Can we decipher indications and outcomes of the PHILOS plate for fractures of the proximal humerus?

Abstract

Sir, I read with interest the recent article by Zhou et al. [1] describing the outcome of the proximal humeral internal locking system (PHILOS) plate in 74 patients. However, I am still unsure of the exact indications for the plate. Despite the escalation in the use and publications regarding the outcome of the PHILOS plate, the evidence as to whether this device is beneficial to patients it was designed for remains difficult to decipher. There have been 27 studies of the 63 cited on PubMed that have used the Constant score as their outcome assessment tool (Table 1). There is, however, marked heterogeneity between these studies, with mean age ranging from 42 to 78 years and reported cohort size ranging from nine to 294 fractures. It is interesting to note the epidemiology of proximal humeral fractures where the mean age is 65 years [2], with three- and four-part fractures occurring in older patients [3]. Mean patient age in those studies reporting the outcome of the PHILOS is 62 years old, which may suggest that there is an inclusion bias for some studies, which may reserve such an intervention for younger patients. However, this does seem to be at odds with the design and intention of the PHILOS plate. Table 1 Twenty-seven studies identified on PubMed (search term “PHILOS plate”) that report the outcome of the proximal humeral internal locking system (PHILOS) plate using the Constant score as their measure Furthermore the Constant score has been demonstrated to diminish with patient age in a normal population [4]. The variation in the reported Constant score after PHILOS plating ranges from 58 to 95 (Table ​(Table1).1). In part, this probably reflects the differing mean age between study cohorts. However, even if these scores are adjusted for patient age, the variation in score ranges from 24 points less than predicted to 21 points greater than predicted for their age (Fig. 1). This variation may also be due to inclusion criteria of the studies, which may reflect that only higher functioning patients were offered surgery. However, it is hard to believe that most patients will regain their prior functional status or even improve relative to their predicted score. Fig. 1 Difference in the reported Constant score relative to patient age-matched score for the 27 identified studies reporting the outcome of the proximal humeral internal locking system (PHILOS) plate The only randomised controlled trial comparing the outcome of the PHILOS plate with conservative management for proximal humeral fractures in elderly patients concluded that there was no statistical difference in any of the outcome measures assessed [5]. The authors also demonstrated a reoperation rate of 17%. The cost and complication risk of operative intervention with a PHILOS plate would seem, on current evidence, to be of no significant benefit to the elderly patients for which this plate was designed. Hence, it would seem the PHILOS plate is not for the masses, and its insertion should be performed by the few.