Keywords

Introduction

The spine consists of 33 vertebrae comprising the cervical, thoracic, lumbar, sacral, and coccygeal segments. Each vertebra unit has attachments to muscles and ligaments, as well as sites of articulation with adjacent vertebrae. The typical vertebra has six joints. Due to this complex network, there are several targets where regenerative medicine can prove to be an effective treatment. This chapter will highlight the current evidence for regenerative medicine to treat common spine pathology.

Epidemiology

Back pain is among the most common patient complaints. In the adult general population, the point prevalence for low back pain is believed to be approximately 12%, the one-month prevalence 23%, the one-year prevalence 38%, and the lifetime prevalence approximately 40% [1]. In regard to neck pain, it is estimated that 20% of the adult population experiences neck pain over a one-year period and around 66% experience neck pain at one point in their lives [2]. In the United States (US), low back pain is the number one cause of years lived with disability and neck pain is ranked sixth [3]. Between 2008 and 2012, a study of the Medicare database illustrated 6,206,578 patients were diagnosed with lumbar and 3,156,215 patients were diagnosed with cervical degenerative conditions [4]. It has been estimated that 10–15% of back pain becomes chronic and, in this subset, can lead to long-lasting disability. Around 80–90% of health care and social costs stemming from back pain result from this small cohort who develops chronic low back pain and disability. Just over 1% of adults in the US are permanently disabled by back pain, and another 1% are on temporary disability [5]. Those with chronic low back pain also have higher odds of unemployment [6]. With an aging population, preventative measures or treatments with the capability to reverse or halt progression are needed.

Most current treatments for back pain focus on targeting the overactive nerves responsible for the pain sensations. The most common are steroid injections and nerve ablations. In the short term many of these treatments provide significant relief; however, if the initial aggravating factors are not improved, the pain will often relapse. This leads to a population with chronic back pain. Furthermore, before the opioid crisis, many of these patients were routinely started on narcotic medications. Opioid use disorders have moved from the 11th leading cause of disability-adjusted life years in 1990 to the 7th leading cause in 2016, representing a 74.5% (95% UI, 42.8–93.9%) increase. Opioid use disorder from 1990 to 2016 went from 52nd place to 15th place on years of life lost due to premature mortality [3]. Back pain and opioid use are often linked; from one population-based survey more than 50% of opioid users reported back pain [7]. Today, the negative effects of chronic opioid use are better understood. Thus, newer treatment methods including regenerative medicine have the opportunity to provide significant relief while also proving to be safer than historical treatments. To understand the targets of regenerative medicine we will discuss the spine anatomy.

Anatomy

The spine consists of 7 cervical vertebrae , 12 thoracic vertebrae, 5 lumbar vertebrae, the sacrum (5 fused sacral vertebrae), and the coccyx (3–4 fused coccygeal vertebrae). Each vertebra unit has attachments to muscles and ligaments, as well as sites of articulation with adjacent vertebrae. The typical vertebra has two symphysis joints and four synovial joints. Each symphysis includes an intervertebral disc: one above connecting the superior vertebra and one below connecting the inferior vertebra. The synovial joints are located posteriorly and are between the articular processes. Connecting the sacrum to the lower body and distributing weight to both lower extremities is the pair of sacroiliac synovial joints. The intervertebral disc (IVD) consists of an outer annulus fibrosus and an inner nucleus pulposus. The annulus fibrosus consists of an outer ring of collagen surrounding a wider zone made of fibrocartilage and configured in a lamellar fashion. This fiber arrangement limits rotation among vertebrae. The nucleus pulposus (NP) is the center of the disc and is a gelatinous substance. It is responsible for absorbing compression forces. Between adjacent vertebra lies the intervertebral foramen through where the spinal nerve emerges from the spinal column [8].

The joints between vertebrae are supported by numerous ligaments. These ligaments reinforce the joints as they pass between vertebral bodies and interconnect structures of the vertebral arches. Along the anterior and posterior vertebral bodies lie the anterior and posterior longitudinal ligaments. The ligamenta flava connect the lamina of adjacent vertebrae. Passing along the posterior tips of the vertebra spinous processes from C7 to the sacrum is the supraspinous ligament. From C7 to the skull the ligament is known as the ligamentum nuchae as it has distinct features responsible for supporting the head. Interspinous ligaments pass between adjacent spinous processes [8].

The sacroiliac (SI) joints act to transmit forces from the lower limbs to the vertebral column. These joints are synovial joints; they lie between the L-shaped articular facets on the lateral surfaces of the sacrum and similar facets on the iliac. The joint is designed with irregular contour and interlocks to resist movement. With age the joint can become fibrous and may become completely ossified. The joint is stabilized by surrounding ligaments: the anterior sacroiliac ligament, interosseous sacroiliac ligament, and posterior sacroiliac ligament . Because this joint is weight-bearing, it is prone to degenerative changes [8].

The back musculature can be divided into the superficial group that aid in movements of the limbs, the intermediate group that may serve in respiration, and the deep group that are related to movements of the vertebral column. This deep group becomes essential for the health of the spine. Thoracolumbar fascia covers these deep muscles. The largest group of these intrinsic back muscles are the erector spinae muscles: the iliocostalis, longissimus, and spinalis. Deep to these muscles are the transversospinalis muscles which include the semispinalis, multifidus, and small rotatores muscles. The deepest group of muscles include the segmental muscles which pass between adjacent spinous processes and transverse processes. All the muscles above provide some form of stabilization or movement to the spine itself. Any derangements in these muscles, or the abdominal muscles that counteract them, can lead to abnormal function of the spine that may progress to pathology that manifests as back pain [8].

In understanding spinal pain , it is important to identify the innervated structures. These include: the vertebrae bony body (by the sinuvertebral nerve anteriorly), the zygapophyseal joint (by the medial branch of the dorsal primary ramus), the external annulus of a healthy disc (posteriorly by the sinuvertebral nerve), the anterior longitudinal ligament and anterior external annulus (by the recurrent branches of rami communicantes), posterior longitudinal ligament (by the sinuvertebral nerve), interspinous ligament (medial branch of the dorsal primary ramus), muscles (specifically the multifidus by the medial branch of the dorsal primary ramus and the paraspinal musculature by small branches of the dorsal primary ramus) and fascia (by small branches of the dorsal primary ramus) , and the nerve roots themselves. The ligamentum flavum and a healthy disc’s internal annulus fibrosus and nucleus pulposus are not innervated and therefore do not transmit pain signals. The posterior longitudinal ligament is often thought to be the cause of pain perception in disc herniation [9].

Overview of Back Pain Pathology and Treatment

As discussed above, the spine is composed of bones that are further supported by a network of ligaments and muscles. The spine’s main role is to protect the sensitive spinal canal and support the upper body. Its secondary function is to provide mobility and movement, which is allowed by the various joint spaces along each vertebra. These include the two zygapophyseal joints (facet joints) and the intervertebral disc itself with each vertebral end plates. The spine has a natural opposing curvature pattern to soften loading forces and disperse force throughout each vertebra. Any abnormality at each of the above structures can lead to inefficient working of this spine complex that often presents as pain. As instability or increased forces target one particular location of the spine, this area often begins to generate pain.

Pain from the spine is traditionally divided into axial or radicular. Axial pain is located primarily at the spine level, where radicular pain is mostly experienced in the extremities. Radicular pain results from irritation or pinching of a spinal nerve as it exits the spinal cord. Axial back pain can be further divided into discogenic, facet mediated, or stenotic. Pain located below L5 can also be caused by the sacroiliac joint. Furthermore, fractures can occur at any boney location, most commonly at the pars interarticularis [9].

Treating back pain proves to be a challenge particularly because there is often no concrete single diagnosis and there is rarely one physical exam, laboratory, or imaging test that gives a precise answer. One study looking at MRIs of asymptomatic, pain-free individuals found that 37% of those 20 years of age showed disc degeneration and >90% of those over 60 years of age had degenerative spine changes on MRI. Similarly, 4% of the asymptomatic patients 20 years of age showed facet degeneration, while 50% of those aged 60 years old did [10]. In another MRI study, disc herniations were strongly associated with low back pain; however, annular fissures, high-intensity zone lesions, Modic changes, and spondylotic defects were all not associated with low back pain severity [11]. Thus, the pain generator often may not correlate with specific findings on imaging or other studies.

Prior to trialing a regenerative intervention, it is essential to first categorize the likely etiology of the pain. While nerve blocks and ablations target the nerves causing pain directly, regenerative medicine has the additional opportunity to target the precipitating cause of the pain. By targeting the reason why the patient has pain in the first place, it is hopeful some form of regenerative medicine will eventually be able to provide a cure, or at least halt the disease progression, rather than temporarily masking the pain.

Back pain is often multifaceted and results from a combination of pathologies. To provide a theoretical framework to highlight the effectiveness of regenerative medicine, we will group similar mechanical pain and pathology into common categories. It is important to keep in mind that the patient may suffer from multiple individual pathologies which when combined together are now the cause of presentation. Similarly, once one element of the vertebral unit is affected this often places abnormal stress on the rest further causing pathology at distinct locations. In chronic low back pain, around 42% are related to the discs, 31% related to SI joint, and 18% related to zygapophyseal joint [12].

Pain Generators of the Spine

Intervertebral Disc Pain

The intervertebral disc itself is crucial to the health of the spine. Several pathologies of the disc itself, including internal disc disruption, tears in the disc, degeneration of the disc, and loss of disc height, can predispose patients to discogenic back pain. The disc acts as a shock absorber. Degeneration often correlates with loss of disc height that can lead to excess motion and unstableness throughout the other joints of the spine. While we will focus on pain related to the disc itself, damage to the disc may lead to excess forces and damage throughout the spine. In order to illustrate the action of regenerative medicine on the disc it is important to understand the disc’s histological makeup.

The disc is a central part to the complex biomechanical system of the spine, which allows for mobility and the spreading of stress. The disc is divided into four separate regions. The outer annulus is highly organized with mostly type 1 collagenous lamellae running in an alternating pattern to assist in strength. The inner annulus is larger and more fibrocartilaginous, with less collagen and lacking the lamellar structure; this collagen is mostly type II . The cells here are both fibroblasts and chondrocytes. The third layer is the transition zone made up of a thin acellular fibrous layer . The final layer, the central nucleus pulposus, is an amorphous matrix of highly hydrated proteoglycans that are embedded in now a loose network of collagen [13].

The disc itself is a sensitive environment as it is avascular at baseline. Thus, it depends on diffusion for nutrients and waste movement. This diffusion capacity is relatively poor and worsens with both age and pathology. In normal discs, nerve endings are limited to the outer one-third of the disc and are not found in the inner annulus or nucleus pulposus region [14]. In degenerated IVDs, nociceptive nerve fibers along with vasculature migrate into the central disc regions [15]. It is theorized that neurotransmitters together with changes within the extracellular matrix itself and the release of cytokines regulate this nerve ingrowth to the IVD. In addition, pain-related peptides and proinflammatory cytokines are increased.

Disc failure can be a result of overloading. Forces may lead to desiccation of the disc and annular tears. The disc itself has a limited capacity for compression and this capacity decreases with decreasing water content—as fluid is not compressible. To improve disc failure , the treatment goal is to regain disc height to reduce the axial nerve compression and to restore the tissue dynamics (fluid content) of the annulus. Secondly, the goal is to reconstitute the central nucleus with a matrix environment that can hold water and improve nutritional flow. A prosthetic disc nucleus has been designed to restore the disc height. However, it fails to fully simulate the compressibility and plasticity of the original disc. Furthermore, this implantation requires a fairly invasive procedure. Using regenerative medicine techniques, the hopes are to “regenerate” the nucleus and disc by injecting the nucleus with a complement of its original cells. In theory, these cells will reconstitute a matrix that will have the capacity to change the damaged internal environment of the disc and eventually reorganize to improve and return disc function [16].

As discussed, the IVD is composed of an interconnected unit of tissues that work together: the nucleus pulposus, annulus fibrosus supporting the nucleus pulposus, and the cartilaginous end plates that connect these tissues to the vertebral bodies providing nutrients. Thus, either of these can be targeted for potential regenerative medicine. Depending on degree of degeneration, different strategies are proposed. In early degeneration, biomolecular treatment strategies (including platelet-rich plasma (PRP), prolotherapy, and hyaluronic acid) are often considered to best support the viable cells remaining in hopes of reverting or halting progression of disease. In intermediate degeneration, cell-based therapies (articular chondrocytes, nucleus pulposus, disc chondrocytes, and stem cells) are required as the numbers of viable cells are decreased. At advanced degeneration, tissue engineering may be required as there is now structural damage and the number of viable cells is severely limited. The literature review by Moriguchi et al. suggests that protein injections are limited due to their relatively short life span. Gene therapy , which involves delivering certain genes through viral or nonviral vectors, has a promising future as it is able to induce modification of the intradiscal expression of genes for a long-term effect. Furthermore, tissue engineering advancements allow for the development of biocompatible and biomimetic scaffolding material to recover extensive loss of matrix cells and structural environment [17]. A meta-analysis by Wu et al. suggests that mesenchymal stem cell (MSC) and chondrocyte therapy for discogenic low back pain correlates with improved pain relief and function metrics. Currently, the authors conclude that there lacks an optimal cell therapy protocol. At this time, cell therapy is not considered a standard treatment; however, it has the potential to be a consideration especially in patients that have not adequately responded to nonoperative management [18].

Much of regenerative medicine studies regarding the spine focus on the intervertebral disc itself. An ex vivo experiment by Pirvu et al. on bovine annular fibrosus cells shows that platelet-rich preparations increased the matrix production and cell number after their injection into an annular fibrosus defect [19]. Another ex vivo study by Kim et al. looked at nucleus pulposus cells from human discs that were cultured in a collagen matrix. PRP administration markedly suppressed cytokine-induced pro-inflammatory degrading enzymes and mediators in the NP cell. As per the authors, it stabilized NP cell differentiation through rescued gene expression concerning matrix synthesis [20]. An additional study by Akeda et al. looked at in vitro porcine IVD cells post-PRP exposure. They concluded PRP had a mild stimulatory effect on cell proliferation. There was a significant upregulation of proteoglycans and collagen synthesis and proteoglycan accumulation compared to platelet-poor plasma [21]. These studies provide data that suggests PRP supports a regenerative-like environment at the cellular level.

One animal study by Wang et al. looked at rabbits that underwent annular needle puncture to simulate early degenerative discs. The rabbits were then injected with BMSCs and PRP, just PRP, and a control group of phosphate-buffered saline. At 8 weeks postinjection, PRP-containing bone marrow-derived mesenchymal stem cells (BMSCs) were more effective than PRP alone as evidenced by an increase in signal intensity over time, and under histological staining the extracellular matrix and cell density as well as type II collagen staining were preserved. Several other animal studies show promising results [22]. A study on platelets and BMSCs by Xu et al. demonstrated effective repair of annulus defects [23]. Another study by Hou et al. on PRP intradiscal injections post needle puncture demonstrated significant recovery of MRI signal intensity. They suggest PRP can enhance the nucleus pulposus cell’s proliferation and anabolic pathway while slowing IVD degeneration in rabbits [24]. A randomized controlled trial (RCT) by Gui et al. investigated intradiscal PRP in rabbits post annulus fibrosus puncture. In the control groups, there were significant IVD height changes compared to the slight decrease in the PRP-treated group [25]. Gullung et al. looked at six rats each in a control, sham, PRP immediately post disc injury with needle, and PRP 2 weeks after disc injury. The PRP groups had fibers that were damaged with empty spaces and inflammatory cells; however overall there was maintenance of the ring structure and the nucleus appeared to keep a healthy central portion. They conclude that immediate injection has a more pronounced effect as the disc height and fluid content on MRI was significantly better in the immediate injection group compared to the sham group at 4 weeks. This study suggests that there may be a time component to treatment effects with regenerative medicine. As most patients often receive injections years after initial injury, this beneficial effect may have limited value in the clinical population [26].

While several of these above studies were RCTs and there seems to be some scientific agreement in favor of regenerative medicine, it is difficult to adopt these animal study results into clinical practice. It is important to highlight that these animals do not reflect the same stressors and pathology that is evident in patients with chronic degenerative disc disease (DDD). Often, the animals are relatively healthy and undergo a single acute stressor event to create disc pathology. Hence the environment of the disc may be more salvageable compared to that of a classic patient presenting with chronic degenerative disc disease.

Focusing on selected human trials, Miller et al. analyzed 76 consecutive patients that received intradiscal prolotherapy who suffered from internal lumbar disc derangements. These patients had undergone two epidural steroid injections (ESIs) 2 weeks apart and had initial good relief followed with later a return of symptoms. Post prolotherapy, 43.4% of patients had at least 20% reduction in pain scores and pain relief was maintained at an average of 18 months. This study provides some support in favor of intradiscal prolotherapy to treat internal disc disease [27].

Focusing on PRP, Akeda et al. led a prospective clinical study evaluating intradiscal PRP releasate in 14 patients who had positive diagnostic discography. More than 50% reduction in low back pain was seen in 71% of patients within 4 weeks of injection, and this relief was generally maintained throughout the 48-week study period. Furthermore, the authors conclude there was no change in T2 imaging and disc height which suggests no negative effects on the disc matrix [28]. In a similar prospective trial, Levi et al. injected 22 subjects intradiscally with PRP and at 2 months 41% had a successful outcome of greater than 50% decrease in visual analogue scale (VAS). At 6 months, 63% had a VAS improvement at least 20 mm [29]. An RCT by Tuakli-Wosornu et al. randomized 29 patients to intradiscal PRP and 18 to the control group receiving intradiscal contrast only. Over 8-week follow-up there were statistically significant improvements in patients who received the intradiscal PRP in pain scale, function, and patient satisfaction compared to controls. Furthermore, those who received PRP were able to maintain significant improvements in the Functional Rating Index (FRI) for at least 1-year follow-up [30].

Other studies examined PRP in patients undergoing spine surgery. Sys et al. randomized 18 patients to undergo spinal fusion with PRP-soaked autologous bone to fill the cages and 18 patients to serve as a control without the PRP soaking. The added PRP in posterior lumbar interbody fusion did not lead to a substantial improvement or deterioration when compared with autologous bone only. The PRP and autologous group trended toward improvements in VAS and Oswestry Disability Index (ODI) . This study points out that there may be little improvement in using PRP during spinal fusions and it can be justified from a clinical and radiological point of view; however it may not be efficient from an economical perspective [31].

Overall PRP is generally safe and has few documented adverse effects besides from local pain temporarily post procedure. Furthermore, as patient’s own blood is used this limits risks of infection and rejection. One downside to PRP is that it only provides the IVD with certain factors that may aid repair. However, if the disc cells are already severely damaged, some suggest that no amount of PRP may make a difference [32]. Thus, PRP may be more efficient when applied at an earlier stage of degeneration, in a patient that has relatively a healthy amount of functioning cells. This is where stem cells theoretically may have the advantage as they may act to replace severely degenerated cells.

Investigating stem cells, Orozco et al. reported on a case series of ten patients with chronic discogenic back pain that were treated with autologous culture-expanded bone marrow MSCs injected into the nucleus pulposus area. This study showed strong safety and feasibility. Patients exhibited improvements in both pain and disability measures. Disc height was not recovered, but water content was significantly improved on MRI [33]. Similarly, Elahd et al. studied five patients with DDD post intradiscal injection of autologous, hypoxic cultured, BMSCs. Post 4–6 years, no adverse events were reported. All patients self-reported overall improvement [34]. Pettine et al. analyzed 26 patients that were candidates for spinal fusion or total disc replacement surgery. Instead, they underwent autologous bone marrow concentrate intradiscal injection into the nucleus pulposus. After 36 months, only six patients progressed to surgery. The remaining 20 other patients reported improvements in ODI and VAS. One year MRI showed that 40% of the subjects improved one modified Pfirrmann grade and no patients worsened. Those with greater concentration of stem cells had better outcomes [35]. Furthermore, Centeno et al. studied autologous MSCs in 37 patients with DDD with secondary radicular pain. The treatments in this study included a preinjection 2 weeks before MSC injection that included a platelet lysate transforaminal epidural injection . Then, MSCs that had been cultured in platelet lysate were injected intradiscal. Two weeks later a second transforaminal epidural with platelet lysate injection was performed. At all-time points from 3 months to 24 months there was significant improvement in pain scores. FRI was statistically improved at all-time points except at 12 months. Twenty patients underwent posttreatment MRI and 85% showed reduction in disc bulge size. Due to a lack of control group, this study is limited in determining the efficacy of these interventions as it is well known that patients often improve with time regardless of treatment [36]. Although the above studies are promising, further RCTs are needed before recommendations can be made.

In an RCT by Noriega et al. 24 patients were randomized so that 12 patients received intradiscal allogeneic (from someone else) MSCs and 12 patients received sham paravertebral musculature local anesthetic treatment. This study demonstrated stem cell-treated subjects had significant improvements in algofunctional indices. However, the improvement seemed to be restricted to a group of responders representing 40% of the cohort. Degeneration graded by MRI and Pfirrmann grading improved in those treated with stem cells and worsened in the controls. This study supports the utilization of allogeneic stem cells which are more convenient than the autologous MSC treatment that must be harvested from the patient [37]. A second RCT by Bae et al. randomized a total of 100 patients to intradiscal injection treatment: 20 patients received hyaluronic acid, 30 patients received allogeneic mesenchymal precursor cells (MPCs) at 6 million dose, 30 patients received allogeneic MPCs at 18 million dose, and 20 received saline to serve as a control. The authors concluded that allogeneic MPCs showed improvements in pain and function and reduced interventions compared to the control group. However, when comparing the stem cell group to the hyaluronic acid group, the results did not reach statistical significance [38].

Secondary to the myriad of components and mixtures that can be utilized in regenerative medicine, studies comparing different mixtures or recipes of injectate present a challenge. Mochida et al. studied mixing autologous NP cells with BMSCs. Nine patients scheduled for fusion underwent harvesting of NP cells. Viable NP cells were co-cultured in direct contact with autologous BMSCs. At 1-week post fusion they underwent transplantation at adjacent levels to the fusion of the now activated NP cells. Imaging revealed improvement in one case, and functional improvement overall was minimal [39]. Studies are needed to determine the best dosage, combination, and type of injectate.

It has been shown that stem cells can also be derived from adipose cells. Kumar et al. looked at adipose tissue-derived mesenchymal stem cells combined with hyaluronic acid in ten patients who had discogenic pain with positive discography. There were no serious adverse effects at 1 year and these patients had significant improvements in VAS, ODI, and Short Form-36 (SF-36). Three of the ten patients were determined to have increased water content in their discs as determined by MRI [40]. Similarly, Comella et al. analyzed 15 patients that underwent adipose tissue-derived stromal vascular fraction (SVF) injection directly into the nucleus pulposus. At 6 months there were no serious adverse effects, and patients improved in flexion and VAS and SF-12. ODI and Dallas pain questionnaire only showed positive trends [41].

Others have utilized stem cells from the umbilical cord. Pang et al. looked at two patients with chronic discogenic pain that were treated with human umbilical cord tissue-derived MSCs. In the two patients, pain and function improved and was maintained for a 2-year follow-up. Furthermore, the water content in the degenerative disc of one patient was found to have significant improvement. This method avoids the invasive procedures required in harvesting stem cells [42]. A separate class of stem cells include hematopoietic stem cells that have the capability to give rise to other blood cells. Huafe et al. looked at ten patients with positive discograms that received intradiscal injection of hematopoietic precursor stem cells obtained from their pelvic bone marrow. Zero patients reported improvement. This study suggests that while there may be benefit for MSCs, HSCs do not appear to have similar efficacy. The authors suggest that perhaps the HSCs are unable to survive the oxygen-poor environment of the inner disc [43]. More studies are necessary to determine which types of stem cells, if any, have the best efficacy for each diagnosis.

Fibrin is another injectate that has been trialed to help those with discogenic pain. Yin et al. reported on 15 adults with confirmed discogenic pain that underwent intradiscal injection of a fibrin sealant. Eighty-seven percent of the subjects achieved at least a 30% reduction in low back VAS compared with baseline at the 26-week primary end point. Although this was not an RCT and only evaluated 15 patients, fibrin may provide benefits in certain patients. Fibrin is composed of purified prothrombin and fibrinogen and reconstituted with aprotinin and calcium. When injected into the annular tears, it has the ability to form a matrix sealant protecting the nucleus pulposus [44].

Intradiscal methylene blue (MB) has also been trialed for patients with discogenic pain. Peng et al. looked at 72 subjects equally randomized to either the MB injection or the control group that received isotonic saline instead. In the MB group, there was a mean reduction in the numeric rating scale (NRS) of 52.4 and ODI by 35.58 and 91.6% patient satisfaction at 24 months. This was a significant improvement over the control group [45]. Once again further studies are needed to replicate these strong findings.

Others have suggested that regeneration of the disc should not be the primary goal when treating these patients with back pain. Adams et al. suggest that we should separate our focus among healing a painful disc and reversing disc degeneration, as these may be two distinct pathways. Discs are often the cause of pain as nerves in the peripheral annulus or vertebral endplate become affected by inflammation and/or radial tears. Adams et al. conclude we should primarily focus on this peripheral region which has the cell density and metabolite transport to improve, rather than the more difficult notion of regenerating the nucleus pulposus. Regardless of the degenerative changes in the nucleus, promoting healing at the periphery can provide significant pain relief. Physical therapy , which employs mechanical loading, can act as a healing stimulus in the peripheral disc. For radial fissure, the authors recommend initial controlled mobilization toward the direction that decreases pain; then after scar formation, stretching should be directed toward the painful direction in hopes of promoting remodeling. In the case of an endplate fracture, initial therapy would include unloading followed by progressive loading and if needed intermittent traction [46].

Although the above studies (Table 7.1) are promising treatments for IVD pain, there is a general lack of comparable RCTs leading to poor overall evidence level. In examining these studies, it is crucial to acknowledge the diagnosis being treated and the precise injectate utilized. Before recommendations for treatment can be more RCTs are needed to support evidenced-based medicine.

Table 7.1 Regenerative medicine for intervertebral disk pain studies reviewed
Full size table

Radiculopathy

Moving from pain resulting from the disc itself, a second generator of pain is caused by irritation or pressure on the nerve root creating a radiculopathy . This can be caused by a bulging disc, herniated disc, and/or stenosis. The classic pain felt is in the distribution of the sensory nerve root. For instance, in the lower back this shooting, electrical type of pain will be reported to be traveling down the lower extremities. Depending on which nerve root is involved, the pain often localizes to a certain extremity or dermatome. Affected cervical nerve roots often will transmit pain down the ipsilateral arm, while affected lumbar nerve roots will have symptoms from the waist down.

For acute radicular pain the routine care is commonly epidural or transforaminal steroid injection. However, although they show some short-term pain relief, they have increasingly been criticized for failure to provide lasting benefit while exposing the patient to potential risks and side effects. To better improve outcomes several clinicians have investigated the efficacy of PRP, dextrose or prolotherapy, and stem cells.

In a pilot study by Bhatia et al. ten patients with prolapsed IVD were injected with 5 ml of PRP with an interlaminar approach into the area of affected nerve root. A significant number of patients showed relief and sustained relief at 3 months. The authors conclude that PRP can be used in replace of steroids; however, a randomly controlled trial comparing the two is needed [47]. In 2017, Cameron et al. reported on PRP injections in 88 total subjects: 38 for cervical, 38 for lumbar, and 12 for both cervical and lumbar disc herniation. PRP was injected in a circumferential manner of the affected area into the lateral masses, facet joints, lateral gutters, and inter- and supraspinatus ligaments, Kambin’s triangle, and spinous process. This prospective nonrandomized clinical study suggested each group of patients showed a significant improvement in pain scores [48].

Similarly, Centeno et al. analyzed a case series of 470 patients who were treated with platelet lysate and nanogram dose hydrocortisone. As per the authors, the nanogram amount of steroid used in the formation of platelet lysate is one million times less than those used in regular epidural steroid injections. At this low level, the steroid provides an anti-inflammatory effect similar to that of endogenous glucocorticoids. The patients showed significant improvements in both their numerical and functional scores. At 24 months posttreatment, patients had a 49.7% rating for their own improvement. Although this was a large study, it lacks both randomization and a control group [49].

In terms of prolotherapy and dextrose injections, Maniquis-Smigel et al. conducted an RCT that looked at 19 patients who received epidural injections of 5% dextrose and 16 who received normal saline into the caudal epidural space. Subjects who received the dextrose reported greater significant pain relief at 15 minutes and up to 48 hours, but not at 2 weeks. Although demonstrating short-term efficacy, this study suggests that dextrose may have positive results and a long-term study should investigate the effects of serial dextrose epidural injections and prolotherapy [50].

Focusing on stem cells, one RCT trial, by Bertagnoli et al., investigated the use of autologous disc-derived chondrocyte transplant in patients undergoing sequestrectomy. Only the interim analysis has been published which looked at 26 patients in each the treated and control group. The results are promising as the control group showed decreases in disc height, while the treated group did not have any cases of disc height loss. Furthermore, the discs treated with chondrocyte cells had adjacent intervertebral discs segments that appeared to retain hydration when compared to the control group. This study suggests that the autologous chondrocyte cells seem to improve disc structure and may even have beneficial effects on neighboring discs. Because the population of this study was only those undergoing discectomy, it is difficult to generalize these findings to the general patient with radicular back pain [51].

While regenerative therapy targeting herniated discs and radiculopathy seem to show promising results (Table 7.2), more long-term RCTs are needed before a general recommendation can be formulated.

Table 7.2 Regenerative medicine for radiculopathy studies reviewed
Full size table

Zygapophyseal Joint (Facet) Arthropathy

The current standard of care for facet-mediated pain includes directly targeting the medial branch nerve that is responsible for the innervation of this joint. This can be done with local anesthetic, steroids, and/or ablation. For each of these above procedures, often the pain returns as the medication wears off or the nerve heals. Furthermore, the root cause of the pain is not addressed. As these facet joints are synovial joints, therapies that have had success in other joints in the extremities have been further investigated. Treatments trialed include PRP, prolotherapy, and viscosupplementation.

In 2016, Wu et al. published on a new technique to treat lumbar facet pain using intra-articular injection with autologous PRP. Nineteen patients had good pain relief outcomes up to 3 months postinjection [52]. Wu later reported a prospective, randomized, controlled study of 46 subjects diagnosed with facet joint arthropathy through positive successful lidocaine blocks. Twenty three subjects underwent PRP injection and 23 underwent standard of care lidocaine with steroid. At 1 week and 1 month the steroid group outperformed the PRP group. However, as time progressed the PRP group began to outperform the steroid group with significantly improved VAS scores from 2 months on. The steroid group peaked around 1 month and relief diminished to 6 months. The PRP group seemed to improve up to 3 months and then plateaued [53]. There seems to be promising evidence for the use of PRP to treat facet-mediated pain.

Prolotherapy has also been studied to treat facet pain. Hooper et al. reported a retrospective case review of 15 patients (three patients treated bilaterally to make 18 total facet joint sides) who were treated with intra-articular prolotherapy after confirmation of cervical facet pain post whiplash injury. This procedure significantly improved the mean neck disability index at months 2, 6, and 12. These results are promising; however this study lacks a control and furthermore may have been confounded as 13 of the patients’ pain was caused by motor vehicle accidents in which they were in litigation. Furthermore, patients had concurrent physiotherapy which may have supported better outcomes [54]. Hooper et al. later reported a retrospective series on 177 patients with chronic spinal pain who each received prolotherapy to the facet capsules of the cervical, thoracic, and lumbar spine in regions that correlated with pain (in addition, the iliolumbar and dorsal sacroiliac ligaments were injected in patients with low back pain). Ninety-one percent of these patients reported reduction in pain. Lumbar and thoracic patients proved to have greater significant relief than compared to cervical [55]. These studies are in favor of prolotherapy for facet-mediated pain.

In terms of viscosupplementation, a pilot prospective study by Cleary et al. recruited 13 patients with symptomatic lumbar-facet joint pain who were treated with injection of hyaluronic acid: 18 facets of the 13 patients were injected. At 6 weeks there was no significant improvement in pain scoring. This study was limited as there was no definitive diagnostic testing for facet arthropathy [56]. A more promising study by DePalma et al. followed 15 patients with identified facet joint pain through successful trial of comparison local blocks. In this prospective uncontrolled pilot study, patients had positive results with significant improvements in VAS and ODI up to 6 months; however results were not sustained at 12 months. However, this study is flawed by its lack of control and blinding [57]. Fuchs et al. followed two groups with axial back pain: one received intra-articular sodium hyaluronate and the control received intra-articular glucocorticoids targeting the facet joints. In this observer-blinded RCT, both groups had positive results, with the hyaluronate group showing prolonged benefits in the long term at 3 and 6 months [58]. An RCT, by Annaswamy et al. investigated 30 subjects with facet pain and injected them either with hyaluronate or with steroid. While the steroid group only providing short-term functional improvement, the hyaluronate group outperformed by providing both short-term and long-term functional improvement, as well as short-term pain relief [59].

In conclusion, for PRP we identified one RCT that suggests it outperforms steroids with its longevity lasting up to 6 months. For prolotherapy, the studies seem to show improvements in pain; however an RCT is lacking. Lastly, the two viscosupplementation RCTs show hyaluronic acid to improve pain up to 6 months. The two other studies showed mixed results, with one trial confirming the positive results and the other showing no significant improvement in pain scoring. Further studies are needed that include a control group and stricter inclusion criteria confirming facet-mediated pain. Each study had a strong safety profile, suggesting these interventions (Table 7.3) can be trialed when evidence-based medicine fails to provide appropriate relief.

Table 7.3 Regenerative medicine for zygapophyseal joint (facet) arthropathy studies reviewed
Full size table

Sacroiliac Joint Dysfunction

Another common cause of chronic low back pain stems from the sacroiliac joints and ligaments. The SI joint acts to transmit forces from the lower limbs to the vertebral column. These synovial joints are prone to degeneration and instability. In addition, other joints of the pelvis including the sacrococcygeal joint can be a source of pain.

One case series investigated the efficacy of viscosupplementation. Srejic et al. reported on four patients treated with viscosupplementation to the SI joint. At 12–16 weeks postinjection pain was reported as 40–67% improved. The authors conclude further studies are needed to look at long-term duration and overall outcomes [60].

Others have examined the effects of prolotherapy theorized to provide stabilization of the painful instable SI joint. In a retrospective cohort study by Hoffman et al. 103 patients received prolotherapy aiming at the SI joint for a total of three injections at approximately 1 month intervals. At an average of 117 day follow-up, 23% of these patients showed a minimum clinically important improvement in ODI. Many of the responders had a median of 2 years of back pain. This suggests prolotherapy could be beneficial in a subset of patients [61]. Similarly, Mitchell et al. reported on prolotherapy in 131 patients injected around the SI joint into the deep interosseous ligament. Over 70% of patients were satisfied with the procedure. The majority of patients demonstrated at least 50% improvement in pelvic/lumbar strength. Two-thirds of patients demonstrated some pain relief with a mean of 51.6% reduction at 12 months [62]. Kim et al. investigated the current routine treatment of steroid injection and compared that to prolotherapy injection to the SI joint. Both groups (23 patients received prolotherapy and 25 patients received steroid) showed similar significant pain relief results at the 2-week follow-up. However, at 15 months the cumulative incidence of greater than 50% pain relief was 58.7% in the prolotherapy group while just 10.2% in the steroid group. This study suggests prolotherapy may have more long-term efficacy compared to steroids [63].

Examining other joints of the pelvis, Khan et al. studied patients with chronic coccygodynia and performed two injections of prolotherapy 15 days apart to the sacrococcygeal joints. Due to the good relief obtained, this prospective observational study recommends that dextrose prolotherapy should be trialed in patients with chronic, recalcitrant coccygodynia prior to undergoing coccygectomy [64].

Focusing on PRP, a case series by Ko et al. reported on four patients who had two sessions of PRP injections to the SI Joint at the three Hackett’s points at the ligament-bone interface. Each of these patients showed significant reduction in pain and improvements in quality of life [65]. A prospective randomized study by Singla et al. treated 20 patients with steroid and 20 patients with PRP to the SI joint. At 6 weeks and 3 months the PRP group had significantly better intensity of pain. The efficacy of steroid injection was reduced to only 25% at 3 months, while efficacy remained at 90% in the PRP group [66]. This study, similar to the prolotherapy study, demonstrates a longer efficacy of the regenerative medicine (PRP group) compared to steroids.

To examine the efficacy of PRP versus prolotherapy , Saunders et al. compared his prospective trial of 45 patients with PRP injection into and around the dorsal interosseous ligament to a control of a prior separate study using prolotherapy to treat presumed SI joint pain. At 3 months the PRP group had good pain and functional improvement without further improvement at 12 months. When this trial was statistically compared to a prior prolotherapy study, the PRP group had better outcomes in pain scores and function and required on average 1.6 injections compared to the three injections of the prolotherapy control group [67].

Viscosupplementation, prolotherapy, and PRP have an excellent safety profile and have shown promising results in treating SI joint pain (Table 7.4). Patient selection, injection target, and injection schedule remain significant variables lacking a gold standard. As previously noted, more well-designed comparative studies are necessary.

Table 7.4 Regenerative medicine for sacroiliac joint dysfunction studies reviewed
Full size table

Back Musculature Atrophy

Pain in the back can also be related to musculature dysfunction. There are several muscles of the back that attach to the spine and act to add strength and support, often stabilizing the spine joints through various movements. When there is a misbalance, pain can result from poor mechanics and additional destructive forces. Furthermore, the general physiological response to back pain is for the muscles to disengage as they inactivate. This leads to atrophy of muscles over time, which further promotes a negative cycle.

A study by Hussein et al. analyzed 104 patients with chronic nonspecific back pain and confirmed muscle atrophy on MRI. These patients were treated with platelet leukocyte-rich plasma (PLRP) into the lumbar multifidus (LMF) muscle weekly for 6 weeks. Patients improved in pain and function as reported on questionnaires. Furthermore, 12-month MRI follow-up showed increased cross-sectional area and decreased fatty degeneration of LMF muscle. This study suggests that PLRP may better pain and function outcomes by improving LMF atrophy. One limitation includes the lack of a control group and the fact the patients were advised to remain active and walk 30 minutes per day. Thus, physical therapy targeting these muscle groups may play an important part in relieving back pain, whether in conjunction with regenerative therapy or on its own. Although this technique had promising outcomes (Table 7.5), again there is a need for RCTs [68].

Table 7.5 Regenerative medicine for back musculature atrophy studies reviewed
Full size table

Back Ligament Dysfunction

Similar to the muscles of the back, the ligaments act to support the spine and its joints. Ligaments can be visualized as guy wires providing strength, reinforcement, and stability. Due to this important role, ligaments can be another target for regenerative therapy. Historically, prolotherapy has been used in theory to strengthen ligaments.

Dechow et al. reported a randomized, double-blind, placebo-controlled trial of 74 mechanical back pain patients, with 36 undergoing three once weekly dextrose injections and 38 in the control group receiving normal saline. Sites injected included the tip of the spinous process of L4 and L5 and associated supraspinous and interspinous ligaments, apophyseal joint capsules at L4–5 and L5–S1, attachment of the iliolumbar ligaments at the transverse processes of L5, attachment of the iliolumbar and dorsolumbar fascia to the iliac crest, and attachments of the long and short fibers of the posterior sacroiliac ligaments and the sacral and iliac attachments of the interosseous sacroiliac ligaments. The authors’ findings showed no statistically significant differences between the control and the prolotherapy group. The authors acknowledge that their inclusion criteria did not evaluate for instability and hence the treatment sample group may not have been the ideal patient cohort that could potentially benefit from prolotherapy [69]. A retrospective case study published by Hauser et al. analyzed 140 patients that received prolotherapy to sites that included the sacroiliac, iliolumbar, sacrotuberous, lumbosacral, supraspinous and interspinous, sacrococcygeal, and sacrospinous ligaments, as well as the gluteal and pyriformis muscle attachments on the iliac crest. On an average of 12-month follow-up, 89% of these patients demonstrated more than 50% pain relief with prolotherapy. Again, this study lacks both a control and blinding [70].

Klein et al. randomized 39 chronic low back pain patients to a xylocaine/proliferant group and 40 to a xylocaine/saline (control group) that received injections into the posterior sacroiliac and interosseous ligaments, iliolumbar ligaments, and dorsolumbar fascia. Although both groups improved, the proliferant (prolotherapy) group showed a statistically significant improvement in number of patients that achieved a 50% or greater diminution in pain or disability scores at 6-month follow-up [71]. Similarly, an RCT by Yelland et al. treated 110 patients with either prolotherapy or normal saline injections into tender lumbo-pelvic ligaments and was then randomized to either flexion/extension exercises or normal activity over 6 months. Although each ligament injection group showed improvement and sustained reductions in pain and disability, no significant attributable difference was seen among the prolotherapy group. This suggests that any needling of these ligaments may provide relief; a different control group that did not receive injectate could better identify these findings [72].

These studies (Table 7.6) show promise that prolotherapy can help in cases of ligament dysfunction. Furthermore, they highlight the importance of identifying the disease pathology one is attempting to treat. For instance, a treatment might fail in an individual patient and have success in another based on the pathology of the patient and the target of the injectate. More RCTs would hopefully define both the optimal patient selection criteria and the ideal injectate target.

Table 7.6 Regenerative medicine for back ligament dysfunction studies reviewed
Full size table

Overall Levels of Evidence

When reviewing the available literature, it is important to objectively determine the evidence level that can be drawn from the authors’ conclusions. It is crucial to define this evidence level prior to adopting the study outcomes into clinical practice. In order to systematically grade the evidence, Manchikanti et al. have developed an interventional specific pain management instrument used in assessing the methodological quality of trials. Randomized controlled trials are often considered the gold standard and superior evidence compared to studies without randomization and/or without controls. Case reports and observational clinical experiences or reports of expert committees are determined to be the lowest level of evidence. This qualified modified approach (Table 7.7) to grading allows us to define the level of evidence for a specific treatment [73]. However, it remains important to remember the specific patient population treated, the exact injectate utilized, and the overall magnitude of results achieved by each study. By utilizing this qualified modified approach to grading, we are better able to categorize the evidence.

Table 7.7 The qualified modified instrument developed by Manchikanti et al. in assessing the methodological quality of trials and grading the overall scientific evidence in interventional specific pain management [73]
Full size table

Limitations of Current Studies and Future Implications

There are several common pitfalls when analyzing regenerative medicine trials. In terms of injectate (whether stem cells, PRP, viscosupplementation, or prolotherapy), there is often not a defined common mixture or recipe. With each injectate the exact effective dose is vital, and each clinician should strive to achieve what is considered the gold standard formula. Further studies should strictly define and list the active dose of the injectate utilized. Although in certain cases the injectate cannot be standardized, for instance, where the injectate is partially derived from the patient (PRP, stem cells), a dose-response relationship should be developed. Furthermore, a combination of regenerative medicine substances should be trialed for greatest benefit. With many pain trials, a control group is essential as often seen some pain may heal with time independent of treatment. In addition, the construct of a control or sham is significant as any injectate or needling may provide some hidden benefit on its own.

The selection criteria in spine pain studies are essential to ensure that the patient has the pathology the clinician is attempting to treat. Poor selection can lead to poor results, as some patients may not have the specific disease that the intervention was designed to treat. Furthermore, injection technique is crucial to ensure the injectate reaches the precise target. Attention must be placed on the scales used to measure a positive result. Regarding pain, function, and quality of life scales it is important that a statistically significant difference makes for a clinical impact on the patient. Lastly, most studies focus on the lumbar spine likely secondary to a lower perceived risk. Although these results have the possibility to be generalized, more data and controlled studies are needed for the thoracic and cervical spine to demonstrate efficacy as well as define a risk profile.

Overall regenerative medicine for the spine appears relatively safe with few side effects or adverse reactions reported from the injectate alone. Prolotherapy, viscosupplementation, and PRP can be used with few risks. Compared to current routine treatments of local anesthetics and steroids, these regenerative treatments may have a superior safety profile with most adverse events coming from the injection technique itself. For stem cells, the safety data is also strong, but longer time frame studies are needed. Most clinical studies have not followed patients for enough time to evaluate long-term safety prognosis. One case report describes a 66-year-old male that, in hopes of curing his deficits from an ischemic stroke, traveled to three separate countries for infusions of mesenchymal, embryonic, and fetal neural stem cells into his spine. He was later found to have a spinal tumor that resulted from the intrathecally introduced exogenous stem cells [74]. Although the pluripotent stem cells this patient received are of a different cell type than those used primarily in regenerative medicine for joint and spine pain, this exceptionally rare and unfortunate case serves as a cautionary tale for possible unforeseen risks.

Regulation Concerns

Currently much of the field of regenerative medicine is regulated in the United States under section 361 by the Federal Drug Administration (FDA). Section 361 of the Public Health Service Act gives the FDA authority to make and enforce regulations if the substance meets certain specifications, is minimally manipulated, is intended for homologous use, is for autologous use, and does not involve combination of the cells or tissues with another article except for water, crystalloids, or a sterilizing, preserving, or storage. Products that do not meet these criteria fall under Section 351 and these products are to be regulated as biologicals. This includes any virus, serum, antitoxin, vaccine, blood component or derivative, allergenic product, or analogous product that is used in the medical care of patients. 351 products are either more than minimally manipulated or used in a nonhomologous manner (different from original function). To summarize, 351 products are defined as a biologic drug and require complete FDA review, including premarket approval and clearance before the biologic drug can be legally marketed. Thus, the average time to market for substances labeled under Section 351 is around 10 years costing millions of dollars similar to the requirements of more traditional chemical drug products. Substances labeled as 351 products fall under the higher regulation and therefore may currently be unattainable for routine clinical use. To the contrary, most of the regenerative medicine substances discussed fall under Section 361 making them easily accessible. One of the substances in a gray area is adipose stem cells. After harvesting the adipose stem cells, preparation requires enzymatic dissociation of the tissue. This would suggest more than minimal manipulation and classify these adipose stem cells as a biologic drug under Section 351 [75]. With advancements in regenerative medicine, it is crucial to understand and follow the evolving regulations set forth.

Future Directions

New technology and advances should allow for better efficacy of regenerative medicine. Currently 3D printing utilizing bio-ink materials creates the ability to provide an optimal artificial extracellular environment to cells which allows for ideal adhesion, proliferation, and differentiation. Cells can now be encapsulated with this 3D printed structure with high viability. Important cell building blocks can be incorporated into this matrix [76]. Furthermore, studies have analyzed the use of exosomes which are extracellular vesicles that carry microRNA, proteins, and other molecules that work to mediate biologic function through gene regulation and intercellular communication. MSC exosomes have in theory the ability to mediate functional recovery by upregulating and promoting repair utilizing the patient’s own intact cells [77].

Conclusion

Traditional pain management treatments target individual pain generators with the main goal of eliminating or masking pain through interrupting the transmission of painful signals. In this approach it is essential to isolate the activated nerves. Similarly, due to this complex network of possible pain generators, there are several targets where regenerative medicine can prove to be an effective treatment. With regenerative medicine, the goal is to create a favorable environment in which the body can jumpstart the healing cascade, promote repair, and hence revitalize itself. Consequently, it may be important to view the entire spine in a holistic approach. Individually deactivating painful nerves may control pain signals, but will not improve the original cause of the pain. Idealistically, through regenerative medicine, the etiology of the pain can be corrected, and pain relief will naturally follow and sustain.

Furthermore, we must use caution with the term regenerative medicine, as the term “regenerative” may not apply to all treatments and may not accurately portray the actual science on a cellular level. While pain and function may improve, this does not necessarily prove that anything has indeed been physically “regenerated.” The pain signal may resolve, but the fundamental pathology itself may remain.

As many of these interventions remain at the investigational level, more quality long-term, randomized, controlled human trials are required if these promising treatments are to become evidence-based medicine and the standard of care in everyday clinical practice. While there is a substantial amount of data on lumbar spine utilization, there is a paucity of studies analyzing intervention at the thoracic and cervical level. As regenerative medicine is a relatively new field with constantly developing technology and biologics, clinicians must continue to judiciously evaluate the evidence. With new evolving therapies, it is vital to remember the first priority remains to do no harm. Overall, due to an increase in promising evidence (Table 7.8) and a relatively good safety profile, regenerative medicine remains an important tool in the physician’s armamentarium to trial on a case-by-case basis: especially (even more so) when routine medical care fails to provide acceptable results. However, at this point regenerative medicine for spinal pain remains a hopefully optimistic treatment to be perfected in the future.

Table 7.8 A summary of the overall level of evidence for the listed regenerative medicine treatments and each specific pain syndrome targeted
Full size table