G Model
JINJ-5554; No. of Pages 5 Injury, Int. J. Care Injured xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures Tobias Harnoss a,*, Bernd Felkel b,1, Stefan Do¨bele c,2, Ulrich Schreiber b,3, Andreas Lenich d,4 a
St. Vinzenz Clinic Pfronten, Department for Surgery and Orthopedics, Kirchenweg 15, 87459 Pfronten, Germany Munich University of Applied Sciences, Lothstr. 34, 80335 Munich, Germany c Department of Traumatology and Reconstructive Surgery, BG Unfallklinik, Eberhard Karls Universita¨t Tu¨bingen, Schnarrenbergstraße 95, 72076 Tu¨bingen, Germany d Clinic of Orthopedics and Traumatology, Department for Sport Orthopedics, Klinikum rechts der Isar, Technische Universita¨t Mu¨nchen, Ismaningerstr. 22, 81675 Munich, Germany b
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 8 October 2013
Proximal humerus fractures treated with intramedullary nails show good results. However, the correct anatomical reconstruction of four-part fractures is demanding especially when using intramedullary nails. We therefore compared different intramedullary nail designs for the proximal humerus in a virtual morphological manner. Three commercially available nailing systems where virtually implanted in virtually generated reproducible four-part fractures of 25 digitised humeri. The objective of this study was to quantify and characterise the anatomical position of the proximal screws in the most vulnerable case of a four-part fracture. Taking into account a minimum distance of 5 mm between the screw head and the fracture line, osteosynthesis was possible in 54 out of 75 cases. Difficulties placing the proximal screws could be observed at the localisation of the lower lesser tubercle or/and at the sulcus intertubercularis. This morphological analysis could be the basis for choosing the most sufficient implant intra operatively or even improving the nail design. ß 2013 Elsevier Ltd. All rights reserved.
Keywords: Proximal humerus Fracture Intramedullary nailing Virtual comparison
Introduction About 4% of all fractures in the human body involve the humeral head with severe fractures patterns often seen in the elderly population [1,2]. Non-displaced or only slightly displaced fractures typically are treated non-operatively [3]. There is a certain consensus to treat displaced four-part fractures associated with a high degree of comminution (involving a split head), with primary hemiarthroplasty [3–5]. However, the recommendations for the treatment of displaced three or four-part fractures remain controversial. Recently, there has been a trend to treat proximal humeral fractures with angular stable systems such as plates and intramedullary nails [6–11]. In addition to the treatment with locking plate systems several types of antegrade interlocking nails
* Corresponding author. Tel.: +49 173 3684176; fax: +49 321 21274439. E-mail addresses: [email protected] (T. Harnoss), [email protected] (B. Felkel), [email protected] (S. Do¨bele), [email protected] (U. Schreiber), [email protected] (A. Lenich). 1 Tel.: +49 176 477 288 05; fax: +49 89 1218 18 3718. 2 Tel.: +49 7071 606 1067; fax: +49 7071 606 1002. 3 Tel.: +49 179 5868 236; fax: +49 89 1218 18 3718. 4 Tel.: +49 89 4140 5135; fax: +49 89 4140 4890.
with uniplanar or multiplanar interlocking modes are currently reported to provide effective means for stabilising proximal humeral fractures [12–17]. According to the manufacturer’s instructions, two to four-part fractures may be treated with intramedullary implants. Several recent studies examining the outcome after intramedullary nailing [14,18,19] noted indications including treatment of displaced two and three-part fractures [14,19]. However, four-part fractures were also treated with the intramedullary implants and yielded relatively good results [14,18,19]. Shortcomings of intramedullary nailing include cutting out of the proximal screws and difficulties in a secure fixation of the bone fragments [14,18,19]. For this reason, we examined the nail design morphologically with a focus on screw configuration, i.e. arrangement, position and orientation of the proximal screws, using a new computer based imaging method. We especially focused on how the osteosynthesis is influenced by the anatomical variance, which can be seen in the European population. We used radiographic landmarks for evaluating the anatomical variance, which can also be seen on images taken preoperatively. This could lead to a better appraisal of operative results. Overall, we compared commercially available intramedullary nail systems used for multi-part proximal humerus fractures. Cadaveric humeri corresponding to the
0020–1383/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.injury.2013.10.039
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
G Model
JINJ-5554; No. of Pages 5 T. Harnoss et al. / Injury, Int. J. Care Injured xxx (2013) xxx–xxx
2
normal distribution in the European population [20] were used to create a virtual model for implant evaluation. Proceeding in that order, it was possible to acquire implantation data for all three nails in all cadaveric specimens. Materials and methods In total 25 fresh frozen cadaver humeri (43–88 years, mean age: 65 years, 16 male, 9 female) were digitised using computer tomography (Cardiac Sensation, Siemens, Erlangen, Germany, 120 kV, 200 mAs, layer thickness of 1 mm) and then reconstructed in 3D using a program specifically developed for medical morphological analyses (Amira, Mercury Computer Systems GmbH, Berlin, Germany). In order to classify the cadaveric specimens anatomically, the software was used to measure the following anatomical landmarks: 1. Diameter of the anatomical neck 2. Diameter of the humeral head 3. Diameter of the shaft Three intramedullary nails currently available for the treatment of proximal humeral fractures, Targon-PHN (left and right version), T2-PHN (left and right version, bend setup) and TriGen-PHN (unilateral setup), were digitised with the help of a CAD system (Catia V5 R14, Dassault Systems, Suresnes Cedex, France). For clarity reasons, the screws were replaced by short bolts allowing analysis of the proximal screw orientations (see Fig. 1). The Amira program was used to simulate a virtual four-part fracture in the humeri using the four-part fracture according to Codman [21] as a basic model and modifying it according to Resch [3]: The fracture line between the two tubercles does not run in the intertubercular groove, but about 5 mm more laterally. The resulting four fragments are the same as described by Neer [22]: a lesser tubercle fragment, a greater tubercle fragment, the remaining shaft the humeral head with an intact anatomical neck and a fracture at the surgical neck (see Fig. 1). The nails were virtually implanted according to the manufacturers’ instructions and recommendations published in recent specialist literature [19,23]. The nail was inserted at the most cranial point of the humeral head along the diaphyseal axis so that the proximal end of the nail came to lie in the subchondral area of the humeral head directly beneath the surface [23]. In that way, the nail did not project from the bone and potential subacromial impingement [19] was avoided. The nail was rotated to optimise the screws’ position in the fragments, i.e. to provide them with a maximum distance from the fragment edge (see Fig. 1). A total of 75 implantations in 25 humeri were conducted in this manner. After the implantation, the position of the implant was evaluated based on selected distances between the fracture lines
Fig. 1. Four-part fracture and implanted nail (Targon-PHN, Aesculap) with safety margins.
Fig. 2. Measurements from the proximal screws (Targon-PHN) to the fracture lines.
and the screws. Distances were measured in 4 directions at each of the tubercles. These were the cranial and caudal direction as well as the medial and lateral on the lesser and the ventral and dorsal direction at the greater tubercle. Current literature considers an implantation to be safe if there is a distance of at least 5 mm between the screw and the fracture line [23]. Therefore a 5 mm band indicated by landmarks (see Fig. 2) was drawn in parallel to the fracture line to visualise this distance. For the evaluation of the osteosynthesis, implantation success was divided into three categories (see Fig. 3): Implantation possible (class A): All inserted screws are at a safe distance of at least 5 mm from the fracture line. Implantation difficult (class B): At least one of the screws is less than 5 mm from a fracture line or the end of a fragment. Implantation impossible (class C): At least one of the screws lies outside its fragment. The normal distribution of the available bone collection was tested with the Kolmogorov–Smirnov-Test and verified based on the epidemiological study [20] by Mall et al. The Mann–Whitney test was performed to see if the implantability depended on the size of the humerus. The next step was to find out whether the available data allowed a prediction of implantability as a function of humerus size using logistic regression and the chi-square test. The results were plotted in a graph showing the probability of a successful implantation as a function of the size (see Fig. 4).
Fig. 3. Evaluation matrix with three categories: implantation possible (A), difficult (B) and impossible (C).
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
G Model
JINJ-5554; No. of Pages 5 T. Harnoss et al. / Injury, Int. J. Care Injured xxx (2013) xxx–xxx
3
Table 1 Implantability of the nails, absolute number (n = 75).
Fig. 4. Implantability as a function of the diameter of the anatomical neck – circles indicate failed (0,00) and successful (1,00) implantations and the accordant diameter of the anatomical neck.
Sensitivity and specificity were evaluated by ROC analysis. A pvalue of less than 0.05 was set as the significance level.
Results The cadaveric specimens represented a normal distribution with respect to anatomical features. According to Mall, the mean diameter of the anatomical neck is 4.70 cm (sd = 0.27). The diameter of the anatomical neck is used as a reference for the humerus size [20,23]. The collection of 25 humeri available for the present study had a mean diameter of the anatomical neck of 4.62 cm (sd = 0.52), which corresponds to the anatomical variance of the European population. According to the measurements, osteosynthesis was possible (class A) in 54 out of 75 cases (72%) based on the evaluation matrix. All screws were more than 5 mm from the fracture line, so that the risk of fixation failure would be minimised. 11 cases (15%) fell into the category of difficult implantation (class B), because the distance between at least one screw and the fracture line was less than 5 mm. In 10 cases (13%), an implantation was not advisable (class C), because at least one of the screws was positioned beyond the fracture line. The implantation measurements differed considerably for the three examined nails is shown in Table 1. An examination of the Class B and C implantations showed that in all cases either the screw for the lesser tubercle was too inferior or that there was perforation of the intertubercular groove by one of the proximal screws compromising the long biceps tendon. The distribution of each nail can be seen in Table 2. The screw in the lesser tubercle is likely to be positioned in the tubercle’s lower part. The risk of the screw to cut out is enhanced.
Implant
Possible (A)
Difficult (B)
Impossible (C)
Targon T2 TriGen
24 15 15
1 7 3
– 3 7
Total
54
11
10
In order to optimise the positioning of this screw we measured the distance of the screw to the fracture line. The mean distances for each nail can be seen in Table 3. The statistical analysis of the data revealed a significant correlation between the size of the humerus as represented by the diameter of the anatomical neck and nail implantability (p < 0.05). The larger the humeral head, the more likely a Class A implantation. This relationship could be correlated in graphical form. The analysis for the T2-PHN and the TriGen-PHN with the specimens used in the present study showed that, with a sensitivity of 100% and a specificity of 80%, either implant will be successful for a humerus size of 42.2 mm or more (Fig. 4).
Discussion Good results have been reported for the treatment of proximal humeral fractures with an intramedullary implant [12–17]. However, some problems remain with both fragment fixation and postoperative stability [12,14,19]. The present study was conducted to examine the angulation and direction of the proximal screws for three intramedullary implants and to evaluate their fragment fixation using a collection of 25 cadaver humeri. The use of a classification for humeral head fractures generally accepted in specialist literature [3,21] improves clinical utility when considering proximal humerus fractures [14,19,21]. There are several classifications for fractures of the humeral head described in the literature. The classification mostly used is the Neer classification [22,24]. It is based on the four-part model described by Codman in 1934 [21]. We used this classification because it is focussed on the fragments of the fracture and modified it according to Resch [3]. He modified the four-parts described by Codman using his clinical experience with a more precise description of the fracture lines. In this way, we were able to perform a precise and reproducible anatomical fracture model, which allowed an exact comparison of the nails. The results we found correspond to the clinical data out of actual literature [19]. Several authors report excellent results in treating even displaced four-part fractures of the proximal humerus with intramedullary nailing, although some difficulties are reported. The virtual model allowed us to examine the herein described four-part fracture and also the two and three part fractures. Treating a two-part fracture, with the fracture line running through the surgical neck, no fixation problems could be obtained. Most authors agree, that two-part fractures can be treated with intramedullary nails [10,13]. In the virtual three-part
Table 2 Localisation for difficult/impossible implantations.
Class B (total) Lesser tubercle Intertubercular groove Class C (total) Lesser tubercle Intertubercular groove
Targon
%
T2
%
TriGen
%
Total
%
1 – 1 – – –
4 – 100 – – –
7 3 6 3 3 1
28 43 86 12 100 33
3 3 1 7 6 2
12 100 33 28 86 29
11 6 8 10 9 3
15 55 73 13 90 30
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
G Model
JINJ-5554; No. of Pages 5 T. Harnoss et al. / Injury, Int. J. Care Injured xxx (2013) xxx–xxx
4 Table 3 Screw distance to fracture line in the lesser tubercle.
Cranial sd Caudal sd
Targon-PHN
T2-PHN
Trigen-PHN
12.5 3.2 9.8 4.1
15.9 3.3 6.9 4.7
17.9 3.3 6.9 4.7
model with a fractured greater tubercle, sufficient fixation could be achieved with all nails although damage of the intertubercular groove was also possible. Relatively good results in treating threepart fractures using intramedullary nails can also be found in current clinical trials [15,19]. In clinical experience, complications in the treatment of four-part fractures are common. In the simulation of four-part fractures many good results could be achieved, although the shortcomings of the intramedullary nails became obvious. Depending on the implant and the patient, it may not be possible to fully engage all fragments with proximal screws. The quality of the fragment fixation differed depending on the humeral anatomy and the implant. Current literature considers an implantation to be safe if there is a distance of at least 5 mm between the screw and the fracture line [23]. If this distance cannot be obtained, the risk of fixation failure is increased. In our results the Targon-PHN screws were most often located within the fracture fragments. Even this nail has exhibited shortcomings as the screw for engaging the lesser tubercle fragment tended to be inserted in an inferior position. The problem becomes even more obvious in the T2-PHN and the TriGen-PHN, which results in a difficult fixation of all available bone fragments. The screw positions near the intertubercular groove also differ between implants. Too close positioning of the screws to the intertubercular groove compromises the long biceps tendon. Again, the best results were achieved by the Targon-PHN, whereas the T2-PHN and particularly the TriGen-PHN performed rather poorly in comparison and exhibited implantation problems in 10 of 25 cases. The screw configuration, in terms of localisation and orientation of the nails could be optimised and validated in further studies. First, the screw fixing the lesser tubercle would lead to a better effect when inserted higher in the humeral head. Second, the screws lying close to the intertubercular grove should be positioned in a way to secure it. The angle between these two screws, namely the screw for the lesser tubercle and the screw inserted most ventral in the greater tubercle should be increased to prevent those screws from perforating into the intertubercular grove. One should keep in mind that in a clinical setting, a comminuted fracture is not uncommon and ideal fixation is often impossible using a standard implant. The modalities evaluated in this study would be easily transferable to actual clinical data evaluation. By choosing a novel approach utilising a virtual implantation system, we were able to examine all three implants in one model and compare them on the basis of a reproducible fracture pattern. In contrast to clinical studies we had the chance to fracture a humerus, implant three different nails in it, and examine the position of the screws in the reduced fracture. Future research will include imaging obtained from patients with complex proximal humerus fractures. In addition it will allow us to place the proximal screws in the safe zone by changing their angulation. The variation of the proximal screw insertional angle allows to optimise the position of the screw in the fracture fragment. This can lead to better load shift and a reduction of cutting out. Further investigation to this subject has to be done virtually and biomechanically with prototypes. In correlation with the diameter of the anatomical neck, the conditions for osteosynthesis were found to become increasingly favourable as the diameter of the anatomical neck increases. This
diameter could easily be determined by evaluating X-rays or CTData of the humerus. This information may allow the surgeon to accurately predict implant capabilities. According to our results, if the diameter of the proximal humerus is larger than 42.2 mm, a successful implantation is possible with a sensitivity of 100% and a specificity of 80% in the examined population. Transferred to the clinical setting a different implant such as angular stable plates, especially polyaxial angular stable plates might be the implant of choice for small patients. Conclusion Excellent results in treating displaced four-part fractures of the proximal humerus with intramedullary nailing are possible, if the positioning of the nail and the proximal screws are in anatomically correct and load relevant stable areas of the fracture fragments. However, some challenges remain with fragment fixation and postoperative stability. The present study was conducted to examine the position of the proximal screws for three intramedullary implants and to evaluate their fragment fixation using a collection of 25 cadaver humeri. We conclude that regarding the anatomy some nailing systems cannot be implantated in humeri with four-part fractures without interference of important anatomical structures like the long biceps tendon. Summarising our results we believe that a proximal humerus nailing system, which allows an individual angulation of the interlocking head screws, would overcome these limitations and in addition have the possibility to address the good bone stock. Conflict of Interest Statement I declare no conflict of interest for the manuscript ‘‘Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures’’ for me and the other authors. Acknowledgments We would like to thank the Dr.-Ing. Leonhard-Lorenz foundation for the financial help and T. Schuster from the Institute for Medical Statistics and Epidemiology of the Technische Universita¨t Mu¨nchen. References [1] Horak J, Nilsson BE. Epidemiology of fracture of the upper end of the humerus. Clin Orthop Relat Res 1975;(112):250–3. [2] Court-Brown CM, Garg A, McQueen MM. The epidemiology of proximal humeral fractures. Acta Orthop Scand 2001;72(4):365–71. [3] Resch H. Fractures of the humeral head. Unfallchirurg 2003;106(8):602–17. [4] Gerber C, Schneeberger AG, Vinh TS. The arterial vascularization of the humeral head. An anatomical study. J Bone Joint Surg Am 1990;72(10): 1486–94. [5] Kontakis G, Koutras C, Tosounidis T, Giannoudis P. Early management of proximal humeral fractures with hemiarthroplasty: a systematic review. J Bone Joint Surg Br 2008;90(11):1407–13. [6] Helmy N, Hintermann B. New trends in the treatment of proximal humerus fractures. Clin Orthop Relat Res 2006;442:100–8. [7] Vallier HA. Treatment of proximal humerus fractures. J Orthop Trauma 2007;21(7):469–76. [8] Sudkamp N, Bayer J, Hepp P, Voigt C, Oestern H, Ka¨a¨b M, et al. Open reduction and internal fixation of proximal humeral fractures with use of the locking proximal humerus plate. Results of a prospective, multicenter, observational study. J Bone Joint Surg Am 2009;91(6):1320–8. [9] 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. [10] 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.
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
G Model
JINJ-5554; No. of Pages 5 T. Harnoss et al. / Injury, Int. J. Care Injured xxx (2013) xxx–xxx [11] 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. [12] Adedapo AO, Ikpeme JO. The results of internal fixation of three- and four-part proximal humeral fractures with the Polarus nail. Injury 2001;32(2):115–21. [13] Agel J, Jones CB, Sanzone AG, Camuso M, Henley MB. Treatment of proximal humeral fractures with Polarus nail fixation. J Shoulder Elbow Surg 2004; 13(2):191–5. [14] Lin J. Effectiveness of locked nailing for displaced three-part proximal humeral fractures. J Trauma 2006;61(2):363–74. [15] Mathews J, Lobenhoffer P. Results of the provision of unstable proximal humeral fractures in geriatric patients with a new angle stabilizing antegrade nail system. Unfallchirurg 2004;107(5):372–80. [16] Tamai K, Ohno W, Takemura M, Mashitori H, Hamada J, Saotome K. Treatment of proximal humeral fractures with a new intramedullary nail. J Orthop Sci 2005;10(2):180–6. [17] van den Broek CM, van den Besselaar M, Coenen JM, Vegt PA. Displaced proximal humeral fractures: intramedullary nailing versus conservative treatment. Arch Orthop Trauma Surg 2007;127(6):459–63.
5
[18] Gradl G, Dietze A, Ka¨a¨b M, Hopfenmu¨ller W, Mittlmeier T. Is locking nailing of humeral head fractures superior to locking plate fixation? Clin Orthop Relat Res 2009;467(11):2986–93. [19] Mittlmeier TW, Stedtfeld HW, 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-A(Suppl. 4):136–46. [20] Mall G, Hubig M, Bu¨ttner A, Kuznik J, Penning R, Graw M. Sex determination and estimation of stature from the long bones of the arm. Forensic Sci Int 2001;117(1–2):23–30. [21] Codman EA. The shoulder: rupture of the supraspinatus tendon and other lesions in or about the subacromial bursa; 1934;p. 285–314. [22] Neer II CS. Displaced proximal humeral fractures. I. Classification and evaluation. J Bone Joint Surg Am 1970;52(6):1077–89. [23] Liew AS, Johnson JA, Patterson SD, King GJ, Chess DG. Effect of screw placement on fixation in the humeral head. J Shoulder Elbow Surg 2000;9(5):423–6. [24] Brunner A, Honigmann P, Treumann T, Babst R. The impact of stereo-visualisation of three-dimensional CT datasets on the inter- and intraobserver reliability of the AO/OTA and Neer classifications in the assessment of fractures of the proximal humerus. J Bone Joint Surg Br 2009;91(6):766–71.
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
JINJ-5554; No. of Pages 5 Injury, Int. J. Care Injured xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures Tobias Harnoss a,*, Bernd Felkel b,1, Stefan Do¨bele c,2, Ulrich Schreiber b,3, Andreas Lenich d,4 a
St. Vinzenz Clinic Pfronten, Department for Surgery and Orthopedics, Kirchenweg 15, 87459 Pfronten, Germany Munich University of Applied Sciences, Lothstr. 34, 80335 Munich, Germany c Department of Traumatology and Reconstructive Surgery, BG Unfallklinik, Eberhard Karls Universita¨t Tu¨bingen, Schnarrenbergstraße 95, 72076 Tu¨bingen, Germany d Clinic of Orthopedics and Traumatology, Department for Sport Orthopedics, Klinikum rechts der Isar, Technische Universita¨t Mu¨nchen, Ismaningerstr. 22, 81675 Munich, Germany b
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 8 October 2013
Proximal humerus fractures treated with intramedullary nails show good results. However, the correct anatomical reconstruction of four-part fractures is demanding especially when using intramedullary nails. We therefore compared different intramedullary nail designs for the proximal humerus in a virtual morphological manner. Three commercially available nailing systems where virtually implanted in virtually generated reproducible four-part fractures of 25 digitised humeri. The objective of this study was to quantify and characterise the anatomical position of the proximal screws in the most vulnerable case of a four-part fracture. Taking into account a minimum distance of 5 mm between the screw head and the fracture line, osteosynthesis was possible in 54 out of 75 cases. Difficulties placing the proximal screws could be observed at the localisation of the lower lesser tubercle or/and at the sulcus intertubercularis. This morphological analysis could be the basis for choosing the most sufficient implant intra operatively or even improving the nail design. ß 2013 Elsevier Ltd. All rights reserved.
Keywords: Proximal humerus Fracture Intramedullary nailing Virtual comparison
Introduction About 4% of all fractures in the human body involve the humeral head with severe fractures patterns often seen in the elderly population [1,2]. Non-displaced or only slightly displaced fractures typically are treated non-operatively [3]. There is a certain consensus to treat displaced four-part fractures associated with a high degree of comminution (involving a split head), with primary hemiarthroplasty [3–5]. However, the recommendations for the treatment of displaced three or four-part fractures remain controversial. Recently, there has been a trend to treat proximal humeral fractures with angular stable systems such as plates and intramedullary nails [6–11]. In addition to the treatment with locking plate systems several types of antegrade interlocking nails
* Corresponding author. Tel.: +49 173 3684176; fax: +49 321 21274439. E-mail addresses: [email protected] (T. Harnoss), [email protected] (B. Felkel), [email protected] (S. Do¨bele), [email protected] (U. Schreiber), [email protected] (A. Lenich). 1 Tel.: +49 176 477 288 05; fax: +49 89 1218 18 3718. 2 Tel.: +49 7071 606 1067; fax: +49 7071 606 1002. 3 Tel.: +49 179 5868 236; fax: +49 89 1218 18 3718. 4 Tel.: +49 89 4140 5135; fax: +49 89 4140 4890.
with uniplanar or multiplanar interlocking modes are currently reported to provide effective means for stabilising proximal humeral fractures [12–17]. According to the manufacturer’s instructions, two to four-part fractures may be treated with intramedullary implants. Several recent studies examining the outcome after intramedullary nailing [14,18,19] noted indications including treatment of displaced two and three-part fractures [14,19]. However, four-part fractures were also treated with the intramedullary implants and yielded relatively good results [14,18,19]. Shortcomings of intramedullary nailing include cutting out of the proximal screws and difficulties in a secure fixation of the bone fragments [14,18,19]. For this reason, we examined the nail design morphologically with a focus on screw configuration, i.e. arrangement, position and orientation of the proximal screws, using a new computer based imaging method. We especially focused on how the osteosynthesis is influenced by the anatomical variance, which can be seen in the European population. We used radiographic landmarks for evaluating the anatomical variance, which can also be seen on images taken preoperatively. This could lead to a better appraisal of operative results. Overall, we compared commercially available intramedullary nail systems used for multi-part proximal humerus fractures. Cadaveric humeri corresponding to the
0020–1383/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.injury.2013.10.039
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
G Model
JINJ-5554; No. of Pages 5 T. Harnoss et al. / Injury, Int. J. Care Injured xxx (2013) xxx–xxx
2
normal distribution in the European population [20] were used to create a virtual model for implant evaluation. Proceeding in that order, it was possible to acquire implantation data for all three nails in all cadaveric specimens. Materials and methods In total 25 fresh frozen cadaver humeri (43–88 years, mean age: 65 years, 16 male, 9 female) were digitised using computer tomography (Cardiac Sensation, Siemens, Erlangen, Germany, 120 kV, 200 mAs, layer thickness of 1 mm) and then reconstructed in 3D using a program specifically developed for medical morphological analyses (Amira, Mercury Computer Systems GmbH, Berlin, Germany). In order to classify the cadaveric specimens anatomically, the software was used to measure the following anatomical landmarks: 1. Diameter of the anatomical neck 2. Diameter of the humeral head 3. Diameter of the shaft Three intramedullary nails currently available for the treatment of proximal humeral fractures, Targon-PHN (left and right version), T2-PHN (left and right version, bend setup) and TriGen-PHN (unilateral setup), were digitised with the help of a CAD system (Catia V5 R14, Dassault Systems, Suresnes Cedex, France). For clarity reasons, the screws were replaced by short bolts allowing analysis of the proximal screw orientations (see Fig. 1). The Amira program was used to simulate a virtual four-part fracture in the humeri using the four-part fracture according to Codman [21] as a basic model and modifying it according to Resch [3]: The fracture line between the two tubercles does not run in the intertubercular groove, but about 5 mm more laterally. The resulting four fragments are the same as described by Neer [22]: a lesser tubercle fragment, a greater tubercle fragment, the remaining shaft the humeral head with an intact anatomical neck and a fracture at the surgical neck (see Fig. 1). The nails were virtually implanted according to the manufacturers’ instructions and recommendations published in recent specialist literature [19,23]. The nail was inserted at the most cranial point of the humeral head along the diaphyseal axis so that the proximal end of the nail came to lie in the subchondral area of the humeral head directly beneath the surface [23]. In that way, the nail did not project from the bone and potential subacromial impingement [19] was avoided. The nail was rotated to optimise the screws’ position in the fragments, i.e. to provide them with a maximum distance from the fragment edge (see Fig. 1). A total of 75 implantations in 25 humeri were conducted in this manner. After the implantation, the position of the implant was evaluated based on selected distances between the fracture lines
Fig. 1. Four-part fracture and implanted nail (Targon-PHN, Aesculap) with safety margins.
Fig. 2. Measurements from the proximal screws (Targon-PHN) to the fracture lines.
and the screws. Distances were measured in 4 directions at each of the tubercles. These were the cranial and caudal direction as well as the medial and lateral on the lesser and the ventral and dorsal direction at the greater tubercle. Current literature considers an implantation to be safe if there is a distance of at least 5 mm between the screw and the fracture line [23]. Therefore a 5 mm band indicated by landmarks (see Fig. 2) was drawn in parallel to the fracture line to visualise this distance. For the evaluation of the osteosynthesis, implantation success was divided into three categories (see Fig. 3): Implantation possible (class A): All inserted screws are at a safe distance of at least 5 mm from the fracture line. Implantation difficult (class B): At least one of the screws is less than 5 mm from a fracture line or the end of a fragment. Implantation impossible (class C): At least one of the screws lies outside its fragment. The normal distribution of the available bone collection was tested with the Kolmogorov–Smirnov-Test and verified based on the epidemiological study [20] by Mall et al. The Mann–Whitney test was performed to see if the implantability depended on the size of the humerus. The next step was to find out whether the available data allowed a prediction of implantability as a function of humerus size using logistic regression and the chi-square test. The results were plotted in a graph showing the probability of a successful implantation as a function of the size (see Fig. 4).
Fig. 3. Evaluation matrix with three categories: implantation possible (A), difficult (B) and impossible (C).
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
G Model
JINJ-5554; No. of Pages 5 T. Harnoss et al. / Injury, Int. J. Care Injured xxx (2013) xxx–xxx
3
Table 1 Implantability of the nails, absolute number (n = 75).
Fig. 4. Implantability as a function of the diameter of the anatomical neck – circles indicate failed (0,00) and successful (1,00) implantations and the accordant diameter of the anatomical neck.
Sensitivity and specificity were evaluated by ROC analysis. A pvalue of less than 0.05 was set as the significance level.
Results The cadaveric specimens represented a normal distribution with respect to anatomical features. According to Mall, the mean diameter of the anatomical neck is 4.70 cm (sd = 0.27). The diameter of the anatomical neck is used as a reference for the humerus size [20,23]. The collection of 25 humeri available for the present study had a mean diameter of the anatomical neck of 4.62 cm (sd = 0.52), which corresponds to the anatomical variance of the European population. According to the measurements, osteosynthesis was possible (class A) in 54 out of 75 cases (72%) based on the evaluation matrix. All screws were more than 5 mm from the fracture line, so that the risk of fixation failure would be minimised. 11 cases (15%) fell into the category of difficult implantation (class B), because the distance between at least one screw and the fracture line was less than 5 mm. In 10 cases (13%), an implantation was not advisable (class C), because at least one of the screws was positioned beyond the fracture line. The implantation measurements differed considerably for the three examined nails is shown in Table 1. An examination of the Class B and C implantations showed that in all cases either the screw for the lesser tubercle was too inferior or that there was perforation of the intertubercular groove by one of the proximal screws compromising the long biceps tendon. The distribution of each nail can be seen in Table 2. The screw in the lesser tubercle is likely to be positioned in the tubercle’s lower part. The risk of the screw to cut out is enhanced.
Implant
Possible (A)
Difficult (B)
Impossible (C)
Targon T2 TriGen
24 15 15
1 7 3
– 3 7
Total
54
11
10
In order to optimise the positioning of this screw we measured the distance of the screw to the fracture line. The mean distances for each nail can be seen in Table 3. The statistical analysis of the data revealed a significant correlation between the size of the humerus as represented by the diameter of the anatomical neck and nail implantability (p < 0.05). The larger the humeral head, the more likely a Class A implantation. This relationship could be correlated in graphical form. The analysis for the T2-PHN and the TriGen-PHN with the specimens used in the present study showed that, with a sensitivity of 100% and a specificity of 80%, either implant will be successful for a humerus size of 42.2 mm or more (Fig. 4).
Discussion Good results have been reported for the treatment of proximal humeral fractures with an intramedullary implant [12–17]. However, some problems remain with both fragment fixation and postoperative stability [12,14,19]. The present study was conducted to examine the angulation and direction of the proximal screws for three intramedullary implants and to evaluate their fragment fixation using a collection of 25 cadaver humeri. The use of a classification for humeral head fractures generally accepted in specialist literature [3,21] improves clinical utility when considering proximal humerus fractures [14,19,21]. There are several classifications for fractures of the humeral head described in the literature. The classification mostly used is the Neer classification [22,24]. It is based on the four-part model described by Codman in 1934 [21]. We used this classification because it is focussed on the fragments of the fracture and modified it according to Resch [3]. He modified the four-parts described by Codman using his clinical experience with a more precise description of the fracture lines. In this way, we were able to perform a precise and reproducible anatomical fracture model, which allowed an exact comparison of the nails. The results we found correspond to the clinical data out of actual literature [19]. Several authors report excellent results in treating even displaced four-part fractures of the proximal humerus with intramedullary nailing, although some difficulties are reported. The virtual model allowed us to examine the herein described four-part fracture and also the two and three part fractures. Treating a two-part fracture, with the fracture line running through the surgical neck, no fixation problems could be obtained. Most authors agree, that two-part fractures can be treated with intramedullary nails [10,13]. In the virtual three-part
Table 2 Localisation for difficult/impossible implantations.
Class B (total) Lesser tubercle Intertubercular groove Class C (total) Lesser tubercle Intertubercular groove
Targon
%
T2
%
TriGen
%
Total
%
1 – 1 – – –
4 – 100 – – –
7 3 6 3 3 1
28 43 86 12 100 33
3 3 1 7 6 2
12 100 33 28 86 29
11 6 8 10 9 3
15 55 73 13 90 30
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
G Model
JINJ-5554; No. of Pages 5 T. Harnoss et al. / Injury, Int. J. Care Injured xxx (2013) xxx–xxx
4 Table 3 Screw distance to fracture line in the lesser tubercle.
Cranial sd Caudal sd
Targon-PHN
T2-PHN
Trigen-PHN
12.5 3.2 9.8 4.1
15.9 3.3 6.9 4.7
17.9 3.3 6.9 4.7
model with a fractured greater tubercle, sufficient fixation could be achieved with all nails although damage of the intertubercular groove was also possible. Relatively good results in treating threepart fractures using intramedullary nails can also be found in current clinical trials [15,19]. In clinical experience, complications in the treatment of four-part fractures are common. In the simulation of four-part fractures many good results could be achieved, although the shortcomings of the intramedullary nails became obvious. Depending on the implant and the patient, it may not be possible to fully engage all fragments with proximal screws. The quality of the fragment fixation differed depending on the humeral anatomy and the implant. Current literature considers an implantation to be safe if there is a distance of at least 5 mm between the screw and the fracture line [23]. If this distance cannot be obtained, the risk of fixation failure is increased. In our results the Targon-PHN screws were most often located within the fracture fragments. Even this nail has exhibited shortcomings as the screw for engaging the lesser tubercle fragment tended to be inserted in an inferior position. The problem becomes even more obvious in the T2-PHN and the TriGen-PHN, which results in a difficult fixation of all available bone fragments. The screw positions near the intertubercular groove also differ between implants. Too close positioning of the screws to the intertubercular groove compromises the long biceps tendon. Again, the best results were achieved by the Targon-PHN, whereas the T2-PHN and particularly the TriGen-PHN performed rather poorly in comparison and exhibited implantation problems in 10 of 25 cases. The screw configuration, in terms of localisation and orientation of the nails could be optimised and validated in further studies. First, the screw fixing the lesser tubercle would lead to a better effect when inserted higher in the humeral head. Second, the screws lying close to the intertubercular grove should be positioned in a way to secure it. The angle between these two screws, namely the screw for the lesser tubercle and the screw inserted most ventral in the greater tubercle should be increased to prevent those screws from perforating into the intertubercular grove. One should keep in mind that in a clinical setting, a comminuted fracture is not uncommon and ideal fixation is often impossible using a standard implant. The modalities evaluated in this study would be easily transferable to actual clinical data evaluation. By choosing a novel approach utilising a virtual implantation system, we were able to examine all three implants in one model and compare them on the basis of a reproducible fracture pattern. In contrast to clinical studies we had the chance to fracture a humerus, implant three different nails in it, and examine the position of the screws in the reduced fracture. Future research will include imaging obtained from patients with complex proximal humerus fractures. In addition it will allow us to place the proximal screws in the safe zone by changing their angulation. The variation of the proximal screw insertional angle allows to optimise the position of the screw in the fracture fragment. This can lead to better load shift and a reduction of cutting out. Further investigation to this subject has to be done virtually and biomechanically with prototypes. In correlation with the diameter of the anatomical neck, the conditions for osteosynthesis were found to become increasingly favourable as the diameter of the anatomical neck increases. This
diameter could easily be determined by evaluating X-rays or CTData of the humerus. This information may allow the surgeon to accurately predict implant capabilities. According to our results, if the diameter of the proximal humerus is larger than 42.2 mm, a successful implantation is possible with a sensitivity of 100% and a specificity of 80% in the examined population. Transferred to the clinical setting a different implant such as angular stable plates, especially polyaxial angular stable plates might be the implant of choice for small patients. Conclusion Excellent results in treating displaced four-part fractures of the proximal humerus with intramedullary nailing are possible, if the positioning of the nail and the proximal screws are in anatomically correct and load relevant stable areas of the fracture fragments. However, some challenges remain with fragment fixation and postoperative stability. The present study was conducted to examine the position of the proximal screws for three intramedullary implants and to evaluate their fragment fixation using a collection of 25 cadaver humeri. We conclude that regarding the anatomy some nailing systems cannot be implantated in humeri with four-part fractures without interference of important anatomical structures like the long biceps tendon. Summarising our results we believe that a proximal humerus nailing system, which allows an individual angulation of the interlocking head screws, would overcome these limitations and in addition have the possibility to address the good bone stock. Conflict of Interest Statement I declare no conflict of interest for the manuscript ‘‘Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures’’ for me and the other authors. Acknowledgments We would like to thank the Dr.-Ing. Leonhard-Lorenz foundation for the financial help and T. Schuster from the Institute for Medical Statistics and Epidemiology of the Technische Universita¨t Mu¨nchen. References [1] Horak J, Nilsson BE. Epidemiology of fracture of the upper end of the humerus. Clin Orthop Relat Res 1975;(112):250–3. [2] Court-Brown CM, Garg A, McQueen MM. The epidemiology of proximal humeral fractures. Acta Orthop Scand 2001;72(4):365–71. [3] Resch H. Fractures of the humeral head. Unfallchirurg 2003;106(8):602–17. [4] Gerber C, Schneeberger AG, Vinh TS. The arterial vascularization of the humeral head. An anatomical study. J Bone Joint Surg Am 1990;72(10): 1486–94. [5] Kontakis G, Koutras C, Tosounidis T, Giannoudis P. Early management of proximal humeral fractures with hemiarthroplasty: a systematic review. J Bone Joint Surg Br 2008;90(11):1407–13. [6] Helmy N, Hintermann B. New trends in the treatment of proximal humerus fractures. Clin Orthop Relat Res 2006;442:100–8. [7] Vallier HA. Treatment of proximal humerus fractures. J Orthop Trauma 2007;21(7):469–76. [8] Sudkamp N, Bayer J, Hepp P, Voigt C, Oestern H, Ka¨a¨b M, et al. Open reduction and internal fixation of proximal humeral fractures with use of the locking proximal humerus plate. Results of a prospective, multicenter, observational study. J Bone Joint Surg Am 2009;91(6):1320–8. [9] 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. [10] 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.
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
G Model
JINJ-5554; No. of Pages 5 T. Harnoss et al. / Injury, Int. J. Care Injured xxx (2013) xxx–xxx [11] 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. [12] Adedapo AO, Ikpeme JO. The results of internal fixation of three- and four-part proximal humeral fractures with the Polarus nail. Injury 2001;32(2):115–21. [13] Agel J, Jones CB, Sanzone AG, Camuso M, Henley MB. Treatment of proximal humeral fractures with Polarus nail fixation. J Shoulder Elbow Surg 2004; 13(2):191–5. [14] Lin J. Effectiveness of locked nailing for displaced three-part proximal humeral fractures. J Trauma 2006;61(2):363–74. [15] Mathews J, Lobenhoffer P. Results of the provision of unstable proximal humeral fractures in geriatric patients with a new angle stabilizing antegrade nail system. Unfallchirurg 2004;107(5):372–80. [16] Tamai K, Ohno W, Takemura M, Mashitori H, Hamada J, Saotome K. Treatment of proximal humeral fractures with a new intramedullary nail. J Orthop Sci 2005;10(2):180–6. [17] van den Broek CM, van den Besselaar M, Coenen JM, Vegt PA. Displaced proximal humeral fractures: intramedullary nailing versus conservative treatment. Arch Orthop Trauma Surg 2007;127(6):459–63.
5
[18] Gradl G, Dietze A, Ka¨a¨b M, Hopfenmu¨ller W, Mittlmeier T. Is locking nailing of humeral head fractures superior to locking plate fixation? Clin Orthop Relat Res 2009;467(11):2986–93. [19] Mittlmeier TW, Stedtfeld HW, 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-A(Suppl. 4):136–46. [20] Mall G, Hubig M, Bu¨ttner A, Kuznik J, Penning R, Graw M. Sex determination and estimation of stature from the long bones of the arm. Forensic Sci Int 2001;117(1–2):23–30. [21] Codman EA. The shoulder: rupture of the supraspinatus tendon and other lesions in or about the subacromial bursa; 1934;p. 285–314. [22] Neer II CS. Displaced proximal humeral fractures. I. Classification and evaluation. J Bone Joint Surg Am 1970;52(6):1077–89. [23] Liew AS, Johnson JA, Patterson SD, King GJ, Chess DG. Effect of screw placement on fixation in the humeral head. J Shoulder Elbow Surg 2000;9(5):423–6. [24] Brunner A, Honigmann P, Treumann T, Babst R. The impact of stereo-visualisation of three-dimensional CT datasets on the inter- and intraobserver reliability of the AO/OTA and Neer classifications in the assessment of fractures of the proximal humerus. J Bone Joint Surg Br 2009;91(6):766–71.
Please cite this article in press as: Harnoss T, et al. Virtual morphological comparison of three intramedullary nailing systems for the treatment of proximal humeral fractures. Injury (2013), http://dx.doi.org/10.1016/j.injury.2013.10.039
