging bone on CT at 1 year, hardware loosening or breakage, or motion on flexion-extension films. Management is to optimise the modifiable factors and revise with improved fixation and graft augmentation, considering a different approach.

The first days. On day 0-1 the priorities are a neurological assessment with particular attention to motor function, multimodal analgesia, DVT prophylaxis and early mobilisation. After ALIF, assess bowel function because ileus is possible; after a lateral approach, assess the hip flexors.

Watching for union. The imaging schedule is genuinely sequential, and each visit asks a different question:

Signs of solid fusion. Bridging bone on CT, no motion on flexion-extension, no hardware loosening, and resolution of symptoms.

View larger

Rehabilitation. Core strengthening waits until the fusion is consolidating; flexibility, conditioning and functional training follow. Smoking cessation is mandatory, and weight optimisation and activity modification are part of the plan.

Fusion rates climb as anterior column support and posterior instrumentation are added, which is the biomechanical argument of the whole topic in a single table.

Clinical success (symptom improvement) depends on the indication: 70-80% good or excellent for spondylolisthesis, 60-70% for DDD with instability, a controversial indication, and a variable 50-70% for revision and pseudarthrosis surgery.

Prognostic factors. The two lists are near mirror images.

• Favourable: clear structural indication, non-smoker, normal BMI, single level, good psychosocial status, no workers compensation claim
• Unfavourable: smoking (the most important), obesity, multi-level fusion, osteoporosis, depression, workers compensation

View larger

Global Epidemiology and Practice Variation

Elective lumbar fusion rates have risen steadily across high-income countries over the past two decades, with marked variation by region, hospital type and surgeon. In a US Nationwide Inpatient Sample analysis of degenerative spondylolisthesis (2001-2010), the choice of fusion technique varied significantly with geographic region, teaching status and hospital size, and combined anterior/posterior or interbody constructs carried higher complication and mortality risk than posterolateral fusion alone in the acute phase. In Norway, the rate of lumbar spinal stenosis surgery more than tripled between 1999 and 2013, with surgery in those over 65 years more than quadrupling, although the proportion involving fusion actually fell from 19.3% to 10.9%. The take-home for any examination: fusion utilisation reflects health-system and surgeon factors as much as pathology, and "more fusion" has not consistently meant "better outcome".

Guidelines Side by Side

Major bodies converge on the same core principle: fusion needs a structural indication (instability, deformity, neural compression requiring destabilising decompression), and is not supported for axial low back pain from degenerative disc disease alone.

The unresolved SLIP versus Swedish Spinal Stenosis Study controversy is a classic viva discussion point: SLIP supports fusion for unstable Grade I spondylolisthesis, while the Swedish trial found no benefit from adding fusion to decompression. The pragmatic synthesis is to fuse when there is demonstrable instability or when decompression itself will destabilise the segment, and to decompress alone when the segment is stable.

Approach Choice and Modifiable Risk - Global Standard of Care

• Technique: TLIF is the most widely used posterior interbody technique; meta-analysis favours it over PLIF on complication grounds. ALIF, LLIF/XLIF and OLIF are valid alternatives selected by level and sagittal goals, with comparable fusion rates.
• Anterior approaches: an access surgeon (vascular/general) for ALIF and continuous EMG neuromonitoring for transpsoas LLIF are widely regarded as standard safeguards.
• Modifiable risk: smoking cessation is the single most important modifiable factor (nonunion relative risk 1.91); diabetic control, weight optimisation and treatment of osteoporosis (including screw cement augmentation) are recommended before elective fusion regardless of health system.

Revision fusion. Scarring obscures the anatomy, hardware may need removal, bone has been lost, and the pseudarthrosis rate is higher. The strategies answer each of those: a different approach (anterior if the previous surgery was posterior), augmented fixation (longer constructs, cement) and an optimised graft (BMP, autograft).

The osteoporotic spine. Poor screw purchase, cage subsidence and a higher failure rate are the problems. The solutions are cement augmentation of the screws, larger or expandable cages, extended fixation and optimising bone health.

View larger

Multi-level fusion. The complication rate is higher and the adjacent segment disease risk greater; combined approaches are considered, and staged procedures may be safer.

Key Takeaways

1. TLIF Has Replaced PLIF: The unilateral transforaminal approach provides adequate disc access with less nerve retraction compared to bilateral PLIF. TLIF is now the most commonly performed interbody fusion technique.

2. ALIF Best at L5-S1: The great vessels bifurcate at L5-S1, providing safe access. Above this level, vascular surgery involvement is recommended. Know the retrograde ejaculation risk from sympathetic plexus injury.

3. Lateral Approach Avoids L5-S1: The iliac crest blocks lateral access to L5-S1. The lumbar plexus runs within the psoas and is at risk - neuromonitoring is essential. Transient symptoms are common (15-25%).

4. Interbody Fusion Improves Fusion Rates: Adding an interbody cage to pedicle screw fixation increases fusion rates from 70-80% to 90-95%. Circumferential fusion provides the highest fusion rate.

5. Smoking is the Most Important Modifiable Factor: A meta-analysis found smoking nearly doubled the nonunion risk (relative risk 1.91). Cessation before elective fusion should be required.

6. Know the Indications: Clear structural indications (spondylolisthesis, instability, deformity) produce the best outcomes. Fusion for degenerative disc disease alone remains controversial.

7. Approach Selection Matters: Match the approach to the level, pathology, and patient factors. L5-S1 suits ALIF or TLIF; L1-L4 suits lateral or TLIF; multi-level deformity may need combined approaches.