Femoral neck fractures after removal of hardware in healed trochanteric fractures

Injury, Int. J. Care Injured 48 (2017) 2619–2624

Contents lists available at ScienceDirect

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

Review

Femoral neck fractures after removal of hardware in healed trochanteric fractures Antonio Barqueta,* , Peter V. Giannoudisb,c,**, Andrés Gelinkd a

Department of Traumatology, Asociación Española Primera de Socorros Mutuos, Bvar. Artigas y Palmar, Montevideo, Uruguay Academic Department of Trauma & Orthopaedic Surgery, University of Leeds, Clarendon Wing, Floor D, Great George Street, Leeds General Infirmary, LS1 3EX Leeds, UK c NIHR Leeds Biomedical Research Center, Chapel Allerton Hospital, Leeds, UK d University Clinic of Traumatology and Orthopaedics, UDELAR, Las Heras 2085, Montevideo, Uruguay b

A R T I C L E I N F O

Keywords: Trochanteric fracture Femoral neck fracture Spontaneous fracture Internal fixation Hardware removal Hip arthroplasty

A B S T R A C T

Introduction: Hardware removal in healed trochanteric fractures (TF) in the absence of infection or significant mechanical complications is rarely indicated. However, in patients with persistent pain, prominent material and discomfort in the activities of daily living, the implant is eventually removed. Publications of ipsilateral femoral neck fracture after removal of implants from healed trochanteric fractures (FNFARIHTF) just because of pain or discomfort are rare. The purpose of this systematic review of the literature is to report on the eventual risk factors, the mechanisms, the clinical presentation, and frequency, and to pay special emphasis in their prevention. Materials and methods: A comprehensive review of the literature was undertaken using the PRISMA guidelines with no language restriction. Case reports of FNFARIHTF and series of TF with cases of FNFARIHTF due to pain or discomfort published between inception of journals to December 2016 were eligible for inclusion. Relevant information was divided in two parts. Part I included the analysis of cases of FNFARIHTF, with the objective of establishing the eventual risk factors, mechanisms and pathoanatomy, clinical presentation and diagnosis, treatment and prevention. Part II analyzed series of TF which included cases of FNFARIHTF for assessing the incidence of femoral neck fractures in this condition. Results: Overall 24 publications with 45 cases of FNFARIHTF met the inclusion criteria. We found that the only prevalent factors for FNFARIHTF were: 1) preexisisting systemic osteoporosis, as most patients were older and elder females, with lower bone mineral density and bone mass; 2) local osteoporosis as a result of preloading by the fixation device in the femoral neck, leading to stress protection, reducing the strain at the neck, and increasing bone loss and weakness; and 3) the removal of hardware from the femoral neck, with reduction of the failure strength of the neck. The femoral neck fractures were spontaneous, i.e. not related to trauma or fall, in 87.5% of the cases, mostly subcapital, and with no prevalence between displaced and undisplaced fractures. The clinical presentation was that of a spontaneous fracture, and most of the patients consulted because of hip pain and presented in the emergency room walking by themselves which led to delayed diagnosis in several instances. Radiological diagnosis was mostly with radiographs, though in some cases CT scans or MRI were necessary. The overall median incidence of this complication was 14.5% after hardware removal because of pain or discomfort in healed trochanteric fractures. Conclusion: The risk factors for FNFARIHTF seem to be preexisisting systemic osteoporosis, local osteoporosis as a result of preloading by the fixation device in the femoral neck, and the removal of hardware from the femoral neck, with reduction of the strength of the neck. The clinical presentation may be obscure as most of the patients complain of hip pain of some days or weeks, and arrive in the hospital walking. Therefore, the attending physician should be alert in order to request the appropriate radiological investigation and if this is not clear, CT scan or MRI should be done in order to diagnose promptly these “spontaneous” fractures. Treatment should be replacement surgery in most cases; however, there is some place for internal fixation especially in undisplaced fractures or younger patients.

* Corresponding author at: Department of Traumatology, Asociación Española Primera de Socorros Mutuos, Bvar. Artigas y Palmar, Montevideo, Uruguay ** Corresponding author at: Academic Department of Trauma & Orthopaedic Surgery, University of Leeds, Clarendon Wing, Floor D, Great George Street, Leeds General Infirmary, LS1 3EX Leeds, UK. E-mail addresses: [email protected] (A. Barquet), [email protected] (P.V. Giannoudis). http://dx.doi.org/10.1016/j.injury.2017.11.031 0020-1383/© 2017 Published by Elsevier Ltd.

2620

A. Barquet et al. / Injury, Int. J. Care Injured 48 (2017) 2619–2624

The occurrence of the femoral neck fracture after hardware removal may be prevented with reosteosynthesis and the use of bone chips or bone substitutes. Finally, the relatively high incidence of this complication should alert orthopaedic surgeons to reduce the removal of hardware in healed trochanteric fractures to very selected cases. © 2017 Published by Elsevier Ltd.

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . Material and methods . . . . . . . . . . . . . . . Criteria for eligibility . . . . . . . . . . . . . Data extraction . . . . . . . . . . . . . . . . . Statistical analysis . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature search . . . . . . . . . . . . . . . . Part 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . Potential risk factors . . . . . . . . . . . . . Femoral neck fracture characteristics Part 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . Conflict of interest . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

Introduction Trochanteric fractures (TF) are a common cause of morbidity and mortality in the elderly. Internal fixation is the treatment of choice, either by intramedullary or extramedullary devices [1–3]. There are formal indications for implant removal in cases of trochanteric nonunion, infection or significant mechanical complications – cutout, implant breakage, etc. Although there is no consensus whether osteosynthesis should be removed or not from a healed trochanteric fracture in the absence of those complications [4], generally the implant in these cases is not removed [5–8]. However, in symptomatic patients with pain, prominent material, skin irritation, and discomfort in the activities of daily living, hardware is eventually removed [2,6–8], even though it not only entails the risks of surgical intervention, especially in older and elderly patients [1], but also because the improvement of symptoms after hardware removal is unpredictable [4], or moderate at best [5]. Reports of ipsilateral femoral neck fracture after removal of implants from healed trochanteric fractures (FNFARIHTF) just for pain or discomfort are rare [8]. The purpose of this systematic review of the literature is to investigate the eventual risk factors, the mechanism and pathoanatomy of the femoral neck fracture, the diagnostic and therapeutic modalities, the outcomes and recommendations for its prevention, and finally to investigate the frequency of this complication. Material and methods This review was conducted in accordance to the PRISMA guidelines [9]. Data were documented according to a standardized protocol, where objectives and inclusion criteria were specified in detail. Searches were conducted using the following databases: Cochrane Library, Pubmed, Embase, Springer, OvidSP, Science Direct, Dialnet, J-Stage, Scielo, and KoreaMed, and also the Google Scholar searcher. The following keywords were used: “femoral neck fracture”, “trochanteric fracture”, and “hardware removal”. Two reviewers selected potentially relevant abstracts and obtained

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

2620 2620 2620 2620 2620 2622 2622 2622 2622 2622 2623 2623 2624 2624

full copies of the articles. Additionally, all references of the retrieved articles were also reviewed. Criteria for eligibility Studies selected were original clinical articles that addressed FNFARIHTF in adult patients – 18 year-old and older–, with no language restriction. Cases with fractures in pathological bone, trochanteric nonunion, or infectious or significant mechanical complications – i.e. cutout, implant breakage, etc. – were excluded. Date limits were set from inception of journals to December 2016. Data extraction Relevant information obtained was divided in two parts. Part I of the study included detailed case reports or cases of FNFARIHTF with useful information, and data were extracted as type of study, age, gender, comorbidities, AO-OTA group of trochanteric fracture [10], fracture reduction, fracture treatment (implant used), position of the implant in the femoral neck and head, postoperative complications, time from internal fixation of trochanteric fracture to implant removal, partial (cervical component) or total hardware removal, time from implant removal to femoral neck fracture, mechanism and severity of injury, site of femoral neck fracture, fracture displacement, symptoms and signs, diagnostic modalities, treatment, and outcome. Part II of the study included series of healed TF with further removal of hardware which included a case or cases of FNF after removal of hardware, with data extracted as type of study, time period of study, number of patients with healed TF, time to followup, and number of cases of FNF, for assessing the incidence of this complication. Statistical analysis As the majority of the data collected were from case reports and few case series statistical analysis was not possible. Descriptive statistics were employed where possible.

Table 1 Femoral neck fractures after removal of hardware in healed trochanteric fractures. A

B

C

D

E

Hunter & Mehta Williams & Parker Mendez et al Mendez et al Mendez et al Mendez et al Taylor et al Barquet et al Buciuto et al Buciuto et al Buciuto et al Buciuto et al Buciuto et al Buciuto et al Buciuto et al Suzuki et al Kukla et al Kukla et al Kukla et al Seibert et al Abe et al Abe et al Hesse & Gachter Hesse & Gachter Hesse & Gachter Hesse & Gachter Lopreite et al Segata et al Yang et al Yang et al George et al Hernandez et al Seibert et al Kim et al Bartonicek Shaer et al El Ibrahimi et al Legnani et al Yoon et al Yoon et al Yoon et al Yoon et al Yoon et al Yoon et al Kato et al

1977 1992 1993 1993 1993 1993 1996 1997 1997 1997 1997 1997 1997 1997 1997 2000 2001 2001 2001 2001 2004 2004 2004 2004 2004 2004 2004 2005 2005 2005 2007 2009 2009 2010 2011 2012 2013 2013 2013 2013 2013 2013 2013 2013 2016

71 83 59 85 50 74 72 71 75 81 81 82 63 79 83 79 64 67 76 74 61 71

F F F F M F F F M F F F F M F F F F M F M M

76 90 64 80 73 82 85 74 60 57 75 67 71 71 71 71 71 71 80

F F F F F F F M F M M F F F F F F F F

A2 A2 A1 A A A A1 A2 A2 A2 A2 A2 A2 A2 A2 A2 A2 A2 A2 A2 A1 A1 A A A A A2 A1 A1 A2 A A2 A2 A1 A1 A1 A2 A1 AU AU AU AU AU AU A2

F

d

ao

G

H

I

SHS Gamma nail SHS SHS SHS SHS SHS Gamma nail SHS SHS SHS Blade plate Blade plate Blade plate Blade plate SHS Gamma nail Gamma nail Gamma nail PFN SHS SHS Gamma nail Gamma nail Gamma nail Gamma nail SHS SHS AP Gamma nail A PFN SHS PFN PFNA blade DYna nail Gamma nail CMN Gamma nail Bi-Nail SHS SHS SHS SHS SHS SHS CMN

G G G

CSch CSch LSch LSch LSch LSch S CSch S S CSch CSch S S S i/o c9

G G Vl G Vl G Vl G Vr G G

J

G G Vl G

SCh S S CSch

p/o c9

G G G G G G G G G Vr G G G

CSch CSch S S L Sch CSch CSch CSch CSch CSch CSch L L L L L L CSch

sm an p/o c9

G

G

an

#

K

L

M

N

O

P

Q

R

S

8 6 21 19 56 8 8 51 24 12 38 26 24 13 14 4 8 13 35 12 12 24

Hip screw & plate Hip screw Hip screw & plate Hip screw & plate Hip screw & plate Hip screw & plate Hip screw & plate Hip screw & nail Hip screw & plate Hip screw & plate Hip screw & plate Blade & plate Blade & plate Blade & plate Blade & plate Hip screw & plate Hip screw & nail Hip screw & nail Hip screw & nail Hip screws Hip screw & plate Hip screw & plate Hip screw & nail Hip screw & nail Hip screw & nail Hip screw & nail Hip screw & plate Hip screw & plate Hip screw & nail Hip screws Hip screw & plate Hip screws & nail Hip blade Hip screw & nail Hip screw & nail Hip screw Hip screw Hip screw & nail Hip screw & plate Hip screw & plate Hip screw & plate Hip screw & plate Hip screw & plate Hip screw & plate Hip screw

120 21 17 17 37 8 21 28 7 14 60 10 7 30 7 8 22 29 36 21 18 21

Sp Sp Sp Sp Sp Sp M Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp M Sp M M M M Sp Sp Sp Sp Sp Sp Sp Sp – Sp Sp Sp Sp Sp

Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Mc Mc Sc Sc Sc

Undisplaced Undisplaced Undisplaced Undisplaced Undisplaced Displaced Displaced Displaced – – Displaced – – displaced – Displaced – – – Displaced Displaced Undisplaced Displaced – – – Displaced Undisplaced Undisplaced Displaced Displaced Displaced Displaced Undisplaced Displaced Displaced Displaced Undisplaced displaced Displaced Undisplaced Undisplaced Undisplaced Undisplaced

Pain Pain Pain Pain Pain Pain Pain Pain Pain Pain Pain Pain Pain Pain Pain Pain – – – – Pain Pain

– Pain Pain Pain Pain Pain

Tg Xr Bs Bs Bs Xr Xr Xr Xr Xr Xr Xr Xr Xr Xr Xr/MRI/Bs Xr Xr Xr Xr Xr Xr Xr Xr Xr Xr Xr Xr MRI Xr Xr Xr Xr Xr/CT Xr Xr Xr CT Xr Xr

Pain

Xr

PHR PHR Restricted w-b Restricted w-b Restricted w-b BHA PHR THR THR THR THR Screws Screws Conservative THR PHR – – – THR PHR Conservative THR THR THR DHS THR BHA ORIF 3 screws Conservative PHR PHR BHA BHA – PHR THR Conservative BHA BHA ORIF 3 screws ORIF 3 screws ¥ Conservative Conservative BHA

7 6 16 18 60 18 15 6 18 22 16 20 19 19 19 19 19 19 20

120 21 14 14 28 14 14 60 49 7 3 9 7

2

Sp

Pain Pain Pain Pain Pain Pain Pain

T

9

17

7

– 30 – 6

U U – U U U U – U – – – – – – – – – – – – L

U – U – – – – – U – U

A. Barquet et al. / Injury, Int. J. Care Injured 48 (2017) 2619–2624

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

FD

(A) author; (B) year; (C) age; (D) sex; (E) fracture type: AO: A1 = pertrochanteric simple; A2 = pertrochanteric multifragmentary; A3 = intertrochanteric; A = trochanteric fracture – unespecified; AU = trochanteric fracture unstable; (F) associated morbidities: a = alcoholism; o = osteomalacia; d = diabetes; (G) fracture reduction (femoral neck-shaft angle): G = good; Vl = valgus; Vr = varus; (H) fracture treatment: SHS = sliding hip screw; S-Nail-Plate = sliding nail plate; PFNA = proximal femoral nail antirotation; AP = Asia Pacific; A = Asian; CMN: cephalomedullary nail unespecified; i/o c9: intraoperative cut through; (I) cervical implant: S = short; CSch = central & subchondral; H = high; L = low; A = anterior; P = posterior; i/o c9 = intraoperative cut through; (J) fracture complications: an = avascular necrosis femoral head; # = lag screw backed out 8 mm; p/o c9 = postoperative cut through; sm = superior migration; (K) time from internal fixation of trochanteric fracture to implant removal in months; (L) hardware removal; (M) time from implant removal to femoral neck fracture in days; (N) mechanism of femoral neck fracture: Sp = spontaneous; M = mild; (O) level of femoral neck fracture: Sc = subcapital; Mc = midcervical; (P) displacement femoral neck fracture; (Q) symptoms and signs; (R) diagnostic modalities: Xr = radiographs; BS = bone scan Tg = tomogram; CT = computed tomography scan; MRI = magnetic resonance imaging; (S) treatment modalities: PHP = partial hip replacement; THP = total hip replacement; BHA = bipolar hip arthroplasty; ¥ = augmented with bone graft substitute; (T) time to follow-up in months; (U) outcome: U = uneventful; L = limp; FD = fracture displacement. 2621

2622

A. Barquet et al. / Injury, Int. J. Care Injured 48 (2017) 2619–2624

Table 2 Frequency of femoral neck fractures in healed trochanteric fractures after hardware removal. Author

Year

Time period

Fracture

Implant

N of fx

N of fx with IR

N of FNF

Percentage

Follow-up (months)

Buciuto et al Abe et al Yang et al Yoon et al

1997 2004 2005 2013

1991–1994 1993–2002 1993–2003 2001–2011

TF TF TF TF

SHS & Fixed angle blade plate SHS AP Gamma Nail & A PFN SHS

233 303 396 598

20 13 14 67 114

7 2 2 6 17

35 15 14 9 m = 14.5 – av = 14.9

12–36 ? ? 29 (12–100)

TF = trochanteric fracture. Implant: similar meaning than in Table 1. N of fx = number of fractures. Na of fx with IR = number of fractures with implant removal. N of FNF = number of femoral neck fractures.

After omitting studies that did not fulfill the inclusion criteria or were repeated in the databases search, there remained 24 references for inclusion in the analysis [11–34]. All studies were retrospective. Nineteen articles [11,13–15,17–20,22–27,30–34] and one book chapter [29] were selected for 30 detailed case/s reports. Four series of fractures of the trochanteric region with 15 cases of FNFARIHTF with detailed information were also selected [12,16,21,28]. Therefore 24 studies with 45 cases were included in Part 1 of the analysis (Table 1). Seventeen were written in English [11,16,18,19,21,24–27,29–33], four were in Japanese [17,20,23,34], two were in Spanish [15,22] and one was written in Korean [28]. For assessing part II of the study we analyzed articles with information of the incidence of FNFRIHTF: four series that had been analyzed for their detailed report of cases in Part I [16,20,24,33] were used for this purpose (Table 2). Additional articles were also reviewed to facilitate the development of the discussion [35–38].

PFN in 1, PFNA blade in 1, DYna nail in 1, Bi-Nail in 1, and an unespecified cephalomedullary nail in 2. Position of the implant in the femoral neck and head. A total of 39 cases had information about the position of the cervical component of the implant used. When implants of one lag screw were used, the lag screw was central and subchondral in 15 cases (38.46%), low and subchondral in 4 (10.25%), short in 10 (25.64%), and undergoing intra-operative cut-through in 1 (2.56%), and when a double lag screw was used in the cases of two proximal femoral nails, the screws were subchondral in both (5.12%). Post-operative complications. There were post-operative complications in 4 of these collected cases, i.e. 8.88%. These were: avascular necrosis of the femoral head in 1, superior migration of the lag screw plus avascular necrosis of the femoral head in 1, post-operative cut-through in 1, and 8 mm back out of the lag screw in 1. Time from internal fixation of trochanteric fracture to implant removal. Forty-one cases reported this interval, which ranged between 4 and 60 months, with a mean of 18 months. Partial (cervical component) or total hardware removal. Among all 45 cases with information, there was total removal of the implant in 38, while in 7 cases only the lag screw or blade were removed in cephalomedullary nails.

Part 1

Femoral neck fracture characteristics

Potential risk factors

Time from implant removal to femoral neck fracture. The mean of this interval, reported in 36 cases, was 18 days, with a range of between 2 and 120 days. Severity of causal trauma. A total of 40 cases reported the severity of the causal trauma of the femoral neck fracture. The fracture was spontaneous, i.e. with no trauma, in 35 cases (87.5%), and mild in 5 (12.5%). There were no cases resulting of severe trauma. Site of femoral neck fracture. In the 45 cases with reported site of the fracture or with useful illustrations, the femoral neck fracture was subcapital in 43 (95.55%) and transcervical in 2 (4.44%). Fracture displacement. A total of 33 had information about fracture displacement. These were undisplaced in 14 cases (42.42%) and displaced in 19 (57.58%). Symptoms and signs. In 31 cases with information the clinical presentation was pain alerting the physician of the fracture. Twenty-nine of these patients had no fall at all. Diagnosis. Forty-one cases had information regarding the method of diagnosis of the femoral neck fracture, which was diagnosed using radiographs in 33 cases (80.48%), bone scan in 3 (7.31%), tomogram in 1 (2.43%), CT scans in 1 (2.43%), MRI in 1 (2.43%), radiographs and CT scans in 1 (2.43%), and finally there was 1 patient who required radiographs, MRI, and bone scans to achieve diagnosis (2.43%). Treatment. A total of 41 cases reported the treatment of the femoral neck fracture. Total hip replacement was performed in 11

Results Literature search

Age. A total of 41 cases reported the age of the patients. The mean age was 74 years (range, 50–90]. There were 3 patients younger than 60 years (7.3%) with a mean age of 57 years (range, 50–59), and 38 patients were 60 years-old or older (92.7%) with a mean age of 76 years (range, 60–90). Gender. A total of 41 cases reported the gender of the patients. There were 9 males (21.9%) and 36 females (79.1%). Comorbidities. Among 45 patients, 2 patients (4.4%) had alcohol abuse and tobacco use. Trochanteric fracture (AO). There were 10 A1 fractures (simple 2fragment) pertrochanteric fractures) (22.22%), 21 A2 fractures (multifragmentary pertrochanteric) (46.66%), 6 unstable trochanteric fractures (13.33%), leaving 8 cases (17.77%) with unespecified fracture group. Fracture reduction. There were 33 cases with information on this issue. There were 27 cases with a normal angle of reduction, 4 cases with reduction in valgus, i.e. an angle greater than 135 degrees, and 2 cases with reduction in varus, i.e. an angle lower than 125 . Fracture treatment (implant used). All 45 cases had information regarding the treatment for the trochanteric fracture. Treatment was operative in every case. Extramedullary implants were used in 25 cases (55.56%): a blade plate was used in 4 cases, and a sliding hip screw in 21 cases. Intramedullary implants were used in 20 cases (44.44%): Gamma nail in 11, AP Gamma nail in 1, PFN in 2, AP

A. Barquet et al. / Injury, Int. J. Care Injured 48 (2017) 2619–2624

cases, partial hip replacement in 8, bipolar hip arthroplasty in 7, open reduction and internal fixation with screws in 5 and with dynamic hip screw in 1. The management was conservative in the remaining 9 cases. Outcome. The follow-up was between 6 and 30 months in 5 cases, and the outcome was uneventful in each case. Part 2 Incidence. The analysis of the incidence of femoral neck fractures following removal of hardware in healed trochanteric fractures based on the study of 4 series of trochanteric fractures [16,20,24,33], which included 17 cases of femoral neck fractures following hardware removal in 114 healed trochanteric fractures showed a median incidence of femoral neck fractures after hardware removal of 14.5% (range, 9–35), with an average of 14.9%. Discussion All aspects of management of proximal femoral fractures continues to attract a lot of attention from both researchers and clinicians [35–43]. In the herein study we investigated the aetiology and incidence of ipsilateral femoral neck fracture after removal of implants from healed trochanteric fractures. Several potential risk factors have been proposed including: 1) preexisting systemic osteoporosis [16,19,23,24,26,29,31,32] or osteomalacia, resulting from alcohol abuse [30]; 2) local osteoporosis as a result of preloading by the fixation device in the femoral neck [1,12,14–16,19– 21,23,25,27,30,31]; 3) a varus deviation of the neck-shaft angle [30]; 4) size of the cervical component of the implant used [18,33]; 5) an inferior and/or anterior position of the implant in the femoral neck [13,25,33,35]; 6) avascular necrosis of the femoral head [19]; and 7) the removal of hardware in a healed trochanteric fracture [1,13,18,20,23–26,30–33,35]. In order to study these eventual risk factors we analyzed age, gender, comorbidities, trochanteric fracture group, fracture reduction, fracture treatment (implant used), position of the implant in the femoral neck and head, postoperative complications, time from internal fixation of trochanteric fracture to implant removal, and partial (cervical component) or total hardware removal, in the collected cases of this review. Moreover, we reviewed additional articles to facilitate the development of the discussion [44–47]. We found that the only prevalent factors for FNFARIHTF were: 1) preexisisting systemic osteoporosis, as most patients were older and elder females, with lower bone mineral density and bone mass; 2) local osteoporosis as a result of preloading by the fixation device in the femoral neck, leading to stress protection, reducing the strain at the neck, and increasing bone loss and weakness; and 3) the removal of hardware from the femoral neck, with reduction of the failure strength of the neck. We did not find the other factors as risky. Kukla et al found in a clinical and experimental study that the size of the cervical component of the implant is a factor increasing the incidence of FNFARIHTF, with more frequency of femoral neck fracture after removal of a gamma nail than a dynamic hip screw [18]. However, Yang et al in an analysis of biomechanical properties of proximal femur after implant removal found that the mechanical stability of the proximal femur does not differ after the removal of proximal femoral nail antirotation and dynamic hip screw [2]. In our analysis there was slight prevalence of this complication after removal of intramedullary or extramedullary implants, with different sizes of cervical components. Most patients consulted because of hip pain, which had appeared without previous trauma or fall. This is why many authors define these fractures as “spontaneous” [16,25,26]: as patients, with this condition, usually arrive to medical consultation walking, physicians

2623

may and have failed to promptly diagnose the femoral neck fracture. Therefore, every patient who has had removal of the implant used to fix a previous trochanteric fracture and complains of hip pain should open the suspicion of a spontaneous femoral neck fracture, and lead to further imaging diagnostic measures. The usual diagnostic modality were simple radiographs of the hip, which permitted to achieve diagnosis in four out of five of the published cases. However, in the remaining cases several other procedures were necessary were necessary in order to discard the femoral neck fracture: these were bone scans, tomograms, computed tomography scans, magnetic resonance images, or even an association of these methods, mainly in those cases of undisplaced fractures which were very difficult to diagnose. Thus, confronted to the suspicion of the femoral neck fracture in this condition the patient should undergo radiologic examination or eventually CT scans or MRI of his or her aching hip in order to confirm or discard a spontaneous femoral neck fracture. The most frequent treatment was by replacement surgery, probably because most patients were older. Total hip replacement is probably the treatment of choice in these patients. Internal fixation of the femoral neck procedures or augmented osteosynthesis using polymethyl methacrylate cement [36–38] may be advised in younger patients and in those with undisplaced or less displaced fractures. Although the outcomes in the reviewed cases seemed to be satisfactory, this cannot be confirmed because of the short number of cases with follow-up. Although there are indications for hardware removal in healed trochanteric fractures, such as infection, cutout, implant breakage, there are no clear indications for metal removal in other conditions, mainly those related to pain in the hip [16,18,31–33]. However, in few cases the persistent pain or some prominence of the of the cervical component of the implant, with major discomfort may lead the attending surgeon to plan the removal of hardware. When and if hardware removal is decided further femoral neck fracture should be prevented and some ways to do it have been recommended. In the case of hardware prominence the hip screw should be changed with a shorter one [30,31]. The use of bone graft chips [11,24,32] or bone substitution material [18,24], i.e. an injectable bone filler such as calciuim phosphate cement [30,37,38], or polymethyl methacrylate cement augmenting reosteosynthesis has been recommended. Although some authors have promoted a careful rehabilitation program with a significantly long period of protective weight-bearing after hardware removal, up to four months, this has been contradicted by others, who consider that there is no compliance for partial weight bearing is not in older and elders [13]. However, there are no series with enough follow-up to confirm that these methods would prevent the spontaneous femoral neck fracture after hardware removal. There have been very few estimations of the incidence of femoral neck fracture after hardware removal in healed trochanteric fractures in the literature. We found a median incidence of 14.5% and an average of 14.9%. This is a somehow high incidence of this complication, and these numbers might indicate only the lowest incidence, as most patients with trochanteric fractures are older or elder and have a high mortality rate few months after the original fracture and the eventual removal of hardware. This finding is important to confirm the relatively high risk of hardware removal in healed trochanteric fractures, and should lead us to advice the patients of the frequency of this complication. In conclusion, the risk factors for FNFARIHTF seem to be preexisisting systemic osteoporosis, local osteoporosis as a result of preloading by the fixation device in the femoral neck, and the removal of hardware from the femoral neck, with reduction of the failure strength of the neck. The clinical presentation may be obscure as most of the patients complain of hip pain of some days or weeks, and arrive in the hospital walking. Therefore, the

2624

A. Barquet et al. / Injury, Int. J. Care Injured 48 (2017) 2619–2624

attending physician should be alert in order to request radiographs of the affected extremity and if this is not clear further investigations such as CT scan or MRI should be obtained in order to make a prompt diagnosis of these ‘spontaneous’ fractures. Treatment should be replacement surgery in most cases; however, there is some place for internal fixation especially in undisplaced fractures or younger patients. The occurrence of the femoral neck fracture after hardware removal may be prevented with reosteosynthesis and the use of bone chips or bone substitutes. Finally, the relatively high incidence of this complication should alert orthopaedic surgeons in order to reduce the removal of hardware in healed trochanteric fractures to very selected cases. A comprehensive search and analysis of the world literature with no year or language limitations, and the usually complete information of the reviewed information, are some of the strengths of this study. The limitations are mainly the small number of cases previously reported. Further studies on this complication might improve our knowledge in the management of trochanteric fractures. Conflict of interest The authors declare no conflict of interest with regard to the content of this article. References [1] Eberle S, Wutte C, Bauer C, von Oldenburg G, Augat P. Should extramedullary fixations for hip fractures be removed after bone union? Clin Biomech (Bristol, Avon) 2011;26(4):410–4. [2] Yang JH, Jung TG, Honnurappa AR, Cha JM, Ham CH, Kim TY, et al. The analysis of biomechanical properties of proximal femur after implant removal. Appl Bionics Biomech 2016;2016:4987831. [3] Parker MJ, Handoll HH. Gamma and other cephalocondylic intramedullary nails versus extramedullary implants for extracapsular hip fractures in adults. Cochrane Database Syst Rev 2010;9(September):CD000093. [4] Busam ML, Esther RJ, Obremskey WT. Hardware removal: indications and expectations. J Am Acad Orthop Surg 2006;14(2):113–20. [5] Hanson B, van der Werken C, Stengel D. Surgeons’ beliefs and perceptions about removal of orthopaedic implants. BMC Musculoskelet Disord 2008;9 (May):73. [6] Gardenbroek TJ, Segers MJ, Simmermacher RK, Hammacher ER. The proximal femur nail antirotation: an identifiable improvement in the treatment of unstable pertrochanteric fractures? J Trauma 2011;71(1):169–74. [7] Vos DI, Verhofstad MH. Indications for implant removal after fracture healing: a review of the literature. Eur J Trauma Emerg Surg 2013;39(4):327–37. [8] Kovar FM, Strasser E, Jaindl M, Endler G, Oberleitner G. Complications following implant removal in patients with proximal femur fractures – an observational study over 16 years. Orthop Traumatol Surg Res 2015;101 (7):785–9. [9] Shamseer L, Moher D, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. PRISMA-P Group. Preferred reporting items for systematic review and metaanalysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ 2015;349(January):g7647. [10] Marsh JL, Slongo TF, Agel J, Broderick JS, Creevey W, DeCoster TA, et al. Fracture and dislocation classification compendium – 2007: Orthopaedic Trauma Association classification, database and outcomes committee. J Orthop Trauma 2007;21(10 Suppl.):S1–S133. [11] Hunter GA, Mehta A. Subcapital fracture of the hip: a rare complication of intertrochanteric fracture of the femur. Can J Surg 1977;20(2)165–9 72. [12] Williams WW, Parker BC. Complications associated with the use of the gamma nail. Injury 1992;23(5):291–2. [13] Mendez AA, Joseph J, Kaufman EE. Stress fractures of the femoral neck following hardware removal from healed intertrochanteric fractures. Orthopedics 1993;16(7):822–5. [14] Taylor PR, Hepple S, Stanley D. Combination subcapital and intertrochanteric fractures of the femoral neck. Injury 1996;27(1):68–71. [15] Barquet A. Fractura subcapital del cuello femoral luego de fractura trocantérica. A propósito de 2 casos. Paper presented at the Annual Meeting of the Uruguayan Society of Traumatology and Orthopaedics. . [16] Buciuto R, Hammer R, Herder A. Spontaneous subcapital femoral neck fracture after healed trochanteric fracture. Clin Orthop Relat Res 1997; (342):156–63. [17] Suzuki M, Yamada H, Washimi D, Seki T, Henmi O, Morita M. Two cases of subcapital fracture of the femur after healed intertrochanteric fractures operated. Orthop Surg Traumatol 2000;43(2):189–95.

[18] Kukla C, Pichl W, Prokesh R, Jacyniak W, Heinze G, Gatterer R, et al. Femoral neck fracture after removal of the standard gamma interlocking nail: a cadaveric study to determine factors influencing the biomechanical properties of the proximal femur. J Biomech 2001;34(12):1519–26. [19] Seibert FJ, Boldin C, Schippinger G, Szyszkowitz R. Subcapital femoral neck fracture after healing of an intertrochanteric fracture stabilized by an AO-PFN. Eur J Orthop Surg Traumatol 2001;11(2):133–5. [20] Abe M, Sasaki H, Miyamoto T. 2 cases of subcapital fracture of the femur after removal of compression hip screw. Kossetsu 2004;26(2):531–3. [21] Hesse B, Gächter A. Complications following the treatment of trochanteric fractures with the gamma nail. Arch Orthop Trauma Surg 2004;124(10):692–8. [22] Lopreite F, Lloyd R, del Sel H. Fractura medial del cuello del femur a continuacion de una fractura lateral: presentacion de dos casos y revisión de la bibliografia. Rev Asoc Argent Ortop Traumatol 2004;69(3):260–2. [23] Segata T, Kato T, Kunitake K, Kikukawa K. Femoral neck fracture after removal of the compression hip screw: a case report. Orthop Traumatol 2005;54 (4):643–6. [24] Yang KH, Choi YW, Won JH, Yoo JH. Subcapital femoral neck fracture after removal of gamma/proximal femoral nails: report of two cases. Injury Extra 2005;36(7):245–8. [25] George B, Gaheer RS, Ratnam A. Spontaneous femoral neck fracture after removal of dynamic hip screw. J Orthopaedics 2007;4(1):e14. [26] Hernandez AJ, Fávaro E, Laraya MH, Zumiotti AV. Spontaneous fracture of the femoral neck following PFN removal. Acta Ortop Bras 2009;17(3):187–9. [27] Seibert FJ, Puchwein P, Lanz P, Tanzer K, Clement HP. Femoral neck fracture after removal of PFNA-blade—case report and review of the literature. Injury Extra 2009;40(11):240–1. [28] Kim YS, Yoon JW, Han SK. [Treatment of intertrochanteric fracture of the femur using a dyna locking trochanteric (DLT) nail]. J Korean Hip Soc 2010;22(3):216– 21. [29] Bartoni9cek J. Complications of trochanteric fractures. In: Waddell JP, editor. Fractures of the proximal femur. Improving outcome. Philadelphia, PA: Saunders; 2011. p. 151–84. [30] Shaer JA, Hileman BM, Newcomer JE, Hanes MC. Femoral neck fracture following hardware removal. Orthopedics 2012;35(1):83–7. [31] El Ibrahimi A, Shimi M, Elmrini A. Femoral neck fracture occurring after gamma nail removal: case report and review of literature. Curr Orthop Prat 2013;24(1):105–7. [32] Legnani C, Dondi A, Pietrogrande L. Conservative treatment of a femoral neck fracture following nail removal. J Midlife Health 2013;4(3):191–4. [33] Yoon PW, Kwon JE, Yoo JJ, Kim HJ, Yoon KS. Femoral neck fracture after removal of the compression hip screw from healed intertrochanteric fractures. J Orthop Trauma 2013;27(12):696–701. [34] Kato H, Tsukamoto I, Inoue S, Mori S, Asada S. Femoral fracture after removal of short femoral nail. Cent Jpn J Orthop Surg Traumatol 2016;59(1):91–2. [35] Xue D, Yu J, Zheng Q, Feng G, Li W, Pan Z, et al. The treatment strategies of intertrochanteric fractures nonunion: an experience of 23 nonunion patients. Injury 2017;48(March (3)):708–14. [36] Takigawa N, Yasui K, Eshiro H, Moriuchi H, Abe M, Tsujinaka S, et al. Clinical results of surgical treatment for femoral neck fractures with the Targon(1) FN. Injury 2016;47(December (Suppl. 7)):S44–8. [37] Lang NW, Arthold C, Joestl J, Gormasz A, Boesmueller S, Hajdu S, et al. Does an additional antirotation U-Blade (RC) lag screw improve treatment of AO/OTA 31 A1–3 fractures with gamma 3 nail? Injury 2016;47(December (12)):2733–8. [38] Song HK, Choi HJ, Yang KH. Risk factors of avascular necrosis of the femoral head and fixation failure in patients with valgus angulated femoral neck fractures over the age of 50 years. Injury 2016;47(December (12)):2743–8. [39] Tucker A, Diamond O, McDonald S, Johnston A, Neil M, Kealey D, et al. Is there any place for the variable angle proximal femoral plate? A case matched cohort study against the Dynamic Hip Screw system. Injury 2016;47(October (10)):2173–81. [40] Konstantinidis L, Helwig P, Hirschmüller A, Langenmair E, Südkamp NP, Augat P. When is the stability of a fracture fixation limited by osteoporotic bone? Injury 2016;47(June (Suppl. 2)):S27–32. [41] Hoshino CM, Christian MW, O’Toole RV, Manson TT. Fixation of displaced femoral neck fractures in young adults: fixed-angle devices or Pauwel screws? Injury 2016;47(August (8)):1676–84. [42] Lin SJ, Huang KC, Chuang PY, Lee CY, Huang TW, Lee MS, et al. The outcome of unstable proximal femoral fracture treated with reverse LISS plates. Injury 2016;47(October (10)):2161–8. [43] Lim JW, Ng GS, Jenkins RC, Ridley D, Jariwala AC, Sripada S. Total hip replacement for neck of femur fracture: comparing outcomes with matched elective cohort. Injury 2016;47(October (10)):2144–8. [44] Mahaisavariya B, Sitthiseripratip K, Suwanprateeb J. Finite element study of the proximal femur with retained trochanteric gamma nail and after removal of nail. Injury 2006;37(8):778–85. [45] Scola A, Gebhard F, Dehner C, Röderer G. The PFNA augmented in revision of proximal femur fractures. Open Orthop J 2014;8:232–6. [46] Larsson S, Bauer TW. Use of injectable calcium phosphate cement for fracture fixation: a review. Clin Orthop Relat Res 2002;(395):23–32. [47] Strauss E, Pahk B, Kummer FJ, Egol K. Calcium phosphate cement augmentation of the femoral neck defect created after dynamic hip screw removal. J Orthop Trauma 2007;21(5):295–300.