Leg-length discrepancy is a relatively frequent disorder that can be real (ie, geometrically detectable) or functional. The indication for surgery depends on the patient's age, the cause and extent of the discrepancy, secondary changes of the hip, and severity of symptoms. In children and adolescents, leg-length discrepancies may have adverse effects on spinal growth.29,37 Despite nonoperative therapy options such as insoles, heel elevation, or orthoses, patients often wish to have the leg-length discrepancy corrected operatively, especially when the discrepancy is substantial. In adults, operative correction often is considered in patients with a discrepancy greater than 3 to 4 cm.19,27
In the 1950s, Ilizarov introduced an external ring fixator for callus distraction.8,15,19,37 For the corticotomy in callus distractions of the proximal tibial metaphysis, Ilizarov used an anterior approach through which callus distraction and segment-shifting of the tibia are still routinely performed.1,8,19 Various factors affect the rate of osteoneo-genesis, such as the latent period between corticotomy and start of distraction and the rapidity and rate of distraction.5,8,18,19,36 The quantity of bone regeneration is further influenced by the location of the corticotomy.9,12,14
A delay or failure of bone consolidation in the distraction area generally is thought to relate to deficient osteo-genesis and occurs in 3% to 8% of patients.3,12,15,28 The regeneration deficiency usually appears locally, mostly in the anteromedial and the anterolateral tibia.12 Such callus defects may result in decreased stability, increased risk of infection, and prolonged healing. However, it is unclear if the development of callus defects is merely a rare but fateful event, or if such a local deficit of bone regeneration is a frequent complication after the anterolateral tibial approach.
We asked whether a posteromedial approach to the tibia for corticotomy would lead to improved callus formation and fewer complications after callus distraction compared with the traditional anterolateral approach. We also questioned whether the two approaches would show different osteogenic reactions. Because blood supply is crucial for bone formation, we performed additional cadaver studies to determine the three-dimensional (3-D) topographic anatomy and individual variances, particularly of the arterial vessels of the proximal tibia.
MATERIALS AND METHODS
We prospectively evaluated 31 patients with substantial leg-length discrepancies scheduled for callus distraction of the proximal metaphyseal tibia from September 1989 to January 1995. Patients were assigned randomly to have either an anterolateral or a posteromedial approach. Corticotomy and insertion of the callus distraction system were performed similarly in both groups. We performed serial clinical, radiographic, and histo-logic evaluations (in case of failing callus formation) to compare the progress of callus formation.
We included 17 male and 14 female patients with congenital, posttraumatic, postoperative, or idiopathic leg-length discrepancies greater than 2.5 cm. Patients were excluded if they had: coagulopathy, diabetes mellitus, renal failure, diseases affecting bone metabolism, inflammatory diseases (especially osteomyelitis), malignant diseases, nicotine abuse, alcohol or analgesic abuse, or current treatment with drugs known to affect bone metabolism (eg, estrogen, glucocorticoids, anticonvulsants, vita-min D, calcium, or phenprocoumon). The mean age of the patients was 22.4 ± 8.6 years in the anterolateral group and 21.6 ± 9.4 years in the posteromedial group. Indications for operative intervention included congenital, posttraumatic, postoperative, postinfectious, and idiopathic causes. Both groups were treated and observed during the same period, and the same two surgeons (MK, CH) performed all operations in both groups. All patients gave informed consent for participation. The assigned physician obtained consent for treatment separately during clinical routine. Eighteen patients (12 males, six females) were operated on using the anterolateral approach, and 13 patients (five males, eight females) were operated on using the newly designed posterome-dial approach (Fig 1A). Preoperatively, we used radiographs to determine the transitional zone from the proximal metaphyseal to the diaphyseal tibia.
The area recommended for the anterolateral approach reaches from the anterior edge of the tibia to the peroneus longus including the tibialis anterior and extensor digitorum longus (Fig 1B).11 An excessively lateral approach should be avoided to prevent the risk of injuring the superficial peroneal nerve, the anterior tibial artery and vein, and the deep peroneal nerve between the tibialis anterior and extensor digitorum longus. Using this approach, the best way to reach the proximal tibia is in the anterior half. After osteotomy of the fibula, we performed a longitudinal incision in the anterior zone, dissected the subcutaneous tissue, and moved the tibialis anterior off the anterior tibial border to reveal the periosteum so the corticotomy could be performed.
The area for the posteromedial approach extends from the medial border of the tibia over the soleus to the middle of the medial head of the gastrocnemius (Fig 1B).11 A transmuscular dorsal approach should not be performed too dorsally as the tibial nerve, the posterior tibial artery and vein, and the peronealartery and vein are at increased risk for injury. Therefore, a posteromedial approach may be safer by sparing the anatomic structures. We made a longitudinal incision within 0° to 70° of the sagittal axis of the tibia, severed the subcutaneous tissues, and longitudinally dissected the medial head of the gastrocnemius and soleus muscles to reach the periosteum for the tibial corticotomy.
We chose the area between the metaphyseal and diaphyseal tibia as the corticotomy site. We first performed a longitudinal incision followed by dissecting the subcutaneous and soft tissues. We cut the periosteum in a z-shape pattern which was elevated to expose the tibia.8,16 The cortical pillars were split step-by-step with the aid of a corticotomy chisel.19,23,27,37 The medullary system was preserved. To verify complete separation, we distracted and realigned the bone ends under radiographic control. This procedure allows an exact and complete corticotomy without medullary disruption or injuring the nutrient periosteal and medullary supply systems.
We used a modified ring fixator (Wiesbaden Ring Fixator, Orthomed Ltd, Lautertal, Germany) for the callus distraction. It was applied using two main fixation planes in the bone close to the distraction site and another main fixation plane at the distal end. We placed three Kirschner (K) wires (1.8 mm diameter per ring) at the widest possible angles. Both sides of the ring were used for wire fixation, and four threaded telescope rods linked the rings. The distraction was started 5 days after the corticotomy at a speed of 1 mm/day. The mean distraction length was 6.6 ± 2.9 cm in the anterolateral approach and 6.4 ± 1.8 cm in the posteromedial approach.
Clinical and radiographic examinations of the distraction zones were performed every 4 weeks during the distraction phase (82.2 ± 7.5 days) and every 6 to 8 weeks during the fixation phase (187.4 ± 16.7 days). Two independent radiologists (DL, CN) blinded to the outcomes evaluated and graded callus defects on serial radiographs according to a grading system developed in our hospital. A defect of as much as 25% of the tibial diameter in the distraction area was classified as Grade 1 (Fig 2A). Defects of as much as 50% and 75% were classified as Grades 2 (Fig 2B and 3) (Fig 2C), respectively, and defects greater than 75% were classified as Grade 4 (Fig 2D). Because the size and grade varied, we used the highest grade during the healing phase for classification and statistical analysis.
All low-grade defects (Grades 1 and 2) healed spontaneously without operative intervention. In contrast, high-grade defects (Grades 3 or 4) showed no signs of callus formation. Therefore, revision surgery was performed when it seemed that the callus defect was not likely to heal spontaneously (ie, 15-21 months after primary surgery depending on distraction length). We took biopsy specimens from these defects during revision surgery to examine the osteogenic reaction and to verify the clinical and radiographic presumptions of failed healing. In the six patients who needed revision surgeries with osteosynthesis and autologous cancellous grafting, three biopsy specimens were taken from the defect and transitional zones from defect to callus and defect to bone. After dehydration, synthetic infiltration (Technovit 7200 VLC®, Kulzer, Friedrichsdorf, Germany) and polymerization by light exposition (wave length, 400-500 nm), we cut 18 samples into 10-to 20-μm thick slices which were stained with hematoxylin and eosin and examined under a light microscope (Axioskop 2 plus®, Zeiss, Jena, Germany). One patholo-gist (UB) analyzed histologic sections in a blinded fashion. The biopsy specimens were assessed for regenerative osteogenic potential leading to spontaneous healing of the defects. By means of light microscopy, the number of osteoblasts, fibroblasts, and the amount of osteoid and vascularization were used as indicators of bone remodeling and bone formation.
Soft tissue samples were taken from six fresh, unfixed human cadavers to determine the 3-D topographic anatomy and individual variances in the arterial vessels to the proximal tibia. The deep femoral artery and the femoral artery were severed, the blood removed from the preparations, and an epoxide resin system (Biodur™ E 20, Biodur™ E 2, Methylethylketon, Biodur-Products®, Heidelberg, Germany) was injected into the arteries, polymerized and heat-treated at 50° to 60°C for 3 days. After the epoxide resin hardened, the arterial vessel systems were dissected and the remains of soft tissue were macerated with potassium lye 10% to show the arteries supplying the proximal tibia (Fig 3).
The association between the operative approach (anterolateral versus posteromedial) and the frequency of callus defects was calculated using the chi square test with contingency tables. We used the Mann-Whitney U test to compare the healing index. We used Spearman's rank correlation coefficient to determine the association between age, callus grade, and healing index (time of healing for 1 cm of the distraction length). The statistical analyses were done using SPSS for Windows (Version 6.1.3®, SPSS Software Ltd, Munich, Germany). Values are expressed as mean ± standard deviation (SD). Differences were significant at the p < 0.05 level.
Callus defects occurred less frequently (p = 0.001) after the posteromedial approach. There were 13 callus defects (41.9%) in 31 patients; 10 occurred in males and three occurred in females. Of the 13 patients with callus defects, 12 were treated using the anterolateral approach, whereas only one patient was treated using the posteromedial approach. Additional radiographic evaluation of the 13 defects showed seven defects (53%) were assessed as Grades 1 and 2 defects, whereas six were assessed as Grades 3 and 4 defects (Fig 2).
The first signs of new bone formation in the distraction space were observed on radiographs of the tibia after a mean of 22.4 ± 4.3 days (both groups) in the shape of faint longitudinal callus strips. The time for detection of callus defects depended on the distraction length and severity. Grades 1 and 2 defects were diagnosed after a mean healing index of 30 days/cm, whereas Grades 3 and 4 defects were diagnosed after a mean healing index of 50 days/cm. Higher-grade defects were associated with a shorter (r = 0.645; p = 0.023) distraction distance and increased healing index. The healing index (time of healing for 1 cm of distraction length) was 43 ± 6.3 days per centimeter using the anterolateral approach compared with 59.98 ± 8.1 days per centimeter using the posteromedial approach. Age did not correlate with the incidence, severity, or healing index of callus defects.
No osteogenic reactions were found inside the defect or along the transitional zones in the patients with Grade 3 and Grade 4 defects (Fig 4) who had surgery. Rather, we observed abundant connective tissue containing numerous capillaries, fibroblasts, and fibrocytes. In contrast, many osteoblasts were observed on the surface of the trabeculae in the callus and corticotomy zones, indicating high osteogenic activity. However, spontaneous healing could not be assumed because none of the defect zones showed signs of osteogenesis.
The vascular anatomy of the cadaver specimens showed the popliteal artery dividing into the anterior and posterior tibial arteries (Fig 3). The anterior tibial artery penetrates the interosseous membrane close to the fibula and runs along the anterior lower leg. The posterior tibial artery with its branches supplies the deep flexors. The anterior tibial artery enters the lodge of the extensors and gives rise to the anterior tibial recurrent artery, which forms anastomoses with branches of genicular arteries. This genicular network receives supply from branches of the popliteal and the genicular arteries and supplies the proximal tibia. High and down-soaring branches of the lower genicular arteries encircle the proximal metaphysis and penetrate the bone on its entire circumference. Arteries of the dorsal muscles rising from the popliteal and the posterior tibial arteries play the major role in the blood supply of the proximal metaphyseal tibia and form a network that is considerably closer at the epiphyseal and metaphyseal tibia compared with the diaphyseal tibia.
Callus distraction as originally described by Ilizarov has become a valuable procedure in the treatment of primary and secondary limb shortness and severe injuries with bony defects.3 Because of critical soft tissue covering and blood supply of the tibia, delay or failure of bone consolidation in the distraction area is a major complication resulting in callus defects and prolonged healing.19 We developed a new posteromedial approach to the tibia that decreased callus defects and improved the healing process. Patients who were treated with the conventional anterolateral approach had callus defects develop more frequently. Higher-grade defects were associated with less osteogenic potential and resulted in an increased healing index.
Our study is limited by several factors. First, it includes a small number of patients. However, we found evidence that the new posteromedial approach has several benefits compared with the conventional anterolateral approach. Additional studies including more patients are warranted to confirm our findings for a larger population. Second, we included patients with different indications leading to the operative intervention (ie, congenital, posttraumatic, postoperative, postinfectious, and idiopathic causes). However, the use of rather restrictive inclusion criteria still allowed us to obtain a homogenous study group. Other limitations were our radiographic and histologic evaluations of callus defects. In some cases it was not possible to make a correct classification even if it was possible to identify the defect. Also, the technique of histologic assessment was developed by our department, but we did not perform any statistical validation.
The requirements for callus distraction and segment shifting are stringent surgical indications, and precise knowledge of the topographic anatomy, physiology, and blood supply.5,18,24,28 In the lower leg, the tibia is in an eccentric anterior position and asymmetrically covered by soft tissues.6,11,12 The anterior border and the medial surface are covered by skin only, so the vascular supply in this area must be assumed to be limited for extensive reparative processes. The powerful flexors on the back of the lower leg, however, provide an excellent cover to the dorsal parts of the tibia, representing a solid foundation for sufficient blood supply to the proximal tibia.11,31,33,35
The local blood supply and the medullary system affect the regionally varying osteogenesis and reparation processes of bone.8,17,26,31 This observation was strengthened by finding better local blood supply and circulation in the posteromedial proximal tibia.13,20,24,32 Brutscher et al attributed the noticeably less bone formation in the anterior and medial tibia to the insufficient soft tissue covering and blood supply of the bone.8 At the same time, callus forms mainly on the dorsal proximal tibia with its density decreasing from posterior to anterior because of the insufficient anterior soft tissue coverage.12,13,30 Soft tissue covering is more luxuriant for osteogenesis in the posterome-dial than in the anterolateral shank, which might explain our results of significantly less callus defects at the dorsal tibia. In contrast, the insufficient soft tissue covering, which is even more damaged by the anterolateral approach, seems to inhibit osteogenesis in the corresponding areas of the bone by delaying the onset of bone formation.
The local blood supply and soft tissue covering the periosteum are important for vascular supply and osteo-genesis. The periosteum consists of the adventitia, fibroelastica, and cambium.10,22 The cambium is the inner layer that fits tightly to the bone and contains preosteoblasts that will differentiate into osteoblasts and produce callus during the distraction phase.34 The periosteum is a well-vascularized structure and is important for blood supply to bone. Blood vessels run through the periosteum before they penetrate the cortex to supply the bone.3,10 Stripping the periosteum should be avoided. A dorsal incision of the periosteum can be much better tolerated than an incision at the anterior tibia, which is covered by a small amount of soft tissue. An injury to these soft tissues may reduce periosteal blood supply. The surgical exposure for corticotomy involves splitting and partially elevating the periosteum. This may lead to impairing the periosteal circulation, which cannot be fully compensated by the intramedullary blood supply.
Numerous authors observed improved osteogenesis after metaphyseal bone transection than after diaphyseal transsection.6,14,25,33 Aronson et al attributed this to the greater osteogenic potency of the metaphysis because of its large share of excellent blood supply of the trabecular bone.2,4,7 For callus distractions, Ilizarov considered the corticotomy to be of primary importance.19 He attributed great significance to the intactness of the nutrient artery.18,19 An osteotomy leads to substantially delayed healing compared with a corticotomy.19 Delloye et al showed the lack of distinction between corticotomy and osteotomy concerning the method of bone healing and the amount of newly formed bone.10 Some authors have shown the efficacy of an osteotomy for callus distraction.21,22,36 It is important for a best-possible osteogenesis to use a chisel instead of a saw to avoid local heat damage to the bone ends that might lead to reduction of bone regeneration; the medullary vessels should remain uninjured during the corticotomy.19
There is no general agreement regarding the best approach for a corticotomy at the proximal metaphyseal tibia.8,19,29,37 Our new posteromedial approach was beneficial in decreasing callus defects and improving the healing index compared with the conventional anterior approach. These advantages may be because the proximal tibia has better dorsal soft tissue coverage. To reduce callus defects at the proximal metaphyseal tibia, we suggest a minimally invasive posteromedial approach for the corticotomy regardless of the fixation system. The periosteum and the periosteal vascular supply should be minimally dissected to avoid injuring the medullary system, which plays a major role in osteogenesis.16,21,26,33
We thank the Laboratory of Experimental Trauma Surgery in Giessen, Germany, for technical assistance, James Kelley (University of Cambridge, UK) for useful comments regarding the manuscript, and Philip Weitnauer (Germany) for invaluable technical support and picture editing.
1. Aprin H. Femoral lengthening by callus distraction and cortical apposition (Z-osteotomy). Clin Orthop Relat Res
. 1994;309: 222-229.
2. Aronson J. Temporal and spatial increases in blood flow during distraction osteogenesis. Clin Orthop Relat Res
3. Aronson J, Good B, Stewart C, Harrison B, Harp J. Preliminary studies of mineralization during distraction osteogenesis. Clin Orthop Relat Res
4. Aronson J, Harp JH. Mechanical forces as predictors of healing during tibial lengthening by distraction osteogenesis. Clin Orthop Relat Res
5. Aronson J, Harrison BH, Stewart CL, Harp JH Jr. The histology of distraction osteogenesis using different external fixators. Clin Orthop Relat Res
6. Aronson J, Shen X. Experimental healing of distraction osteogenesis comparing metaphyseal with diaphyseal sites. Clin Orthop Relat Res
7. Aronson J, Shen XC, Skinner RA, Hogue WR, Badger TM, Lump-kin CK Jr. Rat model of distraction osteogenesis. J Orthop Res
8. Brutscher R, Rahn BA, Ruter A, Perren SM. The role of corticotomy and osteotomy in the treatment of bone defects using the Ilizarov technique. J Orthop Trauma
9. Brutscher R, Ruter A, Rahn B, Perren SM. The significance of cortocotomy or osteotomy in callus distraction. Chirurg
. 1992;63: 124-130.
10. Delloye C, Delefortrie G, Coutelier L, Vincent A. Bone regenerate formation in cortical bone during distraction lengthening: an experimental study. Clin Orthop Relat Res
11. Faure C, Merloz P. Zugaenge fuer die Fixatuer-externe-Osteosynthese
. Berlin, Germany: Spinger Verlag; 1987.
12. Fink B, Krieger M, Strauss JM, Opheys C, Menkhaus S, Fischer J, Ruther W. Osteoneogenesis and its influencing factors during treatment with the Ilizarov method. Clin Orthop Relat Res
. 1996;323: 261-272.
13. Fink B, Neuen-Jacob E, Lienert A, Francke A, Niggemeyer O, Ruther W. Changes in canine skeletal muscles during experimental tibial lengthening. Clin Orthop Relat Res
14. Fischgrund J, Paley D, Suter C. Variables affecting time to bone healing during limb lengthening. Clin Orthop Relat Res
. 1994;301: 31-37.
15. Franke J, Simon M, Hein G. Ilizarov techniques of leg lengthening: problems and results. Orthopade
16. Frierson M, Ibrahim K, Boles M, Bote H, Ganey T. Distraction osteogenesis: a comparison of corticotomy techniques. Clin Orthop Relat Res
17. Ganey TM, Klotch DW, Sasse J, Ogden JA, Garcia T. Basement membrane of blood vessels during distraction osteogenesis. Clin Orthop Relat Res
18. Ilizarov GA. The tension-stress effect on the genesis and growth of tissues. Part I. The influence of stability of fixation and soft-tissue preservation. Clin Orthop Relat Res
19. Ilizarov GA. Transosseous osteosynthesis
. Berlin, Germany: Spinger Verlag; 1992.
20. Ippolito E, Peretti G, Bellocci M, Farsetto P, Tudisco C, Caterini R, De Martino C. Histology and ultrastructure of arteries, veins, and peripheral nerves during limb lengthening. Clin Orthop Relat Res
21. Karaharju EO, Aalto K, Kahri A, Lindberg LA, Kallio T, Karaharju-Suvanto T, Vauhkonen M, Peltonen J. Distraction bone healing. Clin Orthop Relat Res
22. Kojimoto H, Yasui N, Goto T, Matsuda S, Shimomura Y. Bone lengthening in rabbits by callus distraction: the role of periosteum and endosteum. J Bone Joint Surg Br. 1988;70:543-549
23. Meffert RH, Inoue N, Tis JE, Brug E, Chao EY. Distraction osteo-genesis after acute limb-shortening for segmental tibial defects: comparison of a monofocal and a bifocal technique in rabbits. J Bone Joint Surg Am
24. Minematsu K, Tsuchiya H, Taki J, Tomita K. Blood flow measurement during distraction osteogenesis. Clin Orthop Relat Res
25. Monticelli G, Spinelli R. Distraction epiphysiolysis as a method of limb lengthening. I. Experimental study. Clin Orthop Relat Res
26. Mosheiff R, Cordey J, Rahn BA, Perren SM, Stein H. The vascular supply to bone in distraction osteoneogenesis: an experimental study. J Bone Joint Surg Br. 1996;78:497-498
27. Paley D. Current techniques of limb lengthening. J Pediatr Orthop
28. Paley D. Problems, obstacles, and complications of limb lengthening by the Ilizarov technique. Clin Orthop Relat Res
. 1990;250: 81-104.
29. Pfeil J, Niethard FU. Lower leg lengthening using the Ilizarov system: presentation of the various surgical techniques and analysis of lower leg lengthening procedures performed 1986-1989. Orthopade
30. Reichel H, Lebek S, Alter C, Hein W. Biomechanical and densito-metric bone properties after callus distraction in sheep. Clin Orthop Relat Res
31. Rhinelander FW. Circulation in Bone. In: Bourne GH, ed. The Biochemistry and Physiology of Bone
. Vol 2. 2nd ed. New York, NY: Academic Press; 1972:2-76.
32. Rhinelander FW. Tibial blood supply in relation to fracture healing. Clin Orthop Relat Res
33. Steen H, Fjeld TO. Lengthening osteotomy in the metaphysis and diaphysis: an experimental study in the ovine tibia. Clin Orthop Relat Res
34. Takushima A, Kitano Y, Harii K. Osteogenic potential of cultured periosteal cells in a distracted bone gap in rabbits. J Surg Res
35. Trueta J. Blood supply and the rate of healing of tibial fractures. Clin Orthop Relat Res
36. White SH, Kenwright J. The importance of delay in distraction of osteotomies. Orthop Clin North Am
37. Wiedemann M. Callus distraction: a new method? A historical review of limb lengthening. Clin Orthop Relat Res
. 1996;327: 291-304.