J Shoulder Elbow Surg (2011) 20, 827-832

www.elsevier.com/locate/ymse

REVIEW ARTICLES
At the International Congress of Shoulder and Elbow Surgery (ICSES) in Edinburgh, Scotland, in September 2010, Dr Herbert Resch presented his talk
entitled ‘‘Proximal Humeral Fractures: Current Controversies’’ as the prestigious Codman Lecture. We are honored to be able to reproduce Dr Resch’s talk in
the Journal of Shoulder and Elbow Surgery - Bill Mallon, MD, Editor-in-Chief.

Proximal humeral fractures: current controversies

Herbert Resch, MD*

Department of Traumatology and Sports Injuries, University Hospital SalzburgeParacelsus Medical University Salzburg,
Salzburg, Austria

Epidemiology implants with other implants without any information on

either the fracture pattern or the quality of reduction.
According to Horak and Nilsson,13 5% of all fractures of In case of a humeral head fracture, the following issues
the human body are fractures of the proximal humerus. are of interest:
Palvanen et al17 reported an increase in these fractures of 1. Classification
more than 3-fold between 1970 and 2002. Court-Brown 2. Reduction
et al7 found that 70% of all 3- and 4-part fractures are 3. Vascularity
seen in patients aged over 60 years and 50% in patients 4. Implant characteristics
aged over 70 years. These results indicate that poor bone 5. Bone quality
quality or even advanced osteoporosis will be found in the
majority of patients with humeral head fractures. Classification
Conservative treatment of displaced fractures has not
shown consistently satisfactory results.2,6,9,25 Reconstruc- In 1993, Siebenrock and Gerber22 and Sidor et al21 found
tive surgery with locked plating has shown good results in very low interobserver reliability for the existing and
younger patients but was accompanied by a high complica- commonly used classification systems. They concluded that
tion rate in older patients with poor bone quality.23,24,26 it is not valid to compare classified studies from different
Hemiarthroplasty was seen as the treatment of choice for centers. According to these authors, the low reliability is
a long time, but it is associated with a high rate of malunion caused by several factors: first, the amount of displacement
of the tuberosities, which is responsible for poor functional measured in millimeters or degrees; second, a slight change
outcome. The malunion rate has been reported to be even of arm position causes a large change in the radiologic
worse in older patients.14 To date, new prosthetics specially appearance; and third, the use of illustration on just 1 plane
designed for fracture care have not improved the healing rate instead of 2 planes. Therefore, a new classification system
of the tuberosities.4,14,15,23,24 From the literature, it can be should be characterized by 3 features:
summarized that better outcome in terms of function can be
expected with reconstructive surgery than with prosthetic 1. It should be easy to understand.
replacement, despite the high complication rates of both 2. It should include the second plane.
procedures.23,24,26 Unfortunately, most articles compare 3. It should include accepted findings of recent years, such
as varus/valgus deformity23,24 and length and
*Reprint requests: Herbert Resch, MD, Department of Traumatology displacement of the medial hinge.11,12,18-20
and Sports Injuries, LandesklinikeneUniversity Hospital Salzburg,
Muellner-Hauptstrasse 48, 5020 Salzburg, Austria.
All of these factors are possible only for a purely
E-mail address: [email protected] descriptive classification system. The so-called Lego system

1058-2746/$ - see front matter Ó 2011 Journal of Shoulder and Elbow Surgery Board of Trustees.
doi:10.1016/j.jse.2011.01.009
828 H. Resch

of Hertel et al11,12 corresponds to the 4-part system of patients. In comparison to varus deformity, valgus deformity
Codman with the option of 12 possible fracture types. This is better tolerated. According to our own experience,
system fulfils almost all criteria but does not differentiate displacement of the greater tuberosity of more than 5 mm in
between varus- and valgus-type fractures, a distinction that is any direction should not be accepted. For the achievement
crucial for reduction and fixation. In varus-type fractures, the of good reduction, knowledge of the fracture type is
head is disrupted from the shaft and remains in the varus important because this provides information on the
position as a result of the persisting attachment of the rotator preserved periosteum. Varus-impacted fractures are char-
cuff muscles. In the case of an additional fracture of the acterized by residual primary stability, as a result of the
greater tuberosity, the head may follow the subscapularis periosteum still being preserved on the lateral side. The
muscle and rotate into an internally rotated position (3-part calcar on the medial side has to be reduced, which can
fracture according to Neer16). An investigation of 200 usually be achieved just by traction and manipulation of the
consecutive cases showed that 2 varus types could be arm. In contrast to the impaction type, the varus disruption
differentiated. type with additional fracture of the greater tuberosity pres-
ents quite often with the head in an internally rotated
Varus disruption type position (3-part fracture according to Neer16). Reduction of
this fracture type can only be achieved by a step-by-step
The varus disruption type is characterized by complete procedure. At first, the shaft has to be brought into align-
avulsion of the head from the shaft. The shaft is separated ment with the head, and then the head has to be derotated by
from the head in an anteromedial position (Fig. 1). pulling on the lesser tuberosity with a hooked instrument. At
the moment when alignment and derotation are achieved,
Varus impaction type either temporarily or permanently, Humerusblock K-wires
(Synthes, Bettlach, Switzerland) are introduced through the
The varus impaction type is characterized by impaction of shaft into the head. As the last step, the greater tuberosity is
the head on the medial side whereas no disruption occurred pulled downward by means of a hooked instrument and
on the lateral side. In the sagittal plane, the anterior angu- fixed with cannulated screws. All of the maneuvers are
lation angle is increased, but in contrast to the disruption performed percutaneously (but even with an open proce-
type, the shaft is not in a separated position (Fig. 2). dure, the various steps remain the same).
Valgus-type fractures are characterized by the impaction of Valgus-type fractures without lateral displacement are
the head into the metaphysis of the shaft. The fractured easy to reduce, because only the head has to be raised with
tuberosities remain in the normal longitudinal position and are an elevator that is introduced between the fractured tuber-
still attached to the shaft by the undisrupted periosteum. osities. The periosteum on the medial side serves as
Again, the 2 types could be differentiated by factors such as the a mechanical hinge when performing this maneuver. In the
presence/absence of lateral displacement of the head (Fig. 3). case of severe lateral displacement, the mechanical-hinge
Of the 200 investigated consecutive fractures, 43% had periosteum on the medial side is torn and the head fragment
a varus deformity and 31% had a valgus deformity; 25% is very unstable and difficult to reduce. By means of an
had a normal position (<20 displacement). Within the elevator, the hinge has to be reduced first, and then the head
varus group, 25% were of the varus disruption type and fragment is raised until alignment with the tuberosities is
18% were of the varus impaction type. On the basis of this achieved. K-wires (Humerusblock) that have been inserted
investigation, we developed the so-called HCTS classifi- previously are in the so-called waiting position. They can
cation system. H stands for head, C for the medial calcar, T be introduced into the head fragment at the moment when
for the tuberosities, and S for the shaft. Each region is reduction is achieved.
described separately, and all regions are finally assembled.
The system provides information on the expected vascu-
larity and the expected difficulties during reduction and Vascularity
fixation. The HCTS classification system will be published
in a separate article. Gerber et al10 stated that in the case of an existing avascular
necrosis, it is the deformity rather than necrosis that causes
disability. Therefore, the risk of limited blood supply of the
Reduction articular fragment does not influence our decision making
in terms of treatment. Like Gerber et al, we believe that the
Two questions have to be answered. One is what degree of alignment of the tuberosities is very important in cases in
displacement is tolerable, and the second is how reduction which prosthetic replacement might be necessary as
can be achieved. We know from our own experience and a secondary procedure because of head necrosis. In young
from that of the study of Solberg et al23,24 that varus patients who have sustained a 4-part fracture-dislocation, in
deformity of more than 20 should not be left uncorrected, which the head is completely separated, a bone block
because this level of deformity is not well tolerated by procedure for additional blood supply is performed. The
Proximal humeral fractures 829

Figure 1 Varus disruption fracture. (A) Coronal plane. The head fragment is in a varus position and completely separated from the shaft.
(B) Sagittal plane. The shaft fragment is in an anterior position to the head.

Figure 2 Varus impaction fracture. (A) Coronal plane. The head fragment is in a varus position with impaction into the shaft on the
medial side. There is no distance between fragments on the lateral side. (B) Sagittal plane. There is increased anterior apex angulation
between the head and shaft but no separation.

bone block is harvested from the anterior part of the Semi-rigidity

acromion together with the attached muscle pedicle of
the deltoid muscle. From a previous anatomic study, we In porotic bone, using rigid implants such as locked plates
know of an artery that is found in this part of the muscle will destroy the soft bone as the load is transferred from
supplying the end of the acromion. The bone block is the stronger bone of the shaft to the weak bone of the head
inserted right below the central part of the articular by the plate and the angle-stable screws. Semi-rigid
segment. This technique has not been published so far, but implants, such as the K-wires provided with the Humer-
early results are encouraging. usblock implant, show better load distribution at the metal-
bone interface.

Implant
Controlled impaction
For the treatment of fractures with osteoporotic bone
conditions, 2 features seem to be relevant for an implant: According to the studies of Niederberger (A. Niederberger,
semi-rigidity and controlled impaction. personal communication, 2010), who measured the sintering
830 H. Resch

Figure 3 Valgus impaction fractures with high fracture level on medial side. (A) No lateral displacement between head and shaft. (B)
Lateral displacement of head in relation to shaft.

Figure 4 Humerusblock implant. (A) Valgus impaction fracture. (B) Postoperative radiograph immediately after surgery. The K-wires
show perfectly the direction of the load peaks described by Bergmann et al.1

effect of a fractured humeral head fixed with the Humerus- From these 2 studies, we can conclude that in a fractured
block implant in 66 cases, this effect was found in all cases. proximal humerus, the head has a strong tendency toward
A sintering effect of, on average, 5.2 mm ( 4.89 mm) was impaction during the first weeks. To permit the sintering
seen and was significantly correlated to the age of the effect, the implant should be inserted in the direction of the
patients. The sintering effect was also found by Gardner load peaks measured by Bergmann et al.
et al8 in fractures treated with locked plates. Bergmann et al1
published a report on the direction of load peaks entering the
humeral head measured in an in vivo model. According to Humerusblock (Synthes)
their studies, the load peaks enter the head from a superior-
medial direction in the frontal plane and from a superior- The key features of the Humerusblock implant are two
posterior direction in the sagittal plane within a very small 2.5-mm K-wires that are fixed in a cylindrical device. The
range (17 in the frontal plane and 9 in the sagittal plane). 2 K-wires are introduced through the cylindrical device and
Proximal humeral fractures 831

through the cortical bone of the shaft into the humeral head. 4. ‘‘Intelligent’’ K-wires: K-wires are characterized by the
The K-wires, which are inserted in a diverted direction in advantage that they can be introduced into the head up
the sagittal plane, show perfectly the direction of the load to the subchondral bone, where the best bone quality is
peaks described by Bergmann et al1 (Fig. 4). found. On the other hand, K-wires that are fixed in the
cylindrical Humerusblock device will perforate the
Results cartilage when sintering occurs. This perforation may
require another intervention for withdrawal of the
Bogner et al3 published the results of 48 patients with 3- K-wires. Therefore, K-wires migrating together with
and 4-part fractures treated by percutaneous reduction and the sintering head (so-called intelligent K-wires) are
fixation with the Humerusblock implant. All patients were desirable. The concept will be that, based on expected
aged over 70 years, with a mean age of 79 years. Reduction resistance measured by quantitative computed tomog-
was assessed and compared with the radiologic result after raphy scanning, the tips of the K-wires will provide
consolidation. Postoperatively, reduction was assessed as a certain resistance but will not perforate.
good in 39 of the 48 patients and fair in 9 patients. At
consolidation, 35 were assessed as good, 11 as fair, and 2 as
poor. In other words, only 8% of all cases changed from
one group to another. One of the disadvantages of the Disclaimer
Humerusblock implant is that K-wire perforation through
the head requires removal of the implant. In 25% of all The authors, their immediate families, and any research
patients, the K-wires had to be withdrawn but not removed foundations with which they are affiliated have not
before consolidation. received any financial payments or other benefits from
any commercial entity related to the subject of this article.

Bone quality
References
The biomechanical perforation testing of Niederberger
showed that bone mineral density measured by quantitative 1. Bergmann G, Graichen F, Bender A, K€a€ab M, Rohlmann A,
computed tomography correlates positively with the resis- Westerhoff P. In vivo glenohumeral contact forcesdmeasurements in
tance and the failure load. The central part of the head the first patient 7 months postoperatively. J Biomech 2007;40:2139-49.
doi:10.1016/j.jbiomech.2006.10.037
showed the highest resistance against perforation, whereas
2. Bigliani LU, McCluskey GM III. Prosthetic replacement in acute
the lowest was found in the superior-anterior region. fractures of the proximal humerus. Semin Arthroplasty 1990;1:
129-37.
3. Bogner R, Hubner C, Matis N, Auffarth A, Lederer S, Resch H.
Design of future implants Minimally-invasive treatment of three- and four-part fractures of the
proximal humerus in elderly patients. J Bone Joint Surg Br 2008;90:
In my opinion, future implants should be characterized by 4 1602-7. doi:10.1302/0301-620X.90B12.20269
4. Boileau P, Krishnan SG, Tinsi L, Walch G, Coste JS, Mole D.
features: Tuberosity malposition and migration: Reasons for poor outcomes
1. Controlled impaction: As shown by Niederberger and after hemiarthroplasty for displaced fractures of the proximal
humerus. J Shoulder Elbow Surg 2002;11:401-12. doi:10.1067/mse.
Gardner et al,8 the humeral head has a strong tendency 2002.124527
toward impaction. In porotic bone conditions, an 5. Brunner A, Honigmann P, Treumann T, Babst R. The impact of stereo-
implant that does not allow impaction will cause a high visualisation of three-dimensional CT datasets on the inter- and
rate of failure, such as perforation of screws or intraobserver reliability of the AO/OTA and Neer classifications in the
secondary varus displacement.5,23,24,26 In fractures with assessment of fractures of the proximal humerus. J Bone Joint Surg Br
2009;91:766-71. doi:10.1302/0301-620X.91B6.22109
poor bone quality, the implant should be a guiding tool 6. Clifford PC. Fractures of the neck of the humerus: A review of the late
rather than a rigid fixation device, allowing impaction results. Injury 1980;12:91-5.
of the head. 7. Court-Brown CM, Garg A, McQueen MM. The epidemiology of
2. Direction of peak forces: On the basis of the studies of proximal humeral fractures. Acta Orthop Scand 2001;72:365-71.
Bergmann et al1 showing that the peak forces enter the 8. Gardner MJ, Weil Y, Barker JU, Kelly BT, Helfet DL, Lorich DG. The
importance of medial support in locked plating of proximal humerus
head from the superior-medial and superior-posterior fractures. J Orthop Trauma 2007;21:185-91.
directions within a very small angle, the implant 9. Geneste R, Durandeau A, Gauzere JM, Roy J. The treatment of
should be introduced in the same direction, allowing fracture-dislocation of the humeral head by blind pinning [in French].
controlled gliding of the porotic head. Rev Chir Orthop Reparatrice Appar Mot 1980;66:383-6.
3. Semi-rigid implant: To reduce the stress on the bone- 10. Gerber C, Hersche O, Berberat C. The clinical relevance of post-
traumatic avascular necrosis of the humeral head. J Shoulder Elbow
metal interface, the implant should not be rigid. Surg 1998;7:586-90.
K-wires with an elastic 3-point fixation system fulfill 11. Hertel R. Fractures of the proximal humerus in osteoporotic bone. Osteo-
these requirements better than angle-stable plates. poros Int 2005;16(Suppl 2):S65-72. doi:10.1007/s00198-004-01714-2
832 H. Resch

12. Hertel R, Hempfing A, Stiehler M, Leunig M. Predictors of humeral 20. Resch H, Povacz P, Frohlich R, Wambacher M. Percutaneous fixation
head ischemia after intracapsular fracture of the proximal humerus. of three- and four-part fractures of the proximal humerus. J Bone Joint
J Shoulder Elbow Surg 2004;13:427-33. doi:10.1016/j.jse.2004.01.034 Surg Br 1997;79:295-300.
13. Horak J, Nilsson BE. Epidemiology of fracture of the upper end of the 21. Sidor ML, Zuckerman JD, Lyon T, Koval K, Cuomo F, Schoenberg N.
humerus. Clin Orthop Relat Res 1975:250-3. The Neer classification system for proximal humeral fractures. An
14. Kralinger F, Schwaiger R, Wambacher M, Farrell E, Menth-Chiari W, assessment of interobserver reliability and intraobserver reproduc-
Lajtai G, et al. Outcome after primary hemiarthroplasty for fracture of ibility. J Bone Joint Surg Am 1993;75:1745-50.
the head of the humerus. A retrospective multicentre study of 167 22. Siebenrock KA, Gerber C. The reproducibility of classification of
patients. J Bone Joint Surg Br 2004;86:217-9. doi:10.1302/0301- fractures of the proximal end of the humerus. J Bone Joint Surg Am
620X.86B2.14553 1993;75:1751-5.
15. Loew M, Heitkemper S, Parsch D, Schneider S, Rickert M. Influence of 23. Solberg BD, Moon CN, Franco DP, Paiement GD. Surgical treatment
the design of the prosthesis on the outcome after hemiarthroplasty of the of three and four-part proximal humeral fractures. J Bone Joint Surg
shoulder in displaced fractures of the head of the humerus. J Bone Joint Am 2009;91:1689-97. doi:10.2106/JBJS.H.00133
Surg Br 2006;88:345-50. doi:10.1302/0301-620X.88B3.16909 24. Solberg BD, Moon CN, Franco DP, Paiement GD. Locked plating of
16. Neer CS. Displaced proximal humeral fractures. I. Classification and 3- and 4-part proximal humerus fractures in older patients: The effect
evaluation. J Bone Joint Surg Am 1970;52:1077-89. of initial fracture pattern on outcome. J Orthop Trauma 2009;23:
17. Palvanen M, Kannus P, Niemi S, Parkkari J. Update in the epidemi- 113-9.
ology of proximal humeral fractures. Clin Orthop Relat Res 2006;442: 25. Stableforth PG. Four-part fractures of the neck of the humerus. J Bone
87-92. doi:10.1097/01.blo.0000194672.79634.78 Joint Surg Br 1984;66:104-8.
18. Resch H, Beck E, Bayley I. Reconstruction of the valgus-impacted 26. Sudkamp N, Bayer J, Hepp P, Voigt C, Oestern H, Kaab M, et al. Open
humeral head fracture. J Shoulder Elbow Surg 1995;4:73-80. reduction and internal fixation of proximal humeral fractures with use
19. Resch H, Hubner C, Schwaiger R. Minimally invasive reduction and of the locking proximal humerus plate. Results of a prospective,
osteosynthesis of articular fractures of the humeral head. Injury 2001; multicenter, observational study. J Bone Joint Surg Am 2009;91:1320-
32(Suppl 1):SA25-32. 8. doi:10.2106/JBJS.H.00006