The Technique of Intradiscal Injection: A Narrative Review

Background

Low Back Pain (LBP) is one of the most common spine diseases and represents the most frequent reason of absence from work in developed countries. Around 80% of adults suffer from LBP during their lifetime, and 55% suffer from back pain associated with radicular syndrome.¹ Chronic LBP is often responsible for a low quality of life due to pain, for disability and loss of work productivity and, in addition, for high health-care costs for society.^2–4 Regarding its etiology, in literature it has been reported that approximately 40% of chronic LBP has a discogenic origin.^5,6 Currently in the advanced phases of discopathy and in high symptomatic subjects, the elective treatment still remains spinal surgery. In the other less complicated cases, the therapeutical steps could vary from a simple pharmacological therapy to a physical therapy, as the low back traction, over to spinal injection (epidural, periradicular, intradiscal and intra-articular procedures). Moreover, intervertebral disk decompression techniques are minimally invasive outpatient procedures for the treatment of disk herniation. Under imaging guidance and via a percutaneous approach, a needle is inserted in the nucleus pulposus of the herniated disk. A variety of decompressive device of thermic, chemical or mechanical nature are introduced inside the nucleus pulpous with minimal disruption of the surrounding tissues, assuring its partial removal and a significant decrease of intradiscal pressure. Thermal decompression is achieved using laser fiber, plasma, energy electrode, and radiofrequency coil/electrode. Chemical decompression is achieved by alcohol gel or ozone intradiscal injection, which causes dehydration and breakdown of the nucleus pulposus. Lately, there has been a trend for biomaterial implantation (hydrogel, platelet-rich plasma and stem cell therapy) aiming for intervertebral disk regeneration. Symptomatic intervertebral disk herniation (refractory to 4–6 weeks of a conservative therapy course), occupying less than one third of the spinal canal, as confirmed by MRI (magnetic resonance imaging), is an indication for percutaneous decompressive disk therapies. The mean success rate for all techniques is approximately 85%. The mean complication rate (infections like spondylodiscitis, allergic reaction, hemorrhage, neurologic injury) is <0.5%.⁷ The goal of the study is producing a review of literature to describe analytically the actual techniques of intradiscal injections, the type of intervention performed, the used imaging guidance, the inoculated drug, the approach to the intervertebral disc, the patient’s position, the specialty of the operator performing the procedure, the type of anesthesia and the use of antibiotic prophylaxis.

Methods

Search Strategy

The PubMed database was searched for clinical studies with the following key terms: “intradiscal”, “injection”, “steroid” “procedures”, “techniques”, “CT” (computerized tomography), “MRI”, “fluoroscopy”, “fluoroscopic”, “guidance”, “ozone”, “ultrasound”, “images”. We made our research throw the combination of this terms, inserted between the Boolean operators “AND”/“OR”. We limited the research to studies on humans and types of articles were: case reports, clinical trials, controlled clinical trials, reviews, comparative studies, multicenter studies, and randomized controlled trials. The search was expanded through the bibliography within recruited texts (Figure 1).

Inclusion and Exclusion Criteria

For our review, we only considered studies written in English, French, or Italian in which the intradiscal injection represents the main procedure for the low back discopathy treatment, both isolated and in combination. We excluded the articles that did not mention that procedure or those which indicated that the intradiscal injection had happened accidentally during other treatments (ie during transforaminal injection). We excluded the articles where the patient’s pain was due to other causes than the discopathy (facet joint syndrome, tumor, spondylodiscitis), and also those articles that described the treatment performed on an animal species. We checked the bibliography to make sure that the articles were compatible with our research.

Results

Initially using the term “intradiscal injection” as a search key on PubMed, we found 385 articles; the results were reduced when we added other search keys or other selection criteria, as we showed in the description of our strategy of research. Depending on the abstracts or full texts we excluded the studies that did not satisfy the inclusion criteria. Moreover, in this review we included other articles shown as bibliography in previous research. The final result consisted of 31 articles^{4,8-26,35–44} dated from 1969 to 2018, and the examined population was 6843 subjects (Table 1). We did not consider the number of patients treated in the Giurazza et al study^38,40 because being a review, it considered not only intradiscal, but also paravertebral injections. We also decided to cite other authors that described the varied and numerous procedures that are available to the image-guided interventions who may provide these therapies for the spine.^27–31

Table 1 Clinical Characteristics of Trials Employment of Intradiscal Injection

Characteristics of Included Articles (Table 1)

The review includes three observational retrospective studies,^12,18,23 12 observational prospective studies,^{13–15,19,22,26,36-38,41,43,44} two multicenter pilot studies,^4,10 two case–control studies,^11,16 six randomized controlled trials,^{4,8,17,21,24,35,42} one multicenter study²⁵ two pilot studies,^20,43 one case report,⁹ one single arm phase I clinical trial⁴⁴ and one review.⁴⁰

Population

Our population is composed of 6843 subjects, 52.3% male and 47.7% female, with a mean age of 45.9±10.1 years.

End Points

The aim of the study was to review literature for scientific evidence of intradiscal injections, to describe analytically the actual techniques, the type of intervention performed, the used imaging guidance, the inoculated drug, the approach to the intervertebral disc, the patient’s position, the operator who performed the treatment, the type of anesthesia used, antibiotic prophylaxis, if used.

Treated Disease

In all selected articles,^{4,9-26,35–38,39–44} the patients suffered from lumbar discopathy.

Type of Procedure

Different types of treatments are reported in the studies: intradiscal injection,^{4,8-26,35–44} epidural¹⁹ intraforaminal^11,21 and facet joint injection, selective nerve block (SNRB),⁸ intradiscal high pressure injection (IDHP),¹⁵ microendoscopic discectomy (MED).¹⁶

Intradiscal Injection

The technique of intradiscal injection is reported in 31 articles. In 24 studies,^{4,9,10,12,14,16,17,20,22-24,26,35–44} it is the only treatment while, in the remaining seven studies^{11,13,15,18,19,21,25} it is compared to, or in association with, other minimally invasive procedures. (Table 2). In 19 articles the procedure was realized under fluoroscopy,^{4,10-12,14,15,20,22,24,26,35–39,41,42,44} in seven articles under CT guide,^{13,16–18,21,23,43} and in three studies both by fluoroscopy and CT guided, in comparison²⁵ or in association;^19,40 in two articles was not specified^9,44 (Figure 2).

Table 2 Characteristics of Intradiscal Injection Techniques

The patients of the de Seze et al trial,¹³ Levi et al⁴⁶ and Giurazza et al³⁸ works were subjected to neurosedation, in the trials by Khot et al²³ and Oder et al¹⁸ they were subjected to conscious sedation. In the studies by Fayad et al¹⁹ and Andreula et al²⁴ and another five works^34,37–39 the patients were neither sedated nor subjected to local anesthesia. In eleven trials,^{10,11,13,19,21,36,37,41–44} all patients were subjected to local anesthesia. The antibiotic prophylaxis was used in eight trials.^{4,10,19,21,41–44} The interventions were performed by highly experienced operators in the Lehnert’s trial¹² two clinicians (two authors) in the Cao et al’s study;¹⁶ Fayad et al,¹⁹ Benyahya et al,²² Noriega et al,⁴⁴ Giurazza et al,³⁸ and Nguyen et al³⁴ report the experienced radiologists, Gallucci et al²⁰ and Perri et al⁴⁰ neuroradiologists, Tuakli-Wosornu et al³⁹ and Levi^40,46 physiatrist and Khot et al²³ two senior authors.

In 17 articles the required patients position was described according to the procedure: in 11 studies a prone position was used,^{15,18,19,21,36,38,40–44} Nguyen et al³⁴ and Sainoh et al³⁶ propose a lateral decubitus and Zhang et al¹⁰ advised use of a pillow under the waist to get the widest intervertebral spaces.

For the procedure spinal needles of 18- (n=4), 20- (n=2), 21- (n=1), 22- (n=17) and 23- (n=1) gauge were used, with variable length to 7 from 17.8 cm. For example the Muto et al’s study¹⁷ mentioned a 22-gauge spinal needle with paravertebral oblique access, Lehnert et al’s study¹² mentioned an extraspinal lateral approach with a 22-gauge 17.8-cm spinal needle and Gallucci et al²⁰ a paravertebral/interlaminar approach with a 9- or 15-cm 22-gauge spinal needle Five articles^{13,14,21,22,25} have specified that the side of the injection was chosen on the basis of the main location of symptoms.

The percutaneous approach is always posterior for the lumbar access: in 15.^{10,11,14,17,20,23,24,26,35,36,38,40–44} out of 31 articles it was specified as a posterolateral access, in other studies it was extralaminar¹² or paravertebral access,²⁰ posterior-oblique¹⁴ paravertebral-oblique,¹⁷ or anterolateral access. In the Gallucci et al’s study²⁰ the intradiscal and intraforaminal injections were administered with a paravertebral approach in 92.4% of the patients and an interlaminar approach in 7.6% of the patients. The needle was advanced through the intraforaminal space, with an angle usually between 45° and 60°. In seven articles the point of access is not described.^{4,9,12,16,22,39,44} Oder et al’s study¹⁸ specified that the percutaneous approach was about 45° along the lateral margin of the inferior articular process of the vertebra and through the neuroforamen for preserving the nerve root. Muto et al¹⁷ used a needle inclination in a craniocaudal direction in the case of a lower herniation.

The site of injection is the center of the disc in 22 studies,^{7,10,11,13,16-22,35–44} the central third of the disc in the Yin et al’s study⁹ and in the mid portion of the herniated disc in the Fukui et al’s study.¹⁴ The position of the needle was confirmed by fluoroscopy using anteroposterior and lateral views in 20 studies;^{4,8,10–12,14,15,20,22,24,25,35–39,41,42,44} in Yin et al’s trial⁹ the procedures were performed with real time multiplanar fluoroscopy, with CT scan in seven other articles,^{13,16–18,21,23,43} in three articles with both fluoroscopy and CT,^19,25,40 in two articles the position of the needle is not described.^9,40

Some authors recommend the time remaining in supine position after injection: respectively 10 minutes;¹⁰ one hour,¹⁴ one, half an hour,⁹ two hours,^4,21,25,39 three hours,^13,16 four hours,⁴¹ six hours;¹² 12 h,¹⁸ 12/24 h.^11,37

Medicaments Injected

Several drugs have been injected, and were used individually or in association with each other: an oxygen-ozone mixture (O₂O₃) in eight studies,^{10,12,17,18,20,24,38,40} a saline solution in four studies;^14–16,23 in 13 articles the steroids have been administered (methylprednisolone, acetate of prednisolone, hydrocortisone, betamethasone),^{8,10,11,15,16,18–20,22–25,34} and in six trials the local anesthetic (bupivacaine, lidocaine) were injected.^{4,14,20,21,24} For the remaining studies hypertonic dextrose,²¹ fibrin sealant,⁹ blue methylene,^4,37 discogel,¹³ autologous bone marrow concentrate,^35,42 allogenic mesenchymal stem cell and hyaluronic acid,⁴¹ tumour necrosis factor α1 inhibitor³⁶ and songmeile¹⁶ (a kind of synthetic liquid of polypeptidic biological factors extracted from Chinese herbal medical ingredient) were used.

Outcomes Measures

Pain was the most frequently tested variable. It was expressed as percentage of patients with pain relief or as mean improvement on a continuous scale. The outcome measures shown in the studies were: VAS (visual analog score), NRS (numeric rating scale), McGill Pain Questionnaire. Outcome assessment of patient satisfaction are reported by “modified MacNab scale” or using Odom criteria (“Excellent”, “Good”, “Satisfactory” and “Poor”). Back-specific disability is expressed on a back-specific index, such as the Roland Disability Questionnaire or the Oswestry Disability Index and JOA score (widely used in Japan to evaluate disabilities associated with low-back pain and includes the following items: subjective symptoms; clinical signs; restriction of activities of daily living. JOA score ranges from 29 as the most positive score to minus six for the worst a global measure of improvement). Quality of life is measured by the SF-12, SF-36, and EuroQol.

Patients Global Impression of Change (PGIC) measured by a seven-point Likert scale. The evaluations of general health status or well being, disability for work, and patient satisfaction have all so been reported.

The disc volume was evaluated by MRI and CT images.

Clinical and/or radiologic short term follow-up were mainly performed at four or six weeks; the long-term follow-up were performed from 12, 24, 48, weeks up to 4–10 years.

Efficacy

The efficacy of the treatment is the target in 30 articles,^{4,9–25,34–44} The results are reported as clearly satisfactory in 27 out of the 30 articles,^{4,9–20,22,24,34–44} In the Muto et al’s study, for example, the results on 2900 patients, treated for LBP with intradiscal injection of O₂–O₃, were evaluated with the modified MacNab classification, the VAS and the Oswestry Disability Index at six and 12 months. Success rates were 75–80% for soft disc herniation, 70% for multiple-disc herniations and 55% for failed back surgery syndrome. None of the patients suffered early or late neurological or infectious complications.¹⁷ Benyahya et al²² made a retrospective study of medical records of 85 patients (55 women, mean age 49±9 years) to assess the effectiveness of intradiscal injection of acetate of prednisolone for the treatment of LBP. They used the global appreciation of the patient (excellent, good, mild, none, worse) concerning the result of the intradiscal injection, at one, three and six months. For effectiveness of intradiscal injection, the results showed that 71.8% of the patients considered the result good or excellent at one month, 55.3% at three months and 43.5% at six months.

Adverse Events

Six trials have reported the side effects,^{9,12,22,24,26,34} overall 32 cases for 6843 patients (0.47% of patients): three cases of discitis, two after injection of corticosteroid,²⁵ one after injection of fibrin sealant;⁹ 26 patients present impairment of sensitivity in the lower limb ipsilateral to the treatment with injection oxygen-ozone;^12,24 two discs showed a collapse after injection of corticosteroid,²² 1 case of increase in sciatica pain in the 24 hafter the intervention.³⁴

Two trials were performed under CT guidance;^12,22 two studies were performed by fluoroscopy,^9,25 and only one case by both fluoroscopy and CT guide.²⁴ Adverse events occurred in about 0.7% of the patients with CT guided injection and in 0.2% of the patients with fluoroscopic guided injection. In Yin et al’s trials,⁹ the patients have even been subjected to antibiotic prophylaxis, in others articles this was not described.

Yin et al, Lehnert et al, Benyahya et al, Andreula et al and Feffer et al^{9,12,22,24,25} report a posterolateral/extraspinal-lateral approach. Giurazza et al report that

The overall procedural complications rate is estimated around 0.1%. Have been reported in the literature: paresthesia on the anterolateral portion of the left leg and foot, suggesting nerve injury; few temporary episodes of impaired bilateral sensitivity; vitreoretinal hemorrhages; thunderclap headache related to pneumoencephalus as a consequence of inadvertent intrathecal puncture; and 1 case of vertebrobasilar stroke.³⁸

Discussion

For the low back pain management, patients with a small or contained herniated disc with no response to medical treatments, can be candidates for one of the minimally invasive percutaneous techniques. Generally, the minimally invasive techniques offer good results with patient compliance and low cost, showing a very low side effects percentage.²⁰ Only 0.47% of patients have manifested adverse events after intradiscal injection. The procedure is carried out on an outpatient basis by highly experienced operators such as radiologists,^19,22 neuroradiologists,²⁰ physiatrists^46,39 and orthopedics.²³ The procedure is of interest for many medical areas, for this reason standardizing this method allows it to be extended to various practitioners.

For preoperative management there is no consensus regarding sedation, local anesthesia, or antibiotic prophylaxis. Only seven authors mention antibiotic use,^{4,9,20,46–41} only two articles describe conscious sedation^18,23 and three describe a deep sedation.^13,38,46 Some unreviewed medical articles^28,29 do not recommend local or general anesthesia because they could mask the nerve root puncture symptoms; the needle passes very close to the nerve root and may often touch it, causing a strong electric shock sensation which is quite harmless; if the patient is conscious they will feel the pain. About 0.19% of the patients subjected to antibiotic prophylaxis have had adverse events; while without antibiotics about 0.09% of the population have had side effects; current data do not allow a statistical analysis; for this reason prospective clinical trials are needed. Some authors advise setting up an aseptic room for anesthesiology care, ensuring peripheral access to the patient.²⁹

A concordance has emerged about the patient position, the injection site and the needle type. The most included articles^{14,17,20,26,28–30,35,37-44} report a prone position as the best to increase the intervertebral space, also using a support under the abdomen to reduce lumbar lordosis. The lateral decubitus was reported in two works^34,36 and in an unreviewed journal on chemiodiscolysis with ozone.²⁸ According to five of the articles,^{12,13,20,21,24} de Santis et al²⁹ recommend an access side at the same side as the symptoms.

The chosen injection site is the center of the disc, and the injection point was checked by fluoroscopic projections,^{4,9–11,13,14,19,21,23,25,34–39,41,43} CT scans;^{12,15–18,20,23,24,38,40} we highlight the need for trials to evaluate the more effective, safe, and less expensive methods, especially if using a toxic or very expensive drugs. For the safety, the data do not clarify which is the least injurious method, even though we have recorded a greater percentage of adverse events with the CT guided injection (Table 1). Clinical trials with same medication comparing the fluoroscopy and CT guided injections are needed.

During the procedure, the needle can be readily shifted a few millimeters to pass through without damaging the nerve. The approach and the needle inclination are essential criteria for a successful and safe procedure. In some articles it appears that the lumbar approach has a lateral inclination of 45° to 60° with respect to the axial line^18,20 and that for the lower discs an additional cranial-caudal inclination is needed.²⁰ We did not find accurate descriptions on the needle insertion procedure because the needle course was always evaluated radiographically and the access site was chosen accordingly. The authors recommend and/or use a fluoroscopy performed with the C-arm, that allows identification the trajectory of optimal access for needle placement into each disc.^10,29,30 An image-guided procedure handbook³⁰ describes a window of anatomical access to the intradiscal injection delineated by the superior articular process medially, the superior endplate below, and the traversing nerve root laterally and above. Staying close to the superior articular process could keep the needle as far as possible from the traversing nerve root.

The postintervention management was different between treatments, the authors have advised several rest times depending on the procedure (Table 2).

The efficacy and the safety of the intradiscal procedures are not easily comparable because the techniques are highly variable in terms of procedure (different operators, needle guidance, injection sites, drugs, tilt angle of the needle) (Table 2).

Conclusions

The efficacy and the safety of the intradiscal procedures are not easily comparable because of differences in the design of studies and their limited number.

The intradiscal injection is a technique widely used in the LBP management of patients with no response to rehabilitative and medical treatments. Differences of agreement between researchers are present on the technical aspects of the procedure in terms of imaging guidance, of injected substances, and efficacy of evaluation tools.

Further studies are needed in order to standardize the intradiscal injection technique/procedure as well as to improve efficacy, safety, repeatability, and to assess cost-effectiveness.

Figure 1 Flow diagram illustrating published literature on intradiscal injection.

Figure 2 Different type of image guidance for intradiscal injection.