Skip to main content
  • Systematic Review
  • Open access
  • Published:

Cannulated screws versus dynamic hip screw versus hemiarthroplasty versus total hip arthroplasty in patients with displaced and non-displaced femoral neck fractures: a systematic review and frequentist network meta-analysis of 5703 patients

Abstract

Background

Our aim was to determine the best operative procedure in human participants with a displaced or non-displaced femoral neck fracture comparing cannulated screw (CS) fixation, dynamic hip screw (DHS) fixation, hemiarthroplasty (HA), and total hip arthroplasty (THA) in terms of surgical and functional outcomes, reoperation and postoperative complications.

Methods

We searched PubMed, The Cochrane Library, Clinical trials, CINAHL, and Embase for randomized controlled trials (RCTs) or quasi-RCTs up to 31 July 2022. A frequentist network meta-analysis was performed to assess the comparative effects of the four operative procedures, using fixed-effects and random-effects models. Mean differences (MDs) with 95% confidence intervals (CIs) were estimated for continuous variables and odds ratios (ORs) with 95% CIs were estimated for binary variables.

Results

A total of 33 RCTs with 5703 patients were included in our network meta-analysis. CS fixation was best in terms of operation time (CS: MD = − 57.70, 95% CI − 72.78; − 42.62; DHS: MD = − 53.56, 95% CI − 76.17; − 30.95; HA: MD = − 20.90, 95% CI − 30.65; − 11.15; THA: MD = 1.00 reference) and intraoperative blood loss (CS: MD = − 3.67, 95% CI − 4.44; − 2.90; DHS: MD = − 3.20, 95% CI − 4.97; − 1.43; HA: MD = − 1.20, 95% CI − 1.73; − 0.67; THA: MD = 1.00 reference). In life quality and functional outcome, measured at different time points with EQ-5D and the Harris Hip Score (HHS), THA ranked first and HA second (e.g. EQ-5D 2 years postoperatively: CS: MD = − 0.20, 95% CI − 0.29; − 0.11; HA: MD = − 0.09, 95% CI − 0.17; − 0.02; THA: MD = 1.00 reference; HHS 2 years postoperatively: CS: MD = − 5.50, 95% CI − 9.98; − 1.03; DHS: MD = − 8.93, 95% CI − 15.08; − 2.78; HA: MD = − 3.65, 95% CI − 6.74; − 0.57; THA: MD = 1.00 reference). CS fixation had the highest reoperation risk, followed by DHS fixation, HA, and THA (CS: OR = 9.98, 95% CI 4.60; 21.63; DHS: OR = 5.07, 95% CI 2.15; 11.96; HA: OR = 1.60, 95% CI 0.89; 2.89; THA: OR = 1.00 reference).

Conclusion

In our cohort of patients with displaced and non-displaced femoral neck fractures, HHS, EQ-5D, and reoperation risk showed an advantage of THA and HA compared with CS and DHS fixation. Based on these findings, we recommend that hip arthroplasty should be preferred and internal fixation of femoral neck fractures should only be considered in individual cases. Level of evidence I: a systematic review of randomized controlled trials.

Trial registration: PROSPERO on 10 August 2022 (CRD42022350293).

Introduction

One of the most common fractures in elderly patients is the femoral neck fracture [1]. It is associated with a high rate of morbidity and mortality in the elderly population [2]. The most common classifications used to evaluate femoral neck fractures are the Garden [3] and Pauwels [4] classifications. According to Pauwels, fractures are assessed from a mechanical point of view taking into account the orientation of the fracture line, which has an effect on the shear force and varus stress and thus the risk of fracture displacement. The Garden classification provides information on the degree of dislocation, and it is often used in decision-making for the preferred treatment of femoral neck fractures [5]. Garden I and II are defined as non-displaced (or minimally displaced), while Garden III and IV are defined as displaced femoral neck fractures [3]. There is a widespread agreement that displaced fractures should be treated with hip arthroplasty, while non-displaced fractures should be treated with internal fixation to preserve the femoral head [6, 7]. Deviations from this generally accepted procedure are not uncommon in individual cases. In recent years, several studies have shown that head preservation with internal fixation in elderly patients with displaced and non-displaced femoral neck fractures is associated with a high risk of reoperation, implant-related complications such as avascular necrosis of the femoral head, and nonunion [6,7,8,9]. Recent meta-analyses suggested that hip arthroplasty should also be performed in patients with non-displaced femoral neck fractures, as internal fixation was associated with a risk of poor clinical outcomes [6, 8]. On the other hand, numerous meta-analyses already highlighted some disadvantages of hemiarthroplasty (HA) and total hip arthroplasty (THA) in the treatment of patients with femoral neck fractures [10,11,12,13,14]. However, the network meta-analysis by Zhang et al. [15] was the first study that ranked the best operative procedure for femoral neck fractures. This 2017 network meta-analysis compared the outcomes of 7 operative procedures for displaced femoral neck fractures and showed that internal fixation had the highest, unipolar cemented HA had the lowest reoperation incidence; uncemented THA had the highest displacement incidence; and bipolar uncemented HA had the lowest infection incidence. This network meta-analysis focused on differentiating between cemented and uncemented implants, but, unfortunately, it did not differentiate between the type of internal fixation: cannulated screws (CS) or dynamic (sliding) hip screw (DHS). This is a serious limitation as the current literature has shown that DHS was superior to the CS for internal fixation of femoral neck fractures [16, 17]. Furthermore, the study by Zhang et al. [15] is limited to displaced femoral neck fractures. Therefore, there is an urgent need to address these limitations.

We formulated the following PICO (Population, Intervention, Control, and Outcomes) question: In human participants with a displaced or non-displaced femoral neck fracture, what is the best and worst operative procedure among CS fixation, DHS fixation, HA, and THA in terms of surgical and functional outcomes, reoperation and postoperative complications?

Methods

The study protocol was registered in PROSPERO on 10 August 2022 (CRD42022350293). We adhered to the PRISMA Extension Statement for Reporting of Systematic Reviews Incorporating Network Meta-analyses of Health Care Interventions as the basis for the methodology and presentation of the data [18]. The PRISMA Checklist is available in Additional file 1. It must be taken into account that the group of authors of the present study has some experience in the field of meta-analyses and THAs. Similarities between all meta-analyses can only be attributed to the use of the same high-quality methods.

Data sources and search strategies

We searched the following databases for randomized controlled trials (RCTs) or quasi-RCTs up to 31 July 2022: PubMed, The Cochrane Library, Clinical trials, CINAHL, and Embase. We developed a BOOLEAN search strategy, adapted to the syntax of the databases searched, using the following MeSH terms: ‘femoral neck fracture’, ‘displaced’, ‘undisplaced’, ‘non-displaced’, ‘nondisplaced’, ‘Garden’, ‘internal fixation’, ‘cannulated screws’, ‘sliding hip screw’, ‘dynamic hip screw’, ‘hemiarthroplasty’, ‘total hip arthroplasty’, ‘THA’, ‘THR’, ‘hip arthroplasty’, ‘hip replacement’. We additionally checked the reference list of related meta-analyses for relevant records. We did not review grey literature. There were no restrictions to publication language. We did not include RCTs that were older than 30 years.

Screening and selection of RCTs

We searched for titles, abstracts, and finally full-text articles according to our inclusion criteria. RCTs were included on the basis of consensus between two reviewers (NR, PL). The RCT selection process was described in a flowchart diagram.

Inclusion criteria

The types of studies included were only RCTs. The types of participants included were only human participants with a displaced or non-displaced femoral neck fracture. The types of interventions included were the following: cannulated screw (CS) fixation, dynamic (sliding) hip screw (DHS) fixation with or without anti-rotation screws, hemiarthroplasty (HA), and total hip arthroplasty (THA). The types of outcome measures included were: operation time, intraoperative blood loss, EQ-5D [19], Harris Hip Score (HHS) [20], hospital stay, reoperation, mortality, and postoperative complications such as: deep vein thrombosis, hematoma, infection, intraoperative fracture, failure, avascular necrosis of the femoral head (ANFH), dislocation, and nonunion. “Failure” was defined as different types of osteosynthesis failure and loosening, as well as different types of prosthesis failure and loosening. There were no exclusion criteria regarding age and comorbidities of the participants.

Statistical analysis

Data extraction and quality evaluation

Relevant data on RCT characteristics, methods, quality assessment, participant characteristics, operational details, relevant outcomes, and relevant additional information were extracted by two reviewers [NR, PL] independently from each other. The raw data extraction sheet is available in Additional file 2. The agreement between the reviewers was assessed using the Cohen's Kappa coefficient (κ). If the RCTs provided different information on intention to treat (ITT) and per protocol analysis, we adhered to the numbers from the ITT analysis. We compared the characteristics of the patient cohort between the 4 operative procedures, using Kruskal–Wallis tests and a significance level of 5%. Risk of bias was assessed using the Cochrane’s Risk of Bias 2 tool [21].

Direct and indirect comparisons: network meta-analysis

A pairwise and network meta-analysis was performed to simultaneously assess the comparative effects of the four operations: CS fixation, DHS fixation, HA, and THA. All analyses were performed using fixed-effects and random-effects models estimated with the frequentist approach and consistency assumption. However, we only interpreted the results of the random-effects model as we believe that they can be generalized beyond the included studies. Mean differences (MDs) with 95% confidence intervals (CIs) were estimated for continuous variables and odds ratios (ORs) with 95% CIs were estimated for binary variables. In the case of a global and local validation of the consistency assumption, we created a net heat plot to decompose the between-design heterogeneity component into the contribution for each study design. Problematic RCTs were then removed and a network meta-analysis model was repeated. Two-arm comparisons from multi-arm RCTs were included in analyses with adjusted standard errors to account for the fact that comparisons within multi-arm RCTs were correlated. Between-study variance was estimated using the DerSimonian–Laird method, while heterogeneity was assessed using the Cochrane Q statistic test and the Higgins I2 test. Treatment effects together with the 95% CIs were presented on forest plots for separate pairwise comparisons evaluated in the individual studies and for comparison with THA or CS based on the network meta-analysis. The overall effect based on the network meta-analysis included the effect of the direct and indirect comparisons. In addition, we ranked the operations based on the cumulative probabilities for the highest to the lowest priority operations using the surface under the cumulative ranking curve (SUCRA) values obtained with 1000 simulations [22]. Analyses per outcome were performed for all RCTs together and included only RCTs with patients with displaced femoral neck fractures. All statistical analyses were performed using netmeta and metaphor packages in the R software version 4.2.0 [23].

Missing data

Missing information on the standard deviation was imputed using a pooled value of all reported standard deviations calculated for each operation separately. If there was only one RCT for a given operation and it had a missing value, we replaced the standard deviation with a pooled value of all reported standard deviations calculated for all operations together. These imputed values were used for the main analyses [24]. Two sensitivity analyses were performed: (1) if standard deviation was not reported but the minimum and maximum values were given, an approximate standard deviation was obtained using the formula (max–min)/4 and (2) only RCTs that reported complete information were included.

Results

The study selection process is shown in a flowchart (Fig. 1). After removal of duplicates and screening of titles and abstracts with a high inter-reviewer agreement (κ = 0.98), 39 RCTs were assessed for eligibility [25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63]. After the second screening procedure by full-text analysis (κ = 1.0), 6 RCTs were excluded for the following reasons: 4 of these RCTs did not report any of the outcomes of interest [58,59,60,61], 2 RCTs did not differentiate between displaced and non-displaced femoral neck fractures [62, 63]. A total of 33 RCTs [25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57] with 5703 patients were included in our network meta-analysis. Of these patients, 5276 (92%) had a displaced femoral neck fracture, and 427 (8%) had a non-displaced femoral neck fracture. The fractures were operated as follows: CS fixation in 913 (16%) patients, DHS fixation in 372 (6.5%) patients, HA in 2606 (46%) patients, and THA in 1812 (31.5%) patients. Clinical characteristics for age, sex, time to surgery and follow-up period (Tables 1, 2) showed no significant differences between the 4 operative procedures. On average, patients were 79 years of age (range: 59–86), 26% of them were males, their time to surgery was 43 h and they were followed up for 46.3 months. On average, the patients of the CS fixation group were 82 years of age (range: 77–86), 24% were males, their time to surgery was 30 h and they were followed-up for 27 months. On average, the patients of the DHS fixation group were 77 years of age (range: 73–81), 26% were males, their time to surgery was 37 h and they were followed up for 53 months. On average, the patients of the HA group were 78 years of age (range: 63–86), 26% were males, their time to surgery was 46 h and they were followed up for 48 months. On average, the patients of the THA group were 78 years of age (range: 59–85), 27% were males, their time to surgery was 54 h and they were followed up for 51 months. Further details on the patient cohort and the included RCTs are given in Table 1. It is important to notice that two RCTs [42, 46] did not compare 2 operative procedures, but 3 operative procedures: DHS fixation, HA, and THA. Another RCT [30] compared CS fixation with two different types of cemented HA. The distinction between the two types of cemented HA was not relevant to our network meta-analysis and the data from the two HAs were pooled (Table 2). Another RCT [41] investigating CS fixation and DHS fixation provided data that included displaced and non-displaced fractures. Accordingly, 4 different operative procedures (CS fixation for displaced fractures, CS fixation for non-displaced fractures, DHS fixation for displaced fractures, and DHS fixation for non-displaced fractures) were included in the network meta-analysis of this study [41]. RCTs that included non-displaced fractures [31, 38, 41, 57] are marked accordingly in the forest plots. The assessment of the risk of bias is shown in Table 3. Some RCTs had a high risk of bias [25,26,27,28, 31, 33,34,35, 44,45,46,47,48,49,50, 52, 55, 56]. One of them is more likely to be considered a quasi-RCT [48]. The statistical heterogeneity of all outcomes measured is shown in Figs. 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 and 21.

Fig. 1
figure 1

Flow chart of the article selection process

Table 1 Patient and RCT details
Table 2 Clinical characteristics of the patient cohort
Table 3 Risk of bias assessment
Fig. 2
figure 2

Forest plot of operation time. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 3
figure 3

Forest plot of intraoperative blood loss.Results are shown for a unit of 100 ml. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 4
figure 4

Forest plot of EQ 5D 3–4 months postoperatively. CS, cannulated screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 5
figure 5

Forest plot of EQ 5D 12 months postoperatively. CS, cannulated screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 6
figure 6

Forest plot of EQ 5D 2 years postoperatively. CS, cannulated screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 7
figure 7

Forest plot of Harris Hip Score ≤ 6 months postoperatively. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 8
figure 8

Forest plot of Harris Hip Score 12 months postoperatively. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 9
figure 9

Forest plot of Harris Hip Score 2 years postoperatively. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 10
figure 10

Forest plot of Harris Hip Score 3–5 years postoperatively. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 11
figure 11

Forest plot of hospital stay. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 12
figure 12

Forest plot of reoperation. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 13
figure 13

Forest plot of mortality. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 14
figure 14

Forest plot of deep vein thrombosis. CS, cannulated screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval

Fig. 15
figure 15

Forest plot of hematoma. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval

Fig. 16
figure 16

Forest plot of infection. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 17
figure 17

Forest plot of intraoperative fracture. CS, cannulated screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 18
figure 18

Forest plot of failure. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 19
figure 19

Forest plot of dislocation. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval

Fig. 20
figure 20

Forest plot of ANFH. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; OR, odds ratio; CI, confidence interval; *RCT with non-displaced fractures only

Fig. 21
figure 21

Forest plot of nonunion. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval; *RCT with non-displaced fractures only

Network meta-analysis

Operation time

Data on 666 patients, operated with CS fixation, were pooled from 7 RCTs [31, 32, 38, 41, 44, 47, 52], data on 150 patients, operated with DHS fixation, were pooled from 2 RCTs [41, 56], data on 1031 patients, operated with HA, were pooled from 17 RCTs [25, 27,28,29, 31, 32, 35, 38, 39, 44, 45, 47,48,49,50, 54, 56], and data on 517 patients, operated with THA, were pooled from 12 RCTs [25, 27,28,29, 35, 39, 45, 48,49,50, 52, 55]. The operation time of THA was 84.7 min. CS fixation had significantly shorter operation time of 57.7 min. than THA; DHS fixation had significantly shorter operation time of 53.6 min. than THA, HA had significantly shorter operation time of 20.9 min. than THA (CS: MD = − 57.70, 95% CI − 72.78; − 42.62; DHS: MD = − 53.56, 95% CI − 76.17; − 30.95; HA: MD = − 20.90, 95% CI − 30.65; − 11.15; THA: MD = 0.00 reference; Fig. 2). The global consistency assumption of the network meta-analysis model was not met. However, the local consistency assumption was met, suggesting no significant design-by-treatment interaction.

Intraoperative blood loss

Data on 543 patients, operated with CS fixation, were pooled from 5 RCTs [31, 32, 38, 44, 52], data on 21 patients, operated with DHS fixation, were pooled from 1 RCT [56], data on 841 patients, operated with HA, were pooled from 12 RCTs [27, 29, 31, 32, 38, 44, 45, 48,49,50, 54, 56], and data on 383 patients, operated with THA, were pooled from 8 RCTs [27, 29, 45, 48,49,50, 52, 54]. The intraoperative blood loss of THA was 400 ml. CS fixation had significantly lower intraoperative blood loss of 367 ml than THA; DHS fixation had significantly lower intraoperative blood loss of 320 ml than THA, HA had significantly lower intraoperative blood loss of 120 ml than THA (CS: MD = − 3.67, 95% CI − 4.44; − 2.90; DHS: MD = − 3.20, 95% CI − 4.97; − 1.43; HA: MD = − 1.20, 95% CI − 1.73; − 0.67; THA: MD = 0.00 reference; Fig. 3). The global and local consistency assumption of the network meta-analysis model was not met. A net heat plot showed that the inclusion of the RCT by Tidermark et al. [52] resulted in the greatest heterogeneity. However, the model without the RCT by Tidermark et al. [52] showed comparable estimates to the model including all RCTs.

EQ 5D 3–4 months postoperatively

Data on 223 patients, operated with CS fixation, were pooled from 2 RCTs [31, 32], data on 338 patients, operated with HA, were pooled from 4 RCTs [29, 31, 32, 34], and data on 120 patients, operated with THA, were pooled from 2 RCTs [29, 34]. There was no significant difference in EQ 5D 3–4 months postoperatively between CS fixation, HA, and THA (CS: MD = − 0.08, 95% CI − 0.17; 0.00; HA: MD = − 0.01, 95% CI − 0.08; 0.06; THA: MD = 0.00 reference; Fig. 4).

EQ 5D 12 months postoperatively

Data on 223 patients, operated with CS fixation, were pooled from 2 RCTs [31, 32], data on 338 patients, operated with HA, were pooled from 4 RCTs [29, 31, 32, 34], and data on 120 patients, operated with THA, were pooled from 2 RCTs [29, 34]. CS fixation had significantly lower EQ 5D 12 months postoperatively of 0.11 points than THA. There was no significant difference in EQ 5D 12 months postoperatively between HA and THA (CS: MD = − 0.11, 95% CI − 0.21; − 0.02; HA: MD = − 0.03, 95% CI − 0.11; 0.04; THA: MD = 0.00 reference; Fig. 5).

EQ 5D 2 years postoperatively

Data on 223 patients, operated with CS fixation, were pooled from 2 RCTs [31, 32], data on 338 patients, operated with HA, were pooled from 4 RCTs [29, 31, 32, 34], and data on 120 patients, operated with THA, were pooled from 2 RCTs [29, 34]. CS fixation had significantly lower EQ 5D 2 years postoperatively of 0.20 points than THA. HA had significantly lower EQ 5D 2 years postoperatively of 0.09 points than THA (CS: MD = − 0.20, 95% CI − 0.29; − 0.11; HA: MD = − 0.09, 95% CI − 0.17; − 0.02; THA: MD = 0.00 reference; Fig. 6).

Harris Hip Score ≤ 6 months postoperatively

Data on 434 patients, operated with CS fixation, were pooled from 5 RCTs [31, 32, 38, 51, 57], data on 138 patients, operated with DHS fixation, were pooled from 2 RCTs [42, 57], data on 794 patients, operated with HA, were pooled from 11 RCTs [27,28,29, 31, 32, 34, 38, 39, 42, 51, 54], and data on 425 patients, operated with THA, were pooled from 7 RCTs [27,28,29, 34, 39, 42, 54]. Two [54, 57] of those RCTs provided data at different time points (HHS ≤ 3 and 6 months postoperatively). CS fixation had significantly lower HHS ≤ 6 months postoperatively of 6.26 points than THA. DHS fixation had significantly lower HHS ≤ 6 months postoperatively of 5.02 points than THA. There was no significant difference in HHS ≤ 6 months postoperatively between HA and THA (CS: MD = − 6.26, 95% CI − 9.74; − 2.79; DHS: MD = − 5.02, 95% CI − 8.78; − 1.26; HA: MD = − 2.00, 95% CI − 4.10; 0.09; THA: MD = 0.00 reference; Fig. 7). The global and local consistency assumption of the network meta-analysis model was not met. A net heat plot showed that the inclusion of the RCT by Watson et al. [57] resulted in the greatest heterogeneity. The model without the RCT by Watson et al. [57] only showed different results for the comparison of DHA and THA, i.e., a non-significant difference between the two treatments (MD = − 1.80, 95% CI − 4.06; 0.47).

Harris Hip Score 12 months postoperatively

Data on 405 patients, operated with CS fixation, were pooled from 5 RCTs [31, 32, 38, 51, 57], data on 200 patients, operated with DHS fixation, were pooled from 3 RCTs [31, 42, 57], data on 1098 patients, operated with HA, were pooled from 14 RCTs [27,28,29,30,31,32, 34, 38, 39, 42, 49, 51, 54, 55], and data on 520 patients, operated with THA, were pooled from 9 RCTs [27,28,29, 34, 39, 42, 49, 54, 55]. CS fixation had significantly lower HHS 12 months postoperatively of 7.27 points than THA. DHS fixation had significantly lower HHS 12 months postoperatively of 11.56 points than THA. HA had significantly lower HHS 12 months postoperatively of 4.33 points than THA (CS: MD = − 7.27, 95% CI − 11.92; − 2.61; DHS: MD = − 11.56, 95% CI − 16.36; − 6.76; HA: MD = − 4.33, 95% CI − 7.05; − 1.61; THA: MD = 0.00 reference; Fig. 8). The global and local consistency assumption of the network meta-analysis model was not met. A net heat plot showed that the inclusion of the RCTs by Watson et al. [57] and Davison et al. [30] resulted in the greatest heterogeneity. However, the network model without the two RCTs [30, 57] showed comparable estimates to the network model including all RCTs.

Harris Hip Score 2 years postoperatively

Data on 405 patients, operated with CS fixation, were pooled from 5 RCTs [31, 32, 38, 51, 57], data on 124 patients, operated with DHS fixation, were pooled from 2 RCTs [30, 57], data on 765 patients, operated with HA were pooled from 10 RCTs [28,29,30,31,32, 34, 38, 40, 50, 51], and data on 205 patients, operated with THA, were pooled from 5 RCTs [28, 29, 34, 40, 50]. CS fixation had significantly lower HHS 2 years postoperatively of 5.50 points than THA. DHS fixation had significantly lower HHS 2 years postoperatively of 8.93 points than THA. HA had significantly lower HHS 2 years postoperatively of 3.65 points than THA (CS: MD = − 5.50, 95% CI − 9.98; − 1.03; DHS: MD = − 8.93, 95% CI − 15.08; − 2.78; HA: MD = − 3.65, 95% CI − 6.74; − 0.57; THA: MD = 0.00 reference; Fig. 9).

Harris Hip Score 3–5 years postoperatively

Data on 123 patients, operated with CS fixation, were pooled from 1 RCT [38], data on 355 patients, operated with DHS fixation, were pooled from 2 RCTs [30, 42], data on 1035 patients, operated with HA were pooled from 7 RCTs [28, 30, 34, 38, 42, 54, 55], and data on 343 patients, operated with THA were pooled from 5 RCTs [28, 34, 42, 54, 55]. Two [30, 38] of these RCTs provided data at different time points (HHS 3, 4, and 5 years postoperatively). There was no significant difference in HHS 3–5 years postoperatively between CS fixation and THA. DHS fixation had significantly lower HHS 3–5 years postoperatively of 7.17 points than THA. HA had significantly lower HHS 3–5 years postoperatively of 3.73 points than THA. (CS: MD = − 2.92, 95% CI − 8.76; 2.91; DHS: MD = − 7.17, 95% CI − 11.70; − 2.64; HA: MD = − 3.73, 95% CI − 7.05; − 0.42; THA: MD = 0.00 reference; Fig. 10). The global consistency assumption of the network meta-analysis model was not met. However, the local consistency assumption was met, suggesting no significant design-by-treatment interaction.

Hospital stay

Data on 342 patients, operated with CS fixation, were pooled from 5 RCTs [31, 32, 37, 38, 47], data on 169 patients, operated with DHS fixation, were pooled from 2 RCTs [30, 42], data on 941 patients, operated with HA were pooled from 13 RCTs [28, 30,31,32, 35, 38, 39, 42, 45, 47,48,49, 55], and data on 454 patients, operated with THA were pooled from 9 RCTs [28, 35, 37, 39, 42, 45, 48, 49, 55]. CS fixation had significantly shorter hospital stay of 2.97 days than THA. There was no significant difference in hospital stay between DHS fixation, HA, and THA (CS: MD = − 2.97, 95% CI − 5.33; − 0.61; DHS: MD = 1.30, 95% CI − 1.64; 4.23; HA: MD = − 0.83, 95% CI − 2.40; 0.75; THA: MD = 0.00 reference; Fig. 11). The global consistency assumption of the network meta-analysis model was not met. However local consistency assumption was met, suggesting no significant design-by-treatment interaction.

Reoperation

Data on 493 patients, operated with CS fixation, were pooled from 7 RCTs [31, 32, 36,37,38, 41, 47], data on 210 patients, operated with DHS fixation, were pooled from 3 RCTs [41, 46, 56], data on 1590 patients, operated with HA, were pooled from 15 RCTs [25, 27,28,29, 31,32,33, 35, 38, 39, 46,47,48, 55, 56], and data on 1305 patients, operated with THA, were pooled from 12 RCTs [25, 27,28,29, 33, 35,36,37, 39, 46, 48, 55]. CS fixation had a 9.98 times significantly higher reoperation risk than THA; DHS fixation had a 5.07 times significantly higher reoperation risk than THA; there was no significant difference in reoperation risk between HA and THA (CS: OR = 9.98, 95% CI 4.60; 21.63; DHS: OR = 5.07, 95% CI 2.15; 11.96; HA: OR = 1.60, 95% CI 0.89; 2.89; THA: OR = 1.00 reference; Fig. 12). The analysis of all RCTs together showed that the global consistency assumption of the network meta-analysis model was not met. A net heat plot showed that the inclusion of the RCT by Ravikumar et al. [46] resulted in the greatest heterogeneity. However, the model without the RCT by Ravikumar et al. [46] showed comparable estimates to the model including all RCTs.

Mortality

Data on 427 patients, operated with CS fixation, were pooled from 5 RCTs [31, 32, 36, 41, 52], data on 248 patients, operated with DHS fixation, were pooled from 4 RCTs [41, 42, 46, 56], data on 1533 patients, operated with HA, were pooled from 14 RCTs [25, 27, 31,32,33, 35, 39, 42, 45, 46, 48, 54,55,56], and data on 1358 patients, operated with THA, were pooled from 13 RCTs [25, 27, 33, 35, 36, 39, 42, 45, 46, 48, 52, 54, 55]. There was no significant difference in mortality risk between CS fixation, DHS fixation, HA, and THA (CS: OR = 1.16, 95% CI 0.82; 1.64; DHS: OR = 0.98, 95% CI 0.63; 1.53; HA: OR = 0.89, 95% CI 0.73; 1.10; THA: OR = 1.00 reference; Fig. 13).

Deep vein thrombosis

Data on 413 patients, operated with CS fixation, were pooled from 4 RCTs [32, 36, 37, 44], data on 564 patients, operated with HA, were pooled from 6 RCTs [27, 29, 32, 44, 45, 48], and data on 301 patients, operated with THA, were pooled from 6 RCTs [27, 29, 36, 37, 45, 48]. There was no significant difference in deep vein thrombosis risk between CS fixation, HA, and THA (CS: OR = 1.22, 95% CI 0.26; 5.78; HA: OR = 0.96, 95% CI 0.24; 3.84; THA: OR = 1.00 reference; Fig. 14).

Hematoma

Data on 226 patients, operated with CS fixation, were pooled from 1 RCT [44], data on 21 patients, operated with DHS fixation, were pooled from 1 RCT [56], data on 356 patients, operated with HA, were pooled from 4 RCTs [44, 45, 48, 56], and data on 106 patients, operated with THA, were pooled from 2 RCTs [45, 48]. There was no significant difference in hematoma risk between CS fixation, DHS fixation, HA, and THA (CS: OR = 0.14, 95% CI 0.00; 6.15; DHS: OR = 0.45, 95% CI 0.01; 14.10; HA: OR = 0.43, 95% CI 0.06; 3.00; THA: OR = 1.00 reference; Fig. 15).

Infection

Data on 618 patients, operated with CS fixation, were pooled from 7 RCTs [31, 32, 36,37,38, 44, 47], data on 205 patients, operated with DHS fixation, were pooled from 3 RCTs [30, 46, 56], data on 1819 patients, operated with HA, were pooled from 14 RCTs [27, 29,30,31,32,33, 38, 44,45,46,47,48,49, 56], and data on 1148 patients, operated with THA, were pooled from 9 RCTs [27, 28, 34, 36, 37, 45, 46, 48, 49]. There was no significant difference in infection risk between CS fixation, DHS fixation, HA, and THA (CS: OR = 0.61, 95% CI 0.28; 1.35; DHS: OR = 0.87, 95% CI 0.31; 2.44; HA: OR = 1.01, 95% CI 0.64; 1.58; THA: OR = 1.00 reference; Fig. 16).

Intraoperative fracture

Data on 111 patients, operated with CS fixation, were pooled from 1 RCT [31], data on 943 patients, operated with HA, were pooled from 4 RCTs [29, 31, 33, 48], and data on 832 patients, operated with THA, were pooled from 3 RCTs [29, 33, 48]. There was no significant difference in intraoperative fracture risk between CS fixation, HA, and THA (CS: OR = 2.73, 95% CI 0.11; 70.13; HA: OR = 0.93, 95% CI 0.59; 1.47; THA: OR = 1.00 reference; Fig. 17).

Failure

Data on 468 patients, operated with CS fixation, were pooled from 6 RCTs [31, 32, 36, 38, 41, 47], data on 212 patients, operated with DHS fixation, were pooled from 3 RCTs [30, 41, 56], data on 1423 patients, operated with HA, were pooled from 9 RCTs [30,31,32,33, 38, 47, 48, 55, 56], and data on 937 patients, operated with THA, were pooled from 4 RCTs [33, 36, 48, 55]. CS fixation had a 17.15 times significantly higher failure risk than THA; DHS fixation had a 16.62 times significantly higher failure risk than THA; there was no significant difference in failure risk between HA and THA (CS: OR = 17.15, 95% CI 5.48; 53.70; DHS: OR = 16.62, 95% CI 4.54; 60.82; HA: OR = 1.25, 95% CI 0.49; 3.20; THA: OR = 1.00 reference; Fig. 18).

Dislocation

Data on 240 patients, operated with CS fixation, were pooled from 4 RCTs [32, 36, 37, 47], data on 91 patients, operated with DHS fixation, were pooled from 1 RCT [46], data on 1416 patients, operated with HA, were pooled from 13 RCTs [25, 28, 29, 32, 33, 35, 39, 43, 46,47,48, 50, 55], and data on 1295 patients, operated with THA, were pooled from 13 RCTs [25, 28, 29, 33, 35,36,37, 39, 43, 46, 48, 50, 55]. CS fixation had a 94% significantly lower dislocation risk than THA; DHS fixation had a 98% significantly lower dislocation risk than THA; HA had a 52% significantly lower dislocation risk than THA (CS: OR = 0.06, 95% CI 0.01; 0.30; DHS: OR = 0.02, 95% CI 0.00; 0.33; HA: OR = 0.48, 95% CI 0.28; 0.81; THA: OR = 1.00 reference; Fig. 19).

ANFH

Data on 418 patients, operated with CS fixation, were pooled from 5 RCTs [31, 32, 38, 41, 47], and data on 191 patients, operated with DHS fixation, were pooled from 2 RCTs [30, 41]. There was no significant difference in ANFH risk between DHS and CS (DHS: OR = 1.52, 95% CI 0.54; 4.30; CS: OR = 1.00 reference; Fig. 20).

Nonunion

Data on 328 patients, operated with CS fixation, were pooled from 5 RCTs [31, 36,37,38, 41], and data on 191 patients, operated with DHS fixation, were pooled from 2 RCTs [30, 41]. There was no significant difference in nonunion risk between DHS and CS (DHS: OR = 0.73, 95% CI 0.20; 2.59; CS: OR = 1.00 reference; Fig. 21).

Further results

The 2006 RCT by Baker et al. [26] with a 3-year follow-up, which is continued by the 2011 RCT by Avery et al. [25] with a longer follow-up of almost 9 years provided data on acetabular cup erosion. Three years after implantation, 21 (51.22%) of 41 HAs and 0 of 40 THAs had experienced acetabular erosion [26]. The RCT by Tol et al. [53] provided data on reoperation and dislocation. There were no events for any of the outcome parameters in either the HA or THA groups. Therefore, the data from these two RCTs [26, 53] were presented descriptively and could not be meta-analyzed.

Table 4 shows a ranking of the outcome parameters examined and the rankings generally follow the results of the mixed-effects models for all outcomes except reoperation and failure, which could be explained by the very large uncertainty around the ORs estimated with the models.

Table 4 Ranking of probabilities of operative procedures for being best, second, third best, or worst for all investigated outcomes

The subgroup analysis performed only on displaced femoral neck fractures showed no relevant differences to the network meta-analysis of displaced and non-displaced femoral neck fractures. The forest plots of the subgroup analysis are available in the Additional files 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18. There were no RCTs with non-displaced femoral neck fractures for the following outcomes: deep vein thrombosis, hematoma, and dislocation. Therefore, separate subgroup network meta-analyses were not performed. There was only one RCT with non-displaced femoral neck fractures for the intraoperative fracture outcome. Again, separate subgroup network meta-analyses were not performed. The direct comparisons did not show any relevant differences compared with the results of the network meta-analysis. The results of the direct comparisons are shown in the corresponding forest plots (Figs. 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21). In addition, a sensitivity analysis to assess the effect of our imputation procedure also showed no relevant differences.

Discussion

In our study, using a high-quality network meta-analysis, we attempted to rank 4 different operative procedures in patients with femoral neck fractures, distinguishing between displaced and non-displaced fractures. Our overall findings showed that CS fixation was best in terms of operation time and intraoperative blood loss. For quality of life (EQ-5D) and functional outcome (HHS), THA ranked first and HA ranked second. In contrast, CS fixation had the highest reoperation risk, followed by DHS fixation. THA and HA had a low reoperation risk. There was no significant difference in mortality between the 4 operative procedures. The distinction between displaced and non-displaced fractures showed no relevant differences in our network meta-analysis. However, an analysis specifically considering the Pauwels or Garden classification was not possible. On the basis of these findings, we recommend that prosthetic procedures should be preferred in patients with femoral neck fractures. Accordingly, osteosynthesis should only be considered on a case-by-case basis when known factors such as patient age, fracture morphology, patient orientation and compliance, and the expectation of postoperative functional recovery make the preservation of the femoral head possible and absolutely necessary.

To date, numerous studies and meta-analyses have been conducted on this important but still controversial topic. It is striking that the conclusions and recommendations are not consistent. In summary, there is a fair degree of consensus that HA is associated with better overall patient outcomes compared with internal fixation in patients with femoral neck fractures [6, 8, 15]. However, the comparisons between HA and THA do not allow to draw uniform conclusions. Many meta-analyses consider THA to be superior to HA in patients with femoral neck fractures [11,12,13]. On the other hand, many other meta-analyses find no relevant overall difference in outcome between HA and THA in patients with femoral neck fractures [10, 14, 65,66,67].

Our study showed that CS fixation had the shortest operation time, followed by DHS fixation, HA, and finally THA. The mean operation time of the 4 operative procedures was: 26.6 min. for CS fixation, 38.8 min. for DHS fixation, 63.2 min. for HA, and 84.7 min. for THA. This ranking by operation time reflects the complexity of each operative procedure. The importance of the outcome parameter is controversial. The operation time itself may not be a decisive outcome parameter, but on the other hand, it is known that longer operation times are associated with higher infection rates and tissue trauma [64]. In 89,802 cases of THA, Surace et al. suggested an optimal operation time of approximately 80 min with a lower risk of perioperative complications [64]. Our results are consistent with the literature. Four recent meta-analyses found a significantly shorter operation time (mean: 12.28–20.04 min.) for HA compared with THA [14, 65,66,67]. Two other meta-analyses found a significantly shorter operation time (MD: 2.50–36.22 min.) for internal fixation compared with HA [6, 8]. The subgroup analysis showed no relevant differences between displaced and non-displaced femoral neck fractures for the outcome parameter operation time. This may be due to the fact that the reduction of the fracture is usually performed on the traction table shortly before the start of the internal fixation.

The outcome parameter of intraoperative blood loss showed the same ranking of the operative procedures as for operation time, with CS fixation being the best and THA the worst. The mean intraoperative blood loss of the 4 operative procedures was: 30 ml for CS fixation, 200 ml for DHS fixation, 260 ml for HA, and 400 ml for THA. A correlation between the operation time and intraoperative blood loss seems quite likely, as a recent study has convincingly shown [68]. Our results are in line with the literature. Three recent meta-analyses found a significantly lower blood loss (MD: 45.63–69.10 ml) for HA compared with THA [65,66,67]. Another meta-analysis found a significantly less blood loss (MD: 165.84 ml) with CS fixation compared with HA [6]. Subgroup analysis showed no relevant differences in intraoperative blood loss between displaced and non-displaced femoral neck fractures. Our efforts to maintain stable hemoglobin levels begin intraoperatively and should be continued postoperatively, as a recent study has shown [69]. In the management of blood loss, orthogeriatric care may be of great benefit in the elderly population studied [69].

Our study showed that THA had the best EQ-5D score, followed by HA. CS fixation had the worst EQ-5D score. This ranking was repeated in all three intervals examined (EQ-5D 3–4 months, 12 months, and 2 years postoperatively) at which EQ-5D was recorded. None of the RCTs reported EQ-5D scores for DHS fixation. The mean EQ-5D 3–4 months postoperatively of the 3 operative procedures was: 0.5 points for CS fixation, 0.6 points for HA, 0.7 points for THA. The mean EQ-5D 12 months and 2 years postoperatively of the 3 operative procedures were: 0.6 points for CS fixation, 0.6 points for HA, 0.7 points for THA. The EQ-5D is a generic measuring instrument that uses a standardized, preference-based method to assess health status [18]. In a questionnaire, developed by the EuroQol Group, respondents rate their health on scale from 0 (very poor) to 1 (best possible). Our results are in line with the literature. Two recent meta-analyses found a significantly higher EQ-5D score (MD: 0.13 points) for THA compared with HA [10, 14]. Another meta-analysis found no significant difference in EQ-5D between THA and HA up to 1 year after surgery [65]. Our subgroup analysis showed no relevant differences in EQ-5D between displaced and non-displaced femoral neck fractures.

The HHS was developed to assess hip function after surgery [19]. The score ranges from 0 (very poor) to 100 (best possible) points. It accumulates points from the assessment of four aspects: pain, function, degree of deformity, and range of motion of the hip. Hip function was assessed at regular intervals after surgery (HHS ≤ 6 months, 12 months, 2 years, and 3–5 years postoperatively), providing information on short-, middle-, and long-term functional outcomes of the hip after surgery. Notably, THA had the best place in the ranking at every time point measured. HA was ranked second best at almost every time point measured except for HHS 3–5 years postoperatively when CS fixation was ranked second best. In all other cases, the worst and third place was taken by CS fixation and DHS fixation. The mean HHS ≤ 6 months postoperatively of the 4 operative procedures were: 64.5 points for CS fixation, 67.0 points for DHS fixation, 72.5 points for HA, and 78.4 points for THA. The mean HHS 12 months postoperatively of the 4 operative procedures were: 69.8 points for CS fixation, 71.1 points for DHS fixation, 75.2 points for HA, 81.7 points for THA. The mean HHS 2 years postoperatively of the 4 operative procedures was: 71.3 points for CS fixation, 71.8 for DHS fixation, 74.4 points for HA, 80.2 points for THA. The mean HHS 3–5 years postoperatively of the 4 operative procedures was: 80.9 points for CS fixation, 71.1 for DHS fixation, 75.1 points for HA, 81.0 points for THA. The literature shows similar results. Three recent meta-analyses found a significantly higher HHS (MD: 5.05–6.03 points) for THA compared with HA [10, 11, 65]. Another meta-analysis found no significant difference in HHS between HA and THA [67]. In contrast to our results, another meta-analysis showed no significant difference in HHS 1 and 2 years postoperatively between internal fixation and HA [8]. This meta-analysis was limited to patients with non-displaced femoral neck fractures and included data from only 3 studies, one of which was not an RCT [8]. Furthermore, this meta-analysis did not differentiate between CS fixation and DHS fixation in the internal fixation group [8]. Another meta-analysis of non-displaced femoral neck fractures showed a significantly higher HHS 6 months postoperatively for HA compared with CS fixation (MD: 5.05 points) [6]. The same study found no significant difference in HHS 1 and 2 years postoperatively between CS fixation and HA [6]. The importance of the outcome parameter HHS for assessing the functional outcome of hip operations is undisputed. A recent study found that 1 year after hip fracture in elderly patients, significant loss of muscle mass was common, with impaired functional recovery [70]. One treatment approach in this context may be targeted preservation of muscle mass to improve the prognosis in these patients [70]. However, it should be noted that the minimal clinically important difference (MCID) for HHS has been reported in the literature to be no less than 7.9 points [71,72,73,74]. In our study, a significantly higher HHS difference than the MCID was found in only two cases. DHS had significantly lower HHS 12 months postoperatively of 11.56 points compared with THA, and significantly lower HHS 2 years postoperatively of 8.93 points compared with THA.

Our study showed that CS fixation had the shortest hospital stay with 8.4 days. HA was the second best, followed by THA, and finally, DHS fixation, although the differences between these three operative procedures were not significant. A recent meta-analysis found no significant difference in hospital stay between THA and HA [14]. Two other meta-analyses showed a significantly longer hospital stay (MD: 0.47–2.36 days) for HA than for THA [65, 66]. Another meta-analysis, including only non-displaced femoral neck fractures, found a significantly shorter hospital stay (MD: 0.80 days) for internal fixation compared with HA [8]. Another meta-analysis, including only non-displaced femoral neck fractures, found a significantly shorter hospital stay (MD: 3.32 days) for CS fixation compared with HA [6].

Our study showed that CS fixation had the highest reoperation risk with 39.15%, followed by DHS with 27.62%, followed by HA with 9.11% and THA with 6.36%. CS fixation had the highest failure risk with 21.79%, followed by DHS fixation with 15.09%, followed by HA with 1.26% and THA with 0.85%. The difference between HA and THA was not significant. HA had a significantly lower dislocation risk than THA, with rates of 3.11% for HA and 6.33% for THA. The following outcome parameters did not show relevant differences: mortality, deep vein thrombosis, hematoma, infection, intraoperative fracture, ANFH, and nonunion. The subgroup analysis showed no relevant differences between displaced and non-displaced femoral neck fractures for the outcome parameters reoperation, mortality, infection, intraoperative fracture, failure, ANFH, and nonunion. In general, our results were consistent with the literature. We compared our findings with 5 meta-analyses that examined the differences between HA and THA in patients with displaced femoral neck fractures [10,11,12,13,14]. A meta-analysis showed no significant difference in mortality risk between HA (rate: 15.13%) and THA (rate: 13.48%) [10]. The authors Burgers et al. found a 2.53 times significantly higher dislocation risk for THA (rate: 8.94%) compared with HA (rate: 3.40%) [10]. Another meta-analysis showed a 62% significantly lower dislocation risk for HA (rate: 2.70%) compared with THA (rate: 8.12%) [11]. The authors Lewis et al. found no significant difference in infection risk between HA and THA [11]. They also found a 1.54 times significantly higher reoperation risk for HA (rate: 8.76%) compared with THA (rate: 5.72%) [11]. Another meta-analysis found no significant difference in mortality risk between HA (rate: 34.67%) and THA (rate: 30.49%) [12]. The authors Liao et al. found a 60% significantly lower reoperation risk for THA (rate: 5.82%) compared with HA (rate: 14.33%) [12].They also found a 2.02 times significantly higher dislocation risk for THA (rate: 10.67%) compared with HA (rate: 5.18%) [12]. They found no significant difference in infection risk between HA and THA [12]. Another meta-analysis showed a 25% significantly lower mortality risk for THA (rate: 12.22%) compared with HA (rate: 15.37%) [13]. The authors Peng et al. found a 54% significantly lower dislocation risk for THA (rate: 4.15%) compared with HA (rate: 9.22%) [13], which is contradictory to the findings in the specialist literature. Furthermore, they found no significant difference in infection, reoperation, and thromboembolic risk between THA and HA [13]. Another meta-analysis showed a 3.31 times significantly higher reoperation risk for HA (rate: 12.28%) compared with THA (rate: 3.92%) [14]. The authors Wang et al. found no significant difference in infection and dislocation risk [14].

We also compared our findings with 3 meta-analyses that examined the differences in HA and THA in patients with displaced and non-displaced femoral neck fractures [65,66,67]. One meta-analysis found no significant difference in mortality risk between HA and THA [66]. Another meta-analysis comparing dual-mobility THA with HA showed a 3.60 times significantly higher dislocation risk for HA [67]. The authors Ma et al. found a 2.06 times significantly higher reoperation risk for HA compared with THA [67]. Another meta-analysis found no significant differences in thromboembolic, infection, revision, intraoperative fracture, failure, and mortality (1 year postoperatively) risks between HA and THA [65]. The authors Tang et al. found a 1.90 times significantly higher dislocation risk for THA compared with HA [65]. In general, interdisciplinary orthogeriatric care can help us reduce complication rates and mortality [75]. Important perioperative aspects include pain and fluid management, early mobilization, and delirium prevention [75].

A meta-analysis comparing CS fixation with DHS fixation in patients with displaced and non-displaced femoral neck fractures showed a 1.44 times significantly higher reoperation risk for CS fixation (rate: 33.02%) compared with DHS fixation (rate: 21.77%), and a 2.28 times significantly higher failure risk for CS fixation (rate: 13.04%) compared with THA (rate: 5.49%) [76]. Another meta-analysis showed a 4.49 times significantly higher reoperation risk for internal fixation (CS fixation and DHS fixation) compared with HA [8]. The authors Ma et al. found no significant difference in mortality risk (1 year postoperatively) between internal fixation and HA [8]. Another meta-analysis showed a 4.88 times significantly higher reoperation risk for CS fixation compared with HA [6]. The authors Xu et al. found no significant difference in mortality risk between CS fixation and HA [6].

Despite our results that are in line with the literature, we identified several limitations: (1) There was considerable heterogeneity between individual studies for some outcome parameters, which could affect the final results. (2) In some cases, the quality assessment of the studies produced questionable results. (3) Operative skill of the surgeon, intraoperative warming, injection of local anaesthetics and tranexamic acid, use of bone cement, or type of implant could be considered as confounding factors, which may affect the results to some extent. (4) The number of RCTs reporting patients with non-displaced femoral neck fractures was low so some outcome parameters could not be considered. (5) The distinction between closed and open reduction was not considered in the CS fixation and DHS fixation groups.

Conclusions

In our cohort of patient with displaced and non-displaced femoral neck fractures, the more important outcome parameters such as HHS, EQ-5D, and reoperation risk showed an advantage of THA and HA compared with CS fixation and DHS fixation. Based on these findings, we recommend that hip arthroplasty should be preferred and internal fixation of femoral neck fractures should only be considered in individual cases.

Availability of data and materials

Raw data extraction sheet is available in the Additional file 1.

Abbreviations

ANFH:

Avascular necrosis of the femoral head

CS:

Cannulated screw(s)

DHS:

Dynamic hip screw

EQ-5D:

Not an abbreviation

HA:

Hemiarthroplasty

HHS:

Harris Hip Score

ITT:

Intention to treat

MCID:

Minimal clinically important difference

RCT:

Randomized controlled trial

THA:

Total hip arthroplasty

SUCRA:

Surface under the cumulative ranking curve

References

  1. Stevens JA, Rudd RA. The impact of decreasing U.S. hip fracture rates on future hip fracture estimates. Osteoporos Int. 2013;24(10):2725–58. https://doi.org/10.1007/s00198-013-2375-9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Marks R, Allegrante JP, Ronald MacKenzie C, Lane JM. Hip fractures among the elderly: causes, consequences and control. Ageing Res Rev. 2003;2(1):57–93. https://doi.org/10.1016/s1568-1637(02)00045-4.

    Article  PubMed  Google Scholar 

  3. Garden RS. Stability and union in subcapital fractures of the femur. J Bone Joint Surg Br. 1964;46:630–47.

    Article  CAS  PubMed  Google Scholar 

  4. Nandi S. Revisiting Pauwels’ classification of femoral neck fractures. World J Orthop. 2021;12(11):811–5. https://doi.org/10.5312/wjo.v12.i11.811.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Florschutz AV, Langford JR, Haidukewych GJ, Koval KJ. Femoral neck fractures: current management. J Orthop Trauma. 2015;29(3):121–9. https://doi.org/10.1097/BOT.0000000000000291.

    Article  PubMed  Google Scholar 

  6. Xu WN, Xue QY. Long-term efficacy of screw fixation vs hemiarthroplasty for undisplaced femoral neck fracture in patients over 65 years of age: a systematic review and meta-analysis. Orthop Surg. 2021;13(1):3–13. https://doi.org/10.1111/os.12910.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Bigoni M, Turati M, Leone G, Caminita AD, D’Angelo F, Munegato D, Zatti G. Internal fixation of intracapsular femoral neck fractures in elderly patients: mortality and reoperation rate. Aging Clin Exp Res. 2020;32(6):1173–8. https://doi.org/10.1007/s40520-019-01237-z.

    Article  PubMed  Google Scholar 

  8. Ma HH, Chou TA, Tsai SW, Chen CF, Wu PK, Chen WM. Outcomes of internal fixation versus hemiarthroplasty for elderly patients with an undisplaced femoral neck fracture: a systematic review and meta-analysis. J Orthop Surg Res. 2019;14(1):320. https://doi.org/10.1186/s13018-019-1377-5.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Ramadanov N, Toma I, Herkner H, Klein R, Behringer W, Matthes G. Factors that influence the complications and outcomes of femoral neck fractures treated by cannulated screw fixation. Sci Rep. 2020;10(1):758. https://doi.org/10.1038/s41598-020-57696-2.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Burgers PT, Van Geene AR, Van den Bekerom MP, Van Lieshout EM, Blom B, Aleem IS, Bhandari M, Poolman RW. Total hip arthroplasty versus hemiarthroplasty for displaced femoral neck fractures in the healthy elderly: a meta-analysis and systematic review of randomized trials. Int Orthop. 2012;36(8):1549–60. https://doi.org/10.1007/s00264-012-1569-7.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Lewis DP, Wæver D, Thorninger R, Donnelly WJ. Hemiarthroplasty vs total hip arthroplasty for the management of displaced neck of femur fractures: a systematic review and meta-analysis. J Arthroplasty. 2019;34(8):1837-1843.e2. https://doi.org/10.1016/j.arth.2019.03.070.

    Article  PubMed  Google Scholar 

  12. Liao L, Zhao J, Su W, Ding X, Chen L, Luo S. A meta-analysis of total hip arthroplasty and hemiarthroplasty outcomes for displaced femoral neck fractures. Arch Orthop Trauma Surg. 2012;132(7):1021–9. https://doi.org/10.1007/s00402-012-1485-8.

    Article  PubMed  Google Scholar 

  13. Peng W, Bi N, Zheng J, Xi N. Does total hip arthroplasty provide better outcomes than hemiarthroplasty for the femoral neck fracture? A systematic review and meta-analysis. Chin J Traumatol. 2020;23(6):356–62. https://doi.org/10.1016/j.cjtee.2020.09.005.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Wang F, Zhang H, Zhang Z, Ma C, Feng X. Comparison of bipolar hemiarthroplasty and total hip arthroplasty for displaced femoral neck fractures in the healthy elderly: a meta-analysis. BMC Musculoskelet Disord. 2015;16:229. https://doi.org/10.1186/s12891-015-0696-x.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Zhang BF, Wang PF, Huang H, Cong YX, Wang H, Zhuang Y. Interventions for treating displaced intracapsular femoral neck fractures in the elderly: a Bayesian network meta-analysis of randomized controlled trials. Sci Rep. 2017;7(1):13103. https://doi.org/10.1038/s41598-017-13377-1.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Stoffel K, Zderic I, Gras F, Sommer C, Eberli U, Mueller D, Oswald M, Gueorguiev B. Biomechanical evaluation of the femoral neck system in unstable Pauwels III femoral neck fractures: a comparison with the dynamic hip screw and cannulated screws. J Orthop Trauma. 2017;31(3):131–7. https://doi.org/10.1097/BOT.0000000000000739.

    Article  PubMed  Google Scholar 

  17. Lim EJ, Shon HC, Cho JW, Oh JK, Kim J, Kim CH. Dynamic hip screw versus cannulated cancellous screw in Pauwels type II or type III femoral neck fracture: a systematic review and meta-analysis. J Pers Med. 2021;11(10):1017. https://doi.org/10.3390/jpm11101017.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Hutton B, Salanti G, Caldwell DM, Chaimani A, Schmid CH, Cameron C, Ioannidis JP, Straus S, Thorlund K, Jansen JP, Mulrow C, Catalá-López F, Gøtzsche PC, Dickersin K, Boutron I, Altman DG, Moher D. The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations. Ann Intern Med. 2015;162(11):777–84. https://doi.org/10.7326/M14-2385.

    Article  PubMed  Google Scholar 

  19. Brooks R, Boye KS, Slaap B. EQ-5D: a plea for accurate nomenclature. J Patient Rep Outcomes. 2020;4(1):52. https://doi.org/10.1186/s41687-020-00222-9.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Harris WH. Traumatic arthritis of the hip after dislocation and acetabular fractures: treatment by mold arthroplasty. An end-result study using a new method of result evaluation. J Bone Joint Surg Am. 1969;51:737–55. https://doi.org/10.2106/00004623-196951040-00012.

    Article  CAS  PubMed  Google Scholar 

  21. Sterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, Cates CJ, Cheng HY, Corbett MS, Eldridge SM, Emberson JR, Hernán MA, Hopewell S, Hróbjartsson A, Junqueira DR, Jüni P, Kirkham JJ, Lasserson T, Li T, McAleenan A, Reeves BC, Shepperd S, Shrier I, Stewart LA, Tilling K, White IR, Whiting PF, Higgins JPT. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898. https://doi.org/10.1136/bmj.l4898.

    Article  PubMed  Google Scholar 

  22. Rücker G, Schwarzer G. Ranking treatments in frequentist network meta-analysis works without resampling methods. BMC Med Res Methodol. 2015;15:58. https://doi.org/10.1186/s12874-015-0060-8.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Neupane B, Richer D, Bonner AJ, Kibret T, Beyene J. Network meta-analysis using R: a review of currently available automated packages. PLoS ONE. 2014;9(12):e115065. https://doi.org/10.1371/journal.pone.0115065.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Higgins JPT, Deeks J, Altman D. Special topics in statistics. In: Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions version 5.1.0. The Cochrane Collaboration; 2011. chap 16. https://handbook-5-1.cochrane.org/chapter_16/16_1_3_1imputing_standard_deviations.htm.

  25. Avery PP, Baker RP, Walton MJ, Rooker JC, Squires B, Gargan MF, Bannister GC. Total hip replacement and hemiarthroplasty in mobile, independent patients with a displaced intracapsular fracture of the femoral neck: a seven- to ten-year follow-up report of a prospective randomised controlled trial. J Bone Joint Surg Br. 2011;93(8):1045–8. https://doi.org/10.1302/0301-620X.93B8.27132.

    Article  CAS  PubMed  Google Scholar 

  26. Baker RP, Squires B, Gargan MF, Bannister GC. Total hip arthroplasty and hemiarthroplasty in mobile, independent patients with a displaced intracapsular fracture of the femoral neck. A randomized, controlled trial. J Bone Joint Surg Am. 2006;88(12):2583–9. https://doi.org/10.2106/JBJS.E.01373.

    Article  CAS  PubMed  Google Scholar 

  27. Blomfeldt R, Törnkvist H, Eriksson K, Söderqvist A, Ponzer S, Tidermark J. A randomised controlled trial comparing bipolar hemiarthroplasty with total hip replacement for displaced intracapsular fractures of the femoral neck in elderly patients. J Bone Joint Surg Br. 2007;89(2):160–5. https://doi.org/10.1302/0301-620X.89B2.18576.

    Article  CAS  PubMed  Google Scholar 

  28. Cadossi M, Chiarello E, Savarino L, Tedesco G, Baldini N, Faldini C, Giannini S. A comparison of hemiarthroplasty with a novel polycarbonate-urethane acetabular component for displaced intracapsular fractures of the femoral neck: a randomised controlled trial in elderly patients. Bone Joint J. 2013;95-B(5):609–15. https://doi.org/10.1302/0301-620X.95B5.31083.

    Article  CAS  PubMed  Google Scholar 

  29. Chammout G, Kelly-Pettersson P, Hedbeck CJ, Stark A, Mukka S, Sköldenberg O. HOPE-trial: hemiarthroplasty compared with total hip arthroplasty for displaced femoral neck fractures in octogenarians: a randomized controlled trial. JB JS Open Access. 2019;4(2):e0059. https://doi.org/10.2106/JBJS.OA.18.00059.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Davison JN, Calder SJ, Anderson GH, Ward G, Jagger C, Harper WM, Gregg PJ. Treatment for displaced intracapsular fracture of the proximal femur. A prospective, randomised trial in patients aged 65 to 79 years. J Bone Joint Surg Br. 2001;83(2):206–12. https://doi.org/10.1302/0301-620x.83b2.11128.

    Article  CAS  PubMed  Google Scholar 

  31. Dolatowski FC, Frihagen F, Bartels S, Opland V, Šaltytė Benth J, Talsnes O, Hoelsbrekken SE, Utvåg SE. Screw fixation versus hemiarthroplasty for nondisplaced femoral neck fractures in elderly patients: a multicenter randomized controlled trial. J Bone Joint Surg Am. 2019;101(2):136–44. https://doi.org/10.2106/JBJS.18.00316.

    Article  PubMed  Google Scholar 

  32. Frihagen F, Nordsletten L, Madsen JE. Hemiarthroplasty or internal fixation for intracapsular displaced femoral neck fractures: randomised controlled trial. BMJ. 2007;335(7632):1251–4. https://doi.org/10.1136/bmj.39399.456551.25.

    Article  PubMed  PubMed Central  Google Scholar 

  33. HEALTH Investigators, Bhandari M, Einhorn TA, Guyatt G, Schemitsch EH, Zura RD, Sprague S, Frihagen F, Guerra-Farfán E, Kleinlugtenbelt YV, Poolman RW, Rangan A, Bzovsky S, Heels-Ansdell D, Thabane L, Walter SD, Devereaux PJ. Total hip arthroplasty or hemiarthroplasty for hip fracture. N Engl J Med. 2019;381(23):2199–208. https://doi.org/10.1056/NEJMoa1906190.

    Article  Google Scholar 

  34. Hedbeck CJ, Enocson A, Lapidus G, Blomfeldt R, Törnkvist H, Ponzer S, Tidermark J. Comparison of bipolar hemiarthroplasty with total hip arthroplasty for displaced femoral neck fractures: a concise four-year follow-up of a randomized trial. J Bone Joint Surg Am. 2011;93(5):445–50. https://doi.org/10.2106/JBJS.J.00474.

    Article  PubMed  Google Scholar 

  35. Iorio R, Iannotti F, Mazza D, Speranza A, Massafra C, Guzzini M, D’Arrigo C, Ferretti A. Is dual cup mobility better than hemiarthroplasty in patients with dementia and femoral neck fracture? A randomized controlled trial. SICOT J. 2019;5:38. https://doi.org/10.1051/sicotj/2019035.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Johansson T, Jacobsson SA, Ivarsson I, Knutsson A, Wahlström O. Internal fixation versus total hip arthroplasty in the treatment of displaced femoral neck fractures: a prospective randomized study of 100 hips. Acta Orthop Scand. 2000;71(6):597–602. https://doi.org/10.1080/000164700317362235.

    Article  CAS  PubMed  Google Scholar 

  37. Jónsson B, Sernbo I, Carlsson A, Fredin H, Johnell O. Social function after cervical hip fracture. A comparison of hook-pins and total hip replacement in 47 patients. Acta Orthop Scand. 1996;67(5):431–4. https://doi.org/10.3109/17453679608996662.

    Article  PubMed  Google Scholar 

  38. Lu Q, Tang G, Zhao X, Guo S, Cai B, Li Q. Hemiarthroplasty versus internal fixation in super-aged patients with undisplaced femoral neck fractures: a 5-year follow-up of randomized controlled trial. Arch Orthop Trauma Surg. 2017;137(1):27–35. https://doi.org/10.1007/s00402-016-2591-9.

    Article  PubMed  Google Scholar 

  39. Macaulay W, Nellans KW, Iorio R, Garvin KL, Healy WL, Rosenwasser MP, DFACTO Consortium. Total hip arthroplasty is less painful at 12 months compared with hemiarthroplasty in treatment of displaced femoral neck fracture. HSS J. 2008;4(1):48–54. https://doi.org/10.1007/s11420-007-9061-4.

    Article  PubMed  Google Scholar 

  40. Macaulay W, Nellans KW, Garvin KL, Iorio R, Healy WL, Rosenwasser MP, other members of the DFACTO Consortium. Prospective randomized clinical trial comparing hemiarthroplasty to total hip arthroplasty in the treatment of displaced femoral neck fractures: winner of the Dorr Award. J Arthroplast. 2008;23(6 Suppl 1):2–8. https://doi.org/10.1016/j.arth.2008.05.013.

    Article  Google Scholar 

  41. Mjørud J, Skaro O, Solhaug JH, Thorngren KG. A randomised study in all cervical hip fractures osteosynthesis with Hansson hook-pins versus AO-screws in 199 consecutive patients followed for two years. Injury. 2006;37(8):768–77. https://doi.org/10.1016/j.injury.2006.01.001.

    Article  PubMed  Google Scholar 

  42. Mouzopoulos G, Stamatakos M, Arabatzi H, Vasiliadis G, Batanis G, Tsembeli A, Tzurbakis M, Safioleas M. The four-year functional result after a displaced subcapital hip fracture treated with three different surgical options. Int Orthop. 2008;32(3):367–73. https://doi.org/10.1007/s00264-007-0321-1.

    Article  PubMed  Google Scholar 

  43. Narayan KK, George T. Functional outcome of fracture neck of femur treated with total hip replacement versus bipolar arthroplasty in a South Asian population. Arch Orthop Trauma Surg. 2006;126(8):545–8. https://doi.org/10.1007/s00402-006-0182-x.

    Article  PubMed  Google Scholar 

  44. Parker MJ, Khan RJ, Crawford J, Pryor GA. Hemiarthroplasty versus internal fixation for displaced intracapsular hip fractures in the elderly. A randomised trial of 455 patients. J Bone Joint Surg Br. 2002;84(8):1150–5. https://doi.org/10.1302/0301-620x.84b8.13522.

    Article  CAS  PubMed  Google Scholar 

  45. Parker MJ, Cawley S. Treatment of the displaced intracapsular fracture for the “fitter” elderly patients: a randomised trial of total hip arthroplasty versus hemiarthroplasty for 105 patients. Injury. 2019;50(11):2009–13. https://doi.org/10.1016/j.injury.2019.09.018.

    Article  PubMed  Google Scholar 

  46. Ravikumar KJ, Marsh G. Internal fixation versus hemiarthroplasty versus total hip arthroplasty for displaced subcapital fractures of femur–13 year results of a prospective randomised study. Injury. 2000;31(10):793–7. https://doi.org/10.1016/s0020-1383(00)00125-x.

    Article  CAS  PubMed  Google Scholar 

  47. Rödén M, Schön M, Fredin H. Treatment of displaced femoral neck fractures: a randomized minimum 5-year follow-up study of screws and bipolar hemiprostheses in 100 patients. Acta Orthop Scand. 2003;74(1):42–4. https://doi.org/10.1080/00016470310013635.

    Article  PubMed  Google Scholar 

  48. Schleicher I, Kordelle J, Jürgensen I, Haas H, Melzer C. Die Schenkelhalsfraktur beim alten Menschen-Bipolare Hemiendoprothese vs. Totalendoprothese [Femoral neck fractures in the elderly—bipolar hemiarthroplasty in total hip replacement]. Unfallchirurg. 2003;106(6):467–71. https://doi.org/10.1007/s00113-003-0597-6. (German).

    Article  CAS  PubMed  Google Scholar 

  49. Sharma V, Awasthi B, Kumar K, Kohli N, Katoch P. Outcome Analysis of hemiarthroplasty vs. total hip replacement in displaced femoral neck fractures in the elderly. J Clin Diagn Res. 2016;10(5):RC11–3. https://doi.org/10.7860/JCDR/2016/18638.7877.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Sonaje JC, Meena PK, Bansiwal RC, Bobade SS. Comparison of functional outcome of bipolar hip arthroplasty and total hip replacement in displaced femoral neck fractures in elderly in a developing country: a 2-year prospective study. Eur J Orthop Surg Traumatol. 2018;28(3):493–8. https://doi.org/10.1007/s00590-017-2057-y.

    Article  PubMed  Google Scholar 

  51. Støen RØ, Lofthus CM, Nordsletten L, Madsen JE, Frihagen F. Randomized trial of hemiarthroplasty versus internal fixation for femoral neck fractures: no differences at 6 years. Clin Orthop Relat Res. 2014;472(1):360–7. https://doi.org/10.1007/s11999-013-3245-7.

    Article  PubMed  Google Scholar 

  52. Tidermark J, Ponzer S, Svensson O, Söderqvist A, Törnkvist H. Internal fixation compared with total hip replacement for displaced femoral neck fractures in the elderly. A randomised, controlled trial. J Bone Joint Surg Br. 2003;85(3):380–8. https://doi.org/10.1302/0301-620x.85b3.13609.

    Article  CAS  PubMed  Google Scholar 

  53. Tol MC, van den Bekerom MP, Sierevelt IN, Hilverdink EF, Raaymakers EL, Goslings JC. Hemiarthroplasty or total hip arthroplasty for the treatment of a displaced intracapsular fracture in active elderly patients: 12-year follow-up of randomised trial. Bone Joint J. 2017;99-B(2):250–4. https://doi.org/10.1302/0301-620X.99B2.BJJ-2016-0479.R1.

    Article  CAS  PubMed  Google Scholar 

  54. Ukaj S, Zhuri O, Ukaj F, Podvorica V, Grezda K, Caton J, Prudhon JL, Krasniqi S. Dual mobility acetabular cup versus hemiarthroplasty in treatment of displaced femoral neck fractures in elderly patients: comparative study and results at minimum 3-year follow-up. Geriatr Orthop Surg Rehabil. 2019;10:2151459319848610. https://doi.org/10.1177/2151459319848610.

    Article  PubMed  PubMed Central  Google Scholar 

  55. van den Bekerom MP, Hilverdink EF, Sierevelt IN, Reuling EM, Schnater JM, Bonke H, Goslings JC, van Dijk CN, Raaymakers EL. A comparison of hemiarthroplasty with total hip replacement for displaced intracapsular fracture of the femoral neck: a randomised controlled multicentre trial in patients aged 70 years and over. J Bone Joint Surg Br. 2010;92(10):1422–8. https://doi.org/10.1302/0301-620X.92B10.24899.

    Article  PubMed  Google Scholar 

  56. van Vugt AB, Oosterwijk WM, Goris RJ. Osteosynthesis versus endoprosthesis in the treatment of unstable intracapsular hip fractures in the elderly. A randomised clinical trial. Arch Orthop Trauma Surg. 1993;113(1):39–45. https://doi.org/10.1007/BF00440593.

    Article  PubMed  Google Scholar 

  57. Watson A, Zhang Y, Beattie S, Page RS. Prospective randomized controlled trial comparing dynamic hip screw and screw fixation for undisplaced subcapital hip fractures. ANZ J Surg. 2013;83(9):679–83. https://doi.org/10.1111/j.1445-2197.2012.06256.x.

    Article  PubMed  Google Scholar 

  58. Okike K, Udogwu UN, Isaac M, Sprague S, Swiontkowski MF, Bhandari M, Slobogean GP, FAITH Investigators. Not all Garden-I and II femoral neck fractures in the elderly should be fixed: effect of posterior tilt on rates of subsequent arthroplasty. J Bone Joint Surg Am. 2019;101(20):1852–9. https://doi.org/10.2106/JBJS.18.01256.

    Article  PubMed  Google Scholar 

  59. Waaler Bjørnelv GM, Frihagen F, Madsen JE, Nordsletten L, Aas E. Hemiarthroplasty compared to internal fixation with percutaneous cannulated screws as treatment of displaced femoral neck fractures in the elderly: cost-utility analysis performed alongside a randomized, controlled trial. Osteoporos Int. 2012;23(6):1711–9. https://doi.org/10.1007/s00198-011-1772-1.

    Article  PubMed  Google Scholar 

  60. Sikorski JM, Barrington R. Internal fixation versus hemiarthroplasty for the displaced subcapital fracture of the femur. A prospective randomised study. J Bone Joint Surg Br. 1981;63-B(3):357–61. https://doi.org/10.1302/0301-620X.63B3.7263746.

    Article  CAS  PubMed  Google Scholar 

  61. Neander G, Adolphson P, von Sivers K, Dahlborn M, Dalén N. Bone and muscle mass after femoral neck fracture. A controlled quantitative computed tomography study of osteosynthesis versus primary total hip arthroplasty. Arch Orthop Trauma Surg. 1997;116(8):470–4. https://doi.org/10.1007/BF00387579.

    Article  CAS  PubMed  Google Scholar 

  62. Söreide O, Mölster A, Raugstad TS. Internal fixation versus primary prosthetic replacement in acute femoral neck fractures: a prospective, randomized clinical study. Br J Surg. 1979;66(1):56–60. https://doi.org/10.1002/bjs.1800660118(tooold).

    Article  PubMed  Google Scholar 

  63. Griffin XL, Parsons N, Achten J, Costa ML. The Targon femoral neck hip screw versus cannulated screws for internal fixation of intracapsular fractures of the hip: a randomised controlled trial. Bone Joint J. 2014;96-B(5):652–7. https://doi.org/10.1302/0301-620X.96B5.33391.

    Article  CAS  PubMed  Google Scholar 

  64. Surace P, Sultan AA, George J, Samuel LT, Khlopas A, Molloy RM, Stearns KL, Mont MA. The association between operative time and short-term complications in total hip arthroplasty: an analysis of 89,802 surgeries. J Arthroplast. 2019;34(3):426–32. https://doi.org/10.1016/j.arth.2018.11.015.

    Article  Google Scholar 

  65. Tang X, Wang D, Liu Y, Chen J, Zhou Z, Li P, Ning N. The comparison between total hip arthroplasty and hemiarthroplasty in patients with femoral neck fractures: a systematic review and meta-analysis based on 25 randomized controlled trials. J Orthop Surg Res. 2020;15(1):596. https://doi.org/10.1186/s13018-020-02122-6.

    Article  PubMed  PubMed Central  Google Scholar 

  66. Li X, Luo J. Hemiarthroplasty compared to total hip arthroplasty for the treatment of femoral neck fractures: a systematic review and meta-analysis. J Orthop Surg Res. 2021;16(1):172. https://doi.org/10.1186/s13018-020-02186-4.

    Article  PubMed  PubMed Central  Google Scholar 

  67. Ma HH, Chou TA, Pai FY, Tsai SW, Chen CF, Wu PK, Chen WM. Outcomes of dual-mobility total hip arthroplasty versus bipolar hemiarthroplasty for patients with femoral neck fractures: a systematic review and meta-analysis. J Orthop Surg Res. 2021;16(1):152. https://doi.org/10.1186/s13018-021-02316-6.

    Article  PubMed  PubMed Central  Google Scholar 

  68. Ross D, Erkocak O, Rasouli MR, Parvizi J. Operative time directly correlates with blood loss and need for blood transfusion in total joint arthroplasty. Arch Bone Joint Surg. 2019;7(3):229–34.

    PubMed  PubMed Central  Google Scholar 

  69. Quaranta M, Miranda L, Oliva F, Migliorini F, Pezzuti G, Maffulli N. Haemoglobin and transfusions in elderly patients with hip fractures: the effect of a dedicated orthogeriatrician. J Orthop Surg Res. 2021;16(1):387. https://doi.org/10.1186/s13018-021-02524-0.

    Article  PubMed  PubMed Central  Google Scholar 

  70. Chen YP, Kuo YJ, Hung SW, Wen TW, Chien PC, Chiang MH, Maffulli N, Lin CY. Loss of skeletal muscle mass can be predicted by sarcopenia and reflects poor functional recovery at one year after surgery for geriatric hip fractures. Injury. 2021;52(11):3446–52. https://doi.org/10.1016/j.injury.2021.08.007.

    Article  PubMed  Google Scholar 

  71. Singh JA, Schleck C, Harmsen S, Lewallen D. Clinically important improvement thresholds for Harris Hip Score and its ability to predict revision risk after primary total hip arthroplasty. BMC Musculoskelet Disord. 2016;17:256. https://doi.org/10.1186/s12891-016-1106-8.

    Article  PubMed  PubMed Central  Google Scholar 

  72. Kemp JL, Collins NJ, Roos EM, Crossley KM. Psychometric properties of patient-reported outcome measures for hip arthroscopic surgery. Am J Sports Med. 2013;41(9):2065–73. https://doi.org/10.1177/0363546513494173.

    Article  PubMed  Google Scholar 

  73. van der Wees PJ, Wammes JJ, Akkermans RP, Koetsenruijter J, Westert GP, van Kampen A, Hannink G, de Waal-Malefijt M, Schreurs BW. Patient-reported health outcomes after total hip and knee surgery in a Dutch University Hospital setting: results of twenty years clinical registry. BMC Musculoskelet Disord. 2017;18(1):97. https://doi.org/10.1186/s12891-017-1455-y.

    Article  PubMed  PubMed Central  Google Scholar 

  74. Nwachukwu BU, Chang B, Rotter BZ, Kelly BT, Ranawat AS, Nawabi DH. Minimal clinically important difference and substantial clinical benefit after revision hip arthroscopy. Arthroscopy. 2018;34(6):1862–8. https://doi.org/10.1016/j.arthro.2018.01.050.

    Article  PubMed  Google Scholar 

  75. Maffulli N, Aicale R. Proximal femoral fractures in the elderly: a few things to know, and some to forget. Medicina (Kaunas). 2022;58(10):1314. https://doi.org/10.3390/medicina58101314.

    Article  PubMed  Google Scholar 

  76. Zhang LL, Zhang Y, Ma X, Liu Y. Multiple cannulated screws vs. dynamic hip screws for femoral neck fractures: a meta-analysis. Orthopade. 2017;46(11):954–62. https://doi.org/10.1007/s00132-017-3473-8.

    Article  PubMed  Google Scholar 

Download references

Funding

Open Access funding enabled and organized by Projekt DEAL. Funded by the Brandenburg Medical School publication fund supported by the German Research Foundation and the Ministry of Science, Research and Cultural Affairs of the State of Brandenburg.

Author information

Authors and Affiliations

Authors

Contributions

NR, PL performed the literature search. KJ, MH, and NR performed the statistics. KJ and NR created tables and figures. NR wrote the manuscript. All authors supervised the whole process and read the final version.

Corresponding author

Correspondence to Nikolai Ramadanov.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare that they have competing interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1:

Prisma checklist.

Additional file 2:

Raw data extraction sheet.

Additional file 3:

Forest plot of operation time (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 4:

Forest plot of intraoperative blood loss (displaced femoral neck fractures only). Results are shown for a unit of 100 ml. CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 5:

Forest plot of EQ 5D 3-4 months postoperatively (displaced femoral neck fractures only). CS, cannulated screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 6:

Forest plot of EQ 5D 12 months postoperatively (displaced femoral neck fractures only). CS, cannulated screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 7:

Forest plot of EQ 5D 2 years postoperatively (displaced femoral neck fractures only). CS, cannulated screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 8:

Forest plot of Harris Hip Score ≤ 6 months postoperatively (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 9:

Forest plot of Harris Hip Score 12 months postoperatively (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 10:

Forest plot of Harris Hip Score 2 years postoperatively (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 11:

Forest plot of Harris Hip Score 3-5 years postoperatively (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 12:

Forest plot of hospital stay (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; SD, standard deviation; MD, mean difference; CI, confidence interval.

Additional file 13:

Forest plot of reoperation (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval.

Additional file 14:

Forest plot of mortality (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval.

Additional file 15:

Forest plot of infection (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval.

Additional file 16:

Forest plot of failure (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval.

Additional file 17:

Forest plot of ANFH (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; OR, odds ratio; CI, confidence interval.

Additional file 18:

Forest plot of nonunion (displaced femoral neck fractures only). CS, cannulated screw; DHS, dynamic hip screw; HA, hemiarthroplasty; THA, total hip arthroplasty; OR, odds ratio; CI, confidence interval.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ramadanov, N., Jóźwiak, K., Hauptmann, M. et al. Cannulated screws versus dynamic hip screw versus hemiarthroplasty versus total hip arthroplasty in patients with displaced and non-displaced femoral neck fractures: a systematic review and frequentist network meta-analysis of 5703 patients. J Orthop Surg Res 18, 625 (2023). https://doi.org/10.1186/s13018-023-04114-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13018-023-04114-8

Keywords