文献翻译 一种新的肱骨近端三部分骨折内固定技术
推荐这个帖子发布于 14 年零 286 天前，其中的信息可能已发生改变或有所发展。
这篇文章还比较有想法，作者将髓内钉与锁定钢板结合起来，试着翻译了一下全文，明显感觉生物力学方面的内容比较难懂，翻译起来很费力费时，有些地方的句式很拗口，比较难理解，有些地方需停顿下来好好思量一番，有些地方仅是做了一下意译，文中应该有很多错误的地方，请战友指正，特别是生物力学方面的内容，我自已很有必要提高一下。
Biomechanical evaluation of a new fixation technique for internal fixation of three-part proximal humerus fractures in a novel cadaveric model
一种新的肱骨近端三部分骨折内固定技术在新鲜尸体模型上的生物力学评估
a b s t r a c t
Background: The optimal surgical treatment for displaced proximal humeral fractures is still controversial. A new implant for the treatment of three-part fractures has been recently designed. It supplements the existing Expert Humeral Nail with a locking plate. We developed a novel humeral cadaveric model and the existing implant and the prototype were biomechanically compared to determine their ability in maintaining interfragmentary stability.
Methods: The bone mineral density of eight pairs of cadaveric humeri was assessed and a three-part proximal humeral fracture was simulated with a Greater Tuberosity osteotomy and a surgical neck wedge ostectomy. The specimens were randomly assigned to either treatment. A bone anchor simulated part of a rotator cuff tendon pulling on the Greater Tuberosity. Specimens were initially tested in axial compression and afterward with a compound cyclic load to failure. An optical 3D motion tracking system continuously monitored the relative interfragmentary movements.
Findings: The specimen stabilized with the prototype demonstrated higher stiffness (P=0.036) and better interfragmentary stability (P values＜0.028) than the contralateral treated with the existing implant. There was no correlation between the bone mineral density and any of the investigated variables.
Interpretation: The convenience of this new IM-nail and locking plate assembly must be confirmed in vivo but the current study provides a biomechanical rationale for its use in the treatment of three-part proximal humeral fractures. The improved stability could be advantageous in particular when medial buttress is missing, even in osteoporotic bone.
摘要
背景：移位的肱骨近端骨折的理想手术治疗方案仍有争议。最近出现了一种新的用于治疗肱骨近端三部分骨折的植入物。它在现有的EHN（专家级肱骨髓内钉）上加了块锁定钢板。我们设计了一个新鲜的尸体模型，比较现有的内固定物与该新型内固定物生物力学以确定是否能够维持骨折块间的稳定性。
方法：评估8对尸体肱骨的骨密度，通过大结节截骨和外科颈楔形截骨模拟肱骨近端四部分骨折。标本随机分配进入治疗组。以骨锚模拟附着在大结节上的肩袖肌腱。开始时测量标本的轴向加压，然后为复合循环负荷至内固定失败。3D视觉运动追踪系统连续监测骨折块间的相对运动。
发现：与现有的内固定物相比，这种新的内固定显示出较高的(P=0.036)和较好的骨块间稳定性(P values＜0.028)。骨密度与所观察的变量之间没有任何相关性。解读：虽然还需做体内试验以确定这种髓内和锁定钢板相结合的新型内固定物的优势，但本研究已经为它在肱骨近端三部分骨折治疗中的使用提供了一些生物力学的依据。稳定性的提高使其具有一定优势，特别是当失去内侧支撑，即使是骨质疏松病例中（也具有优势）。
1． Introduction
Fracture of the proximal humerus is the third most common injury in people over 65 years of age after hip and Colles´ fracture (Baron et al., 1996). Most of these fractures are stable and can be successfully treated conservatively (Court-Brown CM et al., 2002; Court-Brown and McQueen, 2004; Hintermann et al., 2000; Lill and Josten, 2001; Szyszkowitz and Schippinger, 1999) but about 20% of the patients need surgical intervention. Nowadays the surgical treatment of two and three-part proximal humeral fractures aims for a stable fixation that preserves the blood supply to the fracture fragments and reduces the risk of shoulder stiffness allowing early joint mobilization (Hintermann et al., 2000; Kwon et al., 2002) but the optimal treatment is still controversial. On one side angular stable plating with locking plates has proven to be an effective treatment option (Brunner et al., 2009; Hirschmann et al., 2007; Plecko and Kraus, 2005; Sudkamp et al., 2009) and minimal invasive technique has become popular (Gallo et al., 2005; Gardner et al., 2004, 2005; Laflamme et al., 2008). On the other side recent data support the use of intramedullary nailing with new implants as promising alternatives (Gradl et al., 2007, 2009; Mihara et al., 2008). Recently a new implant, the Expert Proximal Humeral Nail with Locking Plate for Humeral Nail (LPHN, Synthes GmbH, Oberdorf, Switzerland), was designed for proximal three or four-part humeral fractures. It combines an additional locking plate with the existing geometry of the Expert Proximal Humeral Nail (Expert PHN, Synthes GmbH, Oberdorf, Switzerland) (Fig. 1). The rationale behind the additional plate is based on the assumption that augmenting the existing humeral nailing system optimizes bone purchase and leads to additional stability of the humeral head.
1． 介绍
肱骨近端骨折是65岁以上人群中第三常用的损伤，仅次于髋部骨折和Colles’骨折。这些骨折绝大多数是稳定的，可以保守治疗，但约有20%的患者需要手术治疗。现在，两部分和三部分肱骨近端骨折手术治疗的目标是坚强固定，保护骨折块的血供，减少术后肩关节僵硬的风险，允许早期关节活动，但关于最佳治疗仍有争议。 一方面，角度稳定锁定钢板已经证明是一个有效的治疗方法，并且微创技术已很流行。另一方面，近期的研究数据认为带新的植入物的髓内钉装置是一种具有良好的前景的治疗方案。近期，设计出一种新的内固定，肱骨近端带锁定钢板的专家级髓内钉（LPHN）(图1)，来治疗肱骨近端三部分或者四部分骨折。它将现有的专家级肱骨近端髓内钉（Expert PHN）与另一块锁定钢板结合起来。通过增加一块锁定钢板，能加固现有的髓内固定系统从而增加了骨的把持力，增加肱骨头稳定性。
The purpose of this study is to assess if the new implant provides more interfragmentary stability to a three-part proximal humeral fractures when compared to the standard implant in a novel cadaveric model. The information retrieved aims at providing a biomechanical basis for the use of combined plate-nail implants in the treatment of multi-fragmentary proximal humerus fractures.
本研究的目的是在新鲜尸体模型上比较此种新型内固定与标准内固定物，评估前者能否增加肱骨近端三部分骨折骨折块间的稳定性。本实验的目的是为髓内钉-钢板联合系统在肱骨近端多部分骨折中的应用提供生物力学基础。
2. Methods
Eight pairs of fresh frozen skeletally mature human cadaveric humeri were obtained and anterior–posterior X-rays were performed to exclude defects affecting the bone structure integrity. The bones were frozen at −20 °C until further processing.
2.方法：
准备八对骨骼成熟的新鲜冰冻尸体的肱骨，摄前后位片排除影响骨骼完整性的缺陷。骨骼在-20°C下保存备用。
2.1. Bone mineral density measurement The specimens underwent high-resolution bone densitometry measurement (BMD, mg HA/cm3, 123 μm resolution) using threedimensional peripheral quantitative computed tomography (pQCT, XtremeCT, Scanco Medical, Brüttisellen, Switzerland). The volume of interest included those areas where the proximal parts of both implants were to be placed during the following instrumentation.
2.1骨密度测量 使用高分辨率骨密度仪用三维外周量化CT（PQCT）法测量标本的骨密度。测量区域包括术中内固定在肱骨近端植入的区域。2.2. Prototype features
Similarly to the Expert PHN the LPHN provides a fixed-angle blade in the humeral head and an intramedullary load carrier that controls micro movements at the metaphysis. The LPHN has a 23 mm large square medical stainless steel additional plate restrained to the blade through an angular stable locking system. The plate features at each corner a hole hosting a divergent angular stable locking screw.
2.2新的内固定特点
与专家级肱骨近端髓内钉相似，LPHN在肱骨头处具有角度稳定性，同时其应力是髓内传导，可以控制干骺端的微动。LPHN具有一个23mm的方形钢板，其通过锁定系统固定于髓内钉的近端锁钉。该钢板的四个角都可以容纳一枚锁定螺钉。
2.3. The biomechanical model
2.3生物力学模型
2.3.1. Specimen preparation Prior to instrumentation, the specimens were thawed overnight at 4 °C in a refrigerator. During preparation, each specimen was covered with a towel soaked with saline solution and was periodically sprayed to prevent desiccation. The bone pairs were randomly assigned to treatment with either the Expert PHN or the LPHN. The order in which the fixation was applied (Expert PHN versus LPHN) was similarly randomized, but both specimens in each pair were instrumented on the same day always.
2.3.1标本准备
在行内固定前一晚，标本在4 °C冰箱内解冻。制备过程中，每具标本都以浸有盐液的毛巾覆盖，时常洒水以防止干燥。随机将成对的标本分配进入Expert PHN或者LPHN组，先进行Expert PHN内固定还是LPHN也是随机的，但同一对标本总是在同一天进行内固定。
All sampleswere prepared following the samefive steps: 1) marking the reference points and curves on the surface of the bone, 2) performing humeral neck osteotomy creating a lateral fragment (Greater Tuberosity), 3) fixing a custom-made anchor to the lateral–posterior side of this same fragment, 4) anatomical reduction of the fragments and fixation with the Expert PHNor LPHNimplant, 5) humeral surgical neck wedge ostectomy creating the humeral head fragment 6) standardization of the shaft length.
所有标本按以下五步制备：1）在标本表面画出参考点及曲线 2）行肱骨颈截骨造成外侧骨折（大结节）3）将定制的骨锚固定至该骨块的后外侧 4)解剖复位骨折块，Expert PHN或者LPHN固定 5)肱骨外科颈楔形截骨以造成肱骨头骨折块 6)骨干长度标准化 Step 1. The following two groups of reference points and curves were drawn on each humerus using a skin marker (Devon® Skin Marker, TYCO Healthcare LANE COVE, NSW 2066 AUSTRALIA): Greater Tuberosity's reference curves and Surgical Neck's reference curves (Fig. 2). Step 2. A microsagittal saw (Synthes Saw Blade 0.4 mm thick, Synthes GmbH 4500 Solothurn, Switzerland) was used to create a lateral fragment, including the Greater Tuberosity, following the reference curves. Step 3. A custom-made anchor consisting of a 2 mm bolt and two 9 mm external diameter washers was connected to the lateral–posterior side of each Greater Tuberosity fragment after predrilling with a 2 mm drill tip (Fig. 3). A nut and one washer laid outside of the fragment and the screw's head and the second washer inside. In order to improve the holding capability, each anchor was augmented using PMMA (Polymethylmethacrylate, Beracryl, W. Troller AG, Switzerland). Once the anchor was in place the bolt was tightened once leading the inner washer to compact the cancellous bone underneath. The bolt was then released and two 7 metric sutures (TI•CRON 5, Syneture, Covidien Norwalk, CT 06856 USA) were independently connected with surgical knots to that part of the bolt protruding out of the bone and located under the external washer. One milliliter of PMMA at the later hardening stage was applied underneath the internal washer, towards the compacted cancellous bone. The bolt was then retightened and the exceeding PMMA removed. Care was taken not to have any part of the augmented anchor protruding medial to the adjacent fracture plane in order to avoid any additional nonphysiological friction between fragments. Step 4. The randomly assigned implant (Expert PHN or LPHN) was fully fixed following the manufacturer's recommendations including two parallel angular stable locking screws distally. Step 5. The microsagittal oscillating saw was used to perform a surgical neck wedge ostectomy following the most distal and the 15° medially oriented Surgical Neck's reference curves. Care was taken to remove any contact point between the humeral head's fragment and the shaft. Step 6. All the sampleswere reduced to an equal length cutting the shaft perpendicular to its axis 70 mm distal to the center of the most distal angular stable nail screw. All the procedures, including drawings, osteotomy, augmentation, implant fixation and ostectomy, were performed by the same surgeon (M.P.). Fluoroscopy was used to check the length and position of the implants at each step. Each humeral diaphysis was then embedded in the PMMA perpendicular to the ground in a cylindrical stainless steel base-plate up to 40 mm distal to the center of the most distal angular stable nail screw.
步骤一：以皮肤记号笔在肱骨上标出以下两组点和曲线:大结节参考线及外科颈参考线（图2）。步骤二：以小摆锯沿参考线截骨，形成外侧包含大结节的骨折块。步骤三：以2mm钻头预钻孔，将定制的骨锚（由一个直径2mm螺栓及2枚外径为9mm的垫片组成）固定至大结节骨块的后外侧（图3）。一枚螺母及垫片留在骨块外面，螺钉头及另一枚垫片在骨块里面。为了增加把持力，每个螺钉都以骨水泥加固。一旦骨锚到达预定位置，收紧螺栓，挤压垫片间的松质骨。松开螺栓，用两根7#metric缝合线以外科结固定至螺栓突出骨外并位于垫片下的部分。将1mm凝固后期的骨水泥注入到骨块内垫片的下方，再次收紧螺栓，清除溢出的骨水泥。注意防止骨水泥突向内侧相邻的骨块以避免一切非生理的摩擦。步骤四：按照操作手册的要求以Expert PHN或者LPHN固定，包括远端两枚平行的角度稳定的锁定钉。步骤五：沿最远端参考线和向内侧及头侧成角15°的参考线，以小摆锯作外科颈楔形骨切除。注意将肱骨头骨块与肱骨干骨块间的所有接触部分清除。步骤六：所有标本都复位至相同长度，在最远端锁钉中心远端70mm处，垂直肱骨干轴线截断肱骨。所有操作，包括画线，截骨，骨水泥增强，内固定，骨切除，都由同一医生完成。（M.P）.每一步操作都有X光透视确定内固定的长度及位置。每根肱骨干都垂直地面埋入盛骨水泥的圆柱形不锈钢底盘中，底盘距离最远端锁钉的中心40mm。
2.3.2. Test set up
Biomechanical testing was performed on a biaxial servo-hydraulic machine, MTS Mini Bionix II 858 (MTS Systems Corp., Eden Prairie, MN) using a 4kN/20 Nm load cell (HUPPERT 6, HUPPERT GmbHHerrenberg, Germany). Each sample was restrained to the machine with a 25° lateral angulation (Fig. 4) according to the measurement range of Bergmann et al. (2007). Distally a cardan joint connected to the stainless steel base-plate restrained all the sample displacements and the rotation around the shaft axis. The load was applied on the samples by means of a custom-made PMMA cup. An even load distribution on the humeral head was provided via a cavity created in the PMMA cup as negative of a plastic humeral head (5010 Humerus, SYNBONE AG, 7208 Malans, Switzerland) simulating the glenoid. Relative rotation of the cup and the humeral head was prevented applying sand paper strips on that part of the cup in contact with the sample. An adequate notch created in the cup, positioned above the nail entry point, always guaranteed free movements to the nail during testing. The PMMA cup was joined to the machine actuator by means of a custom-made flange and transmitted to the samples the axial load only. Rotational bearing associated with the custom-made flange in fact allowed decoupling the effect of the actuator axial and torsional movements. Once the sample was preloaded, rotations of the actuator did not elicit rotations of the PMMA cup. The torsional actuator was then used to independently generate the force pulling on the Greater Tuberosity fragment via a cable. Fitted in a groove, one side of the cable was anchored to the flange by means of a screw. The cable was passed through two pulleys during testing and connected to the two sutures attached to the augmented bone anchor. The position of the first pulley was such that the pulling direction of the first end of the cable was always tangent to the flange and perpendicular to its axis of rotation. The position of the second pulley was such that the initial direction of the pulling force acting on the anchor inserted in the Greater Tuberosity fragment was angled 110° to the humeral shaft axis in the medial-lateral plane.
2.3.2试验设计
生物力学测试在一个双轴伺服液压器MTS 迷你BionixⅡ858 上进行，使用 4kN/20 Nm 的传感器。根据Bergmann等的测量范围，每个标本都固定于机器上，向外侧成角25°（图4）。远端一个连接在不锈钢底盘上的万向关节限制了标本所有的移位和绕轴旋转。通过一个定制的骨水泥杯将负荷施加于样本。在骨水泥上构成一个如同挖除了肱骨头的臼杯以模拟肩胛盂，从而以使分布于肱骨头表面的力更均匀。用砂皮打磨骨水泥杯与肱骨头的接触部位以防止杯与肱骨头之间的相对旋转。在位于髓内钉进针点上方的骨水泥杯的位置开个槽，以保证在测试过程中骨水泥杯可自由地向髓内钉方向移动。骨水泥杯通过一个定制的轮缘连接至机器的执行器，只向标本传递轴向应力。旋转轴承加上定制的轮缘，实际上就可以将执行器的运动分解成轴向和扭转运动。当预载了标本，执行器的旋转不会引起骨水泥杯旋转。旋转执行器产生独立的拉力通过一根绳作用于大结节。绳装于沟槽内，它的一端通过一枚螺钉固定于轮缘。试验中，绳子通过两个（带滑轮的）把手，并与连接在大结节骨栓上的缝线相连接。一个把手的位置要保证绳的一头总是与轮缘相切，与其旋转轴垂直。另一个把手要使得开始作用于骨栓的力与肱骨干在冠状面上成110°角。
2.3.3. Loading patterns
All specimens were tested according to the following loading patterns:
1. Axial stiffness test: axial loading until 200 N in displacement control (0.02 mm/s).
2. Compound cyclic load till failure: increasing axial compression load, applied at 1 Hz, in phase with pulling on the bone anchor. The axial compression was a filtered interpolation of the pattern recorded by Bergmann et al. (2007) on an instrumented shoulder prosthesis during flexion movement. Starting from 200 N the maximum (peak) of the curve was cyclically increased (Windolf et al., 2009) (0.05 N/cycle) until failure while its minimum (valley) was kept constant at 50 N during the entire test. The pulling force acting on the anchor was controlled in angular displacement control and torque detectors between 1 and 3 Nm. This torque range produced a pulling force on the anchor ranging from 40 to 100 N. The test was automatically stopped when 8 mm actuator axial displacement was reached. At this stage the wedge ostectomy was almost completely closed and every fixation clearly failed.
2.3.3载荷模式
所有标本按以下模式测试
1.轴向刚度试验:在控制移位(0.02 mm/s).轴向加载至200N。
2. 复合循环负荷直到内固定失败：增加轴向载荷，加载频率为1Hz，牵拉骨栓的频率也是1Hz。轴向加压是采用Bergmann等在肩关节假体屈曲运动时所用方法的改良。从200N开始，曲线最大值（峰值）循环增加直到固定失败，其最小值（谷值）始终是50N.通过控制成角并将扭矩控制在1-3NM之间来控制作用于骨栓的力。该范围内的扭矩在骨栓上产生的作用力为40-100N。当执行器轴向位移达到8mm时，试验自动结束。这时，楔形骨间隙已经几乎完全闭合，所有的内固定都失败了。
2.4. Data collection
2.4.1. Data machine Displacement and load were recorded from the MTS controllers at 64 Hz. 2.4.2. Fluoroscopy
At the beginning of the test and every 500 cycles until failure the machine observed a two second pause in unloading (valley) condition (50 N axial load, 1 Nm torque) in order to perform a fluoroscopic assessment (43 kV, 10 mAs — Siemens, ARCO SI 100). The assessment was performed with focus on the identification of the failure mechanism and to associate the loading history with the loss of fragments fixation.
2.4.3. Measurement of the interfragmentary motion During testing an optical 3D motion tracking system consisting of 5 ProReflexMCU digital cameras (Qualisys AB, Gothenburg, Sweden)was used to continuously track the three-dimensional relative motions of the fracture fragments. Reflective marker sets were attached to the humeral shaft, the humeral head and the Greater Tuberosity. Prior to each test the tracking systemwas calibrated and a reference coordinate systemwas defined togetherwith the position of themostmedial aspect of the surgical neck ostectomy, taken as reference point on the humeral head (Fig. 4).
2.4数据采集
2.4.1
从MTS的控制器记录移位及载荷。
2.4.2透视
试验开始时，以及直到内固定失败前，每500个循环机器停2秒，在最小载荷，最小扭矩下透视一次。评估的焦点是内固定失败的机制以及固定失败与所加载荷的关系。
2.4.3骨块间移位的测量
测试中，使用视觉3D运动追踪系统连续追踪骨折块的三维相对运动。反应性标记点固定于肱骨干、肱骨头和大结节。第次测试前，追踪系统都需校准，以楔形截骨时外科颈最内侧作为肱骨头的参考点，设置参考坐标系统（图4）
2.5. Parameters of interest
The construct stiffness was determined between 50 and 100 N on the load-displacement curve via the machine data during the first test. Matching those fluoroscopic images showing the first signs of construct failure with the motion tracking data an arbitrary displacement as failure criterion was defined. The following two datasets of variables were calculated at the failure criterion, at 5000 and at 10000 cycles from the motion tracking data in the reference coordinate system: Humeral Head dataset and Greater Tuberosity fragment dataset (Fig. 5).
The former were the number of cycles until the failure criterion and the humeral head displacements and rotations at 5000 and 10000 cycles. The displacements were calculated along the shaft axis (x axis) and the rotations were calculated with respect to an anterior– posterior axis passing through the humeral head and perpendicular to the shaft axis (y axis) in unloading conditions (valley).
The latter were the rotations of the Greater Tuberosity fragment around a medial-lateral axis passing through the humeral head and perpendicular to the shaft axis (z axis) in unloading condition (valley) at 5000 and 10000 cycles. The 5000th and 10,000th valleys followed the peaks where the samples were respectively axially loaded with 450 N and 700 N in phase with 3 Nm torque applied to the cable. The effect of the BMD on the variables of interest was investigated
2.5感兴趣的参数
第一次测试时，通过机器数据（设定），植入物刚度在载荷-移位曲线上设置为50-100N.综合追踪系统和透视，初步设定内固定失败的移位程度。以下两个变量是根据失败标准，在循环5000和10000次时从视觉运动追踪系统参考坐标的数据计算而来：肱骨头数据集和大结节骨块数据集(图5)。
前者是直到内固定失败标准时循环次数，以及循环5000次和1000次时肱骨头的移位和旋转。移位从X轴上计算，旋转以最小载荷下绕Y轴计算。
后者为循环5000次和1000次时大结节骨块在最小载荷下绕Z轴的旋转。第5000次和第10000次最小载荷前轴向加压分别为450N和700N,同时绳上施加的扭力矩为3Nm.研究骨密度对这些感兴趣的变量的影响。
2.6. Statistical analysis
Datawere evaluatedwithMatlab (MATLAB 6.5.1®, The MathWorks, Natick, Massachusetts). After testing for normality, non parametric paired Wilcoxon Signed-Rank Test was chosen to assess differences between groups concerning the investigated variables. Statistics were performedwith SPSS (SPSS Inc 14.0, 2005, Chicago, Illinois). Significance level of statistical tests was set at p≤0.05.
2.6统计分析
以Matlab分析数据。在测试正常后，使用无参数配对 Wilcoxon Signed-Rank 检验评价组间观察变量是否有显著性差异。统计软件采用SPSS。P≤0.05时，有显著性差异。
3. Results
The BMD was not significantly different between groups (Table 1). In one sample the bone around the custom-made anchor failed immediately not leaving any chance for anchor replacement. For this pair the compound cyclic load till failure ran with just axial compression and only the Humeral Head dataset was included in the statistical analysis.
3 结果
组间的骨密度没有显著性差异（表1）125。在一例标本中，大结节骨栓周围的骨质在装入骨栓骨粉碎，根本无法重新安装骨栓。对此对标本，在做复合循环负荷直至内固定失败的试验时，仅在轴向加压，统计时仅纳入肱骨头数据集。
The axial stiffness of the LPHN was significantly higher than that of the Expert PHN (P=0.036). A qualitative evaluation of the fluoroscopic images taken in unloading (valley) condition shows that the failure mechanism observed in both types of implant was similar and well reproducible throughout tests. Over a critical axial load, the humeral head started sliding along the intramedullary nail due to blade cutting through the cancellous bone. Once in contact with the shaft the humeral head rotated towards medial, partially closing the medial wedge shaped defect. Matching the fluoroscopic images with the machine data we set the arbitrary failure criterion to 3 mm. At this time point the intramedullary nail was already engaging the virtual scapular–humeral joint and signs of cut-out were clearly visible in the fluoroscopic images of all the samples. This failure mode detected via fluoroscopy is similar to that observed in clinical cases with insufficient medial buttress and varus displacement of the humeral head (Fig. 6). At the very end of the test the intramedullary nail was protruding out of the humeral head in the virtual scapular–humeral joint and the ostectomy was almost closed. The combined action of the pulling suture and sliding of the humeral head produced a rotation of the proximal aspect of the Greater Tuberosity fragment around the z axis towards posterior.
LPHN的轴向刚度显著高于Expert PHN(P=0.036)。对最低负荷时所做的透视进行定量分析显示两组的内固定失败机制相似，并且在测试可重现。在临界负荷时，肱骨头由于刃片切出松质骨而开始沿髓内钉滑动。当肱骨头与肱骨干接触后，肱骨头向内转旋转，使内侧楔形骨缺损变小。将透视结果与机器数据配对后，我们们内固定失败标准定为3mm。此时，所有标本的髓内钉已进入虚拟的盂肱关节，透视可见明显的切出。此模型与临床上缺乏内侧支撑，肱骨头内翻的病例相似（图6）。试验结束时，髓内钉切入到虚拟的盂肱关节，内侧截骨间隙几乎消失，拉力缝线和肱骨头滑动的总的结果是使大结节近端绕Z轴向后侧旋转。
Concerning the Humeral Head dataset, the number of cycles necessary to reach the failure criterion was significantly higher with the LPHN (P=0.012). The humeral head displacements were significantly higher (P=0.012; P=0.018) with the Expert PHN fixation at 5000 and at 10,000 cycles. Concerning the head rotation, the angular displacement recorded at 5000 cycles was not significantly different between groups, while at 10,000 cycles the angular displacement was significantly higher with the Expert PHN (P=0.018).
关于肱骨头数据集，LPHN组导致的内固定失败的循环次数明显高（P＝0.012），在循环5000次和10000时，Expert PHN固定组的肱骨头移位明显高（P=0.012，P＝0.018）。
Regarding the Greater Tuberosity dataset, the rotation was not significantly different between groups at 5000 cycles, whereas it was significantly higher with the Expert PHN (P=0.028) at 10,000 cycles. Under these test conditions no correlation between BMD and any of the investigated variables could be observed.
关于肱骨大结节数据集，在循环5000次，肱骨头成角移位在两组间没有显著差异，循环10000次时，Expert PHN组的成角移位显著高（P＝0.018）。组间的大结节数据集在循环5000次时没有显著性差异，在循环10000次时，Expert PHN大结节旋转移位明显多。（P＝0.028）。
在本试验的条件下，未发现骨密度与观察的变量间有相关性。
4. Discussion
Three-part fractures of the proximal humerus are still a challenging problem (Esser, 1994a,b) for trauma surgeons in daily practice. Although augmentation with bone cement and bone substitutes, nails, different plating techniques as well as blade plates (Egol et al., 2008; Kitson et al., 2007; Kwon et al., 2002; Lill et al., 2003; Sanders et al., 2007; Walsh et al., 2006; Weinstein et al., 2006) have shown favorable results there is not a shared optimal surgical treatment for proximal humerus poor bone quality multi fragment fractures. This deficiency shows the potential need for an improvement in techniques and instrumentations.
4 讨论
肱骨近端三部分骨折仍是创伤外科医师日常工作中一个棘手的问题。骨水泥加强、骨替代物、髓内钉、各种钢板技术包括角钢板都已经显示出良好的结果。对肱骨近端多部分骨折且骨量差者，尚没有公认理想的治疗方法。这就需要提高手术技术，改良手术器械。
4.1. The biomechanical model Currently there is no clearly defined and generally accepted model for testing different internal fixation techniques and implants designed for proximal humeral fractures. Although clearly oversimplified compared to the clinical situation, in this study we propose an easily reproducible biomechanical model that resembles some in vivo loading and restrain conditions. We standardized the fracture patterns using reproducible reference curves, osteotomies and ostectomy performed with a microsagittal saw. In example, the purpose of the wedge ostectomy was to simulate a comminuted metaphyseal fracture with deficient cortical support on the medial side. Imitating the Glenoid, the PMMA cup introduced no further nonphysiological constraint, leaving at the same time room for proximal intramedullary nail migration and free access to the Greater Tuberosity. Previous humeral fracture models and relative fixation methods have been subjected to a wide variety of testing conditions in earlier biomechanical studies. Applied as a single cycle, cyclically or to failure, the loading ranged from standard patterns (Foruria et al., 2010; Kitson et al., 2007; Koval et al., 1996; Sanders et al., 2007; Seide et al., 2007; Weinstein et al., 2006) (torsion, axial or ab-axial load and cantilever bending) to more complex patterns simulating shoulder abduction moving the humerus (Kwon et al., 2002) or acting on the remnant of rotator cuff tendons (Walsh et al., 2006). By effectively decoupling the effect of the actuator we were able to apply axial load and torsional motions in phase and to control two independent loading patterns. With the 25° lateral angulated axial pattern we simulated one of the most sever resultant directions of the glenohumeral compressive forces according to Bergmann's (Bergmann et al., 2007) measurements during different exercises (between 14° and 31° in the frontal plane). To the authors knowledge this is the first proximal humeral fracture model where the actuator was driven progressively increasing a load-time glenohumeral contact force curve recorded in vivo. Restraining the samples with a 25° lateral angulation this model put thus the implants under severe test pointing out their efficacy in stabilizing simulated proximal humeral fractures. With the second loading pattern we simulated the destabilizing effect of a part of a rotator cuff tendon on the Greater Tuberosity. The applied pulling force, set according to the assessment of Favre (Favre et al., 2005), resulted in a maximum torque moment on the humeral head of approximately 3 Nm, comparable to that measured in vivo (Bergmann et al., 2007).
4.1生物力学模型 目前尚没有被大家普遍接受的模型可用来测试肱骨近端骨折不同内固定技术和植入物。本试验中的生物力学模型虽然比临床中所见要明显简化，但它是一个简单的，可重复的生物力学模型。可模拟体内的一些负荷与控制情况。我们使用画参考曲线，骨锯截骨和楔形骨块切除使得骨折模型可重现。楔形骨块切除术是为了模拟缺乏内侧支撑的干骺端粉碎性骨折。骨水泥杯用来模拟肩胛盂，骨水泥杯没有增加一些并生理性的限制，为肱骨近端髓内钉移位留下了空间，它可以直接靠近大结节。以前的肱骨骨折模型和相对固定方法已用于多种生物力学测试条件。使用单次循环，多次循环或者直至内固定失败测试模式，负荷从标准模式（扭力，轴向或离心力和悬吊屈曲）到更复杂的通过活动肱骨或在肩袖附着处施加力而模拟肩关节外展的模式。通过解偶作用，我们可以同时施加轴向负荷和扭转运动，并可控制2种独立负荷的模式。通过将肱骨标本向外侧成角25°,我们可以模拟Bergmann在不同运动中测量到的盂肱关节最常见的作用力方向。据作者所知，这是第一个可逐渐增加盂肱关节作用力的体内模型。将标本固定在向外侧成角25°，本模型可测试内固定是否可有效的固定模拟的须骨近端骨折。在第二种负荷模式下，我们模拟附着在大结节的部分肩袖的不稳定作用。根据Favre等的测量结果设定的拉力在肱骨头上产生的最大扭矩约为3Nm,与体内测量结果相当。
4.2. The axial stiffness and compound cyclic load till failure tests
The data collected demonstrate that combining the Locking Plate for Humeral Nail with the existing geometry of the Expert PHN significantly increased the constructs' stiffness and improved the initial stability of the reconstructed simulated proximal humeral fracture. The failure mode observed in the compound cyclic load test confirmed that one of the main limiting factors in proximal humeral fracture fixation is the implant cut-out through the head's cancellous bone. This is especially true in osteoporotic bones where the weakest point of the whole construct is the implant-bone interaction. Despite our simplified and incomplete simulation of the action of the rotator cuff tendons the final position of the Greater Tuberosity fragments somehow reproduced the clinical situation although no major posterior-superior displacement of the Greater tuberosity was observed. The analysis of interfragmentary motion showed that LPHN offered superior stability both in the Humeral Head and in the Greater tuberosity fragments. The different geometry of the implant assembly and the increased number of fixation points could be the key factor slowing down the plastic deformation process distributing more evenly the stress on the fragments' cancellous bone. Despite we did not find any correlation between BMD and the investigated variables better stability can be expected also instru 4.2. The axial stiffness and compound cyclic load till failure tests The data collected demonstrate that combining the Locking Plate for Humeral Nail with the existing geometry of the Expert PHN significantly increased the constructs' stiffness and improved the initial stability of the reconstructed simulated proximal humeral fracture. The failure mode observed in the compound cyclic load test confirmed that one of the main limiting factors in proximal humeral fracture fixation is the implant cut-out through the head's cancellous bone. This is especially true in osteoporotic bones where the weakest point of the whole construct is the implant-bone interaction. Despite our simplified and incomplete simulation of the action of the rotator cuff tendons the final position of the Greater Tuberosity fragments somehow reproduced the clinical situation although no major posterior-superior displacement of the Greater tuberosity was observed. The analysis of interfragmentary motion showed that LPHN offered superior stability both in the Humeral Head and in the Greater tuberosity fragments. The different geometry of the implant assembly and the increased number of fixation points could be the key factor slowing down the plastic deformation process distributing more evenly the stress on the fragments' cancellous bone.
4.2 轴向刚度和复合循环至内固定失败测试
采集的数据显示将锁定钢板与现有的专家级髓内钉结合起来增加内固定物的刚度，提高了模拟的肱骨近端骨折重建后的初始稳定性。复合循环负荷至内固定失败测试证实肱骨近端骨折内固定的一个主要限制因素是内固定切出肱骨头松质骨。这对于骨质疏松性骨折尤其重要，此时内固定最薄弱的地方是内固定-骨相互作用的地方。尽管我们模拟的肩袖作用是简化而且是不完全的，虽然我们没有观察到大结节骨块明显向后上方移位，但大结节骨块的最终位置还是一定程度上再现了临床的情况。骨块间相互运动分析显示LPHN组肱骨头和大结节骨块的稳定性更好。内固定物组装的几何形态不同以及固定点的增加可能是主要的因素，这些因素减缓了塑性形变的过程，使肱骨头内应力分布更均匀。
Despite we did not find any correlation between BMD and the investigated variables better stability can be expected also instrumenting osteoporotic bones with an implant able to share stresses
虽然我们在骨密度与所观察的变量之间未发现有任何的相关性，但在骨质疏松病例中，使用能使应力分布更均匀的内固定时，可以期待更好的稳定性。
4.3. Study limitations
We built an oversimplified model trying to simulate the physiological conditions of three-part proximal humeral fractures but we are aware that the shoulder is a much more complex joint. We recognize that keeping constant the direction of the resultant axial force acting on the humeral head might not exactly reproduce the physiological loading. Trying to simulate a destabilizing force on the Greater Tuberosity fragment we applied a concentrated force of arbitrary magnitude that scarcely resembles the action of the entire rotator cuff. Collecting bones with a narrow range of BMD we were not able to assess which implant offers the best performances when used to fix osteoporotic fractures. The cadaveric model does not provide information about the effect of soft tissue dissection with possible devascularization of the osseous fragments. more evenly.
4.3研究局限性
我们用一个过分简化的模型来模拟肱骨近端三部分骨折的生理状态，但我们明白肩关节要复杂得多。我们认识到使作用在肱骨头上的轴向合力始终在同一方向可能并不能完全模拟生理状态下的负荷。为了试图模拟作用在肱骨大结节骨块上的不稳定的作用力，我们在大结节骨块上施加一个随意大小的集中力，而这几乎不能模拟整个肩袖在作用。由于所采集的骨的骨密度变化范围很小，我们不能确定哪种器械对于骨质疏松性骨折的效果最好。尸体模型还无法提供活体中由于软组织剥离所可能产生的骨块失血管化。
5. Conclusion
In this study we propose a reproducible three-part humeral fracture cadaveric model and we address a compelling question concerning internal fixation. We have shown that fixing three-part proximal humeral fractures combining the Expert Proximal Humeral Nail with a newLocking Plate for Humeral Nail improves interfragmentary stability in a cadaveric model. The significant decrease of varus displacement of the articular fragment in the LPHN group is a convincing advantage of this new device, mainly when there is a lack of medial column support (Gardner et al., 2007). The results of the present study are promising and provide a good biomechanical basis for the use of combined plate-nail implants in the treatment of multi-fragmentary proximal humerus fractures. However the ultimate advantage of this new implant assembly has to be determined in vivo.
5结论
本研究中，我们建立了一个可重现的肱骨近端三部分骨折尸体模型，我们强调与内固定相关的一个重要问题。我们的研究显示带锁定钢板的肱骨近端专家级髓内钉固定肱骨近端三部分骨折时，骨折块间的稳定性提高。在内侧缺乏支撑时，LPHN组关节骨块的内翻移位显著减少即体现出此种新型内固定的优势。本研究的结果是令人振奋的，为混合了钢板-髓内钉的新式内固定物在肱骨近端多部分骨折的使用提供了良好的生物力学基础。但是，这种新内固定物的优势尚需经体内试验评估。
Acknowledgments The AO Research Institute received a partial contribution cost for the study from Synthes GmbH (Oberdorf, Switzerland). None of the authors received payments or other benefits from any commercial entity.
声明：AO研究组织从Synthes GmbH得到了部分研究经费。作者没有从任何经济实体获得报酬或其他利益。
Biomechanical evaluation of a new fixation technique for internal fixation of three-part proximal humerus fractures in a novel cadaveric model
一种新的肱骨近端三部分骨折内固定技术在新鲜尸体模型上的生物力学评估
a b s t r a c t
Background: The optimal surgical treatment for displaced proximal humeral fractures is still controversial. A new implant for the treatment of three-part fractures has been recently designed. It supplements the existing Expert Humeral Nail with a locking plate. We developed a novel humeral cadaveric model and the existing implant and the prototype were biomechanically compared to determine their ability in maintaining interfragmentary stability.
Methods: The bone mineral density of eight pairs of cadaveric humeri was assessed and a three-part proximal humeral fracture was simulated with a Greater Tuberosity osteotomy and a surgical neck wedge ostectomy. The specimens were randomly assigned to either treatment. A bone anchor simulated part of a rotator cuff tendon pulling on the Greater Tuberosity. Specimens were initially tested in axial compression and afterward with a compound cyclic load to failure. An optical 3D motion tracking system continuously monitored the relative interfragmentary movements.
Findings: The specimen stabilized with the prototype demonstrated higher stiffness (P=0.036) and better interfragmentary stability (P values＜0.028) than the contralateral treated with the existing implant. There was no correlation between the bone mineral density and any of the investigated variables.
Interpretation: The convenience of this new IM-nail and locking plate assembly must be confirmed in vivo but the current study provides a biomechanical rationale for its use in the treatment of three-part proximal humeral fractures. The improved stability could be advantageous in particular when medial buttress is missing, even in osteoporotic bone.
摘要
背景：移位的肱骨近端骨折的理想手术治疗方案仍有争议。最近出现了一种新的用于治疗肱骨近端三部分骨折的植入物。它在现有的EHN（专家级肱骨髓内钉）上加了块锁定钢板。我们设计了一个新鲜的尸体模型，比较现有的内固定物与该新型内固定物生物力学以确定是否能够维持骨折块间的稳定性。
方法：评估8对尸体肱骨的骨密度，通过大结节截骨和外科颈楔形截骨模拟肱骨近端四部分骨折。标本随机分配进入治疗组。以骨锚模拟附着在大结节上的肩袖肌腱。开始时测量标本的轴向加压，然后为复合循环负荷至内固定失败。3D视觉运动追踪系统连续监测骨折块间的相对运动。
发现：与现有的内固定物相比，这种新的内固定显示出较高的(P=0.036)和较好的骨块间稳定性(P values＜0.028)。骨密度与所观察的变量之间没有任何相关性。解读：虽然还需做体内试验以确定这种髓内和锁定钢板相结合的新型内固定物的优势，但本研究已经为它在肱骨近端三部分骨折治疗中的使用提供了一些生物力学的依据。稳定性的提高使其具有一定优势，特别是当失去内侧支撑，即使是骨质疏松病例中（也具有优势）。
1． Introduction
Fracture of the proximal humerus is the third most common injury in people over 65 years of age after hip and Colles´ fracture (Baron et al., 1996). Most of these fractures are stable and can be successfully treated conservatively (Court-Brown CM et al., 2002; Court-Brown and McQueen, 2004; Hintermann et al., 2000; Lill and Josten, 2001; Szyszkowitz and Schippinger, 1999) but about 20% of the patients need surgical intervention. Nowadays the surgical treatment of two and three-part proximal humeral fractures aims for a stable fixation that preserves the blood supply to the fracture fragments and reduces the risk of shoulder stiffness allowing early joint mobilization (Hintermann et al., 2000; Kwon et al., 2002) but the optimal treatment is still controversial. On one side angular stable plating with locking plates has proven to be an effective treatment option (Brunner et al., 2009; Hirschmann et al., 2007; Plecko and Kraus, 2005; Sudkamp et al., 2009) and minimal invasive technique has become popular (Gallo et al., 2005; Gardner et al., 2004, 2005; Laflamme et al., 2008). On the other side recent data support the use of intramedullary nailing with new implants as promising alternatives (Gradl et al., 2007, 2009; Mihara et al., 2008). Recently a new implant, the Expert Proximal Humeral Nail with Locking Plate for Humeral Nail (LPHN, Synthes GmbH, Oberdorf, Switzerland), was designed for proximal three or four-part humeral fractures. It combines an additional locking plate with the existing geometry of the Expert Proximal Humeral Nail (Expert PHN, Synthes GmbH, Oberdorf, Switzerland) (Fig. 1). The rationale behind the additional plate is based on the assumption that augmenting the existing humeral nailing system optimizes bone purchase and leads to additional stability of the humeral head.
1． 介绍
肱骨近端骨折是65岁以上人群中第三常用的损伤，仅次于髋部骨折和Colles’骨折。这些骨折绝大多数是稳定的，可以保守治疗，但约有20%的患者需要手术治疗。现在，两部分和三部分肱骨近端骨折手术治疗的目标是坚强固定，保护骨折块的血供，减少术后肩关节僵硬的风险，允许早期关节活动，但关于最佳治疗仍有争议。 一方面，角度稳定锁定钢板已经证明是一个有效的治疗方法，并且微创技术已很流行。另一方面，近期的研究数据认为带新的植入物的髓内钉装置是一种具有良好的前景的治疗方案。近期，设计出一种新的内固定，肱骨近端带锁定钢板的专家级髓内钉（LPHN）(图1)，来治疗肱骨近端三部分或者四部分骨折。它将现有的专家级肱骨近端髓内钉（Expert PHN）与另一块锁定钢板结合起来。通过增加一块锁定钢板，能加固现有的髓内固定系统从而增加了骨的把持力，增加肱骨头稳定性。
The purpose of this study is to assess if the new implant provides more interfragmentary stability to a three-part proximal humeral fractures when compared to the standard implant in a novel cadaveric model. The information retrieved aims at providing a biomechanical basis for the use of combined plate-nail implants in the treatment of multi-fragmentary proximal humerus fractures.
本研究的目的是在新鲜尸体模型上比较此种新型内固定与标准内固定物，评估前者能否增加肱骨近端三部分骨折骨折块间的稳定性。本实验的目的是为髓内钉-钢板联合系统在肱骨近端多部分骨折中的应用提供生物力学基础。
2. Methods
Eight pairs of fresh frozen skeletally mature human cadaveric humeri were obtained and anterior–posterior X-rays were performed to exclude defects affecting the bone structure integrity. The bones were frozen at −20 °C until further processing.
2.方法：
准备八对骨骼成熟的新鲜冰冻尸体的肱骨，摄前后位片排除影响骨骼完整性的缺陷。骨骼在-20°C下保存备用。
2.1. Bone mineral density measurement The specimens underwent high-resolution bone densitometry measurement (BMD, mg HA/cm3, 123 μm resolution) using threedimensional peripheral quantitative computed tomography (pQCT, XtremeCT, Scanco Medical, Brüttisellen, Switzerland). The volume of interest included those areas where the proximal parts of both implants were to be placed during the following instrumentation.
2.1骨密度测量 使用高分辨率骨密度仪用三维外周量化CT（PQCT）法测量标本的骨密度。测量区域包括术中内固定在肱骨近端植入的区域。2.2. Prototype features
Similarly to the Expert PHN the LPHN provides a fixed-angle blade in the humeral head and an intramedullary load carrier that controls micro movements at the metaphysis. The LPHN has a 23 mm large square medical stainless steel additional plate restrained to the blade through an angular stable locking system. The plate features at each corner a hole hosting a divergent angular stable locking screw.
2.2新的内固定特点
与专家级肱骨近端髓内钉相似，LPHN在肱骨头处具有角度稳定性，同时其应力是髓内传导，可以控制干骺端的微动。LPHN具有一个23mm的方形钢板，其通过锁定系统固定于髓内钉的近端锁钉。该钢板的四个角都可以容纳一枚锁定螺钉。
2.3. The biomechanical model
2.3生物力学模型
2.3.1. Specimen preparation Prior to instrumentation, the specimens were thawed overnight at 4 °C in a refrigerator. During preparation, each specimen was covered with a towel soaked with saline solution and was periodically sprayed to prevent desiccation. The bone pairs were randomly assigned to treatment with either the Expert PHN or the LPHN. The order in which the fixation was applied (Expert PHN versus LPHN) was similarly randomized, but both specimens in each pair were instrumented on the same day always.
2.3.1标本准备
在行内固定前一晚，标本在4 °C冰箱内解冻。制备过程中，每具标本都以浸有盐液的毛巾覆盖，时常洒水以防止干燥。随机将成对的标本分配进入Expert PHN或者LPHN组，先进行Expert PHN内固定还是LPHN也是随机的，但同一对标本总是在同一天进行内固定。
All sampleswere prepared following the samefive steps: 1) marking the reference points and curves on the surface of the bone, 2) performing humeral neck osteotomy creating a lateral fragment (Greater Tuberosity), 3) fixing a custom-made anchor to the lateral–posterior side of this same fragment, 4) anatomical reduction of the fragments and fixation with the Expert PHNor LPHNimplant, 5) humeral surgical neck wedge ostectomy creating the humeral head fragment 6) standardization of the shaft length.
所有标本按以下五步制备：1）在标本表面画出参考点及曲线 2）行肱骨颈截骨造成外侧骨折（大结节）3）将定制的骨锚固定至该骨块的后外侧 4)解剖复位骨折块，Expert PHN或者LPHN固定 5)肱骨外科颈楔形截骨以造成肱骨头骨折块 6)骨干长度标准化 Step 1. The following two groups of reference points and curves were drawn on each humerus using a skin marker (Devon® Skin Marker, TYCO Healthcare LANE COVE, NSW 2066 AUSTRALIA): Greater Tuberosity's reference curves and Surgical Neck's reference curves (Fig. 2). Step 2. A microsagittal saw (Synthes Saw Blade 0.4 mm thick, Synthes GmbH 4500 Solothurn, Switzerland) was used to create a lateral fragment, including the Greater Tuberosity, following the reference curves. Step 3. A custom-made anchor consisting of a 2 mm bolt and two 9 mm external diameter washers was connected to the lateral–posterior side of each Greater Tuberosity fragment after predrilling with a 2 mm drill tip (Fig. 3). A nut and one washer laid outside of the fragment and the screw's head and the second washer inside. In order to improve the holding capability, each anchor was augmented using PMMA (Polymethylmethacrylate, Beracryl, W. Troller AG, Switzerland). Once the anchor was in place the bolt was tightened once leading the inner washer to compact the cancellous bone underneath. The bolt was then released and two 7 metric sutures (TI•CRON 5, Syneture, Covidien Norwalk, CT 06856 USA) were independently connected with surgical knots to that part of the bolt protruding out of the bone and located under the external washer. One milliliter of PMMA at the later hardening stage was applied underneath the internal washer, towards the compacted cancellous bone. The bolt was then retightened and the exceeding PMMA removed. Care was taken not to have any part of the augmented anchor protruding medial to the adjacent fracture plane in order to avoid any additional nonphysiological friction between fragments. Step 4. The randomly assigned implant (Expert PHN or LPHN) was fully fixed following the manufacturer's recommendations including two parallel angular stable locking screws distally. Step 5. The microsagittal oscillating saw was used to perform a surgical neck wedge ostectomy following the most distal and the 15° medially oriented Surgical Neck's reference curves. Care was taken to remove any contact point between the humeral head's fragment and the shaft. Step 6. All the sampleswere reduced to an equal length cutting the shaft perpendicular to its axis 70 mm distal to the center of the most distal angular stable nail screw. All the procedures, including drawings, osteotomy, augmentation, implant fixation and ostectomy, were performed by the same surgeon (M.P.). Fluoroscopy was used to check the length and position of the implants at each step. Each humeral diaphysis was then embedded in the PMMA perpendicular to the ground in a cylindrical stainless steel base-plate up to 40 mm distal to the center of the most distal angular stable nail screw.
步骤一：以皮肤记号笔在肱骨上标出以下两组点和曲线:大结节参考线及外科颈参考线（图2）。步骤二：以小摆锯沿参考线截骨，形成外侧包含大结节的骨折块。步骤三：以2mm钻头预钻孔，将定制的骨锚（由一个直径2mm螺栓及2枚外径为9mm的垫片组成）固定至大结节骨块的后外侧（图3）。一枚螺母及垫片留在骨块外面，螺钉头及另一枚垫片在骨块里面。为了增加把持力，每个螺钉都以骨水泥加固。一旦骨锚到达预定位置，收紧螺栓，挤压垫片间的松质骨。松开螺栓，用两根7#metric缝合线以外科结固定至螺栓突出骨外并位于垫片下的部分。将1mm凝固后期的骨水泥注入到骨块内垫片的下方，再次收紧螺栓，清除溢出的骨水泥。注意防止骨水泥突向内侧相邻的骨块以避免一切非生理的摩擦。步骤四：按照操作手册的要求以Expert PHN或者LPHN固定，包括远端两枚平行的角度稳定的锁定钉。步骤五：沿最远端参考线和向内侧及头侧成角15°的参考线，以小摆锯作外科颈楔形骨切除。注意将肱骨头骨块与肱骨干骨块间的所有接触部分清除。步骤六：所有标本都复位至相同长度，在最远端锁钉中心远端70mm处，垂直肱骨干轴线截断肱骨。所有操作，包括画线，截骨，骨水泥增强，内固定，骨切除，都由同一医生完成。（M.P）.每一步操作都有X光透视确定内固定的长度及位置。每根肱骨干都垂直地面埋入盛骨水泥的圆柱形不锈钢底盘中，底盘距离最远端锁钉的中心40mm。
2.3.2. Test set up
Biomechanical testing was performed on a biaxial servo-hydraulic machine, MTS Mini Bionix II 858 (MTS Systems Corp., Eden Prairie, MN) using a 4kN/20 Nm load cell (HUPPERT 6, HUPPERT GmbHHerrenberg, Germany). Each sample was restrained to the machine with a 25° lateral angulation (Fig. 4) according to the measurement range of Bergmann et al. (2007). Distally a cardan joint connected to the stainless steel base-plate restrained all the sample displacements and the rotation around the shaft axis. The load was applied on the samples by means of a custom-made PMMA cup. An even load distribution on the humeral head was provided via a cavity created in the PMMA cup as negative of a plastic humeral head (5010 Humerus, SYNBONE AG, 7208 Malans, Switzerland) simulating the glenoid. Relative rotation of the cup and the humeral head was prevented applying sand paper strips on that part of the cup in contact with the sample. An adequate notch created in the cup, positioned above the nail entry point, always guaranteed free movements to the nail during testing. The PMMA cup was joined to the machine actuator by means of a custom-made flange and transmitted to the samples the axial load only. Rotational bearing associated with the custom-made flange in fact allowed decoupling the effect of the actuator axial and torsional movements. Once the sample was preloaded, rotations of the actuator did not elicit rotations of the PMMA cup. The torsional actuator was then used to independently generate the force pulling on the Greater Tuberosity fragment via a cable. Fitted in a groove, one side of the cable was anchored to the flange by means of a screw. The cable was passed through two pulleys during testing and connected to the two sutures attached to the augmented bone anchor. The position of the first pulley was such that the pulling direction of the first end of the cable was always tangent to the flange and perpendicular to its axis of rotation. The position of the second pulley was such that the initial direction of the pulling force acting on the anchor inserted in the Greater Tuberosity fragment was angled 110° to the humeral shaft axis in the medial-lateral plane.
2.3.2试验设计
生物力学测试在一个双轴伺服液压器MTS 迷你BionixⅡ858 上进行，使用 4kN/20 Nm 的传感器。根据Bergmann等的测量范围，每个标本都固定于机器上，向外侧成角25°（图4）。远端一个连接在不锈钢底盘上的万向关节限制了标本所有的移位和绕轴旋转。通过一个定制的骨水泥杯将负荷施加于样本。在骨水泥上构成一个如同挖除了肱骨头的臼杯以模拟肩胛盂，从而以使分布于肱骨头表面的力更均匀。用砂皮打磨骨水泥杯与肱骨头的接触部位以防止杯与肱骨头之间的相对旋转。在位于髓内钉进针点上方的骨水泥杯的位置开个槽，以保证在测试过程中骨水泥杯可自由地向髓内钉方向移动。骨水泥杯通过一个定制的轮缘连接至机器的执行器，只向标本传递轴向应力。旋转轴承加上定制的轮缘，实际上就可以将执行器的运动分解成轴向和扭转运动。当预载了标本，执行器的旋转不会引起骨水泥杯旋转。旋转执行器产生独立的拉力通过一根绳作用于大结节。绳装于沟槽内，它的一端通过一枚螺钉固定于轮缘。试验中，绳子通过两个（带滑轮的）把手，并与连接在大结节骨栓上的缝线相连接。一个把手的位置要保证绳的一头总是与轮缘相切，与其旋转轴垂直。另一个把手要使得开始作用于骨栓的力与肱骨干在冠状面上成110°角。
2.3.3. Loading patterns
All specimens were tested according to the following loading patterns:
1. Axial stiffness test: axial loading until 200 N in displacement control (0.02 mm/s).
2. Compound cyclic load till failure: increasing axial compression load, applied at 1 Hz, in phase with pulling on the bone anchor. The axial compression was a filtered interpolation of the pattern recorded by Bergmann et al. (2007) on an instrumented shoulder prosthesis during flexion movement. Starting from 200 N the maximum (peak) of the curve was cyclically increased (Windolf et al., 2009) (0.05 N/cycle) until failure while its minimum (valley) was kept constant at 50 N during the entire test. The pulling force acting on the anchor was controlled in angular displacement control and torque detectors between 1 and 3 Nm. This torque range produced a pulling force on the anchor ranging from 40 to 100 N. The test was automatically stopped when 8 mm actuator axial displacement was reached. At this stage the wedge ostectomy was almost completely closed and every fixation clearly failed.
2.3.3载荷模式
所有标本按以下模式测试
1.轴向刚度试验:在控制移位(0.02 mm/s).轴向加载至200N。
2. 复合循环负荷直到内固定失败：增加轴向载荷，加载频率为1Hz，牵拉骨栓的频率也是1Hz。轴向加压是采用Bergmann等在肩关节假体屈曲运动时所用方法的改良。从200N开始，曲线最大值（峰值）循环增加直到固定失败，其最小值（谷值）始终是50N.通过控制成角并将扭矩控制在1-3NM之间来控制作用于骨栓的力。该范围内的扭矩在骨栓上产生的作用力为40-100N。当执行器轴向位移达到8mm时，试验自动结束。这时，楔形骨间隙已经几乎完全闭合，所有的内固定都失败了。
2.4. Data collection
2.4.1. Data machine Displacement and load were recorded from the MTS controllers at 64 Hz. 2.4.2. Fluoroscopy
At the beginning of the test and every 500 cycles until failure the machine observed a two second pause in unloading (valley) condition (50 N axial load, 1 Nm torque) in order to perform a fluoroscopic assessment (43 kV, 10 mAs — Siemens, ARCO SI 100). The assessment was performed with focus on the identification of the failure mechanism and to associate the loading history with the loss of fragments fixation.
2.4.3. Measurement of the interfragmentary motion During testing an optical 3D motion tracking system consisting of 5 ProReflexMCU digital cameras (Qualisys AB, Gothenburg, Sweden)was used to continuously track the three-dimensional relative motions of the fracture fragments. Reflective marker sets were attached to the humeral shaft, the humeral head and the Greater Tuberosity. Prior to each test the tracking systemwas calibrated and a reference coordinate systemwas defined togetherwith the position of themostmedial aspect of the surgical neck ostectomy, taken as reference point on the humeral head (Fig. 4).
2.4数据采集
2.4.1
从MTS的控制器记录移位及载荷。
2.4.2透视
试验开始时，以及直到内固定失败前，每500个循环机器停2秒，在最小载荷，最小扭矩下透视一次。评估的焦点是内固定失败的机制以及固定失败与所加载荷的关系。
2.4.3骨块间移位的测量
测试中，使用视觉3D运动追踪系统连续追踪骨折块的三维相对运动。反应性标记点固定于肱骨干、肱骨头和大结节。第次测试前，追踪系统都需校准，以楔形截骨时外科颈最内侧作为肱骨头的参考点，设置参考坐标系统（图4）
2.5. Parameters of interest
The construct stiffness was determined between 50 and 100 N on the load-displacement curve via the machine data during the first test. Matching those fluoroscopic images showing the first signs of construct failure with the motion tracking data an arbitrary displacement as failure criterion was defined. The following two datasets of variables were calculated at the failure criterion, at 5000 and at 10000 cycles from the motion tracking data in the reference coordinate system: Humeral Head dataset and Greater Tuberosity fragment dataset (Fig. 5).
The former were the number of cycles until the failure criterion and the humeral head displacements and rotations at 5000 and 10000 cycles. The displacements were calculated along the shaft axis (x axis) and the rotations were calculated with respect to an anterior– posterior axis passing through the humeral head and perpendicular to the shaft axis (y axis) in unloading conditions (valley).
The latter were the rotations of the Greater Tuberosity fragment around a medial-lateral axis passing through the humeral head and perpendicular to the shaft axis (z axis) in unloading condition (valley) at 5000 and 10000 cycles. The 5000th and 10,000th valleys followed the peaks where the samples were respectively axially loaded with 450 N and 700 N in phase with 3 Nm torque applied to the cable. The effect of the BMD on the variables of interest was investigated
2.5感兴趣的参数
第一次测试时，通过机器数据（设定），植入物刚度在载荷-移位曲线上设置为50-100N.综合追踪系统和透视，初步设定内固定失败的移位程度。以下两个变量是根据失败标准，在循环5000和10000次时从视觉运动追踪系统参考坐标的数据计算而来：肱骨头数据集和大结节骨块数据集(图5)。
前者是直到内固定失败标准时循环次数，以及循环5000次和1000次时肱骨头的移位和旋转。移位从X轴上计算，旋转以最小载荷下绕Y轴计算。
后者为循环5000次和1000次时大结节骨块在最小载荷下绕Z轴的旋转。第5000次和第10000次最小载荷前轴向加压分别为450N和700N,同时绳上施加的扭力矩为3Nm.研究骨密度对这些感兴趣的变量的影响。
2.6. Statistical analysis
Datawere evaluatedwithMatlab (MATLAB 6.5.1®, The MathWorks, Natick, Massachusetts). After testing for normality, non parametric paired Wilcoxon Signed-Rank Test was chosen to assess differences between groups concerning the investigated variables. Statistics were performedwith SPSS (SPSS Inc 14.0, 2005, Chicago, Illinois). Significance level of statistical tests was set at p≤0.05.
2.6统计分析
以Matlab分析数据。在测试正常后，使用无参数配对 Wilcoxon Signed-Rank 检验评价组间观察变量是否有显著性差异。统计软件采用SPSS。P≤0.05时，有显著性差异。
3. Results
The BMD was not significantly different between groups (Table 1). In one sample the bone around the custom-made anchor failed immediately not leaving any chance for anchor replacement. For this pair the compound cyclic load till failure ran with just axial compression and only the Humeral Head dataset was included in the statistical analysis.
3 结果
组间的骨密度没有显著性差异（表1）125。在一例标本中，大结节骨栓周围的骨质在装入骨栓骨粉碎，根本无法重新安装骨栓。对此对标本，在做复合循环负荷直至内固定失败的试验时，仅在轴向加压，统计时仅纳入肱骨头数据集。
The axial stiffness of the LPHN was significantly higher than that of the Expert PHN (P=0.036). A qualitative evaluation of the fluoroscopic images taken in unloading (valley) condition shows that the failure mechanism observed in both types of implant was similar and well reproducible throughout tests. Over a critical axial load, the humeral head started sliding along the intramedullary nail due to blade cutting through the cancellous bone. Once in contact with the shaft the humeral head rotated towards medial, partially closing the medial wedge shaped defect. Matching the fluoroscopic images with the machine data we set the arbitrary failure criterion to 3 mm. At this time point the intramedullary nail was already engaging the virtual scapular–humeral joint and signs of cut-out were clearly visible in the fluoroscopic images of all the samples. This failure mode detected via fluoroscopy is similar to that observed in clinical cases with insufficient medial buttress and varus displacement of the humeral head (Fig. 6). At the very end of the test the intramedullary nail was protruding out of the humeral head in the virtual scapular–humeral joint and the ostectomy was almost closed. The combined action of the pulling suture and sliding of the humeral head produced a rotation of the proximal aspect of the Greater Tuberosity fragment around the z axis towards posterior.
LPHN的轴向刚度显著高于Expert PHN(P=0.036)。对最低负荷时所做的透视进行定量分析显示两组的内固定失败机制相似，并且在测试可重现。在临界负荷时，肱骨头由于刃片切出松质骨而开始沿髓内钉滑动。当肱骨头与肱骨干接触后，肱骨头向内转旋转，使内侧楔形骨缺损变小。将透视结果与机器数据配对后，我们们内固定失败标准定为3mm。此时，所有标本的髓内钉已进入虚拟的盂肱关节，透视可见明显的切出。此模型与临床上缺乏内侧支撑，肱骨头内翻的病例相似（图6）。试验结束时，髓内钉切入到虚拟的盂肱关节，内侧截骨间隙几乎消失，拉力缝线和肱骨头滑动的总的结果是使大结节近端绕Z轴向后侧旋转。
Concerning the Humeral Head dataset, the number of cycles necessary to reach the failure criterion was significantly higher with the LPHN (P=0.012). The humeral head displacements were significantly higher (P=0.012; P=0.018) with the Expert PHN fixation at 5000 and at 10,000 cycles. Concerning the head rotation, the angular displacement recorded at 5000 cycles was not significantly different between groups, while at 10,000 cycles the angular displacement was significantly higher with the Expert PHN (P=0.018).
关于肱骨头数据集，LPHN组导致的内固定失败的循环次数明显高（P＝0.012），在循环5000次和10000时，Expert PHN固定组的肱骨头移位明显高（P=0.012，P＝0.018）。
Regarding the Greater Tuberosity dataset, the rotation was not significantly different between groups at 5000 cycles, whereas it was significantly higher with the Expert PHN (P=0.028) at 10,000 cycles. Under these test conditions no correlation between BMD and any of the investigated variables could be observed.
关于肱骨大结节数据集，在循环5000次，肱骨头成角移位在两组间没有显著差异，循环10000次时，Expert PHN组的成角移位显著高（P＝0.018）。组间的大结节数据集在循环5000次时没有显著性差异，在循环10000次时，Expert PHN大结节旋转移位明显多。（P＝0.028）。
在本试验的条件下，未发现骨密度与观察的变量间有相关性。
4. Discussion
Three-part fractures of the proximal humerus are still a challenging problem (Esser, 1994a,b) for trauma surgeons in daily practice. Although augmentation with bone cement and bone substitutes, nails, different plating techniques as well as blade plates (Egol et al., 2008; Kitson et al., 2007; Kwon et al., 2002; Lill et al., 2003; Sanders et al., 2007; Walsh et al., 2006; Weinstein et al., 2006) have shown favorable results there is not a shared optimal surgical treatment for proximal humerus poor bone quality multi fragment fractures. This deficiency shows the potential need for an improvement in techniques and instrumentations.
4 讨论
肱骨近端三部分骨折仍是创伤外科医师日常工作中一个棘手的问题。骨水泥加强、骨替代物、髓内钉、各种钢板技术包括角钢板都已经显示出良好的结果。对肱骨近端多部分骨折且骨量差者，尚没有公认理想的治疗方法。这就需要提高手术技术，改良手术器械。
4.1. The biomechanical model Currently there is no clearly defined and generally accepted model for testing different internal fixation techniques and implants designed for proximal humeral fractures. Although clearly oversimplified compared to the clinical situation, in this study we propose an easily reproducible biomechanical model that resembles some in vivo loading and restrain conditions. We standardized the fracture patterns using reproducible reference curves, osteotomies and ostectomy performed with a microsagittal saw. In example, the purpose of the wedge ostectomy was to simulate a comminuted metaphyseal fracture with deficient cortical support on the medial side. Imitating the Glenoid, the PMMA cup introduced no further nonphysiological constraint, leaving at the same time room for proximal intramedullary nail migration and free access to the Greater Tuberosity. Previous humeral fracture models and relative fixation methods have been subjected to a wide variety of testing conditions in earlier biomechanical studies. Applied as a single cycle, cyclically or to failure, the loading ranged from standard patterns (Foruria et al., 2010; Kitson et al., 2007; Koval et al., 1996; Sanders et al., 2007; Seide et al., 2007; Weinstein et al., 2006) (torsion, axial or ab-axial load and cantilever bending) to more complex patterns simulating shoulder abduction moving the humerus (Kwon et al., 2002) or acting on the remnant of rotator cuff tendons (Walsh et al., 2006). By effectively decoupling the effect of the actuator we were able to apply axial load and torsional motions in phase and to control two independent loading patterns. With the 25° lateral angulated axial pattern we simulated one of the most sever resultant directions of the glenohumeral compressive forces according to Bergmann's (Bergmann et al., 2007) measurements during different exercises (between 14° and 31° in the frontal plane). To the authors knowledge this is the first proximal humeral fracture model where the actuator was driven progressively increasing a load-time glenohumeral contact force curve recorded in vivo. Restraining the samples with a 25° lateral angulation this model put thus the implants under severe test pointing out their efficacy in stabilizing simulated proximal humeral fractures. With the second loading pattern we simulated the destabilizing effect of a part of a rotator cuff tendon on the Greater Tuberosity. The applied pulling force, set according to the assessment of Favre (Favre et al., 2005), resulted in a maximum torque moment on the humeral head of approximately 3 Nm, comparable to that measured in vivo (Bergmann et al., 2007).
4.1生物力学模型 目前尚没有被大家普遍接受的模型可用来测试肱骨近端骨折不同内固定技术和植入物。本试验中的生物力学模型虽然比临床中所见要明显简化，但它是一个简单的，可重复的生物力学模型。可模拟体内的一些负荷与控制情况。我们使用画参考曲线，骨锯截骨和楔形骨块切除使得骨折模型可重现。楔形骨块切除术是为了模拟缺乏内侧支撑的干骺端粉碎性骨折。骨水泥杯用来模拟肩胛盂，骨水泥杯没有增加一些并生理性的限制，为肱骨近端髓内钉移位留下了空间，它可以直接靠近大结节。以前的肱骨骨折模型和相对固定方法已用于多种生物力学测试条件。使用单次循环，多次循环或者直至内固定失败测试模式，负荷从标准模式（扭力，轴向或离心力和悬吊屈曲）到更复杂的通过活动肱骨或在肩袖附着处施加力而模拟肩关节外展的模式。通过解偶作用，我们可以同时施加轴向负荷和扭转运动，并可控制2种独立负荷的模式。通过将肱骨标本向外侧成角25°,我们可以模拟Bergmann在不同运动中测量到的盂肱关节最常见的作用力方向。据作者所知，这是第一个可逐渐增加盂肱关节作用力的体内模型。将标本固定在向外侧成角25°，本模型可测试内固定是否可有效的固定模拟的须骨近端骨折。在第二种负荷模式下，我们模拟附着在大结节的部分肩袖的不稳定作用。根据Favre等的测量结果设定的拉力在肱骨头上产生的最大扭矩约为3Nm,与体内测量结果相当。
4.2. The axial stiffness and compound cyclic load till failure tests
The data collected demonstrate that combining the Locking Plate for Humeral Nail with the existing geometry of the Expert PHN significantly increased the constructs' stiffness and improved the initial stability of the reconstructed simulated proximal humeral fracture. The failure mode observed in the compound cyclic load test confirmed that one of the main limiting factors in proximal humeral fracture fixation is the implant cut-out through the head's cancellous bone. This is especially true in osteoporotic bones where the weakest point of the whole construct is the implant-bone interaction. Despite our simplified and incomplete simulation of the action of the rotator cuff tendons the final position of the Greater Tuberosity fragments somehow reproduced the clinical situation although no major posterior-superior displacement of the Greater tuberosity was observed. The analysis of interfragmentary motion showed that LPHN offered superior stability both in the Humeral Head and in the Greater tuberosity fragments. The different geometry of the implant assembly and the increased number of fixation points could be the key factor slowing down the plastic deformation process distributing more evenly the stress on the fragments' cancellous bone. Despite we did not find any correlation between BMD and the investigated variables better stability can be expected also instru 4.2. The axial stiffness and compound cyclic load till failure tests The data collected demonstrate that combining the Locking Plate for Humeral Nail with the existing geometry of the Expert PHN significantly increased the constructs' stiffness and improved the initial stability of the reconstructed simulated proximal humeral fracture. The failure mode observed in the compound cyclic load test confirmed that one of the main limiting factors in proximal humeral fracture fixation is the implant cut-out through the head's cancellous bone. This is especially true in osteoporotic bones where the weakest point of the whole construct is the implant-bone interaction. Despite our simplified and incomplete simulation of the action of the rotator cuff tendons the final position of the Greater Tuberosity fragments somehow reproduced the clinical situation although no major posterior-superior displacement of the Greater tuberosity was observed. The analysis of interfragmentary motion showed that LPHN offered superior stability both in the Humeral Head and in the Greater tuberosity fragments. The different geometry of the implant assembly and the increased number of fixation points could be the key factor slowing down the plastic deformation process distributing more evenly the stress on the fragments' cancellous bone.
4.2 轴向刚度和复合循环至内固定失败测试
采集的数据显示将锁定钢板与现有的专家级髓内钉结合起来增加内固定物的刚度，提高了模拟的肱骨近端骨折重建后的初始稳定性。复合循环负荷至内固定失败测试证实肱骨近端骨折内固定的一个主要限制因素是内固定切出肱骨头松质骨。这对于骨质疏松性骨折尤其重要，此时内固定最薄弱的地方是内固定-骨相互作用的地方。尽管我们模拟的肩袖作用是简化而且是不完全的，虽然我们没有观察到大结节骨块明显向后上方移位，但大结节骨块的最终位置还是一定程度上再现了临床的情况。骨块间相互运动分析显示LPHN组肱骨头和大结节骨块的稳定性更好。内固定物组装的几何形态不同以及固定点的增加可能是主要的因素，这些因素减缓了塑性形变的过程，使肱骨头内应力分布更均匀。
Despite we did not find any correlation between BMD and the investigated variables better stability can be expected also instrumenting osteoporotic bones with an implant able to share stresses
虽然我们在骨密度与所观察的变量之间未发现有任何的相关性，但在骨质疏松病例中，使用能使应力分布更均匀的内固定时，可以期待更好的稳定性。
4.3. Study limitations
We built an oversimplified model trying to simulate the physiological conditions of three-part proximal humeral fractures but we are aware that the shoulder is a much more complex joint. We recognize that keeping constant the direction of the resultant axial force acting on the humeral head might not exactly reproduce the physiological loading. Trying to simulate a destabilizing force on the Greater Tuberosity fragment we applied a concentrated force of arbitrary magnitude that scarcely resembles the action of the entire rotator cuff. Collecting bones with a narrow range of BMD we were not able to assess which implant offers the best performances when used to fix osteoporotic fractures. The cadaveric model does not provide information about the effect of soft tissue dissection with possible devascularization of the osseous fragments. more evenly.
4.3研究局限性
我们用一个过分简化的模型来模拟肱骨近端三部分骨折的生理状态，但我们明白肩关节要复杂得多。我们认识到使作用在肱骨头上的轴向合力始终在同一方向可能并不能完全模拟生理状态下的负荷。为了试图模拟作用在肱骨大结节骨块上的不稳定的作用力，我们在大结节骨块上施加一个随意大小的集中力，而这几乎不能模拟整个肩袖在作用。由于所采集的骨的骨密度变化范围很小，我们不能确定哪种器械对于骨质疏松性骨折的效果最好。尸体模型还无法提供活体中由于软组织剥离所可能产生的骨块失血管化。
5. Conclusion
In this study we propose a reproducible three-part humeral fracture cadaveric model and we address a compelling question concerning internal fixation. We have shown that fixing three-part proximal humeral fractures combining the Expert Proximal Humeral Nail with a newLocking Plate for Humeral Nail improves interfragmentary stability in a cadaveric model. The significant decrease of varus displacement of the articular fragment in the LPHN group is a convincing advantage of this new device, mainly when there is a lack of medial column support (Gardner et al., 2007). The results of the present study are promising and provide a good biomechanical basis for the use of combined plate-nail implants in the treatment of multi-fragmentary proximal humerus fractures. However the ultimate advantage of this new implant assembly has to be determined in vivo.
5结论
本研究中，我们建立了一个可重现的肱骨近端三部分骨折尸体模型，我们强调与内固定相关的一个重要问题。我们的研究显示带锁定钢板的肱骨近端专家级髓内钉固定肱骨近端三部分骨折时，骨折块间的稳定性提高。在内侧缺乏支撑时，LPHN组关节骨块的内翻移位显著减少即体现出此种新型内固定的优势。本研究的结果是令人振奋的，为混合了钢板-髓内钉的新式内固定物在肱骨近端多部分骨折的使用提供了良好的生物力学基础。但是，这种新内固定物的优势尚需经体内试验评估。
Acknowledgments The AO Research Institute received a partial contribution cost for the study from Synthes GmbH (Oberdorf, Switzerland). None of the authors received payments or other benefits from any commercial entity.
声明：AO研究组织从Synthes GmbH得到了部分研究经费。作者没有从任何经济实体获得报酬或其他利益。
