Correction of genu recurvatum by the Ilizarov method

I. H. Choi, C. Y. Chung, T.-J. Cho, S. S. Park

From Seoul National University Childrens Hospital, Seoul, Korea

he Ilizarov apparatus was used to carry out opening-wedge callotasis of the proximal tibia in ten patients who had suffered premature asymmetrical closure of the proximal tibial physis and subsequent genu recurvatum. In four knees, the genu recurvatum was entirely due to osseous deformity, whereas in six it was associated with capsuloligamentous abnormality. Preoperatively, the angle of recurvatum averaged 19.6 (15 to 26), the angle of tilt of the tibial plateau, 76.6 (62 to 90), and the ipsilateral limb shortening, 2.7 cm (0.5 to 8.7). The average time for correction was 49 days (23 to 85). The average duration of external xation was 150 days (88 to 210). Three patients suffered complications including patella infera, pin-track infection and transient peroneal nerve palsy. At a mean follow-up of 4.4 years, all patients, except one, had achieved an excellent or good radiological and functional outcome.

wedge osteotomy, there are no well-documented series 6-8 apart from a few case reports on the use of callotasis in the management of genu recurvatum. Our study was designed to investigate the use of the Ilizarov technique to correct genu recurvatum by an opening-wedge callotasis through the proximal tibia just below the tuberosity.

4,5

Patients and Methods

Between 1991 and 1996, we treated ten patients with genu recurvatum due to asymmetrical premature closure of the proximal tibial physis, using the Ilizarov apparatus. Five had sustained ipsilateral closed fractures of the femoral shaft, which had been treated by plating or skeletal traction followed by immobilisation in a cast. Two had had closed or open fractures of the proximal tibial shaft and had been treated by skeletal traction followed by immobilisation. One had sustained fractures of the ipsilateral femoral and tibial shaft for which intramedullary nailing of the femur and immobilisation of the tibia had been performed. Another had received an injury to the knee and was treated by immobilisation. In these nine patients the injury was due to a trafc accident. The development of gradual progressive genu recurvatum deformity appeared to result from an occult injury to the anterior portion of the proximal tibial physis. The remaining patient (case 8) suffered from multifocal osteomyelitis after meningococcal infection in childhood, and developed a genu recurvatum after a Hauser operation. There were four men and six women with a mean age at the time of injury of 9.9 years. The mean age at the time of operation was 16.3 years (11.0 to 20.5). The mean time between injury and operation was 6.4 years. Pain or discomfort with weakness was present in all knees. Three patients had mild instability of the knee due to stretching of the anterior (case 1) or posterior cruciate ligament (cases 4 and 10). A valgus deformity of 10, 6 and 30 was also present in three knees (cases 1, 5, 8, respectively). The mean shortening of the ipsilateral limb was 2.7 cm (0.5 to 8.7). Tibial shortening averaged 1.6 cm (0.5 to 2.9) in nine limbs. In one patient (case 4) tibial lengthening of 3.0 cm using a unilateral external xator had been carried out elsewhere two years before referral.
769

J Bone Joint Surg [Br] 1999;81-B:769-74. Received 26 October 1998; Accepted after revision 9 December 1998

Premature asymmetrical closure of the proximal tibial physis can result in osseous genu recurvatum. As the deformity increases it leads to pain, weakness, instability of the knee and disturbance of gait. Patella alta, impairment of the quadriceps mechanism and stretching of the posterior capsuloligamentous structure of the knee may also 1,2 occur. Various methods of proximal tibial osteotomy have been suggested to correct osseous recurvatum. Despite many 1-3 reports in the literature of opening-wedge and closing-

I. H. Choi, MD, Professor C. Y. Chung, MD, Assistant Professor T.-J. Cho, MD, Assistant Professor Department of Orthopaedic Surgery, Seoul National University Childrens Hospital, 28 Yongon-dong Chongno-gu, Seoul 110-744, Korea. S. S. Park, MD, Instructor Department of Orthopaedic Surgery, University of Ulsan College of Medicine, Asan Medical Centre, 388-1 Pungnap-dong, Songpa-gu, Seoul 138-736, Korea. Correspondence should be sent to Dr I. H. Choi. 1999 British Editorial Society of Bone and Joint Surgery 0301-620X/99/59589 $2.00
VOL. 81-B, NO. 5, SEPTEMBER 1999

770

I. H. CHOI, C. Y. CHUNG, T.-J. CHO, S. S. PARK

Fig. 1a

Fig. 1b

Fig. 1c

Diagrams showing that a) the angle of recurvatum (RG) is formed by the intersection of the mechanical axis of the tibia and the extension of the mechanical axis of the distal part of the femoral shaft, b) the angle of tilt of the tibial plateau (RT) is formed by the intersection of the line of the plane of the tibial plateau and the mechanical axis of the tibia and c) the angle of tilt of the femoral condyle (RF) is formed by the intersection of the Blumensaat line and the extension of the mechanical axis of the distal part of the femoral shaft.

Fig. 2a

Fig. 2b

Fig. 2c

Determination of the hinge point. Figure 2a Lateral radiographs of the affected and unaffected knee are traced. The gure of the traced proximal tibia of the affected side is superimposed on top of that of the reversed radiograph of the unaffected proximal tibia. Figure 2b The planned osteotomy line at just below the tuberosity is drawn and the tracing paper cut along the osteotomy line. The amount of angulation and posterior translation, when needed, is determined by comparing the two longitudinal lines of the midshaft of the normal and affected sides. The amount of lengthening required can be estimated. Figure 2c Straight lines are drawn between the original and planned position of the anterior and posterior cortices of the affected tibia. These lines are considered to be the tangents to the concentric circles which would be generated by the hinge. The tangent lines are bisected and perpendicular lines drawn. Where the two perpendicular lines cross is taken as the centre for placement of the hinge.

The angle of recurvatum (RG) and the angle of tilt of the tibial plateau (RT) were measured according to the method 1 of Moroni et al using the lateral radiographs of the knee taken with the patient supine and the knee in the position of

maximum possible extension, in neutral, and in 30 exion. The angle of tilt of the femoral condyle (RF) was also measured, since distal femoral physeal injury could inuence the recurvatum deformity (Fig. 1). The amount of
THE JOURNAL OF BONE AND JOINT SURGERY

CORRECTION OF GENU RECURVATUM BY THE ILIZAROV METHOD

771

Table I. Scoring system for the results by Lecuire et al Anatomical (radiographic) results RG in degrees 0 to 3 4 to 6 7 to 9 10 to 12 >12 RT in degrees 92 to 100 88 to 91 or 101 to 104 84 to 87 or 105 to 108 <84 or >108 Patellar height (A:B ratio) 0.66 to 0.94 0.51 to 0.65 or 0.95 to 1.09 0.36 to 0.50 or 1.10 to 1.24 <0.36 or >1.24 Total anatomical score Excellent Good Fair Poor Functional results Pain None Slight Mild Severe Instability None Slight or mild Severe Range of motion Full Decreased 1 to 20 Decreased >20 Weakness None Slight Mild Severe Sports activity Yes No Patients evaluation of the result Excellent Good Fair Poor Total functional score Excellent Good Fair Poor Total combined score (anatomical and functional results) Excellent Good Fair Poor

Points

40 30 20 10 0 30 20 10 0 30 20 10 0 90 to 100 70 to 80 40 to 60 <40

20 10 5 0 15 5 0 20 10 0 15 10 5 0 10 0 20 10 5 0 85 to 100 60 to 80 40 to 55 <40

175 to 200 130 to 170 80 to 125 <80

To determine the effect of genu recurvatum on patella alta or infera, the ratio for patellar height (A:B) was determined from lateral radiographs using the method of 9 Blackburne and Peel. A is the perpendicular distance from the distal osseous margin of the articular surface to the osseous line of the tibial plateau and B is the length of the osseous articular surface of the patella. The results were evaluated according to the scoring 3 system described by Lecuire et al. The anatomical (radiological) and functional results were classied as excellent, good, fair or poor (Table I). Operative technique. We performed opening-wedge callus distraction of the proximal part of the tibia, just distal to the tuberosity and in association with a proximal bular osteotomy. Concomitant tibial lengthening, averaging 1.6 cm (0.4 to 2.5) was undertaken on nine limbs, and concomitant femoral lengthening, averaging 3.2 cm (1.0 to 6.5), in four. Anterior translation of the tibia by a rail system attached to the distal femoral and proximal tibial rings was performed on two unstable knees with posterior displacement. Simultaneous correction of the distal femoral deformity was carried out in two limbs. Two skeletally immature patients had hemiepiphysiodesis of the posterior portion of the proximal tibial physis (Table II). The Ilizarov frame was preconstructed before the operation. The medial and lateral hinge points for correction of the deformity in the proximal tibia were selected as illustrated in Figure 2. When combined correction of deformity and lengthening of the proximal tibia were necessary, the hinge placement was located posterior to the posterior margin of the proximal segment, usually above the level of the osteotomy. Above the level of the osteotomy the proximal metaphysis was fixed by two or three transfixing 1.8 mm wires in addition to two 6 mm half pins attached to one complete ring or one ring and the 5/8 ring. Tibial fixation below the osteotomy was by three transfixing wires and two half pins to two rings. Femoral rings were selectively added when combined femoral lengthening, correction of the distal femoral deformity and reduction of subluxation of the knee were required. A medial and lateral rail system with knee hinges attached to the femoral ring allowed anterior translation of the tibial frame and movement of the knee (Fig. 3). After a period of ve to seven days, distraction of the anterior threaded rod was started based on calculation by the triangulation distraction method described by Herzen10 11 berg and Waanders, and Paley et al, assuming that the distraction gap corresponded to 1 to 1.5 mm/day at the anterior cortical margin.

Results
As shown in Table II, on the basis of radiological measurements, the mean RG angle was 19.6 (15 to 26). The mean RT angle was 76.6 (62 to 90) and the -angle RT 20.3 (12

recurvatum which was due to angulation of the tibial plateau was described as the difference in the angle RT between the normal and affected sides ( -angle RF).
VOL. 81-B, NO. 5, SEPTEMBER 1999

772

I. H. CHOI, C. Y. CHUNG, T.-J. CHO, S. S. PARK

Table II. Clinical details before operation for the ten patients with genu recurvatum Age (yr) 14.9 13.3 17.8 17.7 14.9 11.0 20.5 15.0 19.9 19.4 Leg-length discrepancy (cm) (femur/tibia) 4.6 0.7 2.1 1.0 3.3 1.2 2.9 8.7 2.0 0.5 (4.0/0.6) (0.0/0.7) (0.0/2.1) (1.0/0.0) (1.0/2.3) (0.0/1.2) (0.0/2.9) (6.5/2.2) (0.0/2.0) (0.0/0.5) A:B ratio 1.08 1.08 1.00 1.00 1.11 1.02 1.04 1.00 0.89 0.61 Type of operation* DT, DT, DT, DT, DT, DT, DT, DT, DT, DT, LT, LF LT, HT LT LF, TT LT, LF LT, DF, HT LT LT, LF, DF LT LT, TT

Case 1 2 3 4 5 6 7 8 9 10

Side R R L L R R L R R L

Gender F F F F F F M M M M

RG 20 16 20 26 21 20 17 15 18 23

-RT 14 12 16 28 17 28 13 33 16 26

-RF 0 0 0 0 0 -10 0 -15 0 0

* DT, deformity correction of proximal tibia; LT, lengthening of tibia; LF, lengthening of femur; DF, deformity correction of distal femur; TT, anterior translation of tibia with rail system; HT, posterior hemiepiphysiodesis of the proximal tibia

Fig. 3a

Fig. 3b

Fig. 3c

A 17-year-old girl with recurvatum deformity and posterior subluxation of the knee (a and b). The rail system between the distal femoral and proximal tibial rings for anterior translation of the tibia is shown (c) and the photograph (d) and radiograph (e) three years after the operation indicate a good result.

Fig. 3d

Fig. 3e
THE JOURNAL OF BONE AND JOINT SURGERY

CORRECTION OF GENU RECURVATUM BY THE ILIZAROV METHOD

773

Table III. Clinical details and nal results after operation for the ten patients with genu recurvatum, 3 according to the scoring system by Lecuire et al Leg-length discrepancy (cm) 0.0 +0.3 0.0 0.0 +0.1 0.0 -0.4 +0.3 -0.4 0.0 A:B ratio 0.95 1.00 1.00 0.96 1.05 0.95 1.00 1.57 0.72 0.67 Follow-up (yr) 4.9 4.9 4.2 3.4 5.8 5.9 5.0 5.9 2.2 2.1 Radiological/clinical score Good (70/70) Excellent (90/100) Good (90/80) Good (90/80) Good (90/60) Fair (70/55) Good (80/75) Good (60/100) Excellent (100/100) Excellent (80/100)

Case 1 2 3 4 5 6 7 8 9 10

RG 0 -5 -4 0 -4 -4 0 -4 0 -5

-RT -1 -3 5 1 1 2 5 5 -1 6

-RF 0 0 0 0 0 -3 0 -10 0 0

to 33). The recurvatum deformity was purely osseous in four knees. This was due entirely to tibial deformity in two (cases 4 and 10), and to a combined deformity of the proximal tibia and distal femur in two (cases 6 and 8). In six knees, it was a result of the combination of osseous and soft-tissue changes, the osseous deformity being predominant. The mean time required for correction was 49 days (23 to 85). The mean time of xation was 150 days (88 to 210). After removing the xator, a long-leg cast was applied for a mean of four weeks. In three patients who had instability of the knee, an orthosis was used for six to 12 months after the cast had been removed. Satisfactory correction of the deformity was achieved in all limbs. After operation the mean RG angle was -3 (-5 to 0), while the mean RT angle was 95 (90 to 105) which corresponded to the average RT angle of 97 for the normal contralateral side (Table III). The patellar height ratio was 0.98 preoperatively and 0.99 postoperatively at the last follow-up. There was leg-length discrepancy of less than 5 mm on orthoradiological measurement in all the limbs. There was a denite improvement in function after operation. No pain or discomfort was present in the knees. Weakness disappeared in six, but persisted slightly in four. Sports activities appropriate to the peer group were possible in nine patients. The arc of knee movement was within normal limits in seven patients but three showed a slight decrease. At the last follow-up at a mean of 4.4 years (2.1 to 5.9), three patients had excellent results, six good and one fair, 3 based on the total anatomical and functional score. One patient (case 6) who developed patella infera during the period of treatment had a fair result. Complications. Complications were encountered in three patients. Minor problems included transient palsy of the peroneal nerve and pin-track infection. The nerve palsy, which caused weakness of the extensor hallucis longus, was thought to be due to stretching of a branch of the deep peroneal nerve during retraction of the surrounding muscles at the time of proximal bular osteotomy. The palsy resolved spontaneously within three months of the operation. Pin-track infection was controlled by drainage and administration of antibiotics. Patella infera developed durVOL. 81-B, NO. 5, SEPTEMBER 1999

ing treatment by the external xator in one case, and was considered a major complication. It improved after arthroscopic adhesiolysis and manipulation of the knee under anaesthesia. At the last follow-up, the patient had regained a full range of knee movement.

Discussion
Genu recurvatum can be due to bony deformity of the proximal end of the tibia after premature asymmetrical closure of the proximal tibial physis, or to soft-tissue 1,3 changes (capsuloligamentous recurvatum) or both. In nine out of ten knees in our study, genu recurvatum appeared to be the result of occult injury to the proximal tibia associated with ipsilateral femoral and/or fractures of 12,13 Recogthe tibial shaft, as has been observed by others. nition of the physeal injury was delayed for a mean of 6.4 years until a marked angular deformity appeared as descri13 bed by Hresko and Kasser. In the past, opening-wedge osteotomy of the tibia prox3 imal to the tibial tuberosity, closing-wedge osteotomy 4 proximal to the tuberosity or distal to the tuberosity in 5 association with a bular osteotomy, and opening-wedge osteotomy in combination with the detachment of the tuber1 1 osity, have been described. Moroni et al stressed that in knees in which osteotomy had been carried out proximal to the tuberosity, but in which the patellar ligament had not been detached, the patella was located much too distally. They also stated that when osteotomy was carried out distal to the tuberosity, poor results were obtained because of insufcient correction of the deformity and a prominent anterior curve of the tibial diaphysis was produced. A variety of complications was reported with these techniques. None of these authors discussed the treatment of genu recurvatum associated with genu valgum, posterior subluxation of the knee, and/or ipsilateral shortening of the 3,14 lower limb. In recent years, the Ilizarov technique has gained wide popularity for correcting deformity and lengthening of bone. Apart from a few case reports on physeal distraction or callotasis of the proximal tibia, little has been pub6,7 lished. This technique avoids morbidity from the donor site for bone grafting and removal of plates. Correction of

774

I. H. CHOI, C. Y. CHUNG, T.-J. CHO, S. S. PARK

the deformity can be monitored, and joint movement maintained throughout treatment. Transarticular xation allows control of instability of the knee and prevents subluxation. The hospital stay is short. Correction of the deformity and lengthening can take place simultaneously at the same or at 6,8 different levels. The origin of the recurvatum deformity must be carefully assessed. Appropriate placement of the hinge must be selected when preconstructing the Ilizarov apparatus to achieve angular correction, translation and lengthening. Our results indicate that the Ilizarov apparatus is an effective tool for managing the deformity of osseous recurvatum, with or without shortening. Nine out of ten patients achieved an excellent or good result with satisfactory anatomical and mechanical realignment. These results are sim1 ilar to those reported by Moroni et al with one-stage opening-wedge osteotomy and bone grafting of the proximal tibia if the deformity was entirely or predominantly osseous. The value of the Ilizarov technique, compared with other methods, was demonstrated in two patients who had marked osseous recurvatum and posterior subluxation of the knee; they achieved excellent or good results at the last follow-up examination. Using the special frame, as shown in Figure 3, simultaneous correction of the bony deformity of the proximal tibia and the posterior subluxation of the knee can be obtained, while movement of the knee is maintained. The patellar height ratio was not substantially changed by the procedure. This may be related to the level of the osteotomy, which was carried out just below rather than proximal to the tuberosity. We experienced a major compli15 cation in one patient who developed patella infera during the period of treatment. This may have been due to the transxing pins involving the synovium of the knee, the concomitant femoral osteotomy, and poor compliance with exercises for knee movement. If an opening-wedge callotasis of the proximal tibia is undertaken in a skeletally immature patient to correct genu recurvatum secondary to premature asymmetrical closure of the proximal tibial physis, hemiepiphysiodesis of the posterior portion of the proximal tibial physis would be

necessary to prevent recurrence of the deformity. The anticipated limb-length discrepancy after hemiepiphysiodesis can be treated by primary or secondary tibial lengthening or contralateral epiphysiodesis.
No benets in any form have been received or will be recceived from a commercial party related directly or indirectly to the subject of this article.

References
1. Moroni A, Pezzuto V, Pompili M, Zinghi G. Proximal osteotomy of the tibia for the treatment of genu recurvatum in adults. J Bone Joint Surg [Am] 1992;74-A:577-86. 2. Moroni A, Vicenzi G, Ceccarelli F, Binazzi R, Vaccari V. Surgical treatment of genu recurvatum with procurvatum high tibial osteotomy. Orthop Trans 1987;11:460. 3. Lecuire F, Lerat JL, Bousquet G, Dejour H, Trillat A. The treatment of genu recurvatum. Rev Chir Orthop Reparatrice Appar Mot 1980;66:95-103. 4. Bowen JR, Morley DC, McInerny V, MacEwen GD. Treatment of genu recurvatum by proximal tibial closing-wedge/anterior displacement osteotomy. Clin Orthop 1983;179:194-9. 5. Irwin CE. Genu recurvatum following poliomyelitis: a controlled method of operative correction. J Am Med Assn 1942;120:277-80. 6. ODwyer DJ, MacEachern AG, Pennig D. Corrective tibial osteotomy for genu recurvatum by callus distraction using an external xator. Chir Organi Mov 1991;76:355-8. 7. Olerud C, Danckwardt-Lilliestrom G, Olerud S. Genu recurvatum caused by partial growth arrest of the proximal tibial physis: simultaneous correction and lengthening with physeal distraction: a report of two cases. Arch Orthop Trauma Surg 1986;106:64-8. 8. Pennig D, Baranowski D. Genu recurvatum due to partial growth arrest of the proximal tibial physis: correction by callus distraction: case report. Arch Orthop Trauma Surg 1989;108:119-21. 9. Blackburne JS, Peel TE. A new method of measuring patellar height. J Bone Joint Surg [Br] 1977;59-B:241-2. 10. Herzenberg JE, Waanders NA. Calculating rate and duration of distraction for deformity correction with the Ilizarov technique. Orthop Clin North Am 1991;22:601-11. 11. Paley D, Herzenberg JE, Tetsworth K, Mckie J, Bhave A. Deformity planning for frontal and sagittal plane corrective osteotomies. Orthop Clin North Am 1994;25:425-65. 12. Bowler JR, Mubarak SJ, Wenger DR. Tibial physeal closure and genu recurvatum after femoral fracture: occurrence without a tibial traction pin. J Pediatr Orthop 1990;10:653-7. 13. Hresko MT, Kasser JR. Physeal arrest about the knee associated with non-physeal fractures in the lower extremity. J Bone Joint Surg [Am] 1989;71-A:698-703. 14. Pappas AM, Anas P, Toczylowski Jr HM. Asymmetrical arrest of the proximal tibial physis and genu recurvatum deformity. J Bone Joint Surg [Am] 1984;66-A:575-81. 15. Noyes FR, Wojtys EM, Marshall MT. The early diagnosis and treatment of developmental patella infera syndrome. Clin Orthop 1991;265:241-52.

THE JOURNAL OF BONE AND JOINT SURGERY