A Case-Series of Dry Needling as an Immediate Sensory Integration Intervention

ABSTRACT

Background

Chronic low back pain (CLBP) has been associated with altered cortical mapping in the primary somatosensory cortex. Various sensory discrimination treatments have been explored to positively influence CLBP by targeting cortical maps.

Objectives

To determine if dry needling (DN) applied to patients with CLBP would yield changes in two-point discrimination (TPD) and left-right judgment (LRJ) tasks for the low back. Secondary measurements of pain and limited range of motion (ROM) was also assessed.

Methods

A sample of 15 patients with CLBP were treated with DN to their low back. Prior to and immediately after DN, TPD, LRJ tasks, low back pain, spinal ROM, and straight leg raise (SLR) were measured.

Results

Following DN, there was a significant (p < 0.005) improvement in LRJ for low back images in all measures, except accuracy for the right side. TPD significantly improved at the L3 segment with a moderate effect size. A significant improvement was found for pain and trunk ROM after DN with a large effect in changing pain of 3.33 points and improving SLR by 9.0 degrees on average, which exceeds the minimal detectable change of 5.7 degrees.

Conclusions

This is the first study to explore if DN alters TPD and LRJ tasks in patients with CLBP. Results show an immediate significant positive change in TPD and LRJ tasks, as well as pain ratings and movement.

Introduction

It is well-established that low back pain (LBP) affects many people during their lifetime and is associated with significant pain, disability, and healthcare cost [1–3]. Epidemiological data show that approximately 20% of patients with acute/sub-acute LBP develop chronic LBP (CLBP), which accounts for the largest percentage of disability and healthcare cost associated with LBP [1, 4–6]. Furthermore, it is estimated that 10% of patients with LBP account for nearly 90% of the cost associated with LBP [7]. With an estimated financial cost ranging from $560 to $635 billion, low back pain places a significant burden on the society in addition to individuals [8,9]. This has led to various attempts at understanding LBP better, especially CLBP, and developing strategies to help ease the pain and disability.

Traditionally, CLBP has been tied to the biomedical model regarding the health of the underlying tissues including degenerative disc disease, spinal and foraminal stenosis, bulging discs, and more [10]. In this model, the amount and severity of pathological changes are believed to correlate with the amount and severity of pain [11]. From a biopsychosocial perspective, fear avoidance and pain catastrophizing have also been explored as powerful drivers in the development and maintenance of CLBP [12,13]. An emerging area of interest in the development and maintenance of CLBP is structural changes in the brain, especially the primary somatosensory (S1) cortex [14–18]. It is now well established that the physical body of a person is represented in the brain by a network of neurons [14–18]. These neuronal representations of body parts are dynamically maintained, and it has been shown that patients with chronic pain, including CLBP, display different S1 representations than people with no pain [14–16]. Furthermore, with cortical restructuring, these body representations expand or contract, in essence increasing or decreasing the body map representation in the brain, and these changes in the shape and size of the body maps seem to correlate with increased pain and disability [14–16]. It is currently believed that the altered cortical representation of body maps in S1 may be due to neglect because of the decreased use of the painful body part [19]. Neuroimaging studies such as functional magnetic resonance imaging are often used to study alterations in the S1 in patients with pain [14–16]. When neuroimaging is unavailable or cost prohibitive, such as in select research and clinical practice, distorted body maps in CLBP potentially can be tested with two-point discrimination (TPD), localization testing, and left-right judgment tasks (LRJ) [20–23].

This reorganization of body maps occurs quickly. For example, it has been shown that when fingers are webbed together for 30 minutes, cortical maps associated with the fingers change from distinct separate regions into one [24]. This finding has significant clinical importance as it underscores the importance of strategies such as movement, tactile and visual stimulation of the skin, central nervous system, and brain as a means to help maintain S1 representation [25,26]. Based on these neuroplastic changes, treatment approaches may target strategies to help normalize these altered cortical representations of body maps. One approach is graded motor imagery (GMI) [26–29]. GMI is a collective term describing various ‘brain exercises’ and includes normalizing laterality (LRJ of body parts), motor imagery (visualization), mirror therapy, sensory discrimination, sensory integration, graphesthesia and mirror therapy [26–29]. Various studies have shown that these GMI strategies are able to positively influence pain and movement, including CLBP [25–27].

Recent studies for CLBP started exploring if traditional tactile treatments such as manual therapy would yield any positive effects when applied with an emphasis on cortical remapping [25,26,30]. In a case series of patients with CLBP with a median pain duration of 10 years, a 5-minute tactile localization simulating manual therapy intervention yielded an immediate improvement in pain and lumbar flexion, exceeding the minimal detectable and minimal clinical important difference [25]. In a follow-up randomized clinical trial for CLBP, Ref. 30 showed that a neuroplasticity explanation of manual therapy can be understood by patients and yielded superior results compared to the traditional biomechanical explanation of manual therapy [30]. These findings, along with the other GMI-based tactile acuity studies simulating traditional hands-on treatment, have pushed forth the notion that physical treatments such as manual therapy may yield positive effects by targeting the altered body schema [25,26]. Another localization, tactile, and direct stimulation treatment increasingly used is dry needling (DN). DN is an intervention that uses a thin filiform needle to penetrate the skin and stimulate underlying myofascial trigger points, muscular, and connective tissues for the management of neuromusculoskeletal pain and movement impairments [31–33]. DN has been found to be an effective intervention for alleviating pain and disability in individuals with CLBP [32] and is gaining more interest as a nonpharmacological way of treating individuals with musculoskeletal pain [31,32]. Current literature behind the underlying mechanisms of how DN works to achieve these results is variable and not fully known. Some potential mechanisms may be its effect on trigger points, contractures, ischemia and hypoxia, circulation, muscle contraction, endogenous local tissue (i.e. trigger points), peripheral and central sensitization, enhancing endogenous mechanisms, and more [31–33].

In consideration of the recent studies on the effects of manual treatment as a form of sensory integration and discrimination training and the precise/localized effect of DN, the question arose, Can DN similarly influence altered cortical maps associated with pain and disability? The primary goal of this pilot study was to determine if DN applied to patients with CLBP would yield changes in TPD and LRJ. Secondary measurements of pain and limited range of motion (ROM) were also assessed.

Methods

Patients

A convenience sample of 15 consecutive patients with CLBP, pain persisting for >3 months, referred for physical therapy were utilized for the study. All patients provided informed written consent for treatment including DN intervention. Patient demographic data were collected for program review from the chart and deidentified for analysis. Patient intake forms, including medical history, were reviewed for items indicating serious pathology and warranting referral for further diagnostic testing, thereby making patients ineligible for the study. Patients were excluded if they could not read or understand the English language, were under the age of 18 years (minor), had undergone spine surgery, and had any specific movement-based precautions, (e.g. no active spine flexion). Patients were excluded if they had a fear/phobia of needles, lack of sensation over the lower back, and bleeding disorders, were immunecompromised and pregnant, had epilepsy, were allergic to metals, or had any implanted devices/augmentation in the lower back area. Patients were additionally excluded if they had visual impairment that limited their ability to participate in right/left judgment assessment, known right-left confusion, dyslexia, or mental illness.

Measurements

Prior to the start, patients completed a disability survey to assess their level of disability in order to describe the patient cohort for the study:

Disability (Oswestry disability index – ODI)

The ODI is a 10-item questionnaire used to assess different aspects of physical function. Each item is scored from 0 to 5, with higher values representing greater disability. The total score is multiplied by 2 and expressed as a percentage. The ODI has been shown to be a valid and reliable measure of disability related to LBP [34,35]. The minimal clinical important difference (MCID) for the ODI for chronic low back pain patients is reported to be 12.8 [34].

A series of outcome measures were completed before and immediately after DN

LBP (Numeric pain rating scale – NPRS)

LBP was measured with the use of a NPRS, as has been used in various studies on LBP [36–38]. The MCID for the NPRS for acute/sub-acute LBP is reported to be 2.0 [39] and for chronic pain, to be 1.7 [40].

Lumbar flexion

Active trunk forward flexion was measured from the longest finger on the dominant hand to the floor in centimeters (cm) [25,41]. The MDC for active trunk forward flexion has been reported to be 4.5 cm [42].

Straight leg raise (SLR)

SLR was used as a neurodynamic measurement rather than a test of hamstring length. SLR was measured with an inclinometer placed on the tibial crest 5 cm distal to the inferior border of the patella on the most affected leg [41,43,44]. SLR for this study kept the ankle in neutral (90 degrees) with no added dorsiflexion or plantar flexion, per previous studies [41,43,44]. MDC for SLR has been reported as a 5.7 degree difference [42].

Left/right judgment (LRJ)

The Recognise™ application (www.noigroup.com) was placed on an iPad (Apple Computers™) and patients sorted through 40 images of the lower back in various positions with the iPad placed comfortably in front of them on a table. Consistent with previous studies [23], the app settings were set to ‘basic’ with images presented in the upright position. When instructed to begin, participants viewed the image, touched the screen, and selected the appropriate LRJ image interpretation. The participants had a standard 5 second time limit per image. Accuracy for right, left, and total was calculated and recorded for the 40 images. Finally, the average time per image to the nearest hundredth was recorded for each reaction time. Two trials of LRJ were performed with a 2-min break between each trial.

Two-point discrimination (TPD)

Patients sat on a treatment table with the skin of their lower back exposed to allow for screening of protective sensation via Semmes-Weinstein 5.07 monofilament. After protective sensation was established, each patient was evaluated for tactile acuity of the right and left lower back 5 cm laterally for the L3 spinous process. TPD was performed in a descending-ascending method but modified by Catley et al. with only one series of descending-ascending measurements at each location compared to five [21]. TPD testing with the use of one series of descending-ascending measures shows good to excellent inter-rater reliability [45]. TPD was completed with a Carolina®Two-Point Discriminator from the Carolina Biological Supply. No data are available on the MDC for TPD.

Immediately following the treatment intervention, LBP (NPRS), lumbar flexion, LRJ, and TPD were remeasured to determine the immediate therapeutic effect of DN. Pre- and post-treatment measurements were performed by the therapist who provided the DN intervention. After completion of the pretests, DN and post-tests, the attending therapist continued treatment based on the developed plan of care.

Intervention

If during palpatory assessment of the patient’s physical examination, the patient presented with tenderness to palpation over the iliocostalis lumborum and/or underlying lumbar multifidi that was stated to be their concordant pain, DN with either .30 x 75 mm or .30 x 100 mm Red Coral Myotech® dry needle was performed via a single needle to the most tender area. After insertion, the needle was driven to the desired depth and pistoned for 5-sec to elicit a local twitching response. Dry needling was provided directly over the most tender/provocative/symptom reproducing muscle with a direct posterior to anterior approach within 2 finger widths of the spinous process at the selected spinal level. If a local twitch response was observed or the patient’s concordant pain was recreated within the first 5 seconds, the needle was withdrawn. If local twitching was not observed or the patient did not describe symptom recreation/localization, the needle was left to rest for 20 seconds before another bout of pistoning was performed in a coning/scanning fashion to find the target/desired tissue. After the second bout of needling, the needle was left again to rest for 20 second then withdrawn.

Data analysis

Data were analyzed using IBM SPSS Statistics for Windows, version 24 (IBM Corp., Armonk, N.Y., USA). A sample size a priori determination of 15 participants was made utilizing G*power [46] (version 3.1.9.2) for the matched pairs t-test using a two-tailed test with a large effect size (d = 0.8) and 80% power with the alpha at 0.05. Descriptive statistics of the participants were calculated for means, standard deviations, and frequencies. We used paired sample t tests to analyze the changes in the primary variables of interest of LRJ of the low back along with TPD at L3. Secondary measures of pain, lumbar flexion, and SLR pre- and post-intervention changes were also analyzed with paired sample t tests. The effect size for each of the outcomes of interest was calculated using Cohen’s d formula of the difference in the means divided by the pooled standard deviations. Effect size interpretation is based on Portney and Watkins, with 0.2 being a small effect, 0.5 medium effect, and 0.8 for large effect size [47].

Results

Participant characteristics

The sample consisted of 15 consecutive participants who completed both the pre- and post- testing with no dropouts. Demographics data on the sample are provided in Table 1. The participants ranged in age from 27 to 80 years, with a persistent pain symptom duration between 6 months to 25 years.

Table 1.

Participant demographics.

Characteristics	Sample (n = 15)
Mean age (years) ± Sd	52.87 ± 15.96
Gender, female n (%)	9 (60)
Duration of symptoms (years) ± SD	8.53 ± 8.60
Oswestry Disability Index ± SD	47.93 ± 12.26
Leg pain
Yes – right leg, n (%)	3 (20)
Yes – left leg, n (%)	3 (20)
Yes – bilateral, n (%)	9 (60)
Side needled
Right, n (%)	6 (40)
Left, n (%)	4 (27)
Bilateral, n (%)	5 (33)

SD = standard deviation

The participants showed a significant improvement in their LRJ for low back images in all measures after dry needling except that accuracy for the right side did not reach the significance level. (Table 2) Interestingly, none of the subjects showed any deficits of LRJ norms prior to DN intervention, but they still showed improvement in their scores. Reaction time on both sides demonstrated a moderate effect size change postintervention from preintervention levels, whereas the accuracy score showed a large effect on the left but a small effect on the right. Participants also showed significant improvement in their tactile acuity as measured with TPD at the L3 segment at the moderate effect size (Table 2).

Table 2.

Dry needling change on mental imagery and tactile acuity.

	Mean (SD)	Mean Difference (p-value)	95% Confidence Interval of the Difference	Effect size (d)
Accuracy R pretest	93.20 (8.34)	2.80 (p = .059)	−0.12–5.72	0.36
Accuracy R post-test	96.00 (6.84)
Accuracy L pretest	91.13 (7.58)	5.60 (p = .001)	2.86–8.34	0.92
Accuracy L post-test	96.73 (4.08)
RT R pretest	1.58 (0.31)	−0.17 (p = .002)	−0.27 – −0.08	0.51
RT R post-test	1.41 (0.35)
RT L pretest	1.61 (0.42)	−0.21 (p = .002)	−0.33 – −0.09	0.52
RT L post-test	1.40 (0.38)
TPD pretest	63.73 (16.56)	−7.67 (p = .003)	−12.30 – −3.04	0.49
TPD post-test	56.07 (14.41)

R = right, L = left, RT = reaction time, TPD = two-point discrimination, SD = standard deviation

A significant improvement was found on the secondary measures of pain and ROM for the participants after treatment of dry needling. (Table 3) A large effect was found in changing pain of 3.33 points on the NPRS along with improving SLR by 9.0 degrees on average, exceeding the MDC of 5.7 degrees. A more moderate effect was demonstrated with changing lumbar ROM upon reaching fingertips to floor 4.07 cm further after intervention. Four patients exceeded the MDC for active trunk flexion (4.5 cm), but the overall improvement for the group failed to meet/exceed MDC.

Table 3.

Dry needling change on pain and ROM.

	Mean (SD)	Mean Difference (p-value)	95% Confidence Interval of the Difference	Effect size (d)
NPRS pretest	5.60 (1.99)	−3.33 (<.001)	−3.73 – −2.93	1.63
NPRS post-test	2.27 (2.09)
Lumbar flexion pretest	16.67 (9.51)	−4.07 (p = .001)	−6.13 – −2.00	0.45
Lumbar flexion post-test	12.60 (8.76)
SLR pretest	73.33 (10.34)	9.00 (p < .001)	6.77–11.24	0.83
SLR post-test	82.33 (11.29)

NRS = numeric rating scale, SLR = straight leg raise, SD = standard deviation

Discussion

To the best of our knowledge, this is the first study exploring the effect of DN on TPD and LRJ tasks in patients with CLBP. The results of this study match other studies showing immediate improvements on pain and physical movements [25,26,30]. In addition to those findings, TPD and LRJ tasks also improved, which may be due to changes with either peripheral and/or central sensory processing with the DN intervention.

The significant changes in TPD following DN may provide some insight into potential mechanisms on the efficacy behind the effective outcomes with DN. If the tactile acuity changes are due to body representation deficits being altered by the DN intervention through sensory discrimination training, it would open a new door for the potential study into the mechanism for which DN may provide effective results [18]. These findings need to be viewed with caution since these subjects did not show a large variation from normal TPD measures even before treatment, with many subjects being within normal ranges even before intervention. This brings into question how many individuals with CLBP have true tactile acuity changes and deficits as a result of their CLBP. Future research will need to explore potentially a subset of individuals with CLBP who have tactile acuity deficits as a primary deficit. This result concurs with the study by Wand et al., which showed that acupuncture applied as a sensory discrimination training tool decreases movement-related pain in patients with CLBP more than acupuncture alone [18]. In lieu of the immediate significant changes in TPD after DN, along with the results from the study by Wand et al., a framework is being built whereby studies should further explore if DN could be seen as a form of sensory discrimination training.

Additionally, to the best of our knowledge, this is the first study to explore the effect of DN on LRJ tasks. Although the patient cohort was not above the normal ranges for LRJ (80% accuracy and speed > 2.5 seconds), DN resulted in immediate, significant improvements in LRJ tasks for low back images in all measures except accuracy for the right side. The results from the LRJ tasks, along with the TPD results and Wand’s sensory discrimination findings, once again allude to the possibility that DN may indeed impact cortical maps, which are known to feature in the development and maintenance of a pain experience [18].

Strengths, limitations, and future research

The primary strength of this study is the fact that it is the first study to explore DN’s effect on TPD and LRJ tasks of the low back. Furthermore, the immediate significant reduction in pain and improved movement concur with current evidence that DN is a valuable, nonpharmacological treatment for people suffering from LBP, especially CLBP.

This study has numerous limitations. First, the design of a program review case series limits conclusions since there was no control group to compare the efficacy of DN. Second, the results only report on immediate postintervention findings and no long-term follow-up. In addition, the a priori subject calculation was for large effect sizes, and some of the results showed only moderate effect sizes, so the study may be underpowered. Another significant limitation is that the same therapist who provided the intervention performed the pre- and post-test measures, thus eliminating blinding. Furthermore, the CLBP cohort for this study was a heterogeneous convenience sample, and as such, the generalizability of the results may be limited. Finally, the cohort of patients did not necessarily have abnormal TPD and/or LRJ tasks to start with. Also, potential reasons for improvement in TPD and LRJ may be due to training effects and subjects improving due to repeat testing with the short pre- and post-test design.

Future studies should further explore the notion of DN as a form of sensory discrimination and integration. Neuroimaging studies could be used to directly scan the S1 region of the brain before and after DN to assess alterations in body maps. Clinical studies should aim to include control groups, longer outcome tracking, and potential matching of patients with more pronounced deficits in LRJ and tactile acuity.

Conclusion

This is the first study to explore if DN alters TPD and LRJ tasks in patients with CLBP. Results show an immediate within-subjects positive change in TPD and LRJ tasks as well as pain ratings and movement. More research is needed in the form of randomized clinical trials to determine if DN may be a viable intervention to supplement treatment of CLBP.

Biographies

Matt O’Neill is a staff physical therapist at Catawba Valley Medical Center. He is board certified in orthopedics and has completed fellowship training in pain science.

Adriaan Louw is a senior faculty, pain science director and vice-president of faculty experience for Evidence in Motion. He has completed a master’s degree and PhD in physiotherapy and is adjunt faculty at St. Ambrose University and the University of Nevada Las Vegas.

Jessie Podolak is a physical therapist and owner of Phileo Health. She is a senior faculty member with Evidence in Motion and has completed fellowship training in pain science.

Nicholas Maiers is a phyiscal therapist and faculty clinician at Des Moines University. He has completed fellowship training in pain science.

Terry Cox is a physical therapist and faculty member at Southwest Baptist University. He is board certified in orthopedic physical therapy and has completed a fellowship in orthopaedic manual physical therapy.

Kory Zimney is a physical therapist and faculty member at the University of South Dakota. Kory completed a PhD in physical therapy from Nova Southeastern University.

Funding Statement

The author(s) reported that there is no funding associated with the work featured in this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Ethics

This study was approved by the Internal Review Board of Southwest Baptist University in Bolivar, Mo.