Injury, Int. J. Care Injured 45S (2014) S54–S59
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Proximal third humeral shaft fractures—A fracture entity not fully characterized by conventional AO classification H.W. Stedtfeld, R. Biber * Department of Trauma and Orthopaedic Surgery, Klinikum Nu¨rnberg Klinikum Su¨d, Breslauer Strasse 201, 90471 Nu¨rnberg, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Fracture classification Fracture displacement Meta-diaphyseal fractures Metaphyseal fracture extension
Introduction: The retrospective study was made to evaluate the fracture patterns at the proximal humeral shaft for which the long version of a standard proximal humeral nail (PHNLV) has been used. The indication has been decided by the individual surgeons. Patients and methods: Over a five year period 72 consecutive PHNLV cases of an acute fracture were identified and were included in the study. Mean patient age was 68.9 years. Gender ratio was m/f = 22/ 50. 86.1% of the patients fractured their humerus by a fall, the rest by a high velocity accident. We analysed patient comorbidity, ASA score, osteoporosis, social status before accident, additional injuries affecting local soft tissues or other anatomic regions. We analysed the expansion of the fractures, dividing the humerus into five zones. Fracture morphology was categorized according to the standard AO/ASIF classification (if applicable). Results: Comorbidities were found in 76.4% of the patients. Almost all patients (93.1%) had been living independently at home before the accident. 47.2% of patients had osteoporosis in their medical history. Five patients (6.9%) had a primary palsy of the radial nerve. Six fractures chosen for PHNLV fixation were clearly restricted to the humeral head. The remaining 66 fractures were located in the humeral shaft (AO region 12). There were 5 segmental fractures. Of the remaining 67 fractures affecting the proximal third of the humeral shaft 49.3 percent extended into the humeral head. 98 percent of these fractures displayed spiral morphology. Discussion: Proximal humeral shaft fractures are amazingly similar to subtrochanteric and distal tibial shaft fractures: Spiral fracture types with different grades of comminution are absolutely dominant; a great proportion of the fractures extend into the humeral head with growing tendency of displacement if located closer to the humeral head. Diverging traction of deltoid and pectoralis muscle causes typical displacement if the fracture line runs in between their attachments substantiating the term ‘intermuscular fracture’. A distinct classification system for proximal humeral shaft fractures seems meaningful and is proposed. Conclusions: There is clear evidence of specific characteristics which differentiate proximal third humeral shaft fractures from those of midshaft and distal third. They explain the specific problems of reduction and fixation. If disrespected they will lead to higher rates of therapeutic failure. ß 2013 Elsevier Ltd. All rights reserved.
Introduction Humeral fractures can be treated successfully with both intramedullary and extramedullary implants [1–3]. Analysing the history of implant development in intramedullary nailing one can recognize an expansion of indications from diaphyseal fractures towards metaphyseal fractures. Specific intramedullary fixation systems have been designed for very proximal and distal fractures of femur and tibia. During the last
* Corresponding author. Tel.: +49 911 398 2600; fax: +49 911 398 5830. E-mail address: [email protected] (R. Biber). 0020–1383/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.injury.2013.10.030
decade short proximal humeral nails have also become available. For proximal femoral and proximal humeral fractures involving the shaft, industry offers long versions of these originally short nails [4–9]. It is widely accepted that long versions of proximal femoral nails are well indicated for subtrochanteric fractures [10–12]. However, the spectrum of indications of the long version of proximal humeral nails (PHNLV) remains still unclear. Analysing the injury types and fracture patterns for which these nails have actually been used for in our institution seems to be a suitable first step to determine their adequate indications. Therefore, we conducted a retrospective analysis of injury mechanism and fracture morphology in the respective patients.
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59
Patients and methods We conducted a retrospective analysis on fracture morphology of 92 patients in whom a PHNLV (Targon1 PH long, Aesculap AG, Tuttlingen) was used for fracture fixation over a five years period (February 2000 to February 2005). In order to specifically analyse the epidemiology and biomechanics of acute trauma we excluded pathological fractures (n = 19) and nonunions (n = 1), leaving 72 patients with acute fractures. Mean patient age was 68.9 years (20–93; SD: 15.9). Gender ratio was m/f = 22/50. The right humerus was affected in 38 cases (52.8%); there were no bilateral fractures. 62 patients had broken their humerus by a fall at home (86.1%), 9 fractures had been caused by traffic accident (13.9%) and one by fall from a horse. Using this patient series we analysed patient comorbidity, ASA score [13], osteoporosis, and social status before accident. We also evaluated concomitant injuries affecting other anatomic regions. Regarding the humeral fracture, presence of radial nerve palsy and degree of soft tissue damage was noted [14,15]. For each fracture we analysed its anatomic expansion along the humerus. We divided the humerus into five zones: (1) humeral head, (2) proximal diaphyseal third, (3) midshaft, (4) distal diaphyseal third, and (5) condylar zone. Fracture morphology was categorized according to the standard AO/ASIF classification (if applicable) [16]. Direction and degree of fracture displacement was analysed. Finally we tried to put all the observed fracture types involving the proximal third of the humeral shaft into a systematic and logical order. The study was approved by the local ethics committee.
Results Out of the 72 patients studied, comorbidities were found in 55 patients (76.4%). These included predominantly hypertension (24/ 72), cardiac disease (14/72), diabetes (14/72), alcohol/nicotine abuse (13/72), neurologic disease (9/72), pulmonary disease (6/72) and obesity (6/72). Patient morbidity according to ASA score is shown in Table 1. 67/72 patients (93.1%) had been living independently at home before the accident, 5/72 (6.9%) had been living in a nursery home. 34/72 patients (47.2%) had osteoporosis in their medical history.
S55
Table 1 Patient morbidity according to the ASA score [1].
ASA ASA ASA ASA
S
1 2 3 4
n
Percentage
17 41 13 1 72
23.6% 56.9% 18.1% 1.4% 100%
71/72 fractures (98.6%) were closed without relevant softtissue damage (grade 1). Only one fracture (1.4%) was a grade II open fracture. 42/72 patients (58.3%) had an isolated injury of the upper arm. Additional injuries involved the ipsilateral forearm (n = 3), the contralateral arm (n = 2), leg and foot (n = 10), head (n = 12), spine and pelvis (n = 2), and chest (n = 4). 5 patients (6.9%) showed a primary palsy of the radial nerve. 6/72 fractures affected only the humeral head (zone 1; AO region 11) and could be classified as pure metaphyseal humeral fractures according to the Neer classification. For these we decided that the use of a PHNLV was not indicated (Table 2). The remaining 66 fractures were regarded as shaft fractures (AO region 12). 16 fractures exclusively involved the midshaft. The fracture types were 11 simple spiral (A1), one short oblique (A2), three transverse (A3) and one spiral wedge (B1). Again we decided that, for these 16 fractures, the use of a PHNLV was not indicated (Table 2). The majority (50 fractures) affected the proximal third of the humeral shaft (zone 2). 12/50 fractures were either clearly restricted to the proximal third (zone 2; 9 = 18%) or spanned both proximal third and midshaft (zone 2 + 3; 3 = 6%). 38/50 proximal third fractures showed a fracture extension into the humeral head (76%) (Table 2), mostly undisplaced and hardly verifiable on the X-rays. 13/50 fractures (26%) were classified as type A, 10/50 (20%) type B, and 27/50 (54%) type C. Among the latter group of fractures, there were 5 segmental (AO type C2) fractures (7.6%), each of them involving both the proximal shaft/humeral head and the midshaft (zones 1–3). On each level of these C2 fractures the fracture morphology showed a spiral pattern either being A1, B1 or C1 type.
Table 2 Anatomic zones affected by the fractures fixed with the long version of a proximal humeral nail (PHNLV). The numbers in parentheses indicate the fractures which, after critical review, are judged outside the indication of the PHNLV. Zone 2 fractures are in the focus of this study.
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59
S56
Table 3 AO/ASIF-classification of the observed fractures (6 humeral head fractures of zone 1 excluded) vs. zone distribution of fractures fixed with the long version of a proximal humeral nail. A1 Zones 1–2 Zone 2 Zones 2–3 Zones 1–3 Zone 3
A2
A3
B1
7 2 2
1
1
8 2
11 22
1 2
3 4
1 11
B2
B3
C1
C2
18 3 1 5
S
22 11
28
These three fracture types are the most frequently found types of all the other 45 cases (Table 3). Only two fractures of this group, fully restricted to the upper third (zone 2), were short stable fractures (one short oblique and one transverse). The observed fracture types were different in grades of fracture expansion (from lower shaft to humeral head) and fracture comminution (simple to comminuted). These two parameters are decisive for a basic systematic order into which we have put each individual fracture: we decided for four grades of expansion (Groups A–D) on the basis of the above mentioned different humeral zones and for four grades of comminution (Groups 1–4). Expansion group B (engaging the upper third of the shaft and the humeral head) was divided into two subgroups: (1) the main fracture within the upper third with undisplaced fracture extension towards the greater tubercle and (2) the main fracture directly engaging both head and shaft. Comminution group 4 was specifically reserved for segmental fractures representing two or more separate fracture levels each of which can be categorized by the different preceding grades (1–3) of comminution (Table 4 and Fig. 1). Discussion Do existing classifications adequately describe the observed fracture types? The most common classification for humeral shaft fractures is the AO classification [16]. It gives very thorough and detailed information on fracture morphology. But it does not distinguish between fractures of the proximal humeral shaft and the more distal parts of the diaphysis (AO zone 12). There are several generally acknowledged classification systems for fractures of the humeral head including the most
5
C3
S 33 9 3 5 16 66
27
used ones according to Neer [17] and the AO/ASIF [16]. Yet, in our study we found a wide range of different fractures which involved both the upper third of the humeral shaft and the humeral head without being a true combination of a conventional humeral shaft fracture and a conventional two to four part humeral head fracture. For distal tibia, the tendency of spiral dia-metaphyseal fractures to extend into the articular part of the metaphysis is well known since decades [18–20]. Infra-isthmal tibial spiral fractures frequently have an extension into the ankle joint. Like for the proximal humerus, this type of fracture cannot be classified by either pure diaphyseal or pure metaphyseal fracture classifications. Robinson has, therefore, established a classification of distal tibial metaphyseal fractures [21]. This classifcation differentiates extraarticular stable from spiral fractures types; the latter being, half and half, either extraarticular or associated with intraarticular fractures running towards the medial or dorsal malleolus. In the subtrochanteric area, again, the conventional AO shaft fracture classification [7] does not fully match the fracture reality. There are stable (transverse or short oblique) and spiral fracture types being fully restricted to the proximal end of the shaft. But the spiral fracture types (simple spiral, spiral wedge or spiral comminutes) often involve the trochanteric area affecting either lesser trochanter, greater trochanter, or both. With respect of the involvement of the trochanteric area the AO classification of subtrochanteric fractures was, in our view, a step back from the older Seinsheimer classification [12]. The dominant role of spiral and spiral wedge fractures was better reflected by the subtrochanteric fracture classification of Russell and Taylor from 1992 [11].
Table 4 Distribution of observed fracture types according to the proposed classification. Fractures limited to the humeral head (n = 6) or to the shaft (n = 16) are excluded.
A
B1
B2
C
D
1 2 3 1 2 3 1 2 3 1 2 3 4 1 2 3 4
n
Percentage subtypes
2 2 1 5 11 6 1 2 2 4 1 1 0 2 3 2 5 50
4% 4% 2% 10% 22% 12% 2% 4% 4% 8% 2% 2% 0% 4% 6% 4% 10% 100%
Percentage fracture types 10%
44%
10%
12%
24%
100% Fig. 1. Proposed classification for proximal third (incl. ‘intermuscular’) fractures.
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59
Fig. 2. Typical fracture extension from the proximal humeral shaft into the humeral head. Such extensions were seen frequently (76%), however rarely displaced.
S57
Fig. 3. Displacement of a short transverse fracture (A2) as illustrated by Rockwood [10] (A). Analogous displacement of a spiral wedge fracture (B).
Fracture characteristics of the proximal humeral diaphysis: humeral head involved
Fracture characteristics of the proximal humeral diaphysis: intermuscular displacement
Our analysis of trauma mechanism and fracture morphology indicates that indirect trauma with torsional forces is the predominant mechanism of injury for the proximal third fractures. 48/50 fractures (95.2%), affecting the proximal third of the humeral shaft, presented spiral morphology (type A1, B1 or C1; five C2 fractures with spiral pattern). Only 2/50 fractures (4%) were classified as short and stable (A2 or A3) and therefore did not fit into this pattern. 38/50 proximal fractures extended towards the greater tubercle (76%). However, this fracture extension into the proximal metaphysis was mostly undisplaced. The high proportion of proximal shaft fractures extending into the humeral head seems remarkable. The fracture extension mostly is unobstrusive, detectable only in one single radiograph by a tiny cortical step near the greater tubercle or by a flimsy radiolucent line climbing up towards the greater tubercle (Fig. 2). Only those fractures of zone 2 which were located close to the border of zone 1 showed gross displacement. These fractures presented a pattern which easily may be mixed up with conventional humeral head fractures. The main point which makes the difference is the position of the greater tubercle which is not displaced dorso-proximally by rotator cuff traction but remains part of the latero-proximal shaft fragment (Fig. 1). Very long spiral fractures starting from below the deltoid muscle attachment may wind up to the greater tubercle. We saw five cases of segmental fractures combining shaft fractures below the level of the deltoid attachment with proximal humeral shaft fractures extending towards the greater tubercle as described above.
Literature on conservative treatment of humeral shaft fractures indicates that nonunion rates differ depending on the affected region of the shaft [22–25]. In 2004 Castella´ et al. [26] published a series of 30 humeral nonunions. They reported on 9 cases located in the proximal shaft after suspected spiral or spiral wedge fractures, which they thought to be characteristic for this anatomic area. Yet shaft fractures extending into the humeral head are not reported in this study. Proximal humeral shaft fractures of AO types A and B very often display a typical pattern of displacement: The proximal fragment displaces medially into valgus, losing contact to the distal fragment. The main shaft fragment itself is displaced proximally. In his textbook Rockwood described this pattern of displacement as typical for an ‘intermuscular’ fracture [10]. But regarding fracture morphology there is a difference between his description and our findings: Rockwood’s illustration shows a transverse ‘intermuscular’ fracture (12A3), whereas our series almost exclusively showed spiral fracture patterns of different length and grade of comminution (Fig. 3). The divergent effect of muscular traction obviously is the main reason for less favourable results of conservative treatment. Probably this imposes specific risks and needs of reduction and fixation techniques in operative treatment. Why estimating this intermuscular fracture as separate entity? There are many purposes a fracture classification system can be used for. However experience demonstrates that only those classifications will be generally accepted which help for clinical decision making. Is integration of the intermuscular
S58
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59
fracture into the common classification systems worth the while? For our study we performed a division of the humerus in five zones:
humeral head (proximal metaphysis) proximal third of the diaphysis (‘intermuscular zone’) middle third of the diaphysis (midshaft) distal third of the diaphysis condylar zone (distal metaphysis)
In fact for operative fracture fixation, fracture expansion along the different zones of the humerus has a high impact on the choice of implants. Fractures of each zone represent specific indications for different treatment options: proximal humeral nails, short angular stable plates or arthroplasty may be indicated for humeral head fractures [4,7,8,10]. When it comes to fractures of the proximal third longer implants are needed (plates or nails), but these implants must still be angular stable in order to provide enough torsional and bending stiffness of the fixation construct. Conventional humeral shaft nails are not stable enough in the upper third but they are commonly well indicated for humeral midshaft fractures below the intermuscular zone [6]. Distal third fractures can be fixed with conventional plates and also with interlocking nails if the fracture is not located too much distally. Finally fractures of the supracondylar and condylar zone clearly are the domain of screws and plates. We hope that our proposed classification system for proximal third humeral shaft fractures helps to choose the appropriate implant for the individual fracture, to avoid overuse of the more expensive PHNLV, where conventional implants will do, and to achieve a better understanding for the specific risks and pitfalls of this area. To our knowledge the specific characteristics of the upper third humeral shaft fractures have been denoted only marginally, yet. The exposure to divergent muscle traction and the resulting typical mode of displacement have led to the label of ‘intermuscular fractures’. In addition, the classification is meant to facilitate further studies about these fractures and to make them better comparable. It is the result of our attempt to better reflect the complex fracture situation of the proximal humerus analogously to the closer approach to subtrochanteric fractures by Seinsheimer, Russell and Taylor and to distal tibial fractures by Robinson [11,12,21]. Limitations This is a single-centre study based on a closed collective of patients with proximal humeral shaft fractures fixed by PHNLV. We suggest a new classification for these fractures, which might facilitate further studies on this topic. Certainly the number of patients evaluated is limited. Larger studies are needed in order to prove the validity of our classification. However, even our small series indicates that proximal humeral shaft fractures constitute a distinct entity following specific biomechanical principles, including injury mechanism, fracture morphology, and mode of displacement. Traditional AO classification does not describe this entity adequately. Conclusions There is clear evidence of specific characteristics which differentiate proximal third humeral shaft fractures from those of midshaft and distal third. They explain the specific problems of reduction and fixation. If disregarded they will lead to higher rates of therapeutic failure, as known from literature.
In our collective we described characteristics of a specific injury involving the proximal shaft of the humerus. Since this type of fracture regularly separates the attachments of the deltoid muscle and the greater pectoralis muscle (‘intermuscular fracture’), there is a typical pattern of displacement. Although this mechanism of displacement has been described before, there is still little recognition of this entity in the literature. We showed that spiral fractures caused by indirect trauma prevail at the proximal humeral shaft. However the majority of these fractures is not limited to the diaphysis, but displays extensions into the humeral head. Neither metaphyseal nor diaphyseal fracture classification systems adequately describe this situation. We presented a classification system which is somewhat analogous to Seinsheimer classification of subtrochanteric fractures [17]. This classification respects the involved muscular attachments and the frequent fracture extensions into the metaphyseal or the lower shaft. The proposed classification may serve as basis for further discussion or for clinical outcome studies, which are needed in order to evaluate indications of specific implants such as long proximal humeral nails. Conflict of interest statement Both authors received grants from B.Braun Aesculap (Tuttlingen, Germany) within the past 5 years for consulting activities. HWS is also receiving royalties from the same company. References [1] Brorson S, Rasmussen JV, Frich LH, Olsen BS, Hro´bjartsson A. Benefits and harms of locking plate osteosynthesis in intraarticular (OTA Type C) fractures of the proximal humerus: a systematic review. Injury 2012;43(July (7)):999–1005. [2] Hardeman F, Bollars P, Donnelly M, Bellemans J, Nijs S. Predictive factors for functional outcome and failure in angular stable osteosynthesis of the proximal humerus. Injury 2012;43(February (2)):153–8. [3] Ko¨nigshausen M, Ku¨bler L, Godry H, Citak M, Schildhauer TA, Seybold D. Clinical outcome and complications using a polyaxial locking plate in the treatment of displaced proximal humerus fractures. A reliable system? Injury 2012;43(February (2)):223–31. [4] Mittelmeier ThWF, Stedtfeld H-W, 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 2003;85-A(Suppl 4). [5] Petsatodes G, Karataglis D, Papadopoulos P, Christoforides J, Gigis J, Pournaras J. Antegrade interlocking nailing of humeral shaft fractures. J Orthop Sci 2004;9(3):247–52. [6] Schratz W, Wo¨rsdo¨rfer O, Klo¨ckner C, Go¨tze C. Behandlung der Oberarmschaftfraktur mit dem intramedulla¨ren Verfahren (Seidel-Nagel, Marchetti-VicenziNagel, Prevot-Pins). Unfallchirurg 1998;101:12–7. [7] Stedtfeld H-W, Wick M, Winkler R, Attmanspacher W. Marknagelosteosynthese-Indikation, Technik, sinnvoller Fortschritt (!?), Proximaler Humerus. Trauma Berufskr Suppl 2004;2:241–6. [8] Stedtfeld H-W, Attmanspacher W, Thaler K, Frosch B. Fixation von Humeruskopffrakturen mit anterograder Marknagelung. Zentralbl Chir 2003;128:6–11. [9] Yang KH. Helical plate fixation for treatment of comminuted fractures of the proximal and middle one-third of the humerus. Injury 2005;36(1):75–80. [10] Rockwood CA, Green DP, Bucholz RW. Rockwood and Green’s fractures in adults. 3rd ed. Philadelphia: Lippincott Williams and Wilkins; 1991. [11] Russell TA, Taylor JC. Subtrochanteric fractures of the femur. In: Browner BD, Jupiter JB, Levine AM, Trafton PG, editors. Skeletal trauma. 2nd ed., WB Saunder: Philadelphia, PA; 1992. p. 1832–78. [12] Seinsheimer F. Subtrochanteric fractures of the femur. J Bone Joint Surg Am 1978;60(3):300–6. [13] American Society of Anesthesiologists (ASA): new classification of physical status. Anesthesiology 1963;24:111. [14] Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analysis. J Bone Joint Surg 1976;58-A:453–8. [15] Tscherne H, Oestern HJ. Die Klassifizierung des Weichteilschadens bei offenen und geschlossenen Frakturen. Unfallheilkunde 1982;85:111–5. [16] Mu¨ller ME, Nazarian S, Koch P, et al. The comprehensive classification of fractures of long bones. Berlin, Germany: Springer-Verlag; 1990. [17] Neer CS. Displaced proximal humeral fractures. Part I. Classification and evaluation. By Charles S. Neer, I, 1970. Clin Orthop Relat Res 1987;223:3–10. [18] Bo¨stman OM. Displaced malleolar fractures associated with spiral fractures of the tibial shaft. Clin Orthop Relat Res 1988;228:202–7. [19] Lauge-Hansen N. Laesiones opstaaet ved patologiske bevaegelser af foden. Nord Med 1946;32:2337.
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59 [20] Weber BG. Die Verletzung des oberen Sprunggelenkes. 2nd ed. Bern: Hans Huber Verlag; 1972. [21] Robinson CM, McLauchlan GJ, McLean IP, Court-Brown CM. Distal metaphyseal fractures of the tibia with minimal involvement of the ankle. Classification and treatment by locked intramedullary nailing. J Bone Joint Surg Br 1995;77(5):781–7. [22] Ring D, Chin K, Taghinia AH, Jupiter JB. Nonunion after functional brace treatment of diaphyseal humerus fractures. J Trauma 2007;62(5): 1157–8.
S59
[23] Rutgers M, Ring D. Treatment of diaphyseal fractures of the humerus using a functional brace. J Orthop Trauma 2006;20(9):597–601. [24] Sarmiento A, Latta LL. Closed functional treatment of fractures. Berlin: Springer; 1981. [25] Toivanen JA, Nieminen J, Laine HJ, Honkonen SE, Ja¨rvinen MJ. Functional treatment of closed humeral shaft fractures. Int Orthop 2005;29(1):10–3. [26] Castella´ FB, Garcia FB, Berry EM, Perello´ EB, Sa´nchez-Alepuz E, Gabarda R. Nonunion of the humeral shaft: long lateral butterfly fracture – a nonunion predictive pattern? Clin Orthop Relat Res 2004;(424):227–30.
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Proximal third humeral shaft fractures—A fracture entity not fully characterized by conventional AO classification H.W. Stedtfeld, R. Biber * Department of Trauma and Orthopaedic Surgery, Klinikum Nu¨rnberg Klinikum Su¨d, Breslauer Strasse 201, 90471 Nu¨rnberg, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Fracture classification Fracture displacement Meta-diaphyseal fractures Metaphyseal fracture extension
Introduction: The retrospective study was made to evaluate the fracture patterns at the proximal humeral shaft for which the long version of a standard proximal humeral nail (PHNLV) has been used. The indication has been decided by the individual surgeons. Patients and methods: Over a five year period 72 consecutive PHNLV cases of an acute fracture were identified and were included in the study. Mean patient age was 68.9 years. Gender ratio was m/f = 22/ 50. 86.1% of the patients fractured their humerus by a fall, the rest by a high velocity accident. We analysed patient comorbidity, ASA score, osteoporosis, social status before accident, additional injuries affecting local soft tissues or other anatomic regions. We analysed the expansion of the fractures, dividing the humerus into five zones. Fracture morphology was categorized according to the standard AO/ASIF classification (if applicable). Results: Comorbidities were found in 76.4% of the patients. Almost all patients (93.1%) had been living independently at home before the accident. 47.2% of patients had osteoporosis in their medical history. Five patients (6.9%) had a primary palsy of the radial nerve. Six fractures chosen for PHNLV fixation were clearly restricted to the humeral head. The remaining 66 fractures were located in the humeral shaft (AO region 12). There were 5 segmental fractures. Of the remaining 67 fractures affecting the proximal third of the humeral shaft 49.3 percent extended into the humeral head. 98 percent of these fractures displayed spiral morphology. Discussion: Proximal humeral shaft fractures are amazingly similar to subtrochanteric and distal tibial shaft fractures: Spiral fracture types with different grades of comminution are absolutely dominant; a great proportion of the fractures extend into the humeral head with growing tendency of displacement if located closer to the humeral head. Diverging traction of deltoid and pectoralis muscle causes typical displacement if the fracture line runs in between their attachments substantiating the term ‘intermuscular fracture’. A distinct classification system for proximal humeral shaft fractures seems meaningful and is proposed. Conclusions: There is clear evidence of specific characteristics which differentiate proximal third humeral shaft fractures from those of midshaft and distal third. They explain the specific problems of reduction and fixation. If disrespected they will lead to higher rates of therapeutic failure. ß 2013 Elsevier Ltd. All rights reserved.
Introduction Humeral fractures can be treated successfully with both intramedullary and extramedullary implants [1–3]. Analysing the history of implant development in intramedullary nailing one can recognize an expansion of indications from diaphyseal fractures towards metaphyseal fractures. Specific intramedullary fixation systems have been designed for very proximal and distal fractures of femur and tibia. During the last
* Corresponding author. Tel.: +49 911 398 2600; fax: +49 911 398 5830. E-mail address: [email protected] (R. Biber). 0020–1383/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.injury.2013.10.030
decade short proximal humeral nails have also become available. For proximal femoral and proximal humeral fractures involving the shaft, industry offers long versions of these originally short nails [4–9]. It is widely accepted that long versions of proximal femoral nails are well indicated for subtrochanteric fractures [10–12]. However, the spectrum of indications of the long version of proximal humeral nails (PHNLV) remains still unclear. Analysing the injury types and fracture patterns for which these nails have actually been used for in our institution seems to be a suitable first step to determine their adequate indications. Therefore, we conducted a retrospective analysis of injury mechanism and fracture morphology in the respective patients.
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59
Patients and methods We conducted a retrospective analysis on fracture morphology of 92 patients in whom a PHNLV (Targon1 PH long, Aesculap AG, Tuttlingen) was used for fracture fixation over a five years period (February 2000 to February 2005). In order to specifically analyse the epidemiology and biomechanics of acute trauma we excluded pathological fractures (n = 19) and nonunions (n = 1), leaving 72 patients with acute fractures. Mean patient age was 68.9 years (20–93; SD: 15.9). Gender ratio was m/f = 22/50. The right humerus was affected in 38 cases (52.8%); there were no bilateral fractures. 62 patients had broken their humerus by a fall at home (86.1%), 9 fractures had been caused by traffic accident (13.9%) and one by fall from a horse. Using this patient series we analysed patient comorbidity, ASA score [13], osteoporosis, and social status before accident. We also evaluated concomitant injuries affecting other anatomic regions. Regarding the humeral fracture, presence of radial nerve palsy and degree of soft tissue damage was noted [14,15]. For each fracture we analysed its anatomic expansion along the humerus. We divided the humerus into five zones: (1) humeral head, (2) proximal diaphyseal third, (3) midshaft, (4) distal diaphyseal third, and (5) condylar zone. Fracture morphology was categorized according to the standard AO/ASIF classification (if applicable) [16]. Direction and degree of fracture displacement was analysed. Finally we tried to put all the observed fracture types involving the proximal third of the humeral shaft into a systematic and logical order. The study was approved by the local ethics committee.
Results Out of the 72 patients studied, comorbidities were found in 55 patients (76.4%). These included predominantly hypertension (24/ 72), cardiac disease (14/72), diabetes (14/72), alcohol/nicotine abuse (13/72), neurologic disease (9/72), pulmonary disease (6/72) and obesity (6/72). Patient morbidity according to ASA score is shown in Table 1. 67/72 patients (93.1%) had been living independently at home before the accident, 5/72 (6.9%) had been living in a nursery home. 34/72 patients (47.2%) had osteoporosis in their medical history.
S55
Table 1 Patient morbidity according to the ASA score [1].
ASA ASA ASA ASA
S
1 2 3 4
n
Percentage
17 41 13 1 72
23.6% 56.9% 18.1% 1.4% 100%
71/72 fractures (98.6%) were closed without relevant softtissue damage (grade 1). Only one fracture (1.4%) was a grade II open fracture. 42/72 patients (58.3%) had an isolated injury of the upper arm. Additional injuries involved the ipsilateral forearm (n = 3), the contralateral arm (n = 2), leg and foot (n = 10), head (n = 12), spine and pelvis (n = 2), and chest (n = 4). 5 patients (6.9%) showed a primary palsy of the radial nerve. 6/72 fractures affected only the humeral head (zone 1; AO region 11) and could be classified as pure metaphyseal humeral fractures according to the Neer classification. For these we decided that the use of a PHNLV was not indicated (Table 2). The remaining 66 fractures were regarded as shaft fractures (AO region 12). 16 fractures exclusively involved the midshaft. The fracture types were 11 simple spiral (A1), one short oblique (A2), three transverse (A3) and one spiral wedge (B1). Again we decided that, for these 16 fractures, the use of a PHNLV was not indicated (Table 2). The majority (50 fractures) affected the proximal third of the humeral shaft (zone 2). 12/50 fractures were either clearly restricted to the proximal third (zone 2; 9 = 18%) or spanned both proximal third and midshaft (zone 2 + 3; 3 = 6%). 38/50 proximal third fractures showed a fracture extension into the humeral head (76%) (Table 2), mostly undisplaced and hardly verifiable on the X-rays. 13/50 fractures (26%) were classified as type A, 10/50 (20%) type B, and 27/50 (54%) type C. Among the latter group of fractures, there were 5 segmental (AO type C2) fractures (7.6%), each of them involving both the proximal shaft/humeral head and the midshaft (zones 1–3). On each level of these C2 fractures the fracture morphology showed a spiral pattern either being A1, B1 or C1 type.
Table 2 Anatomic zones affected by the fractures fixed with the long version of a proximal humeral nail (PHNLV). The numbers in parentheses indicate the fractures which, after critical review, are judged outside the indication of the PHNLV. Zone 2 fractures are in the focus of this study.
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59
S56
Table 3 AO/ASIF-classification of the observed fractures (6 humeral head fractures of zone 1 excluded) vs. zone distribution of fractures fixed with the long version of a proximal humeral nail. A1 Zones 1–2 Zone 2 Zones 2–3 Zones 1–3 Zone 3
A2
A3
B1
7 2 2
1
1
8 2
11 22
1 2
3 4
1 11
B2
B3
C1
C2
18 3 1 5
S
22 11
28
These three fracture types are the most frequently found types of all the other 45 cases (Table 3). Only two fractures of this group, fully restricted to the upper third (zone 2), were short stable fractures (one short oblique and one transverse). The observed fracture types were different in grades of fracture expansion (from lower shaft to humeral head) and fracture comminution (simple to comminuted). These two parameters are decisive for a basic systematic order into which we have put each individual fracture: we decided for four grades of expansion (Groups A–D) on the basis of the above mentioned different humeral zones and for four grades of comminution (Groups 1–4). Expansion group B (engaging the upper third of the shaft and the humeral head) was divided into two subgroups: (1) the main fracture within the upper third with undisplaced fracture extension towards the greater tubercle and (2) the main fracture directly engaging both head and shaft. Comminution group 4 was specifically reserved for segmental fractures representing two or more separate fracture levels each of which can be categorized by the different preceding grades (1–3) of comminution (Table 4 and Fig. 1). Discussion Do existing classifications adequately describe the observed fracture types? The most common classification for humeral shaft fractures is the AO classification [16]. It gives very thorough and detailed information on fracture morphology. But it does not distinguish between fractures of the proximal humeral shaft and the more distal parts of the diaphysis (AO zone 12). There are several generally acknowledged classification systems for fractures of the humeral head including the most
5
C3
S 33 9 3 5 16 66
27
used ones according to Neer [17] and the AO/ASIF [16]. Yet, in our study we found a wide range of different fractures which involved both the upper third of the humeral shaft and the humeral head without being a true combination of a conventional humeral shaft fracture and a conventional two to four part humeral head fracture. For distal tibia, the tendency of spiral dia-metaphyseal fractures to extend into the articular part of the metaphysis is well known since decades [18–20]. Infra-isthmal tibial spiral fractures frequently have an extension into the ankle joint. Like for the proximal humerus, this type of fracture cannot be classified by either pure diaphyseal or pure metaphyseal fracture classifications. Robinson has, therefore, established a classification of distal tibial metaphyseal fractures [21]. This classifcation differentiates extraarticular stable from spiral fractures types; the latter being, half and half, either extraarticular or associated with intraarticular fractures running towards the medial or dorsal malleolus. In the subtrochanteric area, again, the conventional AO shaft fracture classification [7] does not fully match the fracture reality. There are stable (transverse or short oblique) and spiral fracture types being fully restricted to the proximal end of the shaft. But the spiral fracture types (simple spiral, spiral wedge or spiral comminutes) often involve the trochanteric area affecting either lesser trochanter, greater trochanter, or both. With respect of the involvement of the trochanteric area the AO classification of subtrochanteric fractures was, in our view, a step back from the older Seinsheimer classification [12]. The dominant role of spiral and spiral wedge fractures was better reflected by the subtrochanteric fracture classification of Russell and Taylor from 1992 [11].
Table 4 Distribution of observed fracture types according to the proposed classification. Fractures limited to the humeral head (n = 6) or to the shaft (n = 16) are excluded.
A
B1
B2
C
D
1 2 3 1 2 3 1 2 3 1 2 3 4 1 2 3 4
n
Percentage subtypes
2 2 1 5 11 6 1 2 2 4 1 1 0 2 3 2 5 50
4% 4% 2% 10% 22% 12% 2% 4% 4% 8% 2% 2% 0% 4% 6% 4% 10% 100%
Percentage fracture types 10%
44%
10%
12%
24%
100% Fig. 1. Proposed classification for proximal third (incl. ‘intermuscular’) fractures.
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59
Fig. 2. Typical fracture extension from the proximal humeral shaft into the humeral head. Such extensions were seen frequently (76%), however rarely displaced.
S57
Fig. 3. Displacement of a short transverse fracture (A2) as illustrated by Rockwood [10] (A). Analogous displacement of a spiral wedge fracture (B).
Fracture characteristics of the proximal humeral diaphysis: humeral head involved
Fracture characteristics of the proximal humeral diaphysis: intermuscular displacement
Our analysis of trauma mechanism and fracture morphology indicates that indirect trauma with torsional forces is the predominant mechanism of injury for the proximal third fractures. 48/50 fractures (95.2%), affecting the proximal third of the humeral shaft, presented spiral morphology (type A1, B1 or C1; five C2 fractures with spiral pattern). Only 2/50 fractures (4%) were classified as short and stable (A2 or A3) and therefore did not fit into this pattern. 38/50 proximal fractures extended towards the greater tubercle (76%). However, this fracture extension into the proximal metaphysis was mostly undisplaced. The high proportion of proximal shaft fractures extending into the humeral head seems remarkable. The fracture extension mostly is unobstrusive, detectable only in one single radiograph by a tiny cortical step near the greater tubercle or by a flimsy radiolucent line climbing up towards the greater tubercle (Fig. 2). Only those fractures of zone 2 which were located close to the border of zone 1 showed gross displacement. These fractures presented a pattern which easily may be mixed up with conventional humeral head fractures. The main point which makes the difference is the position of the greater tubercle which is not displaced dorso-proximally by rotator cuff traction but remains part of the latero-proximal shaft fragment (Fig. 1). Very long spiral fractures starting from below the deltoid muscle attachment may wind up to the greater tubercle. We saw five cases of segmental fractures combining shaft fractures below the level of the deltoid attachment with proximal humeral shaft fractures extending towards the greater tubercle as described above.
Literature on conservative treatment of humeral shaft fractures indicates that nonunion rates differ depending on the affected region of the shaft [22–25]. In 2004 Castella´ et al. [26] published a series of 30 humeral nonunions. They reported on 9 cases located in the proximal shaft after suspected spiral or spiral wedge fractures, which they thought to be characteristic for this anatomic area. Yet shaft fractures extending into the humeral head are not reported in this study. Proximal humeral shaft fractures of AO types A and B very often display a typical pattern of displacement: The proximal fragment displaces medially into valgus, losing contact to the distal fragment. The main shaft fragment itself is displaced proximally. In his textbook Rockwood described this pattern of displacement as typical for an ‘intermuscular’ fracture [10]. But regarding fracture morphology there is a difference between his description and our findings: Rockwood’s illustration shows a transverse ‘intermuscular’ fracture (12A3), whereas our series almost exclusively showed spiral fracture patterns of different length and grade of comminution (Fig. 3). The divergent effect of muscular traction obviously is the main reason for less favourable results of conservative treatment. Probably this imposes specific risks and needs of reduction and fixation techniques in operative treatment. Why estimating this intermuscular fracture as separate entity? There are many purposes a fracture classification system can be used for. However experience demonstrates that only those classifications will be generally accepted which help for clinical decision making. Is integration of the intermuscular
S58
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59
fracture into the common classification systems worth the while? For our study we performed a division of the humerus in five zones:
humeral head (proximal metaphysis) proximal third of the diaphysis (‘intermuscular zone’) middle third of the diaphysis (midshaft) distal third of the diaphysis condylar zone (distal metaphysis)
In fact for operative fracture fixation, fracture expansion along the different zones of the humerus has a high impact on the choice of implants. Fractures of each zone represent specific indications for different treatment options: proximal humeral nails, short angular stable plates or arthroplasty may be indicated for humeral head fractures [4,7,8,10]. When it comes to fractures of the proximal third longer implants are needed (plates or nails), but these implants must still be angular stable in order to provide enough torsional and bending stiffness of the fixation construct. Conventional humeral shaft nails are not stable enough in the upper third but they are commonly well indicated for humeral midshaft fractures below the intermuscular zone [6]. Distal third fractures can be fixed with conventional plates and also with interlocking nails if the fracture is not located too much distally. Finally fractures of the supracondylar and condylar zone clearly are the domain of screws and plates. We hope that our proposed classification system for proximal third humeral shaft fractures helps to choose the appropriate implant for the individual fracture, to avoid overuse of the more expensive PHNLV, where conventional implants will do, and to achieve a better understanding for the specific risks and pitfalls of this area. To our knowledge the specific characteristics of the upper third humeral shaft fractures have been denoted only marginally, yet. The exposure to divergent muscle traction and the resulting typical mode of displacement have led to the label of ‘intermuscular fractures’. In addition, the classification is meant to facilitate further studies about these fractures and to make them better comparable. It is the result of our attempt to better reflect the complex fracture situation of the proximal humerus analogously to the closer approach to subtrochanteric fractures by Seinsheimer, Russell and Taylor and to distal tibial fractures by Robinson [11,12,21]. Limitations This is a single-centre study based on a closed collective of patients with proximal humeral shaft fractures fixed by PHNLV. We suggest a new classification for these fractures, which might facilitate further studies on this topic. Certainly the number of patients evaluated is limited. Larger studies are needed in order to prove the validity of our classification. However, even our small series indicates that proximal humeral shaft fractures constitute a distinct entity following specific biomechanical principles, including injury mechanism, fracture morphology, and mode of displacement. Traditional AO classification does not describe this entity adequately. Conclusions There is clear evidence of specific characteristics which differentiate proximal third humeral shaft fractures from those of midshaft and distal third. They explain the specific problems of reduction and fixation. If disregarded they will lead to higher rates of therapeutic failure, as known from literature.
In our collective we described characteristics of a specific injury involving the proximal shaft of the humerus. Since this type of fracture regularly separates the attachments of the deltoid muscle and the greater pectoralis muscle (‘intermuscular fracture’), there is a typical pattern of displacement. Although this mechanism of displacement has been described before, there is still little recognition of this entity in the literature. We showed that spiral fractures caused by indirect trauma prevail at the proximal humeral shaft. However the majority of these fractures is not limited to the diaphysis, but displays extensions into the humeral head. Neither metaphyseal nor diaphyseal fracture classification systems adequately describe this situation. We presented a classification system which is somewhat analogous to Seinsheimer classification of subtrochanteric fractures [17]. This classification respects the involved muscular attachments and the frequent fracture extensions into the metaphyseal or the lower shaft. The proposed classification may serve as basis for further discussion or for clinical outcome studies, which are needed in order to evaluate indications of specific implants such as long proximal humeral nails. Conflict of interest statement Both authors received grants from B.Braun Aesculap (Tuttlingen, Germany) within the past 5 years for consulting activities. HWS is also receiving royalties from the same company. References [1] Brorson S, Rasmussen JV, Frich LH, Olsen BS, Hro´bjartsson A. Benefits and harms of locking plate osteosynthesis in intraarticular (OTA Type C) fractures of the proximal humerus: a systematic review. Injury 2012;43(July (7)):999–1005. [2] Hardeman F, Bollars P, Donnelly M, Bellemans J, Nijs S. Predictive factors for functional outcome and failure in angular stable osteosynthesis of the proximal humerus. Injury 2012;43(February (2)):153–8. [3] Ko¨nigshausen M, Ku¨bler L, Godry H, Citak M, Schildhauer TA, Seybold D. Clinical outcome and complications using a polyaxial locking plate in the treatment of displaced proximal humerus fractures. A reliable system? Injury 2012;43(February (2)):223–31. [4] Mittelmeier ThWF, Stedtfeld H-W, 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 2003;85-A(Suppl 4). [5] Petsatodes G, Karataglis D, Papadopoulos P, Christoforides J, Gigis J, Pournaras J. Antegrade interlocking nailing of humeral shaft fractures. J Orthop Sci 2004;9(3):247–52. [6] Schratz W, Wo¨rsdo¨rfer O, Klo¨ckner C, Go¨tze C. Behandlung der Oberarmschaftfraktur mit dem intramedulla¨ren Verfahren (Seidel-Nagel, Marchetti-VicenziNagel, Prevot-Pins). Unfallchirurg 1998;101:12–7. [7] Stedtfeld H-W, Wick M, Winkler R, Attmanspacher W. Marknagelosteosynthese-Indikation, Technik, sinnvoller Fortschritt (!?), Proximaler Humerus. Trauma Berufskr Suppl 2004;2:241–6. [8] Stedtfeld H-W, Attmanspacher W, Thaler K, Frosch B. Fixation von Humeruskopffrakturen mit anterograder Marknagelung. Zentralbl Chir 2003;128:6–11. [9] Yang KH. Helical plate fixation for treatment of comminuted fractures of the proximal and middle one-third of the humerus. Injury 2005;36(1):75–80. [10] Rockwood CA, Green DP, Bucholz RW. Rockwood and Green’s fractures in adults. 3rd ed. Philadelphia: Lippincott Williams and Wilkins; 1991. [11] Russell TA, Taylor JC. Subtrochanteric fractures of the femur. In: Browner BD, Jupiter JB, Levine AM, Trafton PG, editors. Skeletal trauma. 2nd ed., WB Saunder: Philadelphia, PA; 1992. p. 1832–78. [12] Seinsheimer F. Subtrochanteric fractures of the femur. J Bone Joint Surg Am 1978;60(3):300–6. [13] American Society of Anesthesiologists (ASA): new classification of physical status. Anesthesiology 1963;24:111. [14] Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analysis. J Bone Joint Surg 1976;58-A:453–8. [15] Tscherne H, Oestern HJ. Die Klassifizierung des Weichteilschadens bei offenen und geschlossenen Frakturen. Unfallheilkunde 1982;85:111–5. [16] Mu¨ller ME, Nazarian S, Koch P, et al. The comprehensive classification of fractures of long bones. Berlin, Germany: Springer-Verlag; 1990. [17] Neer CS. Displaced proximal humeral fractures. Part I. Classification and evaluation. By Charles S. Neer, I, 1970. Clin Orthop Relat Res 1987;223:3–10. [18] Bo¨stman OM. Displaced malleolar fractures associated with spiral fractures of the tibial shaft. Clin Orthop Relat Res 1988;228:202–7. [19] Lauge-Hansen N. Laesiones opstaaet ved patologiske bevaegelser af foden. Nord Med 1946;32:2337.
H.W. Stedtfeld, R. Biber / Injury, Int. J. Care Injured 45S (2014) S54–S59 [20] Weber BG. Die Verletzung des oberen Sprunggelenkes. 2nd ed. Bern: Hans Huber Verlag; 1972. [21] Robinson CM, McLauchlan GJ, McLean IP, Court-Brown CM. Distal metaphyseal fractures of the tibia with minimal involvement of the ankle. Classification and treatment by locked intramedullary nailing. J Bone Joint Surg Br 1995;77(5):781–7. [22] Ring D, Chin K, Taghinia AH, Jupiter JB. Nonunion after functional brace treatment of diaphyseal humerus fractures. J Trauma 2007;62(5): 1157–8.
S59
[23] Rutgers M, Ring D. Treatment of diaphyseal fractures of the humerus using a functional brace. J Orthop Trauma 2006;20(9):597–601. [24] Sarmiento A, Latta LL. Closed functional treatment of fractures. Berlin: Springer; 1981. [25] Toivanen JA, Nieminen J, Laine HJ, Honkonen SE, Ja¨rvinen MJ. Functional treatment of closed humeral shaft fractures. Int Orthop 2005;29(1):10–3. [26] Castella´ FB, Garcia FB, Berry EM, Perello´ EB, Sa´nchez-Alepuz E, Gabarda R. Nonunion of the humeral shaft: long lateral butterfly fracture – a nonunion predictive pattern? Clin Orthop Relat Res 2004;(424):227–30.
