- Email: [email protected]
Undisplaced femoral neck fractures—no problems? A consecutive study of 224 patients treated with internal fixation
Injury, Int. J. Care Injured 40 (2009) 274–276
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Undisplaced femoral neck fractures—no problems? A consecutive study of 224 patients treated with internal fixation Cecilia Rogmark *, Louise Flensburg, Hans Fredin Lund University, Department of Orthopaedics, Malmo¨ University Hospital, SE-205 02, Malmo¨, Sweden
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 19 May 2008
224 patients with undisplaced femoral neck fractures treated with two parallel Hansson hook pins were studied. After a mean follow-up time of 32 months (S.D. 5.2), 15% had a reoperation. 11% were considered failures, mostly avascular necrosis, and 9% had a secondary arthroplasty. Possible risk factors for poor outcome were analysed. Neither high age nor surgical delay was associated with increased failure rate. Survivors received a questionnaire, and 40% stated that they had mild or severe pain in the hip when walking, 25% had pain at rest and 25 stated that they thought ‘‘always’’ or ‘‘often’’ about their injury. The younger the patient, the more frequent the report of subjective pain. 51% of individuals under 80 years reported pain when walking, compared to 27% aged 80 or older (p = 0.016). Corresponding numbers for pain at rest were 32 and 12% (p = 0.034). The failure rate did not differ between the age groups, but the younger patients had more reoperations (p = 0.046) and thought more frequently about their injury (p = 0.016). An undisplaced femoral neck fracture is a major injury with a long-term daily discomfort in about 25% and clinical failure in 11%. ß 2008 Elsevier Ltd. All rights reserved.
Keywords: Hip fracture Femoral neck fracture Internal fixation Osteoporotic fracture Avascular necrosis Non-union Rehabilitation
Introduction The major discussion concerning the undisplaced femoral neck fracture has been whether to treat it with or without operation. Once settled in favour of internal fixation,4,5,11 the fracture type is nowadays regarded as common, easy to treat and with few healing complications. Hence, there are a few studies presented,1,2,12 agreeing on a failure rate of 5–10% after internal fixation. Hui et al.9 suggested that women over 80 years had a worse outcome, with a failure rate of 31%. Eisler et al.6 on the other hand suggested that age did not affect the end result. The aim of the study was to update the knowledge of the clinical results after surgically treated undisplaced femoral neck fracture, and in particular to determine if the ageing of the hip fracture population leads to a worse result in terms of clinical failures and reoperations. Materials and methods 224 consecutive patients operated on between September 2000 and December 2002 at Malmo¨ University Hospital were
* Corresponding author. Tel.: +46 40 331000; fax: +46 40 336200. E-mail address: [email protected] (C. Rogmark). 0020–1383/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2008.05.023
included in a retrospective study. All had an undisplaced femoral neck fracture (Garden 1–2)7 and were treated with two parallel Hansson hook pins1 (Swemac Orthopaedics AB, Sweden) as an emergency procedure. 207 (92%) had spinal anaesthesia. Immediate weight bearing was allowed. Patients with pathological fractures were excluded. After a mean follow-up time of 32 months (S.D. 5.2), all the records were scrutinised for out-patient visits, admissions and reoperations up to the day of follow-up or death. In the Malmo¨ region the only two hospitals able to treat hip complications were checked. Addresses and date of death data were drawn from the national registration. All living patients received a written questionnaire and a prepaid return envelope. If necessary, one reminder was sent out. 132 survivors received the questionnaire and 102 answered (77%). Those who reported walking disability or pain from their hip were invited to a physical and radiological examination. End-points were clinical failure and reoperations. In statistical calculations Chi-square test was used. Results 156 of the 224 patients (69%) were women. Median age was 81 years (31–98). 152 (68%) came from independent living and 82 had no walking aids (37%). 50 individuals (22%) had a diagnosis of senile dementia. 22 fractures (10%) were older than 5 days when
C. Rogmark et al. / Injury, Int. J. Care Injured 40 (2009) 274–276 Table 1 General complications during hospital stay (some patients had more than one) (N = 224).
Urinary tract infection Confusion Heart disease Lung disease Pneumonia Stroke Hematoma/anemia Gastrointestinal disease Urological disease Depression Thrombosis Other
n
Percent
18 15 10 6 6 3 3 3 3 2 1 6
8 7 4 3 3 1 1 1 1 1 0.5 3
275
Table 3 Objective and subjective findings at follow-up.
Objective findings No hip pain No dementia Back to original living No walking aids Subjective findings Thinks never/seldom about the hip fracture Ability to climb stairs No hip pain when walking No hip pain at rest Same walking ability as before fracture No need of orthopaedic consultation
n
N
Percent
63 87 117 39
99 110 144 135
64 79 81 29
37 64 53 74 30 71
102 102 102 102 102 102
36 63 52 73 29 70
Data from medical records and self-administered questionnaire, respectively. n = number of observations. N = number of patient with available data.
diagnosed. 48 (21%) had a contralateral hip fracture, either before the index fracture or during follow-up. In 12 cases (5%) an additional MRI was needed to confirm the fracture diagnosis. One of these fractures finally led to avascular necrosis. Median waiting time for surgery was 21 h (3–65) and median surgery time 19 min (5–55). In both calculations, outliers were excluded. Median hospital stay was 6 days (0–35). 81 (36%) had one or more complications during initial hospital stay (see Table 1). At 32 months follow-up (mean), 92 (41%) were deceased. 1month mortality was 6% and 1-year mortality was 22%. 61 were deceased without any further postoperative contact with the orthopaedic departments. 103 patients (46%) had a late postoperative radiological examination; either as part of an earlier visit or as a 3-year follow-up initiated by their answers to a questionnaire (see below). 34 patients (15%) had any reoperation, defined as open surgery (Table 2). 24 cases (11%) were considered failures, caused by avascular necrosis in 12 cases, non-union in 10 and severe pain in one case. One patient had a re-fracture through the pinholes, and was reoperated with a sliding hip screw. 16 of the patients with failed internal fixation received a secondary total hip arthroplasty, five a hemiarthroplasty. Due to medical reasons, secondary surgery was not performed in two patients with avascular necrosis. The outcome was classified as unknown for 25 patients who died within the first 4 months postoperatively and for two who had clinical symptoms but declined further investigation. 10 had pain but no other clinical or radiological signs of failure. Possible risk factors for poor outcome were analysed. Neither women over 80 years, fractures older than 5 days at diagnosis nor more than 24 h waiting time before surgery were combined with increased failure rate. Objective findings by the treating surgeon during follow-up are shown in Table 3. The patient’s perception of the fracture at follow-up is shown in Table 3. 41 patients (40%) stated that they had mild or severe pain Table 2 Secondary surgery during follow-up (including removal of pins after healed fracture) (N = 224). n Total hip arthroplasty Hemiarthroplasty Extraction of pins New fracture surgery Exchange of pin(s)
16 5 10 2 1
Sum
34
Percent 7 2 4 1 0.4 15
in the hip when walking, 26 (25%) had pain at rest and 26 stated that they thought ‘‘always’’ or ‘‘often’’ about their injury. Only 30 patients had recovered their walking ability, but the major reason for impairment was ‘‘other health problems’’ (43 patients). 27 individuals reported that the hip limited their walking ability. The younger the patient, the more frequent the report of subjective pain. 51% of individuals under 80 years reported pain when walking, compared to 27% aged 80 or older (p = 0.016). Corresponding numbers for pain at rest were 32 and 12% (p = 0.034). On the other hand, pain noted by the doctor was the same in all age groups. The younger patients also had more reoperations (p = 0.046) and reported that they thought more frequently about their injury (p = 0.016). The rate of failure did not show any significant difference between the age groups. Discussion Failure rate after undisplaced femoral neck fractures remains relatively low, 11%. Nevertheless, the fracture afflicts the patients’ lives. After nearly 3 years, one-quarter of the surviving patients had troubles with their injured hip, and four of ten had pain when walking. Younger patients reported significantly more pain. When invited to radiological and clinical examination, the vast majority had normal X-rays and clinical findings. Extraction of pins led to pain relief in half of these patients. One can speculate whether the hip fracture induces altered gait pattern or disturbed propriception, resulting in hip pain when moving. Another explanation may be symptomatic necrotic changes in the femoral head, not visible on plain X-rays. We did not perform MRI means on these patients. Notably, 6 of 16 patients treated with a secondary THA due to avascular necrosis or non-union, continued to have pain afterwards. This may strengthen the theory that extraarticular factors contribute to the discomfort. Another explanation is that an otherwise healthy 60-year-old will focus much more on an injured hip than a 90-year-old with other psychical and cognitive limitations. Clayer and Bauze3 also found that patients under 60 years with any kind of hip fracture reported more pain. In the annual report of the Swedish Hip Fracture Register ‘‘Riksho¨ft’’13 the rate of pain at 4 months was found to be higher after an undisplaced femoral neck fracture than after both displaced femoral neck fractures and stable trochanteric fractures. But when comparing undisplaced and displaced femoral neck fractures, Tidermark et al.14 found better quality of life in patients with healed undisplaced fractures. The patients included were older than 65 years, whereas in the current study we found a worse subjective outcome in the younger group.
276
C. Rogmark et al. / Injury, Int. J. Care Injured 40 (2009) 274–276
Hemiarthroplasty could be a treatment alternative for undisplaced femoral neck fractures, as suggested by Hui et al.9 But the gain in a marginally lower reoperation rate may be outweighed by the longer surgery time, more blood loss and higher infection rate related to arthroplasty surgery. The mortality in our study is the same as in other consecutive femoral neck fracture studies,8,10 but we had no lower age limit in the current study. When the six patients under 50 years are excluded, the 1-year mortality is still 22%. Older studies using social data after hip fractures usually reported that only 50% remain in their original habitat. Our study found that as many as 81% do, to some part reflecting a more benign result after an undisplaced fracture. Moreover, it reflects the modern Swedish elderly care. When an individual is no longer capable of independent living a facilitated living is offered. The person lives there until death, and the intensity of help is increased according to the need, i.e. further changes of living seldom occur. Knowledge of different types of local and national care systems is crucial when comparing socio-economic results after hip fractures. The weakness of the study is its retrospective design, but the Swedish national registration system with unique personal numbers makes tracing of the patients easy and results in exact mortality and habitat data. Any reoperation would certainly have been recorded, since the Swedish health care system provides public hospitals with well-defined catchment areas. The few individuals who moved to other areas during the follow-up were contacted, as well as their new hospitals. There might be some cases of failure not brought to our knowledge, in the group of 61 individuals who died without any further contact with the orthopaedic departments. Normally, the general practitioners refer the patient to us if hip problems develop, but in cases with concomitant severe decease they may leave the hip out of consideration. Regarding the written questionnaire, we consider an response rate of 77% acceptable. The aim of the questionnaire was to describe the patient’s point of view, rather than objective results, as reoperations, that were drawn from medical records and not afflicted by the 23% non-responders. Conclusion The failure rate after internal fixation of an undisplaced femoral neck fracture is 11%, as described in previous studies. The increasing number of very old and osteoporotic patients during the last decades has not altered these findings. The reasons for failure after approximately 3 years are to the same extent avascular necrosis and non-union. When extraction of pins after healing of the fracture is included, 15% had secondary surgery and 9% had a secondary arthroplasty. In the self-administered questionnaire as many as 40% reported pain when walking and 25% had pain at rest. 25% thought ‘‘always’’ or ‘‘often’’ about their injured hip after approximately 3 years. The
high rate of late problems is a new and somewhat surprising finding, especially as orthopaedic surgeons usually consider the undisplaced femoral neck fracture as a less severe hip fracture. In Sweden, due to economical reductions in health care, this group of patients is no longer offered regular out-patient visits after the initial hospital stay. They are referred to their general practitioner or contacted over the telephone by a nurse or physiotherapist after some months. Whatever follow-up is planned; detailed information regarding the expected outcome to the patient at discharge is essential, in order to reduce anxiety about the injured hip. An undisplaced femoral neck fracture is a major injury with a long-term daily discomfort in about one-quarter of patients. Conflict of interest All authors certify that they not have signed any agreement with a commercial interest related to this study which would in any way limit publication of any and all data generated for the study or to delay publication for any reason. Fundings has been given by County Council of Ska˚ne and the Herman Ja¨rnhardts Foundation, neither had any involvement in the work with this study. References 1. Bentley G. Impacted fractures of the femoral neck. J Bone Joint Surg Br 1992;74(3):476–7. 2. Chiu FY, Lo WH, Yu CT, et al. Percutaneous pinning in undisplaced subcapital femoral neck fractures. Injury 1996;27(1):53–5. 3. Clayer MT, Bauze RJ. Morbidity and mortality following fractures of the femoral neck and trochanteric region: analysis of risk factors. J Trauma 1989;29(12): 1673–8. 4. Conn KS, Parker MJ. Undisplaced intracapsular hip fractures: results of internal fixation in 375 patients. Clin Orthop Relat Res 2004;(421):249–54. 5. Cserhati P, Kazar G, Manninger J, et al. Non-operative or operative treatment for undisplaced femoral neck fractures: a comparative study of 122 non-operative and 125 operatively treated cases. Injury 1996;27(8):583–8. 6. Eisler J, Cornwall R, Strauss E, et al. Outcomes of elderly patients with nondisplaced femoral neck fractures. Clin Orthop 2002;(399):52–8. 7. Garden RS. Low-angle fixation in fractures of the femoral neck. J Bone Joint Surg Br 1961;43:647–64. 8. Holmberg S, Thorngren KG. Statistical analysis of femoral neck fractures based on 3053 cases. Clin Orthop 1987;(218):32–41. 9. Hui AC, Anderson GH, Choudhry R, et al. Internal fixation or hemiarthroplasty for undisplaced fractures of the femoral neck in octogenarians. J Bone Joint Surg Br 1994;76(6):891–4. 10. Nilsson LT, Stromqvist B, Thorngren KG. Nailing of femoral neck fracture. Clinical and sociologic 5-year follow-up of 510 consecutive hips. Acta Orthop Scand 1988;59(4):365–71. 11. Raaymakers EL. Undisplaced femoral neck fracture. To operate or not to operate? Acta Orthop Scand 1993;64(2):233–4. 12. Stromqvist B, Nilsson LT, Thorngren KG. Femoral neck fracture fixation with hook-pins. 2-year results and learning curve in 626 prospective cases. Acta Orthop Scand 1992;63(3):282–7. 13. Thorngren KG, Riksho¨ft, http://www.rikshoft.se/online/thePages/index.php. Edited, 2007. 14. Tidermark J, Zethraeus N, Svensson O, et al. Quality of life related to fracture displacement among elderly patients with femoral neck fractures treated with internal fixation. J Orthop Trauma 2002;16(1):34–8.
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Undisplaced femoral neck fractures—no problems? A consecutive study of 224 patients treated with internal fixation Cecilia Rogmark *, Louise Flensburg, Hans Fredin Lund University, Department of Orthopaedics, Malmo¨ University Hospital, SE-205 02, Malmo¨, Sweden
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 19 May 2008
224 patients with undisplaced femoral neck fractures treated with two parallel Hansson hook pins were studied. After a mean follow-up time of 32 months (S.D. 5.2), 15% had a reoperation. 11% were considered failures, mostly avascular necrosis, and 9% had a secondary arthroplasty. Possible risk factors for poor outcome were analysed. Neither high age nor surgical delay was associated with increased failure rate. Survivors received a questionnaire, and 40% stated that they had mild or severe pain in the hip when walking, 25% had pain at rest and 25 stated that they thought ‘‘always’’ or ‘‘often’’ about their injury. The younger the patient, the more frequent the report of subjective pain. 51% of individuals under 80 years reported pain when walking, compared to 27% aged 80 or older (p = 0.016). Corresponding numbers for pain at rest were 32 and 12% (p = 0.034). The failure rate did not differ between the age groups, but the younger patients had more reoperations (p = 0.046) and thought more frequently about their injury (p = 0.016). An undisplaced femoral neck fracture is a major injury with a long-term daily discomfort in about 25% and clinical failure in 11%. ß 2008 Elsevier Ltd. All rights reserved.
Keywords: Hip fracture Femoral neck fracture Internal fixation Osteoporotic fracture Avascular necrosis Non-union Rehabilitation
Introduction The major discussion concerning the undisplaced femoral neck fracture has been whether to treat it with or without operation. Once settled in favour of internal fixation,4,5,11 the fracture type is nowadays regarded as common, easy to treat and with few healing complications. Hence, there are a few studies presented,1,2,12 agreeing on a failure rate of 5–10% after internal fixation. Hui et al.9 suggested that women over 80 years had a worse outcome, with a failure rate of 31%. Eisler et al.6 on the other hand suggested that age did not affect the end result. The aim of the study was to update the knowledge of the clinical results after surgically treated undisplaced femoral neck fracture, and in particular to determine if the ageing of the hip fracture population leads to a worse result in terms of clinical failures and reoperations. Materials and methods 224 consecutive patients operated on between September 2000 and December 2002 at Malmo¨ University Hospital were
* Corresponding author. Tel.: +46 40 331000; fax: +46 40 336200. E-mail address: [email protected] (C. Rogmark). 0020–1383/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2008.05.023
included in a retrospective study. All had an undisplaced femoral neck fracture (Garden 1–2)7 and were treated with two parallel Hansson hook pins1 (Swemac Orthopaedics AB, Sweden) as an emergency procedure. 207 (92%) had spinal anaesthesia. Immediate weight bearing was allowed. Patients with pathological fractures were excluded. After a mean follow-up time of 32 months (S.D. 5.2), all the records were scrutinised for out-patient visits, admissions and reoperations up to the day of follow-up or death. In the Malmo¨ region the only two hospitals able to treat hip complications were checked. Addresses and date of death data were drawn from the national registration. All living patients received a written questionnaire and a prepaid return envelope. If necessary, one reminder was sent out. 132 survivors received the questionnaire and 102 answered (77%). Those who reported walking disability or pain from their hip were invited to a physical and radiological examination. End-points were clinical failure and reoperations. In statistical calculations Chi-square test was used. Results 156 of the 224 patients (69%) were women. Median age was 81 years (31–98). 152 (68%) came from independent living and 82 had no walking aids (37%). 50 individuals (22%) had a diagnosis of senile dementia. 22 fractures (10%) were older than 5 days when
C. Rogmark et al. / Injury, Int. J. Care Injured 40 (2009) 274–276 Table 1 General complications during hospital stay (some patients had more than one) (N = 224).
Urinary tract infection Confusion Heart disease Lung disease Pneumonia Stroke Hematoma/anemia Gastrointestinal disease Urological disease Depression Thrombosis Other
n
Percent
18 15 10 6 6 3 3 3 3 2 1 6
8 7 4 3 3 1 1 1 1 1 0.5 3
275
Table 3 Objective and subjective findings at follow-up.
Objective findings No hip pain No dementia Back to original living No walking aids Subjective findings Thinks never/seldom about the hip fracture Ability to climb stairs No hip pain when walking No hip pain at rest Same walking ability as before fracture No need of orthopaedic consultation
n
N
Percent
63 87 117 39
99 110 144 135
64 79 81 29
37 64 53 74 30 71
102 102 102 102 102 102
36 63 52 73 29 70
Data from medical records and self-administered questionnaire, respectively. n = number of observations. N = number of patient with available data.
diagnosed. 48 (21%) had a contralateral hip fracture, either before the index fracture or during follow-up. In 12 cases (5%) an additional MRI was needed to confirm the fracture diagnosis. One of these fractures finally led to avascular necrosis. Median waiting time for surgery was 21 h (3–65) and median surgery time 19 min (5–55). In both calculations, outliers were excluded. Median hospital stay was 6 days (0–35). 81 (36%) had one or more complications during initial hospital stay (see Table 1). At 32 months follow-up (mean), 92 (41%) were deceased. 1month mortality was 6% and 1-year mortality was 22%. 61 were deceased without any further postoperative contact with the orthopaedic departments. 103 patients (46%) had a late postoperative radiological examination; either as part of an earlier visit or as a 3-year follow-up initiated by their answers to a questionnaire (see below). 34 patients (15%) had any reoperation, defined as open surgery (Table 2). 24 cases (11%) were considered failures, caused by avascular necrosis in 12 cases, non-union in 10 and severe pain in one case. One patient had a re-fracture through the pinholes, and was reoperated with a sliding hip screw. 16 of the patients with failed internal fixation received a secondary total hip arthroplasty, five a hemiarthroplasty. Due to medical reasons, secondary surgery was not performed in two patients with avascular necrosis. The outcome was classified as unknown for 25 patients who died within the first 4 months postoperatively and for two who had clinical symptoms but declined further investigation. 10 had pain but no other clinical or radiological signs of failure. Possible risk factors for poor outcome were analysed. Neither women over 80 years, fractures older than 5 days at diagnosis nor more than 24 h waiting time before surgery were combined with increased failure rate. Objective findings by the treating surgeon during follow-up are shown in Table 3. The patient’s perception of the fracture at follow-up is shown in Table 3. 41 patients (40%) stated that they had mild or severe pain Table 2 Secondary surgery during follow-up (including removal of pins after healed fracture) (N = 224). n Total hip arthroplasty Hemiarthroplasty Extraction of pins New fracture surgery Exchange of pin(s)
16 5 10 2 1
Sum
34
Percent 7 2 4 1 0.4 15
in the hip when walking, 26 (25%) had pain at rest and 26 stated that they thought ‘‘always’’ or ‘‘often’’ about their injury. Only 30 patients had recovered their walking ability, but the major reason for impairment was ‘‘other health problems’’ (43 patients). 27 individuals reported that the hip limited their walking ability. The younger the patient, the more frequent the report of subjective pain. 51% of individuals under 80 years reported pain when walking, compared to 27% aged 80 or older (p = 0.016). Corresponding numbers for pain at rest were 32 and 12% (p = 0.034). On the other hand, pain noted by the doctor was the same in all age groups. The younger patients also had more reoperations (p = 0.046) and reported that they thought more frequently about their injury (p = 0.016). The rate of failure did not show any significant difference between the age groups. Discussion Failure rate after undisplaced femoral neck fractures remains relatively low, 11%. Nevertheless, the fracture afflicts the patients’ lives. After nearly 3 years, one-quarter of the surviving patients had troubles with their injured hip, and four of ten had pain when walking. Younger patients reported significantly more pain. When invited to radiological and clinical examination, the vast majority had normal X-rays and clinical findings. Extraction of pins led to pain relief in half of these patients. One can speculate whether the hip fracture induces altered gait pattern or disturbed propriception, resulting in hip pain when moving. Another explanation may be symptomatic necrotic changes in the femoral head, not visible on plain X-rays. We did not perform MRI means on these patients. Notably, 6 of 16 patients treated with a secondary THA due to avascular necrosis or non-union, continued to have pain afterwards. This may strengthen the theory that extraarticular factors contribute to the discomfort. Another explanation is that an otherwise healthy 60-year-old will focus much more on an injured hip than a 90-year-old with other psychical and cognitive limitations. Clayer and Bauze3 also found that patients under 60 years with any kind of hip fracture reported more pain. In the annual report of the Swedish Hip Fracture Register ‘‘Riksho¨ft’’13 the rate of pain at 4 months was found to be higher after an undisplaced femoral neck fracture than after both displaced femoral neck fractures and stable trochanteric fractures. But when comparing undisplaced and displaced femoral neck fractures, Tidermark et al.14 found better quality of life in patients with healed undisplaced fractures. The patients included were older than 65 years, whereas in the current study we found a worse subjective outcome in the younger group.
276
C. Rogmark et al. / Injury, Int. J. Care Injured 40 (2009) 274–276
Hemiarthroplasty could be a treatment alternative for undisplaced femoral neck fractures, as suggested by Hui et al.9 But the gain in a marginally lower reoperation rate may be outweighed by the longer surgery time, more blood loss and higher infection rate related to arthroplasty surgery. The mortality in our study is the same as in other consecutive femoral neck fracture studies,8,10 but we had no lower age limit in the current study. When the six patients under 50 years are excluded, the 1-year mortality is still 22%. Older studies using social data after hip fractures usually reported that only 50% remain in their original habitat. Our study found that as many as 81% do, to some part reflecting a more benign result after an undisplaced fracture. Moreover, it reflects the modern Swedish elderly care. When an individual is no longer capable of independent living a facilitated living is offered. The person lives there until death, and the intensity of help is increased according to the need, i.e. further changes of living seldom occur. Knowledge of different types of local and national care systems is crucial when comparing socio-economic results after hip fractures. The weakness of the study is its retrospective design, but the Swedish national registration system with unique personal numbers makes tracing of the patients easy and results in exact mortality and habitat data. Any reoperation would certainly have been recorded, since the Swedish health care system provides public hospitals with well-defined catchment areas. The few individuals who moved to other areas during the follow-up were contacted, as well as their new hospitals. There might be some cases of failure not brought to our knowledge, in the group of 61 individuals who died without any further contact with the orthopaedic departments. Normally, the general practitioners refer the patient to us if hip problems develop, but in cases with concomitant severe decease they may leave the hip out of consideration. Regarding the written questionnaire, we consider an response rate of 77% acceptable. The aim of the questionnaire was to describe the patient’s point of view, rather than objective results, as reoperations, that were drawn from medical records and not afflicted by the 23% non-responders. Conclusion The failure rate after internal fixation of an undisplaced femoral neck fracture is 11%, as described in previous studies. The increasing number of very old and osteoporotic patients during the last decades has not altered these findings. The reasons for failure after approximately 3 years are to the same extent avascular necrosis and non-union. When extraction of pins after healing of the fracture is included, 15% had secondary surgery and 9% had a secondary arthroplasty. In the self-administered questionnaire as many as 40% reported pain when walking and 25% had pain at rest. 25% thought ‘‘always’’ or ‘‘often’’ about their injured hip after approximately 3 years. The
high rate of late problems is a new and somewhat surprising finding, especially as orthopaedic surgeons usually consider the undisplaced femoral neck fracture as a less severe hip fracture. In Sweden, due to economical reductions in health care, this group of patients is no longer offered regular out-patient visits after the initial hospital stay. They are referred to their general practitioner or contacted over the telephone by a nurse or physiotherapist after some months. Whatever follow-up is planned; detailed information regarding the expected outcome to the patient at discharge is essential, in order to reduce anxiety about the injured hip. An undisplaced femoral neck fracture is a major injury with a long-term daily discomfort in about one-quarter of patients. Conflict of interest All authors certify that they not have signed any agreement with a commercial interest related to this study which would in any way limit publication of any and all data generated for the study or to delay publication for any reason. Fundings has been given by County Council of Ska˚ne and the Herman Ja¨rnhardts Foundation, neither had any involvement in the work with this study. References 1. Bentley G. Impacted fractures of the femoral neck. J Bone Joint Surg Br 1992;74(3):476–7. 2. Chiu FY, Lo WH, Yu CT, et al. Percutaneous pinning in undisplaced subcapital femoral neck fractures. Injury 1996;27(1):53–5. 3. Clayer MT, Bauze RJ. Morbidity and mortality following fractures of the femoral neck and trochanteric region: analysis of risk factors. J Trauma 1989;29(12): 1673–8. 4. Conn KS, Parker MJ. Undisplaced intracapsular hip fractures: results of internal fixation in 375 patients. Clin Orthop Relat Res 2004;(421):249–54. 5. Cserhati P, Kazar G, Manninger J, et al. Non-operative or operative treatment for undisplaced femoral neck fractures: a comparative study of 122 non-operative and 125 operatively treated cases. Injury 1996;27(8):583–8. 6. Eisler J, Cornwall R, Strauss E, et al. Outcomes of elderly patients with nondisplaced femoral neck fractures. Clin Orthop 2002;(399):52–8. 7. Garden RS. Low-angle fixation in fractures of the femoral neck. J Bone Joint Surg Br 1961;43:647–64. 8. Holmberg S, Thorngren KG. Statistical analysis of femoral neck fractures based on 3053 cases. Clin Orthop 1987;(218):32–41. 9. Hui AC, Anderson GH, Choudhry R, et al. Internal fixation or hemiarthroplasty for undisplaced fractures of the femoral neck in octogenarians. J Bone Joint Surg Br 1994;76(6):891–4. 10. Nilsson LT, Stromqvist B, Thorngren KG. Nailing of femoral neck fracture. Clinical and sociologic 5-year follow-up of 510 consecutive hips. Acta Orthop Scand 1988;59(4):365–71. 11. Raaymakers EL. Undisplaced femoral neck fracture. To operate or not to operate? Acta Orthop Scand 1993;64(2):233–4. 12. Stromqvist B, Nilsson LT, Thorngren KG. Femoral neck fracture fixation with hook-pins. 2-year results and learning curve in 626 prospective cases. Acta Orthop Scand 1992;63(3):282–7. 13. Thorngren KG, Riksho¨ft, http://www.rikshoft.se/online/thePages/index.php. Edited, 2007. 14. Tidermark J, Zethraeus N, Svensson O, et al. Quality of life related to fracture displacement among elderly patients with femoral neck fractures treated with internal fixation. J Orthop Trauma 2002;16(1):34–8.