|Year : 2019 | Volume
| Issue : 2 | Page : 71-78
Ilizarov techniques for upper tibial nonunions: How difficult is it to achieve excellent results?
Milind Madhav Chaudhary, Monish Malhotra, Umamahesh Neeli, Kelvin Vaishnani, Suman Banik, Naeem Jagani
Center for Ilizarov Techniques, Akola, Maharashtra, India
|Date of Submission||21-Dec-2019|
|Date of Decision||23-Dec-2019|
|Date of Acceptance||23-Dec-2019|
|Date of Web Publication||31-Dec-2019|
Dr. Milind Madhav Chaudhary
Center for Ilizarov Techniques, Akola - 444 001, Maharashtra
Source of Support: None, Conflict of Interest: None
Introduction: Upper tibial nonunions (UTNUs) pose challenging difficulties such as soft-tissue scarring, infection, deformity, shortening, and small fragments with intra-articular extension. We present a retrospective analysis of probably the largest series of UTNUs using the Ilizarov techniques. Factors determining the outcome based on difficulty scores and NU scores were also assessed. Patients and Methods: Forty-one patients with UTNU (within 10 cm of joint line) were evaluated retrospectively. Fourteen were aseptic and 27 were septic. The mean NU distance was 5.6 cm from the knee joint. The Association for the Study and Application of Methods of Ilizarov (ASAMI) functional and bone scores were evaluated and compared with NU severity score (NUSS), NU level of difficulty score (NULODS), and infection severity score (ISS). External fixation duration (EFD) was compared to NUSS and NULODS to determine which is better in predicting the outcome. We also compared our results with the literature. Results: We achieved union in all except one. The mean EFD was 333 days. The mean regenerate lengthening was 8.6 cm. The ASAMI bone score was: excellent – 16, good – 22, fair – 2, poor – 1. The ASAMI functional score was: excellent – 16, good – 14, fair – 9, poor – 2, and failure – 0. We found a positive correlation between NUSS, NULODS, and ISS. EFD correlated better with overall NULODS and subset of NULODS gap score compared to NUSS and NUSS gap score, respectively. Conclusions: Ilizarov is an effective method to treat UTNU. It corrects the deformity, provides stable fixation, achieves union, and corrects limb length discrepancy.
Keywords: Association for the Study and Application of Methods of Ilizarov score, distal tibia lengthening, Ilizarov, infected nonunion, infection severity score, nonunion level of difficulty score, nonunion severity score, upper tibial nonunion
|How to cite this article:|
Chaudhary MM, Malhotra M, Neeli U, Vaishnani K, Banik S, Jagani N. Ilizarov techniques for upper tibial nonunions: How difficult is it to achieve excellent results?. J Limb Lengthen Reconstr 2019;5:71-8
|How to cite this URL:|
Chaudhary MM, Malhotra M, Neeli U, Vaishnani K, Banik S, Jagani N. Ilizarov techniques for upper tibial nonunions: How difficult is it to achieve excellent results?. J Limb Lengthen Reconstr [serial online] 2019 [cited 2020 Nov 26];5:71-8. Available from: https://www.jlimblengthrecon.org/text.asp?2019/5/2/71/274569
| Introduction|| |
Upper tibial nonunions (UTNUs) pose several challenges to the treating surgeon. They frequently result from fractures after high-energy trauma. There are varying degrees of soft-tissue loss including fibrosis, scarring, adherence to bone, or even absence of skin cover. The underlying fracture may be intra-articular, comminuted, or have loss of bone, especially anteriorly, after multiple surgeries. Bony deformity, gap, and shortening are common. The bone fragments are likely to be osteoporotic, denying stable fixation. UTNUs – defined as being within 10 cm from the articular surface – also make stable internal fixation rather difficult.
Infection is common after failed surgery, and the presence of suppuration precludes safe internal fixation. The incidence of proximal tibia infection following fixation has been reported in the literature  as 8%–23%. Dormant infection coupled with poor soft tissues can flare up if locking plates are used for stabilization. Infection is difficult to eradicate. With any flare-up of infection, bone grafts may be resorbed.
Under these circumstances, the minimally invasive Ilizarov external fixator successfully allows stable fixation of small fragments, correction of deformities, allows sustained, and progressive compression at NU ends to allow reliable healing. A distal tibial corticotomy  permits regenerate formation to achieve either lengthening or filling up bone gaps at the upper end.
Difficulties , persist despite the use of the Ilizarov fixator. Prolonged fixator duration with pin infections and joint stiffness is common. A NU severity score  (NUSS) has been proposed by Calori which could help decide the modality of treatment.
We asked if NUSS will accurately help predict which UTNU will achieve an excellent result. The score, however, does not consider grades of severity of infection and does not give progressively increasing weightage to larger bony gaps and shortening. We felt the need to predict difficulties in achieving all treatment goals, namely union, eradication of infection, correction of deformities, and equalization of limb lengths as dictated by the Association for the Study and Application of Methods of Ilizarov score (ASAMI)-bone score (BS).
We formulated a NU level of difficulty score (NULODS) [Table 1] which we felt better helps predict the difficulties in achieving an excellent result and compared it with the NUSS. It contains the infection severity score (ISS) which extensively grades infection using seven parameters. It also gives more weightage to bone gap and shortening and area of contact at NU ends. We also compared the results of treatment for the septic versus the aseptic NUs.
In this study, we retrospectively studied 41 consecutive UTNUs over the past 10 years treated at our institute using the Ilizarov techniques. We analyzed our results using the ASAMI-BS and ASAMI functional scores (ASAMI-FS). We also aimed to assess the utility of the NULODS and the ISS , contained within it, in choosing the modality of treatment. We asked if either the NUSS or NULODS would be able to predict which NUs would pose more difficulties – especially duration of external fixation and equalization of limb lengths – or correlate better in achieving excellent results.
| Patients and Methods|| |
We retrospectively evaluated 41 patients with an UTNU. We included patients with a NU within 10 cm from the knee joint. Clinical data were collected from case records and radiographs evaluated using Carestream Image Suite MiniPACS v. A 2017. The mean age was 38.3 years (9–85).
Males dominated (35/41). Most had a right side (25R/16 L) involvement. Twenty-seven (66%) patients were infected. The most common bone geometry at NU surface was comminuted (30), of which 12 had bone loss. Six patients had NU extending into the joint. Four patients had a short oblique NU and only one patient had a transverse NU.
All patients with infection had a two-staged procedure. Debridement was performed in the first stage and 17 patients had antibiotic beads/rods. Due to proximity of NUs to the knee, 18 had a femur frame  and nine required foot frame to add stability to the distal tibial corticotomy construct. We performed a distal tibial corticotomy , in 12 of our patients. Fourteen patients required additional procedures including bone marrow injection  (10) and bone grafting (3) with one who had both bone marrow injection and bone grafting.
The mean distance of the NU from the joint line was 5.7 cm (0–10 cm). Twelve had a NU <5 cm from the joint line. A note of associated injuries along with UTNU was also made [Table 2].
NUSS and NULODS were calculated for all patients before surgery. The ISS was calculated for septic group and considered as zero for aseptic group. We calculated regenerate length, postoperative medial proximal tibial angle (MPTA), and posterior proximal tibial angle (PPTA) on full-length tibia radiographs. We computed the ASAMI-BS and ASAMI-FS for all patients and tried to correlate them with NUSS and NULODS.
Operative technique and postoperative care
Standard operative techniques were used to perform debridement hardware removal, insert antibiotic impregnated cement rods, and beads in infected cases. A waiting period of 6 weeks with reducing cell counts, sedimentation rates, and C-reactive protein levels were used, and a hybrid Ilizarov fixator was applied in the second stage [Figure 1]. Most UT fragments were small and fixated with only wires. Where bone fragments were larger, 6 mm titanium half pins (Uma Surgicals, Mumbai, Maharashtra, India) were used for fixation using drilling with saline cooling. Femur fixations were all done with half pins without wires in selected cases. The knee was put through range of motion on the table. The femoral and foot rings were connected with hinges and removed to allow postoperative range of knee motion. This allowed better fixation as well as ability to correct flexion or equinus contractures.
|Figure 1: (a and b) Preoperative radiograph of a 55-year-old female with an infected upper tibial nonunion (nonunion severity score = 60, nonunion level of difficulty score = 46). She was initially treated by debridement and antibiotic cement rod. (c and d) Three-ring Ilizarov fixator has corrected varus and procurvatum enabled by a proximal fibular osteotomy. (e) High comfort levels maintained in the fixator. (f and g) Postoperative radiographs showing good alignment and sound union. (h and i) The patient can sit cross-legged and has good alignment with excellent Association for the Study and Application of Methods of Ilizarov bone and functional score|
Click here to view
Foot fixation was achieved with one or two half pins in the calcaneum, a wire through the neck of the talus, and two wires through the forefoot attached to a horseshoe ring. Outriggers were attached to allow walking.
Routine postoperative care was taken to ensure pain-free movements and prevent pin-tract infections. Mild-to-moderate pin-tract infections occurred in many but did not interfere with results of treatment.
Patient data were maintained in MS Excel (Microsoft, Redmond, USA), and statistical analysis was performed using IBM SPSS (IBM, Chicago, IL, USA). Means of MPTA, PPTA, NUSS, NULODS, and external fixation duration (EFD) and ISS are calculated and tabulated in [Table 3].
NUSS and NULODS were compared to each other using Pearson's correlation test. NUSS and NULODS were also compared with ASAMI-FS and ASAMI-BS using ANOVA t- test. ISS was compared to NUSS and NULODS using Spearman's correlation test.
| Results|| |
Forty of the 41 (97.5%) patients united. One patient did not have sound union due to poor bony contact from anterior bony defect due to a failed plating, infection, and bony resection. One infected intra-articular NU united, but pain from articular damage required a knee arthrodesis.
Clinically, the patients were assessed by ASAMI-FS. The score was excellent in 16 (40%), good in 14 (35%), fair in 9 (22.5%), and poor in 2 (5%) patients [Table 4]. The mean EFD was 333 days. The mean EFD for septic cases was 353 days and in aseptic cases was 292.5 days. The difference was not statistically significant (P = 0.3).
|Table 4: Association for the Study and Application of Methods of Ilizarov function score|
Click here to view
The mean preoperative MPTA (82.3° ± 8.5°) and preoperative PPTA (83.3° ± 13.6°) were compared to the postoperative measurements (MPTA = 87.4. ± 5.6 and PPTA = 87.1 ± 7.7) Difference of means was statistically significant (P value for MPTA = 0.0004 and for PPTA = 0.0004) [Table 3].
The mean regenerate length was 8.6 cm (2.2–21.3 cm). ASAMI-BS was excellent in 16 (39%), good in 22 (53%), fair in 2 (4.8%), and poor in 1 (2.4%) [Table 5].
|Table 5: Association for the Study and Application of Methods of Ilizarov bone score|
Click here to view
The mean NULODS in aseptic patients was 45.6 ± 15 and septic patients was 56.7 ± 14.1. The difference was statistically significant (P = 0.03).
Difference of means of NUSS in aseptic group (48.7 ± 16.1) and in septic group (71.2 ± 7.2) was statistically significant (P = 0.00008). There was a significant correlation between NUSS and NULODS score (r = 0.765), P = 0.0009.
NUSS and NULODS were compared with ASAMI-FS and ASAMI-BS. Both NUSS (P = 0.009) and NULODS (P = 0.07) correlated with ASAMI-FS, whereas neither correlated with ASAMI-BS (NUSS, P = 0.59; NULODS, P = 0.27).
ISS was calculated for septic NUs (n = 27). The mean value of ISS was 47.03 (20–92). NUSS and NULODS were compared to ISS. Positive correlation was found between them and ISS. The NULODS score correlated better at r = 0.723 (P = 0.00003) with NUSS being r = 0.587 (P = 0.0012).
We performed independent t-test to determine whether NUSS or NULODS influenced the duration of fixator wear (EFD). NULODS had a better correlation (r = 0.495) compared to NUSS (r = 0.303) with EFD [Table 6].
|Table 6: Comparison of external fixation duration with nonunion severity score and nonunion level of difficulty score|
Click here to view
We also compared a correlation of gap and shortening section scores of NUSS and NULODS with EFD. NULODS gap score (r = 0.604, P = 0.0001) trumped NUSS (r = 0.373, P = 0.016) in correlation with EFD [Table 7].
| Discussion|| |
UTNUs have not been commonly reported in literature. High-energy trauma and bone and soft-tissue loss that accompany these injuries make achieving excellent FSs difficult. Ahearn et al. reported a WOMAC score of 70 using either plates or fine wire fixators. When these fail, the NU is usually accompanied by infection, deformity, gap, and shortening and osteoporosis of the small fragments. Internal fixation can be used in aseptic NU with additional bone grafting, demineralized bone matrix, bone marrow aspiration concentrate (BMAC), and procedures for soft-tissue coverage like flaps. Giannoudis et al. reported only two UTNUs in their series of 64 long bone NU using bone morphogenetic protein (BMP)-7, BMAC, and reaming irrigation aspirator graft as part of the diamond concept. Each of these three is very expensive and out of the reach of most of our patients. Most of our self-paying patients bore a treatment cost much lesser than the cost of BMP-7 alone. When possible, this newer modality helps achieve early union using the convenience of internal fixation. Most of the infected patients treated in this manner have first undergone a Masquelet procedure with creation of a vascularized membrane.
We have reservations about the Masquelet techniques as have Morelli et al. and Morris et al. There can be high rates of failure with recrudescence of infection and loss of the extensive and expensive bone grafts, which Ilizarov  external fixation is ideal when risk of recrudescence of infection following internal fixation is high. It also permits correction of deformity, filling the bone gaps with corticotomy lengthening, and correction of overall shortening and helps achieve reliable union due to sustained and gradual compression at the NU [Figure 2]. We could not get excellent results in many of our cases because the postoperative leg length discrepancy was either more than 2.5 cm or the deformity was more than 7°. Many of our cases had anterior bone gaps after debridement of necrotic bone and ended up with a PPTA of mean 87° which is a recurvatum deformity more than 7° compared to a normal of 78°. We have compared our results with the few articles.,,,, We found in English literature which have focused on or included some UTNUs as part of their series and tabulated them [Table 8].
|Figure 2: (a) Comminuted and segmental nonunion of with angulation. The patient had nonunion level of difficulty score and nonunion severity score of 50 each preoperatively. (b) Ilizarov fixator with rings applied perpendicular to the bony fragments. Hinges and motors allow correction to make rings parallel. (c) Postoperative X-rays showing sound union. Anteroposterior view shows medial proximal tibial angle = 93.4° and a recurvatum in lateral view with posterior proximal tibial angle = 101.7°. This patient had good Association for the Study and Application of Methods of Ilizarov bone and functional score|
Click here to view
The NUSS formulated and validated  by Calori helps to grade the severity of presentation of the NU using many parameters. It is indeed a valid and valuable tool to help the surgeon choose treatment modalities. In our experience of treating NUs over 30 years, we feel it does not do justice to at least two facets of the problem.
First, the clinical infection status in NUSS has only three categories (clean, previously infected, and septic). It is almost a binary grading, giving 0, 1, and 4 points, and does not give incremental points though most clinicians see varying grades of severity of infection. Some cases require extensive debridement of both intra- and extramedullary spaces resulting in significant bone and soft-tissue excision as well as significant blood loss. Some require a second surgery for eradication of infection  despite thorough debridement. These could be graded higher with more points. A few need only minimum debridement and are also conducive to immediate external fixation or even internal fixation in the same stage. These could possibly be classified as mild. We feel there is a need to quantify the severity of infection and thus created the ISS.,
Second, NUSS does not thoroughly evaluate the difficulty of achieving an excellent bone result according to ASAMI-BS. In NUSS, bone quality is graded as good, moderate (mildly osteoporotic), poor (severe osteoporosis or bone loss), and very poor (necrotic, appears avascular or septic). Bone gap is classified as 0.5–1 cm, 1–3 cm, and >3 cm. Presumably, this considers overall limb shortening. The ASAMI-BS is excellent if residual shortening is <2.5 cm. If we consider two cases, one having a NU with 4 cm gap and another with 10 cm, the latter is very likely to have more residual shortening and get a “good” score rather than “excellent.” Both these scenarios get 5 points in NUSS but get 2 and 5 points, respectively, in NULODS.
We also found that external fixation time correlated better with NULODS than with NUSS (P = 0.001 vs. 0.054). Calori recommends amputation for scores >75. In fact, 10/41 of our patients had NUSS score >75 and yet none were amputated. ASAMI-FS was excellent in 1, good in 4, fair in 4, and poor in 1 patient among those who had scores >75. Whereas using the NULODS, we had only 4 patients with a score >75. Their results were good in 1, fair in 2, and poor in 1 with none being excellent.
The difficulty of any limb reconstruction treatment using external fixation is always measured explicitly by the EFD. EFD correlates with restriction of activities of daily living, employment, financial hardships, pin-tract infections, and joint stiffness.
More the gap and shortening, more should be the EFD. The mean gap score in NUSS was 3.24 and in NULODS was 1.93 for the same patients. Using paired samples t-test, the correlation between EFD and gap score measured using NULODS was much better at r = 0.604 (P = 0.000029) as compared to EFD and gap score measured using NUSS; r = 0.373 (P = 0.016).
Although NUSS had a slightly better correlation with ASAMI-FS compared to NULODS (P = 0.009 vs. 0.07), we feel the NULODS perhaps better anticipates difficulties in treating these difficult juxta-articular and infected NUs. The NULODS considers small size of bony fragments in the upper tibia, hence necessitating extension of fixation into the femur (18/41 in our series).
Knee joint stiffness following UTNU treatment is usually caused by the intra-articular extension of original or surgical trauma. Spanning external fixation used to augment fixation of small fragments can contribute to stiffness if care is not taken to mobilize the joint with hinges.
Complete eradication of infection in septic UTNU was possible in our series. Staged treatment permits focus on thorough debridement in the first stage. A 6-week gap with specific antibiotics and assisted partial weight-bearing walking improves bone quality and soft-tissue trophicity [Figure 3]. Tolerance to the Ilizarov fixator applied in the second stage is higher and its fixation strength better after the patient has walked. Any residual infection at end of treatment can be graded using the ISS and not labeled binarily as present or absent.
|Figure 3: (a) Flexion contracture of knee with equinus in ankle with multiple discharging sinus, recurvatum at the fracture site, and puckered scars all along the length of tibia. (b) Plain radiograph showing upper tibial nonunion treated initially with antibiotic cement rod. This patient had nonunion level of difficulty score 86 and nonunion severity score 94. (c and d) Ilizarov fixator applied with femur frame for better stability. Fixed flexion deformity corrected at knee and recurvatum corrected at upper end tibia. (e) Compression at fracture site resulted in sound union. The patient had no coronal deformity, though recurvatum was present at the fracture site due to anterior bone loss at the time of injury. (f) Clinical photograph at 1-year 3-month follow-up. He had excellent Association for the Study and Application of Methods of Ilizarov bone score and fair Association for the Study and Application of Methods of Ilizarov functional score|
Click here to view
The strength of our study is that it is possibly the largest study dealing exclusively with UTNU. Most challenges presented by UTNU including infection can be addressed using the Ilizarov techniques without resorting to using expensive ingredients of the diamond concept.
Our study surely has limitations: There is no comparison with other treatment modalities. Our use of newer scores such as NULODS and ISS has yet to be validated.
However, considering the difficulties of the presenting UTNU, we feel the results of using the Ilizarov techniques are quite good. Our rather long external fixator duration could be reduced by adding biological stimulus of grafting or BMAC or even BMP-7 when affordable.
| Conclusions|| |
UTNUs have not been frequently reported in literature. Apart from union, other challenges that need addressing are infection, bone gaps and shortening, poor bone quality, and soft tissue cover. The Ilizarov techniques are reliable and can help achieve good to excellent results in most patients. Preoperative evaluation using NUSS and newer scores such as ISS and NULODS can help the surgeon to better grade severity of the clinical spectrum of UTNU, counsel the patient, and choose modality of treatment.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Bhandari M, Audige L, Ellis T. Operative treatment of extra-articular proximal tibial fractures. J Orthop Trauma 2003;17:591-5.
Ilizarov GA. Transosseous Osteosynthesis. Berlin: Springer Verlag; 1991.
Bhowmick K, Boopalan PRJVC, Gunasekeran C, Livingston A, Jepegnanam TS. Management of Chronic Infected Intra-Articular Fractures of the Proximal Tibia with Ilizarov Ring Fixation. J Knee Surg 2019;16:10.1055/s-0038-1677512.
Paley D, Catagni MA, Argnani F, Villa A, Benedetti GB, Cattaneo R. Ilizarov treatment of tibial nonunions with bone loss. Clin Orthop Relat Res 1989;241:146-65.
Paley D. Problems, obstacles, and complications of limb lengthening by the Ilizarov technique. Clin Orthop Relat Res 1990;250:81-104.
Calori GM, Phillips M, Jeetle S, Tagliabue L, Giannoudis PV. Classification of non-union: Need for a new scoring system? Injury 2008;39 Suppl 2:S59-63.
ASAMI Group. Operative Principles of Ilizarov. Baltimore: Williams and Wilkins; 1991.
Chaudhary Milind M. Infected gap nonunions of femur. In: Kulkarni GS, Babhulkar S, editors. In Guidelines in Fracture Management, Nonunion in Long Bone II. Ch. 25. Noida: Thieme Publisher; 2016. p. 309-24.
Chaudhary MM. Infected nonunion of tibia. Indian J Orthop 2017;51:256-68.
] [Full text]
Paley D, Herzenberg JE. Intramedullary infections treated with antibiotic cement rods: Preliminary results in nine cases. J Orthop Trauma 2002;16:723-9.
Ali F, Saleh M. Treatment of isolated complex distal femoral fractures by external fixation. Injury 2000;31:139-46.
Braly HL, O'Connor DP, Brinker MR. Percutaneous autologous bone marrow injection in the treatment of distal meta-diaphyseal tibial nonunions and delayed unions. J Orthop Trauma 2013;27:527-33.
Ahearn N, Oppy A, Halliday R, Rowett-Harris J, Morris SA, Chesser TJ, et al
. The outcome following fixation of bicondylar tibial plateau fractures. Bone Joint J 2014;96-B: 956-62.
Giannoudis PV, Gudipati S, Harwood P, Kanakaris NK. Long bone non-unions treated with the diamond concept: A case series of 64 patients. Injury 2015;46 Suppl 8:S48-54.
Morelli I, Drago L, George DA, Gallazzi E, Scarponi S, Romanò CL. Masquelet technique: Myth or reality? A systematic review and meta-analysis. Injury 2016;47 Suppl 6:S68-S76.
Morris R, Hossain M, Evans A, Pallister I. Induced membrane technique for treating tibial defects gives mixed results. Bone Joint J 2017;99-B:680-5.
Cattaneo R, Catagni M, Johnson EE. The treatment of infected nonunions and segmental defects of the tibia by the methods of Ilizarov. Clin Orthop Relat Res 1992;280:143-52.
Mahmoud S. Management of post-traumatic extra-articular proximal tibial nonunion by Ilizarov external fixation. Egypt Orthop J 2018;53:119-24. [Full text]
Wani NB, Syed B. Ilizarov ring fixator in the management of infected non-unions of tibia. SICOT J 2015;1:22.
Gardner MJ, Toro-Arbelaez JB, Hansen M, Boraiah S, Lorich DG, Helfet DL. Surgical treatment and outcomes of extra-articular proximal tibial nonunions. Arch Orthop Trauma Surg 2008;128:833-9.
Chan DB, Jeffcoat DM, Lorich DG, Helfet DL. Nonunions around the knee joint. Int Orthop 2010;34:271-81.
Sen RK, Soni A, Saini UC, Singh D. Internal fixation and bone grafting for intra-articular nonunion of tibial plateau: A report of four cases. Chin J Traumatol 2011;14:371-5.
Calori GM, Colombo M, Mazza EL, Mazzola S, Malagoli E, Marelli N, et al
. Validation of the non-union scoring system in 300 long bone non-unions. Injury 2014;45 Suppl 6:S93-7.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8]