Question:
In older patients with a displaced three-part fracture of the proximal part of the humerus, does treatment with a locking plate improve quality of life and functional outcomes more than nonoperative treatment does?
Design:
Randomized (allocation concealed), unblinded, controlled trial with two years of follow-up.
Setting:
Stockholm Söder Hospital in Stockholm, Sweden.
Patients:
Sixty patients who were ≥55 years of age and had an acute, displaced three-part fracture of the surgical neck of the humerus caused by a simple fall. Patients had no previous shoulder problems, no severe cognitive impairment, and were not institutionalized. Patients with an absolute indication for surgery (completely displaced shaft in relation to the head fragment) were excluded. One patient withdrew immediately after randomization, leaving fifty-nine patients (mean age, 74 years; 81% women). Fifty-three patients (88%) were available for the twenty-four month follow-up.
Intervention:
Patients were allocated to open reduction and internal fixation with a locking plate (n = 30) or nonoperative treatment (n = 29). Surgery was done by one of two orthopaedic surgeons using the Philos plate (Synthes, Stockholm, Sweden). The plate was placed ≥8 mm distal to the upper end of the greater tubercle and slightly dorsal to the long head of the biceps. Patients were referred to a physiotherapist, wore a sling for four weeks, and started pendulum exercises and passive elevation and abduction up to 90° on the first postoperative day. Patients in the nonoperative group were immobilized in a sling for two weeks, then were referred to a physiotherapist and began pendulum exercises and passive elevation and abduction up to 90°.
Main outcome measures:
The primary outcome was health-related quality of life as assessed with use of the EuroQol (EQ)-5D index, which rated five dimensions (mobility, self-care, usual activities, pain and/or discomfort, and anxiety and/or depression) as no problem, some problems, or major problems. The EQ-5D score ranged from 0 (worst possible state of health) to 1 (best possible). The secondary outcome was functional outcome as assessed with use of the Constant score, a rating scale used to evaluate shoulder function (pain, 15 points; activities of daily living, 20 points; range of motion, 40 points; and strength, 25 points) and the Disabilities of the Arm, Shoulder and Hand (DASH) score, a rating scale used to measure upper-extremity disability and symptoms. The thirty-item disability/symptom scale was used and ranged from 0 (no disability) to 100 (most severe disability). Pain was assessed on a 100-point visual analog scale (0 = no pain, and 100 = worst possible pain).
Main results:
Because of a lack of reliable data from previous studies, a formal power analysis could not be done. Based on assumptions, a sample size of sixty patients was estimated to be sufficient. The mean prefracture EQ-5D score was 0.85 in both treatment groups. At twenty-four months, the score had decreased to 0.70 (p = 0.006) in the locking plate group and to 0.59 (p = 0.001) in the nonoperative group. The deterioration in the EQ-5D score was less severe in the locking plate group than it was in the nonoperative group, but the difference between groups did not reach significance (0.15 vs. 0.26, p = 0.26). The twenty-four month results for the Constant and DASH scores also showed no significant difference between the locking plate and nonoperative groups (61.0 vs. 58.4, p = 0.64; and 26.4 versus 35.0, p = 0.19, respectively). The mean pain scores did not differ in the locking plate and nonoperative groups (17 vs. 20, p = 0.94). Nine patients (30%) in the locking plate group required additional surgery during the second year of follow-up. In the nonoperative group, one patient had a malunion but declined surgery and one patient required minor surgery.
Conclusion:
In older patients with a displaced three-part fracture of the proximal part of the humerus, there was a statistically nonsignificant trend in health-related quality of life and functional outcomes favoring internal fixation with a locking plate rather than nonoperative treatment.
Sources of funding: Trygg-Hansa Insurance Company and the Stockholm County Council.
For correspondence: Dr. P. Olerud, Department of Orthopaedics, Stockholm Söder Hospital, S-118 83 Stockholm, Sweden. E-mail address: per.olerud@sodersjukhuset.se
For a glossary of terms for evidence-based orthopaedics, go to jbjs.org/ebo_glossary.
Disclosure: The author received payments or services, either directly or indirectly (i.e., via his institution), from a third party in support of an aspect of this work. In addition, the author, or his institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. The author has not had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by the authors are always provided with the online version of the article.
The study by Olerud and colleagues has important ramifications for orthopaedic surgeons who treat patients who have a proximal humeral fracture. Court-Brown and McQueen previously reported that, in patients who are older than seventy-five years, any displacement can be accepted provided that healing can be anticipated1. The study by Olerud and colleagues also concluded that a nonoperative approach in the treatment of displaced fragments about the humeral head can lead to results that are similar to those obtained in fractures treated operatively.
There are several issues to consider. First, the study population appeared to consist of patients with both moderately and substantially displaced fractures. The second issue concerns the early comfort of the patient. Nonoperative treatment is often attempted but cannot be continued due to excessive pain. If patients can tolerate the first twenty-one days of treatment they will have an easier recovery afterwards. The locked plate allows for reconstruction of the fracture and stable fixation and may be used for operative treatment for some patients. However, Olerud and colleagues demonstrated that, while operative reduction may be of benefit in the short term, it does not alter the long-term outcomes. Third, the inclusion of patients who were younger than sixty-five years was surprising. Given that the average age was seventy-four years, it is questionable whether there were sufficient numbers of patients in the fifty-five to sixty-five-year age group to draw any conclusion about younger patients.
In conclusion, patients who are older than sixty-five years and have a moderately displaced fracture of the humeral head do very well with nonoperative treatment. This treatment must include early physiotherapy with pendulum exercises beginning at two weeks, progression to active and active-assisted exercises, and, finally, to strengthening exercises over a period of six to twelve weeks. Failure rates of this treatment are low, and reconstruction can be performed later if indicated by shoulder pain or stiffness.