To The Editor:
In the Instructional Course Lecture "Neurovascular Injury Associated with Hip Arthroplasty" (79-A: 1870—1880, Dec. 1997), Lewallen stated that "the type of operative approach used … has not been shown to influence the observed rates of nerve palsy." However, he then stated that "the sciatic nerve is more at risk [of direct injury] during the posterior approach." Lewallen cited two references3,4 in support of this contention. In fact, Johanson et al.3 reported that the prevalence of neural injury decreased from 0.6 percent to 0.3 percent in the latter half of their study, when their practice changed from the use of a transtrochanteric approach to the use of a posterior approach. The authors thought that this decrease in neural injury reflected "the improvement of technical factors unrelated to the approach." Evidence of the implausibility of this explanation is provided by Robinson et al.9, who had reviewed the same group of patients three years earlier. Interestingly, Robinson et al. emphasized that the results for the posterior approach represented their early experience. They reported no neurological complications in the 160 hips that were treated through a posterior approach and two such complications (1 percent) in the 156 hips that were treated through a transtrochanteric approach. The reported 7.5 percent rate of dislocation (twelve of 160) in the group of hips that were treated through a posterior approach seems inconsistent with the claim made by Johanson et al. regarding the improvement of technical factors that were not related to the operative approach. Other studies have provided additional evidence that the posterior approach is not associated with excessive risk of neural injury when compared with the transtrochanteric approach. Weber et al.11 reported a 70 percent prevalence of neural injury (confirmed electromyographically) in a prospective study of thirty total hip arthroplasties performed through a transtrochanteric approach, whereas Ahlgren et al.1 reported only an 8 percent prevalence of neural injury in a similar prospective, electrophysiological study of fifty total hip arthroplasties performed through the posterior approach. The second reference4 cited by Lewallen in support of his contention that the operative approach has not been shown to influence the prevalence of nerve palsy does not mention neural injury. I assume that he intended to cite the study by Navarro et al.7, who reported a 0.6 percent rate of neural injury in association with the posterior approach and a 1 percent rate in association with the transtrochanteric approach. This difference was not significant, but, again, the available evidence does not seem to incriminate the posterior approach. The direct lateral approach and the posterior approach seem equally safe8.
Lewallen's belief that the sciatic nerve is at greater risk of direct neural injury with the posterior approach is, unfortunately, one that is shared by others2,6,8. However, to the best of my knowledge, no evidence has been provided to support this belief. The posterior approach is quicker and is associated with less blood loss than other approaches9,10; both of these factors have been shown to be negatively correlated with neural injury after hip replacement3. The posterior approach is at least as safe as any other approach—if not safer.
Adrian E. Weale, F.R.C.S., F.R.C.S.(Orth)
Avon Orthopaedic Centre,
Southmead Hospital, Bristol BS10 5NB, England
Dr. Lewallen replies:
Mr. Weale very eloquently documents the variability in the prevalence of nerve palsy following hip arthroplasty and the lack of convincing evidence of a substantial difference in the risk of neurological injury associated with different operative approaches. Discussion of the operative approach used for hip arthroplasty, the choice of which is so often influenced by early training and the habits of the surgeon, is capable of generating heated exchanges between otherwise cordial colleagues, even at my institution. The intent of my Instructional Course Lecture was to convey the message that concern regarding the rate of nerve palsy should not influence the choice of operative approach. This decision should be guided by other factors. However, it is reasonable to recognize that there are important differences between operative approaches with regard to the anatomy through which the dissection occurs. It appears that the source of confusion regarding my article relates to the issue of direct neural injury compared with the overall reported rate of nerve palsy, which is undoubtedly due to both direct and indirect insult to the nerve in different patients. It seems reasonable to be concerned about the greater risk posed to the sciatic nerve by a knife, a cauterizing instrument, or another device when the nerve is directly exposed, visualized, palpated, retracted, and, in some cases, even formally dissected during exposure of the hip compared with the risk posed by an alternative operative approach, such as the anterolateral approach, in which no direct exposure or contact is made. It was my intention to suggest that caution be exercised to avoid direct injury to the sciatic nerve during the posterior approach because the nerve is clearly more proximate to the zone of operative dissection when that approach is used. This certainly has been the practice of the surgeons who perform and teach this operation at my institution, where a large portion of the group routinely uses the posterior approach for primary hip arthroplasty. With appropriate care, the posterior approach is as safe as any other. Without such care, catastrophic direct injury to the sciatic nerve can, and does, occur.
David G. Lewallen, M.D.
Department of Orthopedics, Mayo Clinic,
200 First Street S.W., Rochester, Minnesota 55905