Limb-sparing pelvic resection: zone dictates reconstruction; complications dominate outcome
- Internal hemipelvectomy preserves the limb; external (hindquarter) amputation removes it β the choice hinges on neurovascular involvement and soft-tissue coverage, not tumour size alone
- Biopsy must be along a tract resectable en bloc with the specimen, placed by or after discussion with the resecting surgeon
- Zone I and III resections often need no skeletal reconstruction; zone II demands a decision between biological, endoprosthetic and pseudarthrosis solutions
- Deep infection 20 to 40 percent is the dominant complication and the most common reason endoprosthetic reconstructions fail
- Margins trump reconstruction sophistication: an elegant custom implant after a contaminated margin is an oncological failure
- βHip transposition (Friedel) gives surprisingly good MSTS scores (often 60 to 70 percent) with the lowest implant-related complication burden β never dismiss it as a 'lesser' option
- βSaddle prosthesis is essentially historic: high rates of migration, infection and poor function
- βZone III resections threaten the bladder, urethra, spermatic cord/round ligament and obturator neurovascular bundle β catheterise and consider ureteric stents
- βSciatic nerve encasement plus femoral vessel involvement flips the plan to external hemipelvectomy
A transgluteal or posterior biopsy of a periacetabular tumour can contaminate flaps needed for closure and convert a limb-sparing case to an amputation. Biopsy along the planned incision, resectable en bloc, ideally image-guided core at the treating sarcoma centre.
Deep infection complicates 20 to 40 percent of pelvic resection-reconstructions β long operating time, large dead space, proximity to perineum, irradiated tissue. Plan flaps, drains, antibiotic strategy and warn the patient it may cost the reconstruction.
The common and external iliac vessels, superior gluteal artery (greater sciatic notch) and corona mortis are all at risk. Proximal vascular control of the iliac vessels before osteotomies is a core safety step; massive haemorrhage is the intraoperative killer.
Involvement of the sciatic nerve plus femoral neurovascular bundle, fungating or infected tumour, or unreconstructable soft-tissue loss mandates external hemipelvectomy (covered in the hindquarter amputation topic). A salvaged limb that is insensate, ischaemic and painful is worse than a well-rehabilitated amputation.
Definitions, Indications and Patient Selection
Internal hemipelvectomy is en bloc resection of part or all of the innominate bone with preservation of the lower limb. It contrasts with external hemipelvectomy (hindquarter amputation), where limb and hemipelvis are removed together.
Principal indications
- Primary bone sarcoma of the pelvis: chondrosarcoma (the most common pelvic primary in adults, chemo- and radio-resistant, so surgical margin is everything), Ewing sarcoma (after neoadjuvant chemotherapy), osteosarcoma
- Selected soft-tissue sarcomas invading the pelvis
- Carefully selected solitary metastases or locally aggressive benign lesions (giant cell tumour) where curettage is inadequate
- Occasionally infection or failed prior reconstruction (salvage setting)
Selection for limb salvage requires all of:
- Achievable oncological margin (wide for sarcoma) with the limb retained
- Preservable femoral neurovascular bundle AND at least one major nerve to the limb (classically the sciatic; loss of femoral nerve alone is compatible with useful function with bracing)
- Soft-tissue envelope adequate for closure (with or without flaps)
- Physiology tolerating a long, high-blood-loss procedure
- A motivated patient who understands complication rates and the prolonged rehabilitation
Sciatic nerve encasement combined with femoral vessel or nerve involvement, fungating or infected tumour, prior contaminating surgery/biopsy through all flap options, or tumour progression through chemoradiotherapy with unreconstructable soft tissues β these favour external hemipelvectomy. Discuss and consent for both procedures preoperatively; the final call may be intraoperative.
Enneking-Dunham Zones and What Each Demands
The Enneking-Dunham classification (JBJS 1978) is the universal language for pelvic resections. Zone dictates both technical difficulty and reconstruction.
- Bone resected
- Iliac wing between SI joint and supra-acetabular bone
- Key at-risk structures
- Superior gluteal NV bundle, L5 nerve root, SI joint
- Reconstruction
- None if a bone bridge remains; iliosacral bar/screws with or without vascularised fibula if posterior ring discontinuity
- Bone resected
- Acetabulum plus adjacent ilium/pubis/ischium
- Key at-risk structures
- Femoral NV bundle, sciatic nerve, obturator bundle, hip abductors
- Reconstruction
- Required for hip function: transposition, ice-cream-cone, custom implant, allograft/APC, arthrodesis
- Bone resected
- Pubic rami and/or ischium
- Key at-risk structures
- Bladder, urethra, corona mortis, obturator bundle, spermatic cord
- Reconstruction
- No skeletal reconstruction; pelvic floor/hernia repair with mesh or flap
- Bone resected
- Resection crosses SI joint into sacral ala
- Key at-risk structures
- Lumbosacral trunk, sacral roots, iliac veins
- Reconstruction
- Spinopelvic (lumbopelvic) fixation with or without fibula strut if posterior column of ring lost
Combined resections are common β periacetabular chondrosarcoma frequently requires a zone I plus II (with the osteotomy through the ilium and the pubic root) or zone II plus III resection. Always describe a resection by its zones in the exam.
I A I SEnneking-Dunham zones
Hook:Walk the pelvis from top to bottom to back: Ilium, Acetabulum, Ischiopubic, Sacral.
Workup and Planning
- MRI of the whole pelvis with contrast β local staging: intraosseous and extraosseous extent, relationship to iliac vessels, sciatic and femoral nerves, SI joint, hip joint, bladder/rectum; defines osteotomy planes
- CT chest β pulmonary staging (bone sarcomas metastasise to lung)
- CT pelvis β cortical detail, matrix mineralisation, and the dataset for navigation or 3D-printed cutting guides/custom implants
- Whole-body staging β bone scan or FDG-PET per tumour type; PET valuable for Ewing response assessment
- Bloods, alkaline phosphatase/LDH where relevant; fertility and psychology referral in young patients
Operative Strategy by Zone
Indication: iliac wing tumour sparing acetabulum and SI joint (or taken with SI joint as zone I plus IV).
- Position: floppy lateral or lateral decubitus allowing anterior and posterior access
- Approach: utilitarian pelvic incision along the iliac crest (extensile ilioinguinal-type anteriorly, extending posteriorly toward the PSIS); elevate abductors laterally and iliacus medially off remaining bone only, keeping muscle cuff on tumour
- At-risk: superior gluteal artery at the greater sciatic notch (retraction injury causes torrential bleeding that retracts into the pelvis β control iliac inflow first), L5 nerve root and lumbosacral trunk anterior to the ala, femoral nerve within iliacus
- Osteotomies: supra-acetabular cut and cut adjacent to (or through) the SI joint under guide/navigation
- Reconstruction decision: if a supra-acetabular bone bridge remains, no reconstruction. If the ring is discontinuous between sacrum and acetabulum, options are: accept controlled proximal migration (modest limb shortening, often acceptable function), or restore continuity with an iliosacral bar/screw construct plus vascularised free fibula or structural allograft to fuse sacrum to residual ilium β favoured in young, high-demand patients where limb-length and abductor mechanics matter
Zone II Reconstruction: Options and Selection
Friedel-type hip transposition (iliofemoral pseudarthrosis): the femoral head is moved proximally and stabilised against the residual ilium or lateral sacrum, contained by a soft-tissue sling (capsule, mesh, abductor reattachment); a stable pseudarthrosis forms.
- Advantages: no implant, so no implant infection or loosening; resource-independent; durable; reoperation burden lowest of all options; surprisingly good MSTS scores β commonly 60 to 70 percent in reported series
- Disadvantages: limb shortening 2 to 5 centimetres (shoe raise), Trendelenburg gait, prolonged bracing while the pseudarthrosis stabilises
- Best for: extensive resections with poor bone stock for an implant, resource-limited settings, high infection risk (irradiated field, prior infection), and as the default fallback when endoprosthetic reconstruction fails
- Function (MSTS)
- Often 60 to 70 percent
- Dominant failure mode
- Shortening, Trendelenburg gait β but few reoperations
- Best-fit patient
- Poor bone stock, high infection risk, resource-limited, salvage default
- Function (MSTS)
- Approximately 70 percent in modern series
- Dominant failure mode
- Infection, dislocation
- Best-fit patient
- Preserved iliac bone stock, need for prompt off-the-shelf solution
- Function (MSTS)
- Approximately 70 percent, early weight bearing
- Dominant failure mode
- Infection; cannot adapt if margins change
- Best-fit patient
- Complex combined-zone defect, well-resourced centre
- Function (MSTS)
- Variable
- Dominant failure mode
- Infection, nonunion, graft fracture/resorption
- Best-fit patient
- Young patient, good envelope, bone-banking infrastructure
- Function (MSTS)
- Stable but stiff
- Dominant failure mode
- Nonunion; sitting difficulty
- Best-fit patient
- Heavy manual worker, salvage
- Function (MSTS)
- Poor
- Dominant failure mode
- Migration, iliac fracture, infection β historic
- Best-fit patient
- Essentially none in modern practice
The resection is planned for the margin; the reconstruction is planned for the defect β in that order. Shaving a margin to preserve the acetabulum or fit an implant converts a curative operation into a local-recurrence machine, and local recurrence after pelvic chondrosarcoma is frequently fatal. When forced to choose, choose the margin.
T-CASAZone II reconstruction options
Hook:Rebuilding the acetabulum needs a 'CASA' (house) with a 'T' (the transposed femoral head knocking on the ilium's door).
Complications: Rates, Prevention, Management
The dominant complication. Prevention: minimise operating time (staged procedures for the most complex cases), meticulous dead-space management, flap coverage of implants, drains, extended antibiotic prophylaxis in many units, silver or antibiotic coatings where available. Management: early aggressive debridement; established endoprosthetic infection usually mandates implant removal β conversion to hip transposition or flail hip, or rarely secondary external hemipelvectomy for uncontrolled sepsis.
Long incisions, irradiated tissue, thin posterior flaps. Prevention: preserve perforators, plan pedicled rectus abdominis or gluteal flaps prospectively with plastic surgery, avoid closure under tension. Management: early debridement and flap salvage/re-coverage before implant exposure supervenes.
Endoprosthetic dislocation reflects abductor loss and altered hip centre β dual-mobility or constrained liners, abduction bracing, careful abductor reattachment. Aseptic loosening, stem fracture and iliac stress fracture occur late; revision options narrow with each failure, ending at transposition or amputation.
Massive haemorrhage β proximal iliac control before osteotomy. Sciatic/femoral/obturator/lumbosacral trunk injury β identify and protect early; planned sacrifice is consented, not discovered. Bladder/urethral injury in zone III β catheter, stents, urology on standby. VTE risk is high β extended chemoprophylaxis plus mechanical. Local recurrence tracks with margin status: highest after intralesional or marginal resection; surveillance MRI plus CT chest per sarcoma protocol.
Functional Outcomes and Rehabilitation
- MSTS scores by construct: isolated zone I or III resections β often 80 percent or better; zone II reconstructions cluster around 60 to 70 percent regardless of method, with hip transposition performing comparably to endoprostheses at lower reoperation cost
- All zone II patients have some abductor deficiency: expect a Trendelenburg gait, walking aid in the first year, and counsel accordingly
Drains, flap monitoring, VTE prophylaxis, bed-to-chair transfers; abduction brace after endoprosthesis in many protocols; early isometric quadriceps and ankle work.
Touch or partial weight bearing per construct β earlier full weight bearing after well-fixed custom implants; transposition and fusion constructs protected longer while stabilising. Hip precautions after endoprosthesis.
Wean brace, progress to full weight bearing, abductor and core strengthening, gait retraining; shoe raise for transposition shortening; adjuvant therapy continues in parallel for chemosensitive tumours.
Functional plateau typically 12 to 24 months. Oncological surveillance: local MRI and CT chest at intervals per sarcoma guidelines for a minimum of 5 years (10 for chondrosarcoma in many units).
Guidelines, Registries & Global Practice
- Global epidemiology: pelvic sarcomas are rare β chondrosarcoma is the most common pelvic primary in adults, Ewing sarcoma in children and adolescents; all should be managed in designated sarcoma centres, a principle enshrined in UK NICE sarcoma service guidance, ESMO-EURACAN clinical practice guidelines and equivalent national sarcoma pathways worldwide
- ESMO-EURACAN bone sarcoma guidelines: biopsy at the treating centre along a resectable tract; neoadjuvant chemotherapy for osteosarcoma and Ewing; surgery-first for conventional chondrosarcoma; structured MDT decision-making
- NICE (UK): referral of suspected bone sarcoma to specialist centres; supports the centralisation model that pelvic resection outcomes depend on (volume-outcome relationship is well recognised for these procedures)
- Registry evidence: national arthroplasty registries (NJR, AOANJRR and others) capture little tumour endoprosthesis data; the meaningful datasets are collaborative sarcoma registries and multicentre implant cohorts (for example the European LUMiC collaboration and ISOLS-affiliated series), which consistently report infection as the leading reconstruction failure
- Practice variation by resource setting: custom 3D-printed implants and navigation concentrate in high-resource centres; hip transposition and resection without reconstruction deliver comparable oncological outcomes and respectable function where implant infrastructure, bone banking or revision capacity are limited β an important equity point: limb salvage does not require a custom implant
Controversies & Areas of Uncertainty
- Endoprosthesis versus hip transposition for zone II: no randomised data; endoprostheses restore length and early function but carry infection/dislocation/loosening; transposition avoids implant failure at the cost of shortening. Series-level MSTS scores overlap substantially
- Custom 3D-printed implants: enthusiasm and early series are favourable, but long-term fixation, cost-effectiveness and the inflexibility problem (implant fixed before final margins) remain unresolved
- Navigation and patient-specific guides: reduce intralesional resection rates in comparative series, but survival benefit is unproven
- Cell salvage in sarcoma surgery: theoretical tumour-cell reinfusion risk versus massive transfusion need; leucodepletion filters are used in some centres, avoidance in others
- Reconstruction at all for zone I discontinuity: many patients function well with an unreconstructed defect and modest shortening; when to fuse the ring remains judgement-based
- Thresholds for external hemipelvectomy: with modern flaps and vascular reconstruction, the boundary keeps moving β but survival is driven by tumour biology and margin, not by whether the limb was kept