Rotationplasty

Rotationplasty, the Borggreve-Van Nes procedure, is a limb-salvage technique used primarily for malignant bone tumours of the distal femur or proximal tibia in skeletally immature patients. The tumour-bearing segment is resected en bloc and the distal limb is rotated 180 degrees, so that the ankle joint becomes the functional knee joint. The rotated foot is then fitted with a below-knee prosthesis.

What the patient keeps. The neurovascular structures to the distal limb are preserved, so the new knee has active motion (ankle motion now working as knee flexion and extension) and proprioceptive feedback. The operation avoids the revision burden that expandable endoprostheses carry in growing children. Despite its unusual cosmetic appearance, long-term studies show quality of life and psychosocial functioning approaching those of healthy peers in appropriately selected and counselled patients.

History. The procedure evolved from non-oncological to oncological applications. Borggreve described it first, in 1930, but the eponym "Van Nes rotationplasty" comes from Van Nes, who popularised it for congenital conditions in 1950.

Indications. The primary oncological indications are:

• Osteosarcoma of the distal femur or proximal tibia in skeletally immature patients
• Ewing sarcoma of the distal femur or proximal tibia
• Aggressive benign tumours (GCT, ABC) with extensive bone destruction

Beyond these, it is used in congenital PFFD, for which Van Nes popularised it, and as a salvage after a failed limb-salvage reconstruction.

Who. The typical patient is 6-14 years old and skeletally immature, and the peak incidence matches the demographics of osteosarcoma (10-20 years). The procedure itself has no gender predilection.

The sciatic nerve. It runs posteriorly and supplies the hamstrings and all the leg and foot muscles below the knee. It must be carefully preserved and mobilised to allow 180-degree rotation without tension, and a minimum of 15cm of nerve length is required.

The femoral vessels. The femoral artery and vein become the popliteal vessels behind the knee. They must be mobilised enough to rotate without kinking or tension, and vessel length is critical.

The muscles change jobs. After rotation the ankle is the functional knee, and the muscles that moved it now move the knee:

• Gastrocnemius-soleus (the ankle plantarflexors) become the knee extensors, the "quadriceps equivalent": plantarflexion extends the knee
• Tibialis anterior (the ankle dorsiflexor) becomes the knee flexor, the "hamstring equivalent", preserving active motor control of the new joint
• The peroneal muscles contribute to knee flexion
• The triceps surae reflex becomes functionally equivalent to the quadriceps reflex

The new joint. The ankle provides 20-45 degrees of "knee" flexion and extension. Proprioceptive feedback is maintained through the intact sensory nerves.

The Van Nes classification groups rotationplasty by the location of the tumour resection and the level of osteosynthesis. The letter tells you the bone: A for the proximal tibia, B for the distal femur, C for the proximal femur or hip.

Type A (Van Nes procedure). For proximal tibia tumours. The proximal tibia is resected with the tumour, the distal tibial segment is rotated 180 degrees and fused to the femoral condyles, and the foot is positioned to function as the prosthetic knee joint. The knee joint capsule attachments to the femur are preserved.

Type B (Borggreve procedure). For distal femur tumours, the most common oncological indication, and the most common type for osteosarcoma of the distal femur. The distal femur is resected with the tumour, the entire leg below the knee is rotated 180 degrees, and the proximal tibia is fused to the proximal femur. The subtypes follow the extent of femoral resection:

• B1 - standard distal femur resection
• B2 - extended resection including the diaphysis
• B3 - very proximal femur involvement

A against B. Type A has fewer vascular challenges than Type B: a shorter segment is rotated, so less vessel mobilisation is needed and the vascular management is technically simpler. Fusion is at the distal femur rather than the proximal femur, and overall limb shortening is less than with a femoral resection. Type B needs greater vessel mobilisation and produces more significant shortening.

Type C (modified or proximal). For proximal femur or hip involvement, with a resection that includes the proximal femur or hip joint, tibiopelvic arthrodesis and a complex reconstruction of hip mechanics. It is subtyped C1 at the intertrochanteric level and C2 at the hip disarticulation level with tibiopelvic fusion. Type C is rarely performed for primary bone tumours and more commonly used for congenital conditions (PFFD); it leaves a significant limb length discrepancy and a challenging rehabilitation.

The ideal candidate. A skeletally immature child aged 6-14 years with a distal femur or proximal tibia tumour that can be resected en bloc with clear margins. Open physes are preferred because the skeletally immature patient benefits most, avoiding the multiple revisions an endoprosthesis would need. There should be no metastatic disease, or only controlled oligometastatic disease, and the tumour should respond to neoadjuvant chemotherapy where that is indicated.

The ideal limb. Locally, the ideal case has:

• An intact sciatic nerve, not encased by tumour
• No major vessel or nerve invasion, and femoral vessel length adequate for rotation
• An adequate soft-tissue envelope preserved
• Sufficient residual limb length after resection

Contraindications. Short vessels are a relative contraindication; vessel length inadequate for a 180-degree rotation is an absolute one. Poor understanding or unrealistic expectations about cosmesis, and a patient or family unable to accept the altered body image, are also contraindications.

Vascular failure is the main risk: all three limbs lost to vascular compromise in Sawamura's series of 25 had undergone a vascular anastomosis (see Complications).

The relative contraindications are:

• Skeletal maturity (adolescents near skeletal maturity may prefer other options)
• Extensive soft-tissue involvement requiring skin grafting
• Prior radiation to the surgical field, compromising wound healing
• Ipsilateral foot or ankle pathology limiting function
• Severe psychological concerns despite counselling

Psychological preparation. Psychological assessment and family counselling are mandatory.

What the family must hear. Function is generally favourable against above-knee amputation, but be honest about how wide the reported range is. Psychosocial functioning approaches that of healthy peers in the long term, yet body-image and intimacy concerns persist in a substantial minority. The family keeps the option to decline and choose an alternative.

Planning is built on the same staging workup as any primary bone sarcoma, with added emphasis on the neurovascular anatomy that decides feasibility. Two investigations make or break candidacy, MRI for sciatic nerve status and angiography for vessel length and involvement; state both in the viva before discussing technique.

Local staging. Whole-bone MRI with contrast defines the intramedullary and soft-tissue extent, skip lesions and, critically, the relationship of the tumour to the sciatic and tibial nerve. Nerve encasement is the key contraindication, so MRI directly drives the choice between rotationplasty and amputation.

Vascular assessment. CT or MR angiography, or formal angiography, maps the femoral and popliteal vessels and any tumour involvement. Adequate vessel length and an uninvolved pedicle are prerequisites, and encasement requiring resection generally precludes the procedure.

Systemic staging. CT of the chest and whole-body imaging (bone scan or FDG-PET/CT) exclude pulmonary and skeletal metastases. Active widespread metastatic disease with a poor prognosis is a contraindication.

Biopsy and response. An image-guided core biopsy, through a tract planned for en bloc excision, confirms the histology. After neoadjuvant chemotherapy the percentage tumour necrosis informs prognosis and, per Sawamura, the risk of failure: poor response correlates with vascular failure.

Oncological principles take priority, so the resection is a wide or radical excision with clear margins. The sciatic nerve and femoral vessels are then mobilised widely enough to allow a 180-degree rotation, the limb is rotated, and the bones are stabilised with a plate, nail or external fixator. The heel is set 2-3cm above the contralateral knee, for the growth reasons set out in the next section.

Fixation. The nail, the locking plate, the external fixator and a combination each have their own advantages, drawbacks and best indication.

Pitfalls. The technical errors to avoid:

• Vessel kinking after rotation leads to ischaemia
• Sciatic nerve tension causes paresis
• An incorrect rotation angle (not the full 180 degrees)
• The wrong leg length, too long or too short
• Inadequate tumour margins, which mean oncological failure

The heel is positioned 2-3cm above the contralateral knee. Why is exactly what an examiner will push on, because the answer is the whole reason rotationplasty suits a growing child.

You resect the limb's fastest growth plate. In the common Type B resection, the distal femoral physis, the fastest-growing plate of the lower limb (roughly 9-10mm per year, contributing most of the femur's length), is removed with the tumour. The reconstructed "thigh", the proximal femur fused to the tibia, loses its main growth engine.

But the rotated leg keeps growing. The rotated segment retains its own physes, the proximal and distal tibial growth plates and the foot, and they continue to grow after surgery. Because the tibia now forms the thigh-equivalent above the new knee, this retained growth progressively lengthens the reconstructed thigh and lowers the new "knee", the ankle, over time.

So you deliberately start it too high. Were the two knees made level at operation, the retained tibial growth would leave the reconstructed side too long by maturity. The surgeon therefore overcorrects, placing the new knee about 2-3cm above the contralateral knee, anticipating that growth will bring the knee heights level at skeletal maturity, where the prosthetic knee joint then aligns properly.

How much is a growth prediction. The amount is estimated from remaining growth (skeletal or bone age, remaining-growth or multiplier methods): more overcorrection in a younger child with years of growth ahead, less in one near maturity. Too little leaves the reconstructed thigh too long; too much leaves it short, a leg length inequality.

Rotationplasty is not only a primary operation. Hillmann described it as a "surgical treatment modality after failed limb salvage procedure", and Merkel as "a reconstructive operation after tumor resection".

When. It is a recognised salvage option when a previous limb-salvage reconstruction has failed and the realistic alternative is an above-knee amputation, most often because of a chronically infected or failed distal-femoral endoprosthesis. The classic scenarios are:

• A chronically infected or failed distal-femoral or proximal-tibial endoprosthesis (megaprosthesis)
• A failed allograft or allograft-prosthetic composite
• Local recurrence after prior limb salvage

Why. It converts a failed, infected implant into a durable, implant-free biological reconstruction, removing the hardware and the infected or dead bone while preserving a functional, ankle-driven knee and avoiding amputation. With no implant, the ongoing loosening and revision burden that made the original reconstruction fail is gone.

Who. In the salvage setting it is offered to adolescents and adults as well as children, and the competing option is usually an above-knee amputation rather than another endoprosthesis.

Harder than a primary case. The sciatic nerve and vessels must still be free and of adequate length, which is more difficult after prior surgery, scarring, radiation and infection. The soft-tissue envelope is often compromised and complication rates are correspondingly higher, and candidacy still turns on an uninvolved, mobilisable neurovascular bundle.

The major complication rate is 5-15%. Set against the 2-5% vascular compromise in the table below, the one series carded in the evidence (Sawamura, 25 patients) lost 3 of 25 limbs (12%) to vascular compromise and secondary amputation, all after vascular anastomosis and with less than 95% tumour necrosis. Local recurrence depends on tumour factors and margins, not on the reconstruction itself.

The prosthesis. A below-knee (transtibial) type prosthesis is fitted, with a socket that encompasses the rotated foot, the prosthetic "knee" joint at ankle level and an energy-storing prosthetic foot at the distal end. The calcaneus becomes weight-bearing through the socket, and active ankle motion produces knee function in the prosthesis.

Gait training. The patient learns to plantarflex for knee extension in stance and to dorsiflex to initiate swing (knee flexion). Running and sport are achievable with practice, and gait is typically independent by 3-6 months.

Function. Reported MSTS scores for rotationplasty span ~60-85%, generally favourable against above-knee amputation (60-70%) and comparable to endoprosthetic replacement (70-80%); higher scores reflect better walking, running and activity participation. Long-term series report a wide range: older series scored higher, and Gradl (2015) found a mean of 64% at 14 years. In congenital PFFD, the Floccari gait study in the evidence below found no gait, energy or patient-reported advantage, so the two indications are judged separately.

Activity. Patients can typically run, jump, swim and take part in sport:

• Running is achievable in most patients
• Jumping is possible with training
• Swimming is excellent, and a preferred activity
• Cycling may require adapted equipment
• Participation in team sports such as soccer and basketball is reported

Psychological outcomes. Long-term studies, with a mean follow-up of 6-14 years, show psychosocial functioning and general quality of life approaching those of healthy peers in appropriately selected patients (Veenstra 2000; Gradl 2015). Two-thirds of patients remain actively engaged in sport and wear the prosthesis continuously. Body-image and intimacy concerns persist in a substantial minority, one-third to one-half in Veenstra 2000, which is not trivial, and children generally adapt more readily than adolescents.

Against the alternatives. The table sets rotationplasty against the two reconstructions it competes with. Unlike either, it has no stump or implant-loosening issues.

Global Epidemiology

Osteosarcoma and Ewing sarcoma are the dominant indications. Osteosarcoma has a bimodal age distribution with the larger peak in the second decade and an incidence of roughly 3-4 cases per million per year in children and adolescents; the distal femur and proximal tibia (the classic rotationplasty sites) together account for the majority of lower-limb cases. Ewing sarcoma is somewhat rarer (~1-3 per million) and shows a male predominance. Because rotationplasty is reserved for selected skeletally immature patients in whom the sciatic nerve is spared, it represents a small fraction of all sarcoma reconstructions worldwide and is concentrated in specialist paediatric sarcoma centres.

Guidelines, Side by Side

No society publishes a procedure-specific rotationplasty guideline; recommendations sit within broader bone-sarcoma pathways. The common thread is management within a specialist sarcoma multidisciplinary team (MDT) with limb-salvage decision-making individualised to tumour stage, skeletal maturity and patient preference.

Registries and Practice Variation

There is no dedicated rotationplasty registry; outcome data come from single-centre and pooled sarcoma-unit series. Practice varies markedly with resources and culture: in high-income settings the main competitor is the (expandable) endoprosthesis, and rotationplasty uptake is partly limited by cosmetic acceptability and the availability of revision arthroplasty. In limited-resource settings, rotationplasty's implant-free durability and avoidance of repeated lengthening procedures can make it relatively more attractive, provided skilled microvascular/oncological surgery and quality prosthetic services exist. Cultural attitudes to the visibly rotated foot strongly influence acceptance and must be explored individually rather than assumed.

Controversies and Areas of Uncertainty

Related pages: Osteosarcoma is the disease that generated this operation and sets the resection margin that decides whether the sciatic nerve can be preserved; Ewing Sarcoma for the other paediatric bone sarcoma in which rotationplasty is offered; Limb Salvage Surgery Principles for the margin, staging and reconstruction framework this sits inside; Allograft-Prosthetic Composite Reconstruction and Megaprosthesis Salvage for the endoprosthetic alternatives whose late loosening and revision burden are the argument for a biological reconstruction in a growing child; Proximal Femoral Focal Deficiency for the congenital indication, where the Floccari comparison carded above found no functional advantage and the comparator is a Syme amputation with a working prosthetic knee; Hip Disarticulation and Forequarter Amputation for the ablative end of the spectrum; and Phantom Limb Pain for the complication that rotationplasty is specifically intended to avoid by preserving the neurovascular bundle and the foot.

1. Borggreve J. Kniegelenkersatz durch das in der Beinlangsachse um 180 Grad gedrehte Fussgelenk. Arch Orthop Trauma Surg. 1930;28:175-178.

2. Van Nes CP. Rotation-plasty for congenital defects of the femur. J Bone Joint Surg Br. 1950;32-B(1):12-16.

3. Salzer M, Knahr K, Kotz R, Zielinski C. Treatment of osteosarcomata of the distal femur by rotation-plasty. Arch Orthop Trauma Surg. 1981;99(2):131-136.

4. Winkelmann WW. Rotationplasty. Orthop Clin North Am. 1996;27(3):503-523.

5. Hillmann A, Gosheger G, Hoffmann C, Ozaki T, Winkelmann W. Rotationplasty - surgical treatment modality after failed limb salvage procedure. Arch Orthop Trauma Surg. 2000;120(10):555-558.

6. Veenstra KM, Sprangers MA, van der Eyken JW, Taminiau AH. Quality of life in survivors with a Van Nes-Borggreve rotationplasty after bone tumour resection. J Surg Oncol. 2000;73(4):192-197.

7. Gradl G, Postl LK, Lenze U, et al. Long-term functional outcome and quality of life following rotationplasty for treatment of malignant tumors. BMC Musculoskelet Disord. 2015;16:262.

8. Krajbich JI. Rotation-plasty in the management of proximal femoral focal deficiency. Prosthet Orthot Int. 1991;15(2):100-107.

9. Sawamura C, Hornicek FJ, Gebhardt MC. Complications and risk factors for failure of rotationplasty: review of 25 patients. Clin Orthop Relat Res. 2008;466(6):1302-1308.

10. Kotz R, Salzer M. Rotation-plasty for childhood osteosarcoma of the distal part of the femur. J Bone Joint Surg Am. 1982;64(7):959-969.

11. Merkel KD, Gebhardt MC, Springfield DS. Rotationplasty as a reconstructive operation after tumor resection. Clin Orthop Relat Res. 1991;(270):231-236.

12. Hillmann A, Rosenbaum D, Gosheger G, et al. Rotationplasty type B1 versus type BIIIa. Gait Posture. 2001;14(1):51-59.

13. Floccari LV, Jeans KA, Herring JA, Johnston CE, Karol LA. Comparison of outcomes by reconstructive strategy in patients with prostheses for proximal femoral focal deficiency. J Bone Joint Surg Am. 2021;103(19):1817-1825.