Reverse pedicle subtraction osteotomy for iatrogenic hyperlordosis in a scoliosis patient: illustrative case

Abstract

BACKGROUND

Overcorrection of lumbar lordosis (LL) during deformity surgery can create iatrogenic hyperlordosis with negative sagittal balance, a rare but disabling complication with no standard treatment.

OBSERVATIONS

A 21-year-old woman with adolescent idiopathic scoliosis had undergone T3–L5 fusion, revision L4–5 transforaminal lumbar interbody fusion (TLIF) with carbon fiber instrumentation due to a nickel allergy, and L4–S1 anterior lumbar interbody fusion with L3–pelvis fusion for pseudarthrosis. She developed severe low back and bilateral groin pain that worsened with standing. Imaging demonstrated exaggerated LL (L1–S1 82° vs pelvic incidence 52°), a negative sagittal vertical axis (–96 mm), and a PI-LL mismatch of –30°, consistent with global sagittal imbalance and hip impingement. She underwent revision L1–S1 fusion with bilateral iliac fixation, reverse pedicle subtraction osteotomy (PSO) at L4, and L3–4 TLIF. Postoperatively, LL decreased to 60° with more physiological segmental distribution and sagittal parameters improved toward neutral, with resolution of her symptoms.

LESSONS

Iatrogenic lumbar hyperlordosis with negative sagittal balance can cause debilitating spinopelvic mismatch. Careful attention to patient-specific spinopelvic targets is essential during deformity correction, and reverse PSO is a useful strategy to reduce fixed hyperlordosis, restore alignment, and alleviate symptoms.

https://thejns.org/doi/10.3171/CASE25944

Surgical correction of adult spinal deformity often centers on restoring harmonious sagittal alignment, particularly through augmenting lumbar lordosis (LL) to compensate for global imbalance in the setting of degenerative or iatrogenic flatback.¹ The correlation between positive sagittal malalignment and clinical disability is well established, with normative alignment goals such as sagittal vertical axis (SVA) < 5 cm and pelvic incidence (PI)–LL mismatch within ± 10° linked to improved outcomes.^2,3 Spinal osteotomies, most notably pedicle subtraction osteotomies (PSOs), are frequently used to achieve these corrections, especially in rigid deformities.⁴

In contrast, pathological hyperlordosis and negative sagittal imbalance are rare, underrecognized entities. Excessive LL beyond the requirements dictated by an individual’s PI may result in a negative PI-LL mismatch and fixed negative SVA. The resulting biomechanics impose excessive strain on the posterior elements and can cause lumbosacral impingement, compensatory pelvic anteversion, and pain. Although such conditions are less frequently encountered than their kyphotic counterparts, they are no less functionally limiting.

Overcorrection of lumbar alignment has also been implicated in the development of proximal junctional kyphosis and junctional failure, particularly in long fusions where excessive lordosis creates abrupt transitions to the unfused thoracic spine.⁵ While standard osteotomies are designed to increase lordosis, few reports have addressed intentional reduction of lordosis via reverse osteotomies. Lewis et al. described a reverse Smith-Petersen osteotomy to address fixed hyperlordosis, and more recently, Faundez et al. reported a posterior opening wedge osteotomy to correct junctional kyphosis arising from pathological lumbar hyperlordosis.^5,6 These limited reports underscore the novelty, biomechanical complexity, and technical challenges of performing kyphosing osteotomies for sagittal realignment.

We present a case of iatrogenic hyperlordosis with disabling mechanical symptoms, including apparent hip impingement, successfully managed via a reverse PSO combined with revision fusion and transforaminal lumbar interbody fusion (TLIF). This case adds to the sparse literature on lordosis-reducing interventions and highlights the importance of individualized alignment targets.⁷

Illustrative Case

A 21-year-old female with adolescent idiopathic scoliosis presented with worsening mechanical low back pain and anterior groin symptoms. Her surgical history included T3–L4 posterior spinal fusion for scoliosis at age 14, followed by a revision extending the fusion to L5 for curve progression. At age 16, she underwent an L4–5 TLIF with carbon fiber instrumentation due to a documented nickel allergy. At age 20, she underwent an anterior lumbar interbody fusion from L4 to S1 with posterior instrumentation from L3 to the pelvis to address symptomatic pseudarthrosis.

Postoperatively, she developed worsening pain with standing and walking, described as axial lumbar discomfort and deep bilateral groin pain. These symptoms were relieved by sitting. On examination, she had an exaggerated lumbar sway with a hyperlordotic posture. Neurological examination was normal. Seated hip range of motion was painless.

CT showed anterior column support through L4–S1 cages with solid interbody fusion (Fig. 1). Standing long-cassette radiographs demonstrated significant sagittal overcorrection with a PI of 52°, LL (L1–S1) of 82°, C2 pelvic angle (C2PA) of −13°, and negative SVA of –96 mm (Fig. 2). Notably, the thoracic curve was relatively flat, and there was a fixed compensatory pelvic anteversion (pelvic tilt –8°, sacral slope 60°), consistent with posterior truncal shift and negative balance (Fig. 2). Segmental analysis showed L1–4 lordosis of 46° and L4–S1 lordosis of 54°, indicating excessive curvature distributed across both arcs.

FIG. 1.

Sagittal CT image demonstrating solid interbody fusion across L4–S1.

FIG. 2.

Preoperative standing EOS radiographs demonstrating hyperlordosis and negative sagittal balance.

We hypothesized that the patient’s groin symptoms represented iatrogenic femoroacetabular impingement due to excessive lordosis. With the pelvis fixed in anteversion and the lumbar spine overlordotic, the acetabular rim likely abutted the anterior femoral neck in extension. Her symptoms improved with sitting, consistent with unloading of the impingement vector.

Given her fixed deformity and disabling symptoms, she underwent revision posterior spinal fusion from L1 to S1 with bilateral iliac fixation. Prior instrumentation was removed except for carbon fiber S1 screws, which were retained due to hardware limitations. Freehand pedicle screw instrumentation was then performed at all levels. Iliac screws were placed bilaterally.

A reverse PSO was then performed at L4 (Fig. 3). Bilateral foraminotomies at L3–4 and L4–5 were completed, followed by revision laminectomy and removal of L4 pedicles. Unlike in a standard PSO, an opening wedge technique was used to decrease segmental lordosis. The L3–4 disc was excised, and a TLIF was performed with a titanium cage, autograft, and bone morphogenetic protein (BMP). Controlled reduction was then applied using temporary rods to progressively open the osteotomy and generate kyphosis. Intraoperative monitoring and direct visualization were used to assess L3 and L4 root tension. To reduce stretch, the final rod was intentionally undercontoured and compressed after fixation, giving the nerve roots additional slack. Posterolateral fusion from L1 to S1 was finally performed with autograft, allograft, and BMP. Plastic surgery assisted with complex wound closure, and intraoperative cultures were obtained (negative for infection).

FIG. 3.

Three-dimensional CT reconstructions demonstrating the reverse PSO with interval TLIF at L3–4.

The patient tolerated the procedure well. Early postoperative imaging (Fig. 4B and E) demonstrated improved alignment: LL was reduced to 60°, with L1–4 lordosis decreasing from 46° to 25° and L4–S1 lordosis from 54° to 45°. C2PA improved from −13° to –3°, and SVA was corrected from –96 mm to –58 mm (Table 1). Her immediate postoperative course was notable for acute blood loss anemia requiring one unit of packed red blood cells and transient bilateral thigh numbness that improved with 24 hours of postoperative dexamethasone. She cleared physical therapy and was medically cleared for discharge on postoperative day 4. Six-week postoperative imaging (Fig. 4C and F) demonstrated stability of correction, with L1–S1 lordosis at 60°, L1–4 at 25°, and L4–S1 at 46°. Pelvic tilt shifted to –5° (near neutral) with a sacral slope 57° and apex centered at mid-L4—a favorable outcome highlighting the physiological redistribution of lumbar curvature (Table 1), along with subjective resolution of all preoperative imbalance complaints.

FIG. 4.

Preoperative (A and D), immediate postoperative (B and E), and 6-week postoperative (C and F) standing lateral radiographs demonstrating improved sagittal alignment.

TABLE 1.

Comparison of pre- and postoperative alignment parameters relative to ideal

	Preop	Immediate Postop	6 Wks Postop	Ideal
Pelvic parameters
PT, °	−8	2	−5	—
SS, °	60	50	57	—
PI, °	52	52	52	—
Sagittal balance
SVA, mm	−96	−58	−52	~0
Lumbar alignment
L1–S1, °	82	60	60	—
L1–4, °	46	25	25	—
L4–S1, °	54	45	46	—
Thoracic alignment
T1–12, °	40	35	35	—
Proximal (T1–5), °	37	38	39	—
Distal (T5–12), °	1	2	1	—
Pelvic angles
L1PA, °	−3	6	2	7
T4PA, °	−19	−7	−9	3
C2PA, °	−13	−3	−5	8
Overall
PI-LL, °	−30	−8	−8	0
T4–L1 PA, °	16	−13	−11	−1 to 3

L1PA = L1 pelvic angle; PT = pelvic tilt; SS = sacral slope; T4PA = T4 pelvic angle; — = no established normative value.

Informed Consent

The necessary informed consent was obtained in this study.

Discussion

Observations

This case illustrates a rare but clinically significant complication of spinal deformity surgery: sagittal overcorrection into hyperlordosis with resultant mechanical symptoms. The prevailing paradigm in deformity correction emphasizes restoration of LL, particularly in adult patients with degenerative kyphosis. However, this report demonstrates that excessive lordosis relative to pelvic morphology can be equally deleterious, producing marked negative sagittal balance and iatrogenic hip impingement in a young patient with prior long fusion. Schwab et al. and Glassman et al. underscored the importance of PI-LL matching for achieving optimal functional outcomes, with deviation beyond ± 10° associated with worse pain and function.^2,3

Our patient had a PI-LL mismatch of –30° and a markedly negative SVA, resulting in compensatory pelvic retroversion and disabling symptoms. Hyperlordosis concentrated mechanical load onto the posterior elements, potentially causing facet joint impingement or spinous process abutment. Moreover, fixed retroversion and exaggerated lumbar curvature may predispose to femoroacetabular impingement, particularly in patients with otherwise normal hip morphology. Additionally, our patient’s preoperative segmental distribution was L1–4 lumbar lordosis 46° and L4–S1 lumbar lordosis 54°, reflecting excess curvature in both arcs relative to PI-based expectations and a near-even (rather than distal-predominant) distribution. We therefore selected a reverse PSO at L4 to preferentially decrease the distal arc while allowing concurrent moderation of the proximal arc via the L3–4 TLIF. Postoperatively, L1–4 lumbar lordosis decreased to 25° and L4–S1 lumbar lordosis to 46°, shifting the curve magnitudes toward PI-consistent targets. Importantly, her C2PA also normalized from –13° preoperatively to –5° at 6 weeks, reflecting restoration of global alignment. Similarly, pelvic tilt improved from –8° to –5°. These changes highlight how reverse PSO not only corrected segmental and global lordosis but also restored balanced spinopelvic orientation.

While extensive literature has described undercorrection of alignment and its sequelae, fewer reports have focused on overcorrection. Lewis et al. noted that lumbar hyperlordosis “rarely causes clinically significant negative sagittal imbalance because of thoracic compensation.”⁶ In our patient, thoracic compensation was limited by prior fusion to T3, and her overcorrection indeed manifested as a fixed global imbalance. Merrill et al. described risk factors for postoperative negative sagittal balance, including upper thoracic upper instrumented vertebra, lower BMI, and low preoperative SVA.⁸ Similarly, Le Huec et al. emphasized that individualized alignment targets based on PI, age, and thoracic compensation thresholds are essential to avoid both under- and overcorrection.⁷ In this context, our case expands the small body of work on pathological hyperlordosis by quantitatively demonstrating that correcting a large negative PI-LL mismatch and redistributing lordosis along the lumbar spine can restore spinopelvic harmony and relieve symptoms.

Faundez et al. and Lewis et al. showed that reverse osteotomies can be used to effectively address iatrogenic hyperlordosis and its sequelae.^5,6 Our case expands on these techniques by demonstrating a reverse PSO with concurrent TLIF to reduce segmental and global lordosis. The radiographic improvements in LL, segmental distribution, C2PA, and SVA, coupled with complete resolution of the patient’s preoperative imbalance complaints, underscore the effectiveness of this strategy.

Lessons

This case highlights several important lessons regarding alignment targets and surgical strategy in complex deformity surgery. First, overcorrection of LL can be just as problematic as undercorrection. A large negative PI-LL mismatch and fixed negative SVA may produce debilitating symptoms, particularly when thoracic compensation is constrained by prior fusion. Adhering rigidly to generalized alignment goals without fully accounting for individual pelvic morphology, age, and the capacity for thoracic and pelvic compensation can result in nonphysiological constructs. Incorporating patient-specific spinopelvic parameters and recognizing the risk factors for negative sagittal balance are therefore critical for preoperative planning.

Second, this report demonstrates that reverse PSO is a viable and reproducible technique to reduce fixed hyperlordosis, restore more physiological segmental lordosis distribution, and reestablish global sagittal alignment. Key technical distinctions from a standard PSO include the use of sequential distraction to open the osteotomy, continuous nerve root tension monitoring, and undercontouring with final rod compression to mitigate root stretch. These nuances are essential for safe execution of this less common corrective maneuver and may help minimize neurological risk.

Third, this strategy offers several practical advantages in the revision setting: 1) it preserves the anterior column cages, which have achieved solid fusion; 2) it avoids repositioning or dual approaches, reducing operative time and anesthetic exposure; 3) it allows correction of both segmental and global alignment via a single posterior approach; and 4) it minimizes anterior column violation, reducing risk to vascular structures and avoiding challenges associated with anterior revision in previously operated fields.⁹

This report has limitations inherent to a single illustrative case, including limited follow-up and lack of formal patient-reported outcome measures. As such, the generalizability of these findings is constrained, and further series are needed to validate the safety, durability, and functional impact of reverse PSO for iatrogenic hyperlordosis. Nevertheless, the radiographic and clinical improvements observed in this patient support the concept that carefully planned reverse osteotomy techniques can restore sagittal balance and alleviate symptoms when alignment has been overcorrected. In select patients with fixed hyperlordotic malalignment, reverse PSO should be considered a targeted tool within the deformity surgeon’s armamentarium.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Kim, Dekhne, Kazarian. Acquisition of data: Dekhne, Kazarian, Owusu-Sarpong. Analysis and interpretation of data: Kim, Dekhne, Kazarian. Drafting the article: Dekhne, Kazarian. Critically revising the article: Kim, Dekhne, Owusu-Sarpong. Reviewed submitted version of manuscript: Kim, Dekhne, Owusu-Sarpong. Approved the final version of the manuscript on behalf of all authors: Kim. Study supervision: Kim, Kazarian.

Correspondence

Han Jo Kim: Hospital for Special Surgery, New York, NY. kimh@hss.edu.