Risk factors analysis and risk prediction model for failed back surgery syndrome: a prospective cohort study

doi:10.21203/rs.3.rs-4960039/v1

Download PDF

Article

Risk factors analysis and risk prediction model for failed back surgery syndrome: a prospective cohort study

https://doi.org/10.21203/rs.3.rs-4960039/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Introduction:

With the growing number of posterior open surgery, the incidence of failed back surgery syndrome (FBSS) increases gradually. Currently, there is a lack of predictive systems and scientific evaluation in clinical practice. This study aimed to risk factors analysis of FBSS and develop a risk prediction model.

Materials and Methods

Baseline data were collected from 512 patients. Patients were followed up for one year. Ultimately, 146 patients were classified in the FBSS group, with an incidence rate of 32.5%. Logistic regression was used to screen for independent risk factors influencing the occurrence of FBSS. The diagnostic power of model was evaluated using the ROC curve.

Findings:

Age, smoking, type of pain, revision surgery, surgical technique, quality of life, and psychological status were significantly associated with the incidence of FBSS. The strongest factor in this model was the selected surgical technique, with an odds ratio of 0.095. The area under the ROC curve for the model's diagnostic and classification power was 0.852.

Conclusion

The causes of FBSS can stem from underlying factors, lifestyle, surgical causes, and patients' psychological factors. Therefore, prevention and treatment for each individual should be based on their specific cause to achieve optimal results.

Health sciences/Anatomy

Health sciences/Medical research

Risk factors

Prediction model

Surgery

Back

Syndrome

Failed

Degenerative lumbar disease (DLD) are recognized as the most common cause of low back pain, with prevalence increasing with age (1). The first line of treatment for these patients includes many non-surgical options such as lifestyle modifications, medications, and physiotherapy. However, surgical intervention is recommended if symptoms persist. Over the past two decades, the number of patients eligible for spinal surgery has significantly increased (2). Although initial structural defects are corrected after surgery, persistent pain or limb numbness sometimes continues. Some patients continue to suffer from chronic pain in the lower back and legs, along with ongoing functional limitations (3). Therefore, most patients who do not improve after surgery are classified under the heterogeneous disorder known as Failed Back Surgery Syndrome (FBSS). FBSS is defined as "a diverse and complex set of symptoms including persistent or recurrent chronic pain after one or more spinal surgeries" (4). This term is currently used to describe a heterogeneous group of patients whose surgical outcomes do not meet the pre-surgical expectations of the patient and the surgeon (5). Although the understanding of anatomical structures has increased and minimally invasive techniques have expanded, the prevalence of FBSS is rising due to the complexity of this issue with varied underlying causes (6).

Multiple factors (biological, psychological, and social) are involved in the development of the pain process, necessitating an interdisciplinary approach to clarify its causes further. Researchers believe that age, lifestyle (smoking, obesity, inactivity), the presence of specific comorbidities, the severity of preoperative pain, and psychosocial factors are potential characteristics associated with the incidence of FBSS (7). Additionally, reports indicate that incorrect surgical techniques, surgical complications, instability, recurrent disc herniation, and neuropathic pain have a significant impact on the occurrence of FBSS (8).

Modern medical research has identified multiple complex factors involved in the FBSS process. However, pain management requires a multidimensional approach, making it challenging to determine the exact causes of FBSS (9). Therefore, early screening and effective prevention of FBSS have become critical issues for healthcare professionals. In modern medicine, accurately predicting the occurrence and prognosis of diseases has gained increasing importance, as treatments should be individualized to achieve optimal outcomes. Expectations for outcomes should vary based on the type of structural problem, the number of previous surgeries, and the patient's mental health. Surgeons must convey realistic expectations to patients to align the expectations of both the patient and the surgeon (10). Numerous researchers have demonstrated that developing risk prediction models has effectively reduced disease incidence (11, 12). Despite significant advances in diagnosing and treating FBSS, the lack of baseline epidemiological data and a scientific prediction system hinders the successful evaluation of FBSS prevention and prognosis. Therefore, this study aims to identify the most critical risk factors and develop a robust risk prediction model for FBSS. The primary goal is to assist physicians and patients in enhancing prevention and treatment strategies for FBSS.

2.1 Study Design

This study was conducted as a prospective cohort in Iran (Valiasr Hospital, Qom) in 2023 with a one-year follow-up. The research population included all patients who visited Valiasr Hospital for surgery due to degenerative lumbar disease (DLD) from January 2022 to April 2023. The researcher enrolled all patients. The surgeries were performed by a single surgeon at a single center. The article's writing followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist.

2.2. Inclusion and Exclusion criteria

Inclusion Criteria

Age between 20–60 years
Patients with DLD
Patients who can undergo magnetic resonance imaging (MRI) and computed tomography (CT) scans

Exclusion Criteria

Patients diagnosed with specific diseases such as malignant tumors, vertebral fractures, spinal infections, inflammatory spondylitis
Patients with progressive neurological deficits or severe concomitant neurological symptoms
Patients with pain originating from non-spinal causes and/or soft tissue problems

2.3. Instruments and Data collection

2.3.1 Pre-Operative

Demographic and clinical information forms for the patients were recorded through a questionnaire and an interview before surgery by the researcher. A detailed history was taken regarding the onset of pain, pain characteristics, pain location, pain pattern, and pain source.

Preoperative anxiety was assessed using the Barton et al. (2019) Surgical Anxiety Questionnaire. This questionnaire contains 27 items that measure preoperative anxiety across six dimensions. This questionnaire's minimum and maximum possible scores range from 19 to 95. A score below 38 indicates low surgical anxiety, a score between 39 and 76 indicates moderate anxiety and a score above 77 indicates high anxiety (13).

2.3.2 During Operative

Surgical Technique

Surgery was performed on all patients under general anesthesia in the prone position. A midline incision measuring 5–15 cm (depending on the type of surgery) was made. After muscle dissection, bilateral decompression (laminectomy, medial facetectomy, flavectomy, discectomy) was performed. Following decompression, for the Fixation and Fixation + PLIF groups, pedicle screws were placed unilateral or bilateral (depending on the individual condition), and the pedicle screw anchoring process was completed. Then, in the Fixation + PLIF group, after preparing the endplates, an intervertebral cage was placed and fixed (14).

The amount of blood loss (Blood gases + bottle suction), the time of surgery (from skin incision to the last suture), the type of degenerative lumbar disease (Degenerative disc disease, Spinal stenosis, Spinal instability, Spondylolisthesis, Spondylolysis), the type of surgical procedure (with fixation (unilateral, bilateral), without fixation, with and without interbody fusion), the number of surgical levels, and sacrum fixation were recorded by the researcher.

2.3.3 Postoperative

Postoperative psychological disorders, quality of life, and sleep quality of the patients were assessed using validated questionnaires. The psychological disorders questionnaire (SCL-25) by Najarian and Davoudi (2001) contains 25 questions that measure anxiety, depression, phobic anxiety, paranoid ideation, psychosis, obsessive-compulsive disorder, and somatic complaints. The minimum and maximum scores in this questionnaire range from 25 to 125. Scores between 25–50 indicate low psychological disorders, 50–75 indicate moderate psychological disorders, and scores above 75 indicate high psychological disorders (15).

Patients' quality of life was assessed using the Short Form Health Survey (SF-36). This questionnaire evaluates eight health domains (physical function, role function-physical, bodily pain, general health, vitality, social function, role function-emotional, mental health) in two dimensions: physical and mental health. The minimum and maximum scores range from 0 to 100, indicating the lowest and highest quality of life, respectively (16).

The Pittsburgh Sleep Quality Index (PSQI) contains 19 items that assess sleep quality across seven scales (subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency) sleep disturbances, use of sleeping medication, daytime dysfunction). The total score of these seven scales forms a global score ranging from 0 to 21. A global score of 6 or higher indicates poor sleep quality (17).

After 12 months of follow-up, the patients were divided into two groups. The first group (non-FBSS) consisted of patients who were satisfied with their surgery and completely recovered. The second group (FBSS) included patients who fell into the Failed Back Surgery Syndrome category. Similar studies have provided various definitions of FBSS patients, as outlined below. In FBSS patients, despite successful spinal surgery, chronic radicular pain in the same area recurred or persisted (4). The outcome of lumbar spinal surgery did not meet the preoperative expectations of the patient and the surgeon (1). Patients with chronic neuropathic pain for three months, with a score higher than five on the Visual Analog Scale (VAS) in the lumbar area, with or without radiation to the limbs (5). Patients experienced chronic back or leg pain after successful lumbar surgery without specific reasons, such as compressive lesions or infection (8). Therefore, for the precise diagnosis of patients, considering the above definitions, a quantitative and systematic definition of FBSS was provided as follows: intractable pain or sensory deficit in the back and/or limbs after surgery that is resistant to conservative treatment (for more than three months), leading to dissatisfaction with the surgical outcome.

2.4. Statistical Analysis

Data were analyzed using SPSS version 16. Initially, the equality of variances and the assumption of normality of the data were assessed using Levene's and the Shapiro-Wilk tests. Continuous data were analyzed using the independent t-test and presented as mean (SD). Categorical data were analyzed using the chi-square test and presented as numbers (%). The Omnibus test was used to assess the explanatory and predictive power of the logistic regression model. A binary logistic regression analysis was conducted to build the FBSS risk prediction model, with FBSS occurrence as the dependent variable. The ROC curve (AUC) was used to assess the diagnostic and classification power of the logistic regression model.

3.1. Baseline Characteristics

A total of 512 patients meeting the criteria were enrolled. Initially, 37 patients were excluded due to hospital transfer, diagnosis of fractures, and malignancy. Subsequently, 475 patients underwent lumbar spine surgery. Another 27 patients were excluded due to undergoing surgery in two stages: pregnancy and lack of follow-up interest.... In the end, 302 patients were in the non-FBSS group, and 146 patients were in the FBSS group, with an incidence rate of 32.5% (Fig. 1). In this study, 30 independent variables potentially influencing the occurrence of FBSS were examined. Based on the principle that the required sample size for multivariate analysis should be 5 to 10 times the number of independent variables (18), the size of the sample needed for this study was 150 to 300. Four hundred forty-eight patients were analyzed, exceeding the minimum sample size and meeting the statistical requirements.

Demographic data analysis showed that the mean age in the FBSS group was significantly higher compared to the healthy group (non-FBSS) (65.68 (8.66) vs. 62.62 (9.03), P < 0.05). Smoking was more prevalent in the FBSS group as a potential risk factor (61.6% vs. 49.7%, P < 0.05). No significant differences were observed between the two groups for other variables (P > 0.05) (Table 1).

Table 1

Evaluation of demographic information
	FBSS (N = 146)	Non-FBSS (N = 302)	P-value^‡
Age^† (yr), mean (SD)	64.68 ( 8.66)	62.62 (9.03)	0.022
Gender*, NO(%) Male Female	76 (52.1%) 70 (47.9%)	151 (50.0%) 151(50.0%)	0.683
Height^† (cm), mean ± SD)	176.74 ± 8.74	175.02 ± 9.81	0.062
Weight^† (kg), (mean ± SD)	88.25 ± 10.2	87.26 ± 8.99	0.298
BMI(Kg/m²), (mean ± SD)
Employment status*, NO(%) Unemployed Employed	37 (25.3%) 109 (74.7%)	55 (18.2%) 247 (81.8%)	0.08
Smoking*, NO(%) Yes No	90 (61.6%) 56 (38.4%)	150 (49.7%) 152 (50.3%)	0.017
Opioid*, NO(%) Yes No	12 (8.2%) 134 (91.8%)	28 (9.3%) 274 (90.7%)	0.714
Marital status*, NO(%) Single Married	28 (19.2%) 118 (80.8%)	54 (17.9%) 248 (82.1%)	0.739
Level of education*, NO(%) \(\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\le\:\text{D}\text{i}\text{p}\text{l}\text{o}\text{m}\text{a}\) Callegiate	97 (66.4%) 49 (33.6%)	221 (73.2%) 81 (26.8%)	0.141
BMD^† (T-score), mean(SD)	-1.83 (0.44)	-1.85 (0.65)	0.779

*:Chi-Square, †:t-test, ‡: Statistical significant as p < 0.05. failed back surgery syndrome (FBSS). BMI: body mass index (BMI), bone mineral density (BMD).

3.2. Pre-Surgery Outcome

Analysis of pre-surgery factors indicated that repeated surgeries reduced the success rate of the surgery, and the likelihood of FBSS increased with the number of surgeries (54.8% vs. 35.5%, P < 0.04). As shown in Table 2, the spread of symptoms to the lower limbs may be associated with FBSS. Therefore, patients without leg pain and numbness in the lower limbs had a lower likelihood of developing FBSS (50.7% vs. 19.9%, P < 0.05) (Table 2).

Table 2

Evaluation of pre-surgery factors and underlying diseases
	FBSS (N = 146)	Non-FBSS (N = 302)	P-value^‡
Anxiety before surgery^†, mean (SD)	62.23 ± 12.4	63.60 ± 13.9	0.348
Arthrose*, No(%) Yes No	37 (25.3%) 109 (74.7%)	56 (18.5%) 246 (81.5%)	0.096
Rheumatoid Arthritis*, No(%) Yes No	35 (24.0%) (76.0%)111	(17.9%)54 248 (82.1%)	0.130
Diabetes*, No(%) Yes No	35 (23.97%) (76.02%)111	(23.2%)70 231 (76.5%)	0.774
Hypertension*, No(%) Yes No	41 (28.1%) 105 (71.9%)	73 (24.2%) 229 (75.8%)	0.373
Re-surgery*, No(%) First time Recurrent	66 (45.2%) 80 (54.8%)	195 (64.6%) 107 (35.5%)	0.0001
Symptom location*, No(%) Low back pain without limb symptoms Unilateral limb symptom Bilateral limb symptom	72 (49.3%) 21 (14.4%) 53 (36.3%)	242 (80.1%) 28 (9.3%) 32 (10.6%)	0.0001
Symptom duration (Month) ^†, mean (SD)	11.3 (3.27)	11.8 (3.09)	0.178
Claudication*, No(%) Yes No	21 (14.4%) 125 (85.6%)	54 (17.9%) 248 (82.1%)	0.353
*:Chi-Square, †:t-test, ‡: Statistical significant as p < 0.05. failed back surgery syndrome (FBSS)

3.3. During Surgery Outcome

Analysis of intraoperative factors showed that the incidence of FBSS significantly varied among different surgical techniques. Therefore, the highest incidence of FBSS was observed in patients who underwent fixation with intervertebral fusion (46.6% vs. 13.9%, P < 0.05) (Table 3).

Table 3

Evaluation of factors during surgery
	FBSS (N = 146)	Non-FBSS (N = 302)	P-value^‡
Time of surgery^†, mean (SD)	120.8 (17.3)	117.5 (22.9)	0.076
Type of complication*, No(%) Laminectomy Spinal stenosis Spinal instability Spondylolisthesis Spondylolysis Multiple complications	15 (10.2%) 19 (13.0%) 25 (17.1%) 23 (15.7%) 36 (24.6%) 28 (19.1%)	27 (8.10%) 63 (20.9%) 53 (17.5%) 58 (19.2%) 46 (15.2%) 55 (18.2%)	0.110
Side Fixation*, No(%) Unilateral Bilateral	52 (35.6%) 94 (64.4%)	118 (39.07%) 184 (60.9%)	0.480
Levels of surgery*, No(%) Single level Multilevel	70 (47.9%) 76 (52.1%)	129 (42.7%) 173 (57.3%)	0.296
Type of Surgery*, No(%) Decompressio Fixation alone Fixation + PLIF	25 (17.1%) 53 (36.3%) 68 (46.6%)	164 (54.3%) 96 (31.8%) 42 (13.9%)	0.0001
Sacrum fixation*, No(%) Yes No	36 (24.7%) 110 (75.3%)	95 (31.5%) 207 (68.5%)	0.138
*:Chi-Square, †:t-test, ‡: Statistical significant as p < 0.05. failed back surgery syndrome (FBSS)

3.4. Post-Surgery Outcome

As shown in Table 4, individuals with poor mental health post-surgery are at a higher risk of developing FBSS (70.03(12.6) vs. 40.8(12.1), P < 0.05). Additionally, data from this study indicate that the highest incidence of FBSS is found in individuals with poor quality of life (31.03(11.1) vs. 64.3 (16.4), P < 0.05). No significant differences were observed between the two groups in terms of rest time post-surgery, physiotherapy, and sleep quality (P > 0.05) (Table 4).

Table 4

Evaluation of factors post-surgery
	FBSS (N = 146)	Non-FBSS (N = 302)	P-value^‡
Rest time(day)^†, mean (SD)	42.8 (12.6)	44.6 (11.6)	0.143
Rehabilitation*, No(%) Yes No	49 (33.6%) 97 (66.4%)	90 (29.8%) 212 (70.2%)	0.420
Mental health^†, mean (SD)	70.03 (12.6)	63.5 (12.2)	0.0001
Quality of Life^†, mean (SD)	31.03 (11.1)	64.3 (16.4)	0.0001
Sleep quality^†, mean (SD)	5.87 (1.71)	5.69 (2.15)	0.346

*:Chi-Square, †:t-test, ‡: Statistical significant as p < 0.05. failed back surgery syndrome (FBSS)

Table 5 was used to explain the strength and efficiency of the logistic regression model in distinguishing and classifying individuals with FBSS and non-FBSS. Overall, the classification accuracy of individuals by the fitted logistic regression model was 78.8% (Table 5).

Table 5

Classification Table
		Predicted		Percentage Correct
		Non-FBSS	FBSS	Percentage Correct
Observed	Non-FBSS	267	35	88.4
Observed	FBSS	60	86	58.9
Overall Percentage				78.8
failed back surgery syndrome (FBSS)

In Table 6, the coding of qualitative variables has been carried out. For each variable, one group was designated as the reference category to compare the odds of developing FBSS in other groups relative to the reference category (Table 6).

As shown in Table 7, based on the odds ratio (OR), the selected surgical technique is the strongest factor in the occurrence of FBSS. Therefore, the likelihood of developing FBSS in individuals who underwent Fixation + PLIF surgery is 10.52 times greater than those who underwent decompression surgery. Additionally, patients who underwent Fixation + PLIF surgery are 2.83 times more likely to develop FBSS compared to those who underwent Fixation surgery. In our regression model, the second most influential factor in the occurrence of FBSS was radicular pain in the lower limbs. The likelihood of developing FBSS in individuals who had radicular pain in both lower limbs before surgery is 7.24 times higher than in those who only suffered from back pain. The likelihood of developing FBSS in individuals whose back pain radiated to both legs before surgery is 2.92 times higher than in those whose back pain radiated to only one leg. Smoking was identified as the third most influential factor in the occurrence of FBSS, with smokers having a 2.55 times higher likelihood of developing FBSS compared to non-smokers.

Based on the odds ratio (OR), the likelihood of developing FBSS in revision surgery is 2.50 times higher than in those undergoing surgery for the first time. The odds ratio for the variables of age and psychological disorders is reported to be 1.046 and 1.031, respectively. Thus, an increase in age and a higher score of psychological disorders significantly increase the likelihood of developing FBSS. The odds ratio for the quality of life variable is 0.967, indicating that an increase in the quality of life score significantly reduces the likelihood of developing FBSS (Table 7).

Table 6

Categorical Variables Codings
		Frequency	Parameter coding
		Frequency	(1)	(2)
Type surgery	Decompration	189	1.000	0.000
	Fixation	149	0.000	1.000
	Fixation + PLIF	110	0.000	0.0001
Symptom location	Back-pain	314	1.000	0.0001
	Uni-limb	49	0.000	1.000
	Bi-limb	85	0.000	0.000
Re-surgery	First time	261	1.000
Re-surgery	Recurrent	187	0.000
Smoking	No	208	1.000
Smoking	Yes	240	0.000

Table 7

Multivariate logistic prediction model of FBSS
Variables	Beta	S.E	Wald	Df	OR	95% Confidence Interval for OR		P-value^‡
Variables	Beta	S.E	Wald	Df	OR	Lower	Upper	P-value^‡
Type surgery			51.600	2				0.0001
Type surgery (1)	-2.352	0.327	51.595	1	0.095	0.050	0.181	0.0001
Type surgery (2)	-1.041	0.299	12.115	1	0.353	0.197	0.635	0.001
Symptom location			39.854	2				0.0001
Symptom location (1)	-1.981	0.318	38.847	1	0.138	0.074	0.257	0.0001
Symptom location (2)	-1.072	0.451	5.648	1	0.342	0.141	0.829	0.017
Smoking (1)	-0.936	0.270	11.995	1	0.392	0.231	0.666	0.001
Re-surgery (1)	-0.918	0.254	13.063	1	0.399	0.243	0.657	0.0001
Age	0.045	0.011	15.903	1	1.046	1.023	1.069	0.0001
Psychopathy	0.031	0.008	13.586	1	1.031	1.014	1.048	0.0001
Quality life	-0.033	0.009	14.880	1	0.967	0.951	0.984	0.0001

Coefficient value (Beta), Standard error (S.E), Chi-square value (Wald), Degrees of freedom (Df), Odds ratio (OR). ‡: Statistical significant as p < 0.05.

Finally, the logistic regression model was obtained as follows:

This model calculates the probability that an individual will develop FBSS.

According to the results in Table 8, the AUC (Area Under the ROC Curve) index is reported to be 0.852, indicating that the proposed logistic regression model has a high performance in correctly identifying and classifying individuals with FBSS and non-FBSS. The sensitivity of the prediction model is 58.9%, and the specificity is 88.4% (Table. 8) (F.I.G.2).

Table 8

Area Under the ROC Curve
Predicted probability	Area	S.E	95% Confidence Interval for Area		P-value
			Lower	Upper
	0.852	0.018	0.816	.887	0.0001

F.I.G. 2. The area under the rock curve shows the performance of the logistic regression model in identifying and classifying people with FBSS or Non-FBSS. The x-axis is the sensitivity, which represents the possibility of predicting positive samples but actually negative samples. The y axis is 1-specific, which represents the possibility of predicting positive samples but actually positive samples. The blue real line is the ROC curve of the model, and the red line represents the ROC curve of random gues.

Application example of risk prediction model

Case Study

Consider a non-smoking, 57-year-old married female patient with a BMI of 30.2 kg/m². Six years ago, she underwent fixation surgery at the L3-L4 lumbar spine level. She presented with severe back pain and radicular pain in both lower limbs. Magnetic resonance imaging showed ASD at the L2-L3 vertebrae level. Additionally, flexion and extension radiographs revealed grade 3 spondylolisthesis at the L5-S1 vertebrae level. Thus, she underwent fixation surgery from the L2 to S1 vertebrae levels. Her surgery lasted 3 hours and 45 minutes. Her quality of life and psychological disorder scores were 33 and 73, respectively. Now, we input these values into the logistic regression model.

The predicted probability that this patient will develop FBSS is approximately 0.85 or 85%. Since this probability is more significant than 0.5, according to the logistic regression model, this case is classified as an individual with FBSS.

The present study aimed to analyze the risk factors and develop a risk prediction model for Failed Back Surgery Syndrome (FBSS). A risk prediction model was adopted in this study, which predicts the logistic regression outcomes based on clinical and medical data. This model integrates and compares various determining factors to predict the likelihood of an individual developing this condition. Data analysis revealed that the selected surgical technique is the strongest factor influencing the occurrence of FBSS. Therefore, the highest incidence of FBSS was observed first in the Fixation + PLIF group and then in the Fixation group. Recent reports indicate that pedicle screw fixation techniques, bilateral muscle dissection, removal of posterior elements, and loss of bony structure lead to spinal instability and pain of unknown origin (19, 20). Souslian & Patel (2024) stated that after PLIF surgery, the potential for load distribution among spinal structures increases, resulting in axial pain (21). Cicek et al. (2017) reported similar findings (22). In patients undergoing PLIF + Fixation surgery, the entire bilateral facet joints are removed, the disc is completely evacuated, and a relatively small intervertebral cage acts as a focal pivot point. Thus, despite pedicle screw fixation, the structure may inherently be unstable (23). Contrary to the above results, some studies have shown no significant difference between spinal fixation techniques with and without intervertebral fusion (24). The discrepancy in results may be due to variations in postoperative follow-up duration, levels of spinal fixation, and different surgical techniques.

Data analysis showed that radicular pain in the lower limbs before surgery is the second most influential factor in FBSS. Supporting these findings, Wenbo et al. (2022) believe that preoperative radicular pain in the lower limbs is a significant factor in FBSS development and should be investigated as a marker for identifying at-risk populations (25). Evandro et al. (2020) reported similar results (2). Contrary to these findings, some reports suggest that surgical outcomes are not significantly related to the type and severity of preoperative pain (1). The discrepancy in results may be influenced by the duration of symptoms before surgery and the patient inclusion criteria. The optimal timing for decompression surgery remains unclear. Historically, early surgical intervention for symptomatic spinal stenosis has been recommended based on the view that the condition is always progressive. Unlike peripheral nerves, nerve roots lack a blood-nerve barrier, and prolonged compressive lesions lead to intraneural edema. Over time, this edema causes nerve fibrosis, a process that is irreversible even with surgical intervention (26). Additionally, research has shown that long-term symptomatic radiculopathy is affected by multiple damaging factors and is more complex than simple neural dysfunction caused by physical pressure (27).

Data analysis indicates that smoking is one of the significant risk factors for the development of FBSS. In this context, a study by Mekhail et al. (2020) with a one-year follow-up showed that the incidence of FBSS was significantly higher in current smokers compared to non-smokers or those who had quit smoking (28). Robson et al. (2019) reported similar findings (23). A meta-analysis study found that individuals who quit smoking three months before surgery were less likely to develop FBSS compared to smokers (29). Contrary to these results, some studies did not find a significant difference in the incidence of FBSS between smokers and non-smokers (30). The discrepancy in results may be due to the type of intervention, the extent of iatrogenic tissue damage, and the sample size of the studies. In general, nicotine destroys vascular endothelium, playing a crucial role in vasoconstriction, blood flow disruption, synthesis and secretion of biologically active factors, neovascularization, and immune responses. After spinal surgery, the circulatory system plays a key role in tissue healing and postoperative recovery. Therefore, when the vascular endothelium is damaged, the post-surgical healing process is disrupted and inefficient in smokers. The more extensive the iatrogenic soft tissue damage during surgery, the more significantly postoperative complications increase. These findings are supported by several studies reporting increased inflammatory markers in smokers (31, 32).

Findings from the study indicate that revision surgery is a potential risk factor for developing FBSS. Therefore, patients with previous lumbar surgery were at higher risk. Researchers believe that the number of previous spinal surgeries is a significant predictor of surgical outcomes (33). Montenegro et al. (2021) found that the average functional status score of 46% of patients significantly decreased six months after revision surgery (34). Moaven et al. (2020) stated in their study that revision surgery is much more complicated than primary surgery due to unclear anatomical levels and scars around the nerves, requiring a high skill level (33). In revision surgeries, the absence of spinous processes, bony structures, and fibrous tissue growth in the surgical area leads to significant anatomical changes. Adhesions in the surgical area increase the risk of nerve element damage and dural tears. Therefore, there is concern that the success rate of the second surgery is lower than that of the first surgery. Researchers have noted that the incidence of FBSS increases from 8% in primary surgeries to 48% in revision surgeries. However, the degree of difference may vary depending on surgical techniques and the type of condition (35).

Data analysis showed that age is one of the risk factors for patients undergoing spinal surgery. Early studies have identified age as an important underlying factor in the occurrence of FBSS and believe that it needs special consideration when selecting treatment options (35). Wenbo et al. (2022) argue that older patients have a significant increase in postoperative complications and often require revision surgery (25). Other studies in this field have reported similar findings (3, 36). Changes in pain perception with aging affect the pain experience in various ways, some of which are still unknown. The assumptions are based on the anatomical changes in older individuals, such as spinal canal and foramina narrowing, increased pressure on neural elements, and microcirculation disruption, which exponentially decrease the success rate after surgery. Researchers have found that the loss of lumbar lordosis, increased thoracic kyphosis, and stiffening of the ligamentum flavum are common changes that contribute to spinal degeneration and are frequent sources of back pain after surgery in older individuals (37).

Data from the study indicate that specific psychological factors, including significant levels of depression, anxiety, phobic anxiety, paranoid ideation, psychosis, obsessive-compulsive disorder, and somatic complaints, lead to an increased incidence of FBSS. Recent reports suggest that these psychological factors affect individual changes in pain sensitivity, thereby influencing pain perception (13, 38). The study by Sebaaly et al. (2018) showed a bidirectional relationship between pain and depression (39). Elsamadicy et al. (2018) believe that mental health is a much stronger predictor of disability from back pain than structural abnormalities (3). A recent study identified depression as a risk factor for postoperative pain in specific areas, including the head, neck/shoulder, and back (7). Contrary to these findings, other studies showed no significant difference between psychological disorders and postoperative pain with a one-year follow-up (40). The discrepancy in results may be due to the type of tools used to measure mental health and the duration of follow-up after surgery. In short-term follow-up, it may be difficult to differentiate between nonspecific pain sources (skin incision, iatrogenic tissue damage, reactive spasms, and nerve root inflammation) and pain originating from the central nervous system. In this context, the results of a study by Graham et al. with a five-year follow-up reported similar findings, illustrating this issue well (20).

The existing literature indicates that psychological disorders can result from stressors and disrupt behavioral patterns. In other words, individuals who experience pain due to their spinal condition avoid normal work and recreational activities, significantly reducing their quality of life (41). As noted in the present study, poor quality of life predicts FBSS occurrence. Inoue et al. (2017) found that increased FBSS incidence is associated with low quality of life and high levels of functional disability (40). Therefore, psychological disorders can directly or indirectly increase the prevalence of FBSS. However, it is important to note that awareness of these factors should not prevent patients from undergoing spinal surgery when significant pathology and surgical indications are present. Instead, these risk factors necessitate careful consideration and optimization of surgical timing. Patients with higher surgical risk in spinal surgery may achieve better outcomes with prompt intervention. This is because prolonged pain and discomfort in this population can exacerbate existing psychosocial stressors and reduce the success rate of surgery.

It is logical that "the most effective treatment for FBSS is to avoid or prevent FBSS itself," as aging and the exacerbation of psychological, physical, and mechanical effects experienced by patients transform it into a multifactorial, complex, and painful syndrome. The detrimental effects of FBSS are well known, and current strategies and treatment methods are only available after the onset of the disease. Furthermore, most patients with FBSS experience varying degrees of disability, which brings significant anxiety and economic burden to them and their families. Therefore, early screening for this condition is of particular importance. The prediction model presented in the current study demonstrated good performance and is easy to apply in clinical practice, providing the necessary individual diagnostic and treatment requirements. This model may be useful for preventing and treating FBSS and identifying at-risk populations in clinical practice.

Limitations

There are several limitations to this study. First, a relatively short follow-up period may obscure changes in outcome indicators resulting from other degenerations. Additionally, the patient's risk factors examined in this study were self-reported, which introduces potential recall bias and subjective interpretations. Future research should examine a broader set of clinical variables to enhance the comprehensiveness and completeness of the information obtained by the prediction model. Moreover, using a multicenter approach for sample collection would increase the model's validity and generalizability.

Based on the results of multivariate logistic regression analysis, this study demonstrated that the selected surgical technique, preoperative pain symptoms, smoking, revision surgery, age, mental health, and quality of life are risk factors for FBSS. Increased awareness in this area can serve as a tool for physicians to identify at-risk populations and provide more effective management to reduce discrepancies between patient and physician expectations. Relying on these factors, an initial FBSS risk prediction model was developed. Additionally, the ROC curve indicated that the logistic regression model performs well in accurately identifying and classifying individuals.

Consent to participate

All patients provided written informed consent to participate in the study, publication of radiological images and for their data to be published. The principles of the Helsinki Declaration were adhered to in this study.

Competing interests

The authors have no potential or real conflict of interests

CRediT authorship contribution statement

BI: conceptualization, investigation, Writing-review & esiting,. AM: validation, funding acquisition, resources, methodology. SZ: data curation, investiration, Writing-original draft. PH: project administration, visualization. SK: software, formal analysis, validation.

Funding

The study was funded by Vice-chancellor for Research and Technology, Hamadan University of Medical Sciences (140208237083).

Author Contribution

Acknowledgements

This study has been adapted from an MSc thesis project at Hamadan University of Medical Sciences.

Data Availability

Data associated with the study has not been deposited into a publicly available repository. Data are available from the corresponding author on reasonable request.

Ethics approval

This study was reviewed and approved by Ethics Committee of Hamedan University of Medical Sciences with the approval number: IR.UMSHA.REC.1402.553.

Ho, C-N., Liao, J-C. & Chen, W-J. Instrumented Posterolateral fusion versus instrumented Interbody fusion for degenerative lumbar diseases in uremic patients under hemodialysis. BMC Musculoskelet. Disord. 21, 1–7. https://doi.org/10.1186/s12891-020-03815-z (2020).
Fang, E. F. et al. A research agenda for ageing in China in the 21st century: focusing on basic and translational research, long-term care, policy and social networks. Ageing Res. Rev. 64, 101174. https://doi.org/10.1016/j.arr.2020.101174 (2020).
Elsamadicy, A. A. et al. Drivers and risk factors of unplanned 30-day readmission following spinal cord stimulator implantation. Neuromodulation: Technol. Neural Interface. 21 (1), 87–92. https://doi.org/10.1111/ner.12689 (2018).
Sivaganesan, A. et al. Why are patients dissatisfied after spine surgery when improvements in disability and pain are clinically meaningful? Spine J. 20 (10), 1535–1543. http://dx.doi.org/10.1016/j.spinee.2020.06.008 (2020).
De Andres, J. et al. Prospective, randomized blind effect-on-outcome study of conventional vs high-frequency spinal cord stimulation in patients with pain and disability due to failed back surgery syndrome. Pain Med. 18 (12), 2401–2421. https://doi.org/10.1093/pm/pnx241 (2017).
Vleggeert-Lankamp, C. L., Arts, M. P., Jacobs, W. C. & Peul, W. C. Failed back (surgery) syndrome: time for a paradigm shift. Br. J. pain. 7 (1), 48–55. https://doi.org/10.1177/2049463713479095 (2013).
Tang, B., Meng, W., Hägg, S., Burgess, S. & Jiang, X. Reciprocal interaction between depression and pain: results from a comprehensive bidirectional Mendelian randomization study and functional annotation analysis. Pain. 163 (1), e40–e8. https://doi.org/10.1097%2Fj.pain.0000000000002305 (2022).
Cho, J. H. et al. Treatment outcomes for patients with failed back surgery. Pain physician. 20 (1), E29. http://dx.doi.org/10.36076/ppj.2017.1.E29 (2017).
Miranda, L., Quaranta, M., Oliva, F. & Maffulli, N. Stem cells and discogenic back pain. Br. Med. Bull. 146 (1), 73–87. https://doi.org/10.1093/bmb/ldad008 (2023).
Shirdel, Z., Behzad, I., Manafi, B. & Saheb, M. The interactive effect of preoperative consultation and operating room admission by a counselor on anxiety level and vital signs in patients undergoing Coronary Artery Bypass Grafting surgery. A clinical trial study. Investigación y Educación en Enfermería. ;38(2). (2020). https://doi.org/10.17533/udea.iee.v38n2e07
Dong, Y-M. et al. Development and validation of a nomogram for assessing survival in patients with COVID-19 pneumonia. Clin. Infect. Dis. 72 (4), 652–660. https://doi.org/10.17533/udea.iee.v38n2e07 (2021).
Mahdood, B., Imani, B. & Khazaei, S. Effects of inhalation aromatherapy with Rosa damascena (Damask rose) on the state anxiety and sleep quality of operating room personnel during the COVID-19 pandemic: A randomized controlled trial. J. PeriAnesthesia Nurs. 37 (4), 493–500. https://doi.org/10.1016/j.jopan.2021.09.011 (2022).
Burton, D., King, A., Bartley, J., Petrie, K. J. & Broadbent, E. The surgical anxiety questionnaire (SAQ): development and validation. Psychol. Health. 34 (2), 129–146. https://doi.org/10.1080/08870446.2018.1502770 (2019).
Nomoto, E. K., Fogel, G. R., Rasouli, A., Bundy, J. V. & Turner, A. W. Biomechanical analysis of cortical versus pedicle screw fixation stability in TLIF, PLIF, and XLIF applications. Global Spine J. 9 (2), 162–168. https://doi.org/10.1177/2192568218779991 (2019).
Tanhaye Reshvanloo, F. Construct validity and reliability of Symptom Checklist-25 (SCL-25). J. Fundamentals Mental Health. 18, 48–55. https://doi.org/10.22038/jfmh.2015.6255 (2016).
Lins, L. & Carvalho, F. M. SF-36 total score as a single measure of health-related quality of life: Scoping review. SAGE open. Med. 4, 2050312116671725. https://doi.org/10.1177/2050312116671725 (2016).
Pallesen, N., Omvik, S. & Matthiesen, B. Pittsburgh sleep quality index. Tidsskrift-Norsk Psykologforening. 42 (8), 714. https://doi.org/10.1016/0165-1781(89)90047-4 (2005).
Chen, B. Sample size methodology for multivariate analysis—synthetic estimate method for sample size in multivariate analysis. Injury Med. 1 (4), 58–60. https://doi.org/10.3389%2Ffpubh.2021.679699 (2012).
Shen, J. et al. Comparison between fusion and non-fusion surgery for lumbar spinal stenosis: a meta-analysis. Adv. Therapy. 38, 1404–1414. https://doi.org/10.1007/s12325-020-01604-7 (2021).
Goh, G. S. H. et al. Patients with poor baseline mental health may experience significant improvements in pain and disability after minimally invasive transforaminal lumbar interbody fusion: a 5-year follow-up study. Clin. Spine Surg. 33 (5), 205–214. https://doi.org/10.1097/bsd.0000000000000912 (2020).
Souslian, F. G. & Patel, P. D. Review and analysis of modern lumbar spinal fusion techniques. Br. J. Neurosurg. 38 (1), 61–67. https://doi.org/10.1080/02688697.2021.1881041 (2024).
Cicek, E., Koçak, M. M., Koçak, S., Sağlam, B. C. & Türker, S. A. Postoperative pain intensity after using different instrumentation techniques: a randomized clinical study. J. Appl. Oral Sci. 25, 20–26. https://doi.org/10.1590/1678-77572016-0138 (2017).
Robson, E. K. et al. Healthy Lifestyle Program (HeLP) for low back pain: protocol for a randomised controlled trial. BMJ open. 9 (9), e029290. https://doi.org/10.1136/bmjopen-2019-029290 (2019).
Keorochana, G. et al. Comparative outcomes of cortical screw trajectory fixation and pedicle screw fixation in lumbar spinal fusion: systematic review and meta-analysis. World Neurosurg. 102, 340–349. https://doi.org/10.1016/j.wneu.2017.03.010 (2017).
Xu, W., Ran, B., Zhao, J., Luo, W. & Gu, R. Risk factors for failed back surgery syndrome following open posterior lumbar surgery for degenerative lumbar disease. BMC Musculoskelet. Disord. 23 (1), 1141. https://doi.org/10.1186/s12891-022-06066-2 (2022).
Sheaths, C. T. 5 Nerve Anatomy and Physiology, Compression Neuropathies, and Nerve Injuries. Review of Hand Surgery. E-Book. 85. http://dx.doi.org/10.1016/j.jassh.2004.06.007 (2021).
Urits, I. et al. Low back pain, a comprehensive review: pathophysiology, diagnosis, and treatment. Curr. Pain Headache Rep. 23, 1–10. https://doi.org/10.1007/s11916-019-0757-1 (2019).
Mekhail, N. et al. The impact of tobacco smoking on spinal cord stimulation effectiveness in complex regional pain syndrome patients. Neuromodulation: Technol. Neural Interface. 23 (1), 133–139. https://doi.org/10.1111/ner.13058 (2020).
Mills, E. et al. Smoking Cessation Reduces Postoperative Complications: A Systematic Review and Meta-analysis. Am. J. Med. 124 (2), 144. https://doi.org/10.1016/j.amjmed.2010.09.013 (2011). 54.e8
Alan, N. et al. Smoking and postoperative outcomes in elective cranial surgery. J. Neurosurg. 120 (4), 811–819. https://doi.org/10.3171/2014.1.JNS131852 (2014).
Yılmaz, M. & Kayançiçek, H. A new inflammatory marker: elevated monocyte to HDL cholesterol ratio associated with smoking. J. Clin. Med. 7 (4), 76. https://doi.org/10.3390%2Fjcm7040076 (2018).
Elisia, I. et al. The effect of smoking on chronic inflammation, immune function and blood cell composition. Sci. Rep. 10 (1), 19480. https://doi.org/10.1038/s41598-020-76556-7 (2020).
Moaven, M., Ilkhechi, R. B., Zeinali, M., Hesam, S. & Jamali, K. Study of Re-Operational Risk Factors in Lumbar Herniated Disk Patients Referring to Golestan Hospital, Ahvaz From 2011 to 2015. Jundishapur J. Health Sci. 12 (1). http://dx.doi.org/10.5812/jjhs.99748 (2020).
Montenegro, T. S. et al. Clinical outcomes in revision lumbar spine fusions: an observational cohort study. J. Neurosurgery: Spine. 35 (4), 437–445. https://doi.org/10.3171/2020.12.spine201908 (2021).
Puvanesarajah, V. et al. Risk factors for revision surgery following primary adult spinal deformity surgery in patients 65 years and older. J. Neurosurgery: Spine. 25 (4), 486–493. https://doi.org/10.3171/2016.2.spine151345 (2016).
Vasdev, N. Multicentric validation of nomograms based on BC-116 and BC-106 urine peptide biomarker panels for bladder cancer diagnostics and monitoring in two prospective cohorts of patients. Br. J. Cancer. 128 (6), 929. https://doi.org/10.1038/s41416-023-02142-z (2023).
Cloyd, J. M., Acosta, F. L. Jr & Ames, C. P. Complications and outcomes of lumbar spine surgery in elderly people: a review of the literature. J. Am. Geriatr. Soc. 56 (7), 1318–1327. https://doi.org/10.1111/j.1532-5415.2008.01771.x (2008).
Bener, A. et al. Psychological factors: anxiety, depression, and somatization symptoms in low back pain patients. J. pain Res. 95–101. https://doi.org/10.2147/jpr.s40740 (2013).
Sebaaly, A., Lahoud, M-J., Rizkallah, M., Kreichati, G. & Kharrat, K. Etiology, evaluation, and treatment of failed back surgery syndrome. Asian spine J. 12 (3), 574. https://doi.org/10.4184/asj.2018.12.3.574 (2018).
Inoue, S. et al. Prevalence, characteristics, and burden of failed back surgery syndrome: the influence of various residual symptoms on patient satisfaction and quality of life as assessed by a nationwide Internet survey in Japan. J. Pain Res. 811–823. https://doi.org/10.2147/jpr.s129295 (2017).
Imani, B., Hajilo, P., Zandi, S. & Mehrafshan, A. Comparing the intraoperative and postoperative complications of the scalpel and electrocautery techniques for severing the inner layers of the lumbar disc during discectomy surgery. Front. Surg. 10, 1264519. https://doi.org/10.3389/fsurg.2023.1264519 (2023).

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Risk factors analysis and risk prediction model for failed back surgery syndrome: a prospective cohort study

Status:

Version 1

Abstract

Introduction:

Materials and Methods

Findings:

Conclusion

Figures

1. Introduction

2. Materials and Methods

2.1 Study Design

2.2. Inclusion and Exclusion criteria

2.3. Instruments and Data collection

2.3.1 Pre-Operative

2.3.2 During Operative

2.3.3 Postoperative

2.4. Statistical Analysis

3. Results

3.1. Baseline Characteristics

3.2. Pre-Surgery Outcome

3.3. During Surgery Outcome

3.4. Post-Surgery Outcome

4. Discussion

5. Conclusion

Declarations

Consent to participate

Competing interests

CRediT authorship contribution statement

Funding

Author Contribution

Acknowledgements

Data Availability

Ethics approval

References

Additional Declarations

Status:

Version 1