In relation to the conduct of this study, one or more of the authors is in receipt of a research grant from a non-commercial source.
Humeral hemi arthroplasty has become widely used as a form of surgical management for severe fractures. However there is still no consensus as to the role for prosthetic replacement in displaced proximal humeral fractures. The aim was to assess shoulder hemi arthroplasty for un-reconstructable three and four part proximal humeral fractures at a minimum of twelve months and identify factors that guide to prognosis. Criteria for inclusion were patients with a fracture that went onto shoulder hemi arthroplasty with Constant scoring at a minimum follow up of one year. Patients were treated using a Neer or Osteonics prosthesis, with the decision for hemi arthroplasty being made at the time of surgery. Post-operative management was standardised. An independent functional assessment, record review establishing a physiological index according to comorbidities, and a radiological analysis were carried out. A survival analysis was performed for the one and five year results and data was analysed by linear regression to identify prognostic factors. Of 163 patients there were 138 fitting the criteria, 42 males and 96 females with an average age of 68.5 (range30–90) years and average follow up of 6.3 (range1–15) years. The fracture pattern was three or four part in 133 cases and 5 head split fractures; 58 were associated with a dislocation. Survival was 96.4% at 1 year and 93.6% at 5 years, with no significant difference between prostheses. There were 8 revisions, (1 deep infection, 4 dislocations and 3 peri-prosthetic fractures), most within 12 months. The average Constant score was 67.1 at one year. Prognostic factors on presentation were the age of the patient and their physiological index. Factors at 3 months were any complication, the position of the implant, tuberosity union and persistent neurological deficit. Overall optimum outcome was gained by patients aged 55–60, with minimal comorbidities and an uncomplicated recovery.
All patients over 65 yrs with an ISS greater than 15 attending Edinburgh Royal Infirmary between 1997 and 2000 were prospectively entered into the study. Patients were followed until death or discharge home. The patients were divided into, group 1 [patients who survived], and group 2 [those who died.] A total of 72 patients were included in the study, 42 males and 31 females. 42 patients survived, and 31 died. Group 1 consisted of 29 males and 15 females with an average age of 75.23yrs. Group 2 consisted of 13 males and 18 females with an average age of 78.05yrs. All incidents involved blunt trauma. The three main mechanisms of injury were RTA, Fall less than 2 meters, and Fall greater than 2 meters. Five patients required intubation in group 1 and 12 patients in group 2.The average GCS was lower in group 1 compared to the group 2. All Injuries with AIS of greater than 3 were analysed. The total number of injuries was greater in the group 2. Group 1 required 214 days in HDU/ITU and a total of 943 in-patient days. Group 2 in comparison needed 62 HDU/ITU days and 169 in-patient days. The major cause of death was head and spinal injury 11 (35%), and Multiple injuries 9 (29%). A total number of 1952 days were spent in rehabilitation prior to discharge, with an average of 46.48 days. Post trauma the level of independence was significantly reduced. The injuries are exclusively blunt and in the majority of cases secondary to motor vehicle accidents. Predictors of mortality appear to include, intubation, head and neck injuries, GCS, and chest injuries. Current outcome scores correlate inaccurately. These patients require long hospital stays with a large amount of intensive care input. After discharge rehabilitation is universally required. These patients place a large demand on the NHS and social services; the total cost of their care was approximately £2,500,000.