Proximal Junctional Kyphosis (PJK)

Proximal junctional kyphosis (PJK) is one of the most common mechanical complications of long-segment spinal fusion, particularly after correction of adult spinal deformity (ASD). It is excessive kyphosis developing at the transition zone between the fused and the unfused spine, and most cases appear 3-18 months after surgery.

PJK and PJF. PJK is a radiographic finding: it may be stable, it is often asymptomatic, and it is usually observed. Proximal junctional failure (PJF) is a clinical and radiographic diagnosis, a vertebral fracture, ligament rupture or implant failure at the UIV that presents with pain, deformity or neurology and often requires revision surgery, with a 40% revision rate. Not all PJK progresses to PJF, and many cases can be observed if they are stable and asymptomatic.

Why it matters. PJK may be asymptomatic or cause significant morbidity, and healthcare costs are substantially increased with PJK and PJF. Incidence varies with the population operated on:

The mismatch. PJK results from a mismatch between the mechanical demands at the proximal junction and the capacity of the adjacent tissues to withstand them. A rigid fused construct meets the mobile unfused spine, and that abrupt stiffness transition concentrates stress at the junctional level. Sagittal imbalance adds to it, because a forward trunk shift increases the moment arm at the UIV; when bone, ligament or disc cannot withstand the new loads, the junction gives way.

Where the stress comes from. The junctional zone is loaded by the transition from fused to mobile segments, by the lever arm of a long construct, and by the loss of shock absorption from the fused discs.

What fails. Each tissue at the junction has its own failure pattern:

• Bone - UIV vertebral body compression, UIV+1 vertebral body fracture, superior endplate failure
• Soft tissue - the posterior ligamentous complex (PLC), the interspinous and supraspinous ligaments, and the paraspinal musculature
• Disc - UIV/UIV+1 disc degeneration and accelerated adjacent segment disease

Bone quality. Osteoporotic bone cannot resist vertebral compression, gives UIV screws reduced pull-out strength, and fractures more readily because the cortex is thin.

Compensation. Once PJK develops, the body attempts to compensate with cervical hyperlordosis, pelvic retroversion and knee flexion.

History. Establish when the index operation was, then ask about:

• New or worsening back pain - character, location, severity
• Change in posture or balance
• Neurological symptoms - weakness, numbness, bowel or bladder
• Functional limitation - walking tolerance, activities of daily living

Red flags for PJF. Any of these raises the possibility of failure:

• Acute pain after minor trauma, or sudden onset
• New neurological deficit
• A visible or palpable step-off at the end of the fusion
• Rapidly progressive kyphosis

Look and feel. Standing, the patient shows increased thoracic kyphosis and a forward trunk lean, with compensatory cervical hyperlordosis and a visible prominence at the UIV level. Palpation finds tenderness over the UIV, a step-off or prominence at the junctional level, and muscle spasm.

Neurology and flexibility. Examine motor power in full, especially if PJF is suspected, along with sensation, reflexes and gait. Check whether the kyphosis corrects with prone positioning, and look for a hip flexion contracture with the Thomas test.

Differential diagnosis. New pain or kyphosis above a fusion is not always mechanical PJK, and the alternatives must be considered and excluded.

Standing full-length radiographs first. Compare them with the immediate postoperative films and measure the PJA from UIV to UIV+2. Assess global sagittal alignment, the SVA (C7 plumb line to S1) and PI-LL, evaluate the hardware position, and document the exact UIV level. If instability is suspected, flexion-extension lateral radiographs assess motion at the UIV segment and document any subluxation.

CT, if indicated. CT detects a vertebral fracture at the UIV or UIV+1, assesses the fusion mass for pseudarthrosis, evaluates the hardware for loosening or breakage, and measures bone quality in Hounsfield units.

MRI, if there are neurological symptoms. It assesses cord compression, soft-tissue change, disc pathology and the integrity of the posterior elements.

View larger

View larger

View larger

Bone density. DEXA gives a baseline T-score and is repeated if osteoporosis treatment is started. On CT, the density of the UIV and UIV+1 vertebral bodies adds to it: less than 110 HU suggests osteoporosis and guides the need for cement augmentation at revision.

Complications of PJK and PJF. Each frequency below is per case of PJF or of PJK, as marked:

Complications of revision. Early, the risks are neurological injury (1-5%), dural tear (5-10%), wound infection (5-10%) and significant blood loss. Late, they are recurrent PJK (20-30%), pseudarthrosis (10-20%), adjacent segment disease and chronic pain.

Why it recurs. Advanced age, poor bone quality and a long fusion cannot be changed at revision. Over-correction of alignment, inadequate proximal extension, failure to address osteoporosis and poor soft-tissue preservation can, which is why prevention during primary surgery matters and why every modifiable factor is addressed at revision.

The early phase. Mobilise early under supervision once the construct is judged stable. A thoracolumbosacral orthosis (TLSO) may be used for comfort, though high-level evidence for brace prevention of PJK is lacking. Observe neurology regularly, especially after revision for PJF or after deformity correction, and use multimodal analgesia, expecting a settling period after a revision extension.

Bone health optimisation. Low bone mineral density is an independent predictor of proximal junctional failure (Yagi et al., pmid 28902106), so bone health is a core part of postoperative and perioperative care:

• DEXA assessment and treatment of osteoporosis (bisphosphonates or denosumab)
• Calcium and vitamin D repletion
• In elective primary surgery, consider deferring 3-6 months for anabolic or antiresorptive optimisation in severe osteoporosis
• Anabolic agents (e.g. teriparatide) are of particular interest where bone quality is the limiting factor, though their specific role in PJK prevention remains under study

Surveillance. The first 3-6 months are the peak-risk window for acute failure (mean time to acute PJF 11.4 weeks; Hostin et al., pmid 22986834), so surveillance should be most intensive early:

Untreated PJK. Many cases remain stable, the progression rate is approximately 20-30%, and some patients' symptoms improve.

Untreated PJF. PJF is generally progressive, carries a neurological risk if the cord is involved, and gives poor quality-of-life outcomes.

After revision. Radiographic and clinical results of revision surgery:

Prognosis. Outcomes are better in younger patients with good bone quality, at a first revision, after successful alignment correction and without neurological deficit. Multiple prior revisions, severe osteoporosis, persistent sagittal imbalance, neurological complications and medical comorbidities predict a worse result.

Global Epidemiology

Adult spinal deformity surgery is increasing worldwide as populations age, and PJK is consistently the most common mechanical complication of long-segment fusion across health systems. Reported incidence is broadly consistent internationally: radiographic PJK occurs in roughly a quarter to a half of patients fused to the pelvis (26% in the original Glattes series, 45.1% in a 679-patient multicentre cohort), while acute proximal junctional failure occurs in around 5-6% within the first 6 months (Hostin et al.). Female sex, advanced age and osteoporosis are recurrent risk factors in cohorts from North America, Europe and Asia, and large meta-analyses pooling these international series confirm osteoporosis (OR 1.58) and pelvic fixation (OR 2.08) as drivers.

Society and Guideline Guidance

There is no single randomised-trial-based clinical practice guideline that dictates PJK prevention; guidance is largely consensus and registry-informed:

Key practice principles with the strongest evidence are: individualised, age-adjusted alignment targets (avoid over-correction), proportional correction (GAP concept, acknowledging inconsistent external validation), proximal junction protection (ligamentous augmentation/hooks) and bone optimisation in osteoporotic patients.

Registry and Multicentre Evidence

Most high-quality PJK evidence comes from prospective multicentre deformity databases (e.g. the International Spine Study Group and European multicentre series) rather than national arthroplasty-style registries, because deformity volumes are comparatively low and constructs are heterogeneous. These collaboratives have produced the predictive models (Yagi), proportional scores (GAP) and age-adjusted target work that now guide practice globally. The lack of a dedicated high-volume PJK registry is itself a recognised gap.

Practice Variation

Practice varies internationally in the routine use of preventive adjuncts: prophylactic vertebroplasty/cement augmentation, transverse-process hooks and ligamentous "topping-off" devices are used selectively rather than universally, reflecting cost, availability and the absence of definitive randomised evidence. Thresholds for revising radiographic PJK also differ, with most centres reserving surgery for progressive deformity, intractable pain, instability or neurological compromise. Across health systems, complex deformity and PJK revision surgery is concentrated in tertiary specialist spine units with intraoperative neuromonitoring, advanced imaging and intensive care, with patients from regional or lower-resource settings typically transferred for complex revision.