Floating Knee Injury

A floating knee is an ipsilateral fracture of the femur and the tibia. The knee joint and its surrounding soft tissues become an unstable segment, disconnected from the axial skeleton above by the femoral fracture and below by the tibial fracture, so that the knee "floats" free between two fracture sites.

Who and how. It accounts for 2-4% of lower extremity fractures, with a male predominance of 3:1 and a peak age of 20-40 years. The mechanism is high-energy trauma:

• Motor vehicle accident - 70%
• Motorcycle accident - 20%
• Pedestrian versus vehicle - 5%

The rest of the patient. The large majority of patients have significant associated injuries. No source cited on this page gives a population figure, and the proportion depends entirely on the trauma system reporting it. Mortality is 5-15%, usually from the associated injuries, and 30-40% of the fractures are open.

Associated injuries. The incidence ranges below are conventional teaching ranges from the floating-knee series literature, not single-cohort measurements. The verified anchors on this page are the Evidence Base cards: occult ligament and meniscal injury in Andrade-Silva, and infection, DVT and nerve-palsy rates in Kenmegne.

The popliteal artery. The popliteal vessels and nerves traverse the floating segment. The artery is fixed at the adductor hiatus proximally and the soleus arch distally, and that tethered position makes it vulnerable to traction injury. Intimal tears may cause delayed thrombosis.

Muscles and deformity. The quadriceps mechanism spans the femoral fracture and the gastrocnemius origin crosses the knee, and both contribute to the deforming forces and to stiffness. The proximal femoral fragment is abducted and flexed by the abductors and iliopsoas, while the tibial deformity varies with the level of the fracture. If it is not mobilised early, the knee tends toward a flexion contracture.

Load transmission. Normally, axial load passes from the femur through the knee to the tibia. In the floating knee axial stability is lost completely, so fixation of both fractures is required before the limb can bear weight.

Fraser's classification is descriptive and orders severity. Type I is extra-articular in both bones; type II has articular involvement, subdivided by whether the tibial surface, the femoral surface or both are involved. It remains the most widely used system, but it describes only the bony pattern and has limited prognostic power because it ignores soft-tissue and extensor-mechanism injury (see Controversies).

"Best" is relative. Read "Type I has the best prognosis" as a relative statement only. Andrade-Silva's type I patients, treated with internal fixation of both bones, still had unsatisfactory Karlstrom and Lysholm scores, a 61% peak-torque deficit in the knee extensors, four nonunions and two chronic osteomyelitides. Chouhan found that even Fraser IIA scored significantly worse than matched controls for daily activities, sport and quality of life. No subtype returns to normal.

A child's floating knee is a genuinely different injury, governed by the physis, remodelling potential and growth, and Fraser (an adult, articular-versus-diaphyseal scheme) does not apply. The differential table and the Meccariello series both invoke a paediatric-specific system. The standard descriptive one is the Letts classification (Letts, Vincent and Gouw, 1986; sometimes cited in the literature as "Letts-Ran"), which has five types in increasing severity:

• Type A - both fractures closed and diaphyseal
• Type B - one fracture open
• Type C - intra-articular (or physeal/epiphyseal) extension of one fracture
• Type D - an open fracture with associated intra-articular involvement
• Type E - both fractures open

Open patterns (Letts D and E) carry the worst prognosis, with type E the most severe.

The physis and length. The distal femoral and proximal tibial physes may be involved, risking growth arrest, progressive angular deformity and limb-length discrepancy, which must be tracked to skeletal maturity. Length can change in both directions: post-fracture overgrowth (physeal stimulation) and shortening (arrest or malunion) are both described, so serial length assessment is essential.

Treatment. Remodelling potential allows more diaphyseal fractures to be managed non-operatively than in adults, but Letts' original series showed poor results when both fractures were treated non-operatively, so at least one fracture should be rigidly stabilised. Implant choice favours flexible intramedullary (elastic stable) nailing and physeal-sparing fixation over the rigid reamed antegrade nails used in adults, to protect the open physis.

Outcome. Contemporary paediatric outcome data (Tontanahal et al., 2023) show that a high Injury Severity Score and bone loss greater than 4 cm, not age, open status or comminution alone, are the strongest predictors of poorer sports and transfer/mobility function.

Definitive management of the individual paediatric femoral or tibial fracture, and Salter-Harris grading of physeal injury, are developed in the dedicated paediatric fracture and growth-plate topics; this section covers only the floating-knee-specific principles.

Life before limb. The floating knee is a high-energy polytrauma injury and assessment follows ATLS priorities, starting with haemodynamic stability. The large majority have associated injuries, with the head, chest and abdomen the ATLS priorities, and assessment of the floating segment is secondary to life-saving measures.

The limb. Note obvious deformity and length discrepancy, and inspect the skin, because open fractures are common. Then examine:

• Compartments - monitor both the thigh (3 compartments) and the leg (4 compartments)
• Vascular status - popliteal, dorsalis pedis and posterior tibial pulses, with an ABI if suspicious
• Nerves - the peroneal nerve is the most vulnerable

The ligaments. More than 50% have ACL or PCL injuries, but in the acute phase the knee is often impossible to assess definitively because of pain and swelling. Keep a high index of suspicion and stress the knee under anaesthesia during fixation, then obtain an MRI once the fractures are stabilised and the swelling subsides. Ligamentous stability is often the main determinant of long-term disability.

Radiographs. The single commonest error is under-imaging: an obvious femoral shaft fracture distracts from a subtle proximal tibial or plateau fracture. The standard trauma series therefore covers the whole limb:

• Full-length femur, AP and lateral
• Full-length tibia, AP and lateral
• AP pelvis, to check for associated hip fractures
• Dedicated knee series, AP, lateral and oblique

On these films, identify whether each fracture is diaphyseal or articular, assess comminution and check joint congruity. A true lateral of the knee is essential to rule out occult subluxation.

Advanced imaging. CT angiography deserves particular attention: high-energy floating knee injuries involve significant soft-tissue displacement, and intimal injuries of the popliteal artery can be occult initially.

The diagnosis of a floating knee is radiographic, a fracture of both the ipsilateral femur and tibia, so the real differential is recognising which adjacent injury patterns can masquerade as it, accompany it, or be mistaken for it on initial trauma imaging. The single commonest error is under-imaging, covered under Investigations.

Damage control. In the unstable patient, both the femur and the tibia are stabilised temporarily with spanning external fixation to allow resuscitation. Definitive fixation follows once physiological parameters normalise.

Definitive fixation. In the stable patient, address both fractures at the same sitting if possible, which reduces hospital stay and allows early motion. The strategy must be tailored to the specific fracture pattern, and the usual constructs by Fraser type are:

• Type I - double IM nailing (antegrade femur and tibia)
• Type IIA - femoral nailing and tibial plateau ORIF
• Type IIB - distal femoral ORIF and tibial nailing
• Type IIC - dual ORIF (distal femur and tibial plateau)

Sequence. Most surgeons prefer to fix the femur first. It restores limb length and gross alignment, facilitates the subsequent tibial reduction and makes patient positioning more ergonomic. Tibial fixation follows immediately to finalise the stability of the construct.

Femoral nailing. Entry is piriformis or greater trochanteric. Reamed nailing is preferred for stability, with interlocking screws to control rotation.

Distal femoral ORIF. Through a lateral parapatellar approach, the articular surface is reduced anatomically first and then fixed with a locking plate or a retrograde nail, depending on complexity. Bone quality and comminution dictate the final implant choice.

Tibial nailing. For the shaft, the suprapatellar approach is advantageous for positioning, and static locking maintains length.

Tibial plateau ORIF. Precise articular restoration is the priority, with medial, lateral or dual plating as required. Rigid fixation is required to allow early knee range of motion.

Compartment syndrome. Both the thigh and the leg compartments are affected, and the thigh is often overlooked. Maintain a high index of suspicion with serial evaluations, and keep the threshold for prophylactic fasciotomy low.

Vascular injury. The popliteal artery is the most at risk because of its tethered anatomy, and the injury may present late through an intimal tear. Once diagnosed it requires immediate revascularisation: the limb viability window is 6-8 hours of warm ischaemia.

Infection. Rates are higher than for isolated fractures (10-15%). Open fractures require debridement and antibiotics per the Gustilo protocol, and external fixation brings pin-site infections.

Knee stiffness. The most common long-term problem, in 20-50%, due to periarticular scarring and quadriceps adhesions. Prevention is early range of motion and avoiding prolonged immobilisation; treatment is aggressive physiotherapy, with manipulation under anaesthesia considered.

Malunion and nonunion. Both fracture sites are at risk, and smoking, diabetes and infection increase the risk. Revision fixation or bone grafting may be required.

Post-traumatic arthritis. Particularly prevalent after Fraser type II injuries, which carry a significantly higher burden of post-traumatic OA. Articular damage and ligament instability both contribute, and total knee arthroplasty may be required in the long term.

The first days. From day 0 to 3 the priorities are DVT prophylaxis with LMWH, multimodal analgesia, neurovascular monitoring and wound inspection. If fixation is stable, CPM or active-assisted range of motion starts from day 1, with quadriceps setting exercises and ankle pumps.

Weight-bearing depends on the fracture pattern and the fixation.

Rehabilitation runs in three phases.

What to tell the patient. No source cited on this page reports a return-to-work rate. Kenmegne's series is the closest usable outcome: excellent to good in 61%, acceptable in 6% and fair to poor in 33% by Karlstrom-Olerud at a mean of 51 months. Meccariello's cohort required an average of six operations, and in some cases as many as 23, which is the number patients most need to hear. Manual labourers typically have worse outcomes.

By Fraser type. Higher types correlate with increased morbidity. Read the type I row with the caveat under Classification: even the best subtype leaves substantial residual disability.

Arthroplasty. Severe intra-articular injuries (Type IIC) often lead to debilitating post-traumatic osteoarthritis, and total knee arthroplasty in this setting is technically challenging. Retained hardware, malalignment, metaphyseal bone loss, patella baja and dense periarticular scar make exposure and implant seating demanding, and revision-style implants and stems are often required. Planning points:

• Stage hardware removal
• Screen for low-grade infection before arthroplasty
• Plan an extensile exposure, with tubercle osteotomy if needed
• Anticipate constrained or hinged implants for incompetent ligaments

Almost every floating knee series in this topic, including the Hegazy cohort and the contemporary Kenmegne series, reports results as "excellent / good / acceptable / fair / poor", but the underlying instrument is rarely named. It is the Karlström-Olerud functional grading, described by Karlström and Olerud in their original 1977 ipsilateral femur-and-tibia series, the same paper that first showed rigid fixation of both bones gives fewer complications, shorter hospital stay and better return to work.

What it grades. The score combines patient-reported symptoms and objective limb findings across several domains:

• Pain and symptoms in the thigh and leg
• Symptoms in the knee and ankle joints
• Walking ability
• Ability to return to work or normal activity
• Angular or rotational deformity
• Limb shortening
• Restriction of knee and ankle motion

How it is read. Each domain is graded, and the overall category is conventionally set by the least favourable domain, so a single poorly recovered parameter, such as a stiff knee or a residual malalignment, pulls the whole result down. That makes it a deliberately demanding, patient-relevant measure, and explains why even technically united floating knees often score only "acceptable" or "fair".

Why this score. Because it captures the knee joint and the whole limb together, it suits an injury that isolates the knee between two fractures, which is why it, rather than a generic hip or knee score, remains the reporting standard in this literature. Generic joint scores (Knee Society Score, WOMAC, KOOS, SF-36) are reported alongside it but were not designed for this two-fracture pattern. Newer prognostic schemes (Meccariello, 2024) argue that soft-tissue and extensor-mechanism damage predict these functional scores better than the bony Fraser type does.

Global Epidemiology

• The floating knee is uncommon, accounting for roughly 2-4% of lower-limb trauma; a contemporary single-centre series reported an incidence of 2.32% of all lower-limb injuries.
• Road traffic and motorcycle trauma dominate the mechanism worldwide; the burden is disproportionately high in low- and middle-income countries with rising motorisation and a young, male-predominant cohort.
• Outcomes are governed less by geography than by soft-tissue status and associated injuries, which are consistent across published series.

Side-by-Side Guideline Framework

Where these differ, the divergence is mainly on timing (early total care versus damage control) and is reconciled by patient physiology rather than by national preference.

Registry and Trauma-System Notes

• National joint replacement registries (NJR, AJRR, AOANJRR, SHAR, NZJR) capture the downstream burden of post-traumatic arthritis requiring arthroplasty after Fraser type II injuries, but no registry tracks the acute floating knee itself.
• Mature trauma systems with rapid pre-hospital retrieval and Level 1 centres favour single-stage early total care in stable patients; the EPOFF biological data underpin damage control where physiology is borderline.

High- vs Limited-Resource Practice Variation

• High-resource settings: combined ortho-plastic teams, CT angiography on demand, IM nailing of both bones in one sitting, and early MRI of the knee.
• Limited-resource settings: greater reliance on external fixation (limited implants, contaminated open injuries, delayed presentation), staged conversion, and clinical rather than imaging-based vascular assessment. The principle of stabilising both bones to permit early knee motion remains universal.

• Is Fraser still fit for purpose? Fraser describes only the bony pattern. A 2024 multicentre study of 168 floating knees found Fraser severity did not reliably predict function, whereas soft-tissue and extensor-mechanism damage did, and proposed a new prognostic system. Fraser remains the lingua franca for exams and communication, but is increasingly seen as prognostically weak.
• Early total care vs damage control. Single-stage fixation in the stable patient is well supported, but the threshold for damage control in the "borderline" physiology patient is debated. The EPOFF biological data favour temporising external fixation in borderline patients, while some high-performing trauma centres report good results with early total care in carefully selected borderline cases.
• Femur-first vs tibia-first. Femur-first (length and alignment) is the common teaching, but no high-level evidence proves superiority; some surgeons fix the simpler or more accessible fracture first, or fix the tibia first to allow knee extension for femoral nailing.
• Routine MRI of the knee. Occult ligament and meniscal injuries are common (and may dominate long-term disability), but the optimal timing and whether routine MRI changes management in the acute setting is unresolved; many units image selectively once the knee can be examined.
• How "best" is the best subtype? Even Fraser type I (both extra-articular) carries large isokinetic strength deficits and unsatisfactory function in some series, challenging the assumption that extra-articular equals good outcome.

1. Fraser RD, Hunter GA, Waddell JP. Ipsilateral fracture of the femur and tibia. J Bone Joint Surg Br. 1978;60-B(4):510-515.
2. Veith RG, Winquist RA, Hansen ST Jr. Ipsilateral fractures of the femur and tibia. J Bone Joint Surg Am. 1984;66(7):991-1002. PMID 6480657.
3. Hegazy AM. Surgical management of ipsilateral fracture of the femur and tibia in adults (the floating knee). Clin Orthop Surg. 2011;3(2):133-139. PMID 21629474.
4. Andrade-Silva FB, et al. Functional results and isokinetic muscle strength in Fraser type I floating knee. Injury. 2017;48 Suppl 4:S2-S5. PMID 29145963.
5. Kenmegne GR, et al. The current issues and challenges in the management of floating knee injury. Front Surg. 2023;10:1164032. PMID 37206352.
6. Pape HC, et al. (EPOFF Study Group). Impact of intramedullary instrumentation versus damage control for femoral fractures on immunoinflammatory parameters. J Trauma. 2003;55(1):7-13. PMID 12855874.
7. Chouhan D, et al. Comparison of functional outcomes among subtypes of Fraser's type II floating knee. Chin J Traumatol. 2020;24(1):25-29. PMID 33339679.
8. Meccariello L, et al. Floating knee: a new prognostic classification. Injury. 2024;55 Suppl 4:111471. PMID 39542575.