Evaluation of Clinical and Radiological Outcome of Fracture Proximal Humerus by Locking Compression Plate Taking into Consideration of Biomechanics of the Glenohumeral Joint

Vol 32 | Issue 1 | Jan – June 2017 | page: 50-55 | D C Srivastava, Sachin Yadav,  Alok Gupta


Authors: D C Srivastava [1], Sachin Yadav [1],  Alok Gupta [1] 

[1] Department Of Orthopaedics, M L N Medical College Allahabad.

Address of Correspondence
Dr. Ajai Singh
Department of Orthopaedics Surgery,
K. G. Medical University, Lucknow, India.
Email: as29762@gmail.com


Abstact

Introduction: Although several studies have repeatedly emphasized that PHILOS plating gives better functional outcomes for fracture proximal humerus and that many studies have speculated and concluded various risk factors for failure or poor functional outcomes, none of them gives a clear idea about the important parameters of the biomechanics of the glenohumeral joint and their restoration, while fixation with the PHILOS plate. Our aim of study is to assess and weigh the functional outcomes when PHILOS plating is done for fracture proximal humerus taking into consideration biomechanics of the glenohumeral joint.
Material and Methods: This was an open label prospective study of 30 patients of proximal humerus fracture who were treated with with PHILOS plate in SRN Hospital From Sept 2013 to June 2015. Cases were taken up according to inclusion and exclusion criteria. All the patients were managed by open reduction and internal fixation by PHILOS plate and all patients underwent a rehabilitation program. The patients were examined clinically and radiologically, assessed for range of motion and bony union and complication. Further follow ups were done at 6 weeks,12 weeks, 24 weeks,36 weeks and 1 year.
Results: All the patients were followed up for atleast for 12 months (range 12 to 20 months) with the mean follow up time was 16.5 months. All the fractures united clinically and radiologically around an average of 8 weeks (range7-11weeks). Constant scores were excellent to good in 21 patients (70%), fair in 5 patients (16.66%), poor in 4 patients (13.33%).
Conclusion: Management of proximal humerus fractures with PHILOS plate gives the most satisfactory functional outcome when fixation is done taking into account following parameters: Corrected neck-shaft angle, maintenance of retroversion of the humeral head and acromio-humeral head distance.
Keywords: Biomechanics, neck-shaft angle, retroversion of the humeral head, acromio-humeral head distance.


Introduction
Proximal humerus fractures account for approximately 4 – 5% of the all fractures [1,2,3]. The incidence of fracture is more common in the elderly because of decreased bone density. But it can occur in younger age group following high velocity trauma [3]. Because of increasing incidence of high velocity trauma, the fracture patterns in proximal humerus are becoming complicated. It has been always enigma of management because of numerous muscles attachment and paucity of space for fixing implant in fractures of proximal humerus. While for undisplaced fractures, literature strongly suggests nonoperative treatment [4,5,6] the treatment of displaced fractures is still controversial and challenging. The various methods to treat fractures of the proximal humerus are percutaneous cerclage wiring, k-wire fixation, tension band wiring, intramedullary nailing, plating, and humeral head replacement. Of these methods, recent clinical data have shown favorable results for the treatment of fractures with locking plates. Important for a successful stabilization of the fracture are in the order of anatomic reduction, medial cortical support, proximal humeral bone density, and patient age[7]. Although several studies have repeatedly emphasized that PHILOS plating gives better functional outcomes for fracture proximal humerus and that many studies have speculated and concluded various risk factors for failure or poor functional outcomes, none of them gives a clear idea about the important parameters of the biomechanics of the glenohumeral joint and their restoration, while fixation with the PHILOS plate. Restoration of the normal three-dimensional anatomy of the proximal humerus is desirable. This affects the lever arms of the deltoid and rotator cuff in the vertical plane and soft-tissue balance in the horizontal plane. Changes in this anatomy may produce abnormal kinematics by malpositioning the joint line and by moving the instant centres of rotation [8,9,10]. Our aim of study was to assess and weigh the functional outcomes when PHILOS plating was done for fracture proximal humerus taking into consideration biomechanics of the glenohumeral joint.

Material and Method
From SEPTEMBER 2013 to JUNE 2015, 35 consecutive patients with proximal humerus fractures fulfilling the inclusion criteria were treated with open reduction and internal fixation with PHILOS plate in SRN Hospital. This was an open label prospective study and the sampling technique was consecutive non-probability sampling. A total of 35 patients were included in the study out of which 5 patients could not be followed up for at least 1 year. Patients included in this study were based on the following inclusion criteria: all displaced two part, three part, four part proximal humeral fractures, age above 18, patient had fitness and willingness for surgery and willingness to undergo physiotherapy and posto-perative rehabilitation program. Exclusion criteria were: associated humerus shaft fracture and associated neurovascular injury and acute infection, pathological fractures, old fractures ( more than 3 months old), compound fractures, undisplaced fractures. On admission of the patient, the general condition of the patient and the vital signs were recorded. Methodical examination was done to rule out fractures at other sides. The local examination of injured shoulder was done for swelling, deformity loss of function and altered attitude. Any nerve injury was also looked for and noted. Local neurologic deficit of axillary nerve was also assessed by looking for anaesthetic patch over lateral aspect of shoulder. Radiograph of proximal humerus was taken and fractures were classified according to Neer’s classification.
Of the contralateral shoulder, True AP view and special view as described by Volkan oztuna et al11 were taken. On true AP view, Neck-shaft angle and acromio-humeral distance of the contralateral limb could be determined. Angle created at intersection of lines that are perpendicular to anatomic neck and parallel to shaft of humerus determines the neck shaft angle. Distance between undersurface of acromian and top of humeral head determines the acromio-humeral distance. Special view were taken as described by Volkan oztuna et al [11]. Posteroanterior semiaxial radiographs of humerus bones was taken. This method was performed while the subject was standing with their shoulder in 90° forward elevation and 20° abduction. The beam was positioned 1 m away from the shoulder and centered to the humeral head; and the fully supinated forearm was placed on the cassette. On special view as described by Volkan oztuna et al [11], humeral head retroversion is measured, determining the humeral anatomical neck axis, outlining a perpendicular line to the humeral head joint surface, and a line tangent to the trochlea, the humeral head retroversion angle being the acute angle formed by the intersection of both lines.
Of the 30 patients, there were 16 males (53%) and 14 females (47%). Overall mean age was 53.93 years (range 25 to 92 years). According to Neer’s classification: 12 patients (40%) had Neer’s two part fracture, 9 patients (30%) had Neer’s three part fracture and 9 patients (30%) had Neer’s four part fracture. Mean age of the patients with Two-part fractures was 55.17 years (range 27 to 92 years). Mean age of the patients with Three-part fractures was 49 years(range 25 to 81 years). Mean age of the patients with Four-part fractures was 57.22 years (range 32 to 89 years). 19 patients (63%) had right upper limb involvement and 11 patients (37%) had left upper limb involvement.
All the patients were managed by open reduction and internal fixation by PHILOS plate. Postoperatively all patients were immobilized in shoulder immobilizer, appropriate antibiotics and analgesics were used, immediate postoperative radiographs were taken to determine the bone alignment and maintenance of reduction, sutures were removed depending upon the condition of the surgical wound.


All patients underwent a rehabilitation program in which pendulum exercises were begun immediately depending on pain, passive range of motion were started at 1st week. The active range of motion were started at 2-4 weeks postoperatively, depending on stability of osteosynthesis and bone quality. At 4th to 6th week–immobilization was discontinued. Active assisted movements were started up to 90° abduction with no forced external Rotation. At 6th to 8th week-full range of movements with active exercises started. The patients were examined clinically and radiologically, assessed for range of motion and bony union and complication. Further follow ups were done at 6 weeks and 12 weeks and 24 weeks and 36 weeks and 1 year depending upon the condition of the patient. At 1 year, True AP view and special view of the operated limb as described by Volkan oztuna et al [11] were taken to measure neck-shaft angle, acromiohumeral head distance and retroversion angle of the humeral head of the operated limb. The final result were evaluated using the following score at 1 year follow up : CONSTANT AND MURLEY SCORING SYSTEM.
Surgical technique: General anesthesia was used in all patients. Patients were placed in supine position on operating table with wedge or a sandbag under the spine and medial border of scapula to push the affected side forward while allowing the arm to fall backward. Arm was draped free, because it had to be moved during the approach. A deltopectoral approach was used. Skin was infiltrated with local anaesthetic. The incision was started midway between coracoid and clavicle, and extended distally up to deltoid insertion. Cephalic vein was identified and retracted laterally. Plane was developed between deltoid and pectoralis major. Conjoint tendon was identified and retracted medially. The long head of biceps is the important landmark as it signifies rotator interval. Traction sutures were applied to tuberosity fragments using Ethibond 5 suture to hold and reduce fragments. K-wires were used in humeral head as joysticks and fixed temporarily with the humeral shaft. With the help of c-arm images displaced proximal humerus fracture was provisionally fixed with k-wires taking into consideration biomechanics of glenohumeral joint with the aim to achieve retroversion, neck shaft angle and acromiohumeral distance similar to those of contralateral limb, so that the head was not fixed in valgus or varus rotation in coronal plane and repair the rotator cuff mechanism so as to restore the length between undersurface of acromian and head of the humerus.
Intra operatively after provisionally fixing two, three and four part fracture proximal humerus by k-wires, abduct the arm if possible (as more clear images could be obtained) and flex the elbow to 90° such that horizontal line drawn parallel to the ground overlaps both the condyles of the humerus. In this position by C-ARM machine take an AP view of the shoulder. Now rotate the forearm externally or internally everytime by 10° and take subsequent C-ARM images. At that rotation of the forearm externally or internally where maximum diameter of the head of the humerus is visible will be the version angle of the head of the humerus. If the provisional fixation achieved by k-wires intra-operatively was satisfactory ensuring maximum correction of the various biomechanics parameters of the glenohumeral joint by c-arm images, rigid fixation with PHILOS plate was done and k-wires were removed. PHILOS plate was placed about 5 to 10 mm distal to the tip of greater tuberosity (confirmed under image intensifier) and just lateral to bicipital groove. First a standard 3.5 mm cortical screw was placed in oval hole and tightened lightly. Fracture reduction and plate position were reconfirmed under image before placing other screws. Locking screws were placed in humeral head using drill guide. Drilling was done under sequential fluoroscopic imaging to prevent intraarticular penetration. Subchondral screw position was confirmed under two image views. Finally screws were placed in humeral shaft. Tuberosity sutures were tied to the plate. The wound was closed over a suction drain, which was removed after 24 to 48 hours -postoperatively. Arm was supported with a sling.
Statistical analysis: All data were collected on a form with the patient’s details. Afterwards these data were filled into a Microsoft Excel-2010 sheet for simple calculations. Statistical analysis was performed over the data using the software SPSS 20.0 for Windows. Test used was the One-way ANOVA test and Post Hoc Bonferroni test for comparison and analysis of mean Constant score and mean Neer score between two part, three part and four part fractures which was considered significant at P <0.05.

Results
All the patients were follow up for 12 to 20 months with the mean follow up time was 16.5 months. All the fractures united clinically and radiologically at 8 weeks (range7-11weeks). Complications observed were avascular necrosis of the humerus head in 1 patient (3.33%), superficial infection in 2 patient (6.66%), anterior deltoid muscle atrophy in 3 patients (10%), transiently decreased radial nerve sensations in 3 patients (10%), subacromial impingement in 2 patients (6.66%).
In our study overall mean Constant Murley score (Table 1) was 73.1 (Range 35-96) with standard deviation of 16.34. Mean Constant Murley score for Two- part fracture was 83.33 with standard deviation of 11.51. Mean Constant Murley score for Three- part fracture was 75.67 with standard deviation of 8.94. Mean Constant Murley score for Four- part fracture was 56.89 with standard deviation of 15.67.
Pearson correlation coefficient between Constant Murley Score and Neer Score is 0.98 .
Difference in mean Constant Murley scores (Table 2) of Two-part fracture and Three part fracture was not significant (p value 0.503). Difference in mean Constant Murley scores of Two-part fracture and Four-part fracture was significant (p value 0.000). Difference in mean Constant Murley scores of Three-part fracture and Four-part fracture was significant (p value 0.009). Thus two-part fractures and three-part fractures are amenable to give better functional results than four part results even with the best technique and methods of fixation and post operative rehabilitation.

Grading Of Results (Table 3):In our study at one year follow up, constant scores were excellent to good in 21 patients (70%), fair in 5 patients (16.66%), poor in 4 patients (13.33%)

Comparison between various sub-scores of constant-murley score: (Table 4)There was no significant difference in pain, activities of daily living, strength and range of motion between Two-part fractures and Three- part fractures ( p value 0.335, 0.843, 0.582, 0.774 respectively). There was significant difference in pain, activities of daily living, strength and range of motion between Two-part fractures and Four- part fractures ( p value 0.000, 0.001, 0.000, 0.001 respectively). There was significant difference in pain, activities of daily living, strength and range of motion between Three-part fractures and Four- part fractures ( p value 0.025, 0.023, 0.023, 0.020 respectively)(Table 5).
There is no significant difference between mean acromio-humeral head distance in the uninjured limb and operated limb at one year follow up ( p value 0.92). There is no significant difference between mean neck-shaft angle in the uninjured limb and operated limb at one year follow up ( p value 0.42). There is no significant difference between mean retroversion angle of the humeral head in the uninjured limb and operated limb at one year follow up ( p value 0.08) (Table 6). Thus we have achieved fixation of the proximal humerus fracture with the PHILOS plate and restored the important three dimensional biomechanical parameters nearly as that of the contralateral normal limb.

Discussion
The treatment of proximal humerus fractures with PHILOS plate gives a satisfactory outcome when fixation is done taking into consideration biomechanics of glenohumeral joint. In our study the fixation of the proximal humerus fracture by PHILOS is done taking into consideration various biomechanical parameters of the glenohumeral joints, most important being acromio-humeral head distance, neck-shaft angle of the humerus and retroversion angle of the humeral head. It is observed that there is no significant difference in these biomechanical parameters at one year follow up between the operated limb and uninjured contralateral limb. Thus, we have achieved the various important biomechanical parameters of the injured limb with fracture proximal humerus after fixation with PHILOS plate at one year follow up nearly similar to those of contralateral uninjured limb. Overall mean Constant Murley score is 73.1 (Range 35-96) with standard deviation of 16.3398. Average age and sex adjusted Constant Murley score was 80.934. The functional outcome based on Constant Murley score at one year follow up, is excellent to good in 21 patients (70%), fair in 5 patients (16.66%), poor in 4 patients (13.33%).
It is observed that in young patients, internal fixation using the proximal humeral internal locking system (PHILOS) plate has yielded better functional outcome than in older patients. We speculate the role of osteoporosis in older patients that has lead to above finding. Functional outcome of two part and three part fracture promixal humerus fixed with PHILOS plate are significantly better when compared with four part fracture promixal humerus fixed with PHILOS plate. It is observed that there is no significant difference in pain, activities of daily living, strength and range of motion between Two-part fractures and Three- part fractures (p value 0.335, 0.843, 0.582, 0.774 respectively). While there is significant difference in pain, activities of daily living, strength and range of motion between Two-part fractures and Four- part fractures ( p value 0.000, 0.001, 0.000, 0.001 respectively) & between Three-part fractures and Four- part fractures ( p value 0.025, 0.023, 0.023, 0.020 respectively).

Conclusion
Proximal humerus fractures demands careful evaluation of type of fracture and higher surgical skills to restore three-dimensional anatomy of the gleno-humeral joint. Management of proximal humerus fractures with PHILOS plate gives the most satisfactory functional outcome when fixation is done taking into account following parameters:
A) Corrected neck-shaft angle has been associated with better functional outcome as it prevents lengthening of lever arm of the deltoid and supraspinatus muscles, abductor muscle dysfunction and subsequent subacromial impingement. Thereby associated with early restoration of range of motion especially abduction.
B) Maintenance of retroversion of the humeral head is associated with better functional outcome as it prevents shoulder subluxation and maintains the stability and range of motion especially in horizontal plane.
C) Acromio-humeral head distance if maintained is associated with better functional outcome as it facilitates rotator cuff mechanism and prevents impingement of the supraglenoid structures.
Adequate biomechanical stability is achieved by bone grafts in fractures of the base of the neck of humerus having gross metaphyseal communition and bone gap.


References

1. Gerald R. Williams and Kirk L. Wong: “Two-part and three part fractures-Management of proximal and distal humerus fracture”. Orthop Clin North Am, 2000;31 (1) : 1-21.
2. Anthony F. Depalma and Richards Cautilli. Fractures of the upper end of the humerus.Clin.ortho. 1971;20:73-93.
3. Neer CS II,Rockwood CA: Fractures and dislocations of the shoulder,in RockwoodCA, Green DP (eds):Fracture in adults, Philadelphia, PA, Lippincott, 1984:675-721.
4. Koval KJ, Gallagher MA, Marsicano JG, Cuomo F, McShinawy A, Zuckerman JD. Functional outcome after minimally displaced fractures of the proximal part of the humerus. J Bone Joint Surg Am. 1997;79:203–7.
5. Tejwani NC, Liporace F, Walsh M, France MA, Zuckerman JD, Egol KA. Functional outcome following one-part proximal humeral fractures: A prospective study. J Shoulder Elbow Surg. 2008;17:216–9.
6. Gaebler C, McQueen MM, Court-Brown CM. Minimally displaced proximal humeral fractures: Epidemiology and outcome in 507 cases. Acta Orthop Scand. 2003;74:580–5.
7. Krappinger D, Bizzotto N, Riedmann S, Kammerlander C, Hengg C, Kralinger FS. Predicting failure after surgical fixation of proximal humerus fractures. Injury. 2011;42(11):1283-8.
8. Fischer LP, Carret JP, Gonon GP, Dimnet J. Etude cin´ ´ ematique des mouvements de l’articulation scapulo-hum´erale (Articulatio Humeri). Rev Chir Orthop 1977;63(Suppl II):108-12.
9. Friedman RJ. Biomechanics of the shoulder following total shoulder replacement. In: Post M, Morrey BF, Hawkins RJ, eds. Surgery of the shoulder. St Louis, Mosby-Year Book, 1990:263-6.
10. Ballmer FT, Sidles JA, Lippitt SB, Matsen FA, III. Humeral head prosthetic arthroplasty: surgically relevant geometric considerations. J Shoulder Elbow Surg 1993;2:296-304.
11. Volkan Öztuna et al: Measurement of the humeral head retroversion angle, Archives of Orthopaedic and Trauma Surgery September 2002, 122, 7, 406-409.


How to Cite this Article: Srivastava D C, Yadav S,  Gupta A. Evaluation of Clinical and Radiological Outcome of Fracture Proximal Humerus by Locking Compression Plate Taking into Consideration of Biomechanics of the Glenohumeral Joint. Journal of Bone and Joint Diseases  Jan – June 2017;32(1):50-55.


(Abstract)      (Full Text HTML)   (Download PDF)