Special type of distal junctional failure exhibits pelvic incidence changes: sacroiliac joint-related pain following lumbar spine surgery.

Background: Currently, change in pelvic incidence (PI) in patients after spinal surgery have not been associated with clear clinical symptoms. This study sought to compare changes in the sagittal parameters of different patients before and after thoracolumbar spine surgery, the relationship between PI change and sacroiliac joint pain (SIJP) after surgery was clarified, and the correlation between PI change and sacroiliac joint (SIJ) activity was verified.
Methods: This study retrospectively analyzed the data of patients who underwent thoracolumbar fusion at Sun Yat-sen Memorial Hospital from January 2019 to June 2021. The spinal and pelvic parameters [including pelvic tilt (PT), sacral slope (SS), PI, lumbar lordosis (LL) angle, etc.] of 409 patients with standard standing lateral radiographs before and after surgery were compared and analyzed. Postoperative follow-up of all patients with standardized SIJP assessment. The incidence of postoperative SIJP, and its correlation with sagittal parameters of the spine and pelvis, surgical methods, and the basic characteristics of patients were analyzed. The Chi-square test was used for categorical variables, the independent-sample t-test was used for generally conformed normally distributed continuous variables. Risk factors associated with the development of SIJP were analyzed using logistics regression. Correlations among SS, PI, and the 4 other sagittal parameters were analyzed using the Pearson correlation coefficient (r).
Results: Postoperative PI changes tended to be larger in the lowest instrumented vertebra (LIV) (L4 and above: 1.63°; L5: 2.43°; S1: 3.83°; P<0.05) and longer fixed segment. The risk factors for SIJP included a PI >4° [odds ratio (OR) =13.051; P<0.001], LIV S1 (OR =3.378; P=0.023), and fixed total segment ≥3 (OR =2.632; P=0.038). ∆PI was significantly correlated with ∆SS in patients with non-S1 distal fixation vertebrae (R2=0.388; P<0.01), but no such correlation was found in patients with S1 distal fixation vertebrate. Conclusions: Changes in PI values after thoracolumbar spine surgery can correctly reflect the motion state of the SIJ. Excessive changes in PI (>4°) are similar to the mechanism of distal junctional kyphosis (DJK), while such changes make patients prone to SIJP following lumbar spine surgery. 2022 Annals of Translational Medicine. All rights reserved.

Introduction

Complications due to mechanical stress abnormalities in the adjacent segments after spinal fusion have been extensively examined and described in the literature, but distal junctional kyphosis and failure (DJK/DJF) have received less attention in the literature (1). DJK is defined as a distal junctional angle of more than 10 degrees and 10 or more degrees compared to the preoperative measurement, as measured from the caudal endplate of the lowest instrumented vertebra (LIV) and the caudal endplate of 1 vertebra below (2). The main risk factors for the development of DJK in patients after surgery are currently considered to include: improper distal fixed vertebral selection, the neglect of the overall sagittal balance, the misalignment of spinal-pelvic parameters with pelvic incidence (PI), surgical access, and improper intraoperative manipulation (3-7). However, the current definition of DJK/DJF is not sufficient to explain the problems associated with internal fixation failure that occurs in some cases where S1 fixation has been performed (8).

Previously, PI was considered constant in adulthood (9). However, several studies have reported that PI changes with age and that spinal surgery can also change PI (10-13). PI values may increase with age for a number of reasons, including sacroiliac joint (SIJ) laxity, hip joint deformation, and the long-term weight-bearing morphology of the ilium. Additionally, the magnitude of PI may also change after spinal surgery, and the main cause of its change is SIJ.

The SIJ is a typical diarthrodial synovial joint that is considered the most mechanically stable joint in the entire spinal-pelvic region (14). In adults, the SIJ is a minimally mobile joint; however, its mobility is not negligible in physiological and pathological states. This SIJ motion occurs mainly in the sagittal plane, which is defined as nutation and counternutation, and often ranges from 1–4° and has a translation from 0.5–2 mm (15,16). The PI is the angle between the perpendicular to the upper plate of S1 in its middle and the line joining this point to the bi-coxo-femoral axis. However, as its measurement line passes through the SIJ, the nutation and counternutation motion of the SIJ affect the PI value anatomically (17-19).

Currently, we do not know how much affect SIJ sagittal plane activity has in the occurrence of changes in PI, and the association of SIJ movement with PI changes has also not been reported (20). Previous research suggests that the possible adverse effects of such postoperative PI changes include preoperatively measured sagittal parameters, such as the optimal lumbar lordosis (LL) angle and the sagittal vertical axis, becoming inaccurate due to altered PI values (6,11). New local adverse symptoms of SIJ, such as the exacerbation of SIJ degenerative changes and the development of sacroiliac joint pain (SIJP), may occur. Usually, SIJP has a clear stimulus factor, and in the absence of trauma, tumor or metabolic disease, abnormal local biomechanical changes, including spinal surgery, lower extremity inequality, and pregnancy, are often the underlying causes of SIJP (5). This abnormal stress is often manifested in the sagittal position as abnormal SIJ activity (21). However, no study has clarified the pattern of PI alterations after lumbar spine surgery or its association with SIJ dyskinesia. This study sought to establish the association between postoperative PI changes and SIJP and to verify the correlation between PI changes and SIJ activity. We present the following article in accordance with the STROBE reporting checklist (available at https://atm.amegroups.com/article/view/10.21037/atm-22-2413/rc).

Methods

Subjects

The clinical research design used a cross-sectional study. We performed a retrospective review of the data of consecutive Chinese Han patients who underwent posterior thoracolumbar fusion at Sun Yat-sen Memorial Hospital from January 2019 to June 2021. To be eligible for inclusion in the study, the patients had to meet the following inclusion criteria: (I) be aged >20 years; (II) have preoperative and 3-month postoperative lumbar lateral radiographs in the standing position in which the bilateral femoral head is clearly visualized; and (III) have undergone a single successful surgery with ideal screw placement and no revision or infection. Patients were excluded from the study if they met any of the following exclusion criteria: (I) had significant inequality in both lower extremities, severe injuries, or had undergone bone and joint surgery; (II) had structural damage to the SIJ caused by trauma, surgery, tumors, or other diseases; (III) had been classified as suffering preoperative SIJP based on our diagnostic criteria; and/or (IV) had sacralization and sacral lumbarization.

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by Medical Ethics Committee of Sun Yat-sen Memorial Hospital of Sun Yat-sen University (No. SYSEC-KY-KS-2022-091) and written informed consent was obtained from all patients.

Radiographic measurements

The imaging data of the patients in this cross-sectional study had to meet the following criteria: include standing lateral radiographs of the spine and pelvis in which the bilateral femoral head was clearly visualized. These radiographs were made with the patient in a natural standing position with the knees in natural flexion for adaptive balance to the kyphosis and the hands overlying the ipsilateral clavicles. A senior resident orthopedist and another resident orthopedist with special training performed the measurements of all the parameters, and the average value was taken as the final measurement. The measurement parameters included: (I) LL: the angle between the superior end plates of T12 and the inferior plates S1 [a negative value indicated kyphosis (22)]; (II) PI: the angle between a line perpendicular to the sacral plate and a line joining the middle of the sacral plate and the hip axis (23); (III) pelvic tilt (PT): the angle between a vertical line from the hip axis and a line from the middle of the sacral plate [PT was positive if the hip axis was located anteriorly to the middle of the sacral plate and was negative if it was located posteriorly (23)]; (IV) sacral slope (SS): the angle between a line along the sacral plate and a horizontal line from the posterior corner of the sacral plate (23); (V) ∆PI: the difference between the postoperative PI and the preoperative PI (∆PI was positive if the PI value increased postoperatively and was negative if it decreased postoperatively); (VI) ∆SS: the difference between postoperative SS and preoperative SS; and (VII) PI-LL.

Diagnosing of postoperative SIJ‑related pain

Patients were diagnosed with SIJ-related pain following lumbar spine surgery if they: (I) experienced pain within 2 years of surgery below the L5 spinous process, buttocks, posterior thighs and groin area, and had a SIJ score based on (i) one-finger test (3 scores), (ii) groin pain (2 scores), (iii) pain while sitting on a chair (1 score), (iv) SIJ shear test (1 score), (v) tenderness of posterosuperior iliac spine (1 score), or (vi) tenderness of sacrotuberous ligament (1 score) (24,25), the scores, ranging from 0 to 9 points, had a cutoff value of 4; (II) had no residual compression findings of the nerve roots and cauda equina on lumbar magnetic resonance imaging; and (III) had 3 or more provocation tests that were positive in 6 specialized physical diagnostic tests [i.e., the FABER (flexion, abduction, external rotation), gapping test/distraction test, compression test/approximation test, thigh thrust test/femoral shear test, Gaenslen test/pelvic torsion test, and sacral thrust test/sacral base spring test] (26,27). If necessary, an SIJ intra-articular block was performed for patients with a difficult final diagnosis, and SIJP was diagnosed as 70% pain relief within 3 hours (28-32).

Statistical analysis

SPSS (version 25.0 SPSS Inc., Chicago, IL, USA) was used for the data analysis. The data obtained for lumbar-pelvic parameter generally conformed to a normal distribution. Inter-group differences were evaluated using the independent sample t-test. Differences in parameters among the groups were analyzed using the independent samples t-test and the least significant difference (LSD) t-test. Risk factors associated with the development of SIJP were analyzed using logistics regression. Correlations among SS, PI, and the 4 other sagittal parameters were analyzed using the Pearson correlation coefficient (r). A P value <0.05 was considered significant.

Results

Change in sagittal parameters with surgery

We enrolled 409 patients (see ) in this study. Of the 409 patients, 23 (15 female and 8 male) were diagnosed with SIJP. The patients had a mean age of 61.3±11.2 years. The mean values of each preoperative spinal-pelvic parameter were not statistically different compared to the postoperative values (see ).

Table 1

Patient information summary

Factors	Patients (n=409)
UIV
T4	1
T9	2
T10	5
T11	1
T12	6
L1	18
L2	37
L3	95
L4	194
L5	50
LIV
L2	5
L3	2
L4	16
L5	173
S1	213
Seg
1	159
2	148
3	58
4	25
5	10
6	5
7	1
8	2
13	1
Age (years)
20–40	22
41–60	139
61–75	215
>75	33
Gender
Male	178
Female	231
Etiology
Lumbar spinal stenosis	152
Lumbar disc herniation	126
Spinal deformity	23
Fracture of the spine	15
Lumbar spondylolisthesis	88
Spinal benign neoplasms	5

UIV, upper instrumented vertebra; LIV, lowest instrumented vertebra; Seg, segment.

Table 2

Comparison of preoperative and postoperative mean values of individual parameters in all patients

Radiographic parameter	Preoperative	Postoperative	P value
PI	50.9±11.0	50.7±10.8	0.778
PT	17.6±8.9	17.0±8.2	0.308
SS	33.6±10.9	33.8±8.8	0.775
LL	44.9±15.6	45.1±12.8	0.797
PI-LL	6.0±12.5	5.6±9.8	0.549

Data was present as mean ± SD. A P value <0.05 was considered significant. PI, pelvic incidence; PT, pelvic tilt; SS, sacral slope; LL, lumbar lordosis.

We then examined the difference between the postoperative parameters of each patient compared to the preoperative parameters and found that the absolute value of the mean postoperative PI change for all patients was 3.11±2.76 (n=409). Grouped by LIV, the magnitude of the PI change was 1.63°±1.36° for L4 and above, 2.43°±2.00° for L5, and 3.83°±3.17° for S1 (P<0.05); thus, the closer the LIV was to S1, the greater magnitude of the change. There was a tendency for the magnitude of the PI change to increase as the number of surgical fixed segments increased, and the difference in change between individuals increased, but this was not significant (P=0.145). The magnitude of the PI change in the female group was 3.38°±2.94°, which was greater than that in the male group (2.76°±2.48°; P<0.05). The magnitude of the PI change was significantly greater in patients with SIJP after surgery. Notably, 19 patients showed a positive change (i.e., the postoperative PI value was increased compared to the preoperative PI value). These patients had a mean value of 5.83±2.48. Conversely, 4 patients showed a negative change (i.e., the postoperative PI value was decreased compared to the preoperative PI value). These patients had a mean value of 8.3°±4.08°. The magnitude of positive change in PI was greater than the magnitude of negative change in SIJP patients, and the difference was significant (P<0.05; see ).

Table 3

Effects of different factors on the magnitude of postoperative PI change

Factors	Change of PI (°, n=409)	Positive PI (°, n=186)	Negative PI (°, n=223)
LIV
L4 and above	1.63±1.36 [23]*	2.03±1.40 [12]*	1.21±1.23 [11]**
L5	2.43±2.00 [173]*	2.48±1.95 [78]*	2.39±2.06 [95]**
S1	3.83±3.17 [213]**	3.84±3.22 [96]**	3.81±3.15 [117]**
Seg
1	2.80±2.38 [159]	2.73±2.43 [60]	2.84±2.36 [99]
2	3.28±2.62 [148]	3.19±2.57 [68]	3.35±2.67 [80]
≥3	3.35±3.43 [102]	3.55±3.24 [58]	3.10±3.69 [44]
Gender
Male	2.76±2.48 [178]*	2.73±2.50 [79]	2.79±2.48 [99]
Female	3.38±2.94 [231]*	3.47±2.91 [107]	3.31±2.98 [124]
Age (years)
≤65	2.96±2.55 [253]	3.01±2.68 [112]	2.92±2.45 [141]
>65	3.36±3.07 [156]	3.37±2.87 [74]	3.35±3.25 [82]
SIJP
(−)	2.92±2.87 [386]*	2.85±2.62 [167]*	2.98±2.66 [219]*
(+)	6.26±2.64 [23]*	5.83±2.48 [19]*	8.3±4.08 [4]*

Data was present as mean ± SD [n]. A P value <0.05 was considered significant. *, statistical differences exist; **, statistically different from the other two. PI, pelvic incidence; LIV, lowest instrumented vertebra; Seg, segment; SIJP, sacroiliac joint pain.

SIJP patients

The prevalence of postoperative SIJP in all patients was 5.62% (23/409). There were no significant differences in terms of sex, age (P>0.05). The prevalence of SIJP with 3 or more segments fixed in total was 10.8%, which was significantly greater than single segment fixation (3.8%) and 2 segment fixation (4.1%) (P<0.01). Of the 23 patients who suffered SIJP after surgery, 18 had S1 LIV (8.5%), and 5 had L5 and above LIV (2.6%). However, it is worth noting that the LIV of all five patients was L5. The prevalence of postoperative SIJP was significantly greater in the S1 fixation group than the L5 fixation group (see ). We performed a binary logistics analysis of risk factors (including gender, age, preoperative SS, preoperative PI, preoperative LL, weight, LIV, fixed total segment) for SIJP and similarly concluded that LIV [odds ratio (OR) =3.378; P=0.023], fixed total segment ≥3 (OR =2.632; P=0.038) was associated with postoperative SIJP. We further divided the patients into two groups for chi-square test by PI changes greater than or equal to 4° and less than 4° and concluded that the patients with PI changes greater than or equal to 4°have a higher prevalence of SIJP (OR =13.051; P<0.001) (see ).

Table 4

Multifactorial analysis of whether SIJP occurred after surgery

Factors	All patients (n=409)	Postoperative SIJP (+) (n=23)	Postoperative SIJP (−) (n=386)	P value
Age (years)	61.3±11.2	64.1±8.0	61.2±11.4	0.22
Gender				0.353
Male	178	8 (4.5)	170
Female	231	15 (6.5)	216
LIV				0.009*
L5 and above	196	5 (2.6)	191
S1	213	18 (8.5)	195
Seg				0.033*
1	159	6 (3.8)	153
2	148	6 (4.1)	142
≥3	102	11 (10.8)	91

Data was present as mean ± SD/n/n (%). A P value <0.05 was considered significant. *, statistical differences exist. SIJP, sacroiliac joint pain; LIV, lowest instrumented vertebra; Seg, segment.

Table 5

Postoperative PI changes of more than 4° were more likely to result in SIJP

SIJP	|PI| ≥4°	|PI| <4°	Total
(+)	19	4	23
(−)	103	283	386
Total	122	287

P<0.001. PI, pelvic incidence; SIJP, sacroiliac joint pain.

The relationship between PI and SIJ motion

Counting all patients, we found that there was a correlation between postoperative PI changes and SS changes. Specifically, we found that the PI value increases slightly with increasing SS (n=386; y = 0.531 + 48.189x; R2=0.167; P<0.01). The patients were divided into two groups with LIV of S1and above. There was a significant correlation between the postoperative PI changes and SS changes in patients whose selected L5 and above as the LIV (y = 0.672 + 97.452x; R2=0.388; P<0.01). When we added the exclusion of SIJP-positive patients, this correlation was stronger in female patients, (y = 0.249 + 88.682x; R2=0.457; P<0.01), but there was no correlation in the S1 group (n=195; y = 0.297 + 30.046x; R2=0.086; P<0.01).

Discussion

Research has shown that approximately 10–25% of the time, low back pain or leg pain originates from the SIJ (33-35). Because of the strong correlation between the occurrence of SIJP and local biomechanical changes, easily to associate it with common postoperative spinal adjacent segment degeneration/disease (ASD) in patients who develop SIJP after spinal surgery.

We analyzed the pre- and postoperative sagittal parameters of patients who had undergone posterior thoracolumbar fusion surgery and found that while there was no significant difference in the mean values of PI before and after surgery in all patients, there was a definite magnitude of change in PI after surgery. In the results of our analysis, the magnitude of this change was significantly affected by the lower LIV and gender. The closer inferiorly fixed vertebra is to the sacrum means that the greater postoperative structural stress changes around the SIJ. Notably, when fixed distally to the sacrum, the stress of the internal fixation will act directly on the SIJ and alter the important surrounding structures associated with the SIJ, such as the iliolumbar ligament, anterior/posterior sacroiliac ligament, and the erector spinae musculature. Further, the additional force exerted during pedicle screw insertion and the fixation of the connecting rods may intraoperatively strain and damage the ligaments around the SIJ. In patients with long-segment (≥3) fixation, there is a tendency for the value of PI change to become larger, and for the difference in the magnitude of PI change between individuals to increase, which may be related to the significant reconstruction of sagittal balance after long-segment fixation, the reduced compensatory mobility of the adjacent segment, and the excessive stripping of tissues, such as muscles, leading to excessive nutation or counternutation of the SIJ (36). The magnitude of PI changes was greater in female patients than in males, which is consistent with the greater normal physiological activity of the SIJ in females than in males due to reproductive demands (15,37).

We analyzed the risk factors for the development of SIJP after surgery and found that they included 3 or more fixed segments (P<0.05) and the selection of S1 as the distal fixed vertebra (P<0.05), which were similar to the risk factors for the development of DJK/DJF after spinal fusion previously reported in the literature (36). The SIJP is consistent with the factor of a larger postoperative PI change, which confirms that the occurrence of SIJP after spine surgery is usually due to abnormal sagittal stress in the SIJ.

The relationship between PI and SIJ activity was further verified. Normally, in natural standing, the gravity generated by the upper trunk produces an external torque along the superior sacral edge, which is the main reason why the SIJ is always in the “nutation” position in the standing position (38); the internal torque generated by the ligaments, muscle tension, and joint friction in the posterior oppose it. These results confirm that the postoperative change by which the external torque acts on the sacrum is significantly correlated with the SIJ rotation angle around S2 (see ). There is a gender difference in the form of SIJ motion, such that the nutation is predominantly translational in male and predominantly rotational in female, and thus the correlation between PI and SS changes was further elevated in female patients with non-S1 fixation. When the spinal fusion involves S1, this correlation disappears, as the screw rod system directly involves the bony structure of the SIJ, and the sacrum and the overlying vertebrate form a strong internal fixation unit through the screw rod system, and gravity no longer acts directly on the sacrum to produce shear forces (see ).

Figure 1

In the standing position the pelvis rotates anteriorly around the hip axis and the SIJ will nutation. Orange arrow: direction of hip joint movement; blue arrow: direction of SIJ movement. When the pelvis as the stationary reference, the SIJ is sagittally rotated at an angle of α with S2 as the sagittal axis, and the PT and SS increase at the same time, ΔPT + ΔSS = ΔPI, ΔPI ≈ α ± 0.5°, postoperative PI’ = PI + ΔPI. SS, sacral slope; PT, pelvic tilt; SIJ, sacroiliac joint; PI, pelvic incidence.

Figure 2

Effect of internal fixation on SIJ motion. (A) In the standing position, the gravity of the upper body leads the SIJ nutation and the SIJ tends to stabilize. (B) The upper body gravity no longer acts directly on the sacrum when the fixation concludes S1, which may even allow the SIJ to counternutation if the slope of the superior edge of the superior vertebrate is negative at this time. Orange arrow: the gravity generated by the upper body and its component forces; blue arrow: direction of SIJ movement. BW: the upper body gravity acting to the upper edge of the sacrum; BW’: the upper body gravity acting to the upper edge of the superior vertebrate; BWsin(SS): forward shear force of the upper body gravity acting horizontally on the upper edge of the sacrum; BWcos(SS): the upper body gravity acting perpendicular to the upper edge of the sacrum. SS, sacral slope; SIJ, sacroiliac joint.

Thus, we conclude that the change in PI after spinal fusion reflects the sagittal activity of the SIJ to a greater extent, which can predict the occurrence of SIJP of surgical origin to a certain extent. We analyzed the risk factors and concluded that the characteristics of SIJP of surgical origin are consistent with the mechanism of DJK/DJF, which is defined as a posterior convexity of >10° in the distal junction area after orthopedic surgery and an increase of >10° compared to the preoperative period. The difference is that due to the limitations of the joint itself, the sacrum does not have as much mobility in the sagittal position as the vertebrae above it; however, our analysis showed that the magnitude of the change in PI reflects the sagittal mobility of the SIJ to some extent. These quantifiable results make the sagittal activity of the SIJ measurable on radiograph, and a change in PI of >4° was determined to be a high-risk factor for the development of SIJP. Thus, the DJK in a lower fixed spine of S1 can be defined as a change in PI of >4° from the preoperative value, or DJF if the patient has a new postoperative SIJP. The SIJP caused by spine surgery has been shown in some studies to be prevented by iliac screws; however, further studies on individualized surgical protocol development and the effects of iliac screws on SIJ activity and stress need to be conducted (6,11,39). For patients with a change in PI measurements >4° after surgery, appropriate non-surgical interventions should be considered first, such as standardized anti-osteoporotic treatment and muscle strength training to increase the stability of the SIJ (40,41).

This study had some limitations. The study was a retrospective cross-sectional study and only analyzed sagittal parameters in the standing position; however, the real activity of the SIJ in different positions is quite complex (e.g., exceptionally, some patients may present with nutation on 1 side of the SIJ and counternutation on the opposite side). Further tests are needed to analyze the postoperative SIJ motion in 3 dimensions in different postures.

The article’s supplementary files as