Surgical Treatment of Three and Four-Part Proximal Humeral Fractures

Transcription

1 This is an enhance PDF from The Journal of Bone an Joint Surgery The PDF of the article you requeste follows this cover page. Surgical Treatment of Three an Four-Part Proximal Humeral Fractures Brian D. Solberg, Charles N. Moon, Dennis P. Franco an Guy D. Paiement J Bone Joint Surg Am. 2009;91: oi: /jbjs.h This information is current as of July 1, 2009 Reprints an Permissions Publisher Information Click here to orer reprints or request permission to use material from this article, or locate the article citation on jbjs.org an click on the [Reprints an Permissions] link. The Journal of Bone an Joint Surgery 20 Pickering Street, Neeham, MA

2 1689 COPYRIGHT Ó 2009 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED Surgical Treatment of Three an Four-Part Proximal Humeral Fractures By Brian D. Solberg, MD, Charles N. Moon, MD, Dennis P. Franco, MD, an Guy D. Paiement, MD Investigation performe at Cears Sinai Meical Center, Los Angeles, California Backgroun: Optimal surgical management of three an four-part proximal humeral fractures in osteoporotic patients is controversial, with many avocating prosthetic replacement of the humeral hea. Fixe-angle locke plates that maintain angular stability uner loa have been propose as an alternative to hemiarthroplasty for the treatment of some osteoporotic fracture types. Methos: The recors of 122 consecutive patients who were fifty-five years of age or oler an in whom a Neer three or four-part proximal humeral fracture ha been treate surgically between January 2002 an November 2005 were stuie retrospectively. After exclusions, thirty-eight patients treate with a locke-plate construct were compare with forty-eight patients who ha unergone hemiarthroplasty. All patients ha raiographic an clinical follow-up at a minimum of twenty-four months an an average of thirty-six months. Reuction an implant placement were evaluate raiographically. Clinical outcomes were measure with use of the Constant-Murley system. Results: The mean Constant score (an stanar eviation) at the time of final follow-up was significantly better in the locke-plate group (68.6 ± 9.5 points) than in the hemiarthroplasty group (60.6 ± 5.9 points) (p < 0.001). The Constant scores for the three-part fractures in the locke-plate an hemiarthroplasty groups were 71.6 an 60.4 points (p < 0.001), respectively, an the scores for the four-part fractures in those groups were 64.7 an 60.1 points (p = 0.19), respectively. Patients with an initial varus extension eformity in the locke-plate group ha significantly worse outcomes than those with a valgus impacte pattern (Constant score, 63.8 compare with 74.6 points, respectively; p < 0.001). Complications in the group treate with locke-plate fixation inclue osteonecrosis in six patients, screw perforation of the humeral hea in six patients, loss of fixation in four patients, an woun infection in three patients. Loss of fixation was seen only in patients with >20 of initial varus angulation of the humeral hea. Complications in the hemiarthroplasty group inclue nonunion of the tuberosity in seven patients an woun infection in three patients. Conclusions: In this series, open repair with use of a locke plate resulte in better outcome scores than i hemiarthroplasty in similar patients, especially in those with a three-part fracture, espite a higher overall complication rate. Open reuction an internal fixation of fractures with an initial varus extension pattern shoul be approache with caution. Level of Evience: Therapeutic Level III. See Instructions to Authors for a complete escription of levels of evience. Proximal humeral fractures are common an have a bimoal age istribution 1,2. Fracture-islocations in younger patients result from high-energy injuries, an most surgeons attempt open reuction an internal fixation if at all possible 3-7. Osteoporotic fractures in elerly patients are commonly associate with minor trauma such as groun-level falls, an most are minimally isplace impacte fractures that can be treate successfully with nonoperative means 2,8.However, the optimal surgical management of three an four-part proximal humeral fractures in elerly osteoporotic patients remains controversial, with many avocating prosthetic replacement of the humeral hea Recent avances in fracture fixation technology have le to the evelopment of fixe-angle locke plates that maintain angularstabilityunerloa Biomechanical ata suggest that these implants can resist physiologic loas in osteoporotic Disclosure: The authors i not receive any outsie funing or grants in support of their research for or preparation of this work. Neither they nor a member of their immeiate families receive payments or other benefits or a commitment or agreement to provie such benefits from a commercial entity. No commercial entity pai or irecte, or agree to pay or irect, any benefits to any research fun, founation, ivision, center, clinical practice, or other charitable or nonprofit organization with which the authors, or a member of their immeiate families, are affiliate or associate. J Bone Joint Surg Am. 2009;91: oi: /jbjs.h.00133

3 1690 bone an may provie an alternative to hemiarthroplasty 19. Clinical series have emonstrate some success with the use of these plates for two-part fractures, but their clinical utility for three an four-part fractures remains unclear Substantial rates of complications, incluing loss of fixation, humeral hea perforation, an mechanical impingement, have been reporte We retrospectively compare the outcomes of patients in whom a three or four-part proximal humeral fracture ha been treate with either (1) open reuction an internal fixation with a locke plate or (2) hemiarthroplasty with cement. Materials an Methos The cases of 122 consecutive patients in whom a Neer three or four-part fracture ha been treate surgically at our institution between January 2002 an November 2005 were retrospectively stuie. Institutional review boar approval was obtaine for a retrospective review of patient recors an raiographs, an informe consent was obtaine from all patients in the final stuy group. Inclusion criteria inclue low-energy three or four-part proximal humeral fractures in patients fifty-five years of age or oler treate with a locke plate or a hemiarthroplasty an followe clinically an raiographically for a minimum of twenty-four months. Three trauma-fellowship-traine surgeons were involve in the surgical management of all patients. Patients were exclue if they were younger than fifty-five years of age, ha unergone surgical repair through a eltoi-splitting approach, ha ie uring the review perio, ha a ocumente full-thickness rotator cuff tear at the time of the inex proceure, or ha sustaine a traumatic refracture of the humerus or glenoi or a traumatic islocation uring the follow-up perio. Fractures were classifie accoring to the Neer 2 an the Orthopaeic Trauma Association (OTA) 24 fracture classification systems on the basis of the initial raiographs, which were available for all patients, or compute tomography scans, which were available for ninety-nine of the 122 patients. Displacement of a fracture part was efine, on the basis of Neer s criterion, as greater than either 1 cm of isplacement or 45 of angulation. Fifty-one patients in whom a three or four-part fracture ha been treate with a locke plate were ientifie. Thirteen patients were exclue: eight ha incomplete follow-up, two ha ie, two ha unergone a eltoi-splitting surgical approach, an one ha fallen an sustaine a isplace glenoi fracture uring the follow-up perio. Thirty-eight patients (75%) ha complete clinical an raiographic follow-up at a minimum of twenty-four months. Seventy-one patients who ha unergone a hemiarthroplasty for the treatment of a three or four-part fracture uring the time perio specifie above were ientifie. Twentyfour were exclue: eleven ha incomplete follow-up, five ha a full-thickness rotator cuff tear ocumente at the time of the inex proceure, four ha unergone conversion to a hemiarthroplasty after open reuction an internal fixation, two ha fallen an sustaine a traumatic islocation of the prosthesis, one ha ie, an one ha sustaine a periprosthetic Fig. 1 Postoperative Y view showing superoposterior screw perforation of 3 mm. humeral fracture requiring revision. Forty-eight patients (68%) ha complete clinical an raiographic follow-up at a minimum of twenty-four months. The hemiarthroplasties were performe through a stanar eltopectoral approach with the patient in the beachchair position in all but four cases (later exclue from the stuy), in which the hemiarthroplasty was performe to convert a locke-plate construct an was one with the patient in a supine position. While a cemente humeral stem was use in all cases, the implants were prouce by three ifferent manufacturers: Zimmer (Warsaw, Iniana), DePuy (Warsaw, Iniana), an Stryker (Mahwah, New Jersey). Implant selection was not ranomize an epene on surgeon preference. Tuberosity fractures were repaire with placement of number-2 FiberWire suture (Arthrex, Naples, Floria) through the implant fins an through bone tunnels in the humeral shaft an with irect tuberosity-to-tuberosity apposition. Bone-grafting of the tuberosities was not routinely performe. Intraoperative or immeiate postoperative anteroposterior raiographs were mae with the shouler in 20 of external rotation to assess the implant position. The locke-plate repairs were performe with one of three locke proximal humeral plate systems manufacture by Synthes (West Chester, Pennsylvania), Stryker, or Zimmer. Implant selection was not ranomize an epene on surgeon preference. The surgical repair was performe, with the patient in the supine position on a raiolucent table, through a

4 1691 Fig. 2 Calculation of hea-shaft angulation in an intact humerus. stanar eltopectoral surgical approach an with image intensification brought from the ipsilateral sie. Surgical issection was unertaken between the tuberosities, an articular surface reuctions were carrie out with use of a tamp on the lateral articular margin or a periosteal elevator place uner the meial calcar segment. Tuberosity fragments were manipulate with use of Kirschner wires, sutures through the intact rotator cuff attachment on the tuberosities, or tenaculum clamps. The fractures were reuce an were provisionally fixe with Kirschner wires before the locke plate was applie. Internal fixation was applie uner image intensification in the anteroposterior an axillary projections to verify reuction, plate position, an screw lengths. Tuberosity repair was augmente by placement of nonabsorbable sutures through the rotator cuff tenons an the cephala suture holes in the plate. Raiographs (20 external rotation anteroposterior, an axillary views) were mae intraoperatively or immeiately postoperatively to assess reuction an implant position. Three patients were taken back to the operating room within twenty-four hours because of 3 mm of screw perforation of the humeral hea (Fig. 1). Patients presenting with late-onset screw perforation were examine with compute tomography scans to assess the magnitue an location of the harware prominence. All patients ha anteroposterior raiographs mae with the shouler in 20 of external rotation an axillary raiographs of both the involve an uninvolve, contralateral shouler (for comparison with the treatment sie) to assess fracture reuction, implant position, the integrity of the metaphyseal hinge, an tuberosity isplacement or migration. All raiographs were mae at a stanarize istance with use of a truesize igital format that was calibrate to within 1% of known values every 5000 images. There was no correction for magnification errors introuce by boy habitus or patient positioning. In the locke-plate group, tuberosity isplacement was assesse as the sum of the tuberosity isplacements on the anteroposterior an axillary views by overlaying one of the two true-size images over the other an measuring the greatest extent of isplacement. Hea-shaft angulation was calculate on the 20 external rotation anteroposterior raiograph by comparing a tangent to the articular surface with a line parallel to the long axis of the humeral shaft as previously escribe by Hertel et al. (Fig. 2) 25. The metaphyseal hinge istance was efine as the amount of intact metaphyseal bone (in millimeters) attache to the humeral hea fragment seen on the initial raiographs. In the hemiarthroplasty group, the humeral hea height was recore relative to the intact greater tuberosity on the uninvolve sie, with a positive value recore for a cephala TABLE I Patient Characteristics of Both Cohorts Locke Plate Hemiarthroplasty P Value Patients (no.) Age* (yr) 66.5 ± ± Duration of follow-up* (mo) 36.1 ± ± Male/female (no.) 12/26 14/34 >0.90 Three-part fracture (no.) 23 (61%) 25 (52%) 0.52 Four-part fracture (no.) 15 (39%) 23 (48%) 0.44 Dislocation (no.) 8 (21%) 10 (21%) >0.90 *The values are given as the mean an stanar eviation.

5 1692 TABLE II Comparison of Constant Scores Between Cohorts Constant Score* (points) Locke Plate Hemiarthroplasty P Value All patients Total 68.6 ± ± 5.9 <0.001 Pain 13.2 ± ± Power 15.7 ± ± Range of motion 25.1 ± ± 2.4 <0.001 Activities of aily living 15.0 ± ± 2.1 <0.001 Three-part fractures 71.6 ± ± 5.7 <0.001 Four-part fractures 64.7 ± ± Initial varus extension fracture patterns 63.8 ± 7.8 (32%) 60.2 ± 6.1 (27%) 0.14 Initial valgus impaction fracture patterns 74.6 ± 9.2 (68%) 60.3 ± 6.3 (73%) <0.001 P value < *The values are given as the mean an stanar eviation. The percentage of patients in the subgroup is given in parentheses. For the ifference between subgroups base on the fracture pattern (varus extension or valgus impaction). ifference an a negative value recore for a caua ifference. Tuberosity isplacement was etermine in the same fashion as escribe for the patients in the locke-plate group. Patients with a suspecte nonunion or progressive migration of the tuberosity unerwent a compute tomography scan. All shoulers were immobilize in a sling for the first ten ays postoperatively. Physical therapy was then starte with gentle Coman an active-assiste range-of-motion exercises within the first two weeks postoperatively. Gentle resistive exercises with unrestricte passive motion were begun at six weeks postoperatively. Patients were followe clinically an raiographically at two an six weeks postoperatively an at three-month intervals thereafter. Shouler outcomes were assesse with use of the Constant an Murley scoring system 26 at the last clinical shouler examination. The active range of motion was measure with a goniometer. Power testing was performe with use of a igital ynamometer (Ametek, Largo, Floria) with the elbow extene an the arm abucte 60. The metho of surgical treatment was chosen preoperatively on the basis of raiographs an compute tomography scans. Hemiarthroplasty was selecte for patients with an articular surface fracture or a hea-split pattern (thirty patients), isplacement of the anatomic neck of >2 cm (seventeen), impaction of the articular surface (nine), islocation of the humeral hea for more than twenty-four hours (six), a ocumente previous rotator cuff tear (six), or an inability to perform open reuction an internal fixation with an open technique (three). All other patients unerwent an attempt at open reuction an internal fixation with a locke plate. Four patients unergoing open reuction an internal fixation ha immeiate conversion to a hemiarthroplasty, two because of an inability to reuce the fracture an two because of iatrogenic intraoperative splitting of the hea fragment; these patients were exclue from the stuy. Statistical Analysis The Fisher exact test or Stuent t test was use for comparison of emographic variables between the hemiarthroplasty an locke-plate groups. The Mann-Whitney U test (Wilcoxon rank sum test) was use to compare the outcomes ata between the groups, an correlation was analyze with a Spearman correlation coefficient. The level of significance was set at p Source of Funing No external funing was receive in support of this stuy. Results With the numbers stuie, there were no apparent ifferences between the two groups with regar to age, sex, humeral hea islocation rate, or fracture type (Table I). The patients in the locke-plate group unerwent surgical treatment at an average of 4.8 ays after the injury compare with an average of 6.9 ays in the hemiarthroplasty group (p = 0.24). The locke-plate cohort ha an overall mean Constant score (an stanar eviation) of 68.6 ± 9.5 points at a mean of thirty-six months (range, twenty-four to fifty-two months) postoperatively. The Constant scores in the three-part an four-part-fracture groups were 71.6 ± 7.5 an 64.7 ± 10.8 points, respectively, at the time of the last follow-up (p = 0.11). Accoring to the OTA classification, two fractures were 11-B1, three were 11-B2, seven were 11-B3, twelve were 11- C1, thirteen were 11-C2, an one was 11-C3. Twenty-six (68%) of the thirty-eight patients presente with a valgus impacte humeral hea eformity (OTA C type), an their overall mean Constant score was 74.6 ± 9.2 points; twelve patients (32%) ha a varus extension humeral hea eformity (OTA B type), an their mean Constant score was 63.8 ± 7.8 points (p < 0.001). The ifferences in the outcomes be-

6 1693 Fig. 3 Harware failure with varus collapse at the site of an initially malreuce fracture. tween the locke-plate an hemiarthroplasty groups are summarize in Table II. Complications in the locke-plate group inclue three woun infections (8%), with one of them requiring surgical ébriement with retention of the plate. Osteonecrosis of the humeral hea evelope in six patients (16%), who ha a mean final Constant score of 62.5 ± 4.6 points, which was significantly worse than the mean score for the patients without osteonecrosis in the locke-plate group (68.7 points; p = 0.05) but was comparable with the score in the hemiarthroplasty group (59.7 points; p = 0.25). Six patients ha screw perforation of the humeral hea, an all unerwent a secon operation to reposition or remove the screw. Three of these patients unerwent screw repositioning within twenty-four hours after the inex operation, whereas three patients presente at an average of 4.5 months with varus subsience of the fracture. All of the patients with late-onset screw perforation ha extension of 1 to 3 mm of the screw through the superoposterior part of the humeral hea, an all subsequently unerwent screw removal at an average of 7.4 months. The mean Constant score for the patients with hea perforation was 67.8 ± 9.6 points compare with 71.4 ± 8.5 points for those without perforation. The numbers of patients were not sufficient to allow us to etermine if this ifference was significant. Four patients ha loss of fixation of the implant requiring conversion to a hemiarthroplasty. Three of these patients ha the plate cut out through the humeral hea, an one ha varus loosening of the implant an collapse (Fig. 3). All four patients ha an initial varus malreuction of >20. The harware failures occurre within eight weeks after the initial repair, an the hemiarthroplasties were performe within seventeen weeks after the inex proceure. The patients with conversion of a faile open reuction an internal fixation to a hemiarthroplasty were consiere to be part of the locke-plate group for the final comparison. The mean Constant score for these four patients was only 47.5 ± 3.5 points at the time of the latest follow-up (at a mean of twenty-nine months). There were no fractures of the locke plates. The forty-eight patients in the hemiarthroplasty group ha an overall mean Constant score of 60.6 ± 5.9 points at a mean of thirty-five months (range, twenty-four to fifty-two months) postoperatively. The mean Constant scores for the patients with a three-part fracture an those with a four-part fracture were 60.4 ± 5.7 an 60.1 ± 6.1 points, respectively, at the time of the last follow-up (p > 0.90). Accoring to the OTA classification, three fractures were 11-B1, four were 11-B2, seven were 11-B3, fourteen were 11-C1, seventeen were 11-C2, an three were 11-C3. Thirty-five (73%) of the forty-eight patients presente with a valgus impacte humeral hea fracture, an thirteen ha a varus extension eformity. The patients with valgus impaction ha a mean final Constant score of 60.3 ± 6.3 points compare with 60.2 ± 6.1 for those with varus eformity (p > 0.85). Complications in the hemiarthroplasty group inclue a woun infection in three patients, with one of them requiring surgical irrigation an ébriement with retention of the prosthesis. Compute tomography ocumente nonunion of the greater tuberosity in seven patients (15%), with an average of 8.8 ± 1.4 mm of tuberosity migration. All seven patients unerwent a reoperation with autogenous bone-grafting an revision of the tuberosity fixation. The mean final Constant score in this group (52.9 ± 1.9 points) was significantly lower than that for the patients without a tuberosity nonunion (60.8 ± 6.0 points) (p < 0.001). The patients with a tuberosity nonunion were significantly oler (74.4 years) than the patients who ha primary tuberosity healing (66.5 years) (p < 0.001). In the locke-plate group, the initial tuberosity isplacement was variable, ranging from 0 to 8 mm, an was not progressive over time. Tuberosity isplacement i not appear to have a significant effect on the final Constant score (r = 20.23, p = 0.15). The mean Constant score for the patients Fig. 4 Varus malreuction of approximately 15.

7 1694 Fig. 5 Suboptimal placement of the hemiarthroplasty implant, with negative humeral hea height an >5 mm of meial calcar offset. with >5 mm of tuberosity isplacement i not iffer from that for the patients with 5 mm of isplacement (66.8 compare with 70.2 points), but the ata i not have sufficient power to prevent a type-ii error. The metaphyseal hinge length was not correlate with the final outcome (r = 0.136, p = 0.41) but was strongly correlate with the evelopment of osteonecrosis (r = 20.94, p < 0.001). Osteonecrosis evelope in all patients in whom the metaphyseal hinge length was <2 mm. There was also a significant correlation between initial varus malreuction an the final Constant score (r = , p < 0.001). Of the twenty-four patients who ha <5 of initial varus angulation of the humeral hea, none ha progressive varus subsience over time, an these patients ha a mean final Constant score of 76 points. Ten patients ha an initial varus malreuction ranging from 5 to 20, an all ha progressive varus subsience averaging 7.9 (Fig. 4); their mean Constant TABLE III Influence of Raiographic Parameters on Final Outcome Parameter Correlation Coefficient (Spearman Rho) P Value Locke-plate group Tuberosity isplacement Varus malreuction <0.001 Hemiarthroplasty group Humeral hea height 0.54 <0.001 Meial calcar offset <0.001 Tuberosity isplacement score was 66 points. The thirty-four patients with initial varus angulation of 20 ha a better outcome than the patients in the hemiarthroplasty group (mean Constant scores, 71.4 compare with 59.8 points; p < 0.001). Four patients ha an initial malreuction of >20, an all four ha rapi varus subsience, loss of fixation, an conversion to a hemiarthroplasty; the mean Constant score was 47 points. In the hemiarthroplasty group, the initial tuberosity isplacement i not preict the subsequent evelopment of nonunion an was not correlate with the final outcome (r = , p = 0.49). The humeral hea height ha a significant positive correlation with the outcome (r = 0.54, p < 0.001), an the best outcomes were seen in the patients with a positive variance of humeral hea height. The meial calcar offset ha a significant negative correlation with the final outcome (r = 20.50, p < 0.001), an the best outcomes were seen in the patients in whom the meial calcar offset approache zero (Fig. 5). The correlations between the raiographic measurements in the two groups are summarize in Table III. The comparisons of the Constant scores between the locke-plate an hemiarthroplasty groups (Table II) showe the scores for power to be comparable (15.7 compare with 15.4 points; p = 0.39). The ifference in the pain score reache significance (13.2 compare with 12.3 points; p = 0.05), but the clinical relevance of this fining is questionable. There was a pronounce ifference, which was both statistically significant an clinically relevant, between the two groups with regar to the scores for activities of aily living (15.0 compare with 12.5 points; p < 0.001) an those for range of motion (25.1 compare with 19.9 points; p < 0.001). The Constant scoring scale is heavily weighte towar the range-of-motion an activity components, which explains the overall clinical an significant ifferences between the two groups. Discussion Optimal treatment of three an four-part fractures of the proximal part of the humerus in patients with poor bone quality is controversial 13,14, Open reuction an internal fixation of these fractures with stanar implants has been iscourage 3,14,18. Locke plates, which maintain angular stability in the face of axial loa, have been foun to provie substantial benefit in biomechanical stuies, but their clinical utility has not been wiely accepte 20,21,23. Their clinical benefit in the treatment of two-part fractures has been establishe, but the overall complication rate is substantial, an when they have been use for more complex fracture patterns, loss of fixation an screw perforation of the humeral hea have been challenging complications The authors of most stuies have groupe fracture types together, an there is still ebate over which three an four-part fracture types or patterns o better with open reuction an internal fixation or hemiarthroplasty an what constitutes goo or aequate reuction with a locke plate 20,22. This uncertainty is compoune by ifficulty with classifying many of these fractures with use of the OTA or Neer system, both of which have ha issues with interobserver variability 27.

8 1695 In the present series, isplacement of the greater tuberosity was variable in the locke-plate group but it i not appear to be progressive over time or to influence the final outcome in that group. Patients with initial tuberosity migration of >5 mm ha some mechanical subacromial impingement with a concomitant loss of range of motion, but the Constant scores were not iscernibly ifferent from those with 5 mm of isplacement. However, the numbers in this series were not sufficient for us to clearly istinguish ifferences in the results of these two groups. Defining the impairment from tuberosity-inuce impingement alone was outsie the scope of this stuy, an we are unable to comment on what constitutes a goo reuction of the tuberosity. The egree of humeral hea angulation ha a substantial effect on the final clinical outcome in the locke-plate group. Malreuction of the humeral hea was the most common technical error, an there was a substantial learning curve in the first twelve months of the stuy. In this series, we observe three istinct groups that we eeme goo, satisfactory, an poor reuctions on the basis of the final outcome. Goo reuctions consiste of 5 of initial humeral hea varus angulation an, overall, resulte in goo clinical outcomes. Satisfactory reuctions consiste of >5 but 20 of humeral hea varus malreuction; all of these reuctions were followe by some subsience but resulte in healing an satisfactory outcomes. The patients with a goo or satisfactory reuction ha a better outcome than those treate with a hemiarthroplasty. Patients with a poor reuction ha >20 of varus malreuction, an all ha mechanical loss of fixation an a poor clinical outcome. Hemiarthroplasty for the treatment of proximal humeral fractures has been reporte to have iscouraging clinical results, with significant functional eficits in the range of motion an activities of aily living an with Constant scores typically in the mi 50s to low 60s Pain relief, complication rates, an patient acceptance have been goo espite the relatively poor functional outcomes These finings were reiterate in our series, in which the Constant scores were consistent regarless of the initial fracture pattern. Tuberosity nonunion was the most common complication an resulte in progressive tuberosity migration an the nee for bone-grafting an revision in seven patients. The true number of tuberosity nonunions may have been higher, as compute tomography stuies were performe only for patients with ocumente tuberosity migration over time. The rate of this complication might have been reuce with the use of autogenous bonegrafting, which we i not perform routinely an which has been escribe as a technique that may reuce the incience of this complication 28. Nonetheless, initial tuberosity isplacement i not preict the evelopment of tuberosity nonunion an i not significantly influence the final outcome in the hemiarthroplasty group. Humeral hea height ha a significant positive correlation with the final outcome, whereas meial calcar offset was negatively correlate with the final outcome. Humeral hea height was technically more ifficult to assess than offset intraoperatively as most patients ha comminution in the calcar region. The most common technical error in placing the prosthesis was negative humeral hea variance. We believe that this problem can be reuce somewhat by having images of the normal shouler available for templating. Meial calcar offset was less of a problem an le to the use of humeral hea esigns that were smaller than what we ha template from the raiographs. That effect may have been ue to the slight eccentricity of the humeral hea in the coronal plane, which has been escribe previously 25. Osteonecrosis was the most common complication in the locke-plate group an was strongly correlate with the presence of a islocation initially but seeme inepenent of the Neer fracture type. Osteonecrosis evelope in six patients an was late-onset (after six months of follow-up) in four of them. Because we i not obtain magnetic resonance images routinely, the true prevalence of partial humeral hea osteonecrosis may have been higher. There was a strong correlation between the evelopment of osteonecrosis an the length of the initial metaphyseal hinge attache to the articular fragment. Osteonecrosis evelope in all of the patients in whom the hinge was <2 mm in length, whereas the average hinge length was 6 mm in those without osteonecrosis. Five of the six patients with osteonecrosis also ha ha a islocation at the time of injury; therefore, it is not possible to ientify the specific cause of osteonecrosis in these patients. The patients in whom osteonecrosis evelope ha reasonable clinical outcomes, which were comparable with the results in the hemiarthroplasty group at the time of early follow-up. That suggests that this complication may be better tolerate by the elerly population an shoul not preclue attempts at open reuction an internal fixation. Perforation of the humeral hea has been escribe as one of the more evastating complications of open reuction an internal fixation 22. In our series, the use of ipsilateral fluoroscopy an supine positioning at surgery aie in the accurate assessment of screw length an plate position. Despite this, three patients with >3 mm of screw protrusion through the humeral hea ha to return to the operating room for repositioning of the harware. Three aitional patients ha 1 to 3 mm of screw protrusion through the articular surface that evelope late as a result of varus subsience, an these patients unerwent elective screw removal. The patients with perforation i not have clinically worse outcomes than the seven patients without perforation an similar-quality reuctions, but the numbers in the series were not sufficient for us to raw any conclusions regaring ifferences between these groups. All perforations were in the superoposterior quarant of the hea fragment, an given that the average range of motion in this group preclue abuction of >120 or external rotation of >60, it is oubtful that the perforation of the articular surface in this area resulte in screw contact with the glenoi. We ha ifficulty classifying the fracture patterns an correlating the fracture types with the clinical outcomes. Although in the locke-plate group there was a substantial ifference between the mean Constant score for the patients with a three-part fracture (72 points) an that for the patients with a four-part fracture (65 points), we i not think that this

9 1696 was clinically relevant. Moreover, the Neer classification i not have a particularly useful influence on treatment choice. There were a high number of fractures involving a portion of the articular surface (impaction or hea-split pattern) that were not reaily classifie with either the OTA or the Neer system. Moreover, we foun the OTA classification to be incomplete with regar to its ability to istinguish between the two basic fracture patterns that we observe (varus extension an valgus impaction). Accoring to this classification, three an four-part varus extension patterns woul be groupe as extra-articular (B1 or B2, epening on the amount of metaphyseal impaction), but there is no way to quantify the amounts of tuberosity or hea isplacement. In our series, the initial fracture eformity ha the greatest influence on the final outcome, an this was inepenent of the Neer fracture type. The patients with a valgus impacte fracture ha the best final outcomes (a mean Constant score of 75 points), whereas those with varus extension ha a mean Constant score (64 points) that was comparable with that after hemiarthroplasty. The ifferences in outcomes may be relate to the manner in which the plate functions in each group. In patients with a valgus impaction pattern, the plate acts as a mechanical strut uner compressive forces resisting valgus subsience. In those with a varus extension pattern, the plate functions as a tension ban by pulling the humeral hea out of varus. Therefore, with poor bone quality, varus fractures place the implant at a istinct mechanical isavantage an failure is etermine by the pull-out resistance of the screws rather than by the compressive strength of the bone. We believe that this is the most important istinction to help guie initial surgical ecisionmaking regaring osteoporotic fracture patterns, an it shoul be more clearly efine in the classification systems. The Constant scoring system has been emonstrate to have high rates of intraobserver an interobserver correlation an to correlate strongly with patient satisfaction 29,30. Translating the numerical score into clinical recommenations, however, is less clear-cut. We perceive a 10-point ifference in outcome scores to be clinically relevant, whereas even significant ifferences of 5 points were probably not clinically relevant. Using these criteria, we thought that there was a clinical benefit of open reuction an internal fixation over hemiarthroplasty for the treatment of three-part, but not four-part, fractures. When only the uncomplicate cases were consiere, the scores for the pain an power components of the Constant score were very similar between the locke-plate an hemiarthroplasty groups. We exclue from the stuy patients in whom a rotator cuff tear ha been previously ocumente or was present at the time of the surgery, as the inclusion of such patients woul have confoune these components of the Constant score. The active range of motion was measure in the clinical assessment as we believe that this was the most clinically relevant approach. The main benefits seen in the locke-plate group were a better range of motion an a better ability to perform activities of aily living, which are closely linke. One possible reason for these better results in the locke-plate group was that there was less irect manipulation of the rotator cuff tenons uring the open reuction an internal fixation; however, we coul not quantify this variable. Our stuy ha several limitations. The ifficulty in classifying many of the fractures confoune interpretation of the ata an comparison of OTA an Neer subgroups. For example, hea-splitting an articular impaction were fairly common an were largely iagnose with the use of compute tomography scans as they were often ifficult to visualize on plain raiographs. As previously escribe, the OTA classification of varus extension fracture patterns is incomplete. Many of these fractures ha >1 cm of isplacement of the hea fragment an coul have been characterize as C2 on the basis of the hea isplacement; however, they were classifie as B1 because of the varus angulation an metaphyseal impaction. An initial selection bias against the hemiarthroplasty group may have been introuce by the treatment algorithm. Articular hea-splitting an impaction may represent a more severe fracture pattern, an these cases were all treate with hemiarthroplasty. Any patient with a ocumente history of a torn or repaire rotator cuff was treate with hemiarthroplasty, an this coul have been a negative influence on the final outcome. Moreover, three patients in whom we coul not perform open reuction an internal fixation for technical reasons ha a hemiarthroplasty. Quantification an correction of this bias were beyon the scope of this retrospective stuy. We use plain raiographs to quantify malreuction an implant position an attempte to stanarize the views of the shouler; however, the final resolution of the raiographs probably i not allow us to see ifferences of less than 2 to 3 mm of tuberosity isplacement or 3 to 4 of humeral hea angulation. In aition, we i not attempt to correct for magnification errors introuce by the boy habitus or size of the patient, an we were unable to assess the magnitue of the error introuce by these variables. Finally, arm rotation was not strictly controlle while the raiographs were being mae, an the magnitue of error in the measurement of tuberosity isplacement an metaphyseal hinge length ue to this variable was not quantifie. The use of compute tomography woul have provie more etaile information an an ability to quantify these variables. The routine use of a 20 external rotation view is base on the assumption that the humeral hea or prosthesis is retroverte 20 in all cases; however, we knew that there was some variability in this parameter. Given these issues, the maximum magnitue of the eformity of the humeral hea or tuberosities was probably unerestimate. In conclusion, in this series comparing similar groups of patients with a three or four-part proximal humeral fracture, the overall outcome after locke-plate fixation was better than that after hemiarthroplasty. The initial fracture eformity (varus versus valgus) was the preoperative variable that ha the greatest influence on the outcome, regarless of the fracture type. The quality of the humeral hea reuction was the intraoperative factor with the greatest influence on outcome. On the basis of our observations, we now classify a goo reuction as 5 of varus angulation of the humeral hea relative to the position of the normal, contralateral humeral hea. Repair of

10 1697 three or four-part proximal humeral fractures with an initial varus extension pattern with use of open reuction an internal fixation shoul be approache with caution, whereas valgus impacte patterns are better treate with open reuction an internal fixation than with hemiarthroplasty. Hemiarthroplasty provie more consistent, albeit worse, clinical outcomes than i open reuction an internal fixation, an this fining was inepenent of the Neer or OTA fracture type. Future fracture classification systems shoul inclue information on the initial fracture eformity to help guie the selection of optimal surgical management. n Brian D. Solberg, MD 1414 South Gran Avenue, Suite 123, Los Angeles, CA aress: brian@briansolbergm.com Charles N. Moon, MD Dennis P. Franco, MD Guy D. Paiement, MD Cears Sinai Meical Center, 444 South San Vicente Boulevar, Suite 603, Los Angeles, CA aress for C.N. Moon: charles.moon@cshs.org. aress for D.P. Franco: ennis.franco@cshs.org. aress for G.D. Paiement: guy.paiement@cshs.org References 1. Baron JA, Barrett JA, Karagas MR. The epiemiology of peripheral fractures. Bone. 1996;18(3 Suppl):209S-213S. 2. Neer CS 2n. Displace proximal humeral fractures. I. Classification an evaluation. J Bone Joint Surg Am. 1970;52: Marti R, Lim TE, Jolles CW. On the treatment of comminute fractureislocations of the proximal humerus: internal fixation or prosthetic replacement. In: Kolbel R, Helbig B, Blauth W, eitors. Shouler replacement. Berlin: Springer; p Olsson C, Norquist A, Petersson CJ. Long-term outcome of a proximal humerus fracture preicte after 1 year: a 13-year prospective population-base follow-up stuy of 47 patients. Acta Orthop. 2005;76: Schai P, Imhoff A, Preiss S. Comminute humeral hea fractures: a multicenter analysis. J Shouler Elbow Surg. 1995;4: Vallier HA. Treatment of proximal humerus fractures. J Orthop Trauma. 2007;21: Wijgman AJ, Roolker W, Patt TW, Raaymakers E, Marti RK. Open reuction an internal fixation of three- an four-part fractures of the proximal part of the humerus. J Bone Joint Surg Am. 2002;84: Tejwani NC, Liporace F, Walsh M, France MA, Zuckerman JD, Egol KA. Functional outcome following one-part proximal humeral fractures: a prospective stuy. J Shouler Elbow Surg. 2008;17: Antuña SA, Sperling JW, Cofiel RH. Shouler hemiarthroplasty for acute fractures of the proximal humerus: a minimum five-year follow-up. J Shouler Elbow Surg. 2008;17: Bosch U, Skutek M, Fremery RW, Tscherne H. Outcome after primary an seconary hemiarthroplasty in elerly patients with fractures of the proximal humerus. J Shouler Elbow Surg. 1998;7: Golman RT, Koval KJ, Cuomo F, Gallagher MA, Zuckerman JD. Functional outcome after humeral hea replacement for acute three- an four-part proximal humeral fractures. J Shouler Elbow Surg. 1995;4: Green A, Barnar WL, Limbir RS. Humeral hea replacement for acute fourpart proximal humerus fractures. J Shouler Elbow Surg. 1993;2: Jakob RP, Miniaci A, Anson PS, Jaberg H, Osterwaler A, Ganz R. Four-part valgus impacte fractures of the proximal humerus. J Bone Joint Surg Br. 1991;73: Neer CS 2n. Displace proximal humeral fractures. II. Treatment of three-part an four-part isplacement. J Bone Joint Surg Am. 1970;52: Fankhauser F, Schippinger G, Weber K, Heinz S, Quehenberger F, Bolin C, Bratschitsch G, Szyszkowitz R, Georg L, Frierich A. Caaveric-biomechanical evaluation of bone-implant construct of proximal humerus fractures (Neer type 3). J Trauma. 2003;55: Hessmann MH, Hansen WS, Krummenauer F, Pol TF, Rommens PM. Locke plate fixation an intrameullary nailing for proximal humerus fractures: a biomechanical evaluation. J Trauma. 2005;58: Siffri PC, Peinl RD, Coley ER, Norton J, Connor PM, Kellam JF. Biomechanical analysis of blae plate versus locking plate fixation for a proximal humerus fracture: comparison using caaveric an synthetic humeri. J Orthop Trauma. 2006;20: Walsh S, Reinl R, Harvey E, Berry G, Beckman L, Steffen T. Biomechanical comparison of a unique locking plate versus a stanar plate for internal fixation of proximal humerus fractures in a caaveric moel. Clin Biomech (Bristol, Avon). 2006;21: Lever JP, Aksenov SA, Zero R, Ahn H, McKee MD, Schemitsch EH. Biomechanical analysis of plate osteosynthesis systems for proximal humerus fractures. J Orthop Trauma. 2008;22: Aguelo J, Schürmann M, Stahel P, Helwig P, Morgan SJ, Zechel W, Bahrs C, Parekh A, Ziran B, Williams A, Smith W. Analysis of efficacy an failure in proximal humerus fractures treate with locking plates. J Orthop Trauma. 2007;21: Moonot P, Ashwoo N, Hamlet M. Early results for treatment of three- an four-part fractures of the proximal humerus using the PHILOS plate system. J Bone Joint Surg Br. 2007;89: Owsley K, Gorczyca JT. Fracture isplacement an screw cutout after open reuction an locke plate fixation of proximal humeral fractures. J Bone Joint Surg Am. 2008;90: Strohm PC, Köstler W, Sükamp NP. Locking plate fixation of proximal humerus fractures. Tech Shouler Elbow Surg. 2005;6: Fracture an islocation compenium. Orthopaeic Trauma Association Committee for Coing an Classification. J Orthop Trauma. 1996;10 Suppl 1; v-ix, Hertel R, Knothe U, Ballmer FT. Geometry of the proximal humerus an implications for prosthetic esign. J Shouler Elbow Surg. 2002;11: Constant CR, Murley AH. A clinical metho of functional assessment of the shouler. Clin Orthop Relat Res. 1987;214: Siebenrock KA, Gerber C. The reproucibility of classification of fractures of the proximal en of the humerus. J Bone Joint Surg Am. 1993;75: Krause FG, Huebschle L, Hertel R. Reattachment of the tuberosities with cable wires an bone graft in hemiarthroplasties one for proximal humeral fractures with cable wire an bone graft: 58 patients with a 22-month minimum follow-up. J Orthop Trauma. 2007;21: Erratum in: J Orthop Trauma. 2008; 22: Johansson KM, Aolfsson LE. Intraobserver an interobserver reliability for the strength test in the Constant-Murley shouler assessment. J Shouler Elbow Surg. 2005;14: Rocourt MH, Ralinger L, Kalberer F, Sanavi S, Schmi NS, Jeuing M, Hertel R. Evaluation of intratester an intertester reliability of the Constant-Murley shouler assessment. J Shouler Elbow Surg. 2008;17:364-9.