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 cause of absence from work in developed countries. Approximately 40% of chronic LBP is related to discogenic origin. The goal of the study is producing a review of literature to describe analytically the techniques of intradiscal injections.
METHODS: PubMed database was searched for clinical studies with the different key terms: "intradiscal", "injection", "steroid" "procedures", "techniques", "CT", "MRI", "fluoroscopy", "fluoroscopic", "guidance", "ozone", "ultrasound", "images". Only studies written in English, French, or Italian in which the intradiscal injection represents the main procedure for the low back discopathy treatment on humans were considered. We excluded the articles that do not mention this procedure; those which indicated that the intradiscal injection had happened accidentally during other treatments; those reporting the patient's pain was determined by other causes than the discopathy (facet joint syndrome, tumor, spondylodiscitis).
RESULTS: Thirty-one articles dated from 1969 to 2018 met the criteria. The examined population was 6843 subjects, 52.3% male and 47.7% female, with a mean age of 45.9±10.1 years. The techniques are highly variable in terms of procedure: different operators, needle guidance, injection sites, drugs, tilt angle of the needle).
CONCLUSION: The efficacy and the safety of the intradiscal procedures are not easily comparable due to different types of studies and their limited number. Further studies are needed to standardize the intradiscal injection technique/procedure to improve safety, repeatability and effectiveness, and last but not least to reduce peri- and postoperative care and health-care costs.

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).

Figure 1

Flow diagram illustrating published literature on intradiscal injection.

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 articles4,8-26,35–44 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 study38,40 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

Authors, Date	Study Design	Population	End Points	Disease Treated	Intervention	Medicament	Guidance	Approach	Outcomes	Results	Follow-up	Adverse Events
Kallewaard et al4 2015	Multicenter pilot study	174	Efficacy and safety of the treatment	LBP	ID	Blue methylene	Fluor	ND	VAS, ODI, SF-36, PGIC	Pain relief in 40% of patients	24 weeks	No
Mineta et al8 2014	Case report	1	Association Modic type I-inflammation	LBP	ID	Cs	ND	ND	VAS and Modic change	Modic type switch depends on inflammation	ND	ND
Zhang et al10 2013	Case–control	172	Comparison between 2 groups of the treatments	LBP	ID/IF	Ozone	Fluor	PL	VAS, JOA score	Significant pain relief in both groups	3, 24, 48 weeks	No
					ID/IF	Ozone Cs			Differences between treatments	No statistically significant differences
Beaudreuil et al11 2012	Observational Retrospective study	97	Efficacy in (EG) compared to (C)	LBP±Modic	ID	Cs	Fluor	ND	VAS and self-assessed of pain	Significant pain decrease in EG compared to CG	24 h, 48–56 weeks	No
Lehnert et al12 2012	Observational prospective study	283	Efficacy of the treatment	LBP	ID	Ozone	CT	EL	Disk volume reduction evaluated with CT	Decrease in 91.1% of patients and increased in 3.9% of patients	24 weeks	Impaired sensitivity lower limb 24 cases
					PG
de Seze et al13 2012	Observational prospective study	79	Efficacy and safety of the treatment	LBP	ID	Discogel	Fluor	PL	Verbally numeric scale (from 0 to 10)	Free pain in 60.7% of patients	8, 24 weeks	No
Fukui et al14 2012	Observational prospective study	45	Comparison of effectiveness between 2 treatment		ID	Saline sol LA	Fluor	PO	VAS, JOA score	Pain improvement for IDHP compared to MED	2, 12 weeks	No
					MED	/
Yu et al15 2012	Case–control study	45	Comparison of effectiveness between 2 treatment	LBP	ID	Cs	CT	ND	VAS, ODI	Improvement of pain in the EC compared to CG	1, 4, 12, 24 weeks	ND
						Saline sol
Cao et al16 2011	RCT	120	Comparison of effectiveness between 2 treatment	LBP	ID	Cs ± Songmeile	CT	PL	VAS, ODI	Significant pain improvement in the EC compared to CG	12, 24 weeks	ND
						Saline sol
Muto et al17 2008	ObservationalRetrospective study	2900	Efficacy and safety of the treatment	Lumbar disk herniation	ID/PG/PR	Ozone	CT	PvO	VAS, ODI	Significant pain decrease in 85% of cases	24, 48 weeeks	No
Oder et al18 2007	Observational Prospective study	621	Efficacy of treatment in different groups	LBP	ID/PG/Ep	Ozone, Cs, LA	CT+ Fluor	Post	VAS	Significant pain improvement in 59.4% of all patients	24 weeks	No
Fayad et al19 2007	Pilot study	74	Comparison of effectiveness between Modic type Groups	LBP + Modic	ID	Cs	Fluor	PL	VAS, PGA	Significant pain improvement ModicI >Modic I-2 > ModicII-1	4, 12, 24 weeks	No
Gallucci et al20 2007	RCT	159	Comparison of effectiveness between 2 groups	LBP disk herniation	ID/IF	Cs + LA	CT	Pv	Oswestry Low Back Pain Disability Questionnaire	Significant pain decrease for 47% in group A, 74% in group B	24 weeks	No
					ID/IF	Cs, LA, ozone		IL
Miller et al21 2006	Prospective study	76	Efficacy of the treatment	LBP + leg pain	ID	HyD LA	Fluor	ND	Numeric pain score scale	Pain improvement in 43.4% of patients	6, 164 weeks	ND
Benyahya et al22 2004	Observational retrospective study	85	Efficacy/safety of the treatment	LBP Modic	ID	Cs	CT	PL	Global appreciation “good or excellent” of the patient in %	71.8% at 1 month 55.3% at 3 months 43.5% at 6 months	4, 12.24 weeks	Collapse 2 discs
Khot et al23 2004	RCT	120	Comparison of effectiveness between 2 groups	LBP	ID	Cs	Fluor	PL	VAS, ODI	No improvement in experimental group compared with placebo	48 eeks	No
						Saline sol
Andreula et al24 2002	Multicenter study	600	Comparison of effectiveness between 2 groups	LBP	ID/PG	Ozone + O3	Fluor	EL	Modified MacNab method	Successful of the treatment in 70.3% group A 78.3% group B	24 eeks	Impaired sensitivity lower limb 2 cases
					ID/PG	Ozone, Cs, LA	CT
Feffer et al25 1969	Observational prospective study	244	Efficacy of the treatment	LBP	ID	Cs	Fluor	PL	Back pain (yes/no), radicular pain (yes/no)	Complete remission in 46.7% of patients; “no initial response” in 53.3% of patients.	4–10 years	Discitis 1 case
Pettine et al35 2017	Observational prospective study	26	Efficacy of the treatment	LPB	ID	AT-BMC	Flour	PL	VAS, ODI	VAS: 82.1 (±2.6) at baseline and 21.9±4.4 at 3 Y. ODI: 56.7 (±3.6) at baseline and 17.5±3.2 at 3 years.	3 years	No. Only 6 patients had progression to surgery
Yin et al9 2014	Multicenter study	15	Efficacy and safety of the treatment	LBP	ID	FS	Flour	PL	VAS, RMDQ, Rx/MRI	VAS 72.4, 31.7, 35.4, 33 at baseline, 26, 52, 104 weeks. (significant improvement) RMDQ 15.2, 8.9, 6.2; 5.6 at baseline, 26, 52, 104 weeks. Rx/MRI similar to baseline	104 weeks	Low back muscle spasm 2 cases. Discitis 1 case
Sainoh et al36 2016	Prospective randomized study	60	Efficacy of the treatment	LBP	ID	TNF-α I	Flour	PL	VAS, ODI	Significant pain improvement in 57% of all patients at 8 weeks in patient treated with TNF-α I. No significant difference about ODI	8 weeks	Not
Noriega et al44 2017	Randomized, controlled trial	24	Efficacy and safety of the treatment	LBP	Nd	Nd	Nd	ND	VAS, ODI, SF-12	Average 28% improvement in pain and disability 1 year after the intervention	1 year	ND
Giurazza et al38 2017	Review	ND	Efficacy and safety of the treatment	LBP	ID	Oz	Flour; CT	PL, Pv/IL, TF	VAS, ODI	Improvement in pain and disability	Up to 10 years	0.1% Paraesthesias; temporary impaired bilateral sensitivity; vitreoretinal hemorrhages; thunderclap headache 1 case of vertebrobasilar stroke 1 case of septicemia
Pettine et al42 2015	Observational prospective study	26	Efficacy and safety of the treatment	LBP	ID	AT- BMC	Flour	PL	VAS, ODI	(71% VAS reduction) and ODI improvements (>64%) through 2 years.	2 years	No. only 5 patients had progression to surgery
Nguyen et al34 2017	Prospective, parallel-group, double-blind, randomized, controlled, multicenter study.	135	Efficacy and safety of the treatment	LBP	ID	Cs	Flour	PL	VAS, MRI at 12 months	The percentage of responders (LBP intensity <40) at 1 month was higher in the GC IDI group in than the control group; At 12 months, the groups did not differ in pain intensity or most other secondary outcomes. No difference at MRI	1 year	1 event increase in sciatica pain in the 24 h after the intervention
Perri et al41 2016	Observational prospective study	517	Efficacy and safety of the treatment	LBP	ID	Oz+Cs+LAVs Cs+LA	CT	P	VAS	The study group had a successful outcome in 80% of patients after 6 months, while the control group had a successful outcome in 31.5%	6 months	Not
Zhang et al37 2016	Observational study	33	Efficacy and safety of the treatment	LBP	ID	BM	Flour	PL	NRS, ODI, MRI (apparent diffusion coefficient in T2)	Significant improvement at short-term minimum of 2 points reduction of rating scale scores at 1, 3, and 6 months after treatment, but less than 2 points reduction at 12 months; 50% improvement on the ODI at 1, 3, and 6 months after treatment, but not at 12 months. Apparent diffusion coefficient and T2 value were significantly higher at 6 and12 months after treatment, but no difference at 3 months.	12 months	Not
Levi et al46 2016	Prospective study	22	Efficacy and safety of the treatment	Discogenic LBP	ID	PRP	Flour	PL	VAS, ODI	Significant improvement	6 months	Not
Tuakli-Wosornu et al39 2016	Prospective, double-blind, randomized controlled study.	47	Efficacy and safety of the treatment	Discogenic LBP	ID	PRP	Flour	PL	NRS, FRI, SF-36, NASS	Significant improvement	1 year	Not
Kumar41 2017	Single-arm phase I clinical trial	10	Efficacy and safety of the treatment	LBP	ID	AT-MSC	Flour	PL	VAS, ODI, SF-36, MR-ADC	Significant improvement; 3 patients increased water content based on the ADC map at the 12-month follow-up	1 year	Not
Centeno et al43 2017	Pilot study	33	Efficacy and safety of the treatment	LBP	ID	AT-MSC, LA	Flour	PL	NRS, SANE, FRI, MIDPD	Significant improvement; the patients treated underwent post treatment MRI and 85% had a reduction in disc bulge size, with an average reduction size of 23% post-treatment	6 years	Not

Abbreviations: ADC, apparent diffusion coefficient; AL-MSC, allogenic mesenchymal stem cells; AL, anterolateral; AT-BMC, autologous bone marrow concentrate; AT-MSC, adipose tissue mesenchymal stem cells; BM, blue methylene; C, control group; Cs, corticosteroid; EG, experimental group; EL, extraspinal lateral; Ep, epidural; EQ-5, EuroQol; Fluo, fluoroscopy; FRI, Functional Rating Index; FS, fibrin sealant; HA, hyaluronic acid; HyD, hypertonic dextrose; ID, intradiscal injection; LA, local anesthetic; LBP, low back pain; MGPQ, McGill Pain Questionnaire; MIDPD, measurement of the intervertebral disc posterior dimension; MSC, mesenchymal stem cells; NASS, the modified North American Spine Society; ND, not described; NRS, Numeric Rating Scale; ODI, Oswestry Disability Index; P, pain; PG, periganglionic injection; PL, posterolateral; PO, posterior-oblique; Post, posterior; PR, periradicular injection; PRP, platelet-rich plasma; Pv/IL, paravertebral/interlaminar; PvO, paravertebral-oblique; RCT, randomized controlled trial; RMDQ, Roland–Morris Disability Questionnaire; SANE, modified single assessment numeric evaluation; SF-36, short form-36; SNRB, selective nerve block; TF, transforaminal; TNF-α I, tumor nerosis factoralpha inhibitor; VAS, visual analog scale;.

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 intraforaminal11,21 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 studies11,13,15,18,19,21,25 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 specified9,44 (Figure 2).

Table 2

Characteristics of Intradiscal Injection Techniques

Authors	Intervention	Drugs	Guidance	C-Arm	Approach	Needle	Injection Site	Injection Check	Operator		Patients Position	Sedation	Local Anesthesia	Antib Prophilax	Rest After Injection
Kallewaard et al4	ID	BM, LA	Fluor	No	ND	ND	ND	AP/L	ND		ND	ND	ND	Yes	2 h
Mineta et al8	ID	Cs	ND	No	ND	ND	ND	ND	ND		ND	ND	ND	ND	ND
Yin et al9	ID	FS	Fluor	No	PL	18G	CTh	RTMF	ND		ND	No	Yes	Yes	1, 1/2 h
Zhang et al10,11	ID/IF	Oz	Fluor	Yes	PL	21G	C	AP/L	ND		*	No	Yes	ND	10 min
	ID/IF	Oz Cs
Beaudreuil et al11	ID	Cs	Fluor	No	ND	ND	C	ND	ND		ND	No	Yes	ND	12/24 h
Lehnert et al12	ID	Oz	CT	No	EL	22G	ND	ND	ND		ND	Nt	Yes	ND	6 h
	PG
De Seze et al13	ID	Discogel	Fluor	No	PL	22G	C	AP/L	ND		ND	DS	No	ND	3 h
Fukui et al14	ID	Saline sol, LA	Fluor	No	PO	22G	MPHD	AP/O/L	ND		P	ND	ND	ND	1 h
Yu et al15	ID	Cs	CT	No	ND	ND	ND	ND	ND		ND	ND	ND	ND	ND
		Saline sol
Cao et al16	ID	Cs, Songmeile	CT	No	PL	22G	C	ND	Authors		ND	ND	ND	ND	3 h
		Saline sol
Muto et al17	ID/PG/PR	Oz	CT	No	PvO	18/20G	C	ND	ND		P	ND	ND	ND	ND
Oder et al18	ID/PG/Ep	Cs, LA, Oz	CT/Fluor	No	Post	22G	C	CT scan	ND		P	Yes	Yes	Yes	12 h
Fayad et al19	ID	Cs	Fluor	No	PL	22G	C	AP/L	Radiol		ND	No	No	ND	ND
Gallucci et al20	ID/IF	Cs, LA	CT	No	Pv/IL	22G	C	CT scan	NeuroR		P	No	Yes	Yes	2 h
	ID/IF	Cs, LA, Oz
Miller et al21	ID	HyD, LA	Fluor	No	ND	22G	C	ND	ND		ND	ND	ND	ND	ND
Benyahya et al22	ID	Cs	CT	No	PL	ND	C	ND	Radiol		ND	ND	ND	ND	ND
Khot et al23	ID	Cs	Fluor	No	PL	ND	ND	ND	Authors		ND	AwS	No	ND	ND
Andreula et al24	ID/PG	Oz, O3	Fluor	No	EL	22G	ND	ND		ND	ND	No	No	ND	2 h
	ID/PG	Oz, Cs, LA	CT
Feffer et al25	ID	Cs	Fluor	No	PL	22G	ND	ND		ND	ND	ND	ND	ND	ND
Centeno et al43	Ep/ID	AT-MSC, LA	Fluor	No	PL	ND	ND	ND		ND	P	No	Yes	ND	ND
Zhang et al37	ID	BM	Fluor	Yes	PL	ND	C	Dis		ND	P	ND	ND	ND	24 h
Tuakli-Wosornu et al39	ID	Prp	Fluor	ND	PL	20G/25 G	C	Dis		Physiatrist	P	No	Yes	Yes	ND
Kumar et al41	ID	AT-MSC, HA	Fluor	Yes	PL	22 G	C	CT scan		Spine surgeon	P	ND	Yes	Yes	4 h
Levi et al46	ID	Prp	Fluor	ND	PL	22G/25 G	C	ND		Physiatrist	P	Yes	Yes	Yes	ND
Pettine et al35	ID	AT-BMC	Fluor	ND	PL	22G	C	ND		ND	P	No	Yes	ND	ND
Sainoh et al36	ID	TNF-α I	Fluor	ND	PL	22 G	C	RTMF		ND	LD	No	Yes	ND	ND
Noriega et al44	ND	AL-MSC	ND	ND	ND	ND	ND	ND		Radiologists	ND	ND	Yes	ND	ND
Giurazza et al38	ID	O2-O3	Fluor, CT	ND	PL, Pv/IL, TF	18–22 G	C	ND		Radiologists	P	Yes	ND	ND	ND
Pettine et al42	ID	AT-BMC	Fluor	NDd	PL	22 G	C	ND		ND	P	No	Yes	No	ND
Nguyen et al34	ID	Cs	Fluor	ND	PL	18/22 G	C	ND		Radiologists	LD	No	No	No	ND
Perri et al40	ID	O2-O3	CT	ND	Pv/IL	22 G	C	CT		Neuroradiologists	P	No	Yes	Yes	ND

Note: *A pillow was placed under the waist of patients.

Abbreviations: AL- MSC, allogenic mesenchymal stem cells; AL, anterolateral; AT-BMC, autologous bone marrow concentrate; AT-MSC, adipose tissue mesenchymal stem cells; Aw S, awake sedation; BM, blue methylene; C, center of the disc; Cs, corticosteroid; CTh, central third of the disc; Dis, discography; DP, deep sedation; EL, extraspinal lateral; Ep, epidural; Fluor, fluoroscopy; FS, fibrin sealant; G, gauge; h, hours; HA, hyaluronic acid; HyD, hypertonic dextrose; ID, intradiscal injection; IDHP, intradiscal high pressure injection; L, lateral; LA, local anesthesia; LD, lateral decubitus; MED, microendoscopic discectomy; min, minutes; MPHD, mid portion of herniated disc; MSC, mesenchymal stem cells; ND, not described; NeuroR, neuroradiologists; Oz, ozone; P, prone; PG, periganglionic injection; PL, posterolateral; PO, posterior-oblique; PR, periradicular injection; Prp, platelet-rich plasma; Pv/IL, paravertebral/interlaminar; PvO, paravertebral-oblique; Radiol, radiologists; RTMF, real-time multiplanar fluoroscopic imaging; SNRB, selective nerve block; Songm, Songmelie; TF, transforaminal; TNF-α I, tumor nerosis factor-α.

Figure 2

Different type of image guidance for intradiscal injection.

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 works34,37–39 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 Levi40,46 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 articles13,14,21,22,25 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 al9,12,22,24,25 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 physiatrists46,39 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 sedation18,23 and three describe a deep sedation.13,38,46 Some unreviewed medical articles28,29 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 articles14,17,20,26,28–30,35,37-44 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 works34,36 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 line18,20 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.