Multicenter retrospective study of 71 shoulders undergoing arthroplasty for proximal humerus fracture sequelae. Proposes the Boileau four-type surgical classification and identifies the single factor — greater tuberosity osteotomy — that determines whether results will be good or poor. Average time from fracture to arthroplasty was 5 years 5 months; mean follow-up 19 months.
When you see a patient with proximal humerus fracture sequelae in clinic, your first question should be: can I implant this prosthesis without osteotomizing the greater tuberosity?
Boileau's data make the stakes clear. Types 1–2 (intracapsular, impacted) yield reliable results — 73% adjusted Constant score, mean elevation 123°, 81% satisfaction. Types 3–4 (extracapsular, disimpacted) lock patients into a "limited goals" conversation before they ever go to the OR.
For Type 3 surgical neck nonunions specifically, the authors argue arthroplasty should be abandoned unless the humeral head has totally resorbed. Internal fixation with intramedullary bone grafting is the preferred first-line treatment.
For Types 3–4 where arthroplasty is unavoidable, counsel patients explicitly that elevation will likely remain below 90° and satisfaction rates will be substantially lower. This paper is also why we use modular, adaptable prostheses in post-traumatic cases. Varying inclination and offset to fit the distorted anatomy is what allows us to skip the GT osteotomy in borderline cases.
Multicenter retrospective study of 71 shoulders undergoing arthroplasty for proximal humerus fracture sequelae. Proposes the Boileau four-type surgical classification and identifies the single factor — greater tuberosity osteotomy — that determines whether results will be good or poor. Average time from fracture to arthroplasty was 5 years 5 months; mean follow-up 19 months.
When you see a patient with proximal humerus fracture sequelae in clinic, your first question should be: can I implant this prosthesis without osteotomizing the greater tuberosity?
Boileau's data make the stakes clear. Types 1–2 (intracapsular, impacted) yield reliable results — 73% adjusted Constant score, mean elevation 123°, 81% satisfaction. Types 3–4 (extracapsular, disimpacted) lock patients into a "limited goals" conversation before they ever go to the OR.
For Type 3 surgical neck nonunions specifically, the authors argue arthroplasty should be abandoned unless the humeral head has totally resorbed. Internal fixation with intramedullary bone grafting is the preferred first-line treatment.
For Types 3–4 where arthroplasty is unavoidable, counsel patients explicitly that elevation will likely remain below 90° and satisfaction rates will be substantially lower. This paper is also why we use modular, adaptable prostheses in post-traumatic cases. Varying inclination and offset to fit the distorted anatomy is what allows us to skip the GT osteotomy in borderline cases.