Characteristics and Factors Associated With Spontaneous Regression of Extruded Lumbar Disc Herniation: A Retrospective Analysis of 31 Cases

Our ndings support that patients with extruded or sequestered LDH should receive comprehensive treatment. Early treatment, early bed-rest, and lumbosacral orthosis protection promote the spontaneous regression of herniated discs; however, bed-rest during rehabilitation is a poor prognostic factor. Adherence to rehabilitation exercise will be benecial to the recovery of patients with LDH.


Introduction
Lumbar disc herniation (LDH) is a frequently-occurring disease that imposes a heavy burden on both individuals and society. First line treatment is nonoperative treatment [1], which includes drug therapy, such as oral medicine, epidural anesthesia, also includes non-drug therapy, such as physical therapy, acupuncture, and traction also good options [2] [3]. Since Guinto et al. reported the rst case of spontaneous resorption of LDH in 1984 [4], an increasing number of cases of spontaneous regression have been described, with reported rates of spontaneous resorption varying from 35-100% [5]; for example, the incidence in the United Kingdom was 82.94%, while in Japan it was 62.58%. The overall incidence of spontaneous resorption after LDH, based on analysis of the ndings from 11 cohort studies, was 66.66% [6].
The rate of spontaneous regression is in uenced by numerous factors, including hernia type, hernia volume, age, optimal time of absorption, and other factors. Disc sequestration has the highest reabsorption rate, followed sequentially by extrusion, protrusion, and bulge [5]. The incidence of spontaneous absorption is highest in patients between 20 and 65 years old [6], and magnetic resonance imaging (MRI) showed that the fastest spontaneous absorption occurred within 2 months, and that absorption can still occur after one year [7]. The overall absorption rate was reported as 42.86% in one study [5], and 33.33% in another [8]. Factors including imaging modalities, inclusion criteria, follow-up time, classi cation of herniation, and regression criteria can impact the evaluation of spontaneous regression of LDHs; however, dynamic changes in herniated intervertebral discs should also be considered. Herniated lumbar discs are not always spontaneously absorbed, and both absorption and reprotrusion can occur simultaneously [9]. Family history, smoking, obesity, diabetes, taller height, and heavy work are important factors negatively in uencing the absorption of LDH, while physical exercise and bed-rest are favorable [10,11].
Spontaneous regression generally takes between 3 and 24 months [5] and how to ensure that herniated discs progress towards spontaneous absorption, rather than re-protrusion, is an unsolved problem of interest to both patients and clinicians. There are rare cases of spontaneous resorption of LDH, so a retrospective study was performed. The purpose of this study was to identify characteristics and factors enabling disc resorption by conducting a retrospective clinical analysis of 33 recovered patients with LDH.

Patients and methods
The subjects of this retrospective analysis study were 33 patients with LDH who received conservative treatment during hospitalization at Henan Provincial Luoyang Orthopedic-Traumatological Hospital (Henan Provincial Orthopedic Hospital) from September 2015 to June 2020. Their symptoms were relieved, and spontaneous resorption of the lumbar disc had occurred. Subjects participated in a questionnaire survey. The inclusion criteria for the study were: lumbar disc herniation, which was rst identi ed by MRI, and follow-up MRI showing that the herniated disc was completely or partially absorbed, and that patients received nonoperative treatment, and their clinical symptoms improved signi cantly.

Questionnaire
Attending physicians contacted patients via WeChat or telephone. After the patient consented, they received a paper or electronic version of the questionnaire. Subjects completed the questionnaire themselves, or with the help of family members, and returned it to their attending doctor. The questionnaire consists of closed and open-ended questions, for example, patients need to ll in time in bed each day and number of days spent in bed. The questionnaire had three sections: 1) socialdemographic characteristics (e.g., age, sex, height, weight, education level, occupation, smoking, drinking alcohol); 2) treatment and rehabilitation (e.g., time of symptoms onset, complications, onset-to-treatment interval, received treatments, bed-rest, lumbosacral orthosis, rehabilitation exercise, relapse symptoms, inducements, employment, mood, family support, sequelae); 3) self-rated enabling factors, this is an open-ended question, patients describe factors that they believes are conducive to disc absorption.
The questionnaire de ned and categorized 1) smoking, 2) alcohol consumption, 3) occupation type, 4) mood, and 5) family support as follows: 1) smoking and non-smoking, where smoking was de ned as having smoked at least 1 cigarette per day for at least 1 year; 2) alcohol abuse and no addiction, where alcohol abuse was de ned as the alcohol consumption > 20g per day, or drunkenness at least twice a mont; 3) occupations were de ned as physical workers, mental workers and others; 4) normal and emotional distress, self-evaluated depression, anxiety; and 5) support for nonoperative treatment, support for surgery, and disagreement.

Hernia measurement
The volume, location, and type of herniated disc were assessed using a 3.0 Tesl MRI scanner (Siemens MAGNETOM Skyra); 13 sections were scanned on T1 and T2 weighted sagittal planes, with layer thickness 3 mm and distance 3 mm. Herniated disc volume was calculated by two skilled radiologists, based on the methods previously described by Peng-Fei Yu et al. [12] (Fig. 1). Resorption ratios were calculated as follows: Resorption ratio = (pre-treatment herniated disc volume -post-treatment herniated disc volume/pretreatment herniated disc volume) × 100%.

Statistical analysis
Statistical analysis was performed using SPSS 20.0 software. Continuous data are expressed as mean ± standard deviation (SD), count data are represented as constituent ratio(%). The independent samples ttest and one-way ANOVA were used to determine statistical signi cance (α = 0.05). Variables with signi cant differences and risk factors associated with LDH, based on published articles, were incorporated in multivariate regression analysis. Different categorical variables were treated as dummy variables and then included in regression analysis. The optimal regression equation was generated using multivariable linear regression models (by the backward method), and gradually eliminating nonsigni cant independent variables.

General patient characteristics
A total of 33 subjects received the questionnaire, of which 31 and 2 did and did not complete it, respectively. Of the 31 included patients, 13 (41.94%) were male and 18 (58.06%) female. Mean ± SD (range) age was 36.97 ± 9.48  years. Regarding education, 23 patients (74.19%) had college education or higher. Ten (32.26%) patients had emotional distress and 28 (90.32%) families supported conservative treatment. During the rehabilitation period, 29 patients (93.55%) received follow-up with a clinician or returned for follow-up visits (Table 1). Of 33 herniated lumbar discs from the 31 included patients, 31 were partly or completely reduced in size (Fig. 2). The majority of herniated discs occurred at the L4-L5 (32.26%) and L5-S1(61.29%) levels. The mean spontaneous absorption ratio was 0.69 ± 0.22. During hospitalization, all patients received numerous Traditional Chinese Medicine(TCM) therapies and modern medicine to relieve symptoms, with a mean number of therapies per person of 5.90. TCM fumigation, physiotherapy, traction, massage, and acupuncture were commonly used non-drug therapies, while, oral drugs and sacral anesthesia were commonly used drug therapies. During rehabilitation, the number of subjects receiving outpatient treatment reduced to 21 (67.74%), and the mean number of treatment methods reduced to 1.91 per patient; at this stage, plasters, oral drugs, and patented Chinese medicines became the most frequent treatment methods (Table 2). Self-rating of factors enabling rehabilitation Adherence to rehabilitation exercise, early bed-rest, timely treatment, and proper rest were the top-ranking factors enabling recovery, according to patient self-rating. Con dence, positive attitude, and effective doctor-patient communication were also considered necessary. Further, professional guidance from a physician and appropriate treatment were also considered to be facilitating factors (Table 3). Multivariate analysis of factors enabling reabsorption of herniated lumbar disc Type of herniated disc, duration of wearing a lumbosacral orthosis device each day, number of days wearing a lumbosacral orthosis device, acute bed-rest per day, acute bed-rest duration, rehabilitation bedrest per day, rehabilitation bed-rest duration, exercise per day, exercise duration, and amount of time working were included into multiple linear regression models, and non-signi cant independent variables gradually eliminated to determine the optimal regression equation model.
The mean reabsorption ratio was 0.69 ± 0.22, regression analysis could be performed, as the data followed a normal distribution (df = 31, < 50; Shapiro-Wilk test, P = 0.08 (> 0.05)). Adjusted R 2 and F values indicated that the regression equation was very signi cant, and that the degree of t between the independent and the dependent variables was good. The variance in ation factor values of the independent variables were all < 2 (i.e., << 10), indicating no collinearity between the independent variables in this model (Table 4). The data presented in Table 4 demonstrate that the number of bed-rest days in the acute phase had the greatest impact on the herniated lumbar intervertebral disc reabsorption ratio (standard partial regression coe cient, 0.50), followed by herniated disc type (-0.48), time per day wearing a lumbosacral orthosis device (0.38), onset-to-treatment interval (-0.38), and days of bed-rest during rehabilitation (-0.24). The reabsorption ratio (y) was set as the dependent variable, with bed-rest days in the acute phase (X 1 ), type of herniated disc (X 2 ), time per day wearing a lumbosacral orthosis device (X 3 ), onset-treatment interval (X 4 ), and bed-rest days during rehabilitation (X 5 ) set as independent variables. The optimal linear regression equation was y = 0.616 + 0.497X 1 -0.478X 2 + 0.

Discussion
The main mechanism underlying spontaneous resorption of LDH has been con rmed as contact of the nucleus pulposus perforate posterior longitudinal ligament with blood, which elicits an immune response [9]. Resorption of LDH takes 3-12 months [5]. Patient experience of pain, mood, and family support are important factors that in uence whether a treatment plan can be implemented [13]. The treatments patients receive, as well as their lifestyle, type of work, and rehabilitation exercises, can also affect the disease course [14]. Methods to promote absorption of the herniated disc and shorten the disease course warrant exploration. Because most of LDH patients were treated with surgery, prospective research on resorption of LDH is di cult; therefore, we conducted a retrospective analysis of patients with spontaneous absorption of LDH.
Physical and mental health parameters in patients undergoing conservative treatment Although surgery has advantages in relieving early or mid-term pain and promoting early recovery in patients with LDH, its long-term e cacy does not differ from that of conservative treatment [15][16][17]. The main purpose of conservative treatment is to alleviate pain, increase motor function, improve quality of life, and provide adequate time for reabsorption to occur. Intensive conservative treatment can reduce the number of surgeries required for patients with non-contained LDH within 1 month of onset [18]. During hospitalization, patients in this study received a mean of 5.90 treatment methods each. Most patients (67.74%) also received intensive treatment during the rehabilitation period, with a mean of 1.91 outpatient treatments per patient. In this study, the majority of patients agreed that timely treatment, oral medicine, and hot compress were factors that enabled spontaneous absorption.
Early physical therapy can improve pain relief e cacy, function, and disability indices of patients with low back pain [19]. Acute pain management of non-speci c chronic low back pain is conducive to recovery and return to work, and also reduces costs [19,20]. Our data show that early treatment is bene cial to herniated disc absorption in patients with extrusion or sequestration.
For patients with chronic diseases, especially chronic low back pain, a good doctor-patient relationship is essential to treatment adherence [21]. The impact of the doctor-patient relationship is more obvious in the medium term (6 months after recovery) than in the short term [22]. A good doctor-patient relationship is conducive to recovery via spontaneous absorption in patients with LDH [23]. In this study, doctors and patients could communicate to ensure effective treatment according to patient symptoms during hospitalization, and during rehabilitation, 29 (93.55%) patients received subsequent visit(s) and follow-up by phone or WeChat. Patients also considered effective doctor-patient communication a favorable factor for recovery.
Anxiety or depression are common in patients with chronic low back pain; in one study, 56 (61.5%) such patients were reported to suffer from anxiety or depression before surgery [24]. Further, the prevalence rates of mood change and anxiety in 149 patients with LDH were 16.6% and 35.8%, respectively [25]. The most common speci c diagnoses were major depression (16.9%) and generalized anxiety disorder (12.8%) [25]. Higher pain intensity and longer disease duration are independent risk factors for depression [26]. The use of microsurgery to remove the intervertebral disc can relieve pain, while also reducing physical anxiety and depression [27].
Family support for patients with chronic pain is also an important factor affecting mood. Patients with family support reported signi cantly lower pain intensity, drug dependence, and higher physical activity levels than those without [28]; for example, patients with knee arthritis who have strong family and spousal support had signi cantly fewer depression symptoms, and pain levels decreased with increasing family support [29]. In this study, although only 9.68% of patient family members did not agree with longtime conservative treatment, none rmly opposed this therapy option. Ten patients (32.26%) in this study experienced emotional distress, which is a lower rate than that in previous report [25]. This may be related to the fact that patients were hospitalized and received daily professional help from a physician. A good doctor-patient relationship has an important impact on rehabilitation [30], and 29 (93.55%) patients received follow-up visits via telephone or WeChat during rehabilitation; thus, they had access to professional help.

Symptoms and associated factors
Fourteen subjects (45.16%) were sensitive to factors such as prolonged walking, prolonged sitting, excessive weight bearing, and cold, which provoked waist discomfort, affected limb discomfort, numbness of the affected shank, and sacrococcygeal discomfort. Although patients with LDH exhibit a degree of fear avoidance [31], such beliefs are more common among patients with chronic low back pain than those with acute low back pain [31]; however, fear avoidance cannot explain patient sensitivity to factors such as prolonged sitting, fatigue, and cold. In ammatory mediators released at the herniated disc can alter the expression of sodium, potassium, and calcium ion channels on the surface of the dorsal root nerve, causing ectopic and continuous discharge [32], which is related to sensitization of spinal dorsal horn cells and consequent hyperalgesia [33]. Simultaneously, spinal dorsal horn microglia are activated, and the expression of phosphorylated SRC family kinases up-regulated [34]. Long term nociceptive signals input caused by the herniated disc alters the morphology and/or cell structure of the anterograde projection area [35]. Functional MRI shows that the right dorsolateral prefrontal cortex and left anterior cingulate cortex of the pain-related area are activated in patients with chronic low back pain [36]. Further, patients with chronic low back pain have central sensitization, with sensitivity to noxious stimulus [37], as well as innocuous stimuli [38]. Lower than normal pressure levels can produce mild and severe pain in patients with chronic low back pain [36]. Further, patients with chronic low back pain have lower cold pain thresholds at the waist and in distal body parts [39]. Nevertheless, changes in weather temperature do not increase the risk of low back pain [40]. Since the autoimmune responses to herniated disc exceed 3 months, prolonged painful stimulation can cause central sensitization. Patient sensitivity to fatigue and cold may also be related to central sensitization.

Factors favorable for herniated disc absorption
Hernia type is an important factor affecting the occurrence of spontaneous resorption. Sequestered herniations have the highest proportion of spontaneous resorption, followed by transligamentous, with the lowest rates for subligamentous [8]. Further, the probability of sequestrated discs being fully resorbed is higher than that of extruded discs [5,41]. Three hypotheses, including retraction, absorption, and dehydration, have been proposed to explain the resorption of herniated discs. Matrix metalloproteinases (MMPs) are the main metabolic enzymes involved in the process of apoptosis and self-absorption of intervertebral disc protrusions. Numerous mononuclear macrophages and MMP-3 have been detected in the herniated nucleus pulposus [42]. Levels of MMP-1, 3, and 13 in sequestered discs are signi cantly higher than those in contained discs [43]. In our study, the spontaneous resorption ratio of sequestration type hernia was signi cantly higher than that of extrusion type, con rming that the type of herniation is an important factor in spontaneous absorption.
The ruptured annulus brosus has reduced ability to x the nucleus pulposus, and bed-rest can alleviate the load on spine intervertebral discs, thereby reducing the chance of further nucleus pulposus protrusion [44,45]. When resting in bed, the height and area of the intervertebral disc increase signi cantly; however, the water signal intensity does not alter [46]. A meta-analysis showed that bed-rest is effective in reducing pain and improving lumbar function in patients with acute low back pain [47]. In self-evaluation, patients recommended bed-rest during the acute period as a favorable factor. Further, our regression analysis supported early bed-rest for patients with sequestration or extrusion disc herniation. Although the optimal duration of bed-rest remains uncertain, it has been recommended that patients with LDH remain in bed for < 1 week [48], and 2 weeks has also been suggested [49]. Based on the mean bed-rest day of the 31 patients in this study, we advise that patients with sequestration or extrusion disc herniation should remain in bed for approximately 3 weeks, to alleviate the load on ruptured intervertebral discs. However, long-term bed-rest during the early stages of pain is associated with higher levels of long-term disability [50]. Moderate quality evidence suggests that there is little difference between patients with sciatica who undergo bed-rest and those who stay active, in terms of pain reduction and function improvement [47].
Another study also found that bed-rest does not improve sciatica [51] because radicular pain is caused by various factors, including mechanical compression, in ammatory stimulation, and autoimmunity [35].
For patients with non-speci c low back pain, prolonged bed-rest during the initial stages of symptoms is associated with higher levels of long-term disability [50], and can cause major cardiovascular, respiratory, musculoskeletal, and neuropsychological changes, as well as bone loss, which may delay or prevent recovery from critical illnesses, including disuse muscle atrophy, joint contractures, thromboembolic disease, and insulin resistance [53] [54] [55]. In this study, regression analysis showed that prolonged bedrest duration in rehabilitation was negatively correlated with the reabsorption ratio; hence, prolonged bedrest should be avoided during the rehabilitation period.
MRI analysis showed that the cross-sectional area of the iliopsoas muscle was reduced on the 14th day of strict bed-rest, while the sartorius muscle was reduced on the 42nd day [56]. After living in space weightlessness for 6 months, lumbar lordosis was reduced by an average of 11%, when astronauts were supine, and the functional cross-sectional area of the multi dus muscle and erector spinae were reduced by 20%, with the cross-sectional area reduced 8-9%. These data suggest that prolonged bed-rest causes lumbar muscle atrophy and a decline in spinal stability [57]. Trunk extensor strength is bene cial in increasing spinal stability and improving its function [58]. Early functional exercises, including passive and autonomous activities, can improve patient physiological function following percutaneous transforaminal endoscopic discectomy for LDH, as well as improving emotional function, mental health, and quality of life [59]. Dynamic lumbar stabilization training can relieve pain, reduce functional disabilities, and decrease patient fear avoidance behavior following lumbar microdiscectomy [60]. Early strengthening exercises include Williams exion exercises and/or McKenzie extension exercises, which can increase spine compression by 12-18% and stability by 34-64% [61]. Eight weeks of core stability training improves the static endurance of the trunk muscles of patients with LDH to a normal level [62]. Rehabilitation exercise was the enabling factor rated highest in patient self-evaluation in this study.
Active improvement of trunk extensor strength can help to prevent adverse reactions to prolonged bedrest, enhance patient mood, and improve and maintain spinal function.
Lumbosacral orthosis is widely used to relieve low back pain. Moderate evidence shows that lumbosacral orthosis is more effective in preventing low back pain and improving pain and function than lack of intervention and training [63] [64]. Home care workers who wore lumbosacral orthosis devices had a reduced number of days of pain, but no reduction in sick days [65]; however, another study showed that lumbosacral orthosis did not reduce the incidence of low back pain in workers [66]. Hence, lumbosacral orthosis is effective in improving pain, but ineffective as primary prevention [63]. Rupture of the annulus brosus and decreased pressure on the intervertebral disc increases spinal exion and rotation instability [67]. Lumbosacral orthosis improves the stiffness and stability of the spine by increasing abdominal pressure. The results of this study suggest that lumbosacral orthosis is bene cial for the absorption of herniated discs that break through ligaments and prolapse. Wearing a lumbosacral orthosis device during the acute phase of LDH is a favorable factor, which may be related to increased spine stability and reduced pressure on the intervertebral disc. Nevertheless, lumbosacral orthoses may reduce back muscle activity, leading to a concern that long-term wear may cause trunk muscle weakness; however, wearing lumbosacral orthoses for 6-months does not cause weakening of the paravertebral muscles in patients with chronic low back pain [68]. Further, there is no scienti c evidence that prolonged wearing of lumbosacral orthosis devices results in trunk muscle weakness [69].
BMI is positively correlated with pain and the Roland-Morris Disability and Oswestry Disability indices, and logistic regression analysis demonstrated that higher BMI is a risk factor for LDH recurrence [11]. Here, we did not identify a relationship between BMI and spontaneous regression ratio; however, this does not mean that weight loss is not bene cial for herniated disc regression.
Our survey had several limitations, as follows: rst, there was selection bias, as all subjects included in the study were hospitalized patients from Henan Provincial Luoyang Orthopedic-Traumatological Hospital (Henan Provincial Orthopedic Hospital) who had severe symptoms, hence they are not representative of patients with mild or moderate symptoms; second, this is a retrospective study, spanning a period of 4 years 3 months, hence patients will have exhibited memory bias; third, due to the wide application of spinal surgery techniques, most patients with severe LDH received surgeries, meaning that the number of collected cases was small, leading to potential data bias; fourth, few of the included patients were underweight or overweight, making it di cult to determine the effect of weight on reabsorption; nally, as patients were hospitalized in a Chinese medicine hospital, they received more TCM treatment techniques, which may have resulted in a bias toward Chinese medicine culture and treatment techniques.

Conclusions
To improve the spontaneous absorption ratio, pain management used for lumbar disc sequestration or extrusion should be applied in multiple ways. In the acute phase, patients require early bed-rest and should wear lumbosacral orthosis devices, while the duration of bed-rest days should be reduced during the recovery period. Symptomatic treatments are crucial to symptom relief during the acute and The clinical trial has been approved by the Institutional Review Board of Henan Provincial Luoyang Orthopedic-Traumatological Hospital (Henan Provincial Orthopedic Hospital). This study is to be conducted in accordance with the principles of the Declaration of Helsinki. All participants provided informed consent.

Consent for publication
The authors have received written consent from participants to publish individual images.
Availability of data and materials