Botulinum Toxin A Does Not Decrease Calf Pain or Improve ROM During Limb Lengthening: A Randomized Trial

Abstract

Background

During lower limb lengthening, distraction-induced muscle pain and surrounding joint contractures are frustrating complications for which few effective treatments are available.

Questions/purposes

We evaluated Botulinum Toxin Type A (BtX-A) injection in the calf muscles during human tibial distraction osteogenesis. We hypothesized that it may decrease calf pain and increase ROM of the surrounding joints by reducing muscle stiffness.

Methods

Between April 2010 and January 2011, we evaluated 36 patients undergoing bilateral tibia lengthening who met prespecified inclusion criteria. All patients underwent stature lengthening with lengthening over a nail or lengthening and then nailing. BtX-A (200 IU) was injected at the calf muscle only in one leg for each patient and the same amount of sterile normal saline was injected into the other leg as a control. Selection of the leg receiving the toxin was randomized. Clinical evaluation included a VAS score for calf pain and measurement of ROM of the knees and ankles and calf circumference, with evaluations performed in a double-blinded manner. Side-to-side differences were analyzed until the end of consolidation phase. Minimum followup was 24 months (mean, 30 months; range, 24–39 months). The distraction rate and the final length gain were similar in the treated and control limbs. A priori power analysis suggested that 34 legs were required to achieve statistical significance of 0.05 with 80% of power to detect a 50% difference in treatment effect between treatment and control groups.

Results

There were no differences in calf pain, knee and ankle ROM, and maximal calf circumferences between the two legs at each time point.

Conclusions

Local injection of 200 IU BtX-A at the human calf muscle does not appear to reduce calf pain or help enhance ROM of the knee and ankle during tibial lengthening. However, the small sample size provided sufficient power to detect only relatively large clinical effects; future, larger trials will be needed to determine whether smaller differences are present.

Level of Evidence

Level II, therapeutic study. See Instructions for Authors for a complete description of levels of evidence.

Introduction

Soft tissue problems are one of the major concerns in human distraction osteogenesis. Inadequate adaptation of soft tissues and strength imbalance between the agonist and the antagonist muscles can lead to stiffness of the muscles and tendons, resulting in joint contractures and pain [8, 9, 15, 17, 25]. Distraction-induced muscle pain results in increased analgesic use in patients and even aggravates surrounding joint contractures [9, 15, 26], which then call for strenuous physiotherapy and sometimes even surgical release of the contracture.

Botulinum Toxin Type A (BtX-A) is known to block cholinergic transport at the neuromuscular junction by preventing the release of acetylcholine [19]. It partially denervates the muscle and thus decreases muscle strength and tone. Also, it has been known to alleviate pain through several pathways [1, 2, 5, 11, 14, 23, 31]. It has its maximal effect at about 6 weeks after injection and the effect lasts for about 12 weeks [6]. The local injection of BtX-A has been useful in treating spasticity after stroke or in cerebral palsy, increasing joint ROM by reducing spastic muscle tone and reducing spasticity-related local pain [4, 24, 28, 29, 31]. While a couple of animal studies on the efficacy of this toxin in distraction osteogenesis have been reported [18, 19], one suggesting that the use of this drug in the gastrocnemius before distraction decreased the amount of ankle equinus contracture [19], there has been only one human pilot study on tibial lengthening. That report suggested a positive trend toward enhanced ROM and reduced pain in patients injected with BtX-A, but it did not achieve statistical significance [11].

We therefore determined the functional effect of BtX-A injection at the calf muscle in human tibial distraction osteogenesis in the setting of a properly controlled randomized, double-blind trial. Our hypothesis is that local injection of BtX-A at the calf muscle at the time of distraction osteogenesis increases the ROM of the surrounding joints by reducing calf muscle tone and decrease calf pain, especially during the distraction phase.

Patients and Methods

This study was designed as a randomized, double-blind, placebo-controlled trial. Inclusion criteria were as follows: (1) skeletally mature patients younger than 40 years, (2) no history of medical illness precluding the planned surgery, fracture, soft tissue compromise, bony deformities, or infections of the lower extremity, (3) bilateral tibias that called for similar amounts of lengthening, and (4) deformities that would be amenable to lengthening using the same technique in both legs (either lengthening over nail or lengthening and then nailing). There was no exclusion based on the amount of lengthening planned. Initially, 40 patients who met these criteria were included in the study, but four patients were excluded: two had insufficient radiographic evaluations and two were lost to followup; this left 36 patients with 72 segments of tibia for evaluation. For each patient, BtX-A (200 IU) was injected at the calf muscle in one leg and the same amount of sterile normal saline was injected into the other leg as a control; thus, the BtX-A and placebo groups consisted of 36 legs each.

Each patient underwent bilateral tibial lengthening for familial short stature from April 2010 to January 2011. The mean preoperative height and age were 155 cm (range, 143–165 cm) and 23 years (range, 16–35 years), respectively (Table 1). The minimum followup was 24 months (mean, 30 months; range, 24–39 months). This study was approved by the institutional review board of the author’s institution (BD2011-025D).

Table 1.

Demographic data of patients undergoing bilateral tibial lengthening

Variable	Value
Number of patients	36
Number of tibial segments	72
Male:female (number of tibia)	62:10
Age at time of surgery (years)*	23 (16–35)
BMI (kg/cm2)*	22 (18–27)
Followup (months)*	30 (24–39)

* Values are expressed as mean, with range in parentheses.

All surgeries were performed by the senior author (DHL). The surgical procedure was performed under general anesthesia using techniques based on Herzenberg and Paley [12] for the lengthening over nail technique and Rozbruch et al. [21] for the lengthening and then nailing technique. After detailed explanation of the pros and cons for each surgical procedure, we gave each patient a chance to choose between the two surgical techniques. This was possible because there was no specific tibial deformity except proximal tibia vara in our patient cohort. Each patient was treated with either lengthening over nail (10 patients) or lengthening and then nailing (26 patients), but no patient had a different procedure on each limb.

The injection was placed intraoperatively at the end of all surgical procedures. The 200 IU BtX-A (BOTOX^® Purified Neurotoxin Complex; Allergan, Inc, Irvine, CA, USA) was mixed with 20 mL normal saline. The same amount of sterile normal saline was prepared for the other leg. Selection of the leg receiving the toxin was decided by a toss of a coin. The injection was placed at six different spots evenly at the gastrocnemius and soleus muscles (Fig. 1). Injection was performed manually with no help of instrumentation such as EMG or ultrasound.

Fig. 1.

BtX-A was injected at six different spots (arrow) in the gastrocnemius and soleus muscles.

After the first-stage surgery, each patient had a 7- to 10-day latent period and then entered the distraction period at a rate of about 1 mm/day. The second-stage surgery, completion of internal fixation and removal of the external fixator, followed after the desired length and acceptable limb alignment were achieved. Every patient performed physical exercise as tolerated under a standardized exercise regimen with the help of the same physical therapist (JHP) who specializes in limb lengthening. The patients were allowed full weightbearing anytime during the distraction phase and after removal of the external fixator if there was radiographic evidence of two cortical healings in lengthening over nail and one cortical healing in lengthening and then nailing. Patients were followed up every 2 weeks during the distraction phase and every month thereafter until the end of the consolidation phase.

The mean distraction rate was 0.69 mm/day (range, 0.50–0.96 mm/day) in the BtX-A group and 0.68 mm/day (range, 0.51–0.97 mm/day) in the placebo group (p = 0.88). The mean final tibial length gain was 64 mm (range, 45–81 mm) in the BtX-A group and 64 mm (range, 43–81 mm) in the placebo group (p = 0.91).

For pain management, intravenous patient-controlled analgesia (sufentanil citrate 430 μg mixed with 100 mL normal saline; BC World Pharm Co Ltd, Yeoju, Republic of Korea) was installed at the end of surgery and was maintained until Postoperative Day 2, at which point one tablet of Ultracet^® (acetaminophen 325 mg plus tramadol hydrochloride 37.5 mg; Janssen Korea Ltd, Seoul, Republic of Korea) was administered in a routine manner two times a day. Intravenous Zipan^® (tramadol 50 mg; Ilsung Pharmaceuticals Co Ltd, Seoul, Republic of Korea) was injected when needed and limited to three times a day. For severe pain, Targin^® (oxycodone hydrochloride hydrate 5.25 mg plus naloxone hydrochloride dehydrate 2.73 mg; Mundipharma Korea Ltd, Seoul, Republic of Korea) was given as a rescue medication.

Clinical assessments were performed by one of two orthopaedic residents (KJR, BHK) who were blinded to the patients’ medical records and were recorded on standard data collection sheets. Evaluations included ROM of the knees and ankles, including extension and flexion for each joint measured using a hand-held goniometer, and pain over the gastrocnemius-soleus and tibialis anterior muscle areas based on a 10-point pain VAS. We administered a pain questionnaire, on which each patient was asked to report VAS scores for pain in each leg and the location of pain (anterior shin or calf). The maximal calf circumference (in millimeters) was manually measured with a tape measure. Then, the side-to-side differences were analyzed at each followup. All patients were asked whether they had any injection-related complication, including a flulike syndrome, dysphagia, dry mouth, headache, or nausea, during the followup.

At the beginning of the study, we performed an a priori power analysis for three of the clinical variables we intended to measure, ie, knee and ankle ROM, pain in the gastrocnemius-soleus/tibialis anterior muscle area, and maximal calf circumference, and found that a minimum sample size of 34 legs was required to achieve statistical significance of 0.05 with 80% of power at an effect size of 0.5 (that is, 80% power to detect a 50% of difference in effect size between the treatment and placebo groups); thus, our sample size of 36 legs in each group was sufficient.

All three continuous measurements (knee and ankle ROM, lower leg pain, and maximal calf circumference) were tested for normality using the Shapiro-Wilk test and did not violate the normal distribution assumption. Side-to-side differences (BtX-A group versus placebo group) were analyzed at each followup using the paired t-test. A p value of less than 0.05 was considered statistically significant. The statistical software R (Version 2.12; The R Project for Statistical Computing, Vienna, Austria) was used for all statistical analyses.

Results

Results for the BtX-A group and the placebo group are summarized (Table 2). Knee ROM, including extension and flexion, showed no differences between groups at each time point (Fig. 2). Likewise, ankle ROM, including dorsi- and plantarflexion, showed no differences between groups at each time point (Fig. 3).

Table 2.

Comparison of results between the BtX-A and placebo groups

Variable	BtX-A group	Placebo group	p value
Distraction rate (mm/day)	0.69 (0.50–0.96)	0.68 (0.51–0.97)	0.88
Final length gain (mm)	64 (45–81)	64 (43–81)	0.91
Knee extension (°)
6 weeks	−1 (−10 to 0)	−1 (−10 to 0)	0.78
12 weeks	−4 (−40 to 0)	−5 (−30 to 0)	0.63
24 weeks	−1 (−10 to 0)	−1 (−5 to 0)	0.73
48 weeks	−1 (−4 to 0)	−1 (−3 to 0)	0.82
Knee flexion (°)
6 weeks	125 (115–140)	123 (115–140)	0.99
12 weeks	123 (100–140)	126 (100–140)	0.99
24 weeks	137 (130–140)	136 (131–140)	0.99
48 weeks	135 (130–140)	137 (135–140)	0.99
Ankle dorsiflexion (°)
6 weeks	0 (−15 to 20)	−2 (−10 to 20)	0.92
12 weeks	9 (−5 to 20)	8 (−5 to 20)	0.63
24 weeks	16 (10–20)	15 (12–20)	0.97
48 weeks	18 (10–20)	17 (9–20)	0.98
Ankle plantarflexion (°)
6 weeks	42 (5–50)	43 (10–50)	0.87
12 weeks	39 (0–50)	39 (0–50)	0.99
24 weeks	45 (40–50)	48 (43–50)	0.99
48 weeks	46 (40–50)	46 (43–50)	0.99
Pain VAS score-calf (point)
6 weeks	3 (0–8)	3 (0–10)	0.61
12 weeks	3 (0–7)	3 (0–7)	0.53
24 weeks	1 (0–4)	1 (0–3)	0.91
48 weeks	1 (0–3)	1 (0–3)	0.94
Calf circumference (mm)
Preoperative	35 (32–42)	35 (30–42)	0.92
6 weeks	34 (30–40)	34 (30–40)	0.99
12 weeks	33 (29–39)	33 (28–39)	0.88
24 weeks	35 (31–40)	35 (31–40)	0.89
48 weeks	38 (30–41)	38 (30–42)	0.85

Values are expressed as mean, with range in parentheses; BtX-A = Botulinum Toxin Type A.

Fig. 2A–B.

Serial boxplots show postoperative differences in knee ROM between the BtX-A and placebo groups. (A) Knee extension and (B) flexion show no significant differences between groups at each time point. Box = interquartile range; bars = minimum and maximum values; horizontal line = zero reference line.

Fig. 3A–B.

Serial boxplots show postoperative differences in ankle ROM between the BtX-A and placebo groups. (A) Ankle dorsiflexion and (B) plantarflexion show no differences between groups at each time point. Box = interquartile range; bars = minimum and maximum values; horizontal line = zero reference line.

Calf pain (gastrocnemius-soleus area) based on the pain VAS scores showed no side-to-side differences at each time point (Fig. 4). The pain in the tibialis anterior muscle area also showed no side-to-side differences at each time point. The mean maximal calf circumferences showed no differences between groups at each time point (Fig. 5). No injection-related adverse events were found in either group.

Fig. 4.

Serial boxplots show postoperative differences in pain VAS scores in the BtX-A and placebo groups. The postoperative pain VAS scores show no differences between groups at each time point. Box = interquartile range; bars = minimum and maximum values; horizontal line = zero reference line.

Fig. 5.

Serial boxplots show preoperative and postoperative differences in maximal calf circumferences between the BtX-A and placebo groups. Maximal calf circumferences show no differences between groups at each time point. Preop = preoperative. Box = interquartile range; bars = minimum and maximum values; horizontal line = zero reference line.

Discussion

Tibial lengthening is performed for a variety of indications, including leg length discrepancy, familiar short stature, etc However, the procedure is often complicated and sometimes limited by the tension generated in muscles during lengthening, which can cause pain, stiffness, and contractures [8, 9, 13, 15, 16, 20, 25–27]. While there have been treatments introduced for these muscle-related complications, they have many disadvantages, including complications, inconsistent efficacy associated with analgesics, and risks of surgery. For that reason, prevention, perhaps with botulinum toxin, offers an appealing option, but studies are inconclusive on the efficacy of this toxin for limb lengthening [1–7, 11, 14, 19, 22–24, 28–31]. In this study, we found that BtX-A was not effective at increasing ROM or reducing pain in lower leg lengthening.

Several limitations should be mentioned. First, we used a standard dose. Different dosages could bring different results. Hamdy et al. [10] used a weight-based dose (10 IU/kg, up to maximal 400 IU/kg), which is different from our study. We used 200 IU in all patients because the injections were limited to posterior calf muscles and the patients were relatively similar in size. The actual amount of BtX-A injected was 3.5 IU/kg (range, 2.6–5.0 IU/kg) if translated per kilogram in our patient cohort, and this variability represents a study limitation that should be considered. Second, we had 80% power, but the power was only sufficient to detect a rather large effect size (a 50% difference between treatment and control). This means that more moderate and definitely small effects, which might have been clinically important, likely would be missed by our study. Larger studies, which almost certainly will need to be multicenter efforts, will be needed to try to resolve smaller differences, if indeed such differences exist. Third, injection was performed manually. More accurate injection could be made by an ultrasound-guided method. Fourth, it would be a better design to include a no-injection group, which would help to identify the placebo effect. Fifth, in this study, we asked the patients to report pain in each leg separately, but the degree to which they can do this accurately may be questioned. Sixth, it is difficult to be certain that the pain VAS score at the time they answered the questionnaires completely represented the pain they actually had throughout the followup interval.

As far as we know, there have been only three articles on the effect of BtX-A in distraction osteogenesis [10, 18, 19]. Olabisi et al. [18] first explored an animal research model on the effect of this toxin in distraction osteogenesis using mature rabbits in 2007. They observed that its injection at the rabbit gastrocnemius muscle reduced its strength, preserved the strength of the tibialis anterior muscle, and achieved 22% greater dorsiflexion of the ankle. They expected that it might minimize equinus contracture and tibialis anterior muscle damage from overstretching in human tibial lengthening. But, as they pointed out, their model was designed to lengthen in a worst-case scenario with a faster rate (1.5 mm/day) than that used in human distraction osteogenesis. Another study [19] observed the histologic effect in rabbits and found no differences in the injected gastrocnemius but less fibrosis in the antagonist muscle. In our series, no histologic evaluations were included. We found no enhancement of the ROM of the surrounding joints and no pain relief in either the posterior compartment area or the tibialis anterior muscle, which would be expected from the toxin’s protective effect on the anterior compartment.

In 2009, Hamdy et al. [10] reported the results from a multicenter pilot study using BtX-A in human distraction osteogenesis. They used a single dose of 10 IU/kg body weight, up to a maximum of 400 IU. They observed a tendency toward reduced pain, higher quality of life, and higher functional mobility scores. They observed fewer major adverse events in the injection group but without statistical significance. In the current study, we did not observe even a tendency in the clinical effects of BtX-A. We believe that our study was suitably designed to see the effects of BtX-A during lengthening for several reasons. First, the patients in our study had a homogeneous demographic status; second, all patients were skeletally mature and young (mean age, 23 years; range, 16–35 years); third, all surgeries were performed by a single surgeon with the same operative conditions; and, finally, randomizing the legs within each patient ensured the two study groups were similar. We believe that the continuous and abnormal muscle tension stress from tibial lengthening is greater than the paralytic and analgesic effects from a single injection of 200 IU BtX-A, thus, preventing clinical improvements in the patients.

We found that a single injection of the BtX-A (200 IU) into the gastrocnemius and soleus muscles during the index distraction osteogenesis surgery was not effective in terms of enhancing ROM of the adjacent joint or decreasing pain. Further investigations may find an effect with different dosages, dilution, or volume schedules or repeated injections. A larger sample size, which would almost certainly require multicenter collaborations, might be important as these studies are considered and designed to be able to discern smaller clinical differences, if such differences are present.