Radiologic Insights: Diagnosing Lumbosacral Transitional Vertebrae. Systematic Review of the Literature

Authors: Pedro Andrade‐Andrade, Juan Carlos Acevedo‐González

Abstract

Introduction

The lumbosacral transitional vertebra (LSTV) has been studied since 1876, with Castellvi developing a classification in 1984 based on its anatomy and laterality. It often goes unnoticed, or its diagnosis is limited to a lumbar spine X‐ray for confirmation. This has led to LSTV being underdiagnosed or even ignored. Our aim is to describe and evaluate radiological diagnostic techniques for LSTV and propose a diagnostic methodology to reduce errors in vertebral level identification, useful for percutaneous procedures and/or biomechanical measurement analysis.

Materials and Methods

A systematic literature review was conducted. The search terms “Castellvi,” “Lumbosacral Transitional Vertebra,” “Radiology.” Logical connectors such as “and” and “or” were applied. The following databases were Scopus, PubMed, Ovid, ScienceDirect, EBSCO, and Nature. The timeframe was limited from 2004 to December 2024. Inclusion and exclusion criteria were applied. A total of 419 articles were identified. The “Rayyan” program was used to compile information, and “PRISMA,” “STROBE,” and “CONSORT” were used to facilitate the analysis process.

Results

Forty‐eight articles were included and analyzed (10 CT, 4 PET‐CT, 2 bone scans, 9 MRI, 6 X‐rays, 4 EOS, and 13 mixed). The most common findings highlighted CT as the gold standard for diagnosing LSTV, with spinopelvic parameters correlating with LSTV. Radiography is effective for vertebral numbering. MRI studies utilize anatomical landmarks to identify vertebral levels and LSTV, although they are less sensitive. EOS is also used for vertebral level identification.

Conclusions

Our proposed diagnostic methodology for LSTV first, using plain AP radiography for cranial‐to‐caudal vertebral numbering and evaluating morphological anomalies. Second, if LSTV is suspected, performing CT as the gold standard for diagnosis due to its high sensitivity and specificity, and measuring spinopelvic parameters to correlate with LSTV. Third, using MRI in special cases. Fourth, conducting a morphological analysis and using Jenkins' classification for LSTV categorization.

Introduction

The clinical relevance of Lumbosacral Transitional Vertebra (LSTV) has been a subject of prolonged debate. In 1876, Rosenberg proposed that the spine underwent a shortening process as a mechanism to stabilize the erect column. He suggested that the sacralization of the fifth lumbar vertebra represented an anthropological progression, while the lumbarization of the first sacral vertebra was considered regressive [1]. Subsequently, Bertolotti et al. [2] explored in 1917 the association between low back pain (LBP) and the extension of the fifth lumbar vertebra to the sacrum, giving rise to Bertolotti's syndrome. Later, in 1955, Stinchfield and Sinton [1] conducted a radiological study and referenced herniations at the level of the transitional vertebra.

Years later, in 1984, Castellvi et al. [3] conducted a retrospective study including 200 consecutive cases from 1979 with myelographic evidence of herniated nucleus pulposus. The myelograms were initially interpreted by neuroradiology personnel and later independently reviewed by one of the authors. On the basis of this, they developed the Castellvi classification of LSTV into four categories with seven subcategories (Figure 1) according to the morphological characteristics in radiology and the clinical relevance concerning lumbar herniation [3].

Type I. Dysplastic transverse process: a, unilateral; b, bilateral.

This type is characterized by a large, triangular transverse process measuring at least 19 mm in width, as originally described by Southworth and Bersack in 1950 [3].

Type II. Incomplete lumbarization/sacralization: a, unilateral; b, bilateral.

This type is characterized by a prominent transverse process that appears to fit the shape of the sacral ala. They are considered incomplete due to the presence of a diarthrodial joint between the transverse process and the sacrum [3].

Type III. Complete lumbarization/sacralization: a, unilateral; b, bilateral.

Similar to type II, but distinguished by a true osseous union between the transverse process and the sacrum instead of a diarthrodial joint [3].

Type IV. Mixed: Patients in this category exhibit Type II on one side and Type III on the other.

The terms “lumbarization” and “sacralization” are used because the exact number of vertebrae in the patients' spinal column could not be definitively determined [3].

Given the historical context, the enlargement of the transverse process on one or both sides of the last lumbar vertebra, or the formation of a joint or complete fusion between this process and the sacral ala, constitutes an anomaly that has been and remains common. This anomaly often leads to what is known as LSTV, in which the last lumbar segment may present sacralization, or the first sacral segment may present lumbarization. Occasionally, a spectrum of morphological changes in these vertebral bodies can be observed, described as “squared” and “wedging” [4, 5].

LSTV is a clinically relevant anatomical alteration due to its association with LBP, known as Bertolotti syndrome, which may be of primary origin, caused by intrinsic inflammatory processes, or secondary, resulting from biomechanical alterations in the spine. Its identification is crucial for planning surgical or percutaneous procedures, especially in minimally invasive techniques such as spinal fusions or epidural injections, and for adjusting sagittal balance analysis protocols, which are designed under the premise of five lumbar vertebrae. This number can vary in the presence of LSTV or other anatomical anomalies. Nonetheless, standardizing radiological findings to correctly identify and classify this condition according to Castellvi's criteria remains a challenge, highlighting the need for more consistent and precise diagnostic approaches.

The aim of this systematic review is to describe and evaluate radiological diagnostic techniques for LSTV and propose a diagnostic methodology to reduce errors in identifying vertebral levels, useful for percutaneous procedures and/or biomechanical measurement analysis.

Materials and Methods

Literature Review

A systematic review of medical literature from the last 21 years was conducted. The following search terms were used in “Castellvi,” “Lumbosacral Transitional Vertebra,” and “Radiology,” using logical connectors such as “AND” and “OR.” Databases like Scopus, PubMed, Ovid, ScienceDirect, EBSCO, and Nature were reviewed. The search period was limited from 2004 to December 2024. The selected articles were organized using the systematic review tool “Rayyan.” This application is a platform that allows easy manipulation of the collected information.

Article Selection

A total of 419 articles were identified from the Scopus (n = 149), PubMed (n = 166), Ovid (n = 54), ScienceDirect (n = 35), EBSCO (n = 11), and Nature (n = 4). Each author independently reviewed the abstract of each article found and applied the following inclusion and exclusion criteria.

The inclusion criteria Studies in English or Spanish that included a radiological description of LSTV, either findings or diagnostic technique used.Articles conducted on humans.Articles published between January 1, 2004, and December 18, 2024.

The exclusion criteria Articles not conducted on humans.Cadaveric studies.Technical notes or editor's notes.

A thorough evaluation of all selected articles was performed using specific qualitative checklists for each study Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) for observational studies [6] and Consolidated Standards of Reporting Trials (CONSORT) for randomized clinical trials [7]. All articles included in this analysis are original, ensuring the absence of duplication in the reviewed literature. The systematic review was conducted following the structure and recommendations outlined in the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines [8]. Additionally, the PRISMA diagram was incorporated to visually represent the article selection process (Figure 2).

Finally, 48 articles were included. Among these, there was 1 case series study [9], 2 case reports [10, 11], 2 literature reviews [4, 12], 4 prospective studies [5, 13, 14, 15], 10 cross‐sectional studies [16, 17, 18, 19, 20, 21, 22, 23, 24, 25], and 29 retrospective studies [26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54].

Results

A total of 48 articles were included in the systematic review. A summary of the included studies is shown in Table 1. The results of the selected articles are categorized according to the diagnostic techniques used for the diagnosis of LSTV, such as CT, PET‐CT, scintigraphy, MRI, radiography, EOS, and the combined use of these techniques.

Computed Tomography (CT)

Ten studies published between 2019 and 2024 evaluated the diagnosis of LSTV using CT. Of these, nine [26, 27, 28, 35, 42, 43, 46, 48, 54] studies were retrospective, and one was a cross‐sectional study [17]. These articles describe the use of CT, whether thoracic/abdominal or pelvic. Among the radiological findings to consider are the use of anatomical landmarks such as the superior vena cava, spinopelvic parameters, rib, and thoracic vertebra counts.

Iplikcioglu and Karabg [27, 28] conducted two studies with the same cohort. The authors performed a retrospective study using abdominal CT scans of 1420 patients, of whom 108 had LSTV. The diagnosis was established by identifying unilateral or bilateral pseudoarticulation or complete fusion between the transverse process of the lowest lumbar vertebra and the sacrum, following the Castellvi classification. Both studies emphasized the measurement of pelvic parameters such as pelvic incidence, pelvic tilt, sacral slope, lumbar lordosis, and the difference between pelvic incidence and lumbar lordosis using the superior and inferior endplates of the transitional vertebra. Significant differences were found between upper and lower endplate measurements, although both were highly correlated, and the authors proposed the concept of an “optimum pelvic incidence,” defined as the endplate that yields a value closest to the mean of normal subjects. These findings suggest that LSTV morphology can influence pelvic alignment and that careful selection of the endplate is relevant for accurate measurement.

Similarly, Liebl et al. [26] conducted a retrospective study, including 300 patients with CT images of the spine from four scanner providers. In this study, they aimed to address the large‐scale vertebral segmentation challenge. For the diagnosis of LSTV, they used the well‐established Castellvi classification for LSTV, primarily based on the morphology of the transverse process of the last lumbar vertebra and whether it was fused with the sacrum. Before staging the degree of LSTV, they analyzed the count and variants of cervical (n: 1 and 2), thoracolumbar (n: 11, 12, and 13), and lumbosacral (n: 4, 5, and 6) vertebrae. The study highlighted the importance of complete spinal visualization to ensure accurate classification and labeling. Anatomical variants were frequent, with 33.7% of cases showing six lumbar vertebrae and 28% showing short ribs, reinforcing the high frequency of segmentation anomalies associated with transitional vertebrae. In the study conducted by Doo et al. [48], also retrospective, they evaluated 1553 whole‐spine CTs to determine the prevalence and relationship between thoracolumbar and lumbosacral transitional vertebrae. Regarding imaging, the following parameters were first, vertebral levels were counted cranio‐caudally, using 3D‐CT images of the entire spine from C1, assuming 7 cervical, 12 thoracic, and 5 lumbar vertebrae. Vertebrae 20 and 25 were defined as L1 and S1, respectively. Several investigators proposed that most caudal ribs could be classified into one of four normal rib, hypoplastic rib (or short rib), non‐fused transverse process (or accessory ossification center), and mixed type. Sacralization was defined as the abnormal fusion of L5 (the 24th vertebra) with S1. They found a prevalence of 11.2% for thoracolumbar and 8.3% for LSTV, both frequently associated with abnormal rib counts. Patients with thoracolumbar transitional vertebrae were seven times more likely to have LSTV compared with those without this anomaly. The authors concluded that complete spinal imaging is essential for accurate vertebral numbering, as partial counts based on ribs or fluoroscopic references can easily lead to errors.

Expanding on functional implications, Luo et al. [54] conducted another retrospective study aimed at investigating the impact of LSTV on hip development alterations. In this study, 310 individuals were categorized into three groups based on full‐body CT: a group with sacralization of 23 vertebrae, a group with lumbarization of 25 presacral vertebrae, and a normal control group with 24 presacral vertebrae. All were diagnosed with LSTV via CT. They found that LSTV significantly modified pelvic parameters such as pelvic incidence, pelvic tilt, sacral slope, and lumbar curvature, as well as acetabular morphology. Sacralization was associated with increased sagittal acetabular anteversion and reduced sagittal acetabular coverage, suggesting that LSTV alters lumbopelvic and hip biomechanics. These findings highlight that transitional anatomy should be considered during hip replacement planning to ensure correct implant positioning.

Degenerative changes associated with LSTV were a focus of multiple studies. Hanhivaara et al. [42] conducted a retrospective study using 3855 abdominal CTs and reported a prevalence of 28.6%. Castellvi types II–IV were associated with higher rates of degeneration in intervertebral discs and facet joints, particularly at the L4–L5 level. The authors also proposed two new Castellvi subtypes (IIc and IIIc) to describe unilateral pseudoarticulation or fusion with contralateral transverse process enlargement and reported excellent interobserver agreement. Supporting these findings, the cross‐sectional study conducted by Cheng et al. [17] examined 529 patients, 350 of whom had LSTV confirmed by CT and MRI. LSTV was classified as type I–IV and then marked as “a” if unilateral or “b” if bilateral and classified according to Castellvi. The authors observed that Castellvi type I was not related to degenerative changes, while type II was associated with degeneration affecting both the transitional and adjacent discs. In contrast, types III and IV showed less degeneration at the fused transitional level but more advanced changes in the cranial segment, including disc protrusion, endplate defects, and spondylolisthesis. These results support that higher Castellvi types predispose to adjacent‐segment degeneration, like what occurs after surgical fusion, whereas type II presents a mixed pattern involving both the transitional and adjacent levels. Additionally, in the study performed by Desai et al. [35], a retrospective comparison was made of 172 patients diagnosed with Bertolotti's syndrome (n = 101) or lumbar spondylosis without identified LSTV (n = 71) documented in the clinical history or present in the images. Diagnosis was confirmed by CT following the Castellvi classification, and the study demonstrated that patients with LSTV had higher pelvic incidence and greater adjacent‐segment degeneration at L4–L5 compared with controls. However, after adjusting for age and sex, these associations were no longer significant, suggesting that both anatomical and biomechanical factors contribute to the degenerative process.

Another dimension was explored in musculature‐related changes. In the article authored by Becker et al. [43], a retrospective study was conducted, and 46 matched controls using abdominal and pelvic CT to evaluate muscle morphology. Patients with LSTV showed lower paraspinal and abdominal muscle volume and increased fatty degeneration, except in the most caudal paravertebral muscles. These findings suggest that LSTV leads to muscular atrophy and changes in lumbopelvic biomechanics. Lastly, Gündüz et al. [46] conducted a retrospective study including 113 patients, 58 of whom had LSTV, with the goal of evaluating the reproducibility and accuracy of the iliac crest tangent as a reference point in subjects without disc degeneration. In this study, they also used full‐spine CT for the diagnosis of LSTV. The number of LSTV was evaluated according to the iliac crest tangent measurement without prior assessment of the correct LSTV count. The results concluded that the iliac crest tangent does not appear to be a reliable reference point for the correct numbering of LSTV in patients without intervertebral disc degeneration.

PET‐CT

Four articles were identified in which the diagnosis of LSTV was made using PET‐CT. These articles were published between 2017 and 2023. Of these, there was one case report [10], one case series [9], and two retrospective studies [32, 45].

The studies reported by Usmani et al. [9, 10] involved a case report and a case series with 55 patients, both symptomatic and asymptomatic, diagnosed with incidental LSTV who underwent PET‐CT with 18F‐NaF. Of the 34 symptomatic patients, most showed focal uptake of the tracer at the pseudoarticulation site, demonstrating a linear trend between the intensity of the uptake and the presence of symptoms. Therefore, the degree of uptake was mentioned as a potential biological marker of symptomatic LSTV. Moreover, 18F‐NaF PET‐CT showed high diagnostic performance, with a sensitivity of 82% and a specificity of 86% in differentiating symptomatic from asymptomatic patients. The authors concluded that 18F‐NaF PET‐CT is a sensitive tool for detecting areas of abnormal bone remodeling and provides molecular insight into the biomechanical alterations caused by LSTV [9].

Zhou et al. [32, 45] conducted two retrospective studies using full‐spine CT from PET/CT for the diagnosis of LSTV. In the first study, which included 6097 patients, 210 were diagnosed with LSTV. The authors found that parameters such as pelvic tilt, sacral slope, lumbar alignment, sacral table angle, and pelvic radius were stable and reliable for assessing sagittal balance in LSTV cases, particularly when measured using the Morph S1 reference. They also identified vertebral count as an independent factor influencing sagittal pelvic parameters. In the second study, they reviewed 2845 patients who underwent PET/CT under the same protocol, with 222 included for analysis. They compared new quantitative metrics, such as the anterior‐edge vertebral angle and the ratio of inferior‐to‐superior endplate length, with traditional anatomical landmarks like the iliac crest tangent, iliolumbar ligament, and psoas insertion. The quantitative parameters achieved higher diagnostic accuracy (around 90%) and reproducibility than conventional markers, making them more useful for distinguishing between L5 sacralization and S1 lumbarization in CT imaging.

Scintigraphy

Two articles were identified in which the diagnosis of LSTV was made using bone scintigraphy. These articles were published between 2004 and 2015. Of these, there was one case report [11] and one cross‐sectional study [23].

Kassir et al. [11] published a case report where they diagnosed LSTV using single‐photon emission computed tomography (SPECT) in a 16‐year‐old woman with lumbar pain lasting for 8 months, for which a bone scan was requested due to suspicion of a pars interarticularis defect. The results showed a focus of increased radiotracer activity in the left lumbosacral area over the left sacroiliac joint, which was consistent with LSTV type IIa. The authors concluded that whole‐body bone scintigraphy with hybrid SPECT/CT imaging of the painful area can be a useful tool in diagnosing LSTV.

Pekindil et al. [23] conducted a retrospective cross‐sectional study diagnosing LSTV through SPECT/CT in patients previously diagnosed by radiography. Changes in their conditions were analyzed to compare both imaging findings. The study included 28 patients, and the results showed that planar scans demonstrated mild to normal or non‐focal uptake with mild to moderate increase, while SPECT demonstrated focal uptake from mild to moderate and significantly increased in patients with degenerative changes without lumbar pain and with lumbar pain, respectively. Radiographs showed an association between degenerative changes and lumbar pain, and SPECT results showed focal uptake that was markedly increased. They concluded that bone scintigraphy can be considered for evaluating patients with lumbar pain believed to arise from the LSTV joint.

Magnetic Resonance Imaging (MRI)

Nine articles were identified that made the diagnosis of LSTV using MRI. These articles were published between 2006 and 2024. Among these, there were six retrospective studies [33, 34, 41, 47, 52, 53] and three cross‐sectional studies [16, 21, 25].

Several studies addressed the diagnostic reliability and anatomical challenges of identifying LSTV using MRI. In a cross‐sectional study performed by Landauer et al. [16], 1842 patients were radiologically evaluated and underwent an MRI of the lumbosacral junction. Of the total patients, 115 were diagnosed with Castellvi II–IV and compared with the literature, noting that Castellvi I was excluded from the study. The authors mentioned that the diagnosis of LSTV with only spinal radiography, without a Ferguson view, is ambiguous, and standard lumbar spine radiographs, as well as spinal radiographs, do not provide a reliable orthogonal representation of the transverse process of L5. Since Ferguson view is not taken for radiological hygiene, they state that the alternative is MRI, which provides additional information in all three planes of space. They concluded that the shape of the L5 transverse process might be more indicative of pathology than just the measurement alone, supporting the association between LSTV and LBP.

Similarly, Peker et al. [33] carried out a retrospective study of 143 patients who underwent MRI or CT of the entire spine to evaluate if lumbar vertebrae can be properly numbered using auxiliary parameters. The authors used only MRI to diagnose LSTV in 13 patients based on vertebral morphology and the lumbosacral angle. Vertebral numbering began from the C2 vertebra, counting seven cervical vertebrae and 12 thoracic vertebrae downwards. If there were no differences between the upper and lower terminal plates of the vertebrae, the vertebra was considered square or rectangular. If four square/rectangular vertebrae were present, LSTV was identified as sacralization, and if six square/rectangular vertebrae were present, it was identified as lumbarization. The LSTV was classified using the Castellvi system, and the upper and lower terminal plates of L5 and S1 were also measured. The study concluded that no MRI parameter can precisely indicate the number of vertebrae without counting the levels, and these parameters may only be suggestive of LSTV rather than indicating the correct level. Building on this anatomical theme, Garg et al. [34] conducted a study in which they retrospectively evaluated 260 patients with a confirmed L5 level on MRI, analyzing anatomical landmarks such as the iliolumbar ligament, costal facets, aortic bifurcation, psoas origin, and conus medullaris. The sensitivity of each landmark for identifying L5 was 85%, 78%, 65.8%, 51.2%, and 58.5%, respectively, with specificities of 100%, 97.3%, 82.2%, 58.0%, and 68.0%. They concluded that, although some structures such as the iliolumbar ligament can help orientation, these landmarks are not sufficiently reliable for consistent numbering of L5 in cases of LSTV. Likewise, Chalian et al. [52] conducted a retrospective study involving 100 patients, 50 with LSTV and 50 healthy controls, and proposed two sagittal angle measurements (A and B) that may alert radiologists to the presence of transitional anatomy. Increased angles were associated with LSTV, suggesting their potential as screening signs.

The role of the iliolumbar ligament in LSTV identification was further explored in the article described by Hughes et al. [53], a retrospective study which reviewed 500 MRIs from patients presenting with lumbar pain, lumbar radiculopathy, or both. The study aimed to determine whether identifying iliolumbar ligaments is practically useful for numbering LSTV. Sixty‐seven patients were diagnosed with LSTV using this imaging method. The study concluded that the iliolumbar ligament is easily identifiable on lumbar spine MRI and always arises from L5. The authors suggested that its position can be used to safely assign lumbar levels in patients with LSTV. A similar approach was used by the retrospective study presented by Özbalci [47], which evaluated 1020 MRI images, and 114 patients were diagnosed with LSTV. The study diagnosed LSTV using MRI by identifying the iliolumbar ligament to define L5. They then classified LSTV according to Castellvi into patients with and without dysplasia of the transverse process. The study concluded that lumbosacral variations are common in patients with LBP and that these variations may be associated with degenerative and/or edematous changes that mimic sacroiliitis.

Some studies also focused on the clinical relevance of LSTV in the context of LBP and degenerative changes. Bhagchandani et al. [25] conducted a cross‐sectional study with 2016 patients, dividing them into 1009 with LBP and 1007 with radicular pain; of these, 149 and 174 patients had LSTV, respectively. In this cohort, MRI with T1 and T2 sequences was used. The study concluded that patients with sacralization represented > 80% of patients with LSTV and lumbar pain or radiculopathy, which may be explained by the high incidence of disc degeneration in the three levels immediately proximal to a sacralized vertebra, as well as a high incidence of terminal plate degeneration and facet joint tropism. The authors highlighted the strong association between LSTV and degenerative spine changes, particularly in patients with LBP. In a similar vein, Hanhivaara et al. [21] conducted a cross‐sectional study involving the 1966 northern Finland birth cohort. Of the 1468 cases, 310 were diagnosed with LSTV. Degenerative findings, including facet degeneration, disc protrusion, and Modic type I changes, were more frequent above the transitional segment (L3–L4 and L4–L5). The authors concluded that Castellvi type III is most strongly associated with LBP, supporting the biomechanical model of adjacent‐segment overload.

Finally, the association between LSTV and inflammatory LBP was highlighted by Türk et al. [41] who conducted a retrospective study in which 614 MRI images were reviewed, with 81 patients diagnosed with LSTV. The objective of the study was to assess the presence of LSTV in patients who underwent MRI due to suspected sacroiliitis. The study concluded that LSTV may present with LBP and should be considered in patients clinically suspected of having sacroiliitis. MRI was again noted as a useful modality for identifying coexisting pathologies.

Lumbar Spine X‐Ray (AP, Lateral, and Special Projections)

Six studies were identified that diagnosed LSTV using X‐ray, published between 2015 and 2024. These included one prospective study [5], two cross‐sectional studies [20, 22] and three retrospective studies [30, 31, 49].

Ravikanth and Mahumdar [5] conducted a prospective study that included 500 patients with inclusion criteria of only LBP, of which 134 were classified as positive for sacralization. This study aimed to classify anatomical variations in LSTV and determine, through simple X‐ray, if there is a relationship between sacralization and LBP. Of the total patients, the most common anatomical variant was Castellvi Type IA (7.6%), followed by Type IB (6.0%), Type IIA (1.8%), Type IIB (2.0%), Type IIIA (1.6%), Type IIIB (3.8%), and Type IV (0.8%). Patients with LBP and no malformation showed an average pain level of 2.2 compared to 5.2 in patients with LBP and a transitional vertebra. Similarly focused on LBP and hip pathology, Verhaegen et al. [20] conducted a cross‐sectional study, which included 153 patients, of which 19 had LSTV, with the aim of determining the prevalence of LSTV in young patients with hip pain compared to an asymptomatic volunteer group, the effect of LSTV on static and dynamic spinopelvic characteristics and evaluating the presence of LBP among young adult patients with hip pain and LSTV. In this study, the diagnosis of LSTV was made using an AP pelvic X‐ray. The Castellvi classification system was used to classify LSTV based on the degree of unilateral or bilateral articulation between the transverse processes of L5 and the sacrum. Spinopelvic parameters were measured in standing and sitting positions. This study concluded that LSTV was found in 8.5% of young adults, with no differences between patients with hip pathology and controls; and that individuals with LSTV have greater lumbar lordosis in standing, with altered mechanics at the adjacent cephalic level, which may predispose these individuals to degenerative changes at this level.

Sun et al. [22] conducted a cross‐sectional study with the aim of testing the hypothesis that there is a higher frequency of radiological anomalies in the spine in patients with acetabular dysplasia and evaluating the relationship between the radiological severity of acetabular dysplasia and the frequency of spinal anomalies. This study included 122 patients who presented hip pain and a final diagnosis of acetabular dysplasia and had an LSTV frequency of 39%–43%. The diagnosis and classification of LSTV were made using X‐rays. It was concluded that patients with acetabular dysplasia have a higher frequency of spinal anomalies observed in standard hip X‐rays.

Expanding on biomechanical parameters, in the study carried out by Benlidayi et al. [31], X‐rays were also used for the diagnosis of LSTV, and classification was performed according to Castellvi. Of 1588 patients evaluated retrospectively, 96 patients with Castellvi II or higher were included. The sacral tilt angle was measured on lateral X‐rays for comparison between groups. They concluded that the sacral tilt angle is significantly lower in patients with LSTV than in patients without LSTV, and that the sacral tilt angle does not differ between the types of LSTV. This may be useful for considering or confirming LSTV in patients with a smaller ST angle. Complementing this anatomical and diagnostic focus, Chiu et al. [49] conducted a retrospective study, which included the analysis of 998 X‐rays with the goal of investigating the variation in the number of thoracic and lumbar vertebrae, the prevalence of LSTV, and the prevalence of cervical ribs among surgical patients with adolescent idiopathic scoliosis. In this study, X‐rays were used to diagnose LSTV. They specified that vertebral numbering was performed starting from the C2 vertebra by identifying the odontoid process and continuing caudally to avoid omitting the cervical rib. All vertebrae with costal insertions (except cervical ribs), including full or unilateral ribs, were counted as thoracic vertebrae, and lumbar vertebrae were identified as vertebrae with no costal insertion after the most caudal thoracic vertebra. LSTV was defined as a lumbar vertebra in which one or both transverse processes were fused to the sacrum, through incomplete or complete bony fusion or a diarthrodial joint. This included sacralization of the lowest lumbar vertebral body and lumbarization of the highest sacral segment. In this study, seven different variations were identified in the number of cervical, thoracic, and lumbar vertebrae. The total prevalence of patients with atypical vertebral variation was 15.5%, and LSTV was found in 25.1% of the cohort. Furthermore, they stated that it is important to determine atypical vertebral variations rather than the absolute number of vertebrae. Therefore, it was concluded that due to differences in the number of morphologically thoracic and lumbar vertebrae, there may still be a risk of inaccurate identification.

Finally, in the study presented by Ani et al. [30], a retrospective analysis of 1548 patients was performed, of which 176 were included with LSTV. The aim was to determine whether patients with spinal deformity with sacralization of L5 should have PI and other spinopelvic parameters measured from the L5 terminal plate or S1. The diagnosis of LSTV was made using baseline full body standing X‐rays. The results showed that when measured using the upper terminal plate of L5, pelvic parameters were significantly smaller than those measured in relation to S1. They concluded that measuring PI and spinopelvic parameters in L5 with sacralized anatomy results in an underestimation of spinal deformity. Therefore, they suggest that surgeons may consider measuring PI and spinopelvic parameters in relation to S1 instead of L5 in patients with sacralized L5.

EOS 3D

Four studies were identified that diagnosed LSTV using EOS 3D. These articles were published between 2016 and 2022. Of these, one study was prospective [15] and three retrospective studies [36, 40, 50].

Okamoto et al. [36] conducted a retrospective study with 291 healthy adult patients with no history of spine disease, using biplanar full‐body EOS imaging to evaluate the prevalence of LSTV and its influence on global sagittal alignment. They identified 14 patients with 23 vertebrae (sacralization) and 16 with 25 vertebrae (lumbarization), for an overall prevalence of 10.3%. Compared with normal spines, the sacral base was located higher in the L4 group (sacralization) and lower in the L6 group (lumbarization). Participants with LSTV showed significant differences in global alignment, with greater cervical lordosis (C2–C7) and increased lumbar lordosis in both L4 and L6 groups. Conversely, other spinopelvic parameters were reduced in L4 and increased in L6. On the basis of the analysis of the results, the authors proposed radiographic principles for vertebral numbering based on rib presence, intervertebral disc continuity, and sacral base identification, improving consistency in LSTV assessment. The authors concluded that the spinopelvic parameters of the LSTV population significantly differed from those of the normal spine population due to differences in the location of the sacral base; thus, they proposed an accurate method for numbering the vertebrae using coronal and sagittal full‐body images.

In the article performed by Becker et al. [40], a retrospective study was conducted with 51 patients with L5/S1 osteochondrosis to investigate lumbar spine mobility and the distribution of segmental movement in patients with LSTV. In this study, full spine radiographs obtained from EOS were used to diagnose LSTV and classify it according to Castellvi, following parameters for vertebral numbering and spinopelvic measurements. The authors mention that the number of lumbar vertebral bodies was classified by counting caudally from C1 in full‐spine images. For the cervical spine, 7 vertebrae were assumed, and 12 for the thoracic spine. L1 was defined as the 20th vertebra, and lumbar vertebrae were assumed if there were 25 vertebrae with at least rudimentary discs between them. On the basis of the results, the authors concluded that LSTV has a significant effect on lumbar spine movement patterns with the use of flexion‐extension radiographs. This is because reduced movement was demonstrated in the transitional segment, as well as an increased proportion of mobility in the cranial adjacent segment. Likewise, Zhou et al. [50] retrospectively evaluated 1869 radiographs and identified 70 patients with transitional lumbosacral vertebrae (3.7%). Among these, 82.9% had lumbarized sacral segments and 17.1% had sacralized lumbar segments. The study demonstrated that choosing different sacral endplates (cephalad vs. caudal) for measurements significantly alters pelvic and global alignment parameters. When the caudal transitional segment was selected, mean pelvic incidence, pelvic tilt, and sacral slope were markedly higher (PI: 66.8° vs. 44.3°; PT: 25.1° vs. 12.7°; SS: 41.6° vs. 31.6°, p < 0.001). Similarly, lumbar lordosis and global parameters such as sagittal vertical axis and T1–pelvic angle also increased significantly. Therefore, they mentioned that surgical errors are more likely when lumbar MRI is reported without additional images; as the use of S1–S2 disc morphology, vertebral body shape, and the lumbosacral angle to define vertebral levels leads to the erroneous identification of LSTV and spinal levels.

Price et al. [15] conducted a study based on prospective full‐body EOS spine databases, diagnosing LSTV in 11 patients. This study aimed to provide normative values for spinopelvic measures in S1 lumbarization and to investigate correlations between lumbar lordosis and PI, measured at the first immobile sacral vertebra, S2, and the incidence of L6 in the true S1. Regarding radiological parameters, the authors also followed principles for vertebral numbering. Patients with S1 lumbarization were identified based on the number of lumbar vertebral bodies counted from the cervical spine caudally, assuming seven cervical vertebrae and 12 thoracic vertebrae. Patients were positioned in a standardized standing posture to ensure reproducible spinopelvic measurements. On the basis of the results, it was concluded that incomplete spine images could lead to an erroneous estimate of lumbarization prevalence, and that patients with lumbarization have higher lordosis values, with lordosis now being estimable during preoperative planning for this group.

Mixed

Thirteen articles were identified in which the diagnosis of LSTV was made using a combination of diagnostic techniques. These articles were published between 2010 and 2024. Among these, there were two literature reviews [4, 12], two prospective studies [13, 14], three cross‐sectional studies [18, 19, 24] and six retrospective studies [29, 37, 38, 39, 44, 51].

Two literature reviews focused on vertebral numbering and the clinical relevance of imaging findings in LSTV. Lian et al. [12] reviewed current literature on precise methods for vertebral numbering. In this study, the diagnosis of LSTV was evidenced through various radiological techniques. Their results showed that the gold standard for spinal segment numbering in patients with LSTV remains the acquisition of full‐spine images and caudal numbering starting from C2. If unavailable, the use of the iliac crest tangent sign in coronal MRI has a reliable sensitivity and specificity for accurate LSTV numbering, with a sensitivity and specificity of 81% and 64%–88%, respectively. The role of other anatomical markers such as the right renal artery, superior mesenteric artery, aortic bifurcation, and the conus medullaris in identifying vertebral levels is unreliable and should not be used. In conclusion, the authors highlighted that full‐spine MRI should be considered when transitional anatomy is suspected, as numerical variants are common and may coexist with thoracic or cervical anomalies. Similarly, Konin et al. [4] conducted a literature review aimed at addressing issues related to correctly identifying and numbering LSTV, as well as detecting imaging findings related to lumbar pain genesis. Several diagnostic techniques were used in this study. The results indicated that without high‐quality full‐spine images, no infallible method exists for accurately numbering a transitional segment. Therefore, identification, communication with the referring physician, and correlation of intraoperative and preoperative images become extremely important. The study concluded that LSTV are common spinal anomalies that require accurate identification and numbering of the affected segment. A better understanding of the biomechanical alterations caused by LSTV can help radiologists recognize and interpret imaging findings in patients with lumbar pain and a transitional segment.

Two prospective studies combined multiple imaging modalities, including radiographs and MRI, for accurate LSTV diagnosis. Sencan et al. [13] examined 64 patients with radicular LBP undergoing lumbar transforaminal epidural steroid injection. Sacralization was identified by MRI and classified radiographically by Castellvi type. Pain and disability scores improved significantly in all patients (p < 0.05), but treatment success at 3 months was lower in those with sacralization (44.8%) compared with patients without LSTV (65.6%, p = 0.026). The authors concluded that sacralization may reduce therapeutic response and should be considered when evaluating treatment outcomes. In a related prospective study, Omidi et al. [14] conducted a study in which lumbar sacralization was diagnosed using radiological images, including oblique and lateral views of the spine, radiographs, and T1 and T2 weighted MRI images. Both imaging techniques were required for numbering and LSTV diagnosis. LSTV categorization was reported according to Castellvi's classification system, and the authors reported that transitional anatomy was associated with alterations in pelvic incidence and lumbar curvature, suggesting that combined imaging improves diagnostic accuracy and preoperative planning.

Among the cross‐sectional studies, three focused on combining radiography, MRI, or CT to evaluate LSTV. Shaikh et al. [19] performed a cross‐sectional study aimed at determining the frequency of LSTV in patients with LBP and the role of the L5 iliolumbar ligament origin in LSTV cases. This study evaluated both radiographs and MRIs together to determine the presence of LSTV. Initially, simple radiographs were reviewed, followed by MRI scans. LSTVs were classified according to Castellvi. LSTV was identified in 18% of patients, and in 44% of those cases, the iliolumbar ligament originated anomalously. The study concluded that LSTV occurs frequently in patients with lumbar pain. Furthermore, in the presence of LSTV, the iliolumbar ligament is not a reliable marker for identifying L5. In another cross‐sectional study, Tatara et al. [18] examined 56 patients with the goal of determining the optimal vertebral level for LSTV to measure PI and pelvic tilt. They concluded that if the measured PI is lower than the reference range, it is likely that PI was measured with LSTV as S1, and for this reason, it would be better to remeasure the PI with LSTV as the lowest lumbar vertebra. Conversely, if the measured pelvic tilt is higher than the reference range despite a high PI and no sagittal imbalance, it is likely that pelvic tilt was measured with LSTV as the lowest lumbar vertebra (p < 0.05). Furthermore, the study authored by Hanhivaara et al. [24] was a cross‐sectional study that included 852 patients undergoing lumbar imaging studies using all three modalities, initially evaluated for LSTV presence through CT scans, although all three modalities were employed. This study aimed to compare the diagnostic performance of conventional radiography, CT, and MRI. In total, 100 patients with LSTV anatomy were identified. The results reported superior diagnostic efficacy for CT: the sensitivity, specificity, accuracy, and balanced accuracy were 76%, 93%, 77%, and 84%, respectively. For MRI, the metrics were 54%, 88%, 56%, and 68%, and for radiography 32%, 85%, 42%, and 59%, respectively. CT also demonstrated good inter‐reader reliability (κ = 0.63–0.71), while MRI and radiography showed only fair agreement (κ = 0.24–0.56 and κ = 0.16–0.32, respectively). The conclusion was that CT provided the highest diagnostic performance in all metrics with good reliability among readers. Additionally, MRI and conventional radiography showed poor sensitivity and accuracy and should therefore be interpreted with caution when classifying LSTV.

A group of retrospective studies also relied on mixed modalities for diagnosis and classification of LSTV. Farshad‐Amacker et al. [37] conducted a retrospective study with 155 subjects. The objective was to evaluate inter‐reader reliability for LSTV detection and classification using standard AP radiographs and report its accuracy using intermodality statistics in comparison to MRI as the gold standard. The authors consider that radiographs are insufficient for LSTV diagnosis. Therefore, their study used both radiographs and MRI to identify and classify LSTV according to Castellvi. Agreement between modalities was poor (κ = 0.29), supporting the need for combined imaging to improve diagnostic accuracy. In a related retrospective study, Peckham et al. [38] used both radiographs and MRI to count vertebrae and classify LSTV. First, vertebral numbering considered the first seven vertebrae as cervical and the following 12 as thoracic, even in cases with an abnormal number of ribs. In cases with 13 rib‐bearing vertebrae, it was considered “lumbarization” with L1 having supernumerary ribs. After T12, vertebrae were counted as lumbar, extending to the lumbosacral junction. Vertebra 24 was considered L5 in all cases. LSTV classification considered whether lumbar transverse processes had unilateral or bilateral un‐fused joints with the sacrum (partial sacralization of L5), classifying them as Castellvi 1 or 2. If transverse processes were fused unilaterally or bilaterally with the sacrum (complete sacralization of L5), LSTV was classified as Castellvi 3 or 4. The authors emphasized that numbering errors were common when rib variants were present, underscoring the importance of full‐spine imaging for accurate labeling. Furthermore, Hou et al. [39] also conducted a retrospective study in which LSTV was initially diagnosed using radiographs and confirmed with CT. Both imaging techniques were used for LSTV diagnosis. Simple lumbar AP radiographs were evaluated for LSTV morphological abnormalities (types II–IV according to Castellvi classification). The patient was then examined with spiral CT scans from T12 to the sacrum for LSTV confirmation. The results showed that 35.2% of suspected LSTV types assessed by AP radiographs were inconsistent with the final types assessed by CT. The article concluded that radiographs could detect LSTV correctly but could not provide a precise type of classification, while CT provided detailed information between the pelvic tilt and sacral area and could be considered the gold standard for detecting and classifying LSTV.

Additionally, Tatara et al. [29] conducted a retrospective study where LSTV was detected in 3D CT images, and vertebral numbering was done with spine radiographs and classified according to Castellvi. For presacral vertebrae, numbering was done caudally from C2 using full‐spine radiographs. The lowest lumbar vertebra was identified as lumbar when above the iliac arch line, as normal lumbar vertebrae are located above that line. Additionally, the number of vertebrae and sacral foramina was identified when below the promontory in the CT images. The authors highlighted that correct identification of the lowest lumbar vertebra above the iliac arch line reduces misclassification and improves preoperative accuracy in surgical planning. Additionally, Carrino et al. [51] performed a retrospective review of 147 subjects using spine radiographs as the reference standard to determine total and segmental vertebral count and transitional anatomy. LSTV diagnosis was made with MRI and radiographs. They reported that anomalous total vertebral counts occurred in 8.2% of patients, and that the ligament was identifiable in 85.7% of cases, being located at L5 in 96.8%, though unreliable in anomalous numbering. The study concluded that iliolumbar ligament denotes the lowest lumbar vertebra, which does not always represent L5. Additionally, a well‐formed, complete S1–2 intervertebral disc is associated with LSTV, and LSTV is associated with abnormal vertebral numbering.

Lastly, Byvaltsev et al. [44] conducted a retrospective study. LSTV was detected using MRI, CT, and radiographs. To assess the type of lumbosacral junction anomaly in the study patients, detailed examinations of their lumbar radiographs, MRIs, and CT scans were performed. The number of vertebrae was counted from the cranio‐cervical junction according to radiographic data. The Castellvi classification was applied with a simplified modification grouping the types into three functional enlarged transverse process (types I), pseudoarthrosis (types II), and fused (types III–IV). The authors found that the fused group was the most frequent and correlated with decreased segmental mobility and lumbar pain, reinforcing its biomechanical significance.

Discussion

Historically, opinions on the clinical impact of LSTV have varied widely, from early interpretations as anthropological curiosities to later recognition as a cause of mechanical low back pain. These contrasting views underscore the need for standardized imaging‐based classification and clinical correlation.

In 1981, Wigh's review of 200 positive myelograms revealed no cases of hernia at the LSTV level. However, the variability of findings in different studies led Castellvi and colleagues to investigate the potential association between nucleus pulposus herniation and LSTV by conducting their own study [3]. This study led to the Castellvi classification to organize the types of LSTV. In 1984, Castellvi et al. [3] conducted a retrospective study that included 200 consecutive cases from 1979 with myelographic evidence of nucleus pulposus herniation. The myelograms were initially interpreted by neuroradiology staff and later reviewed independently by one of the authors. From this study, after their own classification and correlation with the myelographic findings of herniated nucleus pulposus, they concluded that type I represents a subtle manifestation of LSTV, with no discernible difference in the appearance of the hernia localization. In types III and IV, no hernias were observed at the LSTV level, nor was there a high incidence of hernias immediately proximal to it. However, the most notable finding was the type II anomaly, which showed a higher prevalence of disc herniation just above the LSTV and presented nucleus pulposus herniation at the transitional level.

The Castellvi classification system represented an initial effort to categorize LSTV, contributing to increasing awareness of this frequently overlooked anatomical variation. Among the advantages of its creation are the promotion of coherence and the standardization of terminology and diagnostic criteria, facilitating communication between neurosurgeons, neuroradiologists, orthopedists, and other healthcare professionals and researchers. Additionally, it offers insight into the natural history of pathologies related to the presence of LSTV, helping in patient counseling and medical decision‐making. However, the Castellvi classification has its limitations. It lacks the ability to predict the need for treatment, as it does not consider certain variables now recognized as significant in Jenkins' classification. Furthermore, the absence of studies linking this classification system to symptoms may limit its clinical usefulness. These deficiencies could increase the risk of misdiagnosis and inappropriate treatment of LSTV‐related conditions [55].

The Castellvi classification has been accepted since its creation. Despite this, Jenkins et al. [55] proposed complementing the Castellvi classification. They conducted a retrospective cohort study of 150 new patients for the treatment of back, hip, groin, and leg pain using MRI, CT, and radiography. From this study, they discovered that the Castellvi classification excludes two types of anatomical the prominent anatomical side and the possible contact of the transverse process and the iliac crest 56. Therefore, their aim was to introduce and validate a new classification system, the “Jenkins classification” for LSTV (Figure 3).

The diagnosis of LSTV is of vital clinical importance due to the multiple implications that this anatomical variation can have on the health and treatment of patients. LSTV is frequently associated with LBP, known as Bertolotti's syndrome. This pain may arise from intrinsic inflammatory processes, biomechanical alterations in the spine, or degenerative changes in the discs and facet joints, particularly in certain types of LSTV according to the Castellvi classification. Its impact is not limited to pain, as it may also influence spinopelvic alignment parameters, essential for evaluating sagittal balance and global spinal stability, which are crucial in patients with deformities or degenerative conditions. From a surgical perspective, accurate identification and proper numbering of the vertebrae are essential to avoid errors in procedures such as spinal fusions or epidural injections, where incorrect localization may compromise treatment effectiveness and lead to complications. Additionally, the presence of LSTV requires adjustments in standard evaluation protocols, designed under the assumption of five lumbar vertebrae, to ensure accurate results in therapeutic planning. Beyond these diagnostic and planning considerations, the clinical relevance of LSTV, particularly sacralization, also extends to its potential biomechanical consequences. Some findings indicate that sacralization significantly increases the risk of higher‐grade spondylolisthesis, especially at the L4–L5 level, due to the altered mechanical stress distribution [56, 57]. The fusion of L5 to the sacrum reduces motion at the lumbosacral junction, consequently placing greater strain on the adjacent superior segment. This biomechanical imbalance predisposes the L4–L5 level to degeneration, vertebral slip, and even motor deficits [57]. Such degenerative changes and instability have direct clinical implications, potentially necessitating earlier or more aggressive intervention. Therefore, recognizing and reporting LSTV is crucial not only for accurate diagnosis but also for anticipating the progression of spinal pathologies and tailoring individualized treatment strategies. However, the lack of standardization in radiological findings and difficulties in properly classifying this anomaly complicate its diagnosis, underscoring the need for a more uniform and precise diagnostic approach to optimize clinical evaluation, guide therapeutic decisions, and improve patient outcomes.

This study demonstrates the different techniques for diagnosing LSTV. Some studies agree with the parameters to consider for its diagnosis, while others vary. It is important to highlight the radiological findings used to arrive at the correct Castellvi classification, and this is discussed considering the type of diagnostic technique and the form or parameters followed. The results of this article suggest analyzing the appropriate measurement level and stable spinopelvic parameters, with the aim of facilitating understanding of the radiological variability in the evaluation and diagnosis of individuals with LSTV, along with its proper staging according to the Castellvi classification.

Usmani et al. [9, 10] described several cases and radiological findings using positron emission tomography (PET‐CT) and sodium fluoride‐18F bone CT to stage LSTV using the Castellvi classification. These cases

LSTV IIa: NaF images show an increased tracer uptake in the pelvis, corresponding to the broad transverse process of the L5 vertebra, which forms a diarthrodial joint with the sacrum (hemisacralized transverse process of L5). The volumetric image shows tracer uptake in the diarthrodial joint. Another case shows that the transaxial and coronal images extend the transverse process of the L5 vertebra, forming a diarthrodial joint with the sacrum, with increased tracer uptake in the fused PET‐CT images. The findings are consistent with osteoblastic activity in the hemisacralized transverse process of L5. LSTV IIb: CT images show the broad bilateral transverse processes of the L5 vertebra forming a diarthrodial joint with the sacrum, without increased tracer uptake in the fused PET/CT images in the hemisacralized transverse process of L5. LSTV IIIa: The CT image shows complete fusion of the transverse process on one side with the sacrum, without the corresponding uptake in the fused images.

Iplikcioglu and Karabg [21, 27, 28] use CT for the diagnosis of LSTV and described that to define the Castellvi classification grade of LSTV. Instead, Hanhivaara et al. [42] described certain subtypes not included in the original Castellvi LSTV type IIa and IIIa with contralateral enlarged transverse processes. Therefore, it is proposed that these new subtypes IIc and IIIc be consistent additions to the Castellvi classification.

Several studies showed the relationship between lumbopelvic parameters in patients with vertebral anomalies or LSTV. In the study by Desai et al.35, lumbopelvic parameters were measured, and they concluded that patients with Bertolotti's syndrome had a significantly higher PI and were more likely to present adjacent‐segment disease compared to control patients (without LSTV). However, after controlling for some variables, no significant association seemed to exist within the cohort of patients with Bertolotti.

On the other hand, Kassir et al. [11] reported a case in which bone scintigraphy with 99m Tc‐methylendiphosphonate was requested, and the planar images showed a focus of increased tracer activity in the lumbosacral area on one side over the same‐side sacroiliac joint. Therefore, a PET‐CT was requested, which showed a focus of increased tracer activity in a transverse process on the same side, consistent with the scintigraphy findings. They mention that, although the presence of a transitional vertebra can be established with simple radiography, CT, or MRI, the demonstration of stress on the transverse‐sacral joint is best achieved using bone scintigraphy with SPECT.

Regarding diagnostic techniques, Ravikanth and mahumdar [5] mention that LSTVs have traditionally been identified through lateral radiographs and Ferguson's views. These “squares” and “wedges” represent a spectrum of morphological changes of the vertebral body and cannot be reliably used to definitively identify an LSTV. Not only is the identification of an LSTV important, but the precise numerical identification of the vertebral segments on MRI is essential before surgery. Similarly, they state that it is important to correlate MRI with intraoperative radiographs to confirm the disc level during surgery. This is because it can often be problematic to determine whether an LSTV is an S1 lumbar or an L5 sacralized vertebra using MRI alone. This can lead to surgical errors when a lumbar spine MRI is reported without accompanying conventional radiographs or MRI localizers. The MRI correlation with other images is supported by Lian et al. [12], who mention that surgical errors are more likely when lumbar MRI is reported without additional images; as the use of S1–S2 disc morphology, vertebral body shape, and lumbosacral angle to define vertebral levels leads to incorrect identification of LSTV and correct spinal levels.

Overall, our review shows that the diagnostic performance of imaging modalities varies, and each has a distinct role in clinical practice. Radiography remains useful for initial vertebral numbering but lacks accuracy for classification. CT provides the highest sensitivity and specificity and should be considered the gold standard for definitive diagnosis and preoperative planning. MRI is valuable for assessing the iliolumbar ligament and coexisting soft‐tissue pathologies, though it is limited for accurate vertebral numbering. EOS offers unique advantages for whole‐spine evaluation and sagittal alignment, but its availability is restricted. Nuclear medicine techniques such as PET‐CT and bone scintigraphy have a niche role in detecting active bone remodeling in symptomatic cases. Taken together, these findings imply that imaging choice should be tailored to the clinical simple numbering in incidental LSTV, but comprehensive methodology when surgical planning, interventional procedures, or biomechanical assessments are required. This tailored approach can reduce diagnostic errors, prevent wrong‐level surgery, and improve patient outcomes.

Taking into account the discussion and the literature review, our proposal for diagnosis and classification of LSTV Initial Identification Using Diagnostic Imaging

Perform craniocaudal AP radiographs of the entire spine from C2 downward. Evaluate morphological anomalies and distinguish hypoplastic ribs from lumbar transverse processes. This allows for determining the number of thoracic segments and correctly assigning vertebral numbering. Consider imaging the entire spine as the most accurate method for numbering the vertebrae, detecting cervical, thoracic, or transitional anomalies that may be present. Define L1 and S1 as vertebrae 20 and 25 only if there are 7 cervical, 12 thoracic, and 5 lumbar vertebrae. Standard AP radiographs alone are insufficient for detecting or classifying LSTV. They should be supplemented with other imaging techniques. 2Confirmation and Classification with Computed Tomography (CT)

We consider CT to be and should remain the gold standard for the diagnosis of LSTV, as it offers the best diagnostic performance and high reliability between readers. It is especially useful for classifying LSTV based on morphology and anatomical variants. Evaluate spinopelvic parameters such as PI, pelvic tilt, lumbar lordosis, using the terminal plate closest to the normative average PI value as a reference. If the measured PI is outside the normative range, review and adjust the measurements, considering the LSTV as the lowest lumbar vertebra or the highest sacral vertebra, as appropriate 3Use of Magnetic Resonance Imaging (MRI) in Specific Cases

Perform MRI of the lumbar spine to identify iliolumbar ligaments, which commonly emerge from the transverse processes of L5. Their position can be used to assign lumbar levels in patients with LSTV. In patients with LSTV, if the iliolumbar ligament is not identified at the superior level, the LSTV may be considered L5. If the ligament arises above the LSTV, it is numbered as S1. Use axial T1 and T2‐weighted images to observe the iliolumbar ligament as a low‐signal intensity structure. 4Morphological Analysis for Classification and Surgical Planning

Identify key morphological features such as hypoplastic or absent facet joints in the L5‐S1 fusion and facet joints between S1‐S2 in cases of lumbarization. Use the Jenkins LSTV classification for typification and therefore for its respective use in surgical planning. 5Functional Evaluation and Biomechanical Effects

Determine the influence of LSTV on the biomechanics of the spine, evaluating its impact on lumbar muscle degeneration and reduced muscle volume. 6Additional Examinations in Complex Cases

If there are anomalous segmentations along with LSTV, identifying the iliolumbar ligament may not be enough. In these cases, consider the possibility of performing additional comprehensive examinations with MRI or CT for a more accurate assessment.

Our proposed diagnostic methodology is intended for radiologists, neurosurgeons, spine surgeons, and interventional pain specialists in cases where precise vertebral level identification is critical for procedures (e.g., spinal fusions, epidural injections, sagittal balance analysis). It is not necessary for incidental, asymptomatic LSTV detected during imaging for unrelated reasons. Instead, it should be reserved for patients with symptomatic LSTV, those undergoing surgical planning, or cases where misidentification could result in procedural errors.

We believe that this approach allows for a comprehensive, accurate diagnosis and is useful for therapeutic planning, reducing errors, and improving clinical outcomes in patients with LSTV.

This study has some limitations, including an incomplete literature search, as other databases may contain articles with a high impact on the objective of this study. Additionally, many of the studies had incomplete information, as they only mentioned the diagnostic technique used to detect LSTV and did not specify how the diagnosis was made or which radiological findings were used. Furthermore, there may be selection bias as the search was limited to title and abstract, suggesting that some relevant articles may have been excluded.

Future research on LSTV should aim to develop a comprehensive meta‐analysis that synthesizes current evidence and supports the creation of standardized imaging protocols and diagnostic criteria, including the consistent application of Castellvi classification. Large‐scale, prospective, multicenter studies are needed to compare the diagnostic performance of CT, MRI, and radiography, using standardized vertebral numbering and detailed spinopelvic measurements. Incorporating artificial intelligence and machine learning into radiological assessment could reduce diagnostic errors and interobserver variability. Additionally, further investigation into the functional impact of different LSTV subtypes, particularly in interventional procedures and biomechanical evaluations, is warranted. Emerging imaging techniques like EOS should be evaluated for their utility in assessing sagittal alignment and spinopelvic parameters. Lastly, as anatomical variability continues to challenge existing classification systems, future studies should explore potential refinements to the Castellvi classification or integrate complementary systems such as Jenkins' to enhance clinical relevance and diagnostic precision. Therefore, this study shows the need for a meta‐analysis on this topic to reach a possible consensus for diagnosing LSTV and correctly using the Castellvi classification.

Conclusions

In conclusion, LSTV is a common condition, and many patients may have anatomical variants in the spine. This has led to varying literature regarding the most appropriate diagnostic technique and the aspects to consider. Each diagnostic technique should be evaluated, considering its sensitivity and specificity; this would provide clearer information for diagnosing LSTV, especially in patients with anatomical variants. When choosing the diagnostic technique to use, healthcare professionals should consider measuring spinopelvic parameters in these patients, identifying anatomical landmarks accurately, and ensuring that the image is taken in the correct position and manner. This would lead to a more accurate diagnosis and allow for a more secure classification using the Castellvi or Jenkins classification. Our proposed methodology for diagnosing LSTV first, using simple AP radiographs to perform craniocaudal vertebral numbering and assess morphological anomalies. Second, if there is suspicion, perform a CT as the gold standard for diagnosing LSTV, considering its high sensitivity and specificity; and measure spinopelvic parameters to correlate with LSTV. Third, use MRI in special cases. Fourth, perform a morphological analysis and utilize the Jenkins classification for LSTV classification.

Author Contributions

Both authors confirm responsibility for the study conception and design, data collection, analysis and interpretation of results, and manuscript preparation. All authors reviewed the results and approved the final version of the manuscript.

Funding

The authors have nothing to report.

Disclosure

All figures included in this paper are the author's own work and have been created specifically for the purposes of this study. No copyrighted or third‐party material has been used in this study.

Ethics Statement

The authors have nothing to report.

Consent

This study did not require informed consent for participation, as there was no direct contact with patients. The research was conducted as a systematic review of previously published studies, all of which obtained informed consent from their respective participants. Since our analysis is based solely on publicly available data and does not involve new data collection or interaction with individuals, there are no ethical concerns or implications requiring additional consent.

Conflicts of Interest

The authors declare no conflicts of interest.