➢ The amount of lengthening performed at one setting should be conservative. The complication rate correlates with the lengthening percentage.
➢ Use of hydroxyapatite pin coatings improves the stability of the pin-bone interface but does not eliminate pin-track infections.
➢ Biologic methods of enhancing bone formation and consolidation are being investigated, but there are no clear recommendations at this time.
Pediatric limb lengthening is an exciting and rapidly developing field. The primary goal is to produce healthy regenerate bone of the desired length without complications. In addition, the experience should be as easy and comfortable as possible for the patient and his or her family. Innovative surgical techniques and devices continue to push surgeons closer to being able to consistently achieve these goals. This article, the second part of a two-part review, will cover advances in techniques designed to decrease the healing index and methods to recognize and avoid complications in pediatric limb lengthening. Part 1 (http://reviews.jbjs.org/content/3/8/e3) covered advances in the preoperative assessment and methods to decrease the time spent in an external fixator1.
Decreasing Bone Healing Index
Although new techniques using integrated fixation or intramedullary lengthening devices have helped to decrease or to eliminate the duration of time in an external fixator, these techniques do not actively change the biology of regenerate bone healing. The consolidation phase of distraction osteogenesis remains the longest portion of time of any limb-lengthening process. The bone healing index represents the time to osseous union in days divided by the amount of lengthening in centimeters. On average, one month of fixator duration is estimated for each centimeter of lengthening required2. Many attempts have been made to promote or to enhance bone formation of the distracted callus, such as motorized distraction, low-intensity pulsed ultrasound, a pulsed electromagnetic field, transplantation of fresh bone marrow cells, platelet-rich plasma, bisphosphonates, or recombinant human bone morphogenetic proteins (BMPs).
One of Ilizarov’s research findings was that smaller, more frequent increments of distraction made healthier bone during distraction osteogenesis3. Mizuta et al. supported this concept by reporting a decreased external fixation time in patients whose bones were lengthened 1 mm/day in eight increments compared with patients whose bones were lengthened 1 mm/day in four increments4. Bright et al. studied motorized high-frequency distraction of the proximal tibial metaphysis in a series of twenty-seven patients5. Their findings suggested that motorized distraction at 1 mm/day in 1440 steps does not improve time to union significantly, nor does it reduce complications of bone fracture, when compared with a traditional manual distraction rate of 1 mm/day in four steps. Motorized distraction may still be helpful in pediatric patients to reduce anxiety over the use of wrenches on manual distraction struts. The study did not comment on any potential benefits to soft tissue (e.g., muscle, nerve, or cartilage) using high-frequency distraction.
Low-intensity pulsed ultrasound delivers high-frequency, low-intensity acoustic pressure waves to the maturing regenerate bone. It has been reported to rescue slow-healing regenerate bone and to accelerate the healing of normal regenerate bone. In a study of 108 children who underwent external fixation using the Ilizarov method, Gebauer and Correll reported seventeen cases of delayed union or nonunion6. Using only low-intensity pulsed ultrasound for twenty minutes per day, they achieved complete union in three to twelve months in each case. At the four-year follow-up, they did not find any negative effects of the ultrasound on the physes. They recommended using low-intensity pulsed ultrasound as a salvage procedure instead of operative intervention. In a case control study of twenty tibial defects managed with distraction osteogenesis, El-Mowafi and Mohsen initiated low-intensity pulsed ultrasound at the beginning of the consolidation phase in ten patients7. The mean healing index was 30 days/cm in the ultrasound group compared with 48 days/cm in the group treated without it.
Another noninvasive modality, pulsed electromagnetic fields, has been used to enhance bone healing. The use of pulsed electromagnetic fields in children is considered a physician-directed (off-label) application. In a study of adolescent patients of short stature who underwent bilateral humeral lengthening, the extremity managed with a pulsed electromagnetic field demonstrated faster callus formation and had greater bone density than the contralateral control side8.
Biologic stimulations designed to decrease the consolidation phase have been gaining attention recently. Kitoh et al. studied the effect of transplanted culture-expanded bone marrow cells and platelet-rich plasma on distraction osteogenesis9. They found that the healing index was significantly lower (p = 0.0005) in a series of adolescent patients with achondroplasia and hypochondroplasia undergoing lower-extremity lengthening when compared with that of a control group. The osteogenic effect seemed to be more effective with femoral lengthening than with tibial lengthening. However, this method requires multiple visits to the operating room and special equipment. In addition, training to learn how to properly harvest the stem cells is necessary.
Two studies have investigated the effect of platelet-rich plasma on the regenerate bone. Latalski et al. injected 30 mL of platelet-rich plasma into the regenerate bone one week after the beginning of the consolidation phase10. Comparing the standard group (38.5 days/cm) with the platelet-rich plasma group (29.8 days/cm), the treatment time in the platelet-rich plasma group was significantly shorter (p = 0.0412). Lee et al. performed a prospective randomized trial testing the clinical efficacy of local injection, at the osteotomy site, of autologous bone marrow aspirate concentrate combined with autologous platelet-rich plasma11. The injection was done at the end of the index surgery in a group of twenty patients undergoing bilateral tibial lengthening over a nail. The authors recommended aspirating 5 mL of the bone marrow from different spots around the anterior superior iliac spine to avoid collecting surrounding peripheral blood. They found no difference in the external fixation index (the number of days that the external fixator is attached to the bone per centimeter of length gained) between the two groups. However, they did find some increased callus regeneration in the treatment group. This allowed earlier weight-bearing in the treatment group (0.99 month/cm) compared with the control group (1.38 months/cm).
Bisphosphonates are a family of anti-catabolic agents that prevent bone resorption. In 2007, Kiely et al. reported that six of seven children treated with parenteral bisphosphonates for poor-quality regenerate bone after distraction osteogenesis eventually healed without further intervention. They used nitrogen-containing bisphosphonates, with three patients receiving pamidronate and four patients receiving zoledronic acid. The patients treated with pamidronate had either two or three infusions (mean total dose, 3.3 mg/kg) for 8.7 weeks. The four patients treated with zoledronic acid had three infusions (mean total dose, 0.09 mg/kg) for 18.2 weeks12. No side effects related to the bisphosphonate therapy were reported.
The development of multiple-growth-factor delivery systems to enhance the healing of regenerate bone holds promise for the future. For example, BMP-2 is known to enhance bone regeneration. However, large doses of BMP-2 can cause uncontrolled ectopic calcification, pain, swelling, and neurological impairment13. One animal study has indicated that, by combining low dose BMP-2 with fibroblast growth factor-2 (FGF-2), their combined action increases the speed and extent of bone healing and may be a solution to the current problems with high-dose BMP-213.
Two recently described surgical techniques are highlighted that have been found to help decrease the bone healing index. A longitudinal (S-Z) osteotomy was found to have a significantly faster healing time (p = 0.03) at 6.3 months compared with transverse osteotomy at 8.1 months in patients undergoing tibial lengthening14 (Fig. 1). The osteotomy is performed percutaneously through two small incisions 5 to 7 cm apart. The lateral half is cut proximally and the medial half is cut distally. The two partial osteotomies are then connected by cutting the anterior cortex subcutaneously between the two sites and rotating the bone to break the posterior cortex. It is believed that the larger exposed surface area and the increased cortical contact between the two segments lead to the rapid healing.
Staged double-level lengthening utilizes a second osteotomy performed six to eight weeks after completion of the first distraction in the same frame15. The second distraction increases the overall blood flow to the entire extremity15. This results in faster consolidation and remodeling of the initial lengthening site. In addition, the speed of lengthening can be maintained at 1 mm/day without concern for overstretching the soft tissues. The first osteotomy is used to correct deformity and achieve moderate lengthening of up to 4 cm. The second osteotomy is 2 to 3 cm above or below the initial osteotomy site. In a series of ten cases, an average of 8.0 cm of lengthening was achieved. The patients spent 2.4 months less time in the frame compared with a control group undergoing single-level lengthening.
Prevention of Complications
Complications during pediatric limb lengthening, such as fractures, infections, deformations of the regenerate bone, nerve injuries, contractures, and hypertension, remain prevalent. Fracture or bending of the regenerate bone is possible after removal of the external fixation. This is especially problematic in patients undergoing lengthening for congenital femoral deficiency, for whom a 34% fracture rate has been reported after removal of the fixator16. The use of a hip spica cast after removal of the fixator does not guarantee that the regenerate bone is sufficiently protected and that a fracture will not occur16. If a fracture does occur, the treatment options include closed reduction and casting, reapplication of an external fixator, and internal fixation. However, these fractures are difficult to manage because they tend to angulate as a result of the tightness of the lengthened soft tissues. In addition, it can be challenging to pass intramedullary fixation through the sclerotic canal. One potential solution is to lengthen through a proximal osteotomy over a Rush nail17. The proximal osteotomy helps to reduce quadriceps scarring and adherence to the regenerate bone. The Rush rod decreases the risk of fracture and deformation of the regenerate bone.
The potential benefits of lengthening over an intramedullary nail inserted at the index procedure have previously been discussed. The insertion of an intramedullary nail or nails at the time of fixator removal has also been found to reduce the risk of fracture and secondary axial deviation. In a series of 101 femoral lengthenings, the regenerate bone fracture rate was decreased to 12% by inserting one or two retrograde titanium elastic nails at the time of frame removal18. The investigators found the one-stage exchange from external to internal fixation to be safe, with a 1% infection rate. The nails were removed at a mean time of 115 days after insertion. The fracture rate was not influenced by the presence of one compared with two elastic nails in that series.
The difference in tension of the lower-extremity muscle groups during lengthening can cause a predictable pattern of regenerate bone deformation19. For example, the proximal part of the femur tends to angulate into varus and procurvatum. Preoperative planning can help to prevent the deformity during distraction from occurring. Placing the uniplanar pins or the circular rings at a 5° to 7° angle in the direction opposite the expected angulation will allow the bone to correct to neutral as the expected deformation takes place19.
Deformities in the sagittal plane are often overlooked. Mean angular deformity of 8.3° and mean displacement of 3.1 mm have been noted in the sagittal plane during femoral lengthening20. These deformities are usually present immediately postoperatively and typically do not worsen significantly with lengthening.
Pin coatings have been designed to help stabilize the pin-bone interface and to decrease pin-track infections. With regard to coatings that improve osseointegration of the pin, hydroxyapatite and bisphosphonate coatings have the most human clinical data. A hydroxyapatite coating helps to decrease pin loosening, especially in metaphyseal bone, by enabling bone formation directly on its surface. The coating creates a chemical bond between the crystals of the coating and those of the newly formed bone21. Hydroxyapatite has been shown to successfully increase the removal torque for half-pins, indicating increased stability of the pin-bone interface over time22,23. However, hydroxyapatite coatings have not been shown to decrease pin-track infection rates24. A review of the literature found the combined pin-track infection rate for hydroxyapatite-coated pins to be 29.6% (seventy-one of 240) compared with 26.0% (1457 of 5609) in uncoated pins25. Despite its ability to improve the pin-bone interface, a hydroxyapatite coating has no direct antibacterial effect and coated pins are still vulnerable to bacterial infection like uncoated pins. External fixator pins coated with a bisphosphonate (zoledronic acid) have also been found to improve fixation in metaphyseal bone, similar to the hydroxyapatite-coated pins21.
Infections of external fixator pins are the result of bacterial adhesion followed by the development of a biofilm26. Changes in half-pin materials and surface coatings continue to be evaluated in an attempt to reduce the risk of infection. Iodine-supported titanium pins represent a promising advancement in half-pin technology27. Iodine has a broad antimicrobial spectrum against common bacteria, fungi, and common viruses and does not cause resistant bacteria. According to the Checketts-Otterburn classification28 (Table I), Shirai et al.27 found, in 476 pin sites, grade-1 infection in 2.5% (twelve pin sites) and grade-2 infection in 1.1% (five pin sites); there were no infections of grade 3 or higher. Iodine-coated pins prevent not only surface infection but also deep infection, because the antimicrobial iodine coating covers the entire pin. Abnormalities of thyroid gland function were not detected. The amount of iodine was maintained for a long time, with approximately 40% remaining after one year.
Pin-care studies are difficult to interpret because of the multiple variables that affect pin-track infections. However, a recent prospective, randomized, controlled, double-blind study demonstrated that the use of gauze impregnated with the antiseptic 0.2% polyhexamethylene biguanide (PHMB) can decrease the risk of pin-track infection compared with plain, normal saline solution-soaked gauze29. In the PHMB group, eleven (1.0%) of 1068 observations showed infection, and in the saline solution group, thirty-nine (4.5%) of 864 observations showed infection.
Postoperative pain management is a concern for all pediatric patients undergoing limb lengthening. Adequate early analgesia and good pain control allow comfortable use of the limb and encourage the maintenance of good range of motion. A continuous peripheral nerve block of the sciatic nerve with use of an infusion of plain levobupivacaine was found to be an effective method of postoperative analgesia in a study of sixty-eight children undergoing limb reconstruction with external fixation30. The patients who received a continuous nerve block had fewer episodes of severe pain than patients who received a morphine infusion. The patients who received a continuous nerve block also had less nausea and vomiting as well as a lower rate of motor blockade than those who received epidural analgesia. Although there is concern that this type of regional analgesia could mask a compartment syndrome in a pediatric patient undergoing osteotomy, no cases of compartment syndrome were encountered in that study.
Intraoperative monitoring of the peripheral nerves using somatosensory evoked potentials has been demonstrated to assist the surgeon to avoid iatrogenic nerve injury during external fixation for limb lengthening31. By tracking the physiologic integrity of the peripheral nerves throughout the procedure, the surgeon can be alerted to nerve compromise before irreversible damage has occurred. The use of intraoperative monitoring is especially helpful in cases in which a substantial risk of neurological injury is anticipated, such as in cases involving acute deformity correction, upper-limb external fixation, congenital deformity with distorted anatomy, or achondroplasia31.
Despite advances in techniques and devices, it is still recommended that the amount of lengthening be conservative. Aston et al. recommended a maximum of 6 cm during one treatment or 19% of the original length of the bone17. The lengthening percentage was found to correlate with the complication rate in a retrospective series of 116 lower limbs of eighty-eight patients undergoing pediatric limb lengthening19. Lengthening the bone by 25% of the original length led to neurological complications in approximately one-third of the patients, and lengthening the bone by 40% of the original length led to neurological complications and joint contractures in approximately 60% of patients. The earlier a nerve injury that occurs during lengthening is detected, the better the prognosis. Prompt intervention, such as slowing the rate of lengthening and/or performing a nerve decompression, may allow continued lengthening without incurring permanent nerve injury19,32. If a high lengthening percentage is expected, a prophylactic peroneal nerve decompression at the time of the index procedure should be done to prevent nerve injury.
Contractures are a frequent problem after limb lengthening. A substantial increase in the lengthening percentage raises the rate of joint contractures19. These contractures are due to increased tension in the quadriceps, hamstrings, triceps surae muscle, and hip adductors. According to Antoci et al.19, if the expected lengthening percentage is to exceed 15%, then the adjacent joint should be protected by adding an extension to the frame that spans the joint. In cases of preexisting joint instability (congenital femoral deficiency, fibular hemimelia), a frame extension across the joint should be used regardless of the expected lengthening percentage. Hinges can be applied at the joint center of rotation, allowing range-of-motion exercises during physical therapy and maintaining some joint motion during the lengthening process. In patients undergoing lengthening with a uniplanar femoral frame, there are now commercially available constructs that allow the surgeon to span the knee and to maintain its range of motion. When spanning the joint, it is recommended to perform a minor joint distraction using the external fixator, which prevents compression of the articular cartilage19.
In a randomized trial of patients undergoing bilateral tibial lengthening, Lee et al. did not find that botulinum toxin type A injection in the calf decreased calf pain32. The knee and ankle ranges of motion were also not improved with the injection when compared with the opposite, untreated side.
A novel quadricepsplasty has been described to improve knee flexion in patients with knee extension contracture following femoral lengthening33. A long oblique incision is made in the quadriceps tendon rather than a classic V-Y lengthening (Fig. 2). The investigators found no quadriceps lag with this technique and the patients maintained active flexion to 120° at an average of six years after surgery.
Hypertension is a complication of limb lengthening that is seldom discussed. It is thought to be caused by distraction of the arteries. The rate of this complication has been found to increase as the lengthening percentage grows19. Thirty percent of patients undergoing a 40% increase in length demonstrated increased blood pressure. Antoci et al. recommended that careful monitoring of the blood pressure during lengthening should become part of the routine19. If hypertension is detected, prompt administration of antihypertensive therapy is recommended. In most cases, slowing the rate of lengthening or even temporarily discontinuing the distraction will help to normalize the blood pressure.
Source of Funding: There was no external funding source in the preparation of this manuscript.
Investigation performed at Nemours Children’s Hospital, Orlando, Florida
Disclosure: The author did not receive payments or services, either directly or indirectly (i.e., via his institution), from a third party in support of any aspect of this work. The author, or his institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. The author has not had any other relationships, nor engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2015 by The Journal of Bone and Joint Surgery, Incorporated