Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Editorial Board
Literature Review
Narrative review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Editorial Board
Literature Review
Narrative review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
View/Download PDF

Translate this page into:

65 (); 204-210
doi:
10.1016/j.jor.2025.05.016

Kinematic vs. mechanical alignment in total knee arthroplasty: A statistical analysis of randomized control trials utilizing dichotomous and continuous fragility metrics

, , , , , , , , ,
a Department of Orthopedic Surgery, Icahn School of Medicine at Mount Sinai, 1 Gustave L. Levy Pl, New York, NY, USA
b Department of Orthopedic Surgery, Royal College of Surgeons in Ireland, 123 St Stephen's Green, Dublin, Ireland

⁎Corresponding author: Adriano Cuadros. adriano.cuadros@icahn.mssm.edu

Received: , Accepted: ,
Professor P K Surendran Memorial Education Foundation
Disclaimer:
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.

Abstract

Abstract

Randomized controlled trials (RCTs) that assess kinematic alignment (KA) versus mechanical alignment (MA) in total knee arthroplasty (TKA) report p-values that influence surgical decision-making. This study utilizes fragility index (FI) and fragility quotient (FQ) metrics to assess the statistical stability of outcomes reported in RCTs comparing KA and MA approaches in TKA.

Pubmed, Embase, and MEDLINE were queried for RCTs evaluating KA vs MA approaches in TKA. The FI and reverse fragility index (rFI) were calculated for dichotomous outcomes and were defined as the number of event reversals needed to alter the statistical significance for significant and non-significant outcomes, respectively. The continuous fragility index (cFI) was used to evaluate statistically significant continuous variables and was calculated using a novel online tool. The FQ was calculated by dividing the FI, rFI, or cFI by sample size.

The median FI across 58 included outcomes was 7.0 (FQ 0.109). Overall, dichotomous variables were more fragile than continuous variables. The 12 dichotomous outcomes were all statistically non-significant, with a median rFI of 4.5 (FQ 0.040). The 46 continuous outcomes were all statistically significant, with a median cFI of 9.1 (FQ 0.113). In 21 of 58 outcomes, the number of patients lost to follow up was greater than the FI for the respective outcome. When analyzing by outcome category, clinical improvement and complications were more fragile, with median FIs of 6.7 and 5.0.

Continuous outcomes were more robust than the relatively fragile dichotomous outcomes. The continuous outcomes in KA versus MA RCTs were also more robust compared to other current studies reporting cFI values. cFI is a novel, valuable tool that allows for assessment of fragility for continuous outcomes, and reporting alongside FI, rFI, and FQ with p-values is recommended to assess the reliability of RCTs.

Abstract

Highlights of The Paper

•First to apply continuous fragility index to RCTs comparing kinematic vs. mechanical alignment in total knee arthroplasty.•Found continuous outcomes are more robust than dichotomous ones, highlighting variable reliability of reported outcomes.•Shows value of reporting fragility metrics alongside p-values to improve interpretation of RCTs in orthopedic literature.

Keywords

Total knee arthroplasty (TKA)
Kinematic alignment
Fragility index
Continuous fragility index
1

1 Introduction

Total knee arthroplasty (TKA) is a common orthopaedic procedure used to relieve pain and restore functionality in those with end-stage osteoarthritis (OA) of the knee. By 2030, the annual demand for TKA in the United States is projected to increase between 1.26 and 3.48 million.1,2 While demand continues to rise, patient dissatisfaction with TKA is not negligible. Recent estimates of patient outcome dissatisfaction with TKA range from 14 to 19 %.1,2 Thus, multiple studies have assessed alternative surgical alignment strategies in an effort to minimize dissatisfaction after TKA).

Two prominent alignment strategies used in TKA are kinematic alignment (KA) and mechanical alignment (MA). MA is the predominantly used surgical technique for TKA.3 It involves making a femoral cut and tibial resection that are perpendicular to the mechanical axis of the respective component and aims to create an even load distribution across the new joint line.4–6 KA TKA alignment aims to maintain the patient's natural knee kinematic axis. This is done by making femoral and tibial bone cuts that attempt to replicate the patient's natural pre-arthritic anatomy. Interest in KA as an alternative to MA TKA has risen in response to the considerable percentage of patients who report dissatisfaction following MA TKA.5 Randomized control trials (RCTs) have been conducted comparing the two alignment approaches, however, the robustness of outcomes reported in such RCTs remains unclear.

RCTs represent the highest level of evidence in determining the superior surgical technique for TKA.7 The P-value is used throughout the orthopaedic literature to determine the statistical significance of a study's findings. However, it has received criticism for overlooking critical aspects of study design.8 The fragility index (FI) was introduced to determine the vulnerability of a reported outcome to the reversal of statistical significance.9 Although fragility analysis has thus far been limited to assessing the robustness of dichotomous categorical variables, the continuous fragility index (cFI) was recently introduced by Caldwell as a way to assess the fragility of statistically significant continuous variables.10 The purpose of this study was to evaluate the statistical fragility of RCTs comparing KA versus MA TKA using FI, rFI, cFI, and FQ metrics. We hypothesized that many study findings originally denoted as statistically significant would prove to be statistically fragile.

2

2 Methods

2.1

2.1 Literature review

This fragility analysis conducted a systematic review of RCTs comparing KA versus MA TKA in the TKA literature. Our systematic review followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). IRB approval was not required for this manuscript as only publicly available data was included in the investigation. The PubMed, Embase, and MEDLINE databases were queried for RCTs published between January 1, 2010 to December 31, 2023. The following keywords were used to identify all articles relevant to KA and MA TKA across the three databases: (knee arthroplasty) AND (mechanical) AND (kinematic).

Studies met inclusion criteria if they were RCTs comparing KA and MA in TKA, had two treatment arms, and reported continuous outcomes or dichotomous categorical outcomes. Studies were excluded if they were not RCTs or had more than two treatment arms. Publications were also excluded if they were not published in English, the full text was unavailable, or if it was an in vitro, animal, cadaveric, or biomechanical study.

The literature review was performed concurrently by two authors and verified by a third, independent reviewer to resolve discrepancies. Bias assessment was performed utilizing the revised Cochrane Risk of Bias tool for evaluating bias in randomized trials. This systematic review analyzed statistical reporting and statistical significance of outcomes rather than direct; outcomes from interventions, and thus did not qualify for PROSPERO international registry (Fig. 1).

PRISMA flow diagram showing identification, screening, and inclusion of eligible studies from PubMed, Embase, and Medline.
Fig. 1 PRISMA flow diagram showing identification, screening, and inclusion of eligible studies from PubMed, Embase, and Medline.
2.2

2.2 Data extraction

Information extracted from selected publications included first author, publication year, journal title, experimental and control group interventions, outcome measures, their respective results, the number of patients lost to follow-up (LTF), and P values if reported. Each recorded outcome measure was categorized into the following categories: radiographic parameters, clinical improvement, and complications.

2.3

2.3 Fragility analysis

Using the recorded outcomes data, P values were confirmed via a Fisher's exact test to ensure accuracy. Fragility analysis was then performed by manipulating outcome events in a 2 x 2 contingency table until a reversal of significance was determined. Statistical significance was set at P < 0.05.

For study outcomes reporting statistical significance, a FI was calculated. The FI is calculated as the number of outcome event manipulations that need to be made for significance to be lost for a study finding.11 The reverse fragility index (rFI) was similarly defined as the number of event manipulations to convert nonsignificant outcomes into statistically significant outcomes.12 The fragility quotient (FQ) accounts for a comparative trial's sample size by dividing the FI by the sample size.11

For statistically significant continuous outcomes, the continuous fragility index (cFI) was calculated using Caldwell's online calculator. The calculator's algorithm utilizes descriptive statistics (e.g. mean, sample size, standard deviation) to simulate full data sets that assume a normal distribution. The algorithm has an option to run multiple iterations to reduce random error, allowing the user to decide how many iterations to run. All cFI values in this study were calculated using 10 algorithmic iterations. The model's tolerance, another user modifiable parameter of the proximity of the simulated data set's mean and standard deviation to the study of interest, was set at 0.01.10

Smaller FIs represent more fragile outcomes. For outcomes that were not statistically significant, an rFI was determined in a similar manner as FI, but with the intention of determining the number of reversals to achieve statistical significance. These values were then used to calculate an FQ, dividing the FI (or rFI, cFI) by the total number of study outcomes, with higher FQs representing more robust results. Subgroup analyses were performed for each of the outcome categories. Interquartile ranges (IQRs) were calculated and reported for both the FIs and FQs. Summary statistics are presented as median FI (IQR).

3

3 Results

After conducting a literature search, a total of 123 studies were screened for inclusion. Ultimately, 12 RCTs across 5 orthopaedic surgery journals were included for fragility analysis. Four studies were from The Bone and Joint Journal, 3 from The Journal of Arthroplasty, 2 each from Clinical Orthopedics and Related Research and Knee Surgery, Sports Traumatology, Arthroscopy, and 1 from Orthopedics (Table 1). After utilizing the revised Cochrane risk-of-bias tool, we found that 4/12 included studies had either “some concerns” or were at “high risk” of bias. Only one (1/12) of these studies was found to be “high risk” (Table 2).

Table 1 Characteristics of included studies.
Author Year Study Title Journal Interventions Assessed Sample Size Lost to Follow-up
Callies et al. 2017 PSI kinematic versus non-PSI mechanical alignment in total knee arthroplasty: a prospective, randomized study Knee Surgery, Sports Traumatology, Arthroscopy KA vs MA TKA 198 1
Dossett and Arthur et al. 2023 A Randomized Controlled Trial of Kinematically and Mechanically Aligned Total Knee Arthroplasties: Long-Term Follow-Up The Journal of Arthroplasty KA vs MA TKA 48 37
Dossett and Estrada et al. 2014 A randomised controlled trial of kinematically and mechanically aligned total knee replacements: two-year clinical results The Bone and Joint Journal KA vs MA TKA 73 15
Dossett and Swartz et al. 2012 Kinematically versus mechanically aligned total knee arthroplasty Orthopedics KA vs MA TKA 82 6
Ettinger et al. 2024 Higher satisfaction and function scores in restricted kinematic alignment versus mechanical alignment with medial pivot design total knee arthroplasty: A prospective randomised controlled trial Knee Surgery, Sports Traumatology, Arthroscopy restricted KA vs MA with medial pivot design TKA 98 22
Laende et al. 2019 A randomized controlled trial of tibial component migration with kinematic alignment using patient-specific instrumentation versus mechanical alignment using computer-assisted surgery in total knee arthroplasty The Bone and Joint Journal KA using patient-specific instrumentation vs MA using computer-assisted surgery 47 1
Matsumoto et al. 2017 Radiological and clinical comparison of kinematically versus mechanically aligned total knee arthroplasty The Bone and Joint Journal KA vs MA TKA 60 0
McEwen et al. 2019 Computer-Assisted Kinematic and Mechanical Axis Total Knee Arthroplasty: A Prospective Randomized Controlled Trial of Bilateral Simultaneous Surgery The Journal of Arthroplasty computer assisted bilateral KA vs computer assisted bilateral MA TKA 82 8
Sarzaeem et al. 2024 Kinematic Alignment Technique Outperforms Mechanical Alignment in Simultaneous Bilateral Total Knee Arthroplasty: A Randomized Controlled Trial The Journal of Arthroplasty KA vs MA TKA 130 8
Waterson et al. 2016 The early outcome of kinematic versus mechanical alignment in total knee arthroplasty: a prospective randomised control trial The Bone and Joint Journal KA vs MA TKA 71 5
Young and Sullivan et al. 2020 No Difference in 5-year Clinical or Radiographic Outcomes Between Kinematic and Mechanical Alignment in TKA: A Randomized Controlled Trial Clinical Orthopedics and Related Research KA vs MA TKA 95 4
Young and Walker et al. 2017 The Chitranjan S. Ranawat Award: No Difference in 2-year Functional Outcomes Using Kinematic versus Mechanical Alignment in TKA: A Randomized Controlled Clinical Trial Clinical Orthopedics and Related Research KA vs MA TKA 99 0
Table 2 Risk of Bias Assessment for Included Studies Evaluated using Revised Cochrane Risk-of-Bias Tool for Randomized Trials.

A total of 58 outcomes were extracted from the 12 included studies. The median FI across the 58 outcomes, both dichotomous and continuous, was 7.0 (IQR 5.0–11.4), with an associated median FQ of 0.109 (IQR 0.062–0.145). Out of the 58 total outcomes, there were 12 dichotomous outcomes that were all statistically non-significant, and 46 continuous outcomes that were all statistically significant. For the 12 non-significant dichotomous outcomes, the median rFI was 4.5 (IQR 2.8–5.0), and the median FQ was 0.040 (IQR 0.020–0.065). For the 46 statistically significant continuous outcomes, the median cFI was 9.1 (IQR 5.9–12.6) and the median FQ was 0.113 (IQR 0.077–0.157). Overall, dichotomous variables were more fragile than continuous variables, with median FIs of 4.5 and 9.1, respectively (Table 3).

Table 3 Fragility Data of Discrete vs. Continuous Outcomes.
Outcome Events FI, median (IQR) FQ, median (IQR) LTF > FI (%)
All Outcomes 58 7 (5.0–11.4) 0.109 (0.062–0.145) 21 (36.21 %)
Discrete Outcomes 12 4.5 (2.8–5.0) 0.040 (0.020–0.065) 6 (50 %)
Continuous Outcomes 46 9.1 (5.9–12.6) 0.113 (0.077–0.157) 15 (32.60 %)

A subgroup analysis was performed by categorizing the reported outcomes. The three outcome classes included: radiographic parameters, clinical improvement, and complications. The 18 radiographic parameters, 17 of which were continuous outcomes, were the least fragile. Radiographic parameters had a median FI of 10.8 (IQR 8.3–13.3) and an associated FQ of 0.143 (IQR 0.111–0.167). Clinical improvement was the most common overall category, with 31 total outcomes in this class. It was also the second least fragile outcome category, with a median FI of 6.7 (IQR 5.0–10.7) and an associated FQ of 0.095 (IQR 0.062–0.129). 29 of the 31 clinical improvement outcomes were statistically significant continuous outcomes. 24 of the 29 (82.7 %) outcomes reported more clinical improvement in MA TKA compared to KA TKA, whilst 5 of 29 (17.3 %) reported more clinical improvement in KA TKA. The most fragile outcome category, complications, was also the class with the least outcomes, with a total of 9 outcomes. All 9 outcomes were non-significant dichotomous outcomes. Complication outcomes had a median FI of 5.0 (IQR 4.0–5.0) and an associated FQ of 0.042 (IQR 0.025–0.068) (Table 4).

Table 4 Fragility data by outcome category.
Outcome Events FI, median (IQR) FQ, median (IQR) LTF > FI (%)
Radiographic Parameters 18 10.8 (8.3–13.3) 0.143 (0.111–0.167) 3 (16.67 %)
Clinical Improvement 31 6.7 (5.0–10.7) 0.095 (0.062–0.129) 12 (38.71 %)
Complications 9 5.0 (4.0–5.0) 0.042 (0.025–0.068) 6 (66.67 %)

Notably, we identified that in 21 of the 58 total outcomes (36.21 %), the number of patients lost to follow-up (LTF) for that particular study was greater than the FI for the respective outcome. This means that for more than a third of all outcomes, simply maintaining complete patient follow-up could have been enough to reverse the outcome significance. When stratified by outcome type, dichotomous variables had a proportionally larger number of outcomes compared to continuous variables in which the patients LTF for the respective study was larger than the FI. In 6 of the 12 (50.00 %) dichotomous outcomes the LTF was greater than FI, compared to 15 of the 46 (32.60 %) of the continuous outcomes (Table 3). When stratified by outcome category, complications were the class with proportionally the largest number of outcomes in which the patients LTF for the respective study was larger than the FI (66.67 %), followed by clinical improvement outcomes (38.71 %), and finally radiographic parameters (16.67 %) (Table 4).

4

4 Discussion

This study aimed to utilize fragility index (FI) and fragility quotient (FQ) metrics to assess the statistical stability of outcomes reported in RCTs comparing KA and MA approaches in TKA. The most notable result of this study was that the median fragility index for dichotomous outcomes was considerably more fragile than those for continuous outcomes, with median FIs of 4.5 and 9.1 respectively. Given that all included dichotomous outcomes were statistically non-significant, this rFI value represents that on average, it would take only 4.5 outcome reversals to make these outcomes significant. Additionally, 50 % of the dichotomous outcomes had an rFI lower than the patients lost to follow-up for the respective study, compared to 32.60 % for continuous variables. This signifies that for half of all non-continuous outcomes, simply maintaining patient follow-up could have made these findings significant. When outcomes were grouped and analyzed by category, the most fragile category was complications, with a median FI of 5.0. Complications were the outcome category with proportionally the most dichotomous outcomes (100 %), which also meant that all these complication outcomes were non-significant. Complications also had proportionally the greatest number of outcomes (66.67 %) in which the patients LTF for the respective study was greater than the FI for that outcome. Clinical improvement outcomes demonstrated superiority of MA TKA compared to KA TKA. 24 of the 29 significant clinical improvement outcomes (82.7 %) reported more clinical improvement in MA TKA compared to KA TKA, whilst 5 of 29 (17.3 %) reported more clinical improvement in KA TKA.

The calculated rFI of 4.5 for dichotomous outcomes in RCTs included in this study is in line with what current orthopaedic fragility studies have considered “fragile”, with FI values under 5 frequently cited.13–31 Considering the amount of patients loss to follow up relative to this FI value, the point arises that for RCTs assessing dichotomous outcomes of KA vs MA in TKA, statistical significance could have been achieved by either simply improving patient follow-up or having a more powerful study. This would reduce the chance of incorrectly accepting a false null hypothesis, known as a type II error. Therefore, it can be surmised that the dichotomous outcomes of KA vs MA in TKA RCTs are relatively fragile. Comparatively, continuous variables had significantly higher FI values (median cFI of 9.1) than the dichotomous outcomes (median rFI of 4.5). Caldwell mentions in his study introducing cFI that this is expected, as the statistical tests utilized to assess continuous variables (e.g. t-tests) are inherently less vulnerable to fluctuations in data than tests for dichotomous variables (e.g. Chi-Squared).10 Therefore, the statistical fragility of the dichotomous outcomes and continuous outcomes within this study should not be compared to each other, as FI (and rFI) is intrinsically a different parameter and impossible to compare directly to cFI. Rather, this study demonstrates how dichotomous outcomes, compared to continuous outcomes, are inherently more fragile and more susceptible to changes in statistical significance in studies with low power or with many patients lost to follow-up.

This study highlights the utility of cFI as a novel tool to begin evaluating the robustness of continuous outcome variables. Previously, fragility analyses have been limited by the capability to assess the robustness of only dichotomous, non-continuous outcomes. Caldwell's novel algorithm provides a standardized method to analyze the fragility of a previously untapped cohort of study outcomes within orthopaedic literature. When comparing the cFI of 9.1 for continuous outcomes of MA vs KA in TKA to the current, small collection of orthopaedic studies utilizing cFI, it is evident that this is a relatively robust cFI. Only 6 studies at the time this study was written have utilized cFI in their fragility analyses, with reported cFIs of 3, 5, 6, 7, 8.2 and 9 10,.32–36 Additionally, Caldwell reported a median cFI of 7 in his survey of continuous sports orthopaedics outcomes.10 Therefore, it can be said that comparatively, the continuous outcomes of MA vs KA in TKA RCTs are relatively robust.

4.1

4.1 Limitations and strengths

Our study is the first to utilize fragility metrics, and now continuous fragility, to assess the robustness of outcomes comparing MA TKA vs KA TKA. Bias and confounding factors were reduced by including studies across three databases, focusing on RCTs only, and by following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. These guidelines ensured that only relevant studies looking at MA vs KA in TKA were included for analysis, by utilizing the appropriate inclusion and exclusion criteria for screening. Additionally, a bias assessment that looked at aspects of trial design, conduct, and reporting, showed that only 1 of the 12 included studies were at “high risk” of bias. Therefore, it is unlikely that the fragility of outcomes in this study was due to bias in study design or reporting.

Our study did not come without limitations. Currently, there are no established guidelines that dictate how to interpret the FI and FQ of a study. It is therefore through comparison to other orthopaedic studies with reported fragility metrics that an FI (or rFI) and FQ value can be said to be robust or fragile. As such, these conclusions are relative, not absolute nor standardized. This is especially true when interpreting the cFI of an outcome, given that there are only six other studies that currently utilize this novel fragility tool. For rFI and FI, there are substantially more studies looking at these dichotomous fragility metrics compared to cFI, making it easier to compare the relative fragility of such studies. cFI is therefore substantially less validated than both rFI and FI, and future studies are needed to further assess the validity of this fragility algorithm. Additionally, cFI calculations were limited to significant continuous outcomes, as cFI is not yet validated for non-significant continuous outcomes. The development of an algorithm to calculate the reverse continuous fragility index, or rcFI, would be useful to assess this cohort of outcomes. Nonetheless, this study adds to the growing body of work that highlights the utility of including FI, and now cFI, when reporting outcomes within orthopaedic studies. Therefore, for the time being, FI, rFI, FQ, and now cFI for continuous outcome variables should continue to be reported alongside P-values in orthopaedic studies. This is with the hope that one day, we may have established fragility guidelines for orthopaedic research that will aid clinicians in interpreting study outcomes and subsequently making sound clinical decisions.

5

5 Conclusion

The fragility of outcomes in RCTs evaluating KA vs MA in TKAs varied depending on outcome category and whether the variable was dichotomous or continuous. Continuous outcomes are more robust than the relatively fragile dichotomous outcomes within KA vs MA in TKA RCTs, but are also more robust compared to the limited cohort of other studies that are also reporting cFI values. Clinical improvement outcomes favored MA TKA over KA TKA. This study highlights cFI as a novel, powerful tool that can assess the fragility of continuous outcomes within the orthopaedic literature, and reporting cFI alongside FI, rFI, and FQ with p-values is recommended to assess the reliability of RCTs.

CRediT authorship contribution statement

Adriano Cuadros: Data curation, Formal analysis, Writing – original draft, Writing – review & editing. Michaela Corvi: Data curation, Writing – original draft, Writing – review & editing. Avanish Yendluri: Data curation, Formal analysis, Writing – original draft, Writing – review & editing. Francesca Docters: Data curation, Writing – original draft, Writing – review & editing. Michael S. Shatkin: Writing – review & editing. John J. Corvi: Writing – review & editing. Suraj A. Dhanjani: Writing – review & editing. Brett L. Hayden: Writing – review & editing. Douglas B. Unis: Writing – review & editing. Robert L. Parisien: Conceptualization, Methodology, Supervision, Writing – review & editing.

Institutional review board

IRB approval was not required for this manuscript as only publicly available data was included in the investigation.

Funding/sponsorship

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

References

  1. , , , , . The John Insall Award: patient expectations affect satisfaction with total knee arthroplasty. Clin Orthop Relat Res. 2006;452:35-43.
    [Google Scholar]
  2. , , , , , . Patient satisfaction after total knee arthroplasty: who is satisfied and who is not? Clin Orthop Relat Res. 2010;468(1):57-63.
    [Google Scholar]
  3. , , , . Kinematic alignment versus mechanical alignment in primary total knee arthroplasty: an updated meta-analysis of randomized controlled trials. J Orthop Surg Res. 2022;17(1):1-18.
    [Google Scholar]
  4. , , , , . Total knee arthroplasty. Clin Orthop Relat Res. 1985;192:13-22.
    [Google Scholar]
  5. , , , , . Mechanical and kinematic alignment in total knee arthroplasty. Ann Transl Med. 2016;4(7):130.
    [Google Scholar]
  6. , , , , , , . Head-to-Head comparison of kinematic alignment versus mechanical alignment for total knee arthroplasty. J Arthroplast. 2022;37(8S):S849-S851.
    [Google Scholar]
  7. , , . Editorial: threshold P values in orthopaedic research-we know the problem. What is the solution? Clin Orthop Relat Res. 2018;476(9):1689-1691.
    [Google Scholar]
  8. , , . Sifting the evidence-what’s wrong with significance tests? Phys Ther. 2001;81(8):1464-1469.
    [Google Scholar]
  9. , . The unit fragility index: an additional appraisal of “statistical significance” for a contrast of two proportions. J Clin Epidemiol. 1990;43(2):201-209.
    [Google Scholar]
  10. A method for calculating the fragility index of continuous outcomes. J Clin Epidemiol. 2021;136:20-25.
    [Google Scholar]
  11. , , , . What is a fragility index?: fragility and reverse fragility index for assessing the significance of results from published randomized controlled trials. Bone Joint Lett J. 2024;106-B(4):319-322.
    [Google Scholar]
  12. , , , et al . Application of the reverse fragility index to statistically nonsignificant randomized clinical trial results. JAMA Netw Open. 2020;3(8)
    [Google Scholar]
  13. , , , et al . The fragility of significance in the hip arthroscopy literature: a systematic review. JB JS Open Access. 2021;6(4)
    [Google Scholar]
  14. , , , et al . Statistical fragility and the role of P values in the sports medicine literature. J Am Acad Orthop Surg. 2019;27(7):e324-e329.
    [Google Scholar]
  15. , , , et al . Comparative studies in the shoulder literature lack statistical robustness: a fragility analysis. Sports Med Arthrosc Rehabil Ther Technol. 2021;3(6):e1899-e1904.
    [Google Scholar]
  16. , , , , , , . The statistical fragility of platelet-rich plasma in rotator cuff surgery: a systematic review and meta-analysis. Am J Sports Med. 2021;49(12):3437-3442.
    [Google Scholar]
  17. , , , , , . Statistical significance in trauma research: too unstable to trust? J Orthop Trauma. 2019;33(12):e466-e470.
    [Google Scholar]
  18. , , , , , , . The fragility of statistical findings in achilles tendon injury research: a systematic review. J Am Acad Orthop Surg Glob Res Rev. 2021;5(9)
    [Google Scholar]
  19. , , , et al . The fragility of statistical significance in cartilage restoration of the knee: a systematic review of randomized controlled trials. Cartilage. 2021;13(1_suppl):147S-155S.
    [Google Scholar]
  20. , , , et al . The statistical fragility of the distal fibula fracture literature: a systematic review of randomized controlled trials. Injury 2023
    [Google Scholar]
  21. , , , et al . The fragility of statistical significance in distal femur fractures: systematic review of randomized controlled trials. Eur J Orthop Surg Traumatol. 2023;33(6):2411-2418.
    [Google Scholar]
  22. , , , , , , . The fragility of statistical findings in distal biceps tendon repairs: a systematic review of randomized controlled trials. J Shoulder Elb Surg. 2023;32(8):e379-e386.
    [Google Scholar]
  23. , , . The fragility of statistical findings in distal radius fractures: a systematic review of randomized controlled trials. Injury. 2022;53(10):3352-3356.
    [Google Scholar]
  24. , , , , , , . Statistical fragility of randomized clinical trials in shoulder arthroplasty. J Shoulder Elb Surg. 2021;30(8):1787-1793.
    [Google Scholar]
  25. , , , , , , . Randomized controlled trials comparing bone-patellar tendon-bone versus hamstring tendon autografts in anterior cruciate ligament reconstruction surgery are statistically fragile: a systematic review. Arthroscopy. 2024;40(3):998-1005.
    [Google Scholar]
  26. , , , et al . The fragility index of total hip arthroplasty randomized control trials: a systematic review. J Am Acad Orthop Surg. 2022;30(9):e741-e750.
    [Google Scholar]
  27. , , , et al . The statistical fragility of operative vs nonoperative management for achilles tendon rupture: a systematic review of comparative studies. Foot Ankle Int. 2022;43(10):1331-1339.
    [Google Scholar]
  28. , , , et al . Statistical fragility of single-row versus double-row anchoring for rotator cuff repair: a systematic review of comparative studies. Orthop J Sports Med. 2022;10(5)
    [Google Scholar]
  29. , , , et al . The fragility of statistically significant findings from randomized trials in spine surgery: a systematic survey. Spine J. 2015;15(10):2188-2197.
    [Google Scholar]
  30. , , , , , , . The fragility of tourniquet use in total knee arthroplasty: a systematic review of randomized controlled trials. J Arthroplast. 2023;38(6):1177-1183.
    [Google Scholar]
  31. , , , , , , . The fragility of statistical significance in patellofemoral instability research: a systematic review. Am J Sports Med. 2022;50(13):3714-3718.
    [Google Scholar]
  32. , , , , . Statistical fragility of ketamine infusion during scoliosis surgery to reduce opioid tolerance and postoperative pain. World Neurosurg. 2022;164:135-142.
    [Google Scholar]
  33. , , , , , , . The statistical fragility of platelet-rich plasma as treatment for plantar fasciitis: a systematic review and simulated fragility analysis. Foot Ankle Orthop. 2022;7(4)
    [Google Scholar]
  34. , . The fragility index for assessing the robustness of the statistically significant results of experimental clinical studies. J Gen Intern Med. 2022;37(1):206-211.
    [Google Scholar]
  35. , , , , , , . The statistical fragility of platelet-rich plasma as treatment for chronic noninsertional achilles tendinopathy: a systematic review and meta-analysis. Foot Ankle Orthop. 2022;7(3)
    [Google Scholar]
  36. , , , et al . The continuous fragility index of statistically significant findings in randomized controlled trials that compare interventions for anterior shoulder instability. Am J Sports Med. 2024;52(10)
    [Google Scholar]

Fulltext Views
114

PDF downloads
2
View/Download PDF
Download Citations
BibTeX
RIS
Show Sections

Suggested read for related articles: