- Email: [email protected]
The Ilizarov method in infected nonunion of fractures
Injury, Int. J. Care Injured 31 (2000) 509±517
www.elsevier.com/locate/injury
The Ilizarov method in infected nonunion of fractures Lalit Maini*, Manish Chadha, Jashan Vishwanath, Sudhir Kapoor, Anil Mehtani, B.K. Dhaon Department of Orthopaedics, Maulana Azad Medical College, New Delhi 110002, India Accepted 14 February 2000
Abstract Thirty patients with infected non-union of long bones were treated with radical resection of the necrotic bone and bone transport or compression/distraction osteosynthesis. Non-union, infection, deformity, bone gap and shortening were all addressed simultaneously using the Ilizarov principles. There were 15 cases with bone loss ranging from 4 to 12 cm (median bone gap of 7 cm), 10 cases of sti non-union (six of which had an associated deformity) and ®ve cases of mobile non-union. The median time in the Ilizarov frame was 150 days. Median follow up time after frame removal was 23.5 months. Bone grafting at the docking site was only required in three cases (10%). There were three cases of refracture (10%) and three cases of recurrence of infection (10%). The bone result was excellent in 21 patients (70%), good in three (10%), fair in none (0%) and poor in six (20%). The functional results were excellent in eight patients (26.7%), good in 12 (40.0%), fair in three (10%) and poor in seven (23.3%). It is dicult to precisely de®ne the indications for preservation and reconstruction of severe injuries. The surgical team has to take into account the length, disability, complications and cost of treatment. Patients must be aware of the limitations of functional results and the possible diculty of return to work despite the reconstructive attempt. 7 2000 Elsevier Science Ltd. All rights reserved.
1. Introduction Infected non-union of fractures is one of the major problems in Orthopaedic surgery. There are many available alternatives in the management of infected non-unions. These include extensive debridement and local soft tissue rotational ¯aps, packing the defects with antibiotic impregnated beads, Papineau type open cancellous bone grafting, tibio®bular synostosis, cancellous allograft in ®brin sealant mixed with antibiotics, and/or free microvascular soft tissue and bone transplants. None of the above mentioned techniques aord the surgeon the ability to correct deformities, eliminate prolonged pre- and postoperative antibiotic therapy, regenerate new bone tissue without the use of bone grafts, progressively lengthen the extremity, and allow * Corresponding author. 22, Samachar Apartment, Mayur Vihar, Phase I, New Delhi 110091, India.
weight bearing during the treatment period simultaneously. All of these capabilities are possible with application of the techniques of Ilizarov. Progressive bone histogenesis after corticotomy and bone transport has bene®ted from the work of Ilizarov [1] and Monticelli and co-workers [2]. It helps to eradicate bone infection and to unite fracture non-unions. It may be a good technique for some dicult post-traumatic conditions. Application of Ilizarov techniques in the treatment of an infected non-union depends on the type of nonunion (hypertrophic or atrophic), the extent of infection and the condition of soft tissues. The wider the area of necrotic bone, the greater the infection. To ensure elimination of infection, it is necessary to perform complete resection of the necrotic bone and infected segments. Bone transport is then used to reconstruct the residual segmental defect. In this study we report on the treatment of infected non-unions using the techniques of debridement, resec-
0020-1383/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 1 3 8 3 ( 0 0 ) 0 0 0 3 6 - X
28,M 36,M 12,M 40,M
18,M 20,M
40,M 42,M 52,F 26,F 38,M 28,M 46,M 50,M 60,M 45,M 40,M 24,M 55,M 65,F 26,M 34,M
9 10 11 12
13 14
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Tibia Tibia Tibia Tibia Humerus Tibia Tibia Femur Tibia Tibia Tibia Tibia Femur Femur Tibia Tibia
Tibia Tibia
Tibia Tibia Femur Tibia
Femur Tibia
Femur Tibia Tibia Tibia Tibia Tibia
Site
No gap/deformity.
25,M 35,M
7 8
a
30,M 19,M 26,M 35,M 30,M 17,M
Age/sex
1 2 3 4 5 6
Case No.
Table 1 Patient pro®le
Quiescent Draining Quiescent Quiescent Draining Quiescent Draining Draining Quiescent Draining Quiescent Quiescent Draining Draining Quiescent Draining
Quiescent Draining
Quiescent Draining Draining Quiescent
Draining Draining
Draining Draining Draining Quiescent Draining Quiescent
Infection
± Staph. Aureus ± ± Mixed ¯ora ± Staph. Aureus Staph. Aureus ± Klebsiella ± ± Mixed ¯ora Staph. Aureus ± E. coli
± E. coli
± Staph. Aureus Mixed ¯ora ±
E. coli Staph. Aureus
Staph. Aureus Staph. Aureus Klebsiella ± Staph. Aureus ±
Organism
3 1 1 2 4 1 5 2 1 1 3 3 1 1 2 2
4 2
3 2 3 2
3 2
4 2 2 2 2 1
Previous surgery
49 38 36 24 25 30 54 30 24 20 19 22 18 19 20 20
48 37
20 24 26 19
19 30
24 30 27 31 42 43
Follow up (month)
a
Varus Equinus
Varus
8 cm loss Val/shortening Valgus shortening
7 cm loss 6 cm loss
10 cm loss 8 cm loss 5 cm loss 6 cm loss
7 cm loss 9 cm loss
6 cm loss 12 cm loss 7.5 cm loss 4 cm loss 5 cm loss 10 cm loss
Gap/deformity
Bifocal Monofocal+deformity correction Monofocal+deformity correction Monofocal+deformity correction Bifocal Monofocal+deformity correction Bifocal Monofocal Monofocal Bifocal Monofocal+deformity correction Bifocal+deformity correction Monofocal Monofocal Monofocal Bifocal
Bifocal Bifocal
Bifocal Bifocal Bifocal Bifocal
Bifocal Bifocal
Bifocal Trifocal Bifocal Bifocal Bifocal Bifocal
Technique
Re-infection, sti knee Axial deviation, sti ankle Poor regenerate, refracture Depression, refracture Re-infection, sti knee Axial deviation, shortening with equinus, sti ankle Poor regenerate, sti knee Poor regenerate, skin invagination, bone graft, sti ankle Axial deviation, sti ankle Skin invagination, bone graft Nil Poor regenerate, wire cut through, sti ankle Nil Skin invagination, bone graft, sti ankle Sti ankle Nil Nil Wire cut through, sti knee Refracture, sti shoulder Sti knee Re-infection Sti knee, wire cut through Nil Nil Sti ankle Sti ankle Edema, sti knee sti knee Nil Sti ankle
Complications
510 L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
tion of non-viable bone and bone transport or compression/distraction osteosynthesis in a series of 30 cases. 2. Material and method Thirty cases of infected non-unions treated by Ilizarov frame were reviewed (Table 1). All patients were assessed at the time of admission to determine the bone involved, shortening, neurovascular de®cit, deformity, previous surgical procedures performed and function of relevant joints. Non-unions were classi®ed as sti or mobile on clinical examination. Preoperative radiological evaluation was done to determine the plane of deformity, assess alignment, classify nonunion as atrophic/hypertrophic and to look for signs of osteomyelitis. In patients with active draining sinuses/wounds either; (a) curettage of the sinus (eight cases), or (b) debridement accompanied by resection of nonviable bone was performed (12 cases). Pre-operative and intra-operative specimens were sent for microbiological assessment and appropriate antibiotics were given. There were 15 cases with bone loss, 10 cases of sti non-union (six of which had an associated deformity) and ®ve cases of mobile non-union (Fig. 1). The Ilizarov frame was assembled preoperatively for all patients. Patients with bone loss were planned for bifocal osteosynthesis and bone transport (14 cases). In one case of bone loss (case no. 2) where the bone loss was of 12 cm, trifocal osteosynthesis with proximal and distal corticotomies was performed (Fig 2a±e). Bifocal osteosynthesis was also planned for all patients with clinically mobile and radiologically atrophic non-unions (®ve cases). Monofocal osteosynthesis was done for all cases of sti nonunion (10 cases, six of which had associated deformity) (Fig. 3). Of a total of 19 cases of bifocal osteosynthesis, proximal tibial corticotomy was done in 14 and distal femoral in three. One patient underwent a distal tibial corticotomy since the proximal fragment was too small. Proximal femoral corticotomy was performed in another patient for similar reasons. Postoperative radiographs were taken to assess the corticotomy and placement of the wires. Ambulation
Fig. 1. Type of non-union.
511
was begun as soon as pain permitted with special emphasis on exercising the relevant joints to prevent contractures and joint stiness. Distraction of the corticotomy site for bone transport at the rate of 1 mm/ day was initiated, on average on the 7th day (5th±10th day). Once discharged, patients were regularly seen in the out patient department and assessed for any infection, loosening of wires, progress of deformity correction/quality of regenerate, docking site problems and neurovascular de®cit. Severe pin track sepsis, major frame reconstruction and bone grafting required further short periods of hospital stay. In all patients with bone loss, once docking was achieved, the frame was retained for at least the time needed for bone transport to ensure proper consolidation of the regenerate. Once consolidation of the regenerate was achieved, the frame was removed and the limb protected in a functional brace for a further period equal to the time from the application of the frame to its removal. In patients without the bone loss alternate compression/distraction at the site of nonunion (Accordian maneuver) was performed (Fig. 4). Once radiological signs of union were seen, the frame was retained for an additional period equal to the time from frame application to the appearance of these signs. Thereafter, the frame was removed and the limb protected in a functional brace for a further period equal to the time from the application of the frame to its removal. All patients were allowed to bear weight to pain tolerance with the frame and subsequently in the brace. 3. Results There were 23 tibial fractures, six femoral fractures and one fracture of the humerus in 27 men and three women with a median age of 34.5 years (12±65 years). The median time between injury and the application of the frame was 7 months and the median number of previous operations was two (Fig. 5). Of the 15 cases with a bone loss, 12 were in the tibia and three in the femur. The median length of segmental defect was 7 cm (4±12 cm). In all the cases the defect was successfully ®lled by external bone transport. Refracture was seen in three cases after frame removal and was due to signi®cant retrauma (cases 3,4, and 19). Union was achieved in all three cases by reapplication of the frame. Six patients with sti non-union and deformity underwent deformity correction followed by alternate compression/distraction (Accordian maneuver) at the non-union site. All the fractures united with correction of the deformity. One patient (case no. 6) required reapplication of the frame after 2 years due to development of limb length discrepancy, which was caused by the initial epiphyseal damage. This patient had a resul-
512
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
tant severe equinus deformity which was also corrected by using a foot plate assembly along with limb lengthening. There were 112 episodes of pin or wire track infection in the 30 patients, 24 episodes (in six patients) of
which required treatment in hospital including parentral antibiotics, antiseptic dressings and change of the oending wires. Others were managed as out patients by culture sensitivity, appropriate antibiotics and dressings. There were no episodes of wire breakage. The
Fig. 2. (a) Pre-operative radiograph of case no. 2 showing 12 cm of post-traumatic tibial bone loss. (b) Immediate postoperative radiograph of the same case with trifocal osteosynthesis. (c) Follow up X-ray before docking (note the axial deviation present at the docking site). (d,e) X-ray (anteroposterior and lateral view respectively) after frame removal showing union at docking site and good regenerate bone.
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
513
Fig. 2 (continued)
median time in the frame was 150 days. Axial deviation was observed at the docking site in three patients (cases 2, 6, and 9) and required the application of a translation assembly. Four episodes of wire cut through were observed in three patients (cases 12, 18, and 22) who had signi®cant preoperative osteoporosis as a result of the length of time since injury and disuse. Poor quality regenerate was seen in four cases (cases 3,7,8, and 12). This was managed eectively by slowing the distraction rate, encouraging the patient to actively use the limb and walk with a pair of crutches. As many as 19 of our patients had stiness of the relevant joint (knee/ankle/shoulder). Most patients had had multiple previous surgeries with resultant soft tissue scarring. Loss of motion averaged 45% at the knee and 60% at the ankle as compared with the opposite side. The median duration of follow-up after frame
Fig. 3. Surgical procedure performed.
removal was 23.5 months. The complications encountered in our study are enumerated in Table 2. The results were divided into bone results and functional results according to the classi®cation of the Association for the Study and Application of the Method of Ilizarov (Table 3). The bone results were determined according to four criteria: union, infection, residual deformity and limb length discrepancy. The fracture was considered united when it appeared so radiologically, when there was no motion at the site of the fracture after loosening of the connecting rods and when the patient was able to walk without pain at the fracture site after dynamisation of the assembly. Union was achieved in all cases. However excision of interposed soft tissue with freshening of the bone ends and cancellous bone grafting was required in three patients (cases 8,10, and 14) while three patients had refracture. In three cases, recurrence of infection occurred within 3 months of frame removal. This was managed by drainage, appropriate antibiotics and antiseptic dressings (cases 1,5, and 21). Infection was eectively controlled with the healing of all discharging sinuses and wounds when last assessed. No patient had a residual deformity of more than 78 and only one patient had a shortening of more than 2.5 cm. According to the protocol of the Association for the Study and Application of the Method of Ilizarov, a bone result cannot be graded as excellent unless union is achieved without the use of a bone graft. This requirement is strict and could cause an
514
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
otherwise excellent bone result to be classi®ed as poor because a relatively simple procedure had been performed [3]. An excellent bone result was de®ned as union, no infection, deformity of <78 and limb length discrepancy of <2.5 cm; a good result, as
union and any two of the other three criteria; a fair result, as union and one of the other criteria and a poor result, as non-union or refracture, or as union but none of the remaining three criteria that are required for an excellent result. According to
Fig. 4. (a,b) Preoperative radiographs of case no. 22 showing an infected intercondylar fracture femur with supracondylar extension. (c,d) Radiographs taken after monofocal osteosynthesis. (e,f) Radiograph after frame removal showing union in good position.
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
515
Fig. 4 (continued)
The functional results were considered excellent if the patient was active (that is, not inactive according to
the de®nition given above) and none of the other four criteria were applicable; good if the patient was active but one or two of the other criteria were applicable; fair if the patient was active but three or four of the other criteria were applicable; and poor if the patient was inactive regardless of whether other criteria were applicable. Of the 27 male patients, nine returned to their previous work (33.3%), 12 required a change from previous work (44.4%) but were gainfully employed while six did not return to any work at all (22.2%). Of the three female patients, two (66.7%) were able to perform some household activities while one (33.3%) could not (Fig. 6). Seventeen patients had a noticeable limp. Nineteen patients had stiness of either the knee or the ankle. Eight patients had signi®cant pain, which resulted in reduced activity or dis-
Fig. 5. Number of previous surgical procedures.
Fig. 6. Activity level of patients at last follow-up.
this system, the bone result was excellent in 20 patients (66.7%), good in four (13.3%), fair in none (0%) and poor in six (20%). The functional results were based on ®ve criteria: 1. A noteworthy limp; 2. Stiness of either the knee or the ankle (a loss of more than 158 of full extension of the knee or of 158 of dorsi¯exion of the ankle in comparison to the contralateral normal ankle); 3. Soft tissue sympathetic dystrophy; 4. Pain that reduced activity or disturbed sleep; 5. Inactivity (unemployment or an inability to return to daily activities because of the injury).
516
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
Table 2 Complications encountered Complications Problemsa
Obstaclesb
True complicationsc
1. 2. 3. 4. 5. 6. 7.
1. Wire cut through(3) 2. Skin invagination(3) 3. Pin track infection(6)
1. 2. 3. 4. 5.
Axial deviation(3) Poor quality regenerate(4) Transient pain(most patients) Transient edema(most patients) Depression(1) Pin track infection(20) Parasthesias(2)
Re-infection(3) Re-fracture(3) Joint stiness(19) Signi®cant pain(8) Edema(1)
a A problem is a diculty that arises during the distraction or consolidation period and that has fully resolved, after non-operative intervention, by the end of the treatment period. b An obstacle is a diculty that arises during the distraction or consolidation period and that has fully resolved, after operative intervention, by the end of the treatment period. c A true complication is any local or systemic diculty that occurs during distraction or consolidation and remains unresolved at the end of the treatment period, or any such diculty that occurs thereafter.
Table 3 Bone and functional results Bone results
Functional results
Case No.
Union
Grade
Limp
Sti joint
Sym. Dystrophy
Paina
Activity level
Grade
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Union,reinfection Union Union,re-fracture Union,re-fracture Union,re-infection Union,re-infection, limb len. Dis > 2.5 cm Union Union, bone graft Union Union, bone graft Union Union Union Union, bone graft Union Union Union Union Union, re-fracture Union Union, re-infection Union Union Union Union Union Union Union Union Union
Good Excellent Poor Poor Good Good Excellent Poor Excellent Poor Excellent Excellent Excellent Poor Excellent Excellent Excellent Excellent Poor Excellent Good Excellent Excellent Excellent Excellent Excellent Excellent Excellent Excellent Excellent
Yes No Yes Yes No Yes Yes Yes No No No Yes No Yes Yes No Yes Yes NA Yes No Yes No No Yes Yes Yes Yes No No
Knee Ankle None None Knee Ankle Knee Ankle Ankle None None Ankle None Ankle Ankle None None Knee Shoulder Knee None Knee None None Ankle Ankle Knee Knee None Ankle
No No No No No No No No No No No No No No No No No No No No No No No No No No Yes No No No
1 2 2 2 2 2 3 3 2 1 1 4 1 2 3 1 2 2 3 2 2 3 3 2 2 1 3 2 2 1
Modi®ed Returned Returned Returned Modi®ed Modi®ed Returned None Returned Modi®ed Returned None Modi®ed Modi®ed None Returned Modi®ed None None None Modi®ed Modi®ed Returned Modi®ed Modi®ed None Returned Modi®ed Modi®ed Modi®ed
Good Good Good Excellent Good Good Fair Poor Good Excellent Excellent Poor Excellent Good Poor Excellent Good Poor Poor Poor Excellent Fair Good Excellent Good Poor Fair Good Excellent Good
a 1 indicates no pain; 2, mild pain that does not interfere with activities or sleep; 3, moderate pain that reduces activities or sleep; 4, severe pain.
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
turbed sleep. According to these criteria, the functional results were excellent in eight patients (26.7%), good in 12 (40.0%), fair in three (10%) and poor in seven (23.3%). 4. Discussion The basic principles of the Ilizarov method Ð stable ®xation, a low energy osteotomy with gradual distraction and bone formation by intramembranous ossi®cation Ð are the same for bone transport as for lengthening [4]. Although the classical Ilizarov method called for healing of the docking site by gradual and prolonged compression [5], Western surgeons have found that supplementation with autogenous bone graft after operative debridement of the ends of the bone accelerates and facilitates healing [6,7]. In our study, operative treatment for interposed soft tissue was required in only three cases of bone transport (20%). Our experience con®rms Ilizarov's assertion that distraction alone is a potent stimulus, at least for non-unions of the hypertrophic type and is in agreement with other more recent work [8,9]. Reconstruction of infected non-union remains a dif®cult problem. We have found the Ilizarov technique to be eective in the treatment of infected non-unions, as it allows for the simultaneous treatment of bone loss, infection, non-union and deformity. In our series the bone results were excellent or good in 80% cases while functional results were excellent or good in 66.7% cases which compares favorably with the results of Dendrinos et al. [3] (78.5 and 64.3%, respectively). An excellent bone result even in a patient who has a severe infection, does not guarantee a good functional result since the functional result is aected by the condition of the nerves, vessels, joints, and Ð to a lesser degree Ð the bone [3]. The dilemma faced by the treating surgeon is to decide when to opt for amputation and prosthetic replacement over salvage. Most of these patients are neglected cases with a history of multiple past surgical procedures with associated complications of sti joints, contractures, persistent infection, bone loss, disuse osteoporosis, neurovascular de®cit etc. The cost of major limb salvage with the Ilizarov method has been compared with that of amputation by Williams [10]. While the cost of the acute care necessitated by the amputation was much lower, the projected costs of the prosthetic care for the remainder of the patients' life resulted in greater overall costs. We believe that all such patients who are psychologically stable and with a sensate foot should be considered for salvage procedure, keeping in mind the ®nancial burden and
517
psychological support needed for the prolonged period of treatment. All our patients willingly opted for salvage in preference to amputation, which is not surprising considering the social stigma attached with amputees in the Indian subcontinent. For patients who could not aord the implants an inventory of the Ilizarov rings and wire ®xation bolts was maintained which were recycled. Since most of the patients were treated on an outpatient basis after frame application, the cost to the hospital was minimal. However, repeated radiological assessment and the long period of treatment requiring absence from work did cause a drain on the ®nancial resources of most patients. Our patients were able to cope with these problems since most of the patients came from joint families, which provided some ®nancial and psychological support. It is dicult to precisely de®ne the indications for preservation and reconstruction of very severe injuries. The surgical team has to appreciate the length, complications and cost of the treatment. Patients must be made aware of the limitations of functional results and the possible diculty of return to work despite the reconstructive attempt.
References [1] Ilizarov GA. The tension±stress eect on the genesis and growth of tissues: Part I: The in¯uence of stability of ®xation and soft tissue preservation. Clin Orthop 1989;238:249. [2] Monticelli G, Spinelli R, Bonnucci E. Distraction epiphysiolysis as a method of limb lengthening: II: Morphologic investigations. Clin Orthop 1981;154:262. [3] Dendrinos GK, Kontos S, Lyrisis E. Use of the Ilizarov technique for treatment of non-union of the tibia associated with infection. JBJS (Am) 1995;77A(6):835±46. [4] de Pablos J, Barrios C, Alfaro C, Canadell J. Large experimental segmental bone defects treated by bone transport with monolateral external ®xator. Clin Orthop 1994;298:259±65. [5] Cattaneo R, Catagni M, Johnson EE. The treatment of infected non-unions and segmental defects of tibia by the method of Ilizarov. Clin Orthop 1992;280:143±52. [6] Cierny G, Zorn KE. Segmental tibial defects, comparing conventional and Ilizarov methodologies. Clin Orthop 1994;301:118±23. [7] Green SA. Skeletal defects. A comparison of bone grafting and bone transport for segmental skeletal defects. Clin Orthop 1994;301:111±7. [8] Marsh DR, Shah S, Elliott J, Kurdy N. The Ilizarov method in non-union, malunion and infection of fractures. J Bone Joint Surg (Br) 1997;79-B:273±9. [9] Saleh M, Royston S. Management of nonunion of fractures by distraction with correction of angulation and shortening. J Bone Joint Surg (Br) 1996;78-B:105±9. [10] Williams MO. Long-term cost comparison of major limb salvage using the Ilizarov method versus amputation. Clin Orthop 1994;301:156±8.
www.elsevier.com/locate/injury
The Ilizarov method in infected nonunion of fractures Lalit Maini*, Manish Chadha, Jashan Vishwanath, Sudhir Kapoor, Anil Mehtani, B.K. Dhaon Department of Orthopaedics, Maulana Azad Medical College, New Delhi 110002, India Accepted 14 February 2000
Abstract Thirty patients with infected non-union of long bones were treated with radical resection of the necrotic bone and bone transport or compression/distraction osteosynthesis. Non-union, infection, deformity, bone gap and shortening were all addressed simultaneously using the Ilizarov principles. There were 15 cases with bone loss ranging from 4 to 12 cm (median bone gap of 7 cm), 10 cases of sti non-union (six of which had an associated deformity) and ®ve cases of mobile non-union. The median time in the Ilizarov frame was 150 days. Median follow up time after frame removal was 23.5 months. Bone grafting at the docking site was only required in three cases (10%). There were three cases of refracture (10%) and three cases of recurrence of infection (10%). The bone result was excellent in 21 patients (70%), good in three (10%), fair in none (0%) and poor in six (20%). The functional results were excellent in eight patients (26.7%), good in 12 (40.0%), fair in three (10%) and poor in seven (23.3%). It is dicult to precisely de®ne the indications for preservation and reconstruction of severe injuries. The surgical team has to take into account the length, disability, complications and cost of treatment. Patients must be aware of the limitations of functional results and the possible diculty of return to work despite the reconstructive attempt. 7 2000 Elsevier Science Ltd. All rights reserved.
1. Introduction Infected non-union of fractures is one of the major problems in Orthopaedic surgery. There are many available alternatives in the management of infected non-unions. These include extensive debridement and local soft tissue rotational ¯aps, packing the defects with antibiotic impregnated beads, Papineau type open cancellous bone grafting, tibio®bular synostosis, cancellous allograft in ®brin sealant mixed with antibiotics, and/or free microvascular soft tissue and bone transplants. None of the above mentioned techniques aord the surgeon the ability to correct deformities, eliminate prolonged pre- and postoperative antibiotic therapy, regenerate new bone tissue without the use of bone grafts, progressively lengthen the extremity, and allow * Corresponding author. 22, Samachar Apartment, Mayur Vihar, Phase I, New Delhi 110091, India.
weight bearing during the treatment period simultaneously. All of these capabilities are possible with application of the techniques of Ilizarov. Progressive bone histogenesis after corticotomy and bone transport has bene®ted from the work of Ilizarov [1] and Monticelli and co-workers [2]. It helps to eradicate bone infection and to unite fracture non-unions. It may be a good technique for some dicult post-traumatic conditions. Application of Ilizarov techniques in the treatment of an infected non-union depends on the type of nonunion (hypertrophic or atrophic), the extent of infection and the condition of soft tissues. The wider the area of necrotic bone, the greater the infection. To ensure elimination of infection, it is necessary to perform complete resection of the necrotic bone and infected segments. Bone transport is then used to reconstruct the residual segmental defect. In this study we report on the treatment of infected non-unions using the techniques of debridement, resec-
0020-1383/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 1 3 8 3 ( 0 0 ) 0 0 0 3 6 - X
28,M 36,M 12,M 40,M
18,M 20,M
40,M 42,M 52,F 26,F 38,M 28,M 46,M 50,M 60,M 45,M 40,M 24,M 55,M 65,F 26,M 34,M
9 10 11 12
13 14
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Tibia Tibia Tibia Tibia Humerus Tibia Tibia Femur Tibia Tibia Tibia Tibia Femur Femur Tibia Tibia
Tibia Tibia
Tibia Tibia Femur Tibia
Femur Tibia
Femur Tibia Tibia Tibia Tibia Tibia
Site
No gap/deformity.
25,M 35,M
7 8
a
30,M 19,M 26,M 35,M 30,M 17,M
Age/sex
1 2 3 4 5 6
Case No.
Table 1 Patient pro®le
Quiescent Draining Quiescent Quiescent Draining Quiescent Draining Draining Quiescent Draining Quiescent Quiescent Draining Draining Quiescent Draining
Quiescent Draining
Quiescent Draining Draining Quiescent
Draining Draining
Draining Draining Draining Quiescent Draining Quiescent
Infection
± Staph. Aureus ± ± Mixed ¯ora ± Staph. Aureus Staph. Aureus ± Klebsiella ± ± Mixed ¯ora Staph. Aureus ± E. coli
± E. coli
± Staph. Aureus Mixed ¯ora ±
E. coli Staph. Aureus
Staph. Aureus Staph. Aureus Klebsiella ± Staph. Aureus ±
Organism
3 1 1 2 4 1 5 2 1 1 3 3 1 1 2 2
4 2
3 2 3 2
3 2
4 2 2 2 2 1
Previous surgery
49 38 36 24 25 30 54 30 24 20 19 22 18 19 20 20
48 37
20 24 26 19
19 30
24 30 27 31 42 43
Follow up (month)
a
Varus Equinus
Varus
8 cm loss Val/shortening Valgus shortening
7 cm loss 6 cm loss
10 cm loss 8 cm loss 5 cm loss 6 cm loss
7 cm loss 9 cm loss
6 cm loss 12 cm loss 7.5 cm loss 4 cm loss 5 cm loss 10 cm loss
Gap/deformity
Bifocal Monofocal+deformity correction Monofocal+deformity correction Monofocal+deformity correction Bifocal Monofocal+deformity correction Bifocal Monofocal Monofocal Bifocal Monofocal+deformity correction Bifocal+deformity correction Monofocal Monofocal Monofocal Bifocal
Bifocal Bifocal
Bifocal Bifocal Bifocal Bifocal
Bifocal Bifocal
Bifocal Trifocal Bifocal Bifocal Bifocal Bifocal
Technique
Re-infection, sti knee Axial deviation, sti ankle Poor regenerate, refracture Depression, refracture Re-infection, sti knee Axial deviation, shortening with equinus, sti ankle Poor regenerate, sti knee Poor regenerate, skin invagination, bone graft, sti ankle Axial deviation, sti ankle Skin invagination, bone graft Nil Poor regenerate, wire cut through, sti ankle Nil Skin invagination, bone graft, sti ankle Sti ankle Nil Nil Wire cut through, sti knee Refracture, sti shoulder Sti knee Re-infection Sti knee, wire cut through Nil Nil Sti ankle Sti ankle Edema, sti knee sti knee Nil Sti ankle
Complications
510 L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
tion of non-viable bone and bone transport or compression/distraction osteosynthesis in a series of 30 cases. 2. Material and method Thirty cases of infected non-unions treated by Ilizarov frame were reviewed (Table 1). All patients were assessed at the time of admission to determine the bone involved, shortening, neurovascular de®cit, deformity, previous surgical procedures performed and function of relevant joints. Non-unions were classi®ed as sti or mobile on clinical examination. Preoperative radiological evaluation was done to determine the plane of deformity, assess alignment, classify nonunion as atrophic/hypertrophic and to look for signs of osteomyelitis. In patients with active draining sinuses/wounds either; (a) curettage of the sinus (eight cases), or (b) debridement accompanied by resection of nonviable bone was performed (12 cases). Pre-operative and intra-operative specimens were sent for microbiological assessment and appropriate antibiotics were given. There were 15 cases with bone loss, 10 cases of sti non-union (six of which had an associated deformity) and ®ve cases of mobile non-union (Fig. 1). The Ilizarov frame was assembled preoperatively for all patients. Patients with bone loss were planned for bifocal osteosynthesis and bone transport (14 cases). In one case of bone loss (case no. 2) where the bone loss was of 12 cm, trifocal osteosynthesis with proximal and distal corticotomies was performed (Fig 2a±e). Bifocal osteosynthesis was also planned for all patients with clinically mobile and radiologically atrophic non-unions (®ve cases). Monofocal osteosynthesis was done for all cases of sti nonunion (10 cases, six of which had associated deformity) (Fig. 3). Of a total of 19 cases of bifocal osteosynthesis, proximal tibial corticotomy was done in 14 and distal femoral in three. One patient underwent a distal tibial corticotomy since the proximal fragment was too small. Proximal femoral corticotomy was performed in another patient for similar reasons. Postoperative radiographs were taken to assess the corticotomy and placement of the wires. Ambulation
Fig. 1. Type of non-union.
511
was begun as soon as pain permitted with special emphasis on exercising the relevant joints to prevent contractures and joint stiness. Distraction of the corticotomy site for bone transport at the rate of 1 mm/ day was initiated, on average on the 7th day (5th±10th day). Once discharged, patients were regularly seen in the out patient department and assessed for any infection, loosening of wires, progress of deformity correction/quality of regenerate, docking site problems and neurovascular de®cit. Severe pin track sepsis, major frame reconstruction and bone grafting required further short periods of hospital stay. In all patients with bone loss, once docking was achieved, the frame was retained for at least the time needed for bone transport to ensure proper consolidation of the regenerate. Once consolidation of the regenerate was achieved, the frame was removed and the limb protected in a functional brace for a further period equal to the time from the application of the frame to its removal. In patients without the bone loss alternate compression/distraction at the site of nonunion (Accordian maneuver) was performed (Fig. 4). Once radiological signs of union were seen, the frame was retained for an additional period equal to the time from frame application to the appearance of these signs. Thereafter, the frame was removed and the limb protected in a functional brace for a further period equal to the time from the application of the frame to its removal. All patients were allowed to bear weight to pain tolerance with the frame and subsequently in the brace. 3. Results There were 23 tibial fractures, six femoral fractures and one fracture of the humerus in 27 men and three women with a median age of 34.5 years (12±65 years). The median time between injury and the application of the frame was 7 months and the median number of previous operations was two (Fig. 5). Of the 15 cases with a bone loss, 12 were in the tibia and three in the femur. The median length of segmental defect was 7 cm (4±12 cm). In all the cases the defect was successfully ®lled by external bone transport. Refracture was seen in three cases after frame removal and was due to signi®cant retrauma (cases 3,4, and 19). Union was achieved in all three cases by reapplication of the frame. Six patients with sti non-union and deformity underwent deformity correction followed by alternate compression/distraction (Accordian maneuver) at the non-union site. All the fractures united with correction of the deformity. One patient (case no. 6) required reapplication of the frame after 2 years due to development of limb length discrepancy, which was caused by the initial epiphyseal damage. This patient had a resul-
512
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
tant severe equinus deformity which was also corrected by using a foot plate assembly along with limb lengthening. There were 112 episodes of pin or wire track infection in the 30 patients, 24 episodes (in six patients) of
which required treatment in hospital including parentral antibiotics, antiseptic dressings and change of the oending wires. Others were managed as out patients by culture sensitivity, appropriate antibiotics and dressings. There were no episodes of wire breakage. The
Fig. 2. (a) Pre-operative radiograph of case no. 2 showing 12 cm of post-traumatic tibial bone loss. (b) Immediate postoperative radiograph of the same case with trifocal osteosynthesis. (c) Follow up X-ray before docking (note the axial deviation present at the docking site). (d,e) X-ray (anteroposterior and lateral view respectively) after frame removal showing union at docking site and good regenerate bone.
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
513
Fig. 2 (continued)
median time in the frame was 150 days. Axial deviation was observed at the docking site in three patients (cases 2, 6, and 9) and required the application of a translation assembly. Four episodes of wire cut through were observed in three patients (cases 12, 18, and 22) who had signi®cant preoperative osteoporosis as a result of the length of time since injury and disuse. Poor quality regenerate was seen in four cases (cases 3,7,8, and 12). This was managed eectively by slowing the distraction rate, encouraging the patient to actively use the limb and walk with a pair of crutches. As many as 19 of our patients had stiness of the relevant joint (knee/ankle/shoulder). Most patients had had multiple previous surgeries with resultant soft tissue scarring. Loss of motion averaged 45% at the knee and 60% at the ankle as compared with the opposite side. The median duration of follow-up after frame
Fig. 3. Surgical procedure performed.
removal was 23.5 months. The complications encountered in our study are enumerated in Table 2. The results were divided into bone results and functional results according to the classi®cation of the Association for the Study and Application of the Method of Ilizarov (Table 3). The bone results were determined according to four criteria: union, infection, residual deformity and limb length discrepancy. The fracture was considered united when it appeared so radiologically, when there was no motion at the site of the fracture after loosening of the connecting rods and when the patient was able to walk without pain at the fracture site after dynamisation of the assembly. Union was achieved in all cases. However excision of interposed soft tissue with freshening of the bone ends and cancellous bone grafting was required in three patients (cases 8,10, and 14) while three patients had refracture. In three cases, recurrence of infection occurred within 3 months of frame removal. This was managed by drainage, appropriate antibiotics and antiseptic dressings (cases 1,5, and 21). Infection was eectively controlled with the healing of all discharging sinuses and wounds when last assessed. No patient had a residual deformity of more than 78 and only one patient had a shortening of more than 2.5 cm. According to the protocol of the Association for the Study and Application of the Method of Ilizarov, a bone result cannot be graded as excellent unless union is achieved without the use of a bone graft. This requirement is strict and could cause an
514
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
otherwise excellent bone result to be classi®ed as poor because a relatively simple procedure had been performed [3]. An excellent bone result was de®ned as union, no infection, deformity of <78 and limb length discrepancy of <2.5 cm; a good result, as
union and any two of the other three criteria; a fair result, as union and one of the other criteria and a poor result, as non-union or refracture, or as union but none of the remaining three criteria that are required for an excellent result. According to
Fig. 4. (a,b) Preoperative radiographs of case no. 22 showing an infected intercondylar fracture femur with supracondylar extension. (c,d) Radiographs taken after monofocal osteosynthesis. (e,f) Radiograph after frame removal showing union in good position.
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
515
Fig. 4 (continued)
The functional results were considered excellent if the patient was active (that is, not inactive according to
the de®nition given above) and none of the other four criteria were applicable; good if the patient was active but one or two of the other criteria were applicable; fair if the patient was active but three or four of the other criteria were applicable; and poor if the patient was inactive regardless of whether other criteria were applicable. Of the 27 male patients, nine returned to their previous work (33.3%), 12 required a change from previous work (44.4%) but were gainfully employed while six did not return to any work at all (22.2%). Of the three female patients, two (66.7%) were able to perform some household activities while one (33.3%) could not (Fig. 6). Seventeen patients had a noticeable limp. Nineteen patients had stiness of either the knee or the ankle. Eight patients had signi®cant pain, which resulted in reduced activity or dis-
Fig. 5. Number of previous surgical procedures.
Fig. 6. Activity level of patients at last follow-up.
this system, the bone result was excellent in 20 patients (66.7%), good in four (13.3%), fair in none (0%) and poor in six (20%). The functional results were based on ®ve criteria: 1. A noteworthy limp; 2. Stiness of either the knee or the ankle (a loss of more than 158 of full extension of the knee or of 158 of dorsi¯exion of the ankle in comparison to the contralateral normal ankle); 3. Soft tissue sympathetic dystrophy; 4. Pain that reduced activity or disturbed sleep; 5. Inactivity (unemployment or an inability to return to daily activities because of the injury).
516
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
Table 2 Complications encountered Complications Problemsa
Obstaclesb
True complicationsc
1. 2. 3. 4. 5. 6. 7.
1. Wire cut through(3) 2. Skin invagination(3) 3. Pin track infection(6)
1. 2. 3. 4. 5.
Axial deviation(3) Poor quality regenerate(4) Transient pain(most patients) Transient edema(most patients) Depression(1) Pin track infection(20) Parasthesias(2)
Re-infection(3) Re-fracture(3) Joint stiness(19) Signi®cant pain(8) Edema(1)
a A problem is a diculty that arises during the distraction or consolidation period and that has fully resolved, after non-operative intervention, by the end of the treatment period. b An obstacle is a diculty that arises during the distraction or consolidation period and that has fully resolved, after operative intervention, by the end of the treatment period. c A true complication is any local or systemic diculty that occurs during distraction or consolidation and remains unresolved at the end of the treatment period, or any such diculty that occurs thereafter.
Table 3 Bone and functional results Bone results
Functional results
Case No.
Union
Grade
Limp
Sti joint
Sym. Dystrophy
Paina
Activity level
Grade
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Union,reinfection Union Union,re-fracture Union,re-fracture Union,re-infection Union,re-infection, limb len. Dis > 2.5 cm Union Union, bone graft Union Union, bone graft Union Union Union Union, bone graft Union Union Union Union Union, re-fracture Union Union, re-infection Union Union Union Union Union Union Union Union Union
Good Excellent Poor Poor Good Good Excellent Poor Excellent Poor Excellent Excellent Excellent Poor Excellent Excellent Excellent Excellent Poor Excellent Good Excellent Excellent Excellent Excellent Excellent Excellent Excellent Excellent Excellent
Yes No Yes Yes No Yes Yes Yes No No No Yes No Yes Yes No Yes Yes NA Yes No Yes No No Yes Yes Yes Yes No No
Knee Ankle None None Knee Ankle Knee Ankle Ankle None None Ankle None Ankle Ankle None None Knee Shoulder Knee None Knee None None Ankle Ankle Knee Knee None Ankle
No No No No No No No No No No No No No No No No No No No No No No No No No No Yes No No No
1 2 2 2 2 2 3 3 2 1 1 4 1 2 3 1 2 2 3 2 2 3 3 2 2 1 3 2 2 1
Modi®ed Returned Returned Returned Modi®ed Modi®ed Returned None Returned Modi®ed Returned None Modi®ed Modi®ed None Returned Modi®ed None None None Modi®ed Modi®ed Returned Modi®ed Modi®ed None Returned Modi®ed Modi®ed Modi®ed
Good Good Good Excellent Good Good Fair Poor Good Excellent Excellent Poor Excellent Good Poor Excellent Good Poor Poor Poor Excellent Fair Good Excellent Good Poor Fair Good Excellent Good
a 1 indicates no pain; 2, mild pain that does not interfere with activities or sleep; 3, moderate pain that reduces activities or sleep; 4, severe pain.
L. Maini et al. / Injury, Int. J. Care Injured 31 (2000) 509±517
turbed sleep. According to these criteria, the functional results were excellent in eight patients (26.7%), good in 12 (40.0%), fair in three (10%) and poor in seven (23.3%). 4. Discussion The basic principles of the Ilizarov method Ð stable ®xation, a low energy osteotomy with gradual distraction and bone formation by intramembranous ossi®cation Ð are the same for bone transport as for lengthening [4]. Although the classical Ilizarov method called for healing of the docking site by gradual and prolonged compression [5], Western surgeons have found that supplementation with autogenous bone graft after operative debridement of the ends of the bone accelerates and facilitates healing [6,7]. In our study, operative treatment for interposed soft tissue was required in only three cases of bone transport (20%). Our experience con®rms Ilizarov's assertion that distraction alone is a potent stimulus, at least for non-unions of the hypertrophic type and is in agreement with other more recent work [8,9]. Reconstruction of infected non-union remains a dif®cult problem. We have found the Ilizarov technique to be eective in the treatment of infected non-unions, as it allows for the simultaneous treatment of bone loss, infection, non-union and deformity. In our series the bone results were excellent or good in 80% cases while functional results were excellent or good in 66.7% cases which compares favorably with the results of Dendrinos et al. [3] (78.5 and 64.3%, respectively). An excellent bone result even in a patient who has a severe infection, does not guarantee a good functional result since the functional result is aected by the condition of the nerves, vessels, joints, and Ð to a lesser degree Ð the bone [3]. The dilemma faced by the treating surgeon is to decide when to opt for amputation and prosthetic replacement over salvage. Most of these patients are neglected cases with a history of multiple past surgical procedures with associated complications of sti joints, contractures, persistent infection, bone loss, disuse osteoporosis, neurovascular de®cit etc. The cost of major limb salvage with the Ilizarov method has been compared with that of amputation by Williams [10]. While the cost of the acute care necessitated by the amputation was much lower, the projected costs of the prosthetic care for the remainder of the patients' life resulted in greater overall costs. We believe that all such patients who are psychologically stable and with a sensate foot should be considered for salvage procedure, keeping in mind the ®nancial burden and
517
psychological support needed for the prolonged period of treatment. All our patients willingly opted for salvage in preference to amputation, which is not surprising considering the social stigma attached with amputees in the Indian subcontinent. For patients who could not aord the implants an inventory of the Ilizarov rings and wire ®xation bolts was maintained which were recycled. Since most of the patients were treated on an outpatient basis after frame application, the cost to the hospital was minimal. However, repeated radiological assessment and the long period of treatment requiring absence from work did cause a drain on the ®nancial resources of most patients. Our patients were able to cope with these problems since most of the patients came from joint families, which provided some ®nancial and psychological support. It is dicult to precisely de®ne the indications for preservation and reconstruction of very severe injuries. The surgical team has to appreciate the length, complications and cost of the treatment. Patients must be made aware of the limitations of functional results and the possible diculty of return to work despite the reconstructive attempt.
References [1] Ilizarov GA. The tension±stress eect on the genesis and growth of tissues: Part I: The in¯uence of stability of ®xation and soft tissue preservation. Clin Orthop 1989;238:249. [2] Monticelli G, Spinelli R, Bonnucci E. Distraction epiphysiolysis as a method of limb lengthening: II: Morphologic investigations. Clin Orthop 1981;154:262. [3] Dendrinos GK, Kontos S, Lyrisis E. Use of the Ilizarov technique for treatment of non-union of the tibia associated with infection. JBJS (Am) 1995;77A(6):835±46. [4] de Pablos J, Barrios C, Alfaro C, Canadell J. Large experimental segmental bone defects treated by bone transport with monolateral external ®xator. Clin Orthop 1994;298:259±65. [5] Cattaneo R, Catagni M, Johnson EE. The treatment of infected non-unions and segmental defects of tibia by the method of Ilizarov. Clin Orthop 1992;280:143±52. [6] Cierny G, Zorn KE. Segmental tibial defects, comparing conventional and Ilizarov methodologies. Clin Orthop 1994;301:118±23. [7] Green SA. Skeletal defects. A comparison of bone grafting and bone transport for segmental skeletal defects. Clin Orthop 1994;301:111±7. [8] Marsh DR, Shah S, Elliott J, Kurdy N. The Ilizarov method in non-union, malunion and infection of fractures. J Bone Joint Surg (Br) 1997;79-B:273±9. [9] Saleh M, Royston S. Management of nonunion of fractures by distraction with correction of angulation and shortening. J Bone Joint Surg (Br) 1996;78-B:105±9. [10] Williams MO. Long-term cost comparison of major limb salvage using the Ilizarov method versus amputation. Clin Orthop 1994;301:156±8.