Mechanisms and management of self-resolving lumbar disc herniation: bridging molecular pathways to non-surgical clinical success

Abstract

Spontaneous resorption of lumbar disk herniation (LDH) presents a promising avenue for the non-surgical management of herniated disks. Here we present a 40-year-old female with severe L5/S1 herniation who experienced spontaneous resorption confirmed by MRI. The patient adhered to a comprehensive non-operative treatment regimen comprising NSAIDs, heat therapy, massage therapy, acupuncture, and kinesitherapy. Over two years, the patient showed substantial clinical improvement, with significant pain reduction and MRI evidence of disk resorption, resulting in a marked decrease in their visual analog scale (VAS) pain scores. Through a systematic review of existing literature, we identify that spontaneous resorption is associated with mechanisms such as inflammation, neovascularization, macrophage infiltration, matrix degradation, disruption of immune privilege, apoptosis and autophagy, and disc dehydration. Clinical predictors for spontaneous resorption of LDH include the size, type and composition of the herniation, rim enhancement on contrast-enhanced MRI, and involvement of the posterior longitudinal ligament (PLL). Future research should focus on elucidating the molecular mechanisms of resorption, regulation of inflammatory response, macrophage polarization, matrix degradation, immune privilege and neovascularization, developing advanced imaging techniques to predict resorption potential, and exploring personalized treatment strategies based on machine learning and deep learning prediction models.

Lumbar disc herniation (LDH) is a common cause of low back pain, radiculopathy, and functional disability. Recent studies have increasingly demonstrated the phenomenon of spontaneous resorption, where herniated disc material shrinks or disappears without surgical intervention [1, 2]. Spontaneous resorption of lumbar disk herniation (LDH) is a phenomenon that has garnered attention due to its implications for non-surgical management in patients with herniated disks [3, 4]. Here, we a patient that underscore the possibility and potential of this non-invasive approach, highlighting the clinical progression, therapeutic measures, and radiological evidence confirming spontaneous disk resorption.

Case presentation

All the procedures of this study were approved by the ethics committee of the Sixth Affiliated Hospital of Xinjiang Medical University, all the procedures followed the guidelines of the declaration of Helsinki.

In 2021, a 40-year-old female presented to the department of spine surgery with a one-year history of severe lower back pain radiating to both legs. She had no history of injuries, chronic diseases, or prior surgeries. On presentation, her Visual Analog Scale (VAS) score for lower back pain was eight. Physical examination revealed that vertebral flexion, extension, and lateral flexion markedly aggravated pain in the buttocks and the rear lateral surface of both legs down to the ankles.

MRI scans at that time confirmed the presence of a large herniated disk at the L5/S1 level (Fig. 1, Panel A). Despite a recommendation for surgical intervention to remove the herniated disk, the patient declined surgery. Instead, she opted for a comprehensive non-operative treatment protocol. This regimen included non-steroidal anti-inflammatory drugs (NSAIDs), heat therapy, massage therapy, acupuncture, and kinesitherapy aimed at strengthening the lower back muscles.

Fig. 1.

MRI Scans of L5/S1 Herniated Disk Before and After Non-Operative Treatment. Panel A: Initial MRI scan revealing a large herniated disk at the L5/S1 level (indicated by the arrow). Panel B: Follow-up MRI scan at 24 months showing spontaneous resorption of the herniated disk at L5/S1 (indicated by the arrow). The patient declined surgical intervention and instead pursued a comprehensive non-operative treatment protocol, including NSAIDs, heat therapy, massage therapy, acupuncture, and kinesitherapy. Over two years, her symptoms resolved completely, with no recurrence of back or leg pain and significant improvement in functional mobility

Over the subsequent two years, the patient’s back pain gradually alleviated, and the radiating pain in her legs completely subsided. At 24 months, she reported no recurrence of back or leg pain, with her VAS pain score for lower back pain reduced to one. Additionally, vertebral flexion, extension, and lateral flexion no longer aggravated pain in the buttocks or legs. Follow-up MRI demonstrated spontaneous resorption of the herniated disk at L5/S1 (Fig. 1, Panel B).

Discussion

Lumbar disc herniation is a prevalent condition in spinal surgery, characterized by low back pain, radicular leg pain, and neurological deficits. Traditional treatment approaches include conservative management and surgical intervention. However, studies have revealed that spontaneous resorption of herniated disc material can occur without surgery, providing a rationale for conservative treatment and altering clinical management strategies [5–34] (see Table 1).

Table 1.

The incidence of spontaneous resorption of LDH

Study	Country	Study design	%(T/R) SRLDH	Age
Teplick et al., 1985 [5]	USA	Retrospective	20.0 (55/11)	NA
Saal et al., 1990 [6]	USA	Prospective	81.8 (11/9)	NA
Bozzao et al., 1992 [7]	Italy	Prospective	66.7 (21/14)	20–64
Bush et al., 1992 [8]	UK	Prospective	64.0 (111/71)	17–72
Maigne et al., 1992 [9]	France	Prospective	81.3 (48/39)	26–75
Ellenberg et al., 1993 [10]	USA	Prospective	78.6 (14/11)	28–67
Gallucci et al., 1995 [11]	Italy	Prospective	73.3 (15/11)	27–62
Matsubara et al., 1995 [12]	Japan	Prospective	64.0 (32/20)	16–52
Yukawa et al., 1996 [13]	Japan	Retrospective	53.3 (30/16)	14–69
Komori et al., 1998 [14]	Japan	Prospective	66.7 (48/32)	20–75
Takada et al., 2001 [15]	Japan	Prospective	88.1 (42/37)	16–64
Ahn et al., 2002 [16]	Korea	Prospective	76.5 (17/13)	19–73
Splendiani et al., 2004 [17]	Italy	Prospective	34.7 (72/25)	21–68
Autio et al., 2006 [18]	Finland	Retrospective	91.9 (74/68)	19–78
Erly et al., 2006 [19]	USA	Retrospective	69.4 (36/25)	NA
Jensen et al., 2006 [20]	Denmark	Prospective	46.8 (139/65)	18–65
Cribb et al., 2007 [21]	UK	Retrospective	93.3 (15/14)	24–73
Benson et al., 2010 [22]	UK	Prospective	100 (28/28)	25–62
Iwabuchi et al., 2010 [23]	Japan	Prospective	61.8 (34/21)	NA
Barzouhi et al., 2013 [24]	Netherlands	RCT	92.6 (95/88)	18–65
Yu et al., 2014 [25]	China	Prospective	50.6 (83/42)	16–60
Hong et al., 2016 [26]	Korea	Retrospective	85.7 (28/24)	26–78
Demirel et al., 2017 [27]	Turkey	RCT	90.0 (20/18)	NA
Lee et al., 2017 [28]	Korea	Retrospective	96.2(505/486)	NA
Kesikburun et al. 2019 [29]	Turkey	Prospective	90.0 (40/36)	39.7–66.7
Dai et al., 2020 [30]	China	Prospective	69.7 (66/46)	25–67
Ma et al., 2021 [31]	China	Retrospective	46.2(409/189)	14–70
Sucuoglu et al., 2021 [32]	Turkey	Prospective	89.1 (55/49)	25–67
Hournung 2023 [33]	China	Retrospective	100 (93/90)	20–51
Budrovac 2023 [34]	Croatia	RCT	66.7 (30/20)	18–65

%(T/R) SRLDH: Percentage (resorption/total) of spontaneous resorption of LDH

Mechanisms of spontaneous resorption in lumbar disc herniation

Inflammation

The inflammatory response is one of the core mechanisms of spontaneous resorption of LDH. When herniated disc tissue enters the epidural space, it triggers an immune response, leading to the infiltration of inflammatory cells and the release of inflammatory mediators. These inflammatory mediators include tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), interleukin-6 (IL-6). They accelerate the degradation and absorption of the herniated tissue by activating matrix metalloproteinases (MMPs) and promoting neovascularization [35–37]. TNF-α, IL-1β, IL-6, and IL-8 are released in the early stages of the inflammatory response, further promoting the infiltration of inflammatory cells and the persistence of the inflammatory response. These inflammatory mediators play a significant role in the spontaneous resorption of LDH. TNF-α and IL-1β can stimulate disc cells to produce more inflammatory mediators and chemokines, such as monocyte chemoattractant protein-1 (MCP-1), thereby attracting more inflammatory cells (e.g., macrophages) to the site of the herniated tissue. The release of inflammatory mediators creates a positive feedback loop, further amplifying the inflammatory response. For example, TNF-α can induce the expression of MCP-1, which in turn attracts more macrophages to the herniated tissue, leading to the release of more TNF-α and IL-1β. In the later stages of the inflammatory response, the release of anti-inflammatory mediators (such as IL-4 and IL-10) gradually suppresses the inflammatory response, promoting tissue repair and absorption.

Neovascularization

Neovascularization is a critical mechanism in the resorption process. New blood vessels form around the herniated disc tissue, providing a pathway for inflammatory cells and degrading enzymes to enter the herniated tissue, thereby promoting the absorption process [38]. Vascular endothelial growth factor (VEGF) and other angiogenic factors are upregulated in herniated disc tissue, facilitating the formation of new blood vessels and accelerating the resorption process [39, 40]. Studies have shown that VEGF is highly expressed in herniated disc tissue, promoting the proliferation and migration of vascular endothelial cells, leading to the formation of new blood vessels. Basic fibroblast growth factor (bFGF) and platelet-derived growth factor (PDGF) also contribute to the formation of new blood vessels. These growth factors stimulate the proliferation and migration of vascular endothelial cells, promoting angiogenesis.

The functions of neovascularization are primarily reflected in the following two aspects: (1) Infiltration of inflammatory cells: New blood vessels provide a pathway for inflammatory cells (e.g., macrophages) to enter the herniated tissue, thereby sustaining the inflammatory response and promoting the degradation of the herniated tissue. (2) Supply of nutrients and oxygen: New blood vessels deliver essential nutrients and oxygen to the herniated tissue, supporting the activity of inflammatory cells and degrading enzymes, thus facilitating the resorption process.

Macrophage infiltration

Macrophages play a critical role in the spontaneous resorption of LDH. When herniated disc tissue enters the epidural space, it triggers the infiltration and activation of macrophages. Macrophages promote the degradation and absorption of herniated tissue by phagocytizing cellular debris and secreting inflammatory mediators.

In the recruitment of macrophages, chemokines and inflammatory mediators play a major role. Chemokines such as MCP-1 and MIP-1α secreted by disc cells and inflammatory cells, attract macrophages to the site of the herniated tissue. Inflammatory mediators such as TNF-α and IL-1βalso promote the recruitment and activation of macrophages. These inflammatory mediators stimulate disc cells and inflammatory cells to produce more chemokines, thereby attracting additional macrophages.

The primary functions of macrophages include phagocytosis and the secretion of inflammatory mediators and matrix-degrading enzymes: (1) Macrophages can phagocytize cellular debris and degradation products in the herniated tissue, thereby reducing the volume of the herniated mass. (2) Macrophages secrete various inflammatory mediators such as TNF-α, IL-1β, and IL-6, which further promote the inflammatory response and matrix degradation. (3) Macrophages secrete matrix metalloproteinases (MMPs), which degrade collagen and proteoglycans in the disc tissue, facilitating the absorption of the herniated tissue.

Matrix degradation

Matrix degradation is the final step in the spontaneous resorption of LDH. Matrix metalloproteinases (MMPs) are key enzymes in matrix degradation, capable of breaking down collagen and proteoglycans in the disc tissue, thereby promoting the absorption of herniated tissue. Studies have shown that herniated disc tissue contains a high density of macrophages, which secrete enzymes such as matrix metalloproteinases (MMPs) that facilitate the degradation and remodeling of the disc material [41–43].

MMP-3 and MMP-7 play a significant role in the spontaneous resorption of LDH. These matrix metalloproteinases degrade collagen and proteoglycans in the disc tissue, promoting the breakdown and absorption of herniated tissue. The activity of MMPs is regulated by various factors, including inflammatory mediators (such as TNF-α and IL-1β) and growth factors (such as VEGF and bFGF). These factors stimulate disc cells and inflammatory cells to produce more MMPs, thereby accelerating matrix degradation.

The progression and resolution of the inflammatory response significantly influence the regulation of matrix degradation. In the early stages of the inflammatory response, the release of inflammatory mediators promotes the production and activity of MMPs, accelerating matrix degradation. In the later stages, the release of anti-inflammatory mediators inhibits the production and activity of MMPs, slowing down matrix degradation. Additionally, the formation and function of new blood vessels play a crucial role in regulating matrix degradation. New blood vessels provide a pathway for inflammatory cells and degrading enzymes to enter the herniated tissue, thereby promoting matrix degradation.

Disruption of immune privilege

Normal disc tissue possesses immune privilege, meaning the immune system does not mount a strong immune response against it. However, when herniated disc tissue enters the epidural space, this immune privilege is disrupted, triggering an immune response. The primary mechanisms of immune privilege include the blood-nucleus pulposus barrier and the Fas/FasL system.

Normal disc tissue is isolated from the immune system by the blood-nucleus pulposus barrier, which maintains immune privilege. This barrier, composed of disc cells and extracellular matrix, prevents the entry of immune cells and inflammatory mediators. When herniated disc tissue enters the epidural space, it disrupts the blood-nucleus pulposus barrier, exposing the tissue to the immune system. This exposure triggers an immune response, leading to the infiltration of inflammatory cells and the release of inflammatory mediators.

The Fas/FasL system also plays a critical role in maintaining the immune privilege of disc tissue. FasL, expressed by disc cells, induces apoptosis in immune cells and vascular endothelial cells, thereby preventing immune cell infiltration and neovascularization. Dysregulation of the Fas/FasL system leads to immune cell infiltration and the formation of new blood vessels, promoting inflammation and matrix degradation.

Apoptosis and autophagy

In herniated disc tissue, the incidence of apoptosis is relatively high. Apoptosis is primarily induced by inflammatory mediators such as TNF-α and IL-1β, and growth factors such as VEGF and bFGF, which promote the degradation and absorption of herniated tissue. The regulation of apoptosis involves multiple signaling pathways, including the p38 MAPK pathway. Activation of the p38 MAPK pathway can induce apoptosis in disc cells, accelerating the absorption of herniated tissue.

Autophagy is a cellular self-degradation process that removes damaged components and metabolic waste within cells. In herniated disc tissue, the incidence of autophagy is relatively high, contributing to the degradation and absorption of herniated tissue. The regulation of autophagy involves multiple signaling pathways, including the PI3K/AKT pathway. Activation of the PI3K/AKT pathway can inhibit autophagy, thereby slowing the absorption of herniated tissue.

Disc dehydration

Disc dehydration is another proposed mechanism of spontaneous resorption. The herniated disc material gradually loses water content, leading to shrinkage and eventual retraction into the annulus fibrosus. MRI studies have shown that the signal intensity of herniated disc material on T2-weighted images decreases over time, indicating a reduction in water content [44–46]. However, the role of disc dehydration in resorption remains controversial and requires further investigation [47–49].

Clinical predictors of spontaneous resorption

Size of the herniation

Larger disc herniations are more likely to undergo spontaneous resorption. Erly et al. found that larger herniations exhibit higher T2-weighted signal intensity on MRI, which correlates positively with resorption [50]. However, other studies have reported no significant relationship between herniation size and resorption, suggesting that this predictor requires further validation [51–53].

Type of herniation

The type of LDH herniation is closely related to the incidence of spontaneous resorption. Based on MRI imaging findings, LDH can be classified into bulging, protrusion, extrusion, and sequestration. Studies have shown that extrusion and sequestration types of LDH have higher rates of spontaneous resorption, while bulging and protrusion types have lower rates. This may be because the nucleus pulposus tissue in extrusion and sequestration types is more likely to come into contact with the bloodstream, triggering immune and inflammatory responses that promote resorption. Ahn et al. found that all sequestrated herniations exhibited partial or complete resorption [54]. A systematic review by Chiu et al. confirmed that the resorption rate for sequestrated herniations is as high as 96% [30].

Composition of herniated disk

The composition of the herniated disc also influences the incidence of spontaneous resorption. Studies have shown that patients with herniated tissue containing a higher proportion of nucleus pulposus are more likely to experience spontaneous resorption, whereas herniated tissue containing more cartilaginous endplate is less likely to be absorbed. This may be because nucleus pulposus tissue has a higher water content, making it more prone to vascularization and dehydration, thereby facilitating resorption.

Rim enhancement on contrast-enhanced MRI

The degree of rim enhancement on contrast-enhanced MRI is a significant predictor of spontaneous resorption. Autio et al. found that herniations with greater rim enhancement thickness were more likely to undergo resorption [55]. Rim enhancement reflects the extent of neovascularization and inflammation, making it a useful predictor of resorption [56–58].

Involvement of the posterior longitudinal ligament (PLL)

The involvement of the PLL is also associated with spontaneous resorption. Ahn et al. found that herniations penetrating the PLL were more likely to resorb [59]. Seo et al. corroborated this finding, suggesting that penetration of the PLL exposes the herniated material to the epidural vascular supply, facilitating resorption [60–62].

Modic changes

Modic changes, which are MRI findings associated with disc degeneration, have been linked to resorption. Shan et al. found that herniations without Modic changes were more likely to undergo resorption [63]. However, Kawaguchi et al. reported no association between Modic changes and resorption, indicating that this predictor requires further investigation [64–66].

Clinical predictive factors

In addition to imaging examinations, certain clinical features can also serve as reference indicators for predicting spontaneous resorption of LDH: (1) Duration of Symptoms: Patients with a shorter duration of symptoms are more likely to experience spontaneous resorption. This may be because the inflammatory response in patients with a shorter symptom duration is more active, promoting the resorption process. (2) Pain Severity: Patients with milder pain are more likely to experience spontaneous resorption. This may be because the inflammatory response in patients with milder pain is more moderate, facilitating the resorption process. (3) Neurological Deficits: Patients with milder neurological deficits are more likely to experience spontaneous resorption. This may be because the degree of compression from the herniated disc is less severe in patients with milder neurological deficits, favoring the resorption process.

Treatment strategies

Conservative treatment is the first-line approach for lumbar disc herniation and includes pharmacological therapy, physical therapy such as mechanical traction [67–69], and rehabilitation. Studies have shown that conservative management can effectively alleviate symptoms and promote spontaneous resorption of herniated disc material [70–72]. A meta-analysis by Wang et al. recommended a follow-up period of 4 to 10.5 months to assess the potential for resorption [73].

Future research directions

Regulation of inflammatory response

The inflammatory response is one of the core mechanisms of spontaneous resorption in LDH [74]. Future research should further explore the regulatory mechanisms of the inflammatory response to promote spontaneous resorption.

1) Targeted regulation of inflammatory mediators: TNF-α and IL-1β are key mediators of the inflammatory response. Future research could develop targeted drugs against these inflammatory mediators to regulate the intensity and duration of the inflammatory response. For example, anti-TNF-α antibodies and IL-1β receptor antagonists may help control the inflammatory response and promote spontaneous resorption. Anti-inflammatory mediators (such as IL-4 and IL-10) play a crucial role in the resolution of the inflammatory response. Future research could explore how exogenous anti-inflammatory mediators (such as recombinant proteins or gene therapy) can facilitate the resolution of the inflammatory response, thereby accelerating spontaneous resorption.

2) Regulation of inflammatory signaling pathways: NF-κB Signaling Pathway: The NF-κB signaling pathway plays a critical role in regulating the inflammatory response. Future research could investigate how to modulate the NF-κB signaling pathway, either by inhibition or activation, to regulate the inflammatory response and promote spontaneous resorption.

MAPK Signaling Pathway: The MAPK signaling pathway (e.g., p38 MAPK) plays an important role in regulating the inflammatory response and apoptosis. Future research could explore how to modulate the MAPK signaling pathway to promote spontaneous resorption.

Regulation of neovascularization

Neovascularization is another important mechanism in the spontaneous resorption of LDH. Future research should further explore the regulatory mechanisms of neovascularization to promote spontaneous resorption.

1) Application of angiogenic factors: VEGF is a key mediator of neovascularization. Future research could investigate how exogenous VEGF (such as recombinant proteins or gene therapy) can be used to promote the formation of new blood vessels, thereby accelerating spontaneous resorption. bFGF and PDGF also play roles in promoting neovascularization. Future research could explore how the combined application of these growth factors can enhance the formation of new blood vessels, facilitating spontaneous resorption.

2) Regulation of angiogenesis inhibitors: In some cases, excessive neovascularization may lead to the persistence of the inflammatory response and tissue damage. Future research could investigate how to modulate anti-angiogenic factors (such as TIMP-3) to balance the formation of new blood vessels, thereby promoting spontaneous resorption.

Regulation of macrophage polarization

Macrophages play a critical role in the spontaneous resorption of LDH. Future research should further explore the regulatory mechanisms of macrophage polarization to promote spontaneous resorption.

1) Balance of M1/M2 macrophages: M1 macrophages primarily secrete pro-inflammatory factors, promoting inflammatory responses and matrix degradation. Future research could investigate how to regulate the activity of M1 macrophages to facilitate spontaneous resorption. M2 macrophages mainly secrete anti-inflammatory factors, promoting tissue repair and absorption. Future research could explore how to modulate the activity of M2 macrophages to enhance spontaneous resorption.

2) Molecular mechanisms of macrophage polarization: Regulation of Signaling Pathways: Macrophage polarization involves multiple signaling pathways, including the PI3K/AKT and STAT pathways. Future research could investigate how to regulate these signaling pathways to promote macrophage polarization, thereby accelerating spontaneous resorption. Epigenetic regulation: Epigenetic mechanisms (such as DNA methylation and histone modification) play a significant role in macrophage polarization. Future research could explore how epigenetic regulation can be utilized to promote macrophage polarization, thus facilitating spontaneous resorption.

Regulation of matrix degradation

Matrix degradation is the final step in the spontaneous resorption of LDH. Future research should further explore the regulatory mechanisms of matrix degradation to promote spontaneous resorption.

1) Regulation of MMPs: MMP-3 and MMP-7 are key enzymes in matrix degradation. Future research could investigate how to regulate the activity of these enzymes to promote matrix degradation, thereby accelerating spontaneous resorption. In some cases, excessive matrix degradation may lead to tissue damage. Future research could explore how to balance matrix degradation by applying MMP inhibitors, thus promoting spontaneous resorption.

2) Molecular mechanisms of matrix degradation: Inflammatory mediators can stimulate the production and activity of MMPs, while growth factors can also stimulate the production and activity of MMPs. Future research could investigate how to regulate these inflammatory mediators and growth factors to promote matrix degradation, thereby accelerating spontaneous resorption.

Regulation of immune privilege

Normal disc tissue possesses immune privilege, and future research should further explore the regulatory mechanisms of immune privilege to promote spontaneous resorption.

1) Regulation of the blood-nucleus pulposus barrier: In some cases, disrupting the blood-nucleus pulposus barrier may help trigger an immune response, thus promoting spontaneous resorption. Future research could investigate how to regulate the disruption of the blood-nucleus pulposus barrier to facilitate spontaneous resorption.

2) Regulation of the Fas/FasL system: The Fas/FasL system plays a critical role in maintaining the immune privilege of disc tissue. In some cases, inhibiting the Fas/FasL system may help trigger an immune response, thereby promoting spontaneous resorption. Future research could investigate how to regulate the inhibition of the Fas/FasL system to facilitate spontaneous resorption.

Regulation of apoptosis and autophagy

Apoptosis and autophagy play important roles in the spontaneous resorption of LDH. Future research should further explore the regulatory mechanisms of these processes to promote spontaneous resorption. Although theoretically, inducing apoptosis and autophagy could promote the degradation and absorption of herniated tissue, in some cases, inhibiting apoptosis and autophagy may help protect disc tissue, thereby facilitating spontaneous resorption. Future research could explore how to regulate the inhibition of apoptosis and autophagy to promote spontaneous resorption.

Pain genetics

Research on pain genetics offers new directions for pain management. Studies have shown that polymorphisms in the OPRM1 gene are associated with pain intensity following lumbar disc herniation [75–77]. Future research should investigate the relationship between pain-related genes and resorption, with the goal of developing personalized pain management strategies [78–79].

Advances in imaging techniques

Advances in imaging technologies, such as chemical exchange saturation transfer (CEST), T1ρ imaging, T2 mapping, ultrashort echo time MRI (UTE-MRI), and sodium MRI, offer new tools for studying disc degeneration and herniation [80–82]. These techniques can reveal biochemical changes in disc tissue, aiding in the prediction of resorption potential [83].

Prediction models based on machine learning and deep learning

With the advancement of artificial intelligence technology, the application of machine learning and deep learning in the medical field has gradually increased. Machine learning is a technology that trains models using data to make predictions and decisions. Deep learning, a subset of machine learning, simulates the learning process of the human brain through multi-layer neural networks, enabling automatic feature extraction from data and complex pattern recognition. In the medical field, machine learning and deep learning are primarily used for disease diagnosis, prognosis prediction, and treatment optimization. For example, machine learning-based image analysis can assist doctors in disease diagnosis, and deep learning-based prediction models can help doctors evaluate patient prognosis.

In recent years, machine learning and deep learning-based prediction models for postoperative recovery in LDH have become a research hotspot. These models analyze patients’ clinical and imaging data to predict postoperative recovery outcomes, thereby providing a basis for clinical decision-making. Future research should further optimize these models, focusing on the following three aspects to improve prediction accuracy: 1) Optimization of data quality: The first step in building an LDH postoperative recovery prediction model is data collection. Typically, researchers collect clinical data from patients, including age, gender, body mass index (BMI), medical history, preoperative examination results, and more. Additionally, imaging data, such as the type, size, and location of disc herniation shown on MRI, are important input variables. Future research should further optimize data standardization methods to improve model prediction accuracy and expand sample sizes to enhance the model’s generalization ability.

2) Optimization of model algorithms: After data preprocessing, researchers use various machine learning and deep learning algorithms to build prediction models. Commonly used machine learning algorithms include Random Forests, Support Vector Machines (SVM), and Logistic Regression. Commonly used deep learning algorithms include Multilayer Perceptron (MLP) and Convolutional Neural Networks (CNN). Future research should explore more machine learning and deep learning algorithms and further optimize model hyperparameters to improve prediction accuracy, and apply those prediction models for personalized treatment and real-time monitoring to dynamically adjust treatment plans, thereby improving treatment outcomes.

Conclusion

Spontaneous resorption in lumbar disc herniation is a complex, multifactorial process involving mechanisms such as inflammation, neovascularization, disc dehydration, and mechanical traction. Clinical predictors of resorption include the size of the herniation, sequestration, rim enhancement on MRI, inflammatory mediators, and involvement of the posterior longitudinal ligament. Future research should focus on elucidating the molecular mechanisms of resorption, developing advanced imaging techniques to predict resorption potential, and exploring personalized treatment strategies based on machine learning and deep learning prediction models (Fig. 2).

Fig. 2.

Mechanisms, clinical predictors and future research trends of spontaneous resorption of lumbar disk herniation

Author contributions

YZ and ZJ; Conceptualization, methodology, software, formal analysis, investigation, resources, data curation. AA writing—original draft preparation, visualization, supervision, validation, AM: writing—review and editing, project administration, funding acquisition. All authors have read and agreed to the published version of the manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.