Femoral Bone Loss in Revision Hip Arthroplasty: Paprosky Classification and Stem Selection

Revision femoral surgery is a problem of achieving durable fixation in compromised bone. Every decision follows from a single question: what bone remains that can support an implant? The Paprosky classification answers it as a fixation-zone ladder, and each rung names a stem.

• Metaphysis intact (Type I). Proximal cancellous bone is supportive and the canal is not widened, so primary-type fixation is still possible with a standard proximally loading uncemented or cemented stem, and the reconstruction is the simplest.
• Metaphysis gone, diaphysis intact (Types II and IIIA). Proximal support is lost but a generous isthmus remains, so fixation moves distally. Scratch-fit in the isthmus carries the load, and either a cylindrical or a tapered design is reliable.
• Diaphysis compromised (IIIB). Fixation must be found in a short, often capacious segment, and only a tapered fluted design does this reliably.
• No isthmus (Type IV). The femur cannot grip an uncemented stem. The choice is to rebuild the tube (impaction grafting), cement into what remains (the low-demand elderly patient), replace the proximal femur with allograft plus stem (allograft-prosthetic composite) or replace it with metal (proximal femoral replacement), chosen by bone stock, host biology and demand.

Bypass every defect. Cortical perforations, stress risers and old screw holes are fracture initiators. The stem must pass each of them by at least two cortical diameters of intact bone to reduce the risk of periprosthetic fracture.

Restore the abductor mechanism. A perfectly fixed stem with a detached trochanter is a failed hip. Trochanteric fixation and the soft-tissue strategy are planned up front, not as an afterthought.

The films. Obtain an AP pelvis and AP and lateral full-length femur views including the knee: they rule out distal deformity and old hardware and define the total canal length. CT adds cortical thickness mapping, version and occult lysis.

View larger

Fixation status of the existing stem. For a cemented stem, grade the cementation (Barrack) and apply the Harris loosening criteria. For a cementless stem, apply the Engh criteria: spot welds, absence of reactive lines around the coated portion, calcar rounding and a pedestal.

Level of bone loss. Is cancellous metaphyseal bone present? Is the calcar intact, thinned or absent? Where does structurally supportive cortex begin?

Isthmus quality. Measure the length of diaphysis with parallel, intact inner cortices distal to the defect. This is the measurement that separates IIIA from IIIB, and therefore a cylindrical or tapered stem from a tapered design only.

Cortical thickness. Cortices thinned to less than 2 mm over the intended fixation zone will not hold a press-fit stem and are at high risk of intraoperative fracture; consider prophylactic cabling or a different strategy.

Canal diameter. A canal wider than roughly 18 to 19 mm with thin cortices (a "stovepipe") predicts poor scratch-fit, and an ectatic canal pushes towards a Type IV solution.

Remodelling into varus. Draw the proximal and distal canal axes. Angulation with a lateralised greater trochanter means straight reamers will exit laterally, and the extended trochanteric osteotomy is planned from this film.

Stress risers. Old screw holes, cement mantle defects and previous perforations are marked now, because the stem length is chosen to bypass them.

Templating the removal. The cement column length, the distal cement plug and any pedestal below a cementless stem dictate whether an extended trochanteric osteotomy is needed and how long it must be. Mark the planned length so that the osteotomy still leaves at least 4 to 5 cm of intact diaphysis for fixation below it.

Templating the new stem. Choose the diameter at the isthmus: line-to-line or 0.5 mm under-ream for tapered titanium, 0.5 mm under-ream for an extensively coated cylinder per the manufacturer. Choose the length to give the required fixation segment plus defect bypass. Leg length and offset are templated from the contralateral hip, and with a modular system distal fixation and proximal reconstruction are templated separately.

The classification grades what is left, reading from the metaphysis downwards, and each grade carries a typical scenario and a preferred reconstruction.

View larger

The four boundaries. Each step down the ladder is a single question, and stating it is how the grade is defended in the exam:

• I versus II: is the metaphyseal cancellous bone supportive?
• II versus III: does the bone loss extend into the metadiaphysis?
• IIIA versus IIIB: greater or less than 4 cm of diaphyseal scratch-fit.
• III versus IV: is there any isthmus capable of gripping a stem at all?

Two workhorses. The fully porous-coated cobalt-chrome cylinder, popularised by Engh and Paprosky, was the historic workhorse; the Wagner conical taper, evolved into modular systems, is the current one. They hold the diaphysis differently, and they need different amounts of it.

How each holds on. The cylinder gains cortical scratch-fit from its distal geometry, with bone ingrowth wherever bone touches the coating, and it needs 4 to 6 cm of intact diaphyseal isthmus to do so. The taper wedges into the cone of the diaphysis and resists subsidence, while its longitudinal flutes cut into the endosteal cortex and resist rotation. The taper stops it sinking and the flutes stop it spinning, and it does both with only 2 to 3 cm of diaphyseal contact, which is its key advantage in a Type IIIB femur.

View larger

The cemented long stem. Fixation is by cement interlock. It retains a role in the elderly, low-demand or irradiated femur where bone ingrowth cannot be expected, and in the staged treatment of infection; its weakness is poor interlock in a sclerotic canal.

Impaction grafting. A cemented polished taper seated in impacted morselised allograft. It is the option for the young Type IV femur because it restores bone stock, at the price of fracture, subsidence and technical demand.

Allograft-prosthetic composite. A stem cemented into a bulk allograft that is then united to the host femur. It suits the young Type IV patient who also needs the abductors reattached, and it fails by nonunion, resorption and infection.

Proximal femoral replacement. An endoprosthesis with distal fixation, for the elderly, low-demand Type IV femur and for salvage. Its weaknesses are instability, which is why it is articulated with dual mobility, and loosening.

Position and equipment. Lateral decubitus on a radiolucent table with well-padded supports and the whole limb draped free, including the knee. Have full-length calibrated femur films, an image intensifier, the company-specific extraction kit, a high-speed burr, flexible osteotomes, trephines, cement removal instruments or an ultrasonic system, cerclage cables, the revision stem system with trials, allograft, and dual mobility or constrained options on the shelf.

Preparation. Exclude infection preoperatively with inflammatory markers and aspiration for culture and synovial white cell count, and send at least five intraoperative tissue samples with separate instruments. Tranexamic acid, cross-match and cell salvage.

Approach. Posterior, with extended proximal and distal exposure; identify and protect the sciatic nerve.

Removal. Dislocate, remove the head and attempt stem extraction in situ. A fixed stem or a long cement column calls for an extended trochanteric osteotomy, templated to reach the stem tip or cement plug: osteotomise one third of the canal circumference laterally, hinge the fragment on vastus lateralis and the anterior soft tissues, and pass a prophylactic cerclage cable distal to the osteotomy before reaming.

Reconstruction. Ream conically by hand, sequentially, down the true canal axis until the endosteal cortex is firmly engaged over the templated segment. Insert the fluted taper to the templated depth, stopping when reproducible advancement per blow ceases, and test axial (impactor load) and rotational stability. Trial proximal bodies for length, offset and version, trial-reduce with stability testing through a full arc, then commit the proximal body and engage the taper cleanly and dry.

At-risk structures. The sciatic nerve, from posterior retraction and from lengthening greater than about 4 cm; the femoral cortex, perforated by straight reamers in a varus femur, which is the ETO indication; the lateral cortex at the ETO hinge; and the remnants of the medial femoral circumflex.

Osteotomy fixation and closure. Reduce the ETO fragment around the proximal body and cable it as described in the next section. Repair the posterior capsule and short external rotators, and close in layers over a drain if preferred.

Aftercare. The check radiograph is the subsidence baseline. Toe-touch to partial weight bearing for 6 weeks protects the osteotomy and permits taper settling, then progressive loading, with posterior precautions throughout. Radiographs at 6 weeks, 3 months and 1 year are measured against the baseline.

Pitfalls and salvage.

• ETO nonunion (uncommon with cables; most unite): revision fixation with grafting
• Intraoperative fracture: cables plus a longer bypass
• Subsidence greater than 10 mm with instability: re-revise with a larger taper or longer stem
• Irreducible instability: dual mobility or a constrained liner

Bypass the calcar. Modern revision philosophy bypasses the calcar rather than reconstructing it: no modern revision stem depends on calcar support, and a calcar-replacing proximal body restores soft-tissue tension and version, not load transfer.

Keep the shell. Residual proximal shell bone is preserved and closed around the stem. Even non-structural bone contributes soft-tissue attachment, vascularity and a start for future revisions, and Wagner-type remodelling can partially restore it. Strut allograft with cables can reinforce a severely thinned medial or lateral cortex, or an ETO in poor bone.

The ETO fragment. Usually two or three cerclage cables, without overtightening on a thin fragment. A trochanteric claw-cable device is considered if the fragment is fractured or the abductors avulsed.

Trochanteric escape or fracture. A claw plate or grip-cable device. In the low-demand patient, accept a fibrous union with reasonable function: repeated fixation attempts in avascular bone often fail.

Absent abductors. Absent abductors, whether in the multiply revised hip or after proximal femoral replacement, carry the highest dislocation risk of any hip scenario. Use dual mobility or a constrained liner, consider abductor reconstruction to the implant or a trochanteric slide, and counsel the patient about limp and lifelong precautions.

View larger

Survivorship by class. The grade predicts the result, and the gap between the two stem designs opens at IIIB:

• Types I and II: excellent — survivorship of diaphyseal-fixing stems above 95 percent at 10 years in classic series.
• IIIA: extensively coated cylindrical stems 96 to 97 percent in Paprosky's own series, and 95 to 97 percent across Types II and IIIA at 10 to 15 years in the classic series; fluted tapered stems comparable at mid-term.
• IIIB: cylindrical stems historically failed in roughly one in five or worse; fluted tapered stems restored survivorship to around 90 to 95 percent at mid-term. This differential is the single most examinable fact in the topic.
• IV: all options carry higher complication rates. Impaction grafting gives durable results in experienced centres with the polished taper technique, but with meaningful fracture and subsidence rates; proximal femoral replacement survivorship is limited more by instability and infection than by fixation.

View larger

Dislocation. The commonest complication of revision hip arthroplasty overall. Restored offset and version through modularity and a sound soft-tissue repair are the first defence, and the abductor-deficient hip is articulated as described in the previous section.

Subsidence in context. Most subsidence of a tapered stem is normal behaviour rather than a complication: in Baldwin's 2024 series 79 per cent settled within the first month, by a mean of 2.9 mm, and then stopped, while 12 per cent exceeded 1 cm. The concerning pattern is subsidence that continues past a year, and the modifiable predictor is canal fill, not stem design.

• Registry evidence: the NJR (England and Wales), AOANJRR (Australia), SHAR (Sweden) and AJRR (United States) consistently report aseptic loosening, infection, dislocation and periprosthetic fracture as the leading reasons for femoral re-revision; registry data show the international migration from cylindrical cobalt-chrome to fluted tapered titanium stems for major bone loss over the last two decades, and steadily rising use of dual mobility in revision.
• AAOS and NICE address hip arthroplasty broadly; neither prescribes a stem for a given Paprosky class — implant choice remains a principles-based surgical decision, which is exactly how examiners frame it.
• BOA/BHS revision network guidance (United Kingdom) supports concentrating complex (Type IIIB and IV) revisions in higher-volume revision units with allograft access and multidisciplinary infection services — a model echoed in several European systems.
• Resource-limited settings: modular revision systems, bone banks and megaprostheses may be unavailable. Pragmatic alternatives include long cemented stems, monoblock Wagner-type tapers (cheaper, no junction), and cement-in-cement techniques; the classification and its fixation logic remain universally applicable even where the implant menu is narrow.
• Fresh-frozen allograft availability (impaction grafting, APC) varies enormously worldwide with regulatory and tissue-banking infrastructure — this drives genuine regional variation in Type IV management, with PFR used earlier where allograft is unavailable.

• Modular versus monoblock tapered stems: modularity aids intraoperative flexibility but introduces junction fracture and corrosion risk; comparative series show similar survivorship, and several units have returned to monoblock tapers for straightforward IIIA/IIIB femurs, reserving modularity for complex version or length problems.
• How short can the fixation segment be? The 2 to 3 cm figure for tapered stems is empirical; distal fixation into a short supracondylar segment blurs into "total femoral" territory and the failure threshold is not precisely defined.
• Impaction grafting versus tapered stems in Type IV: bone-stock restoration versus reproducibility; impaction grafting results are strongly centre-dependent, and many surgeons now attempt a tapered stem with cables even in near-Type IV femurs.
• PFR versus APC in the younger Type IV patient: APC restores bone but risks nonunion and resorption; PFR is reliable early but mortgages the future. No randomised data exist and practice follows unit expertise.
• Routine dual mobility in revision: increasingly adopted given dislocation dominance in registries, but with unresolved questions about intraprosthetic dislocation and long-term wear in younger patients.

Related pages. Getting to the femur: Extended Trochanteric Osteotomy is how a well-fixed stem or distal cement column is removed, and it shares this page's 4 cm rule - the osteotomy must leave enough isthmus below it for the stem chosen here. THA Templating and Pre-operative Planning is where the fixation zone is measured before anyone opens the hip.