The Telegraph nail for proximal humeral fractures: A

prospective four-year study

Christian Cuny, MD,a Marius M. Scarlat, MD,b Mbarek Irrazi, MD,a Patrick Beau, MD,a Valerie Wenger, MD,c
Nicolae Ionescu, MD,a and Aboubekr Berrichi, MD,a Metz and Toulon, France

Proximal humeral fractures in 67 patients older than 50

years treated with the Telegraph nail (FH Orthopedics,
Heimsbrunn, France) were monitored for 4 years to
assess the fracture pattern (weighted Constant score),
ranges of motion, and patient satisfaction. The outcome
was best in patients with extraarticular surgical neck
fractures (mean weighted Constant score, 93.5%);
scores were 85% and 77.5%, respectively, for valgus
impacted fractures and intraarticular displaced or
dislocated fractures. Some or all of the hardware was
removed in 21 patients (31%). Two required implant
removed for mechanical problems related to screw
positioning or migration; 8 were revised because
proximal migration of the implant resulted in subacromial
impingement. Avascular necrosis occurred in 18% of
valgus impacted fractures and in 37.5% of displaced
articular or dislocated fractures. Secondary migration of
the tuberosities occurred in 6 (all 4-part fractures). The
Telegraph nail provides a reproducible and satisfactory
outcome for surgical neck and valgus impacted fractures
in older patients. The outcome was less satisfactory for
unstable articular or dislocated fractures. (J Shoulder
Elbow Surg 2008;17:539-545.)

he treatment of proximal humeral fractures is

challenging. There is no single surgical procedure
that can be successfully used in all cases, and even
the classification of these fractures is controversial.2,10,14,17,19 Accepted surgical treatment includes
various techniques of open reduction and internal fixation, percutaneous pinning, percutaneous or open
From the Departments of aOrthopaedic Surgery and Traumatology
pital
and cMedical Statistics, Centre Hospitalier Regional MetzHo
Bon-Secours, Metz; and bDepartment of Orthopaedic Surgery
and the Shoulder Service, Clinique Chirurgicale Saint Michel,
Toulon.
pital BonCorrespondence: Christian Cuny, MD, C.H.R. MetzHo
Secours, Department of Orthopaedic Surgery and Traumatology,
1 place Philippe de Vigneulles, 57038 Metz Cedex, France
(E-mail: c.cuny@chr-metz-thionville.fr).
Copyright 2008 by Journal of Shoulder and Elbow Surgery
Board of Trustees.
1058-2746/2008/$34.00
doi:10.1016/j.jse.2008.02.004

nailing, and hemiarthroplasty or total shoulder replacement.2,15,18,20-22 Treatment of comminuted fractures,

especially in older patients with osteoporotic bone or
compromised vascular supply to the humeral head, or
both, is particularly challenging, and secondary complications related to bone fragility or implant instability,
or both develop in a significant number of patients.4,8
This article reports a study of a novel implant, the
Telegraph nail5 (FH Orthopedics, Heimsbrunn,
France), developed in 1998 for internal fixation of
proximal humeral fractures (Figure 1). The purpose
of the study was to validate this less invasive fixation
for proximal humerus fractures.
MATERIALS AND METHODS
Implant
The Telegraph nail is 15 cm long, straight, and available
in 3 widths of 7, 8, or 9 mm. It provides fixation within the
humeral head with proximal locking screws that are secured
within the nail, thereby stabilizing the fracture and implant
constructs. This was the first intramedullary device to use
locked screws for humeral head fixation.5 The nail is inserted
through the tendinous portion of the rotator cuff, avoiding the
rotator cuff insertion on the greater tuberosity. Proximal locking is done within the tuberosities. Three self-stabilized
screws allow fixation of the different fragments, including
the humeral head. Optional distal interlocking is performed
at the level of the deltoid tuberosity with 1 or 2 screws.

Surgical technique
Three different techniques were developed, each being
adapted to the anatomic complexity of the fractures: (1)
the minimally invasive (percutaneous) approach, (2) the
standard approach, and (3) the cup-and-ball approach. Operations with all techniques were done with the patients in
the beach chair position and with intraoperative fluoroscopic control.

The minimally invasive technique

The minimally invasive (percutaneous) technique was
used for stabilizing extraarticular surgical neck fractures
and associated surgical neck and greater tuberosity (3-part)
fractures. A prerequisite for the use of this percutaneous technique was the ability to achieve adequate reduction of the
fracture by either external manipulation or percutaneously

539

540

Cuny et al

J Shoulder Elbow Surg

July/August 2008

was placed for stabilizing the lesser tuberosity by using the

ancillary handle from the interlocking jig. Distal locking
was performed, when needed, near the deltoid tuberosity.
With more complex fractures, additional reduction maneuvers were required. The humeral head was reduced
with an elevator and temporarily held with a Kirschner
wire placed between the head and the glenoid (Figure 3).
The balance of the procedure was identical to the previous
description.

The cup-and-ball technique

Figure 1 A, Radiograph and (B) drawing show the humeral fracture repaired with a Telegraph nail. The commonly used assembly
has 2 frontal screws.

with an elevator or pin used as a joystick. A 15- to 20-mm incision was made in the skin over the front of the acromion. A
small incision was made in the rotator cuff, 10 to 15 mm
medial to its insertion, directly over the central axis of the medullary canal of the humerus. An awl and then a reamer were
introduced at the top of the articular surface of the head. With
the fracture reduced, the nail with the attached jig was then
inserted. A C-arm was used to check the reduction of the
fracture and the position and height of the nail. The proximal
interlocking screws were inserted percutaneously using a softtissue sleeve. Distal locking was optional.

The standard technique

This approach was used for irreducible 3-part fractures
and for 3- or 4- part valgus impacted fractures (Figure 2).
An 8- to 10-cm-long anterolateral incision was made over
the anterolateral corner of the acromion, following the direction of the deltoid fibers, with one-third of the incision proximal and two-thirds distal to the acromion. An interval was
developed between the anterior and middle portions of the
deltoid, taking care to detach a periosteal sleeve of the anterior acromion. The coracoacromial ligament was retracted
anteriorly. The fracture fragments, including the tuberosities,
were carefully mobilized and reduced with their attached rotator cuff insertions. Uncommonly, we found the rotator cuff
to be intact. In these patients we opened the rotator interval
to mobilize the tuberosity fragments.
First, the humeral head was reduced, followed by the tuberosities. The desired nail entry point was marked at the top
of the humeral head in the articular area. The entry hole was
created with an awl. The nail was inserted with its jig. The 2
lateral-to-medial proximal interlocking screws were used to
stabilize the greater tuberosity and the humeral head within
the nail. When needed, the anterior-posterior (sagittal) screw

This variety was used for highly displaced, comminuted

fractures with 3 or 4 fragments or dislocation, or both
(Figure 4). The surgical approach was the same as described for the standard technique. The nail was inserted
and first locked distally (Figure 5, B) using 1 or 2 screws.
The jig was removed (Figure 5, C). The humeral head was
then reduced with a bone hook or an elevator and positioned directly over the free proximal part of the nail. The
head was fixed (Figure 5, D) directly to the nail using 1 or
2 screws. The final step involved reduction and fixation of
the tuberosities with intraosseous bone sutures (Figure 5,
E), which were used in 10 patients (8%). Any remaining defects in the cuff were repaired.

Patients
The present series consisted of 67 patients (50 women
and 16 men) aged older than 50 years (average age,
72.1; range, 50-92 years) who were part of a larger prospective analysis of patients who had surgery between
1998 and 2000. The complete prospective series included
122 patients of all ages treated for proximal humeral fractures using this implant. The right shoulder was involved in
54%, and the mechanism of injury was low-energy trauma
in 94%. Associated fractures occurred in 13.8% of the cases
and included proximal femur and wrist fractures. Seventeen
patients were lost to follow-up and 27 died, leaving 78 patients who were followed up after surgery for an average
of 48 months (range, 42-55 months). Of these, 67 fractures
were in patients aged older than 50 years. Special attention
was given to the follow-up of the 27 patients (25 women)
older than 75 years because the outcome in this group
may be influenced by osteoporosis. In this subset of 27 of
the most elderly patients, 17 had extraarticular fractures, 4
had impacted articular injuries, and 4 had unstable fractures
or fracture-dislocations.
The rough Constant score3 was used to evaluate the results of treatment and was adjusted by age and sex. The radiologic results were analyzed from radiographs of the
shoulder using anteroposterior (AP) and lateral scapula
views.8,18,20 The bone healing and the anatomic features
of the repair were assessed by the inclination angle of the humeral head on the AP view (normal, 45 ) and on the lateral
view (normal, 30 ). Any migration of the tuberosities was
noted. The difference in height between the proximal tip of
the greater tuberosity and the outer articular border of the humeral head was measured. The presence of avascular necrosis (AVN) was noted by the loss of head sphericity. Screw
penetration was defined when the distance between the tip
of the implant and the subchondral bone was less than 2 mm.

Cuny et al

J Shoulder Elbow Surg

Volume 17, Number 4

541

Figure 2 A, Radiograph at admission shows a valgus impacted fracture with large posterior displacement in an 83year-old woman. B, First control radiograph after repair with Telegraph nail. C, Radiograph at 5-years follow-up
(Constant score, 54).

with 3 or 4 parts and with large displacement or dislocation

(C2 and C3). Data were collected and statistical correlations
were calculated using the Student t test and the c2 test. Statistical significance was considered for values of P < .05.

RESULTS
Functional results

Figure 3 Reduction of the head with an elevator and temporary fixation with a Kirschner wire inserted on the glenoid.

All fractures were classified using the Arbeitsgemeinschaft Fur Osteosynthesefragen (AO) system16 in 3 groups
of fractures (Table I). The extraarticular fractures, including
the 2- or 3-part surgical neck fractures with varying displacement, are represented by the A2, A3, Bl, and B2 varieties
and accounted for 40 patients (60%). The second group
included articular valgus impacted fractures12 (Cl) and
involved 11 patients (16%). The third group included
16 patients (24%) with complex intraarticular fractures

The average rough Constant score at the 4-year follow-up was 62 (standard deviation, 20). The average
sex- and age-weighted Constant score was 93.5% for
the extraarticular surgical neck fractures, 85% for the
impacted-valgus articular fractures, and 77.5% for the
disengaged or dislocated fractures, or both. The raw
data are reported in Table I. In the patients aged older
than 75 years, the average rough Constant score was
60 and the weighted Constant score was 89% (standard deviation, 20%). Three cases of AVN occurred
in this age group, all after intraarticular fractures.
Radiologic evaluation

The average inclination angle on the AP view was

46.5 6 11 (range, 12 -75 ; normal average,
45 ). The average inclination on the lateral view was
40 6 13 (range, 23 -80 ; normal average, 30 ).
The difference in height of the greater tuberosity to
the lateral border of the articular surface was 5 mm 6
3.4 mm (range, 15 to 9 mm).

542

Cuny et al

J Shoulder Elbow Surg

July/August 2008

Figure 4 A, Admission radiograph shows a complex disengaged fracture in 81-year-old woman. B, First control
radiograph after repair with the Telegraph nail. C, Radiograph at the 3-year follow-up (Constant score 79).

Complications

Three early revisions were necessary due to poor

initial reduction and position of the implant. There
was 1 infection, 3 cases of loss of fixation, and 2 cases
of reflex sympathetic dystrophy that resulted in shoulder stiffness.
Overall, some or all of the hardware was removed
in 21 patients. In 2 patients the implant was removed
for mechanical problems related to screw positioning
or migration, or both, and 8 were revised because of
proximal migration of the implant with screw breakage in 2 and screw joint penetration in 6. Six of these
8 cases were extraarticular fractures. Seven patients
had revision surgery for subacromial impingement
syndrome and secondary stiffness requiring decompression of the subacromial space.
The status of the tuberosities was carefully assessed
on the follow-up radiographs using the lateral
scapular view and the axillary view, when possible.
According to our data, 6 patients showed secondary
displacement with malunion of the tuberosities. The
greater tuberosity was involved in all these patients,
and the lesser tuberosity was also displaced in 1 patient. Malunion was associated with the worst results.
Long-term follow-up showed that AVN developed in
8 patients (11.9%), with deformation of the humeral
head (Table I). AVN developed in 2 of 11 patients
(18%) in the valgus impacted group (Cl) compared

with 6 of 16 (37.5%) in complex fractures with wide

displacement or dislocations, or both (C2/C3). A significant difference was noted in the incidence of AVN
(P .001) in the C2/C3 group compared with the
other groups. Shoulder function in the AVN cases
was poorer than in the overall group, with an average
rough Constant score of 42 (range, 27-64). The adjusted Constant score in patients with AVN averaged
57% (range, 37%-95%), and 13 had Constant scores
of 40 or lower (Table I). These failures did not show
a predictive factor except for the anatomic type of fracture. The poor outcome of patients with C2/C3 injuries was statistically significant (P .01).
DISCUSSION
This article describes our experience with a novel
intramedullary fixation system for different forms of
proximal humeral fractures (Table I). The results
obtained with this implant justify its continued use as
demonstrated by the functional outcome of our patients. The functional results were excellent for the
2-part or 3-part surgical neck fractures, which had
an average 93% weighted Constant score. This finding is promising, because these injuries are considered difficult to manage.4,18 In our experience, these
fractures are predictably repaired using the Telegraph
nail. Most of the cases were done with a less invasive

J Shoulder Elbow Surg

Volume 17, Number 4

Cuny et al

543

Figure 5 The cup-and-ball technique. A, Complex fracture. B, Distal locking without reduction. C, Removal of the
jig. D, Reduction and fixation of the head. E, Reduction and fixation with intraosseous sutures of tuberosities.

percutaneous technique, which may be of benefit

when treating elderly patients.
The average Constant score (85%) was also good
for the valgus-impacted articular fractures. In our experience, these fractures are usually repaired with the
standard open approach because reduction of the
greater tuberosity is typically necessary. The outcome
was worse for the unstable, intraarticular disengaged
or dislocated fractures, which resulted in AVN in
37.5% of patients and an average adjusted Constant
score of 77.5%. However, these functional results are
comparable with those of the hemiarthroplasty in several studies.12,13,17,21,22
The average adjusted Constant score of 89% in
patients aged older than 75 years demonstrates the
quality of the fixation with this implant, even in osteoporotic bone.
The complications encountered in this series were
mild in their consequences and, aside from hardware
removal, were few in number. Hardware removal was
necessary in 8 patients (12%), either because of

a prominent nail, which would cause subacromial

impingement, or due to screw penetration. We noted
that some nails became more prominent with time because of proximal migration owing to impaction or settling of surgical neck fractures. This could cause screw
breakage or intraarticular penetration of the tips of the
screws. In these 8 patients, we removed the screws
and the nail to avoid possible articular injury and posttraumatic arthritis.
From this experience, we recommend that the nail
be inserted at least 0.5 cm below the subchondral
bone of the humeral head. We also recommend checking the length of the proximal screws in the head to
avoid joint protrusion. Migration due to fracture impaction was controlled during the early part of this series by
avoiding distal interlocking screws and resulted in no
complications. The new generation of the implant has
a dynamic oblong distal hole where the distal screw
can move distally during impaction of the fracture.
That feature led to a decrease in the number of patients
who required hardware removal later in the series.

544

Cuny et al

J Shoulder Elbow Surg

July/August 2008

Table I Clinical outcome related to the fracture type

Constant score
Type and designationa
Extraarticular
A2, A3, Bl, B2
Articular impacted
C1
Articular disengaged and/or dislocated
C2, C3

Patients, No. (%)

AVN, No. (%)

40 (59.7)

Rough # 40, No. (%)

4 (8)

11 (16.4)

2 (18)

2 (18)

16 (23.9)

6 (37.5)

7 (43.7)

Weighted

93.5%
85%
77.5%

AVN, Avascular necrosis.

a
AO (Arbeitsgemeinschaft Fur Osteosynthesefragen) classification.

Seven of our patients experienced subacromial

impingement syndrome related to nails that were
placed too prominently or which migrated proximally
during healing. Although in all cases the nail must be
inserted through the rotator cuff, shoulder impingement was not seen in patients with well-positioned
nails. This is contrary to what could be expected after
antegrade nailing of the humerus. An explanation
was provided by the report of Gaullier et al.6 They
showed that the more medial the point where the
cuff is crossed, the less likely it is that symptomatic rotator cuff lesions will be created because the cuff is
well vascularized at this point. The healing potential
of the rotator cuff is greater medially, which is not
the case laterally where the cuff is more fibrous and
less vascular. Because the Telegraph nail is straight,
the entry point is medial compared with other intramedullary implants.
The cases of AVN were generally well tolerated,
with little or no pain and a typical forward elevation
between 80 and 90 . The patients with AVN were
able to resume activities of daily living. These
outcomes were similar to those described by other authors.1,7,8,9,11,17,21 The relatively low incidence of
AVN may also be explained by the mechanism of
creeping substitution described by Lee and Hansen.14
These authors report that interruption of the circulation
to the head in widely displaced fractures frequently occurs. However, a mechanism of contact revascularization occurs that can be visualized by the change in the
osseous structure in serial radiographs. This revascularization does not lead to a deformity or collapse of
the head. Lee and Hansen believe that the mechanism
of creeping substitution is enhanced by the integrity of
the rotator cuff and a solid osteosynthesis. The results
in this series, in association with eventual neovascularization of the humeral head, may suggest that the theoretic risk of AVN does not imply that patients with
severe proximal humeral fractures should be systematically treated with hemiarthroplasty in all cases.
The implant and the procedures that we have described may represent new opportunities in repairs

of shoulder trauma. Our study reports good results

with the Telegraph nail, even in some complex cases.
The different insertion techniques offer a surgical
option for diverse situations. The minimally invasive
technique was easily performed for treating surgical
neck fractures in elderly patients, which is the most
common type of fracture in everyday practice. With increased experience, this implant can be used for the
treatment of more complex fractures.
This study has some weaknesses must be noted.
First, the possible disadvantages of antegrade humeral nailing must be taken into account by the
surgeon. Because the nail unavoidably transfixes the
rotator cuff, there is a risk of rotator cuff damage. As
epidemiologic data show, the most frequent fractures
of the proximal humerus occur in elderly patients. Recent studies show that the natural history of the rotator
cuff evolves to degenerative tears and ruptures in more
than 56% of the asymptomatic subjects older than 75
years.19 However, our study presents objective data
that the shoulder function after osteosynthesis in this
category of patients is very good, probably related
to early mobilization, no matter the status of the rotator
cuff. That may suggest that less invasive surgery and
early mobilization is preferable to anatomic bone
and cuff repair in older patients, but this topic remains
controversial.
Second, the relatively high incidence of screw
penetration may complicate the clinical results, especially at the beginning of ones experience. This feature, however, is common to all interlocking nails,
and even locking proximal humeral plates, and it is
usually easily managed by early removal of the offending screws.
Third, there is no consensus about a universal
method for managing all types of fractures in the
proximal humerus, and this new implant is one more
on the list of the practical possibilities. We would
like to point out that this system was the first to use
a self-stabilizing system to prevent screw migration
and articular penetration, leading to a low number
of mechanical complications.

J Shoulder Elbow Surg

Volume 17, Number 4

Overall, the data presented in this study bring objective information about this type of osteosynthesis,
and we recommend the use of the Telegraph nail in
selected cases.
REFERENCES

1. Bell S, Gschwend N. Clinical experience with total arthroplasty of

the shoulder using the Neer prosthesis. Int Orthop 1986;10:
217-22.
2. Cofield RH. Comminuted fractures of the proximal humerus. Clin
Orthop 1988;230:49-57.
3. Constant CR, Murley AH. A clinical method of functional assessment of the shoulder. J Bone Joint Surg Am 1987;65:135-48.
4. Court-Brown CM, Garg A, McQueen MM. The translated twopart fracture of the proximal humerus: epidemiology and outcome
in the older patient. J Bone Joint Surg Br 2001;83:799-804.
5. Cuny C, Pfeffer F, Irrazi M, et al. Un nouveau clou verrouille pour
les fractures proximales de lhumerus [in French]. Rev Chir Orthop
2002;88:62-7.
6. Gaullier O, Rebai L, Dunaud JL, Moughabghab, Benaissa S.
Traitement des fractures recentes de la diaphyse humerale par enclouage centro-medullaire verrouille selon Seidel [in French]. Rev
Chir Orthop 1999;85:171-8.
7. Gerber C, Hersche O, Berberat C. The clinical relevance of posttraumatic avascular necrosis of the humeral head. J Shoulder
Elbow Surg 1998;7:586-90.
8. Gerber C, Werner CML, Vienne P. Internal fixation of complex
fractures of the proximal humerus. J Bone Joint Surg Br 2004;86:
848-55.
9. Hawkins RJ, Switlyk P. Acute prosthetic replacement for severe fractures of the proximal humerus. Clin Orthop 1993;289:156-60.
10. Hertel R, Hempfing A, Stiehler M, Leunig M. Predictors of humeral
head ischemia after intracapsular fracture of the proximal humerus.
J Shoulder Elbow Surg 2004;13:427-33.

Cuny et al

545

11. Huten D, Duparc J. Larthroplastie prothetique dans les traumatismes complexes recents et anciens de lepaule [in French]. Rev
Chir Orthop 1986;72:517-29.
12. Jacob RP, Miniaci A, Anson PS. Four part valgus impacted fractures
of the proximal humerus. J Bone Joint Surg Br 1991;73:295-8.
13. Kraulis J, Hunter G. The results of the prosthetic replacement in fractures-dislocations of the upper end of the humerus. Injury 1977;8:
129-31.
14. Lee CK, Hansen HR. Post-traumatic avascular necrosis of the humeral head in displaced proximal humeral fractures. J Trauma
1981;21:788-91.
15. Mittlmeier W, Stedtfeld H, Ewert A, Beck M, Frosch B, Gradl G.
Stabilization of proximal humeral fractures with an angular and
sliding stable antegrade locking nail (Targon PH). J Bone Joint
Surg Am 2003;85(suppl 4):136-46.
16. Mueller ME, Nazarian S, Koch P, Schatzker J. The comprehensive
classification of fractures of long bones. New York, NY: Springer;
1990. p. 54-63.
17. Neer CS II. Displaced proximal humeral fractures. Part II. Treatment of three-part and four-part displacement. J Bone Joint Surg
Am 1970;52:1090-103.
18. Resch H. Fractures of the humeral head. Unfallchirurg 2003;106:
602-17.
19. Scarlat M, Florescu AA. Shoulder function and scores in 180
asymptomatic individuals aged over 75 years. Rev Chir Orthop
2005;91:502-7.
20. Scheid KD, Bigliani LV, Gerber C. Proximal humerus fractures: an
unsolved fracture? Presented at American Academy of Orthopaedic Surgeons, Final Program, 66th Annual Meeting, Feb 8, 1999;
Anaheim, California; 1999.
21. Wijgman AJ, Roolker W, Patt TW, Raaymakers EL, Marti RK.
Open reduction and internal fixation of three and four-part fractures
of the proximal part of the humerus. J Bone Joint Surg Am 2002;11:
1919-25.
22. Zyto K, Wallace WA, Frostick SP, Preston BJ. Outcome after hemiarthroplasty for three and four part fractures of the proximal
humerus. J Shoulder Elbow Surg 1998;7:85-9.