Orthonotes Logo
Orthonotes
by the.bonestories

Hertel Predictors — Proximal Humerus Ischemia

6 Views

Category: Trauma

Share Wiki QR Card Download Slides (.pptx)
Medial hinge disruption >2 mm, anatomic neck fracture, head-splitting → high AVN risk. Assists decision towards arthroplasty in ischemic patterns.
Published Feb 28, 2026 • Author: The Bone Stories ✅
🧠 Test Yourself with OrthoMind AI

10 AI-generated high-yield questions by our AI engine



Overview — Proximal Humerus Fractures & AVN Risk

Proximal humerus fractures are the third most common fracture in patients over 65 years of age (after hip and distal radius fractures), accounting for approximately 5–6% of all adult fractures. While the majority can be managed non-operatively or with simple fixation, a significant subset — particularly those involving the surgical neck with varus collapse, head-splitting patterns, and fracture-dislocations — carry a substantial risk of avascular necrosis (AVN) of the humeral head. In 2004, Christian Hertel published a landmark anatomical study identifying specific radiological predictors of humeral head ischaemia that guide the surgeon`s decision between head-preserving internal fixation and primary humeral head replacement (hemiarthroplasty or reverse shoulder arthroplasty). These `Hertel predictors` remain the standard clinical reference for AVN risk assessment in proximal humerus fractures.

  • Neer classification: the universal classification for proximal humerus fractures; based on four anatomical parts (humeral head/articular segment, greater tuberosity, lesser tuberosity, and humeral shaft) and the displacement of each part (>1 cm displacement or >45° angulation = `displaced part`); Neer one-part (undisplaced — 80% of all proximal humerus fractures — non-operative); two-part (displacement of one part — usually surgical neck or greater tuberosity); three-part (displacement of two parts); four-part (displacement of all four parts — highest AVN risk — primary arthroplasty is often recommended over fixation)
  • Why AVN risk matters for surgical decision-making: the primary blood supply to the humeral head is the anterior humeral circumflex artery (AHCA — specifically the ascending branch, which penetrates the bone at the lateral aspect of the bicipital groove — the `arcuate artery`); the posterior circumflex humeral artery (PCHA) via the posteromedial periosteal vessels also contributes; when the humeral head is displaced and the medial metaphyseal periosteum is disrupted, the arcuate artery (AHCA) may be torn → ischaemia of the humeral head; if the surgeon performs internal fixation (plate/nail) and the head is ischaemic, the fixation will fail when the avascular head collapses → a second operation (arthroplasty) is required; it is better to predict AVN risk pre-operatively and select arthroplasty primarily rather than perform fixation that will fail
Hertel`s Anatomical Study — The Five Predictors

Hertel et al. (2004) performed a cadaveric study of 105 proximal humerus fractures, using intraoperative fluorescein dye injection and post-injection laser Doppler perfusion studies to assess humeral head vascularity. They identified five radiological features that predicted ischaemia of the humeral head and correlated these with the Lego classification of proximal humerus fractures (a systematic anatomical classification). The study identified which fracture characteristics most reliably predicted humeral head devascularisation.

Hertel Predictor Description Clinical Measurement / Threshold Risk Level
1. Short medial metaphyseal extension (<8 mm) The `medial metaphyseal extension` is the length of the medial calcar/periosteal hinge remaining attached to the humeral head fragment; the posteromedial periosteal vessels (posterior circumflex humeral artery branches) run along the medial side of the proximal humerus; if the calcar fragment attached to the humeral head is <8 mm long, the posteromedial vessels are disrupted → ischaemia; a longer calcar `tail` preserves the medial periosteal vascularity Measured on the lateral view or 3D CT as the distance from the anatomical neck to the level at which the medial calcar fragment separates from the head; <8 mm calcar `tail` = high ischaemia risk; ≥8 mm = medial periosteal vessels likely intact = lower risk THE SINGLE MOST POWERFUL PREDICTOR; in Hertel`s study, a medial hinge <8 mm was associated with 97% ischaemia rate in head-splitting fractures; the `calcar tail` (medial hinge) is the key vascular attachment; this is why valgus-impacted four-part fractures have LOWER AVN rates than varus four-part fractures — the medial calcar hinge is intact in valgus impaction
2. Disruption of the medial hinge The `medial hinge` is the medial periosteal connection between the humeral head and the shaft at the calcar region; a disrupted medial hinge = the posteromedial periosteal vessels are torn; an intact medial hinge (even in a displaced fracture) = the posteromedial vessels may still be patent; the medial hinge disruption is assessed on CT (loss of bony continuity medially) or at surgery (absence of medial periosteal attachment) On CT: complete loss of medial cortical continuity + no calcar fragment = disrupted medial hinge; on plain X-ray: complete medial displacement of the humeral head without medial bony contact; clinically assessed at surgery by feeling the medial soft tissue tension during reduction HIGH ischaemia risk; a disrupted medial hinge combined with a short medial metaphyseal extension = the two most powerful predictors combined; if both are present → very high probability of AVN with fixation → strong indication for primary arthroplasty
3. Anatomical neck fracture (head-splitting pattern) A fracture at or through the anatomical neck (the waist of the humeral head) — rather than at the surgical neck (below the tuberosities); an anatomical neck fracture detaches the entire articular head fragment from the shaft; the arcuate artery (AHCA) enters the bone at the lateral edge of the bicipital groove — an anatomical neck fracture disrupts this entry point for the dominant blood supply to the humeral head Plain X-ray or CT: the fracture line passes through (or immediately below) the articular margin of the humeral head — separating the `ball` from the `neck`; anatomical neck fractures are distinct from surgical neck fractures (which occur below the tuberosities); anatomical neck fractures are rarer but carry the highest individual risk of complete devascularisation Very high ischaemia risk — the entire humeral head articular segment is a separate fragment without the AHCA; primary arthroplasty is generally preferred for displaced anatomical neck fractures; internal fixation is rarely successful
4. Significant humeral head displacement (>4 mm medial displacement) The degree of medial displacement of the humeral head fragment from the shaft; medial displacement disrupts the posteromedial periosteal vessels and the medial hinge; varus collapse of the humeral head (medial displacement with inferior tilting) is associated with high AVN rates in fixed constructs Measured on the AP X-ray or CT as the distance of medial head displacement from the shaft axis; >4 mm medial displacement = significant disruption of the medial vascular structures; combined with a short calcar tail, this is a high-risk combination Moderate-high ischaemia risk in isolation; combined with calcar <8 mm = very high risk
5. Fracture-dislocation (head dislocated from the glenoid) If the proximal humerus fracture is associated with a dislocation of the humeral head from the glenoid (fracture-dislocation), ALL remaining soft tissue attachments to the head are disrupted by the dislocation; the anterior and posterior circumflex vessels are torn; there is no remaining vascularity Radiological evidence of humeral head dislocation from the glenoid in the presence of a proximal humerus fracture; four-part fracture-dislocation (the classic high-risk pattern); the dislocation confirms complete periosteal disruption Very high ischaemia risk; four-part fracture-dislocation = primary arthroplasty in the elderly; young patients (<50 years) — urgent anatomical reduction and fixation within 6 hours to minimise AVN even though the risk is high (`nothing to lose` by attempting head preservation in the young)
Combined Risk Assessment
  • Hertel`s key finding — the combination rule: in Hertel`s original study, when BOTH the medial metaphyseal extension was <8 mm AND the anatomical neck fracture was present, the positive predictive value for ischaemia was 97%; when only one predictor was present, the risk was lower; when neither was present, ischaemia was rare; the combination of short calcar tail + anatomical neck fracture = near-certain humeral head ischaemia; clinical decision: if both are present → primary arthroplasty; if one or neither present → fixation is likely viable
  • The valgus-impacted four-part fracture exception: valgus-impacted four-part fractures (the medial calcar remains attached to the head and the head is impacted in valgus — `tilted up`) have a LOWER AVN risk than varus four-part fractures; this is because the medial periosteal hinge (and the posteromedial vessels) is usually INTACT in valgus impaction — the head has been tilted superiorly, preserving the medial soft tissue attachment; these fractures can often be fixed with a locking plate with good outcomes; Neer specifically noted this exception in his classification refinements
  • Age-based decision-making: in older patients (>65–70 years), even moderate AVN risk may favour primary arthroplasty (hemiarthroplasty or reverse shoulder arthroplasty) — the outcome of failed fixation with secondary AVN and collapse is worse than primary arthroplasty; in younger patients (<50 years), every attempt should be made to preserve the humeral head (fixation even with high AVN risk) because arthroplasty in young patients has poor longevity; between 50–65 years, the decision requires individualisation based on bone quality, fracture pattern, and patient factors
Surgical Options
Option Indication Key Considerations
Locking plate ORIF (PHILOS plate) Displaced 2/3/4-part fractures with low Hertel risk (long calcar tail, intact medial hinge, no anatomical neck component, no dislocation); young patients (<60 years); good bone quality; valgus-impacted 4-part fractures Locking screws provide angular stability even in osteoporotic bone; calcar screw (inferomedial screw) is critical to prevent varus collapse; avoid lateral impingement by plate position (plate tip should be 5 mm below the greater tuberosity and not impinge the rotator cuff); complications: subacromial impingement, screw cut-out, AVN, non-union
Intramedullary nail Displaced surgical neck fractures (2-part); can be used for 3-part with appropriate technique; limited for comminuted 4-part; avoids the deltoid/rotator cuff split of the plate approach Less soft tissue disruption than plate; entry through the rotator cuff (damage to supraspinatus footprint is a concern); less lateral impingement than plate; limited fixation of the tuberosities; not suitable for comminuted 4-part
Hemiarthroplasty Elderly (>65–70) with 3/4-part fractures with high Hertel risk; anatomical neck fractures; 4-part fracture-dislocation; adequate rotator cuff; primary for selected patients over arthroplasty Pain relief reliable; function depends on tuberosity healing (the most critical technical step — tuberosities must be secured to restore rotator cuff function); if tuberosities fail to heal → poor function; results highly surgeon-dependent; increasingly being replaced by RSA in elderly
Reverse shoulder arthroplasty (RSA) Elderly (>70 years) with comminuted 3/4-part fractures with high AVN risk AND poor rotator cuff; also for failed hemiarthroplasty; tuberosity reconstruction less critical (RSA works without the cuff) RSA provides reliable function even if tuberosities fail to heal (the deltoid drives the reverse mechanism); better functional outcomes than hemiarthroplasty in elderly with comminuted fractures and poor cuff; increasing evidence and adoption for acute fractures in the elderly; complications: instability, nerve injury, notching
Exam Pearls
  • Hertel`s 5 predictors of humeral head ischaemia: (1) medial metaphyseal extension <8 mm (most powerful); (2) disrupted medial hinge; (3) anatomical neck fracture; (4) >4 mm medial displacement; (5) fracture-dislocation; both (1) and (3) together = 97% ischaemia rate → primary arthroplasty
  • Blood supply to the humeral head: anterior humeral circumflex artery (AHCA) via the arcuate artery (enters at the lateral bicipital groove); posterior humeral circumflex artery (PCHA) via posteromedial periosteal vessels; the medial calcar `tail` protects the PCHA contribution
  • Calcar tail (<8 mm = high risk): the medial metaphyseal extension length; measures the intact medial periosteal hinge; <8 mm = posteromedial periosteal vessels disrupted; ≥8 mm = medial vascularity likely preserved; the single most important Hertel predictor
  • Valgus-impacted 4-part exception: medial hinge intact in valgus impaction → lower AVN risk than varus 4-part → can attempt ORIF with plate; varus 4-part fractures have disrupted medial hinge → high AVN risk → primary arthroplasty in elderly
  • Age-based strategy: <50 years = always attempt head preservation (fixation); >70 years with high Hertel risk = primary RSA or hemiarthroplasty; 50–70 years = individualise; PROFHER trial (2015) showed no significant difference between surgery and non-operative treatment for displaced 2/3-part proximal humerus fractures in elderly — non-operative is a valid option for many patients
  • PROFHER trial (UK RCT, 2015): no significant difference in Oxford Shoulder Score between surgery (ORIF or arthroplasty) and non-operative treatment for displaced proximal humerus fractures in patients over 60 years; challenged the routine surgical management of these fractures; non-operative treatment (collar and cuff, early physiotherapy) is appropriate for most elderly patients with 2/3-part fractures
  • Calcar screw in locking plate: the inferomedial locking screw (directed into the posteromedial calcar of the humeral head) is the most important screw in the PHILOS locking plate; prevents varus collapse; must have calcar purchase; its absence or poor placement is the most common technical error leading to varus malunion and screw cut-out
🧠 Test Yourself with OrthoMind AI

10 AI-generated high-yield questions by our AI engine

References

Hertel R et al. Predictors of humeral head ischaemia after intracapsular fracture of the proximal humerus. J Shoulder Elbow Surg. 2004;13(4):427–433.
Neer CS. Four-segment classification of proximal humeral fractures — purpose and reliable use. J Shoulder Elbow Surg. 2002.
Bastian JD, Hertel R. Initial post-fracture humeral head ischaemia does not predict development of necrosis. J Shoulder Elbow Surg. 2008.
PROFHER Trial Collaborators. Surgical vs non-surgical treatment of adults with displaced fractures of the proximal humerus — the PROFHER randomised clinical trial. BMJ. 2015.
Solberg BD et al. Locked plating of 3 and 4 part proximal humerus fractures in older patients — the effect of initial fracture pattern on outcome. J Orthop Trauma. 2009.
Boileau P et al. Tuberosity malposition and migration — reasons for poor outcomes after hemiarthroplasty for displaced fractures of the proximal humerus. J Shoulder Elbow Surg. 2002.
Campbells Operative Orthopaedics. 14th Edition. Elsevier.
Orthobullets — Proximal Humerus Fractures; Hertel Predictors; Neer Classification; PHILOS Plate; RSA for Fracture.