Sciatic neuropathy is one of the most worrisome complications after hip fracture surgery. Remarkably, however, published reports concerning this complication are scarce. Sciatic palsy occurs in approximately 1% of patients who have total hip arthroplasties,9 but the incidence in patients having internal fixation for a fracture of the hip has not been described. From the literature regarding total hip arthroplasties, it is known that females and patients having revision surgery are particularly at risk and that spontaneous recovery may occur, but at least 15% of patients have a poor outcome characterized by persistent dysesthesia or weakness that limits ambulation or both.8
Possible causes of sciatic palsy after hip fracture surgery are surgical manipulation or compression, but in the majority of patients, the origin of the palsy is unknown.12 In one study of femoral neck fractures, preoperative traction was thought to contribute to sciatic palsy.4 This small study, however, did not find a difference between the groups that did and did not receive preoperative traction. Does preoperative traction contribute to postoperative sciatic neuropathy? We suggest that studies to discern this contribution should not focus on the presence or absence of traction, but on its duration. Animal studies have shown that even minimal tension on a nerve, if maintained for a significant amount of time, may result in postoperative weakness.2
We hypothesized that the duration of preoperative traction is related to the occurrence of sciatic nerve palsy and addressed the following questions: is postoperative sciatic neuropathy in patients with hip fractures related to the duration of preoperative traction; is postoperative sciatic neuropathy in patients with hip fractures related to other specific variables; what is the incidence of postoperative sciatic neuropathy in patients with hip fractures; and is the incidence of postoperative sciatic neuropathy in patients treated for hip fractures comparable with the 1% incidence after total hip arthroplasties?
MATERIALS AND METHODS
We performed a retrospective analysis of all consecutive patients who had hip fractures surgically treated from 1995 to 2002 and assessed the incidence of postoperative sciatic neuropathy. The groups with and without postoperative sciatic neuropathy were compared. Because the literature lacks information on the incidence of sciatic neuropathy after hip fracture and the potential influence of skin traction on this incidence, we were unable to perform a power analysis.
We identified all 2369 patients who had hip fractures treated in our hospital from January 1, 1995 to January 1, 2003. During this period, preoperative skin traction with a 3-kg weight, commencing at admission, was part of the treatment protocol of our hospital for all patients with this type of fracture. One hundred sixty-five patients were excluded because of incomplete or missing files, and two were excluded because of a pretraumatic sciatic neuropathy. Of the 2202 patients studied, the median age of the patients was 82 years (interquartile range, 75-88 years); 535 were men and 1667 were women.
The function of the sciatic nerve was routinely evaluated postoperatively in all patients by specifically testing for toe and ankle dorsiflexion. When neuropathy was suspected, a neurology consultation was obtained, and the diagnosis was confirmed or rejected after electrodiagnostic studies.
From the files, we determined the level of pretraumatic mobility (independent, walking aid, or wheelchair-dependent), delay between admission and operation (ie, the duration of traction), the type of fracture (medial femoral neck, lateral femoral neck, peritrochanteric, or subtrochanteric fracture), fracture displacement, the type of operation (arthroplasty or internal fixation by dynamic hip screw, gamma nail, or cannulated screws), operative duration, and experience of the surgeon (consultant or resident).
Because of the asymmetric distribution, outcomes between the groups with and without sciatic palsy are presented as medians with their interquartile ranges (IQR) (ie, the spread from 25th to 75th percentiles). Comparison between data from the groups with and without sciatic palsy was performed using the nonparametric Mann-Whitney test. Two-sided p values < 0.05 were considered to indicate statistical significance.
Before the hip fractures occurred, 1043 patients walked independently, 1049 used a walking aid, and 110 patients were wheelchair-dependent. The median delay between admission and surgery, and therefore, the duration of traction was 0.8 days (interquartile range, 0.6-1.2 days). The fracture included the medial femoral neck in 1280 patients, the lateral femoral neck in 113 patients, and was peritrochanteric in 685 patients and subtrochanteric in 124 patients. Fracture displacement occurred in 1808 patients. The type of operation was an arthroplasty in 1058 patients, a dynamic hip screw in 578 patients, a gamma nail in 316 patients, and internal fixation by cannulated screws in 250 patients. The median duration of the surgery was 56 minutes (interquartile range, 45-72 minutes). The operation was performed by a consultant in 1426 patients and by a resident in 776 patients.
The duration of traction was longer (p < 0.001) in the group that had sciatic palsy (Table 1). The median duration of preoperative traction was 2.6 days in the sciatic palsy group compared with 0.9 days in the group without this complication.
Patients in the group that had sciatic palsy were older (p < 0.04). Patients with a sciatic neuropathy had a median age of 86 years compared with 82 years in the group without neuropathy. We found no other differences between the two groups.
Sixteen of 2202 patients (0.7%) had sciatic palsy develop, which is comparable with the 1% incidence after total hip arthroplasty.
Sciatic neuropathy after surgical treatment of hip fractures is a rare but serious complication. Hip fractures most often occur in elderly patients and foot drop is a burden to their rehabilitation process, especially because at least 15% of patients experience permanent loss of nerve function.8 An incidence of 1% of patients with sciatic palsy after total hip arthroplasties has been reported,9 but this complication in patients having internal fixation for a fracture of the hip is mentioned in only a couple reports.10,11 Preoperative traction has been suggested to contribute to sciatic palsy after hip fracture.4 To find or exclude a relationship between the duration of preoperative traction and postoperative sciatic neuropathy, and to assess the incidence of this complication after hip fracture surgery, a retrospective analysis was performed.
Our study has several limitations. It is possible we underestimated the incidence by missing some cases. However, all patients in our hospital in whom neuropathy is suspected postoperatively are seen by a neurologist and routinely have electrodiagnostic studies. Therefore, we presume few, if any, cases were missed. Another limitation is that we could not determine the number of patients needed to answer our questions. Because our study is the first, to our knowledge, to present the incidence of sciatic neuropathy after hip fracture surgery, we were unable to perform a power analysis. An inclusion of more than 2000 consecutive patients showing an incidence of 0.7%, which is in line with the percentage in published reports of total hip arthroplasties, however, makes bias caused by group size unlikely.
Only one previous study of femoral neck fractures showed that preoperative traction might contribute to paralysis of the sciatic nerve.4 This small study involving only 456 patients, however, did not confirm such an effect. Given the infrequency of the problem, a large group of patients with hip fractures is required for adequate power. The negative effect of traction on nerve function is well known. Fowler et al reported that, in rats, tension on a nerve when maintained long enough, always will lead to loss of function.2 Skin traction in humans is not directly comparable with direct nerve traction in rats because of resistance of skin, bone, and muscle against tension in humans. However, in humans an effect of traction on nerve function has been described.1,3 Our study indicates that the median duration of traction in patients who had sciatic neuropathy develop was approximately 2 days longer than in control patients. Probably, the median duration of traction that a human leg (or sciatic nerve) can resist is between 1 and 3 days. The fact that patients who had sciatic neuropathy develop were older possibly can be explained by the idea that frail legs with minimal muscle can withstand skin traction for a shorter time. Decline of nerve quality with age is another potential explanation.7 Finally, there is a possibility that during traction nerve compression occurs. There are reports of temporary paralyses caused by nerve compression in traction splints,5 metal traction frames,11 or on traction tables.3 However, we do not understand how simple preoperative skin traction can result in nerve compression and do not consider this a likely explanation.
Preoperative skin traction has been a standard approach for treatment of hip fractures for years. An important systemic review on the effectiveness of preoperative traction by Parker and Handoll6 has raised serious doubts concerning the procedure. Their study provided no evidence of benefit from skin or skeletal traction for relief of pain, ease of fracture reduction, or quality of fracture reduction at the time of surgery.6 As a result, our hospital no longer routinely uses preoperative traction in patients with hip fractures. The potential damage to the sciatic nerve, as seen in the patients in the current study, is another reason to avoid preoperative traction. In cases in which traction still seems indicated, we advise a period of traction no longer than 1 day.
Our data suggest sciatic neuropathy after hip fracture surgery is related to the duration of preoperative traction. A weaker relationship exists between postoperative sciatic neuropathy and age. The factors of preoperative mobility, type of operation, operative duration, and experience of the surgeon showed no relationship with postoperative sciatic neuropathy. The overall incidence of sciatic neuropathy after hip fracture surgery was found to be 0.7%. A comparable incidence of sciatic neuropathy has been described after total hip arthroplasties.
1. Fleming P, Lenehan B, O'Rourke S, McHugh P, Kaar K, McCabe JP. Strain on the human sciatic nerve in vivo during movement of the hip and knee. J Bone Joint Surg
2. Fowler SS, Leonetti JP, Banich JC, Lee JM, Wurster R, Young MR. Duration of neuronal stretch correlates with functional loss. Otolaryngol Head Neck Surg
3. Kruger DM, Kayner DC, Hankin FM, Falahee MH, Kaufer H, Matthews LS, Goldstein SA. Traction force profiles associated with the use of a fracture table: a preliminary report. J Orthop Trauma
4. Levi N. Is preoperative tibial traction responsible for peroneal nerve palsy in patients with a fractured hip? Acta Orthop Belg
. 1998;64: 273-276.
5. Mihalko WM, Rohrbacher B, McGrath B. Transient peroneal nerve palsies from injuries placed in traction splints. Am J Emerg Med
6. Parker MJ, Handoll HH. Pre-operative traction for fractures of the proximal femur. Cochrane Database Syst Rev
7. Peters A. The effects of normal aging on myelin and nerve fibers: a review. J Neurocytol
8. Schmalzried TP, Amstutz HC, Dorey FJ. Nerve palsy associated with total hip replacement: risk factors and prognosis. J Bone Joint Surg
9. Schmalzried TP, Noordin S, Amstutz HC. Update on nerve palsy associated with total hip replacement. Clin Orthop Relat Res
10. Simon JP, Van Delm I, Fabry G. Sciatic nerve palsy following hip surgery. Acta Orthop Belg
11. Vermeiren J, Brabants K, Van Hoye M. Paralysis of the peroneal nerve following hip fracture treatment. Acta Orthop Belg
. 1995;61: 122-125.
12. Yuen EC, Olney RK, So YT. Sciatic neuropathy: clinical and prognostic features in 73 patients. Neurology