Conduct of diagnostic cervical and lumbar medial branch blocks

1. Introduction
2. Principles of use
3. Patient selection and consent
4. Conduct of medial branch blocks
5. Outcome assessment and interpretation

1. Introduction

1.1. Local anaesthetic blocks of the medial branches of the dorsal rami are diagnostic procedures used to test whether one or more of these nerves mediate a patient’s spinal pain.


1.2. Of the structures innervated by the medial branches, the zygapophysial (Z, or facet) joints are the most likely to be a discrete source of persistent spinal pain.1-3 For this reason, medial branch blocks are commonly known as Z-joint blocks.


1.3. Medial branch blocks differ from intra-articular injections into the Z-joints with respect to the target structure. There is strong evidence for controlled comparative local anaesthetic facet joint injections and medial branch blocks in the diagnosis of neck and low back pain.4 For pain relief with medial branch blocks, the evidence is moderate for short- and long-term pain relief. 5 With intraarticular facet joint injections, the evidence for short- and long-term pain relief is limited for cervical pain, and moderate for lumbar pain. 5.


1.4 There is medium quality evidence that medial branch blocks of the cervical and lumbar dorsal rami are safe, reasonably accurate, and have diagnostic utility in the investigation of spinal pain in carefully selected adults.6,7 If these blocks are used to select patients for radiofrequency neurotomy of the medial branches, medial branch blocks have potential therapeutic utility.8-11


1.5. For blocks of the thoracic medial branches, there is limited evidence regarding accuracy,12,13 and no evidence of therapeutic utility. Accordingly, at the time of writing, no evidence-based recommendation can be made regarding how or when thoracic medial branch blocks should be incorporated into clinical practice.

2. Principles of use

2.1. The following sections outline the responsibilities of practitioners performing diagnostic medial branch blocks (hereafter termed “proceduralists”) including: patient selection and preparation, conduct of the procedure, outcome assessment and interpretation.


2.2. The proceduralist is responsible for the components of care delivered personally, and also for components of the care delegated to others. In particular:

2.2.1. The proceduralist must ensure that a comprehensive multidimensional assessment has been performed prior to considering diagnostic medial branch blocks (see 3.1 – 3.4);


2.2.2. The proceduralist should consider delegating outcome assessment to an independent observer to reduce bias (see 5.2),6 but remains responsible for overseeing this process.  Therefore, the proceduralist should be satisfied that their delegate has appropriate training, competence and communication skills;


2.2.3. The proceduralist should ensure the results are communicated to an appropriately skilled medical practitioner responsible for co-ordinating the patient’s ongoing care subsequent to the diagnostic blocks.

2.3 It is recommended that proceduralists use the technique(s) for which there is some evidence.


2.4. Patient self-report of pain experience determines the outcome of the block.  Therefore, patients must have adequate baseline cognitive capacity and communication skills. Any medications that impair cognitive function should be avoided.  If a patient requires sedation to withstand the procedure, serious consideration should be given to the likely reliability of the outcome and the propriety of proceeding (if proceeding see ANZCA PS9 (2005) -Guidelines on Conscious Sedation for Diagnostic, Interventional Medical and Surgical Procedures).


2.5. Diagnostic blocks are essential to maximise diagnostic reliability,14,15 therapeutic utility,8-11 and are cost effective.16  In conventional Pain Medicine practice, comparative local anaesthetic blocks can be used.5-7

3. Patient selection and consent

3.1. Prior to considering diagnostic medial branch blocks, patients with spinal pain should undergo comprehensive assessment including medical and pain history, physical and psychosocial assessments.


3.2. In particular, serious possible causes of spinal pain such as tumours, infection, vascular and metabolic disease must be excluded by careful history, examination, and appropriate laboratory and imaging tests.


3.3. Diagnostic medial branch blocks may be considered in patients:

3.3.1. who meet the IASP Taxonomy criteria for somatic cervical or lumbar spinal pain of unknown or uncertain origin (IASP Taxonomy 13X.X8c-eS and 5XX.X8c-eS);17 and


3.3.2. who are able to identify one or more focal areas of pain in a region supplied by one or more medial braches of dorsal rami of cervical or lumbar nerves; and


3.3.3. whose painful disabilities persist for at least 3 months (although there is no a priori reason why these blocks might not be informative in acute spinal pain, there is a lack of evidence supporting therapeutic utility in patients with pain of shorter duration)18 ; who meet the requirements that enable the procedure to be performed awake and without sedation; who able to give clear responses as to efficacy of procedure; who have a full understanding of diagnostic purpose of procedure; and who are psychologically suitable (cognitively intact, not distressed).

3.4. Patients may or may not have had adequate trials of other treatments or management strategies. It is important that the clinician review the patients’ prior access to, and experience of, the various conservative management options before embarking on diagnostic medial branch blocks.


3.5. Patients need to be informed about specific issues related to these blocks:

  • That they are diagnostic only (not therapeutic)
  • That a placebo injection or double blinding or similar techniques will be used.
  • That the patient should not receive sedation during the procedure and needs to be prepared to be awake
  • That the patient will be exposed to radiation
  • Other complications of the blocks



3.6. Indications:

3.6.1. The primary indication for these diagnostic procedures is to influence management.  Therefore, patients must be informed regarding the nature, availability, outcomes and limitations of management options. If a patient would not proceed with management targeting the medial branches/facet joints, then diagnostic medial branch blocks may not be indicated.


3.6.2. An important secondary consideration is that a diagnosis may have utility per se. For some patients, a diagnosis may curtail any further diagnostic pursuit, minimize harm by avoiding therapies for which there is no evidence of efficacy, and facilitate engagement in life-style modification.


3.6.3. Diagnostic medial branch blocks should not be conducted solely to gather “evidence” in a medicolegal matter.  While a positive diagnosis may be seen to be helpful, careful thought should be given to the possible pejorative legal misinterpretations of negative or indeterminate responses. 

3.7. It is recommended that the following information be conveyed to patients in order to ensure informed consent:

3.7.1. The diagnostic process – patients should understand the likely number and types of diagnostic procedures, the intervals between procedures and waiting time for treatment should a positive diagnosis be made.


3.7.2. The procedural experience – patients need to understand the setting, equipment and personnel to be present, what the procedure will feel like, what will be expected of them, likely common side-effects, risks of material adverse events and their management.


3.7.3. The method of assessing outcome – patients need to understand that their experience of pain and their report of this experience constitute the primary means by which the diagnostic outcome is determined.  They need to be familiarised with and competent to use the outcome measures (see 5.3 below).

Implications of positive diagnosis – whilst the patient must be informed about treatment targeting this diagnosis, such as radiofrequency neurotomy ( RFN). Information about RFN should encompass the substance and limitations of published evidence and yield realistic expectations of therapy. 


3.7.4. Implications of negative or indeterminate outcomes – patients need to understand that if results are not positive then their therapeutic options will not include treatment targeting the medial branches or facet joints. 

4. Conduct of medial branch blocks

4.1. The proceduralist should have appropriate knowledge, skills and experience relevant to the performance of cervical and lumbar diagnostic medial branch blocks.  In particular they should:


4.1.1. be aware of the history and evidence base for these procedures;6


4.1.2. be able to clinically assess spinal pain;


4.1.3. be able to identify indications and contra-indications for these procedures;6


4.1.4. know the anatomy of the cervical and lumbar medial branches;1,3


4.1.5. know the radiographic anatomy of the cervical and lumbar spines in health and disease;6


4.1.6. know the radiographic target points for the medial branches;6


4.1.7. know the principles of radiation safety;


4.1.8. know the pharmacology and adverse effects of the agents used;


4.1.9. have skills in aseptic preparation of procedural sites and infection control measures;


4.1.10. have skills in safely and accurately directing needles under radiographic guidance;


4.1.11. have skills in recognising and managing potential adverse events including skills in cardio-pulmonary resuscitation;


4.1.12. understand the principles and interpretation of comparative local anaesthetic blocks;6


4.1.13. have appropriate supervised experience in the conduct of these procedures before undertaking independent practice.

4.2. The facility in which diagnostic medial branch blocks are to be performed should afford all of the following:


4.2.1. a sterile procedure room that complies with radiation safety and relevant occupational health and safety standards;


4.2.2. appropriate supply and storage of sterile goods including gowns, gloves, drapes, needles and other sterile disposables;


4.2.3. appropriate supply and storage of medications including solutions for aseptic hand and skin preparation, local anaesthetic agents, non-ionic radiographic contrast medium, and drugs used in the management of seizures and cardiopulmonary resuscitation;


4.2.4. a fluoroscope (preferably with a C-arm), a qualified fluoroscope operator and a means of documenting accurate placement of needles and contrast – either hard-copy film, an image on specialised paper, or a permanent digital image. Although there is emerging interest in computer tomography and ultrasound-guided procedures, there is no published evidence of comparable efficacy and therapeutic utility. Accordingly, at the time of writing, no evidence-based recommendation can be made regarding the latter imaging modalities.


4.2.5. intravenous access should occur on all patients with ready access to resuscitation equipment and personnel skilled in cardiopulmonary resuscitation and advanced life support;


4.2.6. a supervised recovery area. 

4.3. Despite pre-assessment and consent (Section 3 above), a certain amount of patient preparation is still required on the day of procedure:


4.3.1. A brief history is required immediately prior to the procedure to confirm the persistence, location, and current severity of the index pain and record baseline measures;


4.3.2. The current pain should be sufficiently intense and constant such that any response to the diagnostic block will be more credibly attributable to the block than to chance fluctuation alone. A reasonable guide is that the patient’s present pain should be at least 50% of their pain at its worst and at least 20 points on a 100 point visual analogue scale; 6


4.3.3. The patient should be reminded that the procedure is a diagnostic test, that they may or may not obtain pain relief, and that if relief ensues they will be expected to monitor and report the extent and duration of any relief.


4.3.4. There is no requirement for fasting. Basic physiological monitoring will occur and is more important if the patient is sedated for the procedure (see 2.4 above). Nevertheless, proceduralists may employ these measures at their own discretion.


4.4. It is recommended that proceduralists use the technique(s) for which there is evidence (see 2.3 above). A detailed and illustrated description of the recommended techniques – including equipment, positioning, sterility, radiographic target identification, needle placement and performance parameters – can be found in Reference 6.


4.5. Comparative blocks are essential to exclude false-positive responses.  The most rigorous form of comparison is the use of placebo injection of normal saline under double-blind conditions,15 but logistic and ethical considerations mitigate against their use in conventional practice.  Instead, comparative local anaesthetic blocks are the recommended standard practice:4-7,14,15


4.5.1. If the initial block at a given level results in meaningful pain relief, then a second block of the target nerves should be performed on a separate occasion using a local anaesthetic with a different duration of action;


4.5.2. At least the patient and outcome assessor should remain blind to the order of administration of the local anaesthetics until after the series is evaluated;


4.5.3. The time from the procedure until the return to at least 50% of the pre-procedural pain intensity is taken as the duration of pain relief.


4.6. There must be systems for recording the pre- and post-procedural clinical assessments, technical aspects (needle, technique and agent used, radiographic findings, technical challenges), and images of needle positions, analgesic outcomes, adverse effects and care plan.

5. Outcome assessment and interpretation


5.1. The assessment of the patient’s self-reported pain experience following the procedure is pivotal. This assessment continues for the duration of the patient’s response to the block, or until the effects of the block have been reasonably established beyond doubt. That is: 

5.1.1. Once it has been clearly established that no area of the patient’s pain has been relieved, the patient may be prepared for discharge;


5.1.2. If the patient reports relief, this should be monitored for at least 2 hours or until the relief wears off (whichever occurs first).  Thereafter, the patient may be discharged with instructions for recording the ultimate duration of their response and an arrangement to collect this information by phone or other means.

5.2. Unless the diagnostic block is performed under double-blind conditions, it is recommended that the proceduralist delegate outcome assessment to an independent observer who remains blind to the agent injected.6


5.3. The patient’s response should be assessed in a manner that reflects the multiple dimensions of pain.  The following are suggested components of the assessment:

5.3.1. Pain location compared with immediately before the block. Pre- and post-procedure pain maps are helpful in this regard.


5.3.2. Pain intensity measured prospectively on visual analogue scales. Recommended frequency is at least one score before the procedure, on arrival in the recovery area, half-hourly for the first 2 hours, then hourly thereafter until pain returns.


5.2.3. Duration of relief. This is taken to be the time between the last injection of local anaesthetic and return of pain to at least 50% of baseline intensity.


5.3.4. Improvement or lack of improvement of function with nominated activities of daily living (four such activities).


5.3.5. The patient’s global impression or qualitative response to the block recorded in his or her own words.


5.3.6. Presence of expected effects such as numbness in the cutaneous territory of the target nerve (if relevant).


5.3.7. Presence and duration of adverse effects.

5.4. The patient’s response to a given diagnostic medial branch block procedure may fall into four categories:

5.4.1. No relief at all – categorised as a negative response. It can be inferred that the medial branches targeted do not mediate the patient’s pain in whole or part.


5.4.2. Complete relief of all pain – categorised as a positive response. It can be inferred that the medial branches targeted mediate the patient’s pain.


5.4.3. Complete relief in one topographical area of the patient’s pain with no relief in other areas – categorised as a partial response. It may be hypothesised that the patient has more than one source of pain (an adjacent or contralateral joint or other source) and this hypothesis can then be tested separately.


5.4.4. Fractional relief of all pain – categorised as indeterminate.  If relief is neither complete nor almost complete (allowing for some post-procedural pain), then it cannot be inferred that the target nerves mediate the patient’s pain. 

5.5. The patient’s responses to a series of comparative local anaesthetic blocks may fall into three categories:

5.5.1. Discrepant – the patient has a positive response to the first block but a negative or indeterminate response to the second block of the same target nerves. The series is considered negative;


5.5.2. Concordant – the patient reports longer-lasting relief with the longer-acting local anaesthetic and shorter-lasting relief with the shorter-acting local anaesthetic, with both periods of relief being consistent with the expected duration of action of the agents. If relief following one or both blocks exceeds the expected duration of the action of the agents then the response is qualified as “prolonged concordant”. This constitutes a positive diagnostic series;


5.5.3. Discordant – the patient reports longer-lasting relief following the shorter-acting local anaesthetic and vice versa. This can also be qualified as “prolonged discordant” if relief following one or other agent exceeds its expected duration of action. Under double-blind conditions, the majority of patients who had discordant responses did not respond to a normal saline injection,15 so these responses are considered positive.

5.6. Proceduralists may consider using published Diagnostic Investigation Algorithms such as those published by the International Spinal Injection Society (Reference 6 pp 87-94 for lumbar and pp 152-160 for cervical pain) to help guide decisions and planning regarding initial and subsequent diagnostic blocks.6


5.7. When a series of diagnostic blocks is completed, irrespective of the outcome, the results and interpretation must be communicated to a practitioner responsible for ongoing care and therapeutic planning.


5.8. If the proceduralist does not refer to an independent practitioner to plan therapy, it is important that the proceduralist has processes such as double-blind, or independently assessed blocks to reduce observer bias and, hence, conflict of interest.


5.9. After a therapeutic plan has been implemented, it is important to reassess the patient at regular intervals to establish any residual pain, physical disability and psychosocial issues and to plan supplementary management accordingly. 


  1. Bogduk N. The clinical anatomy of the cervical dorsal rami. Spine 1982;7:319-330.
  2. Barnsley L, Bogduk N. Medial branch blocks are specific for the diagnosis of zygapophysial joint pain. Regional Anesthesia 1993;18:343-350.
  3. Bogduk N, Wilson AS, Tynan W. The human lumbar dorsal rami. J Anat 1982;134:383-397.
  4. Sehgal N, Dunbar EE, Shah RV, Colson J. Systematic review of diagnostic utility of facet (zygapophysial) joint injections in chronic spinal pain: an update. Pain Physician 2007;10 (1):213-228.
  5. Boswell MV, Colson JD, Sehgal N, et al. A systematic review of therapeutic facet joint interventions in chronic spinal pain. Pain Physician 2007;10(1):229-253.
  6. Standards Committee of the International Spinal Intervention Society. Practice Guidelines for Spinal Diagnostic and Treatment Procedures. Edited by Bogduk N. International Spinal Intervention Society, San Francisco, 2004.
  7. Accident Compensation Corporation (ACC). Clinical Guidance on Interventional Pain Management. Published on World Wide Web 17 October 2005 at
  8. Lord SM, Barnsley L, Wallis BJ, McDonald GJ, Bogduk N. Percutaneous radiofrequency neurotomy for chronic cervical zygapophysial joint pain. N Engl J Med 1996;335:1721-1726.
  9. McDonald GJ, Lord SM, Bogduk N. Long-term follow-up of patients treated with cervical radiofrequency neurotomy for chronic neck pain. Neurosurgery 1999;45:61-68.
  10. Dreyfuss P, Halbrook B, Pauza K, Joshi A, McLarty J, Bogduk N. Efficacy and validity of radiofrequency neurotomy for chronic lumbar zygapophysial joint pain. Spine 2000;25:1270-1277.
  11. Niemisto L, Kalso E, Malmivaara A, Seitsalo S, Hurri H. Radiofrequency denervation for neck and back pain. The Cochrane Database of Systematic Reviews Volume (4), 2006 (last updated 23/08/2005).
  12. Chua WH, Bogduk N. The surgical anatomy of thoracic facet denervation. Acta Neurochir 1995;136:140-144.
  13. Manchikanti L, Singh V, Pampati V, Beyer CD, Damron KS. Evaluation of the prevalence of facet joint pain in chronic thoracic pain. Pain Physician 2002;5:354-359.
  14. Barnsley L, Lord SM, Bogduk N. Comparative local anaesthetic blocks in the diagnosis of cervical zygapophysial joint pain. Pain 1993;55:99-106.
  15. Lord SM, Barnsley L, Bogduk N. The utility of comparative local anaesthetic blocks versus placebo-controlled blocks for the diagnosis of cervical zygapophysial joint pain. Clin J Pain 1995;11:208-213.
  16. Bogduk N, Holmes S. Controlled zygapophysial joint blocks: the travesty of cost-effectiveness. Pain Med 2000;1:25-34.
  17. Task Force on Taxonomy of the International Association for the Study of Pain. Classification of Chronic Pain, Second Edition. Edited by Merskey H, Bogduk N. IASP Press, Seattle, 1994.
  18. Lord SM, McDonald GJ, Bogduk N. Review article: Percutaneous radiofrequency neurotomy of the cervical medial branches - a validated treatment for cervical zygapophysial joint pain. Neurosurgery Quarterly 1998;8:288-308.
Copyright © Australian and New Zealand College of Anaesthetists.