I now have comprehensive data from 14+ high-quality systematic reviews and RCTs. Here is the full synthesized review:
Systematic Review: Internal Fixation of Displaced Femoral Neck Fractures
Overview of Evidence Base
This review synthesizes findings from 14 systematic reviews/meta-analyses and 1 landmark RCT (FAITH trial, n=1,108), covering over 15,000 patients. The evidence spans implant comparisons, screw configuration, reduction method, fixation failure risk factors, and patient-specific factors.
SECTION 1 - Burden of Disease and Complication Rates
Slobogean et al. 2015 - Meta-Analysis of Young Patients (n=1,558) [PMID: 25480307]
The foundational reference for complication incidence after internal fixation in patients ≤60 years:
| Complication | Overall Rate | Displaced Fractures |
|---|
| Reoperation | 18.0% | Higher |
| Avascular Necrosis (AVN) | 14.3% | Higher |
| Nonunion | 9.3% | Higher |
| Malunion | 7.1% | - |
| Implant failure | 9.7% | - |
| Surgical site infection | 5.1% | - |
Key finding: Displaced fractures are significantly more likely to result in reoperation, AVN, and nonunion compared to nondisplaced fractures. These complication rates persist even with optimal technique, which is why the choice of implant and reduction quality are so debated.
SECTION 2 - Cannulated Screws (CS) vs. Dynamic Hip Screw (DHS) / Sliding Hip Screw
FAITH Trial 2017 - Landmark RCT (n=1,108, age ≥50) [PMID: 28262269]
The highest-quality direct comparison (Lancet):
- Primary outcome (reoperation at 24 months): Sliding hip screw 20% vs. cannulated screws 22% - no significant difference (HR 0.83, 95% CI 0.63-1.09; p=0.18)
- AVN: Significantly higher with sliding hip screw vs. cancellous screws (9% vs. 5%, HR 1.91, 95% CI 1.06-3.44; p=0.032)
- Subgroup benefits for sliding hip screw: smokers, displaced fractures, basicervical fractures
Conclusion: Neither implant is universally superior for reoperation. The sliding hip screw carries a higher AVN risk but may benefit select subgroups.
Shehata et al. 2019 - Systematic Review + Meta-Analysis (10 RCTs, n=1,934) [PMID: 31165917]
- Blood loss: Significantly more with sliding hip screw (MD = +110 ml; p=0.00002)
- No significant difference in operative time, hip function, nonunion, AVN, reoperation, infection, fracture healing, pain, or mortality
- Cancellous screws preferred for reduced intraoperative blood loss
Xia et al. 2021 - Meta-Analysis [PMID: 33446230]
Consistent with FAITH: sliding hip screws and cannulated screws have comparable outcomes for most endpoints, though sliding hip screw may have a modest advantage in displaced/basicervical subtypes.
SECTION 3 - Network Meta-Analysis: CS vs. DHS vs. Hemiarthroplasty vs. THA
Ramadanov et al. 2023 - Frequentist Network Meta-Analysis (33 RCTs, n=5,703) [PMID: 37626370]
This is the most comprehensive head-to-head comparison of all four operative strategies:
| Outcome | Ranking (Best to Worst) |
|---|
| Shortest operation time | CS > DHS > HA > THA |
| Least intraoperative blood loss | CS > DHS > HA > THA |
| Best functional outcome (HHS at 2 years) | THA > HA > CS > DHS |
| Best quality of life (EQ-5D at 2 years) | THA > HA > CS > DHS |
| Lowest reoperation risk | THA > HA > DHS > CS |
Reoperation ORs (vs. THA as reference):
- CS: OR 9.98 (95% CI 4.60-21.63) - highest reoperation risk
- DHS: intermediate
- HA: lower than CS/DHS
Interpretation: CS and DHS are faster and less invasive, but carry substantially higher reoperation risk for displaced fractures in older patients. THA delivers best long-term function but at higher operative cost.
SECTION 4 - Internal Fixation vs. Arthroplasty (for Displaced Fractures in Elderly)
Bhandari et al. 2003 - Meta-Analysis (14 RCTs, n=1,901) [PMID: 12954824]
The seminal meta-analysis on this question:
- Arthroplasty reduces revision surgery risk: relative risk 0.23 (p=0.0003)
- Internal fixation may have lower early mortality risk (trend toward higher early mortality with arthroplasty, especially DHS vs. screws: RR 1.75; p<0.05)
- Arthroplasty increases infection risk (RR 1.81; p=0.009)
- Greater blood loss and longer operative time with arthroplasty
- Pain and function at 1 year were similar (RR 1.12 and 0.99 respectively)
Ye et al. 2016 - Systematic Review + Meta-Analysis (8 RCTs, n=2,206) [PMID: 27779172]
- Arthroplasty: significantly lower complication rate (RR 0.56), reoperation (RR 0.17), revision (RR 0.11), and better function
- No significant difference in mortality, infection, or DVT
- Recommendation: Arthroplasty as primary treatment for displaced intracapsular fractures in elderly; IF reserved for very frail patients unfit for larger surgery
SECTION 5 - Femoral Neck System (FNS) vs. Cannulated Screws
Jiang et al. 2023 - Systematic Review + Meta-Analysis (8 studies, n=448) [PMID: 37055749]
FNS significantly outperformed CS on all key endpoints:
| Outcome | FNS vs. CS | Result |
|---|
| X-ray exposures | WMD -10.16 (95% CI -11.44 to -8.88) | FNS better |
| Fracture healing time | WMD -1.54 months (95% CI -2.38 to -0.70) | FNS better |
| Femoral neck shortening | WMD -2.01 mm (95% CI -3.11 to -0.91) | FNS better |
| Femoral head necrosis | OR 0.27 (95% CI 0.08-0.83; p=0.02) | FNS better |
| Implant failure/cut-out | OR 0.28 (95% CI 0.10-0.82; p=0.02) | FNS better |
| VAS pain score | WMD -1.27 (95% CI -2.51 to -0.04) | FNS better |
| Harris Hip Score | WMD +4.15 (95% CI 1.00-7.30; p=0.01) | FNS better |
Patel et al. 2023 - Systematic Review + Meta-Analysis (8 studies, n=509, mean age 50.8 yrs) [PMID: 36201031]
- FNS: significantly reduced complications (p<0.001), less neck shortening (p<0.001), faster union (p=0.002), better functional scores (p<0.001)
- Only advantage of CCS: significantly less blood loss (p<0.001)
- Conclusion: FNS is superior to CCS for adult femoral neck fractures
FNS vs. DHS - Jiang et al. 2024 Meta-Analysis (6 studies, n=577) [PMID: 38419236]
- FNS: shorter operation time, less blood loss, lower fixation failure rate, better Harris Hip Score (all p<0.05)
- No significant difference in complication rate or femoral neck shortening
- Both are comparable in complication rates - FNS optimises operative efficiency
SECTION 6 - Fully Threaded vs. Partially Threaded Cannulated Screws
Jia et al. 2024 - Systematic Review + Meta-Analysis [PMID: 39695853]
| Outcome | Fully Threaded (FCS) vs. Partially Threaded (PCS) |
|---|
| Femoral head necrosis | OR 0.60 (95% CI 0.37-0.98; p=0.04) - FCS lower |
| Internal fixation failure | OR 0.37 (95% CI 0.22-0.62; p=0.0002) - FCS lower |
| Femoral neck shortening | OR 0.27 (95% CI 0.19-0.40; p<0.00001) - FCS dramatically lower |
| Harris Hip Score | No significant difference |
| Nonunion rate | No significant difference |
Interpretation: Fully threaded screws significantly reduce AVN, hardware failure, and neck shortening - with no sacrifice in union rates or hip function. This challenges the traditional use of partially threaded screws.
SECTION 7 - Screw Configuration: Parallel vs. Non-Parallel (Inverted Triangle)
Jia et al. 2026 - Systematic Review + Meta-Analysis (20 studies, n=1,508) [PMID: 41827005]
Non-parallel (inverted triangle/divergent) configuration is significantly superior across all outcomes:
| Outcome | Non-Parallel vs. Parallel | Effect |
|---|
| Femoral head necrosis | OR 0.50 (95% CI 0.34-0.74; p=0.0005) | 50% reduction |
| Nonunion | OR 0.41 (95% CI 0.26-0.65; p=0.0001) | 59% reduction |
| Femoral neck shortening | OR 0.40 (95% CI 0.28-0.57; p<0.00001) | 60% reduction |
| Fixation failure | OR 0.34 (95% CI 0.22-0.52; p<0.00001) | 66% reduction |
Clinical implication: The inverted triangle (non-parallel) configuration for cannulated screws is strongly supported over parallel placement. The calcar screw (inferior screw) is the key stability element.
SECTION 8 - Medial Buttress Plate (MBP) as Adjunct
McGarry et al. 2025 - Systematic Review (21 studies, n=642) [PMID: 39827529]
Focus on Pauwels type II/III (high vertical shear) fractures:
- Non-union rate: 6%
- AVN rate: 4%
- Overall failure rate: 17.3%
- Mean time to union: 3.9 ± 1.2 months
- Mean Harris Hip Score: 89.5 ± 5.5 at final follow-up
Fixation comparisons:
- MCS + MBP failure rate: 14.6%
- DHS + MBP failure rate: 26.8% (not statistically significant; p=0.164)
- Medial or anteromedial MBP positioning yields better outcomes; anterior placement is associated with high failure rates
- No studies yet on FNS + MBP
Yuan et al. 2024 - Network Meta-Analysis (34 studies, n=2,291, young adults) [PMID: 39533559]
Comparing CCS, FNS, DHS, and MBP in young adults:
| Outcome | Best Device (SUCRA) |
|---|
| Lowest intraoperative bleeding | Inverted triangle CCS (SUCRA 0.87) |
| Best Harris Hip Score | MBP (SUCRA 0.85) |
| Lowest complication rate | MBP (SUCRA 0.93) |
| Fastest fracture healing | MBP (SUCRA 0.81) |
| Shortest operation time | FNS (SUCRA 0.77) |
| Least neck shortening | FNS (SUCRA 0.79) |
Conclusion: MBP achieves the best clinical outcomes in young adults but requires open surgery. FNS minimises shortening and is the most time-efficient.
SECTION 9 - Basicervical Fractures: CMN vs. DHS
Yoon et al. 2022 - Systematic Review + Meta-Analysis (7 studies, n=353) [PMID: 35914756]
Basicervical fractures are biomechanically distinct (extracapsular tendency, high rotational demands):
- Fracture union time: CMN significantly faster (MD -0.41 months; 95% CI -0.70 to -0.12; p=0.006; I²=0%)
- Cut-out rate: No significant difference (OR 0.54; 95% CI 0.10-2.82; p=0.47)
- Reoperation rate: No significant difference (OR 0.65; 95% CI 0.15-2.86; p=0.57)
- Recommendation: Both DHS and CMN are acceptable; CMN achieves faster union; choice based on surgeon preference and associated injuries
SECTION 10 - Predictors of Fixation Failure
Kalsbeek et al. 2023 - Systematic Review + Meta-Analysis (16 studies) [PMID: 36437167]
Identified 24 predictors; 5 confirmed in meta-analysis:
| Risk Factor | Odds Ratio | 95% CI |
|---|
| Female sex | 1.78 | 1.26-2.52 |
| Smoking | 3.64 | 1.68-7.91 |
| Age >50 years | 3.64 | 1.68-7.91 |
| Inadequate fracture reduction | 2.28 | 1.62-3.22 |
| CS/pins vs. fixed-angle devices | 2.16 | 1.03-4.54 |
Key message: Inadequate reduction is the single most modifiable risk factor. Female sex, smoking, and age >50 each independently predict failure. Fixed-angle devices (DHS, FNS) outperform CS/pins in failure resistance.
SECTION 11 - Open vs. Closed Reduction
Ghayoumi et al. 2015 - Systematic Review (21 studies) [PMID: 25554424]
- Nonunion: CRIF 11.6% vs. ORIF 14.9% (p=0.25) - not significant
- AVN: CRIF 17.2% vs. ORIF 17.7% (p=0.91) - not significant
- Deep infection: CRIF 0.49% vs. ORIF 3.9% (p=0.0019) - significantly lower with closed reduction
- Conclusion: No evidence supporting open over closed reduction for reducing union/AVN; ORIF carries 8x higher infection risk. CRIF preferred when an adequate closed reduction can be achieved.
Synthesis: Key Evidence-Based Conclusions
| Clinical Question | Evidence-Based Answer | Source |
|---|
| IF vs. arthroplasty in elderly displaced FNF | Arthroplasty superior (lower reoperation, better function) | Ye 2016; Ramadanov 2023 |
| IF vs. arthroplasty in young patients | IF preferred (preserves native femoral head) | - |
| CS vs. DHS overall | No difference in reoperation; DHS higher AVN but benefits displaced/basicervical/smokers | FAITH trial 2017 |
| FNS vs. CS | FNS significantly better (AVN, shortening, union, function) | Jiang 2023; Patel 2023 |
| FNS vs. DHS | FNS better operative efficiency; similar complication rates | Jiang 2024 |
| Parallel vs. non-parallel screws | Non-parallel (inverted triangle) dramatically superior | Jia 2026 |
| Fully vs. partially threaded screws | Fully threaded: less AVN, less failure, less shortening | Jia 2024 |
| MBP in Pauwels II/III | Best outcomes in young adults (lowest complications/best HHS) | Yuan 2024; McGarry 2025 |
| CMN vs. DHS in basicervical | Both acceptable; CMN achieves faster union | Yoon 2022 |
| Open vs. closed reduction | CRIF preferred (no difference in AVN/nonunion; ORIF 8x infection risk) | Ghayoumi 2015 |
| Risk factors for fixation failure | Smoking, age >50, female sex, poor reduction, CS use | Kalsbeek 2023 |
Evidence Grading Summary
| Study | Design | N | Evidence Tier |
|---|
| FAITH trial (Lancet 2017) | Multicentre RCT | 1,108 | Tier 1 RCT |
| Ramadanov 2023 | Network meta-analysis (33 RCTs) | 5,703 | Tier 1 |
| Ye 2016 | Systematic review + MA (8 RCTs) | 2,206 | Tier 1 |
| Shehata 2019 | Systematic review + MA (10 RCTs) | 1,934 | Tier 1 |
| Slobogean 2015 | Meta-analysis (41 studies) | 1,558 | Tier 2 |
| Bhandari 2003 | Meta-analysis (14 RCTs) | 1,901 | Tier 2 |
| Jia 2026 (parallel screws) | Systematic review + MA (20 studies) | 1,508 | Tier 1 |
| Yuan 2024 | Network meta-analysis (34 studies) | 2,291 | Tier 1 |
| Patel/Jiang 2023 (FNS vs. CS) | Systematic review + MA | 448-509 | Tier 1 |
| McGarry 2025 (MBP) | Systematic review | 642 | Tier 1 |
| Kalsbeek 2023 | Systematic review + MA (16 studies) | - | Tier 1 |
| Ghayoumi 2015 | Systematic review (21 studies) | - | Tier 1 |
Bottom line from the evidence: For displaced femoral neck fractures treated with internal fixation, the FNS is emerging as the preferred implant over traditional cannulated screws, with the best balance of reduced AVN, less shortening, faster union, and better function. For high-shear Pauwels III fractures in young adults, medial buttress plate augmentation yields the best reported outcomes. The inverted-triangle (non-parallel) screw configuration and fully-threaded screws should be used when CS is chosen. Reduction quality remains the single most modifiable determinant of success across all implants.