Reliability of the Neer classification system in proximal humeral fractures: A systematic review of the literature
Mahadeva D., Mackay DC., Turner SM., Drew S., Costa ML.
Introduction: The Neer classification system for proximal humeral fractures has been in use for more than 30 years. Although it is popular, there have been reports raising doubts regarding its reliability. The purpose of this paper was to systematically analyse the literature addressing the inter/intra-observer agreement of the Neer classification. Method: The OVID (Medline) database was used and the Mesh terms "Neer Classification" AND " Reliability" were inserted. Results: Five studies were retrieved but only three in all of them were found to meet the eligibility criteria for analysis. Three further studies were identified from the reference sections of the papers. Only four of the studies assessed both inter/intra-observer agreement. All studies used Kappa statistics as their principal outcome measure. The Kappa agreements were generally poor (0.21-0.4) to moderate (0.41-0.6), with only one study producing good (0.64) results. Discussion: The outcome from proximal humeral fractures is extremely variable. Inconsistent interpretation of the plain radiographs may be one of the factors that affects the results. The studies included in this review show that there is considerable disagreement with regard to the classification of the injuries and that this problem is not restricted to more junior personnel. The Neer classification is a useful management tool but accurate delineation of fracture pattern requires better imaging modalities. This, however, can never be a direct comparison to the original Neer system. © 2008 Springer-Verlag.