Access to the PPM Journal and newsletters is FREE for clinicians.
8 Articles in Volume 5, Issue #6
Botox Treatment of Chronic Refractory Low Back Pain
DEA Enforcement versus Pain Practice
Group Psychotherapy for Chronic Pain Patients
How Expert Testimony Distorts the Standard of Care
Neurostimulation in Chronic Pain Patients
Physiological Consequences of Guided Imagery
The Role of Tertiary Gain in Pain Disability
Treating Muscular Dysfunction of Upper Limbs

Botox Treatment of Chronic Refractory Low Back Pain

Studies indicate that a specific treatment protocol using Botulinum Neurotoxin A can provide a sustained reduction in chronic, refractory low back pain for about 50% of such patients.

Low back pain is the second cause of physician’s office visits in the United States, second only to respiratory tract infections.1 Chronic low back pain is a major health issue challenging the practicing physician more than ever before, especially since the recent withdrawal of several Cox 2 inhibitors from the market due to unacceptable side effects.

In this article, the author discusses experience with Botulinum Toxin A (Botox-Allergan Inc) in treating chronic, refractory low back pain. Two studies, described below, indicate pain relief in 50% of patients treated with this protocol. The treatment protocol is described in detail following the study results.

The following two studies are the only prospective studies that have evaluated the effect of Botulinum Toxin A in chronic, refractory low back pain:

Study 1

A randomized, placebo controlled, double blind study2

Between the years 1998 and 2000, the author conducted a randomized, placebo controlled double blind study in 31 adult patients with chronic (>six months), unilateral low back pain. This study was conducted based on reported human observations and animal data which indicated an analgesic effect for Botox. In particular, several randomized, blinded studies of human subjects had shown that injecting Botox into neck muscles significantly reduced neck pain.3-5 Recent animal data have also suggested a variety of analgesic functions for Botox including:

  1. reduction of muscle spasm due to blocking acetylcholine release from pre-synaptic vesicles,
  2. decreased release of pain neurotransmitters (substance P and bradykinin) from the dorsal root ganglia,6
  3. reduction of muscle spindle discharge leading to decreased central sensitization,7 and
  4. decreased discharge of sympathetic neurons8 possibly interrupting the role of sympathetic nervous system in maintaining the chronic pain.

In the author’s study, the level of pain and its influence on the activities of daily living was assessed in each patient using the visual analog scale (VAS) and the Oswestry Low Back Pain Questionnaire (OLBPQ). Values were documented at baseline, 3 weeks, and 2 months after Botox injection.

Study 1 Results: Results were interpreted as a significant response if there was either a 50% or more decrease in pain intensity depicted by VAS, or a two step or more improvement in at least one function (sitting, standing, walking, etc) in OLBPQ in addition to the improvement of a pain subset. Note that OLBPQ has 10 subsets, each with six steps ranging from normal function to total dysfunction. The total injected dose was 200 units/patient. The details of Botox preparation and technique of injection—similar in both blinded and prospective studies—is presented at the end of this article.

Twenty-eight of 31 patients completed the two-month assessment. At two months, 9 of 15 (60%) patients in the botox group had a significant reduction in pain intensity (VAS) versus 2 of 16 (18.8%) in the control (saline) group (p=0.011). An improvement in the activities of daily living (OLBPQ) was noted in 10 of 15 (66%) of the Botox group versus 3 of 16 (18.8%) in the saline group (P=0.009). No patient experienced side effects.

Table 1. Mean VAS max of Botox responders over a period of 12 months Pain intensity(VAS) is represented from 0-100mm. Mean VAS scores: baseline=82 mm, 1 and 2 months =50mm, 12 months = 40mm. Table 2. Mean Pain days in the preceding month for baseline (b), 1, 2, 4, 6, 8, 10 and 12 months

Study 2

Prospective 14 month study evaluating benefits and safety of repeated Botox treatment of paraspinal muscles in chronic, refractory low back pain

This prospective long-term study began in 2002 and, like the first study, was conducted at Walter Reed Army Medial Center, Washington, DC. The plan of this study and its first six months results (with assessment after second treatment) were detailed in a previous communication.9 In brief, the effect of Botox was assessed on chronic low back pain (> 6 months) in 75 patients. The technique of Botox administration was similar to the aforementioned blinded study (see below). All patients had an MRI to exclude a pathology necessitating urgent surgical intervention. Each of the following rating scales were used at baseline, and at 1, 2, 4, 6, 8, 10, 12 and 14 months:

  • pain intensity: VAS (Visual analog scale)
  • pain frequency: PIQ (Pain impact questionnnaire)
  • activities of daily living OLBPQ (Oswestry low back pain questionnaire)
  • patient’s perception of change PIQ

Botox was injected after the baseline evaluation and at four month intervals. However, if patient was a responder and at four months reported continued significant pain relief, treatment was deferred for two months pending the results of the next evaluation.

Study 2 Results. A significant initial (at one month) response was noted in 39 of 75 patients (53%). Of 39 responders, 33 completed the 12 month follow up. One of the 33 became a non-responder after the second injection. The remaining 32 continued a favorable response with repeat treatments over a period of 12 months.

A student’s t-test was used to assess the statistical difference between means at baseline and following treatments. The difference was statistically significant (P

In 33% of responders, the initial response to Botox lasted more than 4 months. In one patient a single treatment resulted in sustained relief for 12 months. Three patients experienced side effects. In all three it was a mild flu-like reaction which lasted from 2-5 days.

More Recent Animal Studies

Among the these studies, the work of Cui et al10 is of special interest. These investigators studied the effect of Botox pre-treatment on the formalin model of pain. Injection of formalin into the paw of a rat produces a two-peak pain response: the first one being related to irritation of C-fibers and the second (more intense) due to inflammation and accumulation of glutamate. Pretreatment of the paw with Botox a week prior to injection of formalin, reduced the inflammatory pain significantly in a dose-related fashion and reduced accumulation of glutamate. Figure 1 shows the effect of pretreatment with saline and two doses of Botox on the inflammatory pain caused by formalin injection.

Treatment Protocol

Botulinum toxin A was prepared by combining vacuum-dried BoNT-A with preservative-free 0.9% saline to a concentration of 100 units/ml. Injections were made using a 1 cc -tuberculin syringe, unilaterally or bilaterally based on the predominant pattern of pain distribution.

The first site of injection selected was at the vertebral level of most intense pain (defined by the patient and physician using deep finger pressure). Subsequent injection sites extended to at least one (and often two) levels above and below the pain location. Thus, most patients were injected at five sites into para-spinal muscles between L1 to S1 vertebral levels (see Figure 2). The rational for this methodology is to cover the length of the para-spinal muscles as much as possible since superficial recti muscles are long and extend over several vertebral levels. Furthermore, it is known that in patients with low back pain, the source of pain may be located a few segments higher than the level of perceived pain. When the area of pain extended laterally, one additional injection (same dose) was administered at the same level to a more lateral aspect of the para-spinal muscles.

In thin patients, with little adipose tissue in the back, a ¾ inch needle with penetration to the needle hub was used to deliver the entire site dose in one depot. In patients with more abundant adipose tissue in the back, a 1.5 inch needle was used and the drug was injected at the depth of 1 or 1.5 inch. Injections were performed without electromyographic guidance. The dose per injection site was 50 units (40 units for very thin patients). The total dose per session ranged from 200 to 500 units depending on the presence of unilateral or bilateral pain. The reasons for not exceeding 500 units in bilateral pain (300 unilateral pain) are twofold: safety (i.e to minimize any potential CNS effects that might have resulted from spread of the toxin to the spinal cord through nerve roots) and to avoid antibody formation after repeated treatments which has been linked to administration of large doses per session and short intervals between injections.

Figure 1. Animal pain study after pretreatment with Botox, from Cui et al, Pain 2004. Figure 2. Showing the site of injections in a patient with unilateral low back pain. The dose is 40-50 units per site with an additional 40 to 50 units injected more laterally at the level of laterally-extended pain (if present).


In the author’s experience, treatment of low back pain with botulinum toxin A has been found to relieve low back pain in approximately half of the patients. The results of long-term follow up indicate that 90% of the responders continue to respond over a period of 12 months. The side effects are few, mild, and transient. The success may rest entirely on methodology which covers the whole low back area and uses a sufficient dose per site to exert the maximal effect.

At the present time, this treatment is off-label and not FDA approved. Due to the high cost of Botox, this approach should be reserved for refractory patients and performed only by physicians with significant skill and experience in Botox treatment of myofascial pain syndromes.n

Last updated on: January 10, 2012
close X