SECTION II: ORIGINAL ARTICLES: Trauma
Femoral neck fractures are a heavy economic burden on healthcare systems with reported global numbers of 1.3 million in 1990.10 Approximately half of these fractures are intracapsular. This type of fracture has been termed the “unsolved fracture” with continuing debate as to whether the femoral head should be replaced with an arthroplasty or retained using internal fixation.8,14 Recent randomized trials on this subject have favored arthroplasty for displaced fractures in the elderly.2,11 However, debate continues regarding what constitutes an elderly patient, with no clear consensus on treatment methods.3 Nonetheless, the patient's age generally is used as a determinant of treatment by many orthopaedic surgeons with internal fixation being used for younger patients and arthroplasty for the elderly. In the absence of this information, the surgeon is unable to make a well-informed choice of treatment.
Internal fixation has clear advantages; it is a quicker operation, involves less operative blood loss, and has lower infection rates.2 The main concerns regarding internal fixation are nonunion and avascular necrosis. An estimation of the risk of these complications occurring is essential before a surgeon can make a balanced judgment of the relative merits of internal fixation over arthroplasty. The overall incidence of these complications has been reported in numerous studies and these results have been summarized in review articles. For undisplaced fractures, an expected incidence of nonunion of 4.3% and avascular necrosis of 2.2% has been reported.7 For displaced fractures, the figures are 33% and 16%, respectively.12 These figures, however, only relate to patients of all ages and give no reference to how this figure may change with the age of the patient.
We undertook a Medline literature search and review of our collection of all articles on hip fractures up to 2005. We also scrutinized references from these articles to identify articles that reported the incidence of fracture healing complications related to age and/or gender. The only study identified with sufficient patient numbers to adequately address the relationship between the age and gender of the patient and the risk of nonunion was from patients treated in the 1960s.1 Subsequent articles lacked sufficient patient numbers to provide this information other than suggesting a relationship among age, gender, and fracture-healing complications.4,18,19 Assuming the results for displaced and undisplaced fractures are analyzed separately, using a power calculation for linear trend, a minimum of 270 patients is required to detect a significant difference at the 90% level.13
What is the incidence of nonunion related to either age or gender? The aim of our study was to determine the incidence of nonunion related to age and gender for a large series of patients with intracapsular fractures treated by internal fixation.
MATERIALS AND METHODS
Between January 1989 and September 2004, we admitted 2559 patients with an intracapsular fracture of the hip. Data regarding treatment and outcome for these patients were prospectively recorded. We excluded those treated by arthroplasty (1353 cases [52.8%]) and those treated nonoperatively (49 cases [1.9%]). We also excluded 17 patients with a pathologic fracture from tumor or Paget's disease treated by internal fixation, three patients with an associated femoral shaft fracture, and one patient treated with blade plate fixation. This left 1133 patients included in this study. The average age of the patients was 76 years (range, 16-100 years); 258 (22.7%) were male, 862 (76.1%) lived in their own home, 162 (14.3%) lived in residential homes, 61 (5.4%) lived in nursing homes, and 48 (4.2%) sustained the injury while a hospital inpatient. Four patients were lost to the 1-year followup; the followup for these patients was 43, 100, 193, and 290 days. Two hundred forty-eight patients died before the 1-year followup. For the 881 surviving patients, the average clinical followup was 474 days (range, 300-3670 days) and the average radiographic followup was 287days (range, 0-3433 days).
Fractures were classified as either undisplaced (Garden Grade I or II) or displaced (Garden Grade III or IV) by one of us (MJP) and this was related to the patient's age (Fig 1).9 Surgery entailed the use of the fracture table and radiograph image intensification. Undisplaced fractures were fixed in situ and displaced fractures were reduced closed.16 There were no open reductions or cap-sulotomies. All fractures were fixed with either a sliding hip screw (14 cases) or three cannulated cancellous screws (1119 cases). The implant was inserted either percutaneously or through a minimum skin incision. The grade of surgeon was a specialist hip fracture surgeon for 848 cases (MJP), orthopaedic consultant for 43 cases (nine different surgeons), and trainee orthopaedic surgeon for 242 cases (approximately 47 different surgeons). After surgery, the postoperative regimen consisted of immediate full weightbearing, except for the subgroup of patients younger than 60 years with a displaced intracapsular fracture in which partial weightbearing for 6 to 8 weeks was recommended. After discharge from the hospital, patients were followed up in a hip fracture clinic until the fracture showed radiographic evidence of fracture union. Referrals from the patients' general practitioner also were made to this clinic for any later complications related to the hip after discharge from followup. For patients unable to attend the clinic because of distance to travel or frailty, followup was by phone interview 1 year after injury with the patient or their caregivers. Patients with significant symptoms related to the hip were invited to return to the clinic for radiographic review.
We defined nonunion as failure of the fracture to show signs of bony union on the anteroposterior or lateral radiograph 1 year after surgery or redisplacement of the fracture for which surgical treatment was undertaken. All radiographs were assessed by one observer (MJP).
Statistical evaluation was performed using Fisher's exact test to compare groups for the outcomes of nonunion related to fracture displacement and gender and the chi square test for trend to identify a trend for increase or decrease in incidence in nonunion related to age. A p value less than 0.05 was considered significant. For the chi square test for trend, only patients between 40 and 100 years of age were included because there were few patients younger than 39 years.
The overall incidence of nonunion for all patients was 19.3%. Fracture nonunion was less common (p < 0.0001) for undisplaced fractures than for displaced fractures (48 of 565 [8.5%] versus 171 of 568 [30.1%]) (Table 1). At the final followup, surgical treatment generally with an arthroplasty was performed for 195 of 219 (89.0%) patients with nonunions.
What was the relationship between age and fracture nonunion? We found no association between the age of the patient and the incidence of nonunion in either displaced (p = 0.17) or undisplaced (p = 0.43) fractures (Fig 2). However, when patients who died before the 1-year assessment were excluded, there was an increase in the incidence of nonunion in displaced (p = 0.001) and undisplaced (p = 0.0029) fractures (Fig 3).
What was the relationship between gender and fracture nonunion? For undisplaced fractures, the incidence of nonunion was eight of 130 (6.1%) for men versus 40 of 435 (9.2%) for women (p = 0.37). The incidence of nonunion for displaced fractures was higher (p = 0.001) in women than in men (144 of 427 [33.7%] versus 27 of 141 [19.2%]).
The aim of this study was to determine the incidence of failure of an intracapsular fracture to heal in a contemporary setting and relate this to age and gender. Fracture nonunion and avascular necrosis are the two main healing complications, but it is nonunion that is the most problematic because it has a high prevalence and generally requires revision surgery. Avascular necrosis is less common and causes significant and disabling symptoms in only approximately ⅓ of cases.1
We note several limitations. Whereas the study of Barnes et al1 had multiple observers for the radiographs, all radiographic observations in our study were made by one observer (MJP). This eliminates interobserver variation for assessment of this outcome but increases the risk of systematic bias. The indications for internal fixation for patients in our study were essentially the preference of the attending physician under which the patient initially was admitted. Internal fixation generally was used in undisplaced fractures and displaced fractures in patients younger than approximately 70 years and frequently used for displaced fractures in very frail patients to avoid the larger operation of arthroplasty and for those on oral an-ticoagulation medication. In addition, 226 of the patients were treated by internal fixation in a randomized trial of internal fixation versus arthroplasty in the elderly.15 These factors enabled us to have a broad age range of patients treated by internal fixation.
Although intracapsular fracture of the femoral neck is a common injury, few reports in the literature provide sufficient numbers of cases studied to adequately answer the question. Barnes et al1 studied 1254 women and 249 men and, similar to our study, reported an increased risk of nonunion with age. For displaced fractures, the nonunion rate was 28% for patients younger than 75 years versus 41% for patients older than 75 years. In our literature search, we found no other studies on this topic that had sufficient patient numbers to make definite conclusions regarding the relationship of fracture-healing complications with age and gender. Three studies of 228, 118, and 300 patients were identified that reported an increased incidence of nonunion with age,4,18,19 whereas three studies of 195, 200, and 52 patients failed to find any such association.5,6,20
The patients in the study of Barnes et al1 were treated by numerous fixation implants. In this study, our implant of choice was three cancellous screws, although the sliding hip screw device was used in 14 cases at the discretion of the attending surgeon. Given the absence of differences in fracture-healing complications between these two implants,17 this should not result in any bias to the results presented.
Unlike in our study, Barnes et al1 failed to find any difference in the occurrence of nonunion related to gender even when the figures were corrected for age. However, they did report an increased occurrence of avascular necrosis for women. A smaller study of 195 patients showed an increased risk of nonunion for women,5 but other smaller studies failed to find any such association.18,20
Numerous other factors that may influence later development of fracture-healing complications have been studied. These include patient factors such as the presence of rheumatoid arthritis and metabolic bone disease and surgical factors such as timing of surgery, quality of the reduction and fixation, and positioning of the implant. These factors, although relevant in determining the best treatment method or outcome for a patient with an intracapsular fracture, are outside the scope of this study.
The main complication of internal fixation of intracapsular fractures is nonunion of the fracture. This term is used for fractures in which the fixation fails to hold and the fracture is displaced before union has occurred or the more unusual situation of the fracture failing to heal with the fixation intact. The median time from surgery to revision surgery for this complication was 12 weeks (mean, 19 weeks). This suggests the main cause for this complication is mechanical with failure of the implant to resist the deforming forces around the fracture. The incidence of nonunion decreased for patients older than 80 years because there were more deaths in the elderly before the fracture had time to develop a nonunion. Nonunion was lower in the male patients and this may be attributable to the improved bone strength enabling the fixation device to have greater hold on the fracture.
This study provides the clinician with the predicted incidence of fracture-healing complications of nonunion related to age and gender. The main complication is nonunion with an overall incidence of 8% for undisplaced fractures and 30% for displaced fractures. This complication was generally followed by surgical treatment. Age and gender were related to the occurrence of fracture-healing complications. Knowledge of the predicted incidence of the fracture-healing complications, particularly that of nonunion, should enable the surgeon to make a more balanced decision between internal fixation and arthroplasty for this condition.
1. Barnes R, Brown JT, Garden RS, Nicoll EA. Subcapital fractures of the femur: a prospective review. J Bone Joint Surg Br
. 1976;58: 2-24.
2. Bhandari M, Devereaux PJ, Swiontkowski MF, Tornetta P3rd
, Obremskey W, Koval KJ, Nork S, Sprague S, Schemitsch EH, Guyatt GH. Internal fixation compared with arthroplasty for displaced fractures of the femoral neck: a meta-analysis. J Bone Joint Surg Am
3. Bhandari M, Devereaux PJ, Tornetta P3rd
, Swiontkowski MF, Berry DJ, Haidukewych G, Schemitsch EH, Hanson BP, Koval K, Dirschl D, Leece P, Keel M, Petrisor B, Heetveld M, Guyatt GH. Operative management of displaced femoral neck fractures in elderly patients: an international survey. J Bone Joint Surg Am
. 2005; 87:2122-2130.
4. Boyd HB, Salvatore JE. Acute fracture of the femoral neck: internal fixation or prosthesis? J Bone Joint Surg Am
5. Brown JT, Abrami G. Transcervical femoral fracture: a review of 195 patients treated by sliding nail-plate fixation. J Bone Joint Surg Br
6. Brown TI, Court-Brown C. Failure of sliding nail-plate fixation in subcapital fractures of the femoral neck. J Bone Joint Surg Br
7. Conn KS, Parker MJ. Undisplaced intracapsular hip fractures: results of internal fixation in 375 patients. Clin Orthop Relat Res
8. Dickson JA. The unsolved fracture: a protest against defeatism. J Bone Joint Surg Am
9. Garden RS. Low-angle fixation in fractures of the femoral neck. J Bone Joint Surg Br
10. Johnell O, Kanis JA. An estimate of the worldwide prevalence, mortality and disability associated with hip fracture. Osteoporos Int
11. Keating JF, Grant A, Masson M, Scott NW, Forbes JF. Randomized comparison of reduction and fixation, bipolar hemiarthroplasty, and total hip arthroplasty: treatment of displaced intracapsular hip fractures in healthy older patients. J Bone Joint Surg Am
. 2006;88: 249-260.
12. Lu-Yao GL, Keller RB, Littenberg B, Wennberg JE. Outcomes after displaced fractures of the femoral neck: a meta-analysis of one hundred and six published reports. J Bone Joint Surg Am
. 1994;76: 15-25.
13. Nam J. A simple approximation for calculating sample sizes for detecting linear trend in proportions. Biometrics
14. Nicoll EA. The unsolved fracture. J Bone Joint Surg Br
. 1963;45: 239-241.
15. Parker MJ, Khan RJ, Crawford J, Pryor GA. Hemiarthroplasty versus internal fixation for displaced intracapsular hip fractures in the elderly: a randomised trial of 455 patients. J Bone Joint Surg Br
16. Parker MJ, Pryor GA, Thorngren KG. Handbook of Hip Fracture Surgery
. Oxford, UK: Butterworth-Heinemann Publications; 1997.
17. Parker MJ, Stockton G. Internal fixation implants for intracapsular proximal femoral fractures in adults. Cochrane Database Syst Rev
18. Stappaerts KH, Broos PLO. Internal fixation of femoral neck fractures: a follow-up study of 118 cases. Acta Chir Belg
. 1987;87: 247-251.
19. Stromqvist B, Hansson LI, Nilsson LT, Thorngren KG. Hook-pin fixation in femoral neck fractures: a two-year follow up study of 300 cases. Clin Orthop Relat Res
© 2007 Lippincott Williams & Wilkins, Inc.
20. Unger AS, Shuster HF. Predicting the healing of displaced subcapital hip fractures via postoperative roentgenographic factors. Orthop Rev