ORIGINAL ARTICLE Three-Column Fixation for Complex Tibial Plateau Fractures Cong-Feng Luo, MD, PhD, Hui Sun, MD, Bo Zhang, MD, and Bing-Fang Zeng, MD Objectives: 1) To introduce a computed tomography-based ‘‘three- column fixation’’ concept; and 2) to evaluate clinical outcomes (by using a column-specific fixation technique) for complex tibial plateau fractures (Schatzker classification Types V and VI). Design: Prospective cohort study. Setting: Level 1 trauma center. Patients: Twenty-nine cases of complex tibial plateau fractures were included. Based on routine x-ray and computed tomography images, all the fractures were classified as a ‘‘three-column fracture,’’ which means at least one separate fragment was found in lateral, medial, and posterior columns in the proximal tibia (Schatzker classification Types V and VI). Intervention: The patients were operated on in a ‘‘floating position’’ with a combined approach, an inverted L-shaped posterior approach combined with an anterior–lateral approach. All three columns of fractures were fixed. Outcome Measures: Operative time, blood loss, quality of reduction and alignment, fracture healing, complications, and functional outcomes based on Hospital for Special Surgery score and lower-extremity measure were recorded. Results: All the cases were followed for average 27.3 months (range, 24–36 months). All the cases had satisfactory reduction except one case, which had a 4-mm stepoff at the anterior ridge of the tibial plateau postoperatively. No case of secondary articular depres- sion was found. One case had secondary varus deformity, one case had secondary valgus deformity, and two cases of screw loosening occurred postoperatively. No revision surgery was performed. Two cases had culture-negative wound drainage. No infection was noted. The average radiographic bony union time and full weightbearing time were 13.1 weeks (range, 11–16 weeks) and 16.7 weeks (range, 12–24 weeks), respectively. The mean Short Form 36, Hospital for Special Surgery score, and lower-extremity measure at 24 months postoperatively were 89 (range, 80–98), 90 (range, 84–98), and 87 (range, 80–95), respectively. The average range of motion of the affected knee was 2.7° to 123.4° at 2 years after the operation. Conclusion: Three-column fixation is a new fixation concept in treating complex tibial plateau fractures, which is especially useful for multiplanar fractures involving the posterior column. The combination of posterior and anterior–lateral approaches is a safe and effective way to have direct reduction and satisfactory fixation for such difficult tibial plateau fractures. Key Words: tibial plateau fracture, three-column fixation, combined approach, floating position (J Orthop Trauma 2010;24:683–692) INTRODUCTION Complex tibial plateau fracture management remains clinically challenging. These fractures are usually described as Schatzker Type Vand VI or as a C type injury when using the AO/Orthopaedic Trauma Association classification. 1,2 Bilat- eral dual plating is usually recommended as the definite fixation for this kind of fracture. 3–6 However, this technique sometimes is not applicable to work in fractures with multi- planar articular comminution. This is especially true when there is posterior shearing or a coronal fracture. 7,8 Tradition- ally, the treatment for tibial plateau fractures is based on two-dimensional classification systems. Several authors have noted computed tomography (CT)-based three-dimensional consideration of the fracture pattern was important in the treatment of tibial plateau fractures. 9–11 In recent years, we developed a ‘‘three-column fixation’’ technique to treat the multiplanar complex tibial plateau fractures, which is based on three-dimensional understanding of the fractures. In this article, we report on the clinical results of using a ‘‘three-column fixation’’ technique through combined approaches: the anterolateral and the posterior approaches. A special ‘‘floating position’’ was designed to perform the surgery, which was based on a lateral decubitus, and the lower leg was rotated to a prone position when the posterior approach to the tibial plateau was performed. PATIENTS AND METHODS The patients’ data were collected prospectively. Patient demographics and the preinjury status were recorded at Accepted for publication January 14, 2010. From the Department of Orthopaedic Surgery, Shanghai Sixth People’s Hospital, Shanghai Jiaotong University, Shanghai, China. The authors did not receive grants or outside funding in support of their research or preparation of this manuscript. This study was presented in part as a poster presentation at the Annual Meeting of the Orthopaedic Trauma Association, San Diego, CA, 2009. Reprints: Cong-Feng Luo, MD, PhD, Department of Orthopaedic Surgery, Shanghai Sixth People’s Hospital, Shanghai Jiaotong University, 600 YiShan Road, Shanghai 200233, China (e-mail: cong_fengl@yahoo. com.cn). Copyright Ó 2010 by Lippincott Williams & Wilkins J Orthop Trauma Volume 24, Number 11, November 2010 www.jorthotrauma.com | 683
Transcript
ORIGINAL ARTICLE
Three-Column Fixation for ComplexTibial Plateau Fractures
Cong-Feng Luo, MD, PhD, Hui Sun, MD, Bo Zhang, MD, and Bing-Fang Zeng, MD
Objectives: 1) To introduce a computed tomography-based ‘‘three-
column fixation’’ concept; and 2) to evaluate clinical outcomes (by
using a column-specific fixation technique) for complex tibial plateau
fractures (Schatzker classification Types V and VI).
Design: Prospective cohort study.
Setting: Level 1 trauma center.
Patients: Twenty-nine cases of complex tibial plateau fractures were
included. Based on routine x-ray and computed tomography images,
all the fractures were classified as a ‘‘three-column fracture,’’ which
means at least one separate fragment was found in lateral, medial, and
posterior columns in the proximal tibia (Schatzker classification
Types V and VI).
Intervention: The patients were operated on in a ‘‘floating
position’’ with a combined approach, an inverted L-shaped posterior
approach combined with an anterior–lateral approach. All three
columns of fractures were fixed.
Outcome Measures: Operative time, blood loss, quality of
reduction and alignment, fracture healing, complications, and
functional outcomes based on Hospital for Special Surgery score
and lower-extremity measure were recorded.
Results: All the cases were followed for average 27.3 months
(range, 24–36 months). All the cases had satisfactory reduction
except one case, which had a 4-mm stepoff at the anterior ridge of the
tibial plateau postoperatively. No case of secondary articular depres-
sion was found. One case had secondary varus deformity, one case
had secondary valgus deformity, and two cases of screw loosening
occurred postoperatively. No revision surgery was performed. Two
cases had culture-negative wound drainage. No infection was noted.
The average radiographic bony union time and full weightbearing
time were 13.1 weeks (range, 11–16 weeks) and 16.7 weeks (range,
12–24 weeks), respectively. The mean Short Form 36, Hospital for
Special Surgery score, and lower-extremity measure at 24 months
postoperatively were 89 (range, 80–98), 90 (range, 84–98), and 87
(range, 80–95), respectively. The average range of motion of the
affected knee was 2.7� to 123.4� at 2 years after the operation.
Conclusion: Three-column fixation is a new fixation concept in
treating complex tibial plateau fractures, which is especially useful
for multiplanar fractures involving the posterior column. The
combination of posterior and anterior–lateral approaches is a safe
and effective way to have direct reduction and satisfactory fixation for
clinically challenging. These fractures are usually described asSchatzker Type V and VI or as a C type injury when using theAO/Orthopaedic Trauma Association classification.1,2 Bilat-eral dual plating is usually recommended as the definitefixation for this kind of fracture.3–6 However, this techniquesometimes is not applicable to work in fractures with multi-planar articular comminution. This is especially true whenthere is posterior shearing or a coronal fracture.7,8 Tradition-ally, the treatment for tibial plateau fractures is based ontwo-dimensional classification systems. Several authors havenoted computed tomography (CT)-based three-dimensionalconsideration of the fracture pattern was important in thetreatment of tibial plateau fractures.9–11 In recent years, wedeveloped a ‘‘three-column fixation’’ technique to treat themultiplanar complex tibial plateau fractures, which is based onthree-dimensional understanding of the fractures.
In this article, we report on the clinical results of usinga ‘‘three-column fixation’’ technique through combinedapproaches: the anterolateral and the posterior approaches.A special ‘‘floating position’’ was designed to perform thesurgery, which was based on a lateral decubitus, and the lowerleg was rotated to a prone position when the posteriorapproach to the tibial plateau was performed.
PATIENTS AND METHODSThe patients’ data were collected prospectively. Patient
demographics and the preinjury status were recorded at
Accepted for publication January 14, 2010.From the Department of Orthopaedic Surgery, Shanghai Sixth People’s
Hospital, Shanghai Jiaotong University, Shanghai, China.The authors did not receive grants or outside funding in support of their
research or preparation of this manuscript.This study was presented in part as a poster presentation at the Annual
Meeting of the Orthopaedic Trauma Association, San Diego, CA, 2009.Reprints: Cong-Feng Luo, MD, PhD, Department of Orthopaedic Surgery,
J Orthop Trauma � Volume 24, Number 11, November 2010 www.jorthotrauma.com | 683
admission. Preliminary management included distal bonyskeletal traction or bridging external fixation where reductionneeded to be maintained preoperatively. The external fixatorwas used in emergency situations in which there was high-energy injury to the soft tissues. The fixation bridged acrossthe knee and the pin(s) in the tibial shaft were placed to avoidthe site of future operative incisions. All the patients haddefinitive operative procedures after the soft tissue conditionwas stable. The mean time from presentation to definitivefixation was 8.5 days (range, 1–18 days).
On admission, all the patients underwent a standardradiologic protocol of x-rays and CT scans. The CT scans wereperformed after bony traction or bridging external fixation hadbeen applied; this was much more informative for decision-making. Besides Schatzker classification, all the fractures werealso classified with a ‘‘three-column’’ concept (Fig. 1); on thetransverse view, the tibial plateau is divided into three areas,which are defined as the lateral column, the medial column,and the posterior column. These three columns are separatedby three connecting lines, namely OA, OC, and OD. Point O isthe center of the knee (midpoint of two tibial spines); Point Arepresents the anterior tibial tuberosity; Point D is theposteromedial ridge of proximal tibia; and Point C is themost anterior point of the fibular head. Point B is the posteriorsulcus of the tibial plateau, which intersects the posteriorcolumn into the medial and lateral parts. Besides the transverseview, the accurate classification usually was done with the helpof frontal view and three-dimensional reconstruction.
A ‘‘three-column classification’’ was used for decision-making. According to this classification, one independent
articular depression with a break of the column wall is definedas a fracture of the relevant column. Pure articular depression(Schatzker Type III) was defined as a ‘‘zero-column fracture.’’Most of the simple lateral split and split depression fractures(Schatzker Types I and II) belong to a ‘‘one-column (lateralcolumn) fracture.’’ However, when there is an anterolateralfracture and a separate posterior–lateral articular depressionwith a break of the posterior wall, the fracture is defined asa ‘‘two-column (lateral and posterior column) fracture.’’Articular depression in the posterior column with a break ofthe posterior wall is also defined as a ‘‘one-column (posteriorcolumn) fracture’’ (not included in any type of the Schatzkerclassification). The other typical ‘‘two-column fracture’’ is theanteromedial fracture with a separate posteromedial fragment(medial and posterior column fracture), which traditionallybelongs to Schatzker Type IV (medial condylar fracture). The‘‘three-column fracture’’ is defined as at least one independentarticular fragment in each column. The most common three-column fracture is a traditional ‘‘bicondylar fracture’’(Schatzker Type V or Type IV) combined with a separateposterolateral articular fragment.
All the cases were assessed by different team leaders (wehave seven trauma teams). If the case was considered to needthe ‘‘three-column fixation’’ technique, he or she wastransferred to the authors’ team. All of these 29 cases wereoperated on by the authors (C.F.L. and B.F.Z.).
Postoperatively, anteroposterior x-rays of the knee weretaken in the immediate postoperative period, 6 weeks, 12weeks, and every 6 to 8 weeks until bony union occurred andthen 2 years after the index operation. Tibial plateau angle(TPA), the femorotibial angle, and the medial and lateralposterior slope angle (PA) were measured by one surgeon(B.Z.). Malreduction was defined as intra-articular stepoff of2 mm or more, a TPA $95�/TPA #80�, or PA $15�/PA #–5�.11
Secondary loss of reduction was defined as an increase of 5�malalignment or an articular depression of 2 mm whencompared with the first postoperative radiograph at final followup. Bony union was defined as radiologically finding at leastthree healed cortices. Full weightbearing was defined as the timethat patients could have painless walking without any aids.
Operative TechniqueAll patients were treated by open reduction and internal
fixation with the same surgical team. After induction ofgeneral anesthesia and antibiotic prophylaxis, the procedurewas performed in the ‘‘floating position,’’ which was based ona lateral decubitus, and the lower leg was rotated to a proneposition when the posterior approach to the tibial plateau wasperformed (Fig. 2). A combined approach was used for all thecases. The bridging fixator was removed before surgerystarted.
A posterior inverted L-shaped approach was indicated todeal with medial column and posterior column fractures(Fig. 2). With the patient prone on a radiolucent table, the kneewas slightly flexed by a bump under the ankle. An invertedL-shaped incision begins at the center of popliteus parallel toLangers line superiorly and medial. Distally it turns at themedial corner of the popliteal fossa and is carried down toFIGURE 1. Three-column classification.
684 | www.jorthotrauma.com q 2010 Lippincott Williams & Wilkins
Luo et al J Orthop Trauma � Volume 24, Number 11, November 2010
deep fascia. Full-thickness fasciocutaneous flaps were raisedpaying attention to protecting the sural nerve and shortsaphenous vein. The tendon of the medial head of thegastrocnemius was then visualized with blunt dissection andthen retracted laterally, protecting the neurovascular bundleand displaying the back of the knee capsule (Fig. 3). To avoidinjury to the neurovascular bundle in popliteal space, all thedissection from medial to lateral should be done beneathpopliteus muscle in the proximal part. Overdissection laterallytoward the tibial shaft should be avoided, because it is easy toinjure the posterior tibial recurrent artery (a branch from theproximal part of the anterior tibial artery and bifurcation of thetibial arteries).
The popliteus and soleus origin are then elevated off theposteromedial aspect of the proximal tibia from medial tolateral as needed to gain exposure of the fracture of posteriorcolumn. In most situations, under general anesthesia, the entireposterior aspect of the tibia can be exposed without cutting themedial head of the gastrocnemius. The articular surface waselevated by working through the ‘‘fracture window’’ at thefracture site by using a periosteum elevator (Fig. 4). Thereduced articular surface was temporarily fixed with severalsubchondral Kirschner wires. From the posterior approach, itis not easy to manipulate the anterolateral part of the articularsurface, which can only be reduced and fixed through theanterolateral approach in the later stage of the operation.Flexing the knee to relax the posterior soft tissue can helpexposure; however, full reduction and buttress plate fixation ofthe articular surface in the posterior column can only be donewith the knee in extension. Because there is no standard
implant for posterior column fractures, an undercontoured3.5-mm LC-DCP, 3.5-mm T-plate, or a 3.5-mm cloverleaf plate(Synthes, Oberdorf, Switzerland) with the central tip cut offwas used for posterior column fixation. For the posteromedialfragment (between Points D and B in Fig. 1), the buttress platewas usually put in longitudinally (parallel to the medial ridgeof the tibia). An oblique posterior plate (from proximal lateralto the distal medial) was usually used to buttress the postero-lateral fragment (between Points B and C in Fig. 1) (Fig. 5).
To expose the anteromedial (medial column) fracture,anterior dissection can be done along the medial edge of thisincision. The fascia was incised between the medialgastrocnemius and the pes anserinus anteriorly. The medialcollateral ligament remains intact anteriorly and deep to thepes anserinus. The semimembranosus insertion was releasedoff the bone. Both pes anserinus and semimembranosus can bereattached with nonabsorbable sutures after fracture fixation.A buttress plate (usually 3.5-mm LC-DCP) was put on themedial ridge of the proximal tibia to support the medialcolumn. It is important to not put this plate too posteriorly orthe buttress effect will decrease.
A conventional anterior approach was used to reduceand fixate the fracture in the lateral column. The arthrotomywas performed through a submeniscal approach. The articularsurface was elevated through the ‘‘fracture window’’ and fixedwith a lateral plate (L-plate or LISS-PT; Synthes).
The quality of reduction, the location of the plates, andthe length of the screws were confirmed under fluoroscopicguidance. The deep fascia was left open. Subcutaneous tissueand skin were closed over suction drainages.
q 2010 Lippincott Williams & Wilkins www.jorthotrauma.com | 685
J Orthop Trauma � Volume 24, Number 11, November 2010 Three-Column Fixation for Complex Tibial Plateau Fractures
Postoperative Management and Follow UpA continuous passive motion machine was used in the
hospital for 3 days after the surgery. Partial weightbearingbegan at the fourth to sixth postoperative week. Full
weightbearing was delayed until the fracture was healed andcallus appeared on radiographs.
Standard anteroposterior and lateral radiographs weretaken at follow up and were evaluated for fracture healing andjoint congruity. Bony union time and full weightbearing timewere recorded. Both TPA and PA on the radiographsimmediately postoperatively and 24 months postoperativelywere measured and recorded. At 24-month follow up, patientswere administered the Short Form 36 general health survey,Hospital for Special Surgery, score and lower-extremitymeasure.
Statistical MethodsAll data analysis was done using SPSS 11.0 (SPSS Inc.,
Chicago IL). Descriptive statistics were used to determineranges, means, and standard deviations. One-way analysis ofvariance and Student t tests were used to determine thedifference between two means. Correlations were analyzed byusing the Pearson correlation coefficient. P , 0.05 wasconsidered statistically significant.
RESULTSFrom December 2004 to July 2006, 266 cases of tibial
plateau fractures were operated on in our center. Among those,there were 32 cases diagnosed as ‘‘three-column fractures,’’which needed ‘‘three-column fixation.’’ Three patients wereexcluded because they could not be contacted during followup, leaving 29 cases for the study. The patient demographicsand fracture types are shown in Table 1. There were six womenand 23 men with an average age of 46.8 years (range, 22–62years). Thirteen fractures were on the left side and 16 on theright. All fractures in this series were closed fractures withoutany distal neurovascular injury or compartment syndromes.
The total mean operation time was 140 minutes (range,110–180 minutes). The mean blood loss was 327 mL (range,200–800 mL). Two cases had blood transfusion (No. 4 and
FIGURE 4. Intraoperative photograph. Depressed posterolat-eral articular surface (arrow) can be seen from the ‘‘fracturewindow.’’ Direct reduction and fixation was performedthrough this window.
FIGURE 5. Intraoperative photograph. The posterior columnwas reconstructed and buttressed with two separate plates.
FIGURE 3. Schematic diagram of the operative approach to theposterior aspect of the tibial plateau.
686 | www.jorthotrauma.com q 2010 Lippincott Williams & Wilkins
Luo et al J Orthop Trauma � Volume 24, Number 11, November 2010
No. 5). All the cases were followed up for at least 24 months;the mean follow-up time was months 27.3 (range, 24–36months). The average radiographic bony union time was 13.1weeks (range, 11–16 weeks) and the average full weightbear-ing time was 16.7 weeks (range, 12–24 weeks). One case (No.21) had secondary varus deformity, one case (No. 26) hadsecondary valgus deformity, and two cases (No. 15 and No.17) had screw loosening postoperatively. No revision surgerywas performed for these complications. There were two cases(No. 10 and No. 12) of wound drainage with negative bacterialculture, and these healed with nonoperative wound manage-ment. No infection was noted.
Patient scores for the Short Form 36, Hospital for Spe-cial Surgery score, and lower-extremity measure at 24 monthspostoperatively were 89 (range, 80–98), 90 (range, 84–98),and 87 (range, 80–95), respectively. The average range ofmotion of the affected knees was 2.7� to 123.4�. There were nosignificant differences in either TPA or PA on the radiographsimmediately postoperatively and 24 months postoperatively
(P = 0.840 for TPA, 0.060 for medial posterior-slope-angle,and 0.061 for lateral posterior-slope-angle) (Table 2).
DISCUSSIONMost of the current classification systems for tibial
plateau fractures use two-dimensional images, which usuallydirect surgeons to pay attention to medial and lateral fixationwithout thinking of posterior fixation. With careful reviewand application of the CT scan for the evaluation for thesefractures,7,8,10–12 some surgeons have realized the importanceof considering posterior fixation in tibial plateau fractures,especially for the posteromedial fragment.8,13 In this article,we reported on a column specific fixation concept: three-column fixation, which is dependent on the understanding ofthe fractures using CT scans. The authors believe this fixationconcept for tibial plateau fractures has been poorly reportedon in the English literature. Multiplanar complex tibialplateau fractures, especially those involving the posterior
q 2010 Lippincott Williams & Wilkins www.jorthotrauma.com | 687
J Orthop Trauma � Volume 24, Number 11, November 2010 Three-Column Fixation for Complex Tibial Plateau Fractures
column, are quite difficult to manage clinically. With ourtechnique, posterior column fixation is stressed when thefractures involve the posterior aspect of the plateau. Insteadof classic bilateral (medial and lateral) approaches, a newcombination of posterior and anterolateral approaches usingcareful patient positioning (the ‘‘floating position’’) isintroduced to treat such a fracture. This new approach issafe and effective in managing complex Schatzker V and VItibial plateau fractures.
Most complex tibial plateau fractures are a result ofhigh-energy injury. Resulting comminution makes interpretingof fracture patterns difficult. Fully understanding thesefractures is the basis for successful treatment. Both theSchatzker and AO/Orthopaedic Trauma Association systemsclassify these fractures according to the appearance onanteroposterior radiographs.14 Some of these fractures areeasy to misunderstand, especially fractures involving theposterior aspect of the tibial plateau. Wicky et al11 reporteda cohort of 42 cases with tibial plateau fractures, which wereassessed by plain radiographs and three-dimensional CT
separately. As a result, 43% (18 of 42) of the fractures wereunderevaluated by plain radiographs. On the other hand, suchfractures can be difficult to fit into the classification systemscurrently used, which makes diagnosis and preoperativeplanning difficult. Macarini et al15 studied 25 cases of tibialplateau fractures. After CT scan, only 48% of the cases had thesame classification as before the CT scan and 60% of the caseshad changes in the operative plan. Most authors agree that CTscanning adds invaluable information to the treatment of tibialplateau fractures.7,8,16 We think the CT-based ‘‘three-columnconcept’’ can help surgeons analyze these fractures threedimensionally providing a better approach and fixationmethods.
Although Khan et al12 had listed coronal splits at theposterior tibial plateau as a separate group in their classifica-tion system, this group of the fracture has been underappre-ciated in other commonly used classification systems.7,16 Thisis partly because this type of fracture usually appears con-fusing on initial radiograph and only can be clearly identifiedon a CT scan. For example, when the fracture involves the
TABLE 2. 24-Month Postoperative Data
PatientNo.*
X-rayUnion
(weeks)FWB
(weeks)2-Year FTA
(degree)
2-YearTPA
(degree)
2-YearPAM
(degree)
2-YearPAL
(degree)
2-YearStepoff(mm)
2-YearExtension
2-YearFlexion
2-YearHSS
2-YearLEM
2-YearSF-36
1 13 17 175.2 87.2 7.6 5.2 0 1 132 93 90 92
2 12 16 173.6 84.7 13.6 10.1 1 6 121 90 85 88
3 15 17 172.6 86.7 6.0 3.8 0 5 120 88 85 86
4 11 12 171.8 89.6 7.8 4.3 0 2 129 94 90 93
5 14 16 170.8 86.3 13.4 1.2 1 3 119 87 85 93
6 12 12 170.0 91.0 5.0 3.0 0 2 122 93 90 96
7 15 24 170.6 90.6 10.4 8.4 0 4 120 86 85 89
8 12 20 172.2 86.6 7.0 3.3 1 3 120 86 90 84
9 11 12 176.4 85.8 2.3 1.3 0 0 128 98 85 98
10 12 17 174.8 88.5 3.0 2.5 4 5 120 84 80 80
11 16 24 173.6 86.8 6.9 5.0 2 4 122 85 85 88
12 14 17 176.2 85.3 11.3 7.6 0 2 126 90 85 80
13 13 15 178.3 88.3 3.2 2.0 1 0 120 90 90 84
14 13 16 175.6 86.5 7.6 6.4 1 3 122 84 80 88
15 14 17 172.0 87.8 3.5 2.7 0 4 110 84 85 92
16 11 13 174.9 88.2 13.2 9.6 0 6 121 86 85 88
17 15 18 173.7 87.7 10.2 8.1 0 3 124 88 85 88
18 13 16 177.2 89.1 11.2 7.8 0 2 120 90 85 88
19 13 17 172.5 87.4 11.4 8.1 0 2 120 93 90 90
20 13 14 178.3 84.6 12.1 7.9 1 0 126 92 95 96
21 14 22 182.3 87.3 9.4 8.6 0 4 121 85 90 88
22 13 16 174.2 86.5 12.1 12.9 0 2 128 90 85 90
23 11 16 176.7 86.7 8.2 6.5 0 1 125 93 90 90
24 15 17 173.9 90.3 11.6 7.7 0 2 127 90 85 88
25 11 13 173.2 88.4 12.6 7.8 1 2 130 93 90 86
26 14 16 167.8 88.6 7.2 6.8 2 6 122 88 80 88
27 13 15 175.6 86.9 6.2 4.6 1 2 128 96 90 90
28 12 16 173.9 87.9 10.6 8.4 1 3 123 94 90 88
29 14 24 173.5 85.0 13.5 11.1 0 0 134 98 95 96
*Patients 30–32 were lost to followup.FWB, full weightbearing; FTA, femorotibial angle; TPA, tibial plateau angle; PAM, medial posterior-slope-angle; PAL, lateral posterior-slope-angle; HSS, Hospital for Special
Surgery, LEM, lower-extremity measure; SF-36, Short Form 36.
688 | www.jorthotrauma.com q 2010 Lippincott Williams & Wilkins
Luo et al J Orthop Trauma � Volume 24, Number 11, November 2010
posterior–lateral aspect of the plateau, the fracture might bevisible through an anterior approach, but the reduction andfixation are quite difficult, especially for those without anintact posterior cortex (disruption of the posterior column).Direct reduction through posterior approaches and posteriorbuttress plating have been recommended by several authors.16–19
Other authors have also used this theory to produce betterclinical results than older less safe approaches.18,19
A bilateral dual plating technique using a posterior–medial approach combined with an anterior–lateral approachhas been suggested by several authors.8,20 This posteromedialapproach, in the supine position, can deal with the poster-omedial fragment,21 but it is impossible to obtain a directreduction when there is an articular depression in the lateralpart of the posterior column (between Points B and C inFig. 1). Posterior–lateral depressed fracture fragments areimpossible to deal with in the supine position and can only bereduced and buttressed posteriorly in the prone position. Theamount of posterior dissection and the number of buttressplates can be determined from preoperative CT.
Unilateral locking plates have also been used to treatcomplex tibial plateau fractures. Some of the proximal tibiallocking plates have special design features with a ‘‘posterior–medial fragment screw’’ aiming from anterolateral to poster-omedial. Clinically, they are not strong enough to hold thesefragments and prevent secondary varus when compared witha direct posterior–medial buttress plate.7,9,22 Barei et al8
investigated 57 bicondylar fractures with CT scans and foundthe occurrence of the posteromedial fragment in approxi-mately one third of the cases. The different shapes andmorphologic features of these fragments implicated supple-mentary fixation when managing such a fracture. A com-bination of the ‘‘reversed L-shaped’’ posterior approach and
FIGURE 6. Female, 59 years of age, the victim of a trafficaccident, with a complex tibial plateau fracture of the right leg.
FIGURE 7. Computed tomographyscan after emergent bridging exter-nal fixator.
q 2010 Lippincott Williams & Wilkins www.jorthotrauma.com | 689
J Orthop Trauma � Volume 24, Number 11, November 2010 Three-Column Fixation for Complex Tibial Plateau Fractures
the anterior–lateral approach is recommended by the authorsfor those fractures that have a bicondylar fracture, but also forthose that have a separate fragment/articular depression in theposterior column. Such fractures will not reduce well with
conventional bilateral dual plating techniques. It is importantto note that only a small percentage of tibial plateau fracturesneed the ‘‘three-column fixation’’ technique in our series (12%of fractures, or 33 of 266 of the cases). Without carefulplanning, these ‘‘three-column fractures’’ usually proceed tofailure of reduction and fixation.
In our technique, through the ‘‘reversed L-shaped’’posterior approach, both posterior–lateral and posterior–medialfragments can be directly reduced and buttressed (reconstructionof posterior column) (Figs. 6–9). In the authors’ opinion, thisapproach also obviates the use of a second posterolateral
FIGURE 8. Coronal computed tomography scan indicatingboth medial and lateral column are involved.
FIGURE 10. Postoperative x-ray aftera combined approach with three-column fixation.
690 | www.jorthotrauma.com q 2010 Lippincott Williams & Wilkins
Luo et al J Orthop Trauma � Volume 24, Number 11, November 2010
incision as advocated by Carlson.23 The posterolateral approachcreates problems around the exposure of the common peronealnerve and management of posterior tibial recurrent artery (abranch from the proximal part of the anterior tibial artery).Through the inverted L-shaped posterior approach, the re-duction of the posterolateral articular surface can be achievedwith an elevator through a posterior ‘‘fracture window.’’ Werecommend application of a buttress plate to the posterolateral
plateau and it can be placed in an oblique fashion (fromproximal posterolateral to distal posteromedial) (Fig. 10).
As the last step, the anterior–lateral aspect of proximaltibia (the lateral column) is approached through an anterolateralincision, which can be performed on a patient in the ‘‘floatingposition’’ without a second draping (Fig. 2). The lateral columnfracture is usually manipulated with minimal invasive techni-ques; the small proximal incision is used to reduce the articularsurface and the metaphyseal area is plated percutaneously.Because ‘‘three-column fractures’’ are usually quite commi-nuted and the fragments are small, 3.5-/4.5-mm systems insteadof the conventional 4.5-/6.5-mm systems are recommended forfixation. To the authors’ understanding, this is the first time thatthis ‘‘floating position’’ for the combined approach has beenreported in the English literature.
The weaknesses of this article include the fact that this isa small series of patients and it represents a single center’s
q 2010 Lippincott Williams & Wilkins www.jorthotrauma.com | 691
J Orthop Trauma � Volume 24, Number 11, November 2010 Three-Column Fixation for Complex Tibial Plateau Fractures
experience. There is no postoperative CT scanning toaccurately quantitate articular reductions. We also feel thata full length standing x-ray is necessary for more precisealignment at long-term follow up (Figs. 11–14).
CONCLUSIONSThe ‘‘three-column concept’’ is a new and useful sup-
plement to the present classification systems for tibial plateaufractures. ‘‘Three-column fixation’’ seems to be an effectiveand a safe way for the treatment of multiplanar complex tibialplateau fractures.
ACKNOWLEDGMENTWe thank Dr. Richard Buckley from Calgary, Canada,
for his kind review of our paper.
REFERENCES1. Muller M, Allgower M, Schneider R, et al. Patella and tibia. In: Allgower
M, ed. Manual of Internal Fixation. New York: Springer-Verlag; 1979:553–94.
3. Krieg JC. Proximal tibial fracture: current treatment, results, andproblems. Injury. 2003;34:A2–A10.
4. Ricci WM, Rudzki JR, Borrelli J. Treatment of complex proximal tibiafractures with the less invasive skeletal stabilization system. J OrthopTrauma. 2004;18:521–527.
6. Barei DP, Nork SE, Mills WJ, et al. Complications associated with internalfixation of high-energy bicondylar tibial plateau fractures utilizing a two-incision technique. J Orthop Trauma. 2004;18:649–657.
7. Weil YA, Gardner MJ, Boraiah SB, et al. Posteromedial supine approachfor reduction and fixation of medial and bicondylar tibial plateaufractures. J Orthop Trauma. 2008;22:357–362.
8. Barei DP, O’Mara TJ, Taitsman LA, et al. Frequency and fracturemorphology of the posteromedial fragment in bicondylar tibial plateaufracture patterns. J Orthop Trauma. 2008;22:176–182.
9. Luo CF, Jiang R, Hu CF, et al. Medial double-plating for fracturedislocations involving the proximal tibia. Knee. 2006;13:389–394.
10. Hackl W, Riedl J, Reichkendler M, et al. Preoperative computerizedtomography diagnosis of fractures of the tibial plateau. Unfallchirurg.2001;104:519–523.
11. Wicky S, Blaser PF, Blanc CH, et al. Comparison between standardradiography and spiral CT with 3D reconstruction in the evaluation,classification and management of tibial plateau fractures. Eur Radiol.2000;10:1227–1232.
12. Khan RM, Khan SH, Ahmad AJ, et al. Tibial plateau fractures. A newclassification scheme. Clin Orthop Relat Res. 2000;375:231–242.
13. Lobenhoffer P, Gerich T, Bertram T, et al. Particular posteromedial andposterolateral approaches for the treatment of tibial head fractures.Unfallchirurg. 1997;100:957–967.
14. Schatzker J, Tile M. The Rationale of Operative Fracture Care, 2nd ed.Berlin: Springer-Verlag; 1996:390–391.
15. Macarini L, Murrone M, Marini S, et al. Tibial plateau fractures:evaluation with multidetector-CT. Radiol Med. 2004;108:503–514.
16. Bhattacharyya T, McCarty LP 3rd, Harris MB, et al. The posteriorshearing tibial plateau fracture: treatment and results via a posteriorapproach. J Orthop Trauma. 2005;19:305–310.
17. Georgiadis GM. Combined anterior and posterior approaches for complextibial plateau fractures. J Bone Joint Surg Br. 1994;76:285–289.
18. De Boeck H, Opdecam P. Posteromedial tibial plateau fractures. Operativetreatment by posterior approach. Clin Orthop Relat Res. 1995;320:125–128.
