G.J. Haidukewych replies:
I appreciate Dr. Parker's comments on the difficulty of measuring the Pauwels angle and understand the lack of literature on the interobserver reliability of the classification. I am concerned, however, that Dr. Parker feels that our data are not clinically relevant. I must point out that our series is the largest contemporary multicenter study of the "worst-case scenario," that is, the most vertical femoral neck fractures treated with modern internal fixation devices. We chose to study these since we thought that it is relatively simple to determine the verticality of angles of >70° compared with the lower shear angles of 30° or 40°, for example. Also, this cohort would be most likely to elucidate any differences in performance of various fixation strategies since, theoretically, these fractures would experience the most shear.
The overall nonunion rate in our study was about fourfold higher than recent nonunion rates reported in series that did not subanalyze shear angle1. Our patient group was young, implying good bone quality, and had a high rate of early, accurate reductions. Although we did not demonstrate a significant difference between fixed-angle devices and screws alone, the nonunion rate for screws alone was 19%. We consider a nonunion rate of 19% in a relatively young cohort to be very clinically relevant. We concede that the ideal fixation strategy has not been defined; however, Dr. Parker must admit that a nonunion in about one in five young patients is clinically relevant. Anecdotally, in our practices, when we see a femoral neck nonunion in a younger patient, the typical clinical scenario involved a very vertical fracture that was treated with screws alone. We are aware that other centers are using antiglide plates applied through an anterior approach to neutralize shear in vertical fractures.
In their series reported in 1998, Parker and Dynan evaluated the utility of the Pauwels angle in predicting outcome in 335 patients with a femoral neck fracture2. However, it should be noted that only eight patients had a displaced fracture with a shear angle of >60° in that series, and half went on to nonunion. Making conclusions on high-shear-angle fractures on the basis of only eight patients is impossible. Also, making conclusions on the basis of pooled data compiled from the literature of the 1960s and 1970s is not relevant to modern fracture management, involving fixation with cannulated screws, use of fluoroscopy, and the understanding of the importance of reduction quality and timely fixation.
Obviously, further research is necessary to determine the ideal fixation strategy, but we still contend that Pauwels type-3 high-shear-angle fractures are problematic to treat and result in a nonunion rate that is substantially higher than historical controls. The Pauwels classification may not be perfect, but we maintain that fracture verticality matters.
These letters originally appeared, in slightly different form, on . They are still available on the web site in conjunction with the article to which they refer.