Management and Risk Factors of Fracture-Type Proximal Junctional Kyphosis: Focus on Myelopathy and Intervention

Abstract

Study Design

Retrospective observational study.

Objectives

To identify radiographic and clinical predictors of fracture-type proximal junctional kyphosis (PJFx) and associated myelopathy (PJFx-M), with the goal of facilitating timely surgical intervention.

Methods

We retrospectively reviewed 438 patients with adult spinal deformity who underwent long-segment fusion surgery at our institution between 2013 and 2023. Risk factors for PJFx and PJFx-M were analyzed, focusing on demographic and radiographic parameters. In PJFx-M cases, we specifically assessed sliding length, screw protrusion, and superior articular facet fractures.

Results

PJFx occurred in 102 patients (23.3%). Significant risk factors for PJFx included a lower Hounsfield unit (HU) value at the upper instrumented vertebra (UIV) and reduced preoperative lumbar lordosis. PJFx-M was identified in 10 patients (2.3%). Radiographic predictors of PJFx-M included increased sliding length, screw protrusion into the upper vertebra, and bilateral superior articular facet fractures (all P < 0.001). Additionally, a fall within 1 month of symptom onset was frequently observed in PJFx-M cases. PJFx-M was not associated with demographic factors or body mass index.

Conclusions

Lower HU at the UIV and reduced preoperative lumbar lordosis were key predictors of PJFx. PJFx-M should be suspected in the presence of increased sliding length, screw protrusion into the upper vertebra, superior articular facet fractures, and a recent fall. These findings highlight the importance of vigilant postoperative radiographic and clinical monitoring to enable early detection and intervention, potentially preventing neurological deterioration.

Keywords

proximal junctional kyphosis proximal junctional failure adult spinal deformity myelopathy radiographic predictive factors

Background

Adult spinal deformity (ASD) is characterized by abnormal spinal alignment that develops in adulthood.¹ Patients with ASD experience significant back pain, radiculopathy, and gait disturbances as a result of coronal and sagittal misalignment.^2,3 These symptoms substantially impact quality of life and are important when determining the need for corrective surgery. Surgical interventions has been shown to improve health-related quality of life more effectively than non-surgical treatments, particularly in patients with severe deformity.⁴ Nevertheless, these procedures come with high complication rates, with revision rates reaching up to 47%.^5-7 These rates may be even higher with extended postoperative follow-up.

Proximal junctional kyphosis (PJK) is one of the most frequent complications following ASD surgery^8-10 and typically results from fractures at the upper instrumented vertebra (UIV) or the adjacent vertebrae or from disruption of the posterior ligament. The main cause of PJK is proximal junctional fracture (PJFx), which usually occurs within the first 3 months after surgery.^11,12 Identified risk factors include osteoporosis,¹³ overcorrection of alignment,¹⁴ and severe preoperative deformity,^15,16 although findings have been inconsistent and not fully understood.

In most cases, symptoms of PJFx are mild and repeat surgery is not required. However, in some cases, PJFx leads to myelopathy and paralysis, necessitating immediate surgical intervention. Despite prompt treatment, paralysis may persist and significantly impact quality of life.

This study aimed to clarify the risk factors for PJFx by analyzing a large dataset from a single spine center. It also sought to identify risk factors for myelopathy resulting from PJFx (PJFx-M) and potential risk factors, which has not been well characterized in previous studies.

Methods

We retrospectively reviewed all patients who underwent surgery for ASD at our hospital between January 1, 2013 and December 31, 2023. Institutional review board approval was secured, permitting patient enrollment and data collection. The inclusion criteria were age 21 years or older at the time of surgery, a minimum postoperative follow-up of 1 year, surgery that involved posterior instrumentation spanning 5 or more levels, and complete radiographic data available. At the UIV, bilateral pedicle screws were placed in all cases, resulting in a uniform screw density of two per vertebra. Screw trajectory, angulation, and insertion depth were determined intraoperatively by each surgeon based on anatomical landmarks and bone quality. The underlying causes identified included degenerative kyphosis/kyphoscoliosis, complications following lumbar surgery, and vertebral fractures. Any deformity caused by tumor or infectious disease was excluded.

Three-column osteotomy entailed use of pedicle subtraction osteotomy or vertebral column resection. PJK was defined as kyphosis exceeding 10° between the UIV and the 2 vertebrae above it (UIV+2), while PJFx was defined as PJK resulting from a vertebral fracture, with the specific angle termed the proximal junctional angle. Myelopathy resulting from PJFx was categorized as PJFx-M. The decision to pursue surgical vs conservative treatment for PJFx was made based on patient condition and shared decision-making between surgeon and patient. All patients who developed PJFx-M had initially received conservative management without immediate revision surgery.

To evaluate the influence of screw placement on PJFx and PJFx-M, we conducted a detailed radiographic analysis of pedicle screws at the upper instrumented vertebra (UIV). The following parameters were assessed on postoperative sagittal CT images: Screw trajectory angle in the sagittal plane, measured relative to the superior endplate. Positive values indicate cranial direction, negative values indicate caudal angulation. Screw tip–superior endplate distance (%): the vertical distance from the screw tip to the superior endplate, normalized to the total vertebral body height. Screw tip–anterior cortex distance (%): the distance from the screw tip to the anterior cortical margin of the vertebral body, normalized to its anteroposterior diameter. All parameters were measured on both right and left screws. For analysis, the mean values of bilateral screws were used to represent each patient.

We collected data for age, sex, body mass index (BMI), medical comorbidities, bone mineral density at the femoral neck (T-score), and details on use of medication for osteoporosis before and after surgery. Medical comorbidities included diabetes, renal dysfunction, and cerebrovascular, cardiovascular, and respiratory diseases. Radiographic measurements included the sagittal vertical axis (SVA), thoracic kyphosis (TK) from T4 to T10, thoracolumbar kyphosis from T10 to L2, lumbar lordosis from L1 to S1, sacral slope, pelvic incidence, and pelvic tilt (PT). Global tilt (GT) was defined as the sum of the PT and the C7 vertical tilt, the angle created by the intersection of 2 lines; the first line is drawn from the center of the C7 vertebra to the center of the sacral endplate, and the second line is drawn from the center of the femoral heads to the center of the sacral endplate. The positioning of the lower instrumented vertebra, whether fixed to the sacrum or not, was also examined. These parameters were assessed in a standing position before and 2 weeks after surgery and at the latest radiographic follow-up. Ideal spinal alignment was defined as a pelvic incidence–lumbar lordosis mismatch (PI–LL) angle within –10° to +10° according to the SRS-Schwab ASD classification.¹⁷ Referring to a previous report,¹⁸ we also measured the relative pelvic version (RPV; equal to the measured minus the ideal SS), the relative LL (RLL; equal to the measured minus the ideal LL), the lordosis distribution index (LDI; equal to the L4-S1 lordosis divided by the L1-S1 lordosis multiplied by 100), and the relative spinopelvic alignment (RSA; equal to the measured minus the ideal GT). The ideal values of SS, LL, and GT were defined according to previous reports that analyzed their relationship with PI using asymptomatic population records. Age was stratified into 2 subgroups: <60 years and 60 years or older. The GAP score was calculated by adding the scores for RPV, RLL, LDI, RSA, and the age factor, according to a previous report. The Hounsfield Unit (HU) value at the UIV was assessed in the midsagittal view using preoperative CT imaging.

In patients with PJFx, we analyzed the proximal junctional angle at the time of the last follow-up or immediately before repeat surgery. Additional assessments included examination of screw back-out into the caudal side disc or upper vertebra, the slip distance between the posterocranial point of the UIV and posterocaudal point of the UIV+1, and fractures of the superior articular process (none, unilateral, or bilateral, Figure 1).

Figure 1.

Mechanism and Radiographic Findings of Myelopathy Following Proximal Junctional Failure. Abbreviations: UIV, Upper Instrumented Vertebra; UIV+1, the Vertebra One Level Above the UIV; UIV+2, the Vertebra Two Levels Above the UIV

Patients were categorized into 2 groups, namely, those with PJFx [PJFx(+)] and those without (PJFx (−)), with further division of the PJFx(+) group into those with myelopathy (PJFx-M (+)) and those without myelopathy (PJFx-M (−)). To address our 2 main objectives—first, to identify risk factors for PJFx, and second, to explore which cases of PJFx progressed to PJFx-M—we performed 2 separate sets of analyses. Specifically, we compared PJFx(+) and PJFx(−) groups to identify PJFx risk factors, and then compared PJFx-M(+) and PJFx-M(−) within the PJFx group to assess predictors of myelopathy. Continuous variables were compared between groups using t-tests and categorical variables using chi-squared tests. Risk factor analysis for PJFx was performed using multivariate logistic regression with a stepwise forward procedure (P < 0.1 for entry).Missing values were handled using the last observation carried forward method. All statistical analyses were performed using IBM SPSS Statistics for Macintosh, version 25.0 (IBM Corp., Armonk, NY). A P-value of <0.05 was deemed statistically significant.

Results

A total of 438 patients with full preoperative and postoperative radiographic data available and a minimum 1 year of follow-up were included in the study. Table 1 shows the patient characteristics. The mean age at the time of surgery was 75.4 ± 7.5 years, and 368 patients (84.0%) were female. PJFx was identified in 102 patients (23.3%). Thirty-five (8.0%) of the patients with PJFx underwent revision surgery. Myelopathy caused by PJFx was treated with immediate surgery in 10 cases (2.3%).

Table 1.

Demographic and Clinical Data for Patients Who Underwent Surgery for Adult Spinal Deformity

Cases, n	438
Age at surgery (years)	73.1 ± 8.3
Male/female sex, n	70/368
Body mass index	22.5 ± 3.8
Levels fixed, n	7.4 ± 1.7
3CO (yes/no)	197/241
Fixed to sacrum (yes/no)	353/85
UIV (T6 or above/below T6)	14/424
PJFx (yes/no)	102/336
PJFx-failure (yes/no)	35/403
PJFx-M (yes/no)	10/438

Abbreviations: 3CO, three-column osteotomy; PJFx, proximal junction fracture; PJFx-failure, proximal junctional fracture requiring repeat surgery; PJFx-M, proximal junctional fracture resulting in myelopathy; UIV, upper instrumented vertebra.

The demographic characteristics of the patients in the PJFx(+) group and those in the PJFx(−) group are shown in Table 2. T-scores and HU value at UIV were significantly lower in the PJFx(+) group than in the PJFx(−) group (T-score −2.2 ± 1.0 vs −1.8 ± 0.9, P = 0.018; HU value at UIV 102.4 ± 32.2 vs 127.5 ± 47.3, P < 0.001). Among the screw-related parameters at the UIV, only tip-anterior cortex distance (expressed as a percentage of vertebral body height) was significantly smaller in the PJFx(+) group than in the PJFx(−) group (13.0 ± 10.5% vs 18.7 ± 11.1%, P < 0.001). However, there was no significant difference in sex, BMI, frequency of 3-column osteotomy, or UIV >T7 between the 2 groups. The relationship between each radiographic parameter and the incidence of PJFx was evaluated (Table 2). Preoperative LL and SS were significantly lower, and preoperative PT and PI-LL were significantly higher in PJFx(+) group (Pre LL -1.2 ± 20.1vs 7.2 ± 20.4, P < 0.001; Pre SS 11.6 ± 11.1 vs 14.7 ± 11.1, P = 0.030; pre PT 39.7 ± 10.1 vs 35.0 ± 13.1 P = 0.032; pre PI-LL 54.7 ± 19.2 vs 44.1 ± 18.6 P = 0.010). Change in SVA and LLvalues were significantly higher in the PJFx(+) group than in the PJFx(−) group (ΔSVA 111.8 ± 71.7 vs 90.3 ± 66.2, P = 0.020; ΔLL 36.5 ± 18.2 vs 28.6 ± 19.6, P = 0.020). The postoperative TK was significantly higher in the PJFx(+) group (post TK 41.1 ± 11.9 vs 34.6 ± 11.9, P = 0.003). The other parameters did not have a significant impact on the likelihood of PJFx. We investigated the risk factors for PJFx by using multivariate analysis. Forward stepwise logistic regression indicated that the preoperative LL and HU value at UIV were significant risk factors for PJFx.

Table 2.

Demographic and Radiographic Data for Patients With and Without Proximal Junction Fracture

	PJFx(+)	PJFx(−)	Univariate analysis	Multivariate analysis
	PJFx(+)	PJFx(−)	P-value	OR	95%CI	P-value
Cases, n	102	336	-
Age at surgery (years)	74.0 ± 8.4	72.8 ± 8.4	0.23
Male/female sex, n	11/91	59/277	0.12
Body mass index	22.7 ± 3.8	22.4 ± 3.8	0.50
Bone mineral density (T-score)	−2.2 ± 1.0	−1.8 ± 0.9	<0.001
Hyper-unit value at UIV	102.4 ± 32.2	127.5 ± 47.3	<0.001	0.997	0.994-0.999	0.015
Before use of agents for osteoporosis
None/SERM/denosumab/BP/TP/romosozumab	74/4/4/17/3/0	249/12/7/47/21/4	0.66
After use of osteoporosis drugs
None/SERM/denosumab/BP/TP/romosozumab	41/2/6/23/27/3	191/13/12/69/38/13	0.10
Etiology (degenerative/vertebral fracture/previous lumbar fusion)	55/21/26	213/50/73	0.21
History of any lumbar surgery (yes/no)	29/73	82/254	0.42
Levels fixed, n	7.5 ± 1.7	7.3 ± 1.6	0.41
3CO (yes/no)	51/51	146/190	0.26
UIV (T6 or above/below T6)	4/98	14/322	0.63
Fixed to sacrum (yes/no)	88/14	262/94	0.063
Screw angulation at UIV (°, vs upper endplate)	−1.5 ± 3.8	−0.7 ± 4.9	0.088
Tip-anterior cortex distance at UIV (% of VB)	13.0 ± 10.5	18.7 ± 11.1	<0.001
Tip–Upper endplate distance at UIV (% of VB height)	38.6 ± 18.8	37.6 ± 15.8	0.32
Pre SVA	144.6. ± 62.7	131.7 ± 69.8	0.14
Post SVA	32.8.4 ± 44.4	42.0 ± 46.0	0.115
ΔSVA	111.8 ± 71.7	90.3 ± 66.2	0.020
Pre LL	−1.2 ± 20.1	7.2 ± 20.4	<0.001	0.993	0.988-0998	0.010
Post LL	35.3 ± 11.5	35.7 ± 11.9	0.81
ΔLL	36.5 ± 18.2	28.6 ± 19.6	<0.001
Pre TLK	23.7 ± 15.1	20.7 ± 16.3	0.31
Post TLK	15.4 ± 10.7	14.3 ± 9.8	0.57
Pre TK	30.7 ± 18.9	28.1 ± 17.6	0.42
Post TK	41.1 ± 11.9	34.6 ± 11.9	0.003
Pre SS	11.6 ± 11.1	14.7 ± 11.1	0.030
Post SS	22.0 ± 9.3	23.9 ± 8.8	0.083
PI	53.1 ± 11.9	49.9 ± 11.8	0.13
Pre PT	39.7 ± 10.1	35.0 ± 13.1	0.032
Post PT	28.6 ± 11.3	25.2 ± 12.0	0.12
Pre PI–LL	54.7 ± 19.2	44.1 ± 18.6	0.010
Post PI–LL	17.8 ± 13.6	15.5 ± 14.1	0.20
Gap score	8.9 ± 3.0	8.4 ± 3.0	0.36
RPV	2.4 ± 0.9	2.1 ± 1.1	0.051
RLL	2.3 ± 1.1	2.1 ± 1.0	0.52
LDI	1.1 ± 1.2	1.3 ± 1.1	0.27
RSA	2.2 ± 1.2	2.0 ± 1.2	0.31

Abbreviations: 3CO, 3 column osteotomy; BP, bisphosphonate; LL, lumbar lordosis; PI, pelvic incidence; PI–LL, pelvic incidence–lumbar lordosis mismatch; PJFx, proximal junctional fracture; PT, pelvic tilt; SERM, Selective estrogen receptor modulator; SS, sacral slope; SVA, sagittal vertical axis; TLK, thoracolumbar kyphosis; TK, thoracic kyphosis; TP, teriparatide; UIV, upper instrumented vertebra; RPV, relative pelvic version; RLL, relative lumbar lordosis, LDI, lordosis distribution index, RSA, relative spinopelvic alignment; VB, vertebral body.

The demographic characteristics of the patients in the PJFx-M(+) group and those in the PJFx-M(−) group are shown in Table 3. A proportion of patients with a history of any lumbar surgery was significantly greater in the PJFx-M(+) group (100% vs 21.5%, P < 0.001). Change in SS and PT were significantly less in the PJFx-M(+) group than in the PJFx-M(−) group (ΔSS 3.3 ± 5.5 vs 11.1 ± 10.2, P = 0.021; ΔPT -4.6 ± 4.4 vs −11.0 ± 10.7, P = 0.005).

Table 3.

Demographic and Radiographic Data for the Group With PJFx-M and the Group Without PJFx-M

	PJFx-M(+)	PJFx-M(−)	P-value
Cases, n	10	92	-
Age at surgery (years)	77.1 ± 7.9	73.7 ± 8.3	0.22
Male/female sex, n	1/9	10/82	0.97
Body mass index	24.9 ± 3.2	22.4 ± 3.8	0.041
Bone mineral density (t-score)	−2.1 ± 1.2	−2.4 ± 0.9	0.52
Hounsfield unit value at UIV	116.4 ± 27.0	100.4 ± 32.1	0.14
History of any lumbar surgery (yes/no)	10/0	20/72	<0.001
Levels fixed, n	7.6 ± 1.0	7.4 ± 1.8	0.64
3CO (yes/no)	7/3	45/47	0.32
UIV (T6 or above/below T6)	1/8	4/88	0.38
Screw angulation at UIV (°, vs upper endplate)	−4.3 ± 4.6	−1.1 ± 3.5	0.062
Tip-anterior cortex distance at UIV (% of VB)	8.4 ± 9.5	13.6 ± 10.6	0.13
Tip–Upper endplate distance at UIV (% of VB height)	46.6 ± 11.9	37.7 ± 19.2	0.056
Pre SVA	140.0 ± 34.8	145.2 ± 64.6	0.80
Post SVA	56.8 ± 45.0	30.3 ± 43.3	0.11
ΔSVA	83.2 ± 41.0	115.0 ± 73.8	0.19
Pre LL	3.4 ± 21.8	−1.8 ± 19.7	0.48
Post LL	31.2±10.7	35.5 ± 11.6	0.25
ΔLL	27.8 ± 18.2	37.4 ± 18.1	0.14
Pre TLK	24.8 ± 15.6	22.7 ± 15.7	0.70
Post TLK	14.9 ± 7.9	15.5 ± 10.9	0.85
Pre TK	22.2 ± 17.2	27.2 ± 18.7	0.41
Post TK	37.4 ± 8.5	38.0 ± 13.5	0.47
Pre SS	16.1 ± 5.6	10.8 ± 11.5	0.16
Post SS	19.4 ± 7.1	22.1 ± 9.5	0.29
ΔSS	3.3 ± 5.5	11.1 ± 10.2	0.002
PI	52.3 ± 7.3	52.9 ± 11.4	0.87
Pre PT	36.1 ± 7.5	40.5 ± 10.9	0.15
Post PT	31.6 ± 7.9	29.5 ± 12.5	0.52
ΔPT	−4.6 ± 4.4	−11.0 ± 10.7	0.005
Pre PI-LL	45.9 ± 23.3	54.3 ± 19.4	0.30
Post PI-LL	21.4 ± 13.7	12.7 ± 13.9	0.11
Gap score	10.4 ± 2.8	8.6 ± 3.0	0.08
RPV	2.8 ± 0.4	2.3 ± 1.0	0.15
RLL	2.7 ± 0.5	2.2 ± 1.2	0.17
LDI	1.4 ± 1.3	1.0 ± 1.2	0.36
RSA	2.2 ± 1.2	2.0v1.2	0.34

Abbreviations: 3CO, 3 column osteotomy; GT, global tilt; LL, lumbar lordosis; PJFx-M, proximal junctional fracture that developed into myelopathy; PI, pelvic incidence; PT, pelvic tilt; SS, sacral slope; SVA, sagittal vertical axis; TK, thoracic kyphosis; TLK, thoracolumbar kyphosis; UIV, upper instrumented vertebra; VB, vertebral body; RPV, relative pelvic version; RLL, relative lumbar lordosis, LDI, lordosis distribution index, RSA, relative spinopelvic alignment.

We also compared 4 additional parameters between the PJFx-M(+) and PJFx-M(−) groups (Table 4). The slip distance was significantly greater in the PJFx-M(+) group (4.7 ± 2.1 vs 0.6 ± 1.1, P < 0.001), as was the rate of screw protrusion into the upper vertebra (44.4% vs 4.3% P < 0.001) and the rate of fracture of both superior articular processes (78.0% vs 3.4%, P < 0.001). However, there was no significant difference between the groups in the proximal junctional angle.

Table 4.

Radiographic Indicators of PJFx-M

	PJFx-M(+)	PJFx-M(−)	P-value
Cases, n	10	92	-
Slip length (mm)	4.7 ± 2.7	0.6 ± 1.1	<0.001
Superior articular process fracture (none/unilateral/ bilateral)^a, n	2/0/8	44/12/2	<0.001
Screw protrusion (none/endplate/cranial vertebra), n	4/2/4	61/27/4	<0.001
Proximal junctional angle, degrees	30.0 ± 8.2	27.4 ± 9.7	0.48

Abbreviations: PJFx-M, proximal junctional fracture that developed into myelopathy.

^aComputed tomography scans were obtained in only 58 patients in the PJFx-M(−) group. PJFx-M, proximal junctional fracture with myelopathy.

Discussion

There have been few studies on the classification of PJK. Yagi et al.¹² classified PJK based on type and grade, including spondylolisthesis, providing a simple and effective way for clinicians to discuss the types of PJK. In this study, we focused specifically on fracture-type PJK (PJFx), also known as type 2 PJK, given that most cases of PJK are caused by fractures at the UIV or the vertebra immediately above (UIV+1). Severe PJK is typically caused by these fractures, which usually occur within the first 3 months post-surgery.¹⁹ We reviewed all our cases of PJFx and noted inconsistency in the criteria used for revision, which we attributed to decisions being influenced by patient resilience or surgeon preference. Therefore, we further analyzed cases of myelopathy that required immediate revision surgery, where the timing of intervention was critical.

First, we analyzed risk factors for PJFx. A lower preoperative LL and HU value at UIV was identified as a significant risk factor for PJK, whereas age, sex, and BMI did not show any correlation with the incidence of PJK. Previous studies have identified older age,^20-22 osteoporosis,¹³ a high BMI,²¹ and sarcopenia²³ to be risk factors for PJK following surgery for correction of ASD. Although age ≥50 years is often cited as a risk factor, the average age of our patients was ≥70 years, which may explain why age did not significantly affect the PJK rate. Another study found that patients with low bone mineral density, including those with osteoporosis or osteopenia, had a 30.9% increase in risk of developing PJK,¹⁹ which aligns with our present findings. Yagi et al reported that the incidence of type 2 PJK in patients with osteopenia was significantly lower in those treated with teriparatide than in controls (4.6% vs 15.2%, P = 0.02).²⁴ In our study, only a limited number of patients received teriparatide preoperatively and postoperatively, so we could not examine its potential protective effect. Sacral fixation was assessed as a potential risk factor for PJFx. Although the incidence of PJFx was slightly higher in patients with sacral fixation, this difference was not statistically significant. Therefore, sacral fixation was not treated as a stratified variable in our multivariate analysis. Nonetheless, rigid distal fixation may biomechanically contribute to increased stress at the proximal junction, and this possibility should be considered in future studies.

In terms of radiographic parameters, forward stepwise logistic regression indicated that the preoperative LL was a significant risk factor for PJFx. Maruo et al found that a preoperative TK of ≥30° was a predictor of PJK, and a meta-analysis identified a pre TK >40° as a risk factor for PJK.²⁵ Other studies have also found that a preoperative proximal junctional angle of ≥10° and pelvic incidence >55° are associated with an increased risk of PJK.¹⁴ Helgeson et al noted that postoperative changes in SVA of ≥50 mm increased the incidence of PJK,¹⁵ while other studies found that a postoperative SVA of <50 mm significantly increased the risk of proximal junctional failure.²⁶ A change in lumbar lordosis of ≥30° has also been identified as a risk factor for PJK.¹⁴ However, analysis of the risk factor of fracture type PJFx is limited.

We investigated risk factors of PJFx-M. A change in SS and PT were significantly lower in PJFx-M(+) group and all patients had a history of previous lumbar fixation surgery (Table 5). We estimated that previous fixation surgery limited pelvic retroversion adjustment, which can lead to worse outcome. 10 cases of myelopathy following proximal junctional failure further and found that 2 were caused by massive herniation and 8 by subluxation between UIV and UIV+1. All 8 patients with subluxation had experienced a fall within 1 month prior to the gradual onset of paralysis. Radiographically, these patients showed an increased sliding length, higher rates of screw protrusion into UIV+1, and fractures of the superior articular facet of UIV. We hypothesize that loosening of the proximal screw weakens the superior articular facet, and a fall leads to its fracture, causing instability between UIV and UIV+1 that results in subluxation. If a fall occurs, immediate repeat surgery is recommended if increased sliding length, screw protrusion into UIV+1, or superior articular facet fractures are detected, considering that these conditions can lead to myelopathy (Figure 1).

Table 5.

Characteristics of Individual Patients With PJFx-M

Age (years)/Sex	BMI	T-score	UIV	Surgical procedure	Previous lumbar surgery	Fall episode	Pathogenesis	PJA	Slip length (mm)	Screw protrusion	SAP fracture	Outcome
64, F	26.4	−0.9	T5	L3PSO	L4/5/sTLIF	Yes	Instability	30	5.5	Endplate	Both	Wheelchair
78, F	24.1	−0.2	T10	L2PSO	L4/5/sTLIF	Yes	Instability	42	4.7	UIV+1	Both	T-cane
69, F	30.3	−2.3	T11	L3PSO	L2/3/4/5PLIF	Yes	Instability	24	4.2	None	Both	Independent gait
80, F	22.0	−2.9	T12	L3PSO	L4/5/sTLIF	Yes	Instability	42	4.4	Endplate	Both	Independent gait
73, F	21.0	−3.2	T10	L3PSO	L4/5PLF	Yes	Instability	18	5.9	UIV+1	Both	Independent gait
73, F	22.7	−1.6	T11	L3PSO	L4/5TLIF	Yes	Instability	25	5.1	UIV+1	Both	T-cane
89, F	24.8	−1.9	Th9	Multiple TLIF	L3/4/5TLIF	Yes	Instability	35	7.8	UIV+1	Both	T-cane
89, F	24.7	−3.6	Th11	Multiple TLIF	L1/2,4/5TLIF,L2/3.3/4OLIF	No	Hernia	25	0	None	None	Pick-up walker
79, M	23.1	−1.4	Th11	L3PSO	L4/5/sTLIF	Yes	Hernia	26	3.9	None	None	T-cane
77, F	30.4	−2.3	Th9	Multiple TLIF	L3-STLIF	Yes	Instability	27	4.9	None	Both	Pick-up walker

Abbreviations: BMI, bone mass index; UIV, upper instrumented vertebra; PJA, proximal junctional angle; SAP, superior articular process; PSO, pedicle subtraction osteotomy; TLIF, transpedicular lumbar interbody fusion; OLIF, oblique lateral lumbar interbody fusion.

Several studies have explored the use of cement augmentation to prevent PJK and proximal junctional failure in patients with osteoporosis.^27-30 1 study found that prophylactic vertebroplasty at UIV and UIV+1 reduced the incidence of PJK by 8% and that of proximal junctional failure by 5% at 6 months.²⁷ Spinal hooks have also been used to prevent PJK, with 1 study showing a 0% incidence of PJK in a hook group compared with 29.6% in a pedicle screw group.¹⁵ Another technique involves reconstructing the posterior ligamentous complex using Mersilene tape to reduce loading stress, with a systematic review showing a lower incidence of PJK and a reduction in disc pressure in comparison with use of pedicle screws.^31,32 These methods reduce violations of the superior articular facet at the UIV, potentially lowering the risk of myelopathy.

To date, no studies have clearly identified risk factors for myelopathy or the optimal timing of intervention in patients with PJFx. In our study, all patients with PJFx-M had a history of prior lumbar fusion surgery, and changes in sacral slope (SS) and pelvic tilt (PT) were significantly smaller in these patients. These findings suggest that prior lumbar fusion may restrict compensatory pelvic alignment, thereby contributing to the onset of myelopathy. As for the timing of intervention, increased sliding length of UIV+1, screw protrusion into UIV+1, and bilateral fractures of the UIV’s superior articular process were key radiographic predictors of impending myelopathy. Additionally, a fall episode was observed within 1 month prior to symptom onset in many patients. These findings underscore the importance of vigilant postoperative radiographic and clinical surveillance to allow early detection and timely intervention, potentially preventing neurological deterioration.

This study has several limitations. First, its retrospective design inherently limits causal inference. Second, all data were obtained from a single institution, which may affect the generalizability of the findings. Finally, the number of patients with PJFx-M was relatively small, which may limit the statistical power for detecting significant differences.

Conclusion

In this study, lower preoperative LL and lower HU value at UIV were identified as key risk factors for PJFx. Age and BMI did not show a significant correlation in our cohort. If a fall occurs in a patient with PJFx, immediate intervention is recommended when increased sliding length, screw protrusion into the UIV+1, or superior articular facet fractures is present, given that these factors can lead to myelopathy.

Footnotes

ORCID iDs

Atsuyuki Kawabata

Takashi Hirai

Toshitaka Yoshii

Ethical Approval

The study was approved by the ethics committee of all institutions involved. Informed consent was waived by the above ethics committee as the present retrospective cohort study involved already existing data and records at the time of investigation and did not retain personal identifiers of the gathered information. The ethical committee, Kudanzaka Hospital; 2019-5.

Author Contributions

Atsuyuki Kawabata analysed the data and wrote the original draft. Yu Matsukura, Takuya Oyaizu, Hiroaki Onuma, and Toshitaka Yoshii conceived of, reviewed, and edited the paper. Keisuke Hide, Takayuki Motoyoshi, Jun Hashimoto, Kazuo Kusano, Kazuyuki Otani, Satoru Egawa, Kentaro Yamada and Shigeo Shindo acquired the data. Shingo Morishita performed the investigation. Toshitaka Yoshii validated the paper. Kazuyuki Otani, Shigeo Shindo, and Toshitaka Yoshii supervised the research. All authors contributed to the writing of the final manuscript. All authors approved the manuscript to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Kim

Yang

Chang

, et al. Adult spinal deformity: current concepts and decision-making strategies for management. Asian Spine J. 2020;14(6):886-897. doi:10.31616/asj.2020.0568

Smith

Shaffrey

Ames

Lenke

. Treatment of adult thoracolumbar spinal deformity: past, present, and future. J Neurosurg Spine. 2019;30(5):551-567. doi:10.3171/2019.1.SPINE18149

Lord

Ayres

Woo

, et al. The impact of global alignment and proportion score and bracing on proximal junctional kyphosis in adult spinal deformity. Glob Spine J. 2023;13(3):651-658. doi:10.1177/21925682211001812

Scheer

Hostin

Robinson

, et al. Operative management of adult spinal deformity results in significant increases in QALYs gained compared to nonoperative management: analysis of 479 patients with minimum 2-year follow-up. Spine. 2018;43(5):339-347. doi:10.1097/BRS.0000000000001626

Charosky

Guigui

Blamoutier

Roussouly

Chopin

Study Group on Scoliosis . Complications and risk factors of primary adult scoliosis surgery: a multicenter study of 306 patients. Spine. 2012;37(8):693-700. doi:10.1097/BRS.0b013e31822ff5c1

Smith

Sansur

Donaldson

, et al. Short-term morbidity and mortality associated with correction of thoracolumbar fixed sagittal plane deformity: a report from the Scoliosis Research Society Morbidity and Mortality Committee. Spine. 2011;36(12):958-964. doi:10.1097/BRS.0b013e3181eabb26

Pichelmann

Lenke

Bridwell

Good

O’Leary

Sides

. Revision rates following primary adult spinal deformity surgery: six hundred forty-three consecutive patients followed-up to twenty-two years postoperative. Spine. 2010;35(2):219-226. doi:10.1097/BRS.0b013e3181c91180

Smith

Annis

Lawrence

Daubs

Brodke

. Acute proximal junctional failure in patients with preoperative sagittal imbalance. Spine J. 2015;15(10):2142-2148. doi:10.1016/j.spinee.2015.05.028

Kim

Lenke

Shaffrey

Van Alstyne

Skelly

. Proximal junctional kyphosis as a distinct form of adjacent segment pathology after spinal deformity surgery: a systematic review. Spine. 2012;37(22 Suppl):S144-S164. doi:10.1097/BRS.0b013e31826d611b

10.

Pinter

Mikula

Townsley

, et al. Lower Hounsfield units and severe multifidus sarcopenia are independent predictors of increased risk for proximal junctional kyphosis and failure following thoracolumbar fusion. Spine. 2023;48(4):223-231. doi:10.1097/BRS.0000000000004517

11.

Yagi

Fujita

Okada

, et al. Fine-tuning the predictive model for proximal junctional failure in surgically treated patients with adult spinal deformity. Spine. 2018;43(11):767-773. doi:10.1097/BRS.0000000000002415

12.

Yagi

Akilah

Boachie-Adjei

. Incidence, risk factors and classification of proximal junctional kyphosis: surgical outcomes review of adult idiopathic scoliosis. Spine. 2011;36(1):E60-E68. doi:10.1097/BRS.0b013e3181eeaee2

13.

Yagi

Fujita

Tsuji

, et al. Low bone-mineral density is a significant risk for proximal junctional failure after surgical correction of adult spinal deformity: a propensity score-matched analysis. Spine. 2018;43(7):485-491. doi:10.1097/BRS.0000000000002355

14.

Yuan

Zeng

Chen

Zhang

Mai

. Degenerative lumbar scoliosis patients with proximal junctional kyphosis have lower muscularity, fatty degeneration at the lumbar area. Eur Spine J. 2021;30(5):1133-1143. doi:10.1007/s00586-020-06394-8

15.

Helgeson

Shah

Newton

, et al. Evaluation of proximal junctional kyphosis in adolescent idiopathic scoliosis following pedicle screw, hook, or hybrid instrumentation. Spine. 2010;35(2):177-181. doi:10.1097/BRS.0b013e3181c77f8c

16.

Miller

Neuman

Jain

, et al. An assessment of frailty as a tool for risk stratification in adult spinal deformity surgery. Neurosurg Focus. 2017;43(6):E3. doi:10.3171/2017.10.FOCUS17472

17.

Schwab

Patel

Ungar

Farcy

Lafage

. Adult spinal deformity-postoperative standing imbalance: how much can you tolerate? An overview of key parameters in assessing alignment and planning corrective surgery. Spine. 2010;35(25):2224-2231. doi:10.1097/BRS.0b013e3181ee6bd4

18.

Yilgor

Sogunmez

Boissiere

, et al. Global alignment and proportion (GAP) score: development and validation of a new method of analyzing spinopelvic alignment to predict mechanical complications after adult spinal deformity surgery. J Bone Joint Surg Am. 2017;99:1661-1672.

19.

Yagi

King

Boachie-Adjei

. Incidence, risk factors, and natural course of proximal junctional kyphosis: surgical outcomes review of adult idiopathic scoliosis. Minimum 5 years of follow-up. Spine. 2012;37(17):1479-1489. doi:10.1097/BRS.0b013e31824e4888

20.

Hyun

Lee

Park

Kim

Jahng

Kim

. Proximal junctional kyphosis and proximal junctional failure following adult spinal deformity surgery. Korean J Spine. 2017;14(4):126-132. doi:10.14245/kjs.2017.14.4.126

21.

Bridwell

Lenke

Cho

, et al. Proximal junctional kyphosis in primary adult deformity surgery: evaluation of 20 degrees as a critical angle. Neurosurgery. 2013;72(6):899-906. doi:10.1227/NEU.0b013e31828bacd8

22.

Kim

Bridwell

Lenke

Glattes

Rhim

Cheh

. Proximal junctional kyphosis in adult spinal deformity after segmental posterior spinal instrumentation and fusion: minimum five-year follow-up. Spine. 2008;33(20):2179-2184. doi:10.1097/BRS.0b013e31817c0428

23.

Hyun

Kim

Rhim

. Patients with proximal junctional kyphosis after stopping at thoracolumbar junction have lower muscularity, fatty degeneration at the thoracolumbar area. Spine J. 2016;16(9):1095-1101. doi:10.1016/j.spinee.2016.05.008

24.

Yagi

Ohne

Konomi

, et al. Teriparatide improves volumetric bone mineral density and fine bone structure in the UIV+1 vertebra, and reduces bone failure type PJK after surgery for adult spinal deformity. Osteoporos Int. 2016;27(12):3495-3502. doi:10.1007/s00198-016-3676-6

25.

Maruo

Inoue

, et al. Predictive factors for proximal junctional kyphosis in long fusions to the sacrum in adult spinal deformity. Spine. 2013;38(23):E1469-E1476. doi:10.1097/BRS.0b013e3182a51d43

26.

Manini

Zhang

Gao

Liu

YuXiang

. Proximal junction kyphosis in adult scoliosis: best postoperative radiological predictors-a retrospective cohort study. Adv Orthop. 2022;2022:9814416. doi:10.1155/2022/9814416

27.

Martin

Skolasky

Mohamed

Kebaish

. Preliminary results of the effect of prophylactic vertebroplasty on the incidence of proximal junctional complications after posterior spinal fusion to the low thoracic spine. Spine Deform. 2013;1(2):132-138. doi:10.1016/j.jspd.2013.01.005

28.

Maruo

Racine

, et al. Proximal junctional kyphosis and clinical outcomes in adult spinal deformity surgery with fusion from the thoracic spine to the sacrum: a comparison of proximal and distal upper instrumented vertebrae. J Neurosurg Spine. 2013;19(3):360-369. doi:10.3171/2013.5.SPINE12737

29.

Han

Hyun

Kim

Jahng

Kim

. Comparative study between cobalt chrome and titanium alloy rods for multilevel spinal fusion: proximal junctional kyphosis more frequently occurred in patients having cobalt chrome rods. World Neurosurg. 2017;103:404-409. doi:10.1016/j.wneu.2017.04.031

30.

Aydogan

Ozturk

Karatoprak

Tezer

Aksu

Hamzaoglu

. The pedicle screw fixation with vertebroplasty augmentation in the surgical treatment of the severe osteoporotic spines. J Spinal Disord Tech. 2009;22(6):444-447. doi:10.1097/BSD.0b013e31818e0945

31.

Doodkorte

RJP

Roth

Arts

Lataster

LMA

van Rhijn

Willems

. Biomechanical comparison of semirigid junctional fixation techniques to prevent proximal junctional failure after thoracolumbar adult spinal deformity correction. Spine J. 2021;21(5):855-864. doi:10.1016/j.spinee.2021.01.017

32.

Doodkorte

RJP

Vercoulen

TFG

Roth

de Bie

Willems

. Instrumentation techniques to prevent proximal junctional kyphosis and proximal junctional failure in adult spinal deformity correction-a systematic review of biomechanical studies. Spine J. 2021;21(5):842-854. doi:10.1016/j.spinee.2021.01.011