Sagittal Double-Hump Deformity of the Lumbosacral Spine: An Anatomical Risk Factor for Surgical Landmark Obscuration During Laparoscopic Sacrocolpopexy.

Hirotaka Sato, Miki Kurita, and Shota Otsuka contributed to protocol and project development data collection, data management, and analysis. Hirotaka Sato, Hirokazu Abe and Sachiyuki Tsukada contributed to manuscript writing and editing. All authors approved the final manuscript.

This study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (approval number: 2024‐097).

The authors received no specific funding for this work.

All patients provided written informed consent prior to participation. The consent process included a detailed explanation of the study objectives, procedures, and potential risks, including the use of contrast‐enhanced CT imaging. The study protocol was approved by the Institutional Review Board of Hokusuikai Kinen Hospital (Approval No. 2024‐097) and was conducted in accordance with the ethical standards of the Declaration of Helsinki and its subsequent amendments.

A total of 197 patients diagnosed with POP and scheduled to undergo LSC between October 2023 and December 2024 underwent preoperative CT. After applying the exclusion criteria, which included conversion to vaginal surgery ( n  = 7), obliteration of the pouch of Douglas because of endometriosis ( n  = 3), extensive peritoneal adhesions resulting from prior laparotomy ( n  = 4), refusal to participate ( n  = 2), indications for native tissue repair ( n  = 2), and intraoperative conversion to conservative management because of severe adhesions ( n  = 2), 184 female patients were included in the final analysis (Figure  3 ). Among these patients, 52 (28%) met the criteria for the double‐hump deformity based on sagittal CT imaging results. Demographic and perioperative characteristics were comparable between groups (Table  1 ), although the double‐hump deformity group demonstrated significantly greater estimated blood loss than that of the control group (median: 10 mL [IQR, 9–16 mL] vs. 10 mL [IQR, 5–10 mL]; p  = 0.008). Distribution of the study patients. Baseline demographic and operative characteristics of the study patients. Note: Values are presented as mean ± SD, median (IQR), or n (%). Abbreviations: BMI, body mass index; CI, confidence interval; EBL, estimated blood loss; IQR, interquartile range; LAVH, laparoscopic assisted vaginal hysterectomy; OT, operative time; POP‐Q, Pelvic Organ Prolapse Quantification; SCH, supracervical hysterectomy; SD, standard deviation. CT‐based anatomical measurements are summarized in Table  2 . Compared with the control group, the double‐hump deformity group exhibited a significantly reduced L5–S1 disk height (8.2 ± 3.1 mm vs. 10.9 ± 3.8 mm; p  < 0.001) and L5–S1 angle (39.0 ± 9.8° vs. 56.1 ± 8.8°; p  < 0.001). Conversely, the L4–L5 angle of the double‐hump deformity group was significantly greater than that of the control group (20.6 ± 4.9° vs. 16.4 ± 6.4°; p  < 0.001); however, the L4–L5 disk height was similar between groups. Anatomical measurements of the study patients derived from computed tomography images. Note: Values are presented as mean ± SD, median (IQR), or n (%). Abbreviations: CI, confidence interval; IQR, interquartile range; SD, standard deviation. Among radiographic signs of degeneration, L5–S1 disk narrowing (44% vs. 27%; p  = 0.044) and L5 spondylolisthesis (14% vs. 2%; p  = 0.006) were significantly more common in the double‐hump deformity group than in the control group. Osteophyte formation was observed in 38% of the double‐hump deformity group and 23% of the control group; however, this difference was not statistically significant ( p  = 0.083) (Figure  4 ). Degenerative spinal changes, including vertebral osteophytes, intervertebral disk space narrowing, and L5–S1 spondylolisthesis, were assessed using preoperative computed tomography (CT). Left: Normal lumbosacral anatomy (labeled for reference). Right: Abnormal lumbosacral anatomy. (a) Intervertebral disk space narrowing. (b) L5–S1 spondylolisthesis. (c) Vertebral osteophytes. A Pearson correlation analysis demonstrated a moderate positive association between the L5–S1 angle and L5–S1 disk height (r = 0.52; 95% CI, 0.40–0.62; p  < 0.001) and a weak inverse correlation between the L4–L5 angle and L5–S1 angle (r = −0.15; 95% CI, −0.29 to −0.0011; p  = 0.048) (Table  3 ). Pearson's correlation between the L5–S1 angle and L5–S1 disk height and between the L4‐L5 angle and L5–S1 angle. Abbreviations: CI, confidence interval; r , Pearson's correlation coefficient. According to the multivariable ANCOVA, the presence of double‐hump deformity was significantly associated with a smaller L5–S1 angle (estimate, –14.1°; 95% CI, –17.3 to –10.8; p  < 0.001) (Table  4 ), whereas greater L5–S1 disk height ( p  < 0.001) and L4–L5 disk height ( p  = 0.028) were significantly associated with a larger L5–S1 angle. Among the degenerative parameters, L5–S1 osteophytes approached significance ( p  = 0.053), whereas narrowing and spondylolisthesis were not significantly associated in the multivariable model. Factors associated with the L5–S1 angle: analysis of covariance of patients who underwent preoperative computed tomography prior to laparoscopic sacrocolpopexy. Abbreviations: BMI, body mass index; CI, confidence interval. An intraoperative video review of the 52 double‐hump deformity cases revealed that sacral promontory visualization was rated as “difficult” in 26 patients (50%), “moderate” in 16 patients (31%), and “easy” in 10 patients (19%) by blinded laparoscopic surgeons. Figure  5 presents laparoscopic views and corresponding CT images of cases with normal spinal anatomy and those with double‐hump deformity. Laparoscopic views (upper panels) and corresponding sagittal computed tomography (CT) images (lower panels) of normal anatomy (left) and the double‐hump deformity (right). Left: Normal anatomy. The white arrow indicates the sacral promontory. Right: Double‐hump deformity. The sacral promontory is indicated by a white arrow. The iliac arteries and iliac veins are annotated where clearly visible. Postoperative CT imaging allowed clear visualization of the mesh course in 40 of 52 (77%) patients with double‐hump deformity. Although complete tracking of the entire mesh path was not possible for all cases, no evidence of intervertebral fixation was observed with postoperative CT in 52 double‐hump cases. All anchors were located at or below the S1 superior endplate and sacral promontory without radiodense tracks entering the L4–L5 disk space. Figure  6 presents representative CT images of the mesh course. Postoperative computed tomography (CT) images demonstrating the mesh course in patients with a double‐hump deformity. Representative cases in which the mesh pathway was clearly visualized are shown. Arrows and dots delineate the mesh course relative to the sacral promontory and S1 endplate. Mesh tracking was possible in 40 of the 52 (77%) patients. No cases fulfilled the predefined CT criteria for malposition (disk fixation or radiodense track entering the L4–L5 disk space). These results suggest that double‐hump deformity and its associated spinal degeneration significantly affect L5–S1 angulation and may influence surgical landmark visibility during LSC.

This study demonstrated that the presence of double‐hump deformity—a bimodal curvature pattern with anterior protrusions at both the L4–L5 and L5–S1 levels—is significantly associated with reduced L5–S1 angulation and advanced spinal degeneration. Based on the ANCOVA results, the double‐hump deformity was independently associated with a 14.1°　reduction in L5–S1 angle, whereas greater L5–S1 and L4–L5 disk heights were associated with increases of 1.01° and 0.58°, respectively. These findings highlight an important anatomical consideration for LSC because reduced L5–S1 angulation may obscure the sacral promontory and complicate intraoperative orientation. The importance of this configuration was further underscored by our intraoperative video review during which sacral promontory visualization was considered “difficult” in 50% of double‐hump deformity cases by blinded laparoscopic surgeons. This finding provides direct clinical evidence linking the deformity to landmark visibility challenges during LSC. The anatomical basis of this difficulty may be associated with the relative prominence of the L4–L5 intervertebral disk, which can appear more anteriorly projected than the true sacral promontory in patients with double‐hump deformity. Such misidentification may lead to inadvertent mesh fixation to the L4–L5 disk, which is a serious surgical error associated with complications such as discitis, osteomyelitis, and chronic pelvic pain [ 16 ]. Although postoperative CT imaging enabled full mesh tracking in only 77% (40/52) of the double‐hump cases, radiological evidence of mesh misplacement based on our predefined CT criteria (intervertebral disk fixation or radiodense track entering the L4–L5 disk space) was not observed in any patient. These findings support the notion that preoperative recognition of double‐hump anatomy combined with careful dissection and tactile verification during surgery can mitigate the risk of erroneous fixation. An additional surgical consideration is vascular variability at the sacral promontory. Atypical courses of the common iliac veins or arteries can obscure the true bony landmark and increase the risk of inadvertent vascular injury during mesh fixation. Recognition of these variants in patients with double‐hump deformity is particularly important because their anatomical orientation is already compromised. Therefore, careful preoperative imaging and meticulous dissection are essential to avoiding complications. Additionally, our analysis revealed a significant positive correlation between the L5–S1 angle and L5–S1 disk height ( r  = 0.52; p  < 0.001), reinforcing the anatomical relationship between sagittal alignment and disk morphology [ 10 ]. A weak but statistically significant negative correlation was also found between the L4–L5 and L5–S1 angles ( r = −0.15; p  = 0.048), suggesting a compensatory curvature mechanism underlying the double‐hump configuration [ 17 ]. These data add to the previous literature that described the variability of sacral anatomy and its impact on pelvic surgery. Giraudet et al. emphasized the anatomical complexities of the sacral promontory and their implications for safe mesh fixation during LSC [ 4 ]. Additionally, varying pelvic and sacral slopes have been recognized as significant determinants of sacral landmark exposure during lumbar‐pelvic procedures [ 18 ]. Unlike previous studies that primarily focused on the sacral slope and pelvic tilt, our study provides quantitative criteria for identifying the double‐hump deformity and clarifies its association with surgical difficulty. Our findings emphasize the need to preoperatively assess lumbosacral curvature, particularly in those with radiographic signs of degeneration such as narrowing of the disk and spondylolisthesis. Agrawal et al. demonstrated that the aorta‐sacral promontory distance significantly decreases with age and hypertension, suggesting that anatomical variation may impact surgical landmark visibility [ 19 ]. Similarly, Baker et al. highlighted that variability in sacral and pelvic tilt angles is a critical determinant of sacral promontory exposure during lumbar‐pelvic procedures [ 20 ]. Importantly, we propose the following simple radiological definition of double‐hump deformity: an L4–L5 angle > 15°, indicating moderate to marked lordosis at the upper lumbar level, and an L5–S1 angle < 50°, indicating a tendency toward flattening at the lumbosacral junction. These thresholds were based on the distribution of measurements of our cohort. This definition allows consistent identification of the double‐hump deformity and may assist with preoperative risk stratification. Limitations of our study include the subjective nature of intraoperative video scoring, absence of interobserver reliability for the double‐hump deformity diagnosis, and incomplete mesh tracking for a subset of patients. Additionally, CT was the sole imaging modality used in this study. Although CT provides high spatial resolution for disk height and disk angle measurements and postoperative mesh visualization, it involves radiation exposure, which is particularly relevant for premenopausal women. MRI is a radiation‐free alternative; however, its visualization of bony landmarks for angular measurements lacks clarity, and ultrasound is limited in lumbosacral junction assessments. In selected cases, non‐contrast or low‐dose CT protocols may mitigate radiation exposure. Nonetheless, the consistent radiographic trends and observed intraoperative challenges provide compelling support of the clinical relevance of this anatomical variant. Future studies that incorporate three‐dimensional preoperative planning and intraoperative navigation may further enhance the safety of sacrocolpopexy for patients with atypical spinal anatomy. Additionally, between‐group comparisons of the final anchoring position were not performed because postoperative CT was limited to double‐hump cases. This design choice was based on our a priori focus on anatomical risk characterization. Furthermore, standardized postoperative pain scores were not prospectively collected. This absence precluded between‐group comparisons, but we highlighted pain as a non‐primary endpoint for future prospective studies. Moreover, short‐term surgical success (e.g., early anatomical outcomes or symptom relief) was outside the prespecified scope of this study. Uniform early postoperative POP‐Q assessment results were not available for all patients; therefore, between‐group comparisons could not be performed. The current analysis was limited to preoperative anatomy and intraoperative visibility in relation to L5–S1 angulation and the mesh course. However, this study was not powered to detect between‐group differences in non‐primary clinical endpoints such as postoperative pain or short‐term success. Thus, conclusions regarding these outcomes should be interpreted with caution.

The double‐hump deformity—defined radiologically as an L4–L5 angle > 15° and an L5–S1 angle < 50°—is significantly associated with reduced L5–S1 angulation and an increased prevalence of spinal degenerative changes. This anatomical configuration may impair intraoperative visualization of the sacral promontory during LSC, thus increasing the risk of surgical difficulty. Preoperative identification of this deformity using CT imaging may aid in surgical planning and reduce the likelihood of mesh misplacement, particularly for patients with atypical spinal anatomy.

Laparoscopic sacrocolpopexy (LSC) is a well‐established surgical treatment for pelvic organ prolapse (POP) that provides durable anatomical and functional outcomes through mesh attachment to the sacral promontory [ 1 , 2 ]. Despite its high success rate, accurate identification of the sacral promontory and surrounding structures is essential to ensuring safe and effective mesh placement [ 3 , 4 ]. Spinal degenerative changes can lead to variations in lumbosacral morphology, potentially obscuring key surgical landmarks and increasing the technical difficulty of LSC [ 5 , 6 ]. Recent studies have underscored the relevance of sagittal spinal alignment during pelvic reconstructive surgery because altered lumbosacral angulation may hinder identification of the L5–S1 junction [ 7 ]. Such variations may increase the risk of inadvertent mesh fixation to intervertebral disks, which can result in serious complications, including discitis, osteomyelitis, and chronic pelvic pain [ 5 , 8 , 9 ]. However, the effects of spinal anatomical variations on intraoperative landmark visibility during LSC have not been systematically investigated. One specific configuration of interest is the “double‐hump deformity,” which is defined by anterior protrusions at both the L4–L5 and L5–S1 disks. This pattern reflects a combination of L4–L5 hyperlordosis (primary hump) and L5–S1 segmental flattening (secondary hump) that often results from disk degeneration [ 10 , 11 ]. This bimodal sagittal imbalance may impair sacral promontory exposure, particularly in cases with reduced L5–S1 angulation. Although previous studies have examined the influence of pelvic floor morphology and sacral slope on pelvic disorders [ 12 ], the specific anatomical relationship between the double‐hump deformity and factors relevant to LSC remains unclear. This study aimed to assess the prevalence and clinical implications of the double‐hump deformity using sagittal computed tomography (CT) imaging for women who underwent LSC for POP. We hypothesized that this deformity would be associated with reduced L5–S1 angulation and limited visibility of the sacral promontory, thereby potentially increasing the risk of improper mesh placement. Through detailed radiographic assessments and multivariable analyses, we sought to identify anatomical predictors of L5–S1 angulation that may support improved preoperative planning for pelvic reconstructive surgery.

The authors declare no conflicts of interest.

This prospective observational study was conducted at Hokusuikai Kinen Hospital (Mito, Ibaraki, Japan) between October 2023 and December 2024. Consecutive female patients scheduled to undergo LSC were enrolled. All participants underwent preoperative CT to assess lumbar spine morphology. Patients with a history of lumbar spine surgery or incomplete CT data were excluded. The study protocol was approved by the local ethics committee (approval number: 2024‐097), and written informed consent was obtained from all participants. The consent process included explanations regarding the purpose of the study, potential risks and benefits of contrast‐enhanced CT, and institutional safety policies. The study was conducted in accordance with the Declaration of Helsinki. The L5–S1 intervertebral angle measured using sagittal CT was the primary outcome. Intraoperative sacral promontory visibility, postoperative CT findings of the mesh course and mesh malposition, and perioperative surgical measures (operative time, estimated blood loss) were key secondary outcomes. Degenerative spinal findings (osteophytes, disk narrowing, spondylolisthesis) and their associations with the L5–S1 angle were exploratory outcomes. Patient‐reported postoperative pain was not a predefined outcome in this exploratory imaging–anatomy study. All participants were informed about radiation exposure during CT. Radiation exposure was minimized using standard low‐dose protocols when applicable. The study protocol was approved by the ethics committee. Baseline demographic information and clinical information, including age, height, weight, body mass index (BMI), parity, history of tobacco use, comorbidities (hypertension and diabetes), and prior abdominal or gynecological surgery, were prospectively collected at the time of surgical scheduling. Surgical data included operative time, estimated blood loss, type of hysterectomy (if applicable), and Pelvic Organ Prolapse Quantification (POP‐Q) stage [ 13 ]. Preoperative CT imaging was conducted as part of the routine preoperative assessment. All CT images were reviewed using dedicated radiology software. Sagittal images (1‐mm slices, urographic phase) were used to measure intervertebral angles and disk heights at L4–L5 and L5–S1 as well as the distance from the midpoint of the pubic symphysis to the center of the sacral promontory. The presence of a double‐hump deformity was determined when two independent reviewers with clinical and radiological expertise reached a consensus based on predefined imaging criteria. All data were anonymized and stored in a secure electronic database. Data entry was performed by a trained research physician who was not involved in the study to ensure completeness and accuracy. CT was selected because it offers high spatial resolution to quantify disk heights and angles and visualize the postoperative mesh course. Alternatives such as magnetic resonance imaging (MRI) do not provide sufficient clarity of bony landmarks for angular measurements, and ultrasound is limited in L4–S1 alignment assessments. Sagittal CT images were used to measure the following parameters (Figure  1 ): L4–L5 disk height (Figure  1a ). L5–S1 disk height (Figure  1b ). L4–L5 angle (Figure  1c ). L5–S1 angle (Figure  1d ). Distance from the midpoint of the upper edge of the pubic symphysis to the center of the sacral promontory (Figure  1e ). L4–L5 disk height (Figure  1a ). L5–S1 disk height (Figure  1b ). L4–L5 angle (Figure  1c ). L5–S1 angle (Figure  1d ). Distance from the midpoint of the upper edge of the pubic symphysis to the center of the sacral promontory (Figure  1e ). The following parameters were measured using sagittal computed tomography (CT) images: (a) intervertebral disk height at L4–L5; (b) intervertebral disk height at L5–S1; (c) intervertebral angle at L4–L5; (d) intervertebral angle at L5–S1; and (e) distance from the midpoint of the upper edge of the pubic symphysis to the center of the sacral promontory. All patients underwent multi‐phase CT urography using an 80‐detector row scanner. The protocol included non‐contrast, arterial, nephrographic, and excretory phases. Non‐contrast and nephrographic phase images were reconstructed with a thickness of 5 mm. Excretory phase images were reconstructed with thicknesses of 5 mm and 1 mm to improve measurement precision. A dedicated three‐dimensional workstation was used to generate multiplanar reconstructions. The midpoint of the sacral promontory was identified in the sagittal view and served as a reference for all measurements. The double‐hump deformity was defined based on combined morphological and angular characteristics observed on sagittal CT images. Specifically, the double‐hump deformity was considered present when both of the following conditions were met: L4–L5 angle > 15° (indicating moderate to marked lordosis at the upper lumbar level). L5–S1 angle  15° (indicating moderate to marked lordosis at the upper lumbar level). L5–S1 angle < 50° (indicating a trend toward flattening at the lumbosacral junction). These slightly relaxed thresholds were selected based on distributional characteristics of our cohort to improve sensitivity while preserving anatomical specificity. This definition was established to objectively identify the characteristic bimodal curvature pattern that may complicate surgical visualization of the sacral promontory. Figure  2 presents representative CT images of double‐hump cases with the maximum, average, and minimum L5–S1 angles to aid in the visual interpretation of the diagnostic criteria. Computed tomography (CT) sagittal images illustrating the anatomical characteristics of the double‐hump deformity showing the minimum, maximum, and average angles of descent at the L4–L5 and L5–S1 levels. (Left) Maximum L5–S1 angle of 49.5° and minimum L4–L5 angle of 15.5°. (Middle) Average L5–S1 angle of 39.5° and average L4–L5 angle of 26.0°. (Right) Minimum L5–S1 angle of 24.0° and maximum L4–L5 angle of 28.0°. (a) L5–S1 angle. (b) L4–L5 angle. Measurements were independently performed by a radiologist (M.K.) and a urologist (H.S.). The average of their two measurements was used in the analysis. Each reviewer independently assessed the images, and cases were classified as double‐hump deformity only when the reviewers agreed. The sacral promontory was defined as the uppermost anterior point of the S1 vertebra. Intraoperative videos of patients with double‐hump deformity ( n  = 52) were independently reviewed by a blinded laparoscopic surgery expert who was not involved in the procedures [ 14 ]. The visibility of the sacral promontory was rated using a 3‐point scale (easy, moderate, or difficult). Postoperative CT imaging of double‐hump cases was performed to evaluate mesh malposition. The final anchoring level was defined relative to the S1 superior endplate and the sacral promontory [ 15 ]. Prespecified findings that resulted in suspicion of intervertebral fixation were an anchor or sutures traversing the L4–L5 disk region or a radiodense track entering the disk space. All images were independently reviewed by two radiologists, and discrepancies were adjudicated by a consensus. In accordance with the study protocol, postoperative CT images were only obtained from the double‐hump group to characterize potential risks. Therefore, between‐group comparisons of the anchoring level were not feasible. Additional degenerative spinal changes, including the presence of vertebral osteophytes, intervertebral disk space narrowing, and L5–S1 spondylolisthesis, were evaluated using preoperative CT. The relationship between these findings and the presence of double‐hump deformity was analyzed. This exploratory, hypothesis‐generating study aimed to identify anatomical predictors of reduced L5–S1 angulation. Continuous variables were reported as means ± standard deviations (SDs), medians with interquartile ranges (IQRs), or 95% confidence intervals (CIs). Categorical variables were summarized as counts and percentages. Pearson's correlation coefficients were calculated to assess associations between anatomical parameters, including the relationships between the L4–L5 and L5–S1 angles and between the L5–S1 angle and disk height. An analysis of covariance (ANCOVA) was performed to evaluate the association between the double‐hump deformity and L5–S1 angle after adjusting for age, BMI, smoking history, L4–L5 angle, L4–L5 disk height, L5–S1 disk height, and additional degenerative findings, including L5–S1 disk space narrowing, L5–S1 osteophytes, and L5–S1 spondylolisthesis. Multicollinearity was assessed using variance inflation factors (all of which were < 2). All tests were two‐sided, and statistical significance was set at p  < 0.05. Analyses were conducted using R software (R Foundation for Statistical Computing, Vienna, Austria) and EZR (Saitama Medical Center, Jichi Medical University, Japan). Sensitivity analyses were not performed because this was an exploratory study. Although a formal a priori sample size calculation was not performed, precision‐based post hoc considerations indicated that a sample size of 184 (double‐hump group, n  = 52; control group, n  = 132) and an observed L5–S1 angle with an SD of 9° to 12° would ensure sufficient precision to detect the observed between‐group difference (approximately 17°) with a two‐sided α value of 0.05.