G Model

JINJ-6214; No. of Pages 6 Injury, Int. J. Care Injured xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Injury journal homepage: www.elsevier.com/locate/injury

Intertrochanteric femur fractures in the elderly treated with either proximal femur nailing or hemiarthroplasty: A prospective randomised clinical study§ ¨ zkayın a,*, Gu¨venir Okc¸u b, Kemal Aktug˘lu a Nadir O a b

Department of Orthopaedics and Traumatology, University of Ege, Medical School, I˙zmir, Turkey Department of Orthopaedics and Traumatology, University of Celal Bayar, Medical School, Manisa, Turkey

A R T I C L E I N F O

A B S T R A C T

Keywords: Intertrochanteric femur fractures Proximal femoral nail Hemiarthroplasty Image intensifier

Introduction: The purpose of this prospective randomised study is to compare in the elderly the functional results of intertrochanteric femur fractures treated either with closed reduction and internal fixation with proximal femoral nailing or cemented hemiarthroplasty. Materials and methods: The study included patients above the age of 75 who were diagnosed with intertrochanteric femur fracture and admitted to the Department of Orthopaedics and Traumatology, Ege University Hospital between October 2006 and December 2012. After informed consent was obtained from the patients, they were randomised via sealed opaque envelops into two groups. Patients in Group 1 were internally fixated utilizing proximal femoral nail, whilst the patients in Group 2 were treated with a cemented hemiarthroplasty. Complications were recorded and functional results were evaluated using the Harris Hip score. The mean time of follow up was 31.72 months (min. 18–max. 47, std. dev. 10.68). Results: A total of 54 patients were included in the study. 21 of them (38.9%) received a proximal femoral nail whilst 33 (61.1%) were treated with hemiarthroplasty. Average age of the patients was 82.24 (min. 75–max. 97). Average age in Group 1 was 79.57 (min. 75–max. 91), whilst it was 83.94 in Group 2 (min. 75–max. 97). Harris Hip score analysis revealed that the difference between the patients treated with hemiarthroplasty and proximal femoral nailing was statistically significant in favour of the hemiarthroplasty group within the first 3 months. However, this difference diminished at the 6th month time point, and even reversed as of the 12th month postoperatively. Discussion and conclusions: Although cases with hemiarthroplasty achieved a better level of activity in the beginning, cases with proximal femoral nailing reached a comparable level of activity within a short period of time, faster than those treated with hemiarthroplasty, displaying a better level of activity in the end. ß 2015 Elsevier Ltd. All rights reserved.

Introduction

§ The authors of this manuscript do not have financial or proprietary interest in the subject matter or materials discussed in the manuscript, including (but not limited to) employment, consultancies, stock ownership, honoraria, and paid expert testimony. No funds were received in support of this study. The legal and regulatory status of the device that is the subject of this manuscript is not known by the authors. This study is supported by Ege University Scientific Research Project Commission. * Corresponding author at: Zafer Cad., No:15/11, 35040 Bornova-Izmir, Turkey. Tel.: +90 532 367 4022; fax: +90 232 388 1357. ¨ zkayın). E-mail address: [email protected] (N. O

Proximal femur fractures are frequently observed in the elderly population [1]. The incidence of these fractures increase with age. After the age of 50 years, the incidence rate doubles each decade. The incidence is two to three times higher in women than in men [2,3]. The presence of osteoporosis is an important cofounding parameter in the occurrence of these fractures where almost 90% of these injuires in the geriatric population are sustained by following a simple fall [4]. Hip fractures are grouped under three categories based on the anatomical location of the injury being intracapsular, intertrochanteric, and subthrochanteric femur fractures [5]. Intracapsular

http://dx.doi.org/10.1016/j.injury.2015.05.024 0020–1383/ß 2015 Elsevier Ltd. All rights reserved.

¨ zkayın N, et al. Intertrochanteric femur fractures in the elderly treated with either proximal femur Please cite this article in press as: O nailing or hemiarthroplasty: A prospective randomised clinical study. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.024

G Model

JINJ-6214; No. of Pages 6 N. O¨zkayın et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx

2

fractures make 40% of the total number of proximal femur fractures with non-union being the most common complication reported [6]. Intertrochanteric fractures represent nearly 50% of proximal femur fractures with malunion and shortening due to poor bone quality being the most prominent complications observed [6,7]. Mortality rates for intertrochantric femur fractures are reported between 15 and 20%. These fractures are generally observed above the age of 70. Selection of the implant type is vitally important as it affects the survival, and functional results of the patient. Yet, a review of literature proves the lack of consensus regarding this issue [8]. Cemented hemiartroplasty and/or internal fixation with proximal femur nailing after closed reduction currently represents the mainstay of treatment of intertrochanteric fractures [9]. Implant selection remains to be a conflicting issue in the surgical treatment of these fractures. Intramedullary fixation with minimal invasive surgical method is considered more appropriate for geriatric patients. Closed reduction protects the fracture hematoma. Intramedullary fixation reduces complications occurring in relation to surgical trauma, blood loss, infection and injury location [8]. Gotze et al. [10] did a comparison of osteosynthesis between unstable sub and pertrochanteric femur fractures in an experimental study using the proximal femur nail and found out that this device was the most load bearing implant among the other implants tested [10]. However, cemented hemiarthroplasty provides the advantage of load bearing ambulation in the early post-operative period [11]. The purpose of this prospective randomised study was to compare the results of treatment of intertrochanteric femur fractures managed either with internal fixation (IM nailing after closed reduction) or with cemented hemiarthroplasty.

Materials and methods The research included those patients at the age of 75 or above with intertrochanteric femur fractures who were admitted to Ege University, Department of Orthopedics between October 2006 and December 2012. A statement of approval dated 11.16.2006 and numbered 06-4.1/5 was obtained for the study from the Research Ethics Committee of Medical Faculty, Ege University, Bornova, I˙zmir, Turkey. Exclusion criteria were patients who did not want to be included in the study and patients with pathological fractures. Sealed opaque equivalent envelopes were used for randomisation of patients. Patients in Group 1 were assigned to internal fixation and were treated with proximal femur nailing whereas patients in Group 2 were treated with hemiarthroplasty. Operations were performed ¨ ). Patients were given either only by two surgeons (K.A and N.O general or regional block anaesthesia. Standard technique was used for the proximal femorla nailing. The patient was positioned on the fracture table. The fracture was reduced by closed reduction. All patients were administered prophylactic antibiotics (1gr second generation of cephalosporins) and low molecular heparin for 4 weeks as a means of thromboprophylaxis. All patients were mobilised full weight bearing within 48 h after surgery. Perioperative complications were recorded and analysed. After hospital discharge, patients were followed up in the outpatient clinics for clinical and radiological evaluation at 1.5, 3, 6, 12, 18 months. The Harris Hip score was used for functional evaluation during examination. Statistical analyses included T-Test, Fisher’s Exact Test, ChiSquare Test, Shapiro-Wilk Test, Independent Samples T Test and Paired Samples T Test. p < 0.005 value was accepted statistically significant.

Results A total number of 54 patients participated in the study. 21 of them (38.9%) were randomised in Group 1 (treated with proximal femoral nail) and 33 (61.1%) in Group 2 (treated with cemented hemiarthroplasty). Average age of all patients was 82.24 (min. 75–max. 97, std. dev. 5.298). Average age was 79.57 in Group 1 (min. 75–max. 91, std. dev. 4.833), and 83.94 (min. 75–max. 97, std. dev. 4.924) in Group 2. The difference between the two groups was statistically significant with respect to average age (T-Test, p = 0.002). Although the difference is statistically significant we consider it clinically disregardable (Table 1). Average follow-up period of all patients was 31.72 months (min. 18–max. 47, std. dev. 10.68) where such time was 32.33 months (min. 19–max. 47, std. dev. 10.97) in Group 1, and 31.33 months (min. 18–max. 47, std. dev. 10.65) in Group 2. The difference between the two groups was not statistically significant with respect to follow-up period (T-Test, p = 0.741), (Table 1). Of all the patients, 34 (63%) suffered a right and 20 a left hip fracture respectively. The difference between the two groups was not statistically significant with respect to the side localisation of the hip fracture (Fisher’s Exact Test, p = 0.775) (Table 1). Of all the patients 35 were (64.8%) female, 19 were (35.2%) male. The difference between the two groups was not statistically significant with respect to sex (Fisher’s Exact Test, p = 0.392) (Table 1). According to AO classification, thee were 8 (14.8%) A1, 23 (42.6%) A2, 23 (42.6%) A3. The difference between the two groups was not statistically significant with respect to AO fracture classification. (Chi-Square Test, p = 0.519), (Table 1). 53 (98.1%) of all patients were given regional anaesthesia, 1 (1.9%) was given general anaesthesia. The difference between the two groups was not statistically significant with respect to the type of anaesthesia administered. (Fisher’s Exact Test, p = 0.389), (Table 1). No statistical difference was also observed between the two groups with regard to the operating surgeon (Fisher’s Exact Test, p = 1.000), (Table 1). Average operative time length in all patients was 42.94 min (min. 8–max. 75, std. dev. 16.183). Whilst in Group 1, average operative time was 28.19 min (min. 8–max. 50, std. dev. 12.044), in Group 2 it was 52.33 min (min. 30–max. 75, std. dev. 10.457). The difference between the two groups was statistically significant with respect to operative time (T-Test, p < 0.001). Neither in Group 1 nor in Group 2, any femur shaft fracture occurred during implant application in any one of the patients. No problem was encountered during cement application in Group 2 as well. Average post-operative hospitalisation time of all patients was 6.02 days (min. 1–max. 15, std. dev. 2.858). Whilst it was 6.76 days (min. 2–max. 15, std. dev. 3.846) in Group 1, it was 5.55 days in Group 2 (min. 1–max. 11, std. dev. 1.922). The difference between the two groups with respect to post-operative hospitalisation time was not statistically significant (T-Test, p = 0.190). For all patients, the time length from admission to operating theatre was 7.61 days (min. 1–max. 30, std. dev. 5.413). The distribution included 7.81 days in Group 1 (min. 2–max. 30, std. dev. 6.462), and 7.48 days (min. 1–max. 20, std. dev. 4.731) in Group 2. The difference between the two groups was not statistically significant (T-Test, p = 0.832), (Table 1). Among all the patients, one superficial infection was documented (1.9%) in group 2. This difference was not statistically significant (Fisher’s Exact Test, p = 1.000). Late deep infection was observed only in 1 (1.9%) of all patients in Group 1. The difference was not statistically significant between the two groups (Fisher’s Exact Test, p = 0.389).

¨ zkayın N, et al. Intertrochanteric femur fractures in the elderly treated with either proximal femur Please cite this article in press as: O nailing or hemiarthroplasty: A prospective randomised clinical study. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.024

G Model

JINJ-6214; No. of Pages 6 N. O¨zkayın et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx

3

Table 1 Statistical matching between the two Groups was inspected. Group 1 (N: of cases: 21)

Group 2 (N: of cases: 33)

p value

83.94  4.924 31.33  10.65 20

0.0021 0.7411

Right

79.57  4.833 32.33  10.97 14

Left

7

13

Female

12

23

Male

9

10

A1

4

4

A2 A3

7 10

16 13

Regional

20

33

General

1

0

KA

3

5

¨ NO

18

28

7.81  6.462 3

7.48  4.731 8

2

Age (year) Follow-up period (month)2 Side distribution3

0.7754 Sex distribution3

0.3924 Fracture classification3

0.5195

Type of anaesthesia administered3

0.3894 Operating surgeon3

1.0004 Pre-operative hospitalisation time2 Support use before fracture3

0.8321 0.4974

1

T-Test. Value indicates the arithmetic average and standard deviation. 3 Value indicates the number of cases. 4 Fisher’s Exact Test. 5 Chi-Square Test. 2

Nonunion was not observed in patients in Group 1. The average fracture union time was 11.9 weeks (min. 6–max. 18, std. dev. 2.862). No prothesis dislocation was detected in any of the patients in Group 2. In none of the patients in Group 2, any loosing was detected in the femoral stem of the prosthesis at the final follow up. No cut-out was identified in any of the patients in Group 1. Shortening was detected in 1 (1.9%) of the patients postoperatively. Whilst shortening was detected in 1 (4.8%) patient in Group 1, no shortening was detected in any of the patients in Group 2. The difference between the two groups was not statistically significant (Fisher’s Exact Test, p = 0.389). Except for one patient in Group 1 who was diagnosed with deep infection, none of the patients developed any complication, which would require reoperation. By the time that the patient was diagnosed with infection, union of the fracture had occurred. The implant was removed and the infection settled down with antibiotic therapy. The difference between the two groups was not statistically significant with respect to reoperation (Fisher’s Exact Test, p = 0.389). Whilst 11 of all patients (20.4%) were able to walk with support before fracture, the distribution included three (14.3%) in Group 1 and eight (24.2%) in Group 2. The difference between the two groups was not statistically significant (Fisher’s Exact Test, p = 0.497) (Table 1). Fig. 1 and Table 2 show Group 1 Average, Group 2 Average and Total Average of patients on the basis of Harris Hip score during post-operative evaluations at 1.5, 3, 6, 12 and 18 months. At 1.5 months, Harris Hip score average was 33.52 in Group 1 whereas it was 48.81 in Group 2. The difference was statistically significant (Independent Samples T Test, p < 0.001) (Table 2). At 3 months, Harris Hip score average was 45.24 in Group 1 and 63.38 in Group 2 respectively. The difference was statistically significant (Independent Samples T Test, p < 0.001), (Table 2). At 6 months, Harris Hip score average was 68.14 in Group 1 and 66.91 in Group 2. The difference was not statistically significant (Independent Samples T Test, p = 0.604), (Table 2).

At 12 months, Harris Hip score average was 75.95 in Group 1 and 68.44 in Group 2. The difference was statistically significant (Independent Samples T Test, p < 0.001), (Table 2). Finally, at 18 months, Harris Hip score average was 77.57 in Group 1 whereas it was 68.63 in Group 2. The difference was statistically significant (Independent Samples T Test, p < 0.001), (Table 2). Overall, whilst the difference between the cases with hemiarthroplasty and the cases with proximal femoral nailing was statistically significant in favour of those with hemiarthroplasty until 6 months, after that time point, this difference became reverse in favour of those treated with proximal femoral nailing (Fig. 1). With regard to the Harris Hip score, in Group 1 the average score was 33.52 at 1.5 months, 45.24 at 3 months, the difference being statistically significant (Paired Samples T Test, p < 0.001). The difference between 6 months and 12 months was statistically significant (Paired Samples T Test, p < 0.001) as well as the difference between 12 months and 18 months (Paired Samples T Test, p < 0.001). The difference between 1.5 months and 12 months was also statistically significant (Paired Samples T Test, p < 0.001). In other words, the functional results of Group 1 indicated positive recovery of patients, which was statistically significant at each inspection date when compared to the former date of inspection (Table 2, Fig. 1). In Group 2 the average Harris Hip score was 48.81 at 1.5 months, 63.38 at 3 months, the difference being statistically significant (Paired Samples T Test, p < 0.001). The difference between 6 months and 12 months was also statistically significant (Paired Samples T Test, p < 0.001). However, the difference between 12 months and 18 months was not statistically significant (Paired Samples T Test, p = 0.607). The difference between 1.5 months and 12 months was also statistically significant (Paired Samples T Test, p < 0.001). In other words, the functional results of Group 2 indicated positive recovery of patients, which was statistically significant at each inspection date when compared to the former date of inspection until the 12 months point where the increase noted previously was ceased (Table 2, Fig. 1).

¨ zkayın N, et al. Intertrochanteric femur fractures in the elderly treated with either proximal femur Please cite this article in press as: O nailing or hemiarthroplasty: A prospective randomised clinical study. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.024

G Model

JINJ-6214; No. of Pages 6 N. O¨zkayın et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx

4

Fig. 1. Schematic drawing of the increase rate of Functional Harris Hip scores of both groups at inspection dates.

The increase rate of the functional scores was investigated with respect to the groups (Table 3, Fig. 1). For this reason, the differences between the average Harris Hip scores were averaged. Although in Group 1, the average of differences was 11.7143 between 1.5 and 3 months, in Group 2 it was 14.5625. This difference was not statistically significant (Independent Samples T Test, p = 0.072). In other words, whilst in the first 1.5–3 months period there was a significant difference between the functional score averages of the two groups in favour of Group 2, no statistical difference was

Table 2 Distribution of Harris Hip score averages for Group 1, Group 2 and total patients at different inspection periods. Average

Min.

Max.

Std. Dev.

1.5

Group 1 Group 2 Total

33.52 48.81 42.75

20 29 20

48 67 67

7.521 9.00 11.464

3

Group 1 Group 2 Total

45.24 63.38 56.19

30 31 30

62 85 85

8.859 12.471 14.252

6

Group 1 Group 2 Total

68.14 66.91 67.40

56 48 48

80 83 83

6.544 10.672 9.206

12

Group 1 Group 2 Total

75.95 68.44 71.42

59 50 50

90 86 90

7.324 10.549 10.037

18

Group 1 Group 2 Total

77.57 68.63 72.17

60 49 49

92 87 92

7.724 10.219 10.233

Month

detected with respect to increase rate. However, the difference in the increase with respect to 3–6 months was statistically significant between the two groups (Independent Samples T Test, p < 0.001), statistically significant between 6 and 12 months (Independent Samples T Test, p < 0.001), statistically significant between 12 and 18 months (Independent Samples T Test, p < 0.001), statistically significant between 1.5 and 18 months in favour of Group 1 (Independent Samples T Test, p < 0.001). Whilst no difference was detected with respect to recovery increase rate between the two groups in the first 1.5–3 months period, in the periods to follow the recovery increase rate in Group 1 was higher than that of Group 2. The increase rate in Group 1 was maximum between 3 and 6 months, then between 1.5 and 3 months, then 6–12 months, and then 12–18 months. In Group 2 the increase rate was maximum between 1.5 and 3 months, then 3–6 months, then 6–12 months, and then between 12 and 18 months, however the increase rate between 6–12 months and 12–18 months were found statistically insignificant in this group (Table 3, Fig. 1). Discussion Intertrochanteric fractures represent the most common geriatric fracture and remain a topic of vivid discussion for the trauma surgeon [12–17]. The issue of implant selection for unstable intertrochanteric fractures is still under discussion. Following treatment with dynamic hip screw there is high prevelance of insufficient functionality, unacceptable shortening and external rotation deformity in osteoporised geriatric patients [7,18–21]. Some surgeons advise arthroplasty or intramedullary hip nailing in

¨ zkayın N, et al. Intertrochanteric femur fractures in the elderly treated with either proximal femur Please cite this article in press as: O nailing or hemiarthroplasty: A prospective randomised clinical study. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.024

G Model

JINJ-6214; No. of Pages 6 N. O¨zkayın et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx

5

Table 3 Statistical research on the increase rate of functional scores on the basis of groups. Month

Group

Average of Harris Hip score average differences

1.5–3

1

11.7143

2

Min.

Max.

Std. Dev.

4.00

19.00

3.67618

14.5625

2.00

28.00

7.49597

1

22.9048

8.00

34.00

6.91307

2

3.5312

3.00

18.00

3.80987

1

7.8095

2.00

18.00

3.80288

2

1.5312

2.00

6.00

2.15503

1

1.6190

3.00

10.00

2.43877

2

0.1875

4.00

4.00

2.03894

p =0.072

3–6

Intertrochanteric femur fractures in the elderly treated with either proximal femur nailing or hemiarthroplasty: A prospective randomised clinical study.

The purpose of this prospective randomised study is to compare in the elderly the functional results of intertrochanteric femur fractures treated eith...
400KB Sizes 4 Downloads 37 Views