A review of electromyography techniques of the cervical paraspinal muscles

Electrodiagnosis for cervical radiculopathy often involves exploration of the cervical paraspinal muscles. Accurate and reproducible results require a technique with specific anatomic localization, direction of insertion, extent of insertion, scoring system for insertion, and criteria for determining abnormality. We sought to understand if a published technique met these criteria. A Medline search found 39 articles with original research and 10 review articles involving the cervical paraspinals. A library search found 19 textbooks since 2000, but 9 were not available. Only two studies were specific to the question. Neither had reproducible techniques and they contradicted each other. Studies in which the paraspinals were used for comparison or inclusion did not provide any specific technique. The review articles and textbooks typically met none of our criteria and the few that discussed technique at all provided no reproducible methods. Despite 80 years of electrodiagnostic testing, there is no useful, reproducible technique for exploring the cervical paraspinal muscles. Yet such a paraspinal mapping technique has proven invaluable in the lumbar region. For cervical electromyography to be of value, the next step is to understand the anatomy and propose a reproducible technique. Subsequent research will determine whether the neck muscles are helpful in the diagnosis of cervical radiculopathy. The absence of a valid reproducible cervical paraspinal technique impedes clinical and scientific understanding of cervical radiculopathy.


INTRODUCTION
For most of a century, electrodiagnosis has been used to clarify the differential diagnosis, confirm the presence of cervical radiculopathy, localize the lesion, and grade its severity and duration. 1,2Although there are general standards for the needle examination, there is no single accepted standard approach.Among the important muscles for consideration are the cervical paraspinal muscles.
If experience with the lumbar paraspinals is any indication, it is possible that the cervical paraspinal muscles hold the key to important diagnostic information.][5][6][7][8][9][10] Numerous benefits included the demonstration that the paraspinal muscles are the most sensitive and specific aspect of the electromyography (EMG) examination, research showing that PM is superior to advanced imaging in diagnosing some spinal disorders, and evidence that the use of PM improves spinal procedures and surgical interventions.A valid, standardized cervical paraspinal protocol will improve interpretation of testing at a minimum, but may lead to improved sensitivity, specificity and localization of lesions.
Regarding needle EMG of the cervical paraspinal muscles, there seems to have been little movement down this pathway.The current article reviews publications of the past to determine gaps in our knowledge.
An accompanying article examines current knowledge of the anatomy of the cervical paraspinal muscles and uses this combined information to propose a technique for cervical PM. 11

METHODS
The study involved a review of the medical literature for original research, review articles, and textbook content regarding the cervical spine needle EMG procedure.Methods are fully described in the text.Data collection templates and other information can be obtained from the senior author.
We determined at the onset that a viable technique would require the following information: anatomic location for needle insertion, direction of insertion, extent of insertion, scoring system, and norms or criteria for determining abnormality.
A Medline search was performed in December 2020 on the terms "electrodiagnosis" (n = 471,101), "cervical vertebrae" or "neck" (n = 69,168), and "radiculopathy" (n = 6206).The combination of these resulted in 62 articles.Case reports and irrelevant articles were eliminated (n = 32).Of the remainder, 24 were primary research sources and 6 were reviews without primary data.References from these articles and other work known to the authors was added to these.Review of the 62 articles and of subsequent possible inclusions was performed by both authors, with final decision on inclusion made by the senior author.
Articles presenting primary research were evaluated for the following three criteria: those that made cervical paraspinal muscles an item of study, those that were comparing EMG to some other standard for cervical radiculopathy, and those in which EMG was done as part of inclusion or exclusion criteria for another purpose.Mention of cervical radiculopathy, mention of paraspinals in the discussion of cervical radiculopathy, any mention of a cervical technique, any mention of a scoring system, any indication of reference values, and then actual report of cervical paraspinal findings were recorded.
Review articles that discuss EMG in cervical radiculopathy were also evaluated for mention of the cervical paraspinals, of any specific technique, any scoring system, norms, and any other illustrative comments.
A second search was performed for textbooks of electrodiagnosis published after 2000.This was done through a search of the university online catalogue and the Amazon commercial site.Texts were evaluated for mention of cervical paraspinal EMG, any discussion of the multifidus or paraspinal anatomy, advocacy for any technique, scoring system, norms, or illustrative comments.
Publications were accessed through download from the University of Michigan and the University of Vermont libraries, with Michigan's library known as one of the most extensive collections in the world.Some articles, especially very old articles, were not available online during the pandemic, and some textbooks were not available.These are tabulated in the results.

Original research publications
Thirty-nine articles of original research were found, as listed in Table 1; 12 were not available for review.Of the 12 that were not available, two from the Physical Medicine and Rehabilitation Clinics of North America may well have been reviews without new data.The remaining unavailable articles were citations that were decades old.
Among the rest of the original articles, only two had a primary purpose of examining the cervical paraspinals.Gilad et al. tested 100 asymptomatic volunteers stating, "The needle was inserted into the multifidus muscle at the level of the C6 spinal process, sampling four separate muscle insertion points." 39The exam was considered abnormal if fibrillations lasted more than 0.5 seconds after insertion at two or more of these four sites.They found no spontaneous abnormalities in persons younger than 40 years, 8% in persons age 40 to 60, and 92% in persons older than 60.Gilad et al. did not describe the method of approaching the multifidus in any reproducible way.A second paper by Date et al. mentioned no anatomic technique and no scoring system, but they reported that 40.8% (20 of 49) participants had cervical paraspinal abnormalities. 40oth groups concluded from their data that paraspinal fibrillations were not useful in the diagnosis of radiculopathy, yet the authors made no comparison to persons with actual pathology.
The paraspinals were used as part of comparison to some other variable in 21 publications and used as part of inclusion criteria in five original publications.Two of these made passing mention of technique (Khan et al. investigated "the middle and lower cervical paraspinals," whereas Lauder et al. tested "multiple cervical levels." 13,42).None presented any scoring system or norms by which they judged their findings.Only four mentioned paraspinal results.Dillingham et al. found abnormal paraspinal spontaneous activity in 47% of radiculopathies, and the addition of motor unit neurogenic changes (not quantified or defined) increased this sensitivity to 56%.5][16][17] These articles compared EMG results to other diagnostic tests and sometimes drew their conclusions from as few as three limb muscles.

Review articles
Ten review article citations were found and eight were available for review (see Table 1).Only two of these mentioned a technique for cervical paraspinal EMG.The second part of a recent review by Dillingham et al.  in Muscle & Nerve provides an extensive discussion of the importance of the paraspinal muscles. 32No cervical Han 41 Not available Lauder 42 Not available Johnson 43 Not available Partanen 44 Not available Waylonis 45 Not available Hong 46 Not available Berger 47 Not available Tamura 48 Not available Honet 12 Not available Maranacci 49 Not available Negrin 50 Not available technique is described but the authors instruct clinicians that proper placement occurs when muscle membrane sounds are heard and then the needle contacts bone, presumably the lamina.Fisher wrote, "[Needle EMG] of the paraspinal muscles is usually performed by inserting the needle to the lamina 1-2 cm lateral to a spinous process and then withdrawing slightly." 64utside of our review of the technical literature, many review articles about neck-related pathology were found to make unsubstantiated, but perhaps insightful, claims.For example, Carrette and Feldings stated, "Examination of the paraspinal muscles increases the sensitivity of the test, since insertional activity can be seen as early as 10 days after the nerve injury.In addition, the presence of abnormal findings in paraspinal muscles differentiates cervical radiculopathy from brachial plexopathy." 71ao et al. stated, "[T]he ultimate site of paraspinal muscle innervation may be anywhere from 3 to 6 segments beyond their level of spinal cord origin," and "Fibrillation potentials … may disappear completely in the paraspinal muscles after 6 weeks." 72However, this statement referenced a retrospective study of only the lumbar region by Johnson and Fletcher, done in the 1981. 43Tsao et al. quote the same article by Johnson to declare that fibrillations may disappear in the first 8 weeks. 73,74Johnson and Fletcher mentioned this time frame in their discussion, but provided no data to support the claim.

Textbooks
Textbooks can provide more basic science background in addition to the expertise and experience of the author and any scientific data.Nineteen textbooks were thought appropriate, but nine were not accessible (see Table 2).
Two of the 11 available books did not mention the cervical paraspinals at all.Nine of the 11 did not mention the pertinent muscle anatomy.Five alluded to some technique, but none mentioned a way to score the findings or how to determine if the findings were abnormal.For example, Leis and Trepani and Pohlman et al. both instruct the reader to insert the needle perpendicular to the body at 2 cm lateral to the spinous process, a strategy that would most likely miss the multifidus altogether. 54,55Cooper et al. and Preston and Compte both treat the cervical, thoracic, and lumbar paraspinals in the same way despite clear anatomic differences. 53,57Only Perotto and Delagi, in a book dedicated exclusively to the anatomy of needle placement, approximates what is known about the muscle: "If the electrode is placed in the angle between the lamina of the vertebra and the spinous process, it will be in the multifidus." 51egarding interpretation of findings, three books specifically stated that the paraspinals are not specific to a root. 58,59,62One stated that the multifidus travels down two levels. 65Weiss et al. stated that a Fishman 64 Not available Pease 65 Not available EMG Pearls 66 Not available Binnie 67 Not available Misulis 68 Not available Bertorini 69 Not available Dumitru 70 Not available definitive diagnosis of a radiculopathy requires abnormal paraspinal muscles. 56Siao et al. pointed out that sometimes the paraspinals are the only finding in a radiculopathy. 59

DISCUSSION
This examination of original research, review articles, and textbooks finds that research and teaching about electrodiagnostic testing of the cervical paraspinals is sparse, contradictory, and inadequate.The implications for clinical practice are substantial, yet the recommended next steps are feasible.
The literature search was systematic; however, it is always possible that other original research is hidden in studies that had purposes not directly related to cervical radiculopathy.It is unfortunate that some of the older articles were not available online, but reference to these articles in the more modern literature provides no indication that these studies provide more specific details about technique or scoring.A number of textbooks were not available in the two university libraries and online resources available during the pandemic.
The reporting of Gilad et al. and Date et al., the only two original research reports addressing the issue of cervical paraspinal EMG technique, failed our criteria in many ways: the anatomy targeted, describing a reproducible technique to target that anatomy, the sampling, and the scoring or lack of scoring are some issues to be discussed later. 39,40Dillingham et al., on reviewing these two, declared, "The marked difference in results are difficult to explain." 32one of the other original research trials came close to approximating the criteria that we considered necessary to accept a protocol as valid and useful.These included specifying the anatomic location for needle insertion, direction of insertion, extent of insertion, scoring system for insertion, and criteria for determining abnormality.It is no wonder that data and thus opinions regarding the utility of the cervical paraspinal examination were so highly varied.
Among the review articles and textbooks, there was a paucity of discussion of the cervical paraspinal muscles overall.Opinions varied about the utility of these muscles.In the few instances in which any technique was recommended, the descriptions were limited to one or two of our requirements for an understandable, useful process.Many of the authors seemed to repeat the thoughts of other authors, and some clearly extrapolated from whatever they thought was useful in the lumbar area to the cervical region.
Location, direction, extent of insertion, scoring system, and criteria for abnormality are all missing from almost all of these studies.The few exceptions do not provide enough information for one to reproduce their techniques or scoring.We must conclude that there is no technique that approaches our criteria for a useful cervical paraspinal EMG technique.
The review articles and texts about radiculopathy are markedly limited in the mention of the paraspinals.The writing is replete with dogma from the 80-year history of electrodiagnosis.Some of it is overtly wrong: Long ago research dismissed the idea that "any" spontaneous activity means symptomatic disease, 6,[75][76][77] but this belief is propagated in the reviewed texts.Repeated claims that the paraspinals are useless because everyone has fibrillations, or because older individuals have a higher rate of fibrillations, continue despite research that quantifies the range of normal fibrillations by age, thus proving an abnormal paraspinal EMG highly specific to important disease. 6Some authors indicate that the multifidus is innervated by many segments, citing the finding of fibrillations five or six levels below the bony level of a cervical spinal cord injury, 78 despite our knowledge of the anatomy: Just as the biceps is anatomically below the C6 nerve root the cervical multifidus travel below their origins.
In the lumbar region this dogma has been slowly replaced by a logical cascade of research data-for the betterment of the test.First, a better understanding of the anatomy of the paraspinals as outlined by Bogduk. 79hen particular technique, which included specific locations of insertion, depth and technique of insertion, and scoring system were proposed. 4This was modified with experience. 80,81Normative values were established. 6nter-rater reliability was established 82 and the potential for clinical bias was explored. 83That paraspinal mapping technique was then used in a population of persons with possible radiculopathy in the lumbar spine to determine the range of possible scores. 73[9][10] For the cervical spine, the current article does not yet take the first step in this process.We have found that much of the knowledge regarding cervical paraspinal EMG testing is heavily reliant on anecdotal evidence.Critical questions remain unanswered: How long do fibrillations need to be present to be considered present?What percentage of people without radiculopathy have fibrillations in their cervical paraspinals?What is the innervation of the multifidus muscle?How many insertions and or vectors should be used for cervical paraspinal evaluation?Where should the needle be placed when evaluating the cervical paraspinals?
This status quo is inadequate.A more careful approach may improve the overall usefulness of EMG in the analysis of cervical disorders.The first step is to understand the anatomy, which will allow the development of a technique based on that understanding.This journal-based CME activity is designated for 1.0 AMA PRA Category 1 Credit™.Effective January 2024, learners are no longer required to correctly answer a multiple-choice question to receive CME credit.Completion of an evaluation is required, which can be accessed using this link, https://onlinelearning.aapmr.org/.This activity is FREE to AAPM&R members and available to nonmembers for a nominal fee.CME is available for 3 years after publication date.For assistance with claiming CME for this activity, please contact (847) 737-6000.All financial disclosures and CME information related to this article can be found on the Online Learning Portal (https://onlinelearning.aapmr.org/) prior to accessing the activity.

DISCLOSURES
Dr. Haig is president of Haig Consulting LLC, president of the not-for-profit International Rehabilitation Forum, and holds leadership roles in the International Society for Physical and Rehabilitation Medicine and the Association of Academic Physiatrists.How to cite this article: McGuire T, Haig AJ.A review of electromyography techniques of the cervical paraspinal muscles.PM&R.2024;16(3): 287-294.doi:10.1002/pmrj.13047 Honet et al. found 1 of 42 persons having abnormal cervical paraspinal EMG, Levin et al. claimed 47% sensitivity for radiculopathy diagnosis, and Czyrny et al. found that 40.8% (20 of 49) of persons with presumed radiculopathy had only cervical paraspinal findings.
Cervical paraspinal EMG in the peer-reviewed literature.
T A B L E 1