Comparative Study Between Closed Reduction Internal Fixation Alone Versus Closed Reduction Internal Fixation and Addition of Muscle Pedicle Bone Graft for Prevention of Osteonecrosis in Cases of Displaced Femoral Neck Fractures

Abstract

Background

In patients of less than 60 years of age, osteosynthesis is the treatment of choice for displaced femoral neck fractures (Garden 3 and 4). Most of these fractures unite well but chances of osteonecrosis are quite high even after early operations and till date there is no well-defined method of prediction and prevention of osteonecrosis. Vascularized bone graft may prevent the development of osteonecrosis.

Materials and Methods

This is a retrospective case control study done in an urban Medical College in eastern India and tried to evaluate efficacy of addition of quadratus femoris muscle pedicle bone graft in cases of osteosynthesis comparing the results of cases treated with closed reduction and internal fixation alone for femoral neck fractures We have taken into consideration 45 consecutive cases from each group with identical demographic background and of very similar age pattern of 20 years to 60 years.

For further reference, we will describe Group I for closed reduction and CHS/DHS only, and Group II for closed reduction CHS and quadratus femoris muscle pedicle graft.

Results

Satisfactory bony union occurred in 42 out of 45 follow-up patients in Group 1 (93.3%) and 43 out of 45 patients in Group 2. (95.5%) In 2- to 10-year follow-up period, eight patients (18.6%) developed osteonecrosis in Group I, and in Group 2, in the same period, there were no cases of osteonecrosis.

Conclusion

Addition of quadratus femoris muscle pedicle bone graft is a very effective procedure for prevention of osteonecrosis when we attempt osteosynthesis by closed reduction and internal fixation for displaced femoral neck fractures.

Introduction

Treating femoral neck fractures is a challenge for all orthopedic surgeons. This is more so when the fracture occurs in young patients due to high-energy trauma and the fracture is of Garden type 3 and 4 and Pauwel’s angle type 3. Option of arthroplasty is not suitable for them. A good number of these patients are destined to suffer from osteonecrosis of femoral head within 2 to 5 years of operation. Very early operation within 8 h of injury could bring it down to 20% [1]. Improved methods of fixation of fracture adopted during last few decades have resulted in more than 90% union [2, 3]. But till date, there is no well-defined methodology assuring prevention of occurrence of osteonecrosis which is quite prevalent and remained an unsolved problem [4–6]. Improved results regarding union along with lesser incidence of osteonecrosis was reported when quadratus femoris muscle pedicle graft added after open reduction and internal fixation (ORIF) [6, 7]. In cases of fractures of femoral neck treated within 1 week of injury, it was shown that the procedure of closed reduction internal fixation and addition of quadratus femoris muscle pedicle bone graft might prevent the development of osteonecrosis [8]. In young patients, osteonecrosis is likely to lead to severe morbidity and the best way to tackle it is by prevention. Till date, there is no conclusive way to pre-identify the fractures leading to osteonecrosis and to take timely measures for its prevention. This study is concerned with prevention of osteonecrosis.

The hypothesis of our study: the addition of a vascularized quadratus femoris muscle pedicle graft after CRIF in displaced fractures of femoral neck may prevent osteonecrosis which cannot be ruled out when only CRIF is done.

Materials and Methods

Study Design

This is a retrospective study.

Study Area

The study was conducted in an urban Medical College of eastern India.

Study Population

The study was conducted in an urban Medical College of eastern India. Ninety consecutive patients, 45 from each group, Group1 (CRIF alone), Group 2 (CRIF and MPBG) who satisfied our inclusion criteria.

Study Period

From October 2009 to December 2016 were finally evaluated.

Inclusion Criteria

• Age group 20 years to 60 years,

• Operated within 48 h of injury,

• Could be followed up for more than 2 years.

Exclusion Criteria

• Existing pathological condition of hip, Sing’s osteoporosis [9] index 1–4,

• Polytrauma patients,

• History of long continued steroid intake.

Sample size: 90 Patients 45 from each group 1. Group 1 (CRIF alone), (2) Group 2 (CRIF and MPBG).

Sample Size Justification

The incidence of hip fractures population has 4.4%. Therefore, for this study, p = 0.044.

The formula used for sample size calculation was as follows: $n\text{is 4pq}/({\text{L}}^2)$ where n is the required sample size, p = 0.044 (as per the study), q = 1 – p, and L is the loss % (loss of information).

Calculation

$$\text{Here},p\text{is}0.044,Q=1-\text{p}=1-0.044=0.96$$

$$\text{4pq}=4\times 0.044\times 0.96=0.1690$$

$$L^2=0.00187$$

$$L=0.0432$$

$$\text{Loss}\%=4.32\%$$

$$n=\text{4pq}/\left(L^2\right)=0.1690/0.00187=90.3=90$$

90 patients were taken in this study.

45 patients in Group 1 (CRIF alone).

45 patients in Group 2 (CRIF and MPBG).

Thus, the number of patients required for this study was 90 with power 89% with 95% confidence of interval.

Study Design

It is a retrospective study.

Statistical Software for Sample Size

Sample size has been calculated with help of Epi Info (TM) 3.5.3. EPI INFO which is a trademark of the Centers for Disease Control and Prevention (CDC). For statistical analysis, data were entered into a Microsoft excel spreadsheet and then analyzed by SPSS 27.0. and Graph Pad Prism version 5. Data had been summarized as mean and standard deviation for numerical variables and count and percentages for categorical variables. Unpaired proportions were compared by Chi-square test or Fischer’s exact test, as appropriate. Two-sample t-tests for a difference in mean involved independent samples or unpaired samples. p value ≤ 0.05 was considered for statistically significant.

Methodology

Evaluation was done by two other surgeons who were not involved in the operation. Though retrospective, the permission of the Ethical Committee was duly obtained for conducting the study. Consecutive Ninety (male = 63, female = 27) patients of displaced (Garden type 3 and type 4) femoral neck fracture were included in this study. The average age of the patients was 42.6 years (range = 21 to 60 years). All of them had identical demographic features.

In our institution, the closed reduction internal fixation and muscle pedicle bone grafting (CRIF and MPBG, Group 2) was being done by the first author, while in another two units, closed reduction and internal fixation were being done with cannulated hip screw or by DHS (CRIF, Group 1) by other senior colleagues. After completion of 10 years, it was found that the number of cases done under the two procedures which satisfied our inclusion–exclusion criteria and also having at least 2-year follow-up, were adequately sufficient for carrying out the study. These patients were retrospectively evaluated from early 2019.

Surgical Procedure

Closed reduction internal fixation and addition of quadratus femoris muscle pedicle bone grafting described in details.

Under regional or general anesthesia, the patient was placed prone on the fracture table. Reduction was done in extension (Fig. 1 and confirmed by C-arm image intensifier in both AP and lateral projections. It was assessed according to Garden’s alignment index and Lowel’s/Lowel,s method. [10, 11] A 12 cm long incision (Fig. 1) was given from just above and medial to the tip of greater trochanter and along the greater trochanter to the shaft of the femur. The fascia lata was incised in the same line and the vastus lateralis was split to expose the base of trochanter and upper shaft of femur. Three cannulated hip screws were introduced in a triangular configuration with two upper screws and one lower screw. To minimize the amount of bleeding, the vastus lateralis was closed and upper part of the incision was explored by splitting the gluteus maximus. The bursa over trochanter was split to expose quadratus femoris and its bony attachment. This muscle was properly delineated with help of a thin bone lever (Fig. 2). The exact size and shape of the graft was then marked at the insertion of quadratus femoris in the trochanteric crest with the scalpel. Harvesting of pedicular bone started 1 cm proximal to the upper border of this muscle so that this excess amount of bone could be tapered later on to get inserted into the slot made in the femoral head. The distal end of the graft extended just above the level of the lesser trochanter. The graft was then harvested with the help of straight and curved osteotomes taking care to have its depth and width of about 1 cm. each. Harvesting thicker graft might injure medial circumflex femoral artery while thinner one likely to break. We must avoid invading the base of the neck to safeguard the artery. The graft length was about 3 cm in this series (Fig. 3). The graft was held with a tissue forceps and quadratus femoris was further mobilized to shift the graft medially. Other short rotators were retracted by two bone levers placed on either side of the femoral neck. Posterior neck, head and capsule were identified. L shaped incision was made at the middle of head neck junction and extending proximally while the horizontal limb of the L was directed inferiorly to safeguard lateral epiphyseal vessels. The articular surface of the head became visible. A number 1 vicryl stay suture was passed through the capsule for securing the graft later on (Fig. 4). A slot was made in the head of the femur with osteotome and gouge starting at the head neck junction and going 1 cm proximally. (Fig. 5) The cephalad end of the graft was denuded of any soft tissues, made slightly tapered and was introduced into the slot. The graft was then anchored with capsule with the help of vicryl stay suture already placed in the capsule, and it was ensured that the graft was firmly secured in the slot (Figs. 6 and 7). Gel foam was applied over donor site at trochanteric crest. The fracture site was never opened. The wound was closed in layers over a suction drain. No patient required per operative blood transfusion.

Fig. 1.

Reduction in extension in prone position. Incision line shown

Fig. 2.

Vastus lateralis sutured and quadratus femoris delineated

Fig. 3.

Quadratus femoris harvested

Fig. 4.

Opening of the hip joint capsule—head of the femur visible. Stay suture passing through the capsule

Fig. 5.

Preparing the slot in the head of femur

Fig. 6.

Quadratus femoris muscle pedicle bone graft placed in the slot prepared in the head of the femur and being secured by suturing with the capsule

Fig. 7.

Diagrammatic representation of donor site of the graft and placement of the graft in the slot in femoral head—actual per operative picture of the same

Surgical technique used in patients of Group 1 was the standard technique of CRIF using CHS/ DHS in supine position in fracture table.

Post-operative Care

The patients were allowed to sit up in bed after 24 h. The suction drain was removed after 24 h. Quadriceps drill and gentle knee bending exercise started from the 2nd post-operative day. They were allowed non-weight bearing crutch walking after three days under proper supervision of physiotherapists. Sutures were removed on the 10th day and they were discharged on the 11th day. Non-weight bearing crutch walking continued at their residence under care of physiotherapists. Partial weight bearing started from 6th week. Subsequent follow-ups were done, respectively, at the 6th week, the 12th week, the 24th week and then at an interval of 3 months for 1 year and after that every 6 months.

Clinical and Radiological Assessment

At the 6 month post-operation, the final conclusion on union was arrived at using the RUSH score [12]^. It was based on four measures of healing: cortical bridging, cortical fracture disappearance, trabecular consolidation and trabecular fracture disappearance. In this study, in Group 1, bony union occurred in 42 out of 45 patients, and in Group 2, it was 43 out of 45. The next follow-ups were done, respectively, at 9 months and 1 year. During these follow-ups, the focus was on the range of movements, ability to walk without support and on leg length discrepancy if any. From 1 year onwards, the follow-up typically put thrust on whether there was any evidence of aseptic necrosis both clinically and radiologically in terms of appearance of pain and or restriction of movements and in X-ray changes in bone density, sclerotic patches, cystic changes, subchondral collapse and flattening of femoral head. Osteonecrosis was detected between 18 months and 3 years. In Group 1, eight cases developed osteonecrosis, and in Group 2, no case of osteonecrosis was seen in 3- to 10-year follow-up period.

Statistical Analysis

To compare the patient features in two treatment groups (Group 1: CHS/DHS alone, Group 2: CHS + bone graft) categorized across three factors: union, nonunion and osteonecrosis. In other words, the test involves the following set of hypotheses:

• I.Null hypothesis: no difference in the incidence rate of osteonecrosis across groups. Alternative hypothesis: incidence of osteonecrosis in Group 1 > that in Group 2.
• II.Null hypothesis: no difference in the incidence rates of union and nonunion across groups.

Alternative hypothesis: higher in one compared to the other

Test on Osteonecrosis

Age profile of patients (in years)	21–40	41–60
Group 1 No necrosis	14	23
Necrosis	5	3
Group 2 No necrosis	18	27
Necrosis	0	0

A 3-way contingency table analysis is used to test the homogeneity of proportions across the groups defined by each of the factors above, while correcting for age effect. We have used R 3.5.1 for performing the Fisher’s Exact Test for comparing the two groups (Table 1).

Table 1.

Comparison of outcome between Group I and Group II

Age group	CRIF alone Group I	CRIF with MPBG Group II
20–40
Group I 15 M 04 F	Nonunion 01	Nonunion 00
Group II 16 M 02 F	Osteonecrosis 05	Osteonecrosis 00
41–60
Group I 14 M 12 F	Nonunion 02	Nonunion 02
Group II 18 M 09 F	Osteonecrosis 03	Osteonecrosis 00

P value = 0.00332 Alternative hypothesis: true odds ratio is not equal to 1.

Here, p is 0.044, Q = 1.

95 percent confidence interval: 1.859021.

The P value for this test being $0.00332<0.05$ clearly indicates that there is a statistically significant difference in the incidence rates of osteonecrosis between Group 1 and Group 2.

Union

Unlike necrosis, the difference in union does not show a statistically significant difference (that is the P value is not small enough)—which is not unexpected as the difference in percentages was not too high relative to the number of cases.

The high P value indicates we do not have sufficient information to reject the null hypothesis of no difference between the groups.

Results

The follow-up period of patients in the two groups varied from 2 years to 10 years (average being 5.2 years). Following are the observed outcomes:

• Superficial infection occurred in two cases which were healed within 2 weeks by regular dressing and continued antibiotic therapy.

• Slight shortening of 0.5 cm occurred in six patients which was due to collapse at the fracture site. In the younger patients, there were no shortening.

• Operation in prone position in fracture table with traction maintained for more than 1 h may cause compression of pudendal nerve, but in our series, it did not happen. Our OT staffs also took care of male genitalia particularly in prone position.

• Modified Harris hip scoring system [13] was used for analyzing the results at the end of 2 years. In Group 1, it was excellent in 30 cases, good in 6, fair in 2 and poor in 7 cases. In Group 2, it was excellent in 35 cases and good in 8 cases and poor in 2 cases (Figs. 8, 9, 10, 11 and 12).

• In Group 1

  • i.
    Osteonecrosis—found in eight cases
  • ii.
    Pain, restriction of movements (evident from 18 months to 2 years) with radiological features also evident—found in six cases, all of them less than 50 years old. All had to undergo total hip replacement within 4 years.
  • iii.
    In two other patients of more than 50 years who developed osteonecrosis, it took more than 3 years to become radiologically evident. One of them underwent total hip replacement.

Fig. 8.

23-year-old male with displaced femoral neck fracture

Fig. 9.

Eight years’ post-operative result of the same patient of Fig. 8

Fig. 10.

56-year-old female with displaced femoral neck fracture treated with CHS and muscle pedicle bone graft and having 5-year follow-up. Graft visible in the lateral view

Fig. 11.

Three-year follow-up in a 42-year-old male with displaced femoral neck fracture showing bony union of the fracture by CHS—osteonecrosis developed and also segmental collapse

Fig. 12.

50-year-old female operated within 24 h developed osteonecrosis of femoral head after 2 years

Discussion

Morbidity resulting from impairment of vascular supply as a consequence of fracture is very often evident in displaced fractures of the femoral neck. As we know, the three sources of blood supply to the femoral head are capsular vessels, intramedullary vessels and a contribution from ligamentum teres [14]. In the adult, most important source of supply to the femoral head comes from the capsular vessels. Within the capsule, these are known as retinacular vessels and are of four types—anterior, medial, lateral and posterior. The lateral retinacular vessel, a branch of medial femoral circumflex artery, is the main source of supply and it courses along the postero superior aspect of the femoral neck to supply the femoral head. In displaced fracture of the femoral neck, these vessels are kinked and compromised and some additional vascular supply which might reach the head via medullary bone in the neck are totally cut off. Early operation with anatomical reduction and internal fixation tries to restore the vascularity of femoral head and is likely to increase the rate of union and minimize the chance of osteonecrosis. In young adult patients, urgent operation may bring down the occurrence of osteonecrosis from 86 percent to 18.5 percent but incidence still remains high and more frequent in younger patients even after urgent operation [4, 13, 14],. For early prediction of osteonecrosis lots of invasive and non-invasive procedures [15–20] were used. With the advent of dynamic MRI, an attempt to identify fractures which might lead towards osteonecrosis came into being. It revealed how in almost all displaced fractures vascularity was impaired. Using dynamic MRI within 48 h of injury, it was found that in more than 50% cases of femoral neck fractures, femoral head perfusion was absent and most of them developed osteonecrosis. The use of advanced MRI came later, but till date it is not 100% specific and does not give any clue for prevention. Histological study of femoral head after 16 days of femoral neck fracture in 109 cases showed diminished vascularity in all the cases [19, 21].

To find ways to prevent osteonecrosis, animal experiment was done on dogs using muscle pedicle bone graft. Experimental studies were conducted using distal iliopsoas muscle pedicle bone graft in dogs after displaced femoral neck fractures and dislocation of hip. It was proved by microangiographic and histologic studies this procedure can maintain vascularity of femoral head [20, 22] While treating femoral neck fractures with multiple CHS, the presence of continued bleeding from drill holes for 5 min was considered to be a good indicator of femoral head perfusion. In all cases where bleeding stopped before 5 min, osteonecrosis developed, and in one case where bleeding continued, osteonecrosis still developed [21, 23]. Therefore, obviously 100% predictability and clue for prevention is missing.

Recently, post-operative bone SPECT/CT [22, 24] started to be done after 2 weeks and if cold defects were found, follow-up SPECT/CT were done 2 to 10 months later. Three out of eight patients of Garden type 3 fracture and six out of six patients of Garden type 4 fracture developed osteonecrosis. In many/cases, revascularization occurred in cold defects and final conclusion could be made only after 1 year. In another recent study [23, 25], it was shown that 22.6% of patients with displaced femoral neck fracture developed osteonecrosis and operation within 24 h did not alter the outcome. Judet [24, 26] first proposed the idea of adding muscle pedicle bone graft for treatment of femoral neck fractures. In 1973, Meyer’s published encouraging results in respect of union and prevention of osteonecrosis by his technique of open reduction, internal fixation and quadratus femoris muscle pedicle bone graft. In his series, there was 89 percent union and osteonecrosis were 8 percent. Subsequently other authors could not duplicate Meyer’s result and it went into disrepute as a procedure on the ground that it deformed the anatomy and might cause harm to medial circumflex femoral artery. In our study, it was seen that careful harvesting of the graft never led to injury to medial circumflex femoral artery. Even in recent times, the incidence of osteonecrosis revealed in different reports were quite high up to the extent of about 25% in displaced femoral neck fractures and very early operative interference could not make any significant difference [23, 25–30]. Vascularity of the relatively avascular heads can be restored by vascularized muscle pedicle graft like quadratus femoris. In 3- to 4-week-old fractures treated by open reduction internal fixation and muscle pedicle graft, osteonecrosis was 10% only [29, 31]. In another study with average 5-week-old fractures, incidence of osteonecrosis was 6.89% only after addition of quadratus femoris muscle pedicle bone graft [30, 32]. In fractures treated within 1 week, it was shown that by the method of closed reduction internal fixation and addition of quadratus femoris pedicle graft, there were no occurrence of osteonecrosis [7, 8]. In a recent systematic review, it has been shown that muscle pedicle bone flap transportation definitely improves union rate and reduces chances of osteonecrosis [31, 33]. Grant’s [32, 34] atlas of anatomy description was strictly adhered to during the operative procedure particularly during harvesting of the graft, use of curved osteotome must ensure graft width not more than 1 cm to avoid injury to medial circumflex artery. Cadaveric dissection of hip helped a lot in this matter.

This study is in quest of an optimally effective treatment for displaced femoral neck fractures in young and middle-aged patients. In the present study, male patients belonging to 20–40 years of age group were more in number and the fractures were mostly due to road traffic accidents. If we fail to prevent osteonecrosis in them, they are likely to suffer from morbidity for the rest of their life. It is important to highlight the following:

• Group 1 (retrospective control group):

  • i.
    In 45 cases followed up for more than 3 years, 03 cases developed nonunion
  • ii.
    Out of 42 cases of fracture union, 08 cases developed osteonecrosis (18.6%)
• Group 2

  • i.
    Out of 45 cases, 02 cases went into nonunion
  • ii.
    In 43 patients followed up for more than 3 years, not a single case developed osteonecrosis.

This study shows in fractures treated by early operation and stable internal fixation addition of muscle pedicle bone graft is a necessity for prevention of osteonecrosis in subsequent years. We did not want to take any chance but simply extended our procedure for another 25 to 30 min to allow the head of the femur to recover from vascular insufficiency caused by the displaced femoral neck fracture.

Limitations of This Study

As this is a retrospective study, we could not ensure more than 2-year follow-up in all cases operated during this period so we have taken 90 consecutive cases only.

We have depended on clinical and plain radiological features only to ascertain osteonecrosis. Recently, SPECT/CT was used 2 weeks after operation to find cold defects but final conclusion could not be drawn before 1 year as many cold defects were normalized subsequently.

Quadratus femoris pedicle graft usually increases the femoral neck girth resulting in restoration of vascularity of femoral head, but in our series, neck girth measurement pre- and post-operative were not done.

There was no mention about measures taken for revascularization of femoral head when osteonecrosis detected at 18 months in very young patients. In the last 2 years, we did tensor fascia lata muscle pedicle graft in two patients of less than 35 years old who developed osteonecrosis after union of fracture. Till date, results are encouraging but we have to wait for some more time to come to any conclusion.

Conclusion

In the end, the evidence-based conclusion of the study may once again be highlighted as that the addition of quadratus femoris muscle pedicle bone graft will bring new ray of hope to the life of younger patients of displaced femoral neck fracture by minimizing to the extent of naught the possibility of the incidence of osteonecrosis.

We all know prevention is better than cure, and if we add 25 min more to the duration of the operative procedure, we are most likely to safeguard the return of our younger patients back to normalcy for the rest of their life. With not much complexity in the procedure, an average orthopedic surgeon will be able to perform this operation.

Author Contributions

SC: did CHS and muscle pedicle graft in all of his cases; BMP and AS: did CHS/DHS only in all of their cases; SS, KG and RC: did the comparative analysis.

Declarations

Conflict of Interest

On behalf of all the authors, the corresponding author states that there is no conflict of interest in this study.

Ethical approval

This article does not contain any studies with human or animal subjects performed by the any of the authors.

Informed consent

For this type of study informed consent is not required.