publishing staff surgical treatment of … “four-part” pattern 8–10(figure1) seen in some...

$: PUBLISHING STAFF Surgical Treatment of … “four-part” pattern 8–10(Figure1) seen in some fractures of the proximal humerus deserves particular attention. Subtle impaction of$
Endorsed by the Association for HospitalMedical Education

The Association for Hospital Medical Educationendorses HOSPITAL PHYSICIAN for the pur-pose of presenting the latest developments inmedical education as they affect residency pro-grams and clinical hospital practice.

®

ORTHOPAEDIC SURGERY BOARD REVIEW MANUAL

Orthopaedic Surgery Volume 7, Part 4 1

PUBLISHING STAFF

PRESIDENT, GROUP PUBLISHERBruce M. White

EXECUTIVE EDITORDebra Dreger

SENIOR EDITORBecky Krumm, ELS

CONTRIBUTING EDITOREllen M. McDonald, PhD, ELS

ASSISTANT EDITORJennifer M. Vander Bush

EDITORIAL ASSISTANTNora H. Landon

EXECUTIVE VICE PRESIDENTBarbara T. White, MBA

PRODUCTION DIRECTORSuzanne S. Banish

PRODUCTION ASSOCIATESTish Berchtold KlusMary Beth Cunney

PRODUCTION ASSISTANTStacey Caiazzo

ADVERTISING/PROJECT MANAGERPatricia Payne Castle

Copyright 2001, Turner White Communications, Inc., 125 Strafford Avenue, Suite 220, Wayne, PA 19087-3391, www.turner-white.com. Allrights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means,mechanical, electronic, photocopying, recording, or otherwise, without the prior written permission of Turner White Communications, Inc.The editors are solely responsible for selecting content. Although the editors take great care to ensure accuracy, Turner WhiteCommunications, Inc., will not be liable for any errors of omission or inaccuracies in this publication. Opinions expressed are those of theauthors and do not necessarily reflect those of Turner White Communications, Inc.

NOTE FROM THE PUBLISHER:This publication has been developed withoutinvolvement of or review by the AmericanBoard of Orthopaedic Surgery.

Surgical Treatment ofFractures of the ProximalHumerusSeries Editor: Robert T. Trousdale, MDAssociate Professor of Orthopaedic Surgery, Mayo Graduate School ofMedicine, Consultant, Department of Orthopaedic Surgery, MayoClinic, Rochester, MN

Contributing Author:Michael E. Torchia, MDAssociate Professor of Orthopaedic Surgery, Mayo Graduate School ofMedicine, Consultant, Department of Orthopaedic Surgery, MayoClinic, Rochester, MN

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Surgical Indications . . . . . . . . . . . . . . . . . . . . . . . . . 2

Surgical Procedures . . . . . . . . . . . . . . . . . . . . . . . . . 4

Humeral Head Replacement in Acute Fractures . . . . 6

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Table of Contents

Cover Illustration by Marc Galindo

2 Hospital Physician Board Review Manual

INTRODUCTION

Despite recent advances in imaging and fixation tech-niques, the treatment of displaced fractures affecting theproximal humerus remains a challenge. Many such frac-tures occur in elderly patients who may have poor gen-eral health, poor bone quality, and poor postoperativecompliance. The remainder tend to occur in youngerpatients with better bone quality who have experiencedhigh-energy trauma that can cause associated severe soft-tissue injuries. Specifically, the labrum, capsule, rotatorcuff, brachial plexus, peripheral nerves, and blood vesselscan all be injured in cases of high-energy trauma. As withthe treatment of almost any fracture, the goal remains toobtain and maintain an acceptable reduction while heal-ing progresses. Although this goal often can be achievedwith limited internal fixation and a period of prolongedimmobilization, the subsequent stiffness can be disabling.Optimal treatment involves providing fixation that willwithstand the stress of early passive movement, an idealthat cannot always be attained.

In situations in which the humeral head is not recon-structable, replacement arthroplasty may be indicated.However, unlike hemiarthroplasty of the hip for fem-oral neck fractures, successful humeral head replace-ment after trauma still requires bone-to-bone healing.The tuberosities must unite to the shaft if any degree offunction is to be restored.

This manual will discuss surgical treatment of frac-tures of the proximal humerus, focusing on indications,approaches, specific techniques, and outcomes. Themanual will conclude with a specific, detailed review ofhumeral head replacement in cases of acute fracture ofthe proximal humerus.

SURGICAL INDICATIONS

DISPLACED FRACTURES

Most fractures of the proximal humerus (approx-imately 80%) are minimally displaced, impacted, rela-tively stable, and able to be treated conservatively.1

Fractures with displaced, unstable patterns are typical-

ly reduced, with either a closed or open method, andthen stabilized or fixed with various types of implants.2

Although Neer’s classification system of fractures of theproximal humerus3,4 is well known, the central impor-tance of displacement has probably not been empha-sized enough.

Determining whether or not a patient’s fracturedbone segments are displaced to a significant degree re-quires high-quality radiographs, and these images can bedifficult to obtain in acutely injured patients. Often, it ishelpful for the operating surgeon to assist the radiologytechnician in supporting the arm, positioning the pa-tient, and confirming appropriate trajectory of the beam.In most cases, high-quality radiographs will allow thefracture pattern to be understood. When they do not,thin-cut computed tomography (CT) scanning can bedone. This modality allows the option of 3-dimensionalreformatting, which can be particularly helpful in theevaluation of complex fractures and malunions.

NEER’S CLASSIFICATION SYSTEM

If the necessary high-quality radiographs are obtained,Neer’s classification system3,4 is a practical and usefulguide to treatment, particularly when applied to fracturesof the surgical neck. Fractures with less than 1 cm of dis-placement or 45 degrees of angulation of the head frag-ment (relative to the shaft) can generally be treated by aclosed method. However, these traditional guidelines3,4

may not apply to the greater tuberosity. Others have sug-gested that less than 1 cm of displacement of the greatertuberosity (approximately 5–10 mm) may be significantin some patients.5–7 This small degree of displacementcan be difficult to determine, given the variability in indi-vidual anatomy. Therefore, comparison views of the oppo-site shoulder with the arm held in a similar degree ofexternal rotation can be useful.

The most common fracture patterns that require sur-gical treatment are 2-part fractures (with shaft displace-ment), 3-part fractures (with shaft and tuberosity dis-placement), and 4-part fracture-dislocations (typicallyoccurring in elderly patients). Less common patterns in-clude the isolated 2-part greater tuberosity fracture, the 2-part anatomic neck fracture (with displacement of thehead segment only), and the so-called valgus-impactedpattern.

ORTHOPAEDIC SURGERY BOARD REVIEW MANUAL

Surgical Treatment of Fractures of the Proximal Humerus

This valgus-impacted “four-part” pattern8–10 (Figure 1)seen in some fractures of the proximal humerus deservesparticular attention. Subtle impaction of the lateral aspectof the head can often mimic greater tuberosity displace-ment. If this fracture pattern is not appreciated prior tosurgery, the greater tuberosity may be irreducible. Al-though all 4 segments are fractured, the displacementstypical of this pattern are generally not enough to qualifyas true 4-part fractures, according to Neer’s criteria. Therate of avascular necrosis after this type of fracture is low.

ANATOMIC CONSIDERATIONS

The blood supply to the humeral head has been welldescribed11,12 and should be respected during internalfixation. During surgery, this means avoiding soft-tissuedisruption in the region of the arcuate artery that runsnear the lateral aspect of the bicipital groove. Equallyimportant is an awareness of the course of the axillarynerve and its vulnerability during exposure and fixationof fractures that involve displacement of the greatertuberosity. Protection of the axillary nerve is most easilyaccomplished by abducting the arm slightly to relax thedeltoid muscle. These maneuvers become particularlyuseful in cases that are treated late and in cases of revi-sion surgery in which extensive callus or scar tissueobscures the anatomical planes.

The points available for fixation in the proximal hu-merus include the shaft, head fragment, and tuberosi-ties. The shaft is characterized by its hollow tubularstructure and constant availability as a point of fixationfor plates and screws, intramedullary devices, and pros-thetic stems (in combination with bone cement). Thetuberosities are characterized by their fragility, frag-mentation, and consistency of rotator cuff tendon at-tachment. These tendon attachments serve as a secondpoint of consistent and reliable fixation. The quality ofthe humeral head fragment varies with the age of thepatient and the pattern of the fracture. In elderlypatients, the head fragment is most often a thin, hollowshell of bone that precludes rigid fixation but can besupported by a bone graft or a fixed-angle device suchas a blade plate.13 In contrast, younger patients oftenhave ample high-quality cancellous bone available inthe humeral head that allows adequate purchase andfixation with standard bone screws.

OTHER AGE-RELATED CONSIDERATIONS

In addition to bone quality, other patient charac-teristics associated with age influence decision makingin the management of fractures of the proximal hu-merus. For example, younger patients tend to experi-ence high-energy trauma that often results in associat-

ed dislocations and soft-tissue trauma. In this pop-ulation, there is a high propensity of postoperative stiffness, so the goal of achieving stable fixation be-comes paramount. Fortunately, stable fixation can beachieved consistently when there is adequate bonequality and good postoperative compliance with pas-sive exercises. In contrast, older patients are moreoften subject to low-energy trauma (eg, falls), typicallyhave poorer bone quality, and often have limitations ofboth cognitive capacity and the social support neces-sary to protect the fixation by avoiding active use of thearm during the early postoperative period.

A patient’s ability to comply with postoperative in-structions becomes critical when fractures of the proxi-mal humerus are treated with replacement arthroplasty,because this surgical procedure also mandates protectionand rehabilitation postoperatively to allow the opportu-nity for healing between the greater tuberosity and shaft.Thus, it is not surprising that results of hemiarthroplastyin elderly patients have been only fair at best.


S u r g i c a l T r e a t m e n t o f F r a c t u r e s o f t h e P r o x i m a l H u m e r u s

Figure 1. An example of a valgus-impacted fracture pattern.Although there are 4 fractured segments, each is not displacedenough to qualify as a true 4-part fracture according to Neer’s sys-tem of classification. Note that the greater tuberosity is minimallydisplaced relative to the shaft. The head has been impacted into avalgus position relative to the shaft. (Reprinted with permissionfrom Moehring H, Greenspan A, editors. Fractures: diagnosis andtreatment. New York: McGraw-Hill/Appleton & Lange; 2000:218.)

SURGICAL PROCEDURES

INTRAOPERATIVE SETUP AND POSITIONING

Regardless of the planned procedure (eg, closed re-duction with percutaneous fixation, open reduction withinternal fixation, hemiarthroplasty), the intraoperativesetup and positioning used in repair of fractures of theproximal humerus remain consistent. Key features in-clude neutral positioning of the cervical spine and lateralpositioning of the shoulder to allow access to the humeralcanal (Figure 2). Prior to preparing and draping thepatient, the surgical team should confirm that fluoroscop-ic images of the entire humeral head can be obtained, inorder to ensure that similar images will be available asneeded during the process of internal fixation.

Particular attention should be directed to cleansingof the skin. Many fractures are treated surgically severaldays after the injury, and most patients find mainte-nance of axillary hygiene impossible during this inter-val. Subsequent bacterial growth in the region of thefracture can become substantial and theoretically con-tribute to deep infection postoperatively.14

SURGICAL APPROACHESPercutaneous Approach

In certain cases, a percutaneous approach is valu-able. Terminally threaded wires (eg, 2.5 mm Schanzpins) are typically used.15 Although other devices (eg,cannulated screws, nails) can be applied in a percuta-neous fashion, the following guidelines recently setforth by Gerber16 still apply: good bone quality, the abil-ity to obtain a satisfactory closed reduction, satisfactorystability during fluoroscopic examination, and goodpatient compliance with the necessary period of post-operative immobilization should be present.

A percutaneous approach is well suited for fractureswith minimal comminution. In younger patients, thistype of fracture typically means a 2-part surgical neckfracture. The application of this technique to 3- and 4-part fractures is more challenging and may not bepossible in all cases.

In patients with poor bone quality, percutaneous orminimally invasive fixation can still be possible with useof an intramedullary device. As with any nailing proce-dure, satisfactory closed reduction should ideally beobtained prior to nail insertion. Equally essential is theestablishment of an appropriate starting point for thedevice. If a straight implant is placed through the greatertuberosity, the nail will invariably push the head segmentinto varus alignment as it enters the intramedullarycanal of the humerus.

A unique and relatively simple intramedullary devicewas developed by Evans approximately 2 decades ago17

when he devised a long prestressed stainless steel staplethat could be implanted through a small incision and del-toid split. This implant provides rotational stability bygaining purchase in the subchondral bone of the humer-al head and endosteal surface of the humeral shaft.

Anterosuperior Approach

Depending on the pattern and complexity of thefracture, a deltoid-splitting incision or even a formalanterosuperior approach with removal of a portion ofthe deltoid origin from the acromion can be applied.If this approach is selected for the treatment of frac-tures of the greater tuberosity, care should be taken, asalways, to avoid excessive distal propagation of the split



Figure 2. A view from behind the operating table demonstrat-ing the important features of patient positioning: lateral posi-tioning of the shoulder to allow access to the intramedullarycanal of the humerus and neutral positioning of the cervicalspine. (Reprinted with permission from Moehring H, GreenspanA, editors. Fractures: diagnosis and treatment. New York:McGraw-Hill/Appleton & Lange; 2000:220.)

and subsequent injury to the axillary nerve from over-zealous retraction.

Long Deltopectoral Approach

Almost all fractures of the proximal humerus can be addressed through a long deltopectoral approach(Figure 3), particularly if the approach is modified bythe following 6 steps: (1) the incision is placed slightlylateral to the deltopectoral interval, extending from theclavicle to the deltoid origin; (2) the deltopectoralinterval is developed in its entirety from the claviclethrough the deltoid origin; (3) the plane between thedeltoid and the coracoacromial ligament is developed;(4) a subbursal exposure is performed with completerelease of adhesions that can limit mobility of the rota-tor cuff, humeral shaft, and fracture fragments of theproximal humerus; (5) a deltoid retractor is used toeffectively displace the deltoid muscle mass posteriorlyand lever the proximal humerus forward; (6) retentionsutures are placed in the supraspinatus, infraspinatus,and subscapularis muscles, allowing the tuberosity frac-ture fragments to be controlled and reduced.

IMPLANTS

A variety of implants are available for use in surgicaltreatment of fractures of the proximal humerus, includ-ing sutures, flexible wires, rigid pins, screws, intramed-ullary nails, and plates. In actuality, combinations of thesedevices are often used to achieve stable fixation.

Sutures

Sutures have the desirable qualities of radiolucency,simplicity of use, and softness of texture. Heavy suturematerial is perhaps the material of choice for fixation ofthe tuberosities (with or without a prosthesis), particu-larly when the tuberosities are comminuted and thesuture is placed at the bone-tendon junction. However,suture fixation alone typically does not provide ade-quate rigidity to control larger segments of bone, suchas the head relative to the shaft in surgical neck frac-tures. In these situations, tension-band wiring as de-scribed by Hawkins and colleagues,18 with or withoutadditional intramedullary fixation, has proved effective.The latter method is particularly attractive in elderlypatients with poor bone quality in whom the only avail-able points of fixation are the shaft and the rotator cuff.

Intramedullary Nails

Although intramedullary nails can be effective inrepair of fractures of the proximal humerus, they cannecessitate violation of the rotator cuff. At times, thenails can become very prominent and cause subacro-

mial impingement and wear of the supraspinatus ten-don (Figure 4).

Plates

Plate fixation has a definite role in the treatment offractures of the proximal humerus. Indications for useof plates include fixation of fractures extending distallyinto the diaphysis, severe metaphyseal comminution,and support of a fragile head segment, particularly afterelevation from a valgus-impacted position.

Some authors have suggested that plate fixation ofthe proximal humerus can cause avascular necrosis.19

This potential complication may be related more to in-jury to the arcuate artery (supplying blood to the hu-meral head) than to the implants. Esser recently report-ed a very low incidence of avascular necrosis with platefixation.20 Of course, posttraumatic avascular necrosisof the humeral head can and does occur in the absenceof surgical treatment. If plates are used, they should bepositioned appropriately low on the tuberosity to avoidiatrogenic subacromial impingement. Plates can alsoserve as an important point of attachment for multipletension band sutures. This additional fixation can neu-tralize the pull of the rotator cuff and prevent late hard-ware failure and varus deformity.

RESULTS OF INTERNAL FIXATION

Analysis of the available literature on internal fixationof fractures of the proximal humerus is somewhat limited



Figure 3. An example dissection in a fresh cadaver specimendemonstrating how the entire proximal humerus (including theposterior aspect of the greater tuberosity) can be exposed with amodified extensile deltopectoral approach. The deltoid origin ispreserved in its entirety. (Reprinted with permission fromMoehring H, Greenspan A, editors. Fractures: diagnosis and treat-ment. New York: McGraw-Hill/Appleton & Lange; 2000:221.)

by variations in the fracture patterns reported, theimplants used, the follow-up period established, and thegrading scales applied. Despite these variations, however,

satisfactory results are achieved in approximately 80% ofcases. Poor prognostic factors include advanced age, poorbone quality, and more severe fracture patterns (ie, 3- and4-part fractures)—particularly when associated with a dis-location of the humeral head fragment.

Regardless of the severity of the injury or the type offixation device selected, anatomic reduction is crucial.21

Gerber’s work suggests that if the tuberosities and headsegment can be maintained in an anatomic or near-anatomic position through fracture healing, the clinicalresults are consistently satisfactory, even in the presenceof posttraumatic avascular necrosis.

HUMERAL HEAD REPLACEMENT IN ACUTEFRACTURES

GENERAL CONSIDERATIONS

In 1970, Neer published his classic articles on fracturesof the proximal humerus.3,4 In part II of the series, theresults of treatment of 4-part displacements were report-ed.4 With this severe pattern of injury, both nonoperativetreatment and open reductions led to dismal results inevery case3,4; only humeral head replacement (HHR) pro-duced a satisfactory outcome consistently. Although oth-ers have also published their experience,22–30 Neer’s workremains influential and has directed treatment for thepast 30 years.

During the past 2 decades, most 4-part fracturesrequiring HHR have been treated with implants original-ly designed to treat glenohumeral arthritis. The resultshave been variable. More recently, implants specificallydesigned to treat fractures of the proximal humerus havebeen developed in hope of improving clinical results.

INDICATIONS FOR HUMERAL HEAD REPLACEMENT

Recently, the indications for HHR in the setting ofacute trauma have become more refined. In general,the use of HHR is reserved for elderly patients with poorbone quality and fracture patterns that preclude salvageof the humeral head or stable internal fixation (eg,head-splitting fractures, Neer 4-part fractures, selectedNeer 3-part fractures, fracture-dislocations in patientswith poor bone quality and a very small head fragment,selected severe impression fractures that involve morethan 40% of the articular surface, selected anatomicneck fractures in which internal fixation is not possible).

Typically, in patients younger than 60 years, an at-tempt is made to salvage the proximal humerus ratherthan to replace it. Moreover, the advent of improvedimplants for internal fixation, a new understanding of



Figure 4. Radiograph showing iatrogenic subacromial impinge-ment from an intramedullary nail (A) and intraoperative photo-graph of a rotator cuff tear caused by an intramedullary nail (B).(Reprinted with permission from Moehring H, Greenspan A,editors. Fractures: diagnosis and treatment. New York:McGraw-Hill/Appleton & Lange; 2000:221.)

A

B

the clinical significance of posttraumatic avascular necro-sis,21 and the relatively high rate of complications report-ed with HHR in cases of acute trauma14,31 have resultedin a trend away from HHR toward internal fixation inthese patients. Additionally, a group from Paris has re-cently developed a new implant, the Bilboquet device,which may further limit the indications for HHR foracute fractures, even in the setting of poor bone quality.32

SURGICAL TECHNIQUEPreliminary Considerations

If HHR is attempted, the objective should be toreconstruct the proximal humerus anatomically. Theimportance of prosthetic head size, implant position,and tuberosity position have been emphasized by sever-al authors.33–37 These studies suggest that HHR has lit-tle chance of restoring satisfactory function of theshoulder unless the height of the humeral head, thedegree of retroversion, the humeral head size, and thetuberosity position are anatomic or near-anatomic (ie,within a few millimeters or degrees). Achieving this goalconsistently requires good planning, instrumentation,and surgical technique.

Preoperative Planning

Careful preoperative planning is essential beforeHHR. A high-quality radiograph of the opposite shoul-der is useful. The single best view is an anteroposteriorview of the scapular plane with the arm held in externalrotation. Because the goal of surgery is to recreate nor-mal anatomy, templating the opposite uninjured proxi-mal humerus helps determine appropriate stem size,head size, head offset, and tuberosity position. Preop-erative templating complements intraoperative guide-lines such as soft-tissue tension. A full-length view of thehumerus with radiographic markers also can be usefulto assist in determining true arm length.

Patient Positioning

The beach-chair position is recommended forHHR. Several details are worth emphasizing. The legsand feet should be elevated, with hip flexion limited to90 degrees (or less) and the table placed in a slightTrendelenburg position. These steps minimize lowerextremity venous stasis and lessen the tendency of larg-er patients to slide down the table during the proce-dure. The cervical spine is positioned neutrally to min-imize the chance of stretch injury to the cervical cordand brachial plexus. Lateral positioning of the shoulderallows surgical access to the humeral canal and im-proved fluoroscopic imaging.

Skin Preparation

Particular attention should be directed to cleansingof the skin before HHR. As previously stated, manyfractures are repaired surgically several days after theinitial injury, and most patients are unable to maintainaxillary hygiene during the intervening period. Sub-sequent microbial growth in the region of the fracturecan be substantial and can contribute to a postopera-tive deep infection.14 For this reason, after anesthesia isinduced, prescrubbing the axilla with alcohol is rec-ommended prior to proceeding with the standard sur-gical preparation.

Surgical Approach and Exposure

A modification of the deltopectoral approach is usedin HHR. The entire deltopectoral interval from theclavicle to the humerus is developed. Specifically, allveins draining the deltoid (between the medial borderof the muscle and the cephalic vein) are cauterized orligated. Leaving the cephalic vein with the pectoralismajor muscle allows better mobility of the deltoid andgreater ease in exposing the posterior aspect of thegreater tuberosity and shaft. The clavipectoral fascia isincised, and the subacromial/subdeltoid bursa is en-tered. Slight abduction of the arm at this point relaxesthe deltoid muscle and protects the terminal branchesof the axillary nerve. Periosteal attachments of the shaftand greater tuberosity are preserved to enhance heal-ing after repair.

Deeper exposure of the dislocated humeral headfragment typically is achieved by working through a lon-gitudinal fracture line in the greater tuberosity. Thisexposure is most often associated with a small longitudi-nal split supraspinatus tendon, which can be surgicallyenlarged towards the superior glenoid labrum, as need-ed, to gain access to the head fragment, glenoid, andupper portion of the humeral shaft. Although manytexts illustrate 4-part fractures as occurring through thebicipital groove, most often the hard bone of the grooveremains intact, with a coronal fracture line occurringjust posterior to it.

Once exposed, the glenoid articular surface isassessed. Unless the patient suffers from an asso-ciated disease (eg, inflammatory arthritis), the gle-noid cartilage is usually well preserved and does notmerit replacement. Next, the shaft of the humerus isgently exposed, taking care to avoid iatrogenic strip-ping of the periosteum. At this point, drill holes andsutures should be placed in the upper portion of theshaft in anticipation of tuberosity repair after cemen-tation.



Trial Reduction

The trial reduction process in cases of HHR can bedivided into at least 3 steps: (1) restoration of humer-al length, (2) restoration of appropriate humeral headretroversion, and (3) restoration of humeral head off-set or size.

The length of the arm is determined by the heightof the prosthesis. When the humeral head is posi-tioned too low, permanent inferior subluxation canoccur. If the implant is too high, the tuberosity androtator cuff repair will come under excessive tensionand may fail. Perhaps the most accurate method ofrestoring humeral length is to use a ruler. To this end,Boileau and Walch have developed an external frac-ture jig that has enabled surgeons more consistently torestore the length of the humerus during arthroplastyfor fractures,34 and Williams and Rockwood have devel-oped an internal jig that clamps to the diaphysis. Bothjigs hold the trial implant in place during trial reduc-tion and allow clinical assessment of height, retrover-sion, and tension of the rotator cuff with the tube-rosities reduced. Similarly, Frankle has developed amodular system that allows the trial stem to be press fitfor stability. The trial implant bodies are etched in 5-mm increments. Based on the preoperative plan, asurgeon can calculate how far above the shaft frag-ment to set the head. If these newer systems are notavailable, the trial implant is either held in place witha clamp or press fit with a sponge; the height and retro-version are then marked.

Several intraoperative guidelines can also be used toassess humeral height: (1) the tension in the long head

of the biceps, (2) the tension of the reduced tuberosi-ties, (3) the distance between the supraspinatus tendonand the coracoacromial ligament, and (4) the upperborder of the anatomically reduced greater tuberosity.The latter is perhaps the most useful landmark. In mostshoulders, the apex of the humeral head rests justabove the greater tuberosity. This relationship and itsvariations can be appreciated on the radiograph of theopposite shoulder used for templating. During surgery,with the trial implant in place and the greater tuberosi-ty reduced and provisionally fixed, a radiograph or flu-oroscopic image of the shoulder can be obtained toconfirm clinical impressions about the position of theimplant and tuberosities.

Retroversion

In the setting of HHR to treat acute fractures of thehumeral head, reliable landmarks are often absent,resulting in a tendency of the surgeon to malrotate thehumeral component. It should be recognized that greatvariability exists in the degree of retroversion of thehumerus,34,38–41 with the average value measuring20 degrees relative to the epicondylar axis.34,41 If theforearm is used as a reference point, approximately25 to 30 degrees of retroversion would be appropriate inmost cases. Boileau and Walch have pointed out thenegative consequences of excessive retroversion.34 If thegreater tuberosity is reduced and repaired under a pros-thetic head that is placed in too much retroversion, therepair will come under excessive tension when the armis brought into internal rotation against the abdomen.Posterior tuberosity displacement cannot always be ap-preciated on radiographs and may be one cause oftuberosity resorption. Perhaps a more common cause oftuberosity migration is the failure of the shaft of thehumerus to heal.

Cementation

Cement is used to gain immediate stability of theimplant. However, the proximal metaphysis should notbe filled with cement, because it is only the area avail-able for tuberosity healing (Figure 5). Given the rela-tively high risk of infection, the use of antibiotics in thecement is reasonable.

Tuberosity Repair

Conditions for healing of the tuberosity to the shaftcan be optimized by avoiding the tendency to excessive-ly trim or thin the tuberosity or to fill the metaphysis withmetal and cement and by selecting a humeral head sizesimilar to that of the fractured head or of the humeral



Figure 5. A radiograph showing a well healed tuberosity follow-ing hemiarthroplasty. (Reprinted with permission from MoehringH, Greenspan A, editors. Fractures: diagnosis and treatment.New York: McGraw-Hill/Appleton & Lange; 2000:223.)

head of the opposite side. Adjunctive cancellous bonegrafting from the resected humeral head is advisable, andsecure repair of the tuberosities is essential. The heavysuture placed in the configuration described by Hawkinsand colleagues18 typically provides satisfactory stability.Additional fixation in compression can be achieved byplacing cerclage sutures or wires around the medial neckof the implant and the tuberosities.

REHABILITATION AFTER HUMERAL HEAD REPLACEMENT

Once true anatomic reconstruction has been achievedin HHR (ie, the length of the arm is appropriate, thehumeral head size and retroversion are appropriate, thetuberosity position is correct, the osteosynthesis is secure),near full passive range of motion should be possible. Inpractice, rehabilitation must be individualized in eachcase because of variations in bone quality and patientunderstanding. After repair of the tuberosities is achieved,the arm is taken through passive range of motion whilethe osteosynthesis site is examined directly. The limit atwhich passive motion brings the repair under stress isrecorded. Postoperatively, supine passive elevation and ex-ternal rotation are prescribed, to a limit of 5 to 10 degreesless than the intraoperative measurements. Occasionally,the tuberosity is so comminuted that a secure repair can-not be achieved, in which case a rehabilitation programwith more limited goals is prescribed.3,4

RESULTS OF HUMERAL HEAD REPLACEMENTAssessing Outcome

Many grading systems have been used to assess theoutcome of shoulder arthroplasty as treatment of frac-tures of the proximal humerus. Although this variabilityprecludes a highly accurate comparison of results, moststudies do report at least some shoulder range of motionand comfort. In our review of published series, there wasa wide range in the values reported for mean elevationrange of motion (ranging from 58 to 131 degrees).27,42

The severity of pain experienced by patients after HHRwas also variable; patients with no pain or mild painranged from 60% to 95%.

A review of the Mayo Clinic’s experience with HHRfor acute fractures was recently updated. Including casesfrom 2 previously published series,14,28 75 HHRs were per-formed at Mayo between 1970 and 1998. The average follow-up interval was 40 months. Of the 75 patients in-volved in these procedures, 66 (88%) reported mild orno pain. Range of motion on active elevation and exter-nal rotation averaged 101 and 38 degrees, respectively.

These data and those of others25,43,44 suggest thatpatients treated with HHR for acute fractures are usu-

ally free of severe pain. Their functional outcomeseems to be less predictable. Many patients are leftwith a restricted range of motion, especially on activearm elevation. Newer implants and techniques41 mayallow average results to improve to the relatively highstandard set by Neer some 30 years ago. The factorsaffecting outcome have been extensively discussed inthe literature and include fracture pattern, timing ofsurgery, patient’s age, position of the implant, andaftercare.22,24,25,34,45

Fracture Pattern

Two reports suggested better functional outcome for3-part fractures treated primarily with HHR than for 4-part fractures.25,44 These results are not entirely unex-pected, given the larger fragments of tuberosity avail-able for repair in the setting of 3-part fractures.

Timing of Surgery

There is some rationale for early treatment of frac-tures of the proximal humerus with HHR, prior to thedevelopment of an unwanted scar, contractures, andbone deformity. Tanner and Cofield compared the re-sults of HHR in patients with acute and chronic frac-ture, including patients with malunion; the complica-tion rate was higher for cases in which HHR wasdelayed.28 In a study by Bosch and colleagues, the lengthof time between injury and hemiarthroplasty was thebest predictor of outcome in most cases; the outcomewas inversely proportional to the interval between injuryand hemiarthroplasty.46 The results were significantlybetter after early operation, despite the fact that thefracture pattern was more severe in patients undergoingearly operation.46

Age

Patients of advanced age (ie, older than 70 years)who undergo HHR also tend to have less satisfactoryresults.25 This finding possibly results from the fact thatelderly patients may be less determined than areyounger patients to pursue a rigorous rehabilitation pro-gram after HHR because of their limited functionalneeds. Wretenberg and Ekelund29 also noted inferiorresults in their series in which patients who were, onaverage, age 82 years obtained only approximately55 degrees of elevation; they added that these patientsusually had decreased elevation in the contralateral non-operated shoulder as well, indicating that this decreasedrange of motion was somewhat related to physiologicage changes. Finally, in older patients, tuberosity frag-ments are more comminuted and osteoporotic, making



fixation and union less predictable. Poor functionalresults of arthroplasty in elderly patients are typically off-set by their high rate of satisfaction, because pain reliefplays a much greater role than does the demand forphysical performance. Of interest, the mean age of pa-tients in Neer’s study was 55 years. This considerationmay have contributed greatly to the high rate of satisfac-tory results.

Prosthesis Malposition

Perhaps the most common reason for failure inHHR is prosthetic malposition. Boileau and Walch34

found that differences between the original anatomyand the prosthetic design led to altered biomechanics.In their clinical series, the most significant factor asso-ciated with a poor functional outcome was malpositionof the prosthesis, especially when the prosthesis wasplaced too proud or in excessive retroversion. Theyalso found a clear correlation between prosthesis posi-tioning and greater tuberosity migration, leading tononunion, malunion and late bone resorption. A re-cent study by Frankle42 found a correlation betweenhead-to-tuberosity distance and patient functional out-come and satisfaction. A distance of 13 mm or morewas associated with poorer results.

COMPLICATIONS OF HUMERAL HEAD REPLACEMENT

Many authors have recognized the challenges ofachieving good results with HHR for acute trauma. Ananalysis of a series of 30 failed prosthetic replacementsfor displaced fractures of the proximal humerusshowed that the single greatest cause of failure wasdetachment of the greater tuberosity. Loosening of thehumeral prosthesis occurred in 13 shoulders, two ofwhich were also infected; nerve injury was diagnosed in30%, and glenoid erosion and malposition of thehumeral prosthesis occurred in 23%. Ectopic forma-tion was present in 16% of these failures.47

In a recent review by Muldoon and Cofield,14 com-plications of HHR were diverse. Wound healing prob-lems, infection, nerve injury, periprosthetic humeralfracture, component malposition, instability, tuberosityhealing problems, rotator cuff tearing, reflex sympa-thetic dystrophy, heterotopic ossification, periarticularfibrosis, component loosening, glenoid arthritis withinstability, tuberosity malunion or nonunion, and het-erotopic bone formation were the most frequent com-plications.

In the Mayo combined series of 75 cases occurringbetween 1970 and 1998, complications occurred in 25 pa-tients (33%), most often involving healing of the tuberos-ity and rotator cuff. There were 6 cases of nonunion of

one or both tuberosities, 13 cases of mild-to-moderateanterosuperior instability related to rotator cuff insuf-ficiency, 4 cases of infection (1 superficial, 3 deep), 1 caseof wound hematoma, and 1 case of reflex sympatheticdystrophy. Five reoperations were necessary (3 to treat in-fections, 1 for tuberosity repair, 1 for drainage of a hema-toma).

CONCLUSIONS

Treatment of fractures of the proximal humerus canbe challenging. Although there is a potential for frequentcomplications, many can be avoided with careful pre-operative planning and meticulous surgical techniqueand aftercare. Optimal function of the shoulder afteracute trauma requires anatomic or near-anatomic recon-struction. Newer techniques, instruments, implants mayimprove clinical results.

REFERENCES

1. Young TB, Wallace WA. Conservative treatment of frac-tures and fracture-dislocations of the upper end of thehumerus. J Bone Joint Surg Br 1985;67:373–7.

2. Tile M. Fracture of the proximal humerus. In: SchatzkerJ, Tile M, editors. The rationale of operative fracturecare. New York: Springer-Verlag; 1996.

3. Neer CS 2nd. Displaced proximal humeral fractures. I.Classification and evaluation. J Bone Joint Surg Am 1970;52:1077–89.

4. Neer CS 2nd. Displaced proximal humeral fractures. II.Treatment of three-part and four-part displacement.J Bone Joint Surg Am 1970;52:1090–3.

5. McLaughlin HL. Dislocation of the shoulder with tuberos-ity fracture. Surg Clin North Am 1963;43:1615–20.

6. Weber SC. The ten millimeter rule revisited: nonopera-tive treatment of the displaced greater tuberosity frac-tures. Paper presented at: 14th Open Meeting of theAmerican Shoulder and Elbow Surgeons; February 15,1997; San Francisco, CA.

7. Rodosky MW, Puraldi X, Pollock R, et al. Operative treat-ment of malunions of the proximal humerus. Paper pre-sented at: 13th Annual Meeting of the MinnesotaOrthopaedic Society; May 1997; Minneapolis MN.

8. Jakob RP, Miniaci A, Anson PS, et al. Four-part valgusimpacted fractures of the proximal humerus. J BoneJoint Surg Br 1991;73:295–8.

9. Resch H, Beck E, Bayley I. Reconstruction of the valgus-impacted humeral head fracture. J Shoulder Elbow Surg1995;4:73–80.



10. Resch H, Povacz P, Fröhlich R, Wambacher M. Per-cutaneous fixation of three- and four-part fractures ofthe proximal humerus. J Bone Joint Surg Br 1997;79:295–300.

11. Gerber C, Schneeberger AG, Vinh TS. The arterial vas-cularization of the humeral head. An anatomical study.J Bone Joint Surg Am 1990;72:1486–94.

12. Brooks CH, Revell WJ, Heatley FW. Vascularity of thehumeral head after proximal humeral fractures. Ananatomical cadaver study. J Bone Joint Surg Br 1993;75:132–6.

13. Jupiter JB, Mullaji AB. Blade plate fixation of proximalhumeral non-unions. Injury 1994;25:301–3.

14. Muldoon MP, Cofield RH. Complications of humeralhead replacement for proximal humeral fractures. InstrCourse Lect 1997;46:15–24.

15. Jaberg H, Warner JJ, Jakob RP. Percutaneous stabilizationof unstable fractures of the humerus. J Bone Joint SurgAm 1992;74:508–15.

16. Gerber C. Alternatives to hemiarthroplasty for complexproximal humeral fractures. In: Warner JJP, Iannotti JP,Gerber C, editors. Complex and revision problems inshoulder surgery. Philadelphia: Lippincott-Raven; 1997.

17. Evans DC, Katz I. Pre-stressed humeral staple for fullydisplaced surgical neck fractures (poster). Presented at:American Academy of Orthopaedic Surgery AnnualMeeting; February 9–14, 1989; Las Vegas, NV.

18. Hawkins RJ, Bell RH, Gurr K. The three-part fracture ofthe proximal part of the humerus. Operative treatment.J Bone Joint Surg Am 1986;68:1410–4.

19. Sturzenegger M, Fornaro E, Jakob RP. Results of surgicaltreatment of multifragmented fractures of the humeralhead. Arch Orthop Trauma Surg 1982;100:249–59.

20. Esser RD. Treatment of three- and four-part fractures ofthe proximal humerus with a modified cloverleaf plate.J Orthop Trauma 1994;8:15–22.

21. Gerber C, Hersche O, Berberat C. The clinical relevanceof posttraumatic avascular necrosis of the humeral headJ Shoulder Elbow Surg 1998;7:586–90.

22. Ambacher T, Erli H, Parr O. Significance of rehabilita-tion to the functional outcome after primary hemi-arthroplasty of the humeral head fractures. AcktuelleTraumatology 2000;30:20–5.

23. Boss AP, Hintermann B. Primary endoprosthesis in com-minuted humeral head fractures in patients over 60 yearsof age. Int Orthop 1999;23:172–4.

24. Dines DM, Warren RF. Modular shoulder hemiarthro-plasty for acute fractures. Surgical considerations. ClinOrthop 1994;307:18–26.

25. Goldman RT, Koval KJ, Cuomo F, et al. Functional out-come after humeral head replacement for acute three–and four-part proximal humeral fractures. J Shoulder

Elbow Surg 1995;4:81–6.

26. Moeckel BH, Dines DM , Warren RF, Altchek DW.Modular hemiarthroplasty for fractures of the proxi-mal part of the humerus. J Bone Joint Surg Am 1992;74:884–9.

27. Movin T, Sjoden O, Ahregart L. Poor function aftershoulder replacement in fracture patients. A retrospec-tive evaluation of 29 patients followed for 2–7 years. ActaOrthop Scand 1998;69:392–6.

28. Tanner MW, Cofield RH. Prosthetic arthroplasty for frac-tures and fracture-dislocations of the proximal humerus.Clin Orthop 1983;179:116–28.

29. Wretenberg P, Ekelund A. Acute hemiarthroplasty afterproximal humerus fracture in old patients. A retrospec-tive evaluation of 18 patients followed for 2–7 years. ActaOrthop Scand 1997;68:121–3.

30. Zyto K, Wallace WA, Frostick SP, Preston BJ. Outcomeafter hemiarthroplasty for three- and four-part fracturesof the proximal humerus. J Shoulder Elbow Surg 1998;7:85–9.

31. Brown TD, Bigliani LU. Complications with humeral headreplacement. Orthop Clin North Am 2000;31:77–90.

32. Doursounian L, Grimberg J, Cazeau C, et al. A new in-ternal fixation technique for fractures of the proximalhumerus—the Bilboquet device: a report on 26 cases.J Shoulder Elbow Surg 2000;9:279–88.

33. Williams GR Wong K Pepe M, et al. Effect of humeralarticular malpostion after total shoulder arthoplasty onimpingement range of motion and translation. Paper pre-sented at: 15th Open Meeting of the American Shoulderand Elbow Surgeons; February 7, 1999; Anaheim, CA.

34. Boileau P, Walch G. Shoulder arthroplasty for proximalhumeral fractures: problems and solutions. In: Walch G,Boileau P, editors. Shoulder arthroplasty. New York:Springer-Verlag; 1998:297–314.

35. Nicholoson GP, Duckworth MA. Operative treatment ofproximal humeral malunions radiographic and clinicalassessment. Paper presented at: 15th Open Meeting ofthe American Shoulder and Elbow Surgeons; February 7,1999; Anaheim, CA.

36. Coumo F, Lodenberg M, Jones D, Zuckerman JD. Theeffect of greater tuberosity placement on active range ofmotion after hemiarhroplasty for acute fractures of theproximal humerus. Paper presented at: 16th OpenMeeting of the American Shoulder and Elbow Surgeons;March 2000; Orlando, FL.

37. Frankle M, Greenwald D, Markee B, Ondrovic L. Thebiomechanical effects of malposition of tuberosity frag-ments on the humeral prosthetic reconstruction forfour-part proximal humeral fractures. Paper presentedat: 17th Open Meeting of the American Shoulder andElbow Surgeons; March, 2001; San Fancisco, CA.

38. Krahl VE, Evans FG. Humeral torsion in man. Am J Phys



Anthrop 1945;3:229–53.

39. Roberts SN, Foley AP, Swallow HM, et al. The geometryof the humeral head and the design of prostheses. J BoneJoint Surg Br 1991;73:647–50.

40. Walch G, Boileau P. Morphological study of the proximalhumerus. J Bone Joint Surg Br 1991;74(Suppl):14.

41. Neer CS. Articular replacement for the humeral head.J Bone Joint Surg Am 1955;37:215–28.

42. Frankle M. Outcomes of prosthetic replacements for acutefractures of the proximal humerus. Paper presented at:17th Open Meeting of the American Shoulder and ElbowSurgeons; March 2001; San Francisco, CA.

43. Frich LH, Sojbjerg JO, Sneppen O. Shoulder arthroplas-ty in complex acute and chronic proximal humeral frac-

tures. Orthopedics 1991;14:949–54.

44. Kristiansen B, Christensen SW. Proximal humeral frac-tures. Late results in relation to classification and treat-ment. Acta Orthop Scand 1987;58:124–7.

45. Compito CA, Self EB, Bigliani LU. Arthroplasty andacute shoulder trauma. Reasons for success and failure.Clinical Orthop 1994;307:27–36.

46. Bosch U, Skutek M, Fremerey RW, Tscherne H. Out-come after primary and secondary hemiarthroplasty inelderly patients with fractures of the proximal humerus.J Shoulder Elbow Surg 1998;7:479–84.

47. Bigliani LU, Flatow EL, McCluskey GM, Fischer RA.Failed prosthetic replacement in displaced proximal hu-merus fractures. Orthop Trans 1991;15:747–8.



Copyright 2001 by Turner White Communications Inc., Wayne, PA. All rights reserved.

publishing staff surgical treatment of … “four-part” pattern 8–10(figure1) seen in some...

Documents