Sex Does Not Impact Outcomes of Arthroscopic Treatment of Hip Labral Tears: A Systematic Review and Meta‐analysis Authors: Anne Gordon, Jacob Adams, Reed Coda, Grace Tideman, Damon Mar, Lisa Vopat, Bryan G Vopat, John Paul Schroeppel, Rachel Long, Sterre Van Den Anker Published: 2026 ABSTRACT AbstractPurposeTo conduct a systematic review and meta‐analysis to evaluate sex‐specific outcomes after arthroscopic treatments for hip labral tears.MethodsA review of literature was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses guidelines. Risk of Bias in Non‐Randomized Studies‐Intervention tool was used for the risk‐of‐bias assessment. Eligible studies included patients who underwent arthroscopic hip labral repair or reconstruction, reported sex‐based outcomes, and were published between 2005 and 2024. Exclusion criteria included studies not written in English, cadaveric and animal studies, and studies that failed to stratify outcomes based on sex. The outcomes of interest were modified Harris Hip Score, Hip Outcome Score‐Sport Scale score, Visual Analog Scale for Pain score, revision rates, and return‐to‐sport (RTS) rates.ResultsThere were no statically significant differences in Hip Outcome Score‐Sport Scale, modified Harris Hip Score, revision rates, and RTS rates. Men and women had similar RTS rates (odds ratio, 1.27 [95% confidence interval, 0.65; 2.48], P = .45, I 2 = 0%). Men had lower revision rates between 0% and 7.9% compared with female rates between 0% and 17%, but this difference was not significant (odds ratio, 0.44 [95% confidence interval, 0.23; 0.83], P = .19, I 2 = 38%). There was a statistically significant difference in Visual Analog Scale for Pain scores, with slightly higher scores in women; however, the standardized mean difference was small (mean difference, 0.08 [95% confidence interval, −0.24; 0.41], P = .05, I 2 = 67%).ConclusionsThe equality of outcomes between sexes and the low rates of revision, high rates of RTS, low pain scores, and high functional scores suggest that arthroscopic treatments are beneficial to both sexes. There was no significant difference in functional scores, revision rates, or RTS rates between men and women undergoing arthroscopic treatment for labral tears. The practical relevance of the statistically significant difference in Visual Analog Scale for Pain scores is negligible.Level of EvidenceLevel IV, systematic review and meta‐analysis of Level I to IV studies. Acetabular labral tears are a common pathology, especially in athletes, with 22% to 55% of patients with hip or groin pain receiving a diagnosis of an acetabular labral tear. 1 It has been suggested that femoroacetabular impingement syndrome, the abnormal articulation of the femoral head with the acetabular rim, contributes to acetabular labral tears and other chondrolabral injuries. 1 , 2 Although there are a number of nonsurgical treatment options for labral tears, hip arthroscopy is an established, effective surgery that has been shown to improve patient‐reported outcomes. 3 , 4 According to Colvin et al., there was an 18‐fold increase between 1999 and 2009 in hip arthroscopies. 5 Notwithstanding the prevalence of the procedure and the evidenced success, few studies take into consideration sex‐specific differences in reporting patient outcomes. 6 Despite biological and phenotypical sexual dimorphisms, there is currently a lack of female‐specific research in sports medicine, and instead, male‐dominant research guides treatment protocols for both sexes. 6 It is problematic to extrapolate male‐dominant data to women because these sexual dimorphisms can influence biomechanics, athletic performance, and the body's response to exercise. 7 Past studies have investigated sex differences in hip arthroscopy. In 1 study, preoperative anatomy, hip function scores, and self‐reported impairment scores differed significantly between male and female patients, with women having lower alpha angles, lower functionality scores, and higher pain scores. 3 These differences may impact postoperative outcomes. Specifically researching outcomes after hip arthroscopy, Beck et al. found that female patients reached minimal clinically important difference in patient‐reported outcomes at higher rates than male patients after hip arthroscopy for femoroacetabular impingement syndrome. 8 Conflictingly, Cvetanovich et al. found that patients of both sexes reported similar improvements in outcomes and both had low complication rates. 9 A similar study reviewing sex‐based differences in outcomes after hip arthroscopy with labral reconstruction found that women who underwent hip arthroscopy reported greater improvements in outcome. 10 However, both men and women surpassed the minimal clinically significant difference threshold; thus, the study concluded that both sexes benefit from hip arthroscopy. 10 The conflicting evidence of sex‐based differences in hip labral injuries and outcomes on hip labral repair necessitates greater research. The purpose of this study was to conduct a systematic review and meta‐analysis to evaluate sex‐specific outcomes after arthroscopic treatments for hip labral tears. We hypothesized that postoperative outcomes are equivalent between male and female patients after arthroscopic labral repair and reconstruction. METHODS Search Strategy and Study Selection A systematic review of the literature was conducted using PubMed, Cochrane, Embase, and Web of Science databases. A systematic review will be conducted according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses guidelines. 11 A search was performed in November 2024 using the following search terms: (hip OR acetabulum OR acetabular) AND (labrum* OR labral) AND (arthroscopy* OR arthroscopic repair OR arthroscopic reconstruction) AND (outcomes* OR prognosis* OR complication*) AND (male* OR female* OR male vs female OR male vs female). The University of Kansas Medical Center Institutional Review Board (IRB) considers systematic reviews exempt from needing IRB approval. Therefore, IRB approval was not required to publish this case. Eligibility Criteria Studies were included if they met the following criteria: focus on hip labral repair and/or reconstruction, focus on living human participants, outcomes are differentiated by sex, and studies written originally in the English language. Exclusion criteria included studies with a biomechanical focus, not written in English, cadaveric studies, animal studies, failure to stratify subjects based on sex, systematic reviews, editorials, case reports, and studies published before 2005. It was chosen to select studies published after 2005 to ensure similar operative techniques, and equipment was used in the arthroscopies conducted among the different studies. 12 Data Extraction/Analysis Articles were exported to Rayyan systematic review software, and duplicates were removed. Titles and abstracts were screened by A.G. and J.A. according to the inclusion and exclusion criteria. The studies included were then reviewed in full text by A.G., J.A., and G.T. according to the eligibility criteria. If any conflicts arose during this process, the senior author (B.G.V.) was consulted for the final decision on inclusion or exclusion. Data from the articles was recorded in a spreadsheet, including demographics, measures of outcome, surgical technique, and follow‐up time. Random‐effects meta‐analyses were conducted on 9 studies that included both male and female participants and had over a 2‐year follow‐up period. Two‐year postoperative outcomes were recorded based on the modified Harris Hip Score (mHHS), Hip Outcome Score‐Sport Scale (HOS‐SS) score, Visual Analog Scale for Pain (VAS Pain) score, revision rates, and return to sport (RTS). A meta‐analysis of 5‐year postoperative scores of HOS‐SS and mHHS was also included. The remaining 10 studies were reviewed and compared with the meta‐analysis values. Risk‐of‐Bias Assessment To assess the risk of bias for the studies included in the meta‐analysis, the authors used the Risk of Bias in Non‐randomized Studies‐Intervention (ROBINS‐I) assessment tool. 13 Each study was ranked as having a low, moderate, serious, or critical risk of bias in 7 domains. A.G. and G.T. separately completed the assessments, and discrepancies were reviewed and discussed to come to a consensus conclusion. RESULTS Literature Search A total of 1681 articles were identified in the initial search including 535 from PubMed, 38 from Cochrane, 809 from Embase, and 299 from Web of Science. A total of 710 duplicates were excluded, leaving 971 articles for further review. After abstract screening was complete, 686 articles were excluded, leaving 285 articles for full‐text review. After the full‐text manuscripts were reviewed, 16 articles met the inclusion criteria and were included in this review, as shown in Table 1. Nine of these studies were included in the meta‐analyses for several outcome measures. The meta‐analyses included a total of 2232 participants (mean age 31.4 years), 910 men and 1322 women. Figure 1 denotes the flow diagram of this literature search. TABLE 1Studies Included in Systematic Review Author, Year Level of Evidence Study Design Total Number of Hips Mean Age, yr Number of Men Number of Women Treatment Modality Mean Length of Follow‐Up Inclusion in Meta‐analysis Alter et al. 2022 14 IIICohort9833.1 ± 12.82969Repair, debridement2 yrYesGlein et al. 2021 17 IIICohort13625.67373Repair, reconstruction, few debridement2 yrYesLebus et al. 2018 21 IVCase series31135.3170141Reconstruction3.7 yr (range, 2.0‐11.3)YesLocks et al. 2018 15 IVCase series1135 (range, 20‐51)65Reconstruction5.4 yr (range, 1‐10)YesMaldonado et al. 2022 22 IIIRetrospective comparative observation trial132631.6466860Repair4.9 ± 2.4 yrYesMenge et al. 2021 23 IVCase series7016 ± 1.22149Repair10 yrYesOwens et al. 2022; “Elite female athletes” 18 IIICohort13024.66565Repair, reconstruction, few debridement5.6 ± 0.5 yrYesOwens et al. 2022; “Sex‐based differences” 10 IIICohort5840.42929Reconstruction2 yrYesPhilippon et al. 2020 16 IVTherapeutic8238.7 ± 11.45131Reconstruction10 yrYesBoykin et al. 2013 19 IVCase series2328230Reconstruction3.5 yr (range, 1.8‐6.2)NoBrooks et al. 2012 25 IIProspective cohort1023‐36010Repair4 moNoChahla et al. 2019 26 IIICohort60033.5 ± 12.3213387Reconstruction2 yrNoDean et al. 2024 24 IIICohort17437.8 ± 11.28688Repair10 yrNoDrager et al. 2020 27 IIICase‐control34631.497249Repair1 yrNoKrych et al. 2013 28 IProspective, randomized1838018Repair1 yrNoPhilippon et al. 2010 20 IVCase series2827280Repair2 yrNomo, months; yr, years. Studies Included in Systematic Review mo, months; yr, years. Flow diagram of study selection following the PRISMA guidelines. (PRISMA, Preferred Reporting Items for Systematic Reviews and Meta‐Analyses.) Study Quality and Risk‐of‐Bias Assessment The ROBINS‐I tool indicated that most of the studies included in the meta‐analyses have a low‐moderate risk of bias (see Figure 2). 13 The tool uses 4 levels of risk of bias—low, moderate, serious, and critical. Three studies, Atler et al., Locks et al., and Philippon et al., had a serious risk of bias. 14 , 15 , 16 These studies did not propensity‐matched subjects, leaving room for bias because of confounding variables. However, the age ranges are relatively narrow, 20 to 51 years for Locks et al., 33.1 ± 12.8 years for Alter et al., and 38.7 ± 11.4 years for Philippon et al. Therefore, the lack of matching is unlikely to be a significant concern for confounding bias. Additionally, several studies performed reconstructions or repairs, depending on the medical necessity of the patient, and a few elected to only selectively debride. This adds another confounding variable to patient outcomes, increasing the risk of bias. A table of the results for each study is shown in Table 2. Forest plot of studies included in the meta‐analysis of RTS rates in players attempting to RTS. (CI, confidence interval; OR, odds ratio; RTS, return to sport.) TABLE 2Risk‐of‐Bias Assessment Using the ROBINS‐I Tool Study Risk‐of‐Bias Domains D1 D2 D3 D4 D5 D6 D7 Overall Alter et al. 2022 14 SeriousLowLowLowLowLowLowSeriousGlein et al. 2021 17 ModerateLowLowLowLowLowLowModerateLebus et al. 2018 21 ModerateLowLowLowLowLowLowModerateLocks et al. 2018 15 SeriousLowLowLowLowLowLowSeriousMaldonado et al. 2022 22 ModerateLowLowLowModerateLowLowModerateMenge et al. 2021 23 ModerateLowLowLowLowLowLowModerateOwens et al. 2022 Elite female athletes 18 ModerateLowLowLowLowLowLowModerateOwens et al. 2022 Sex differences 10 LowLowLowLowLowLowLowLowPhilippon et al. 2020 16 SeriousLowLowLowLowLowLowSerious Note: Domains: D1, bias because of confounding; D2, bias because of selection of participants; D3, bias in classification of the interventions; D4, bias because of deviations from intended interventions; D5, bias because of missing data; D6, bias in measurement of outcomes; D7, bias in selection of the reported result.ROBINS‐I, Risk of Bias in Non‐randomized Studies‐Intervention. Risk‐of‐Bias Assessment Using the ROBINS‐I Tool Note: Domains: D1, bias because of confounding; D2, bias because of selection of participants; D3, bias in classification of the interventions; D4, bias because of deviations from intended interventions; D5, bias because of missing data; D6, bias in measurement of outcomes; D7, bias in selection of the reported result. ROBINS‐I, Risk of Bias in Non‐randomized Studies‐Intervention. Another limitation to the risk‐of‐bias assessment is that we were unable to find scoring criteria that align well with this project's comparison of male and female outcomes because gender is not an intervention nor an exposure. The use of the ROBINS‐I tool applied the intervention of surgery; however, there could be additional bias in the comparison of the effects of gender that was not illuminated by this tool. 13 Therefore, the overall risk of bias for each source may be different than indicated with the ROBINS‐I tool. Objective Outcomes RTS Rates Men and women were found to RTS at equal rates, with male rates ranging from 77.9% to 96.6% compared with female rates between 78.1% and 93.1% (odds ratio, 1.27 [95% confidence interval (CI), 0.65; 2.48], P = .45, I 2 = 0%). 10 , 17 , 18 For the meta‐analyses, RTS was defined as athletes returning to their previous sport at any level and only included those who attempted to RTS. In 2 male‐only studies from Boykin et al. and Philippon et al., the RTS was 85.7% and 100%, respectively. 19 , 20 Philippon et al. found that their cohort of professional hockey players returned to skating and drills in an average of 3.8 months after surgery. 20 A forest plot showing the results and weight for RTS rates is shown in Figure 2. Revision Rate Across 5 studies, although men had a lower mean revision rate, ranging from 0% to 7.9% compared with 0% to 17% revision rates in women, there was no significant difference in revision rates between sexes (odds ratio, 0.44 [95% CI, 0.23; 0.83], I 2 = 38%, P = .19). 15 , 16 , 21 , 22 , 23 This was further divided by surgery type, repair, or reconstruction to ascertain whether the type of surgery influences the outcomes. The test for subgroup differences using the random‐effects model showed no significant difference between reconstruction and repair (χ 2 = 3.38, P = .31). Additionally, the low male revision rates were corroborated by 2 other studies of male participants, Boykin et al. and Philippon et al., with male revision rates of 4.3% and 7.14%, respectively. Philippon et al. found that both the participants requiring revision surgery reinjured the hip rather than the revision being from a complication or lack of healing from the first labral repair. 19 , 20 A forest plot showing the results and weight for RTS rates is shown in Figure 3. Forest plot of studies included in the meta‐analysis of revision rates, subdivided by treatment modality (labral reconstruction or repair). (CI, confidence interval; OR, odds ratio.) Conversion Rates to Total Hip Arthroplasty Only 6 studies reported conversion rate to Total Hip Arthroplasty (THA), and 5 of them had mean lengths of follow‐up less than 6 years. In Dean et al., men and women had similar conversion rates to THA; 20.9% (18/86) of men and 18.2% (16/88) of women underwent conversion to THA within 10 years of having an arthroscopic hip labral repair. 24 The 5 other studies, with shorter follow‐up periods, reported rates of THA between 0% and 17.6% for men and 0% and 8.5% for women. 15 , 18 , 19 , 21 , 22 Table 3 presents the THA rates and mean follow‐up times. TABLE 3Rates of Conversion to Total Hip Arthroplasty After Hip Labral Repair or Reconstruction Author, Year Follow‐Up Avg, yr Surgery Male Total Female Total Male THA Female THA Boykin et al. 2013 19 3.5 (range, 1.7‐6.2)Reconstruction232 (4.3%)Dean et al. 2024 24 11.1 ± 2.5Repair868818 (20.9%)16 (18.2%)Lebus et al. 2018 21 3.7 (range, 2.0‐11.3)Reconstruction17014130 (17.6%)12 (8.5%)Locks et al. 2018 15 5.4 (range, 1‐10)Reconstruction650 (0.0%)0 (0.0%)Maldonado et al. 2022 22 4.9 ± 2.4Repair46686029 (6.2%)38 (4.4%)Owens et al. 2022, “Elite female athletes” 18 5.6 ± 0.5Reconstruction, repair, debridement65653 (4.6%)2 (3.08%)Avg, average; THA, total hip arthroplasty; yr, years. Rates of Conversion to Total Hip Arthroplasty After Hip Labral Repair or Reconstruction Avg, average; THA, total hip arthroplasty; yr, years. Subjective Outcomes Modified Harris Hip Score and Hip Outcome Score‐Sport Scale Based on meta‐analyses of 3 studies, men and women reported similar scores for mHHS and HOS‐SS at 2 years postoperatively, and there was no significant difference. 14 , 17 , 22 The mHHS assesses the general hip functionality, and HOS‐SS assesses hip function in sports‐related activities. Both scores are from 0 to 100, with 100 being the best possible score. Forest plots showing the results and weight for mHHS and HOS‐SS are shown in Figures 4 and 5, respectively. Forest plot of studies included in the meta‐analysis of mHHS. (CI, confidence interval; mHHS, modified Harris Hip Score; SD, standard deviation; SMD, standardized mean difference.) Forest plot of studies included in the meta‐analysis of HOS‐SS. (CI, confidence interval; HOS‐SS, Hip Outcome Score‐Sports Scale; SD, standard deviation; SMD, standardized mean difference.) Two years after surgery, mHHS scores were high and ranged from 78.9 to 89.1 in men and 85.4 to 86.2 in women. Using the random‐effects model, the standardized mean difference (SMD) for mHHS was very small (0.03 [95% CI, −0.26; 0.33], P = .07, I 2 = 62%). The meta‐analysis of HOS‐SS also resulted in a very small SMD and minimal heterogeneity (0.05 [95% CI, −0.06; 0.15], P = .43, I 2 = 0%). HOS‐SS scores were moderately high, ranging from 69.1 to 78.3 for men and 75.2 to 78 for women. The 5‐year follow‐up cohort from Maldonado et al. reported slightly higher mean scores for mHHS (90.1 for men, 87.7 for women) and HOS‐SS (81.5 for men, 78.3 for women). 22 The other reviewed studies reported mHHS and HOS‐SS value similar to those of meta‐analysis. 19 , 20 , 25 , 26 , 27 , 28 In a study on postpartum runners, Brooks et al. reported that women had a mean mHHS of 84.3 only 4 months after surgery. 25 In a 2010 study by Philippon et al., elite male hockey players 2 years postsurgery had a mean mHHS of 95. 20 Visual Analog Scale for Pain The VAS Pain scores measure pain intensity from 0 to 10, with 0 being no pain and 10 being the most pain. At 2 years postoperatively, there was a statistically significant difference in VAS Pain scores between men and women in 3 studies, with slightly higher scores in women. 14 , 17 , 22 However, the SMD was small (SMD, 0.08 [95% CI, −0.24; 0.41], P = .05, I 2 = 67%); thus, this finding is likely not clinically significant. Glein and Maldonado showed close pain scores between men and women (within 0.2), with women reporting slightly higher pain. Conversely, Alter et al. reported higher pain in men than women (2.8 vs 1.7). 14 The Drager et al. article, not included in the meta‐analysis because of a short follow‐up time, found similar VAS Pain scores for men and women 1 year after surgery (mean scores for normal labrums were 1.36 ± 1.41 for men and 2.03 ± 1.93 for women), suggesting a negligible difference in pain after 1 year postoperation. 27 A forest plot of the results and weight for VAS Pain is shown in Figure 6. Forest plot of studies included in the meta‐analysis of VAS Pain scores. (CI, confidence interval; SD, standard deviation; SMD, standardized mean difference; VAS Pain, Visual Analog Scale for Pain.) DISCUSSION The findings of this systematic review and meta‐analysis suggest that men and women have advantageous and equal outcomes—assessed through RTS rate, revision rate, VAS Pain, mHHS, and HOS‐SS—after undergoing arthroscopic treatment for hip labral tears. Patients who underwent arthroscopic treatment for hip labral tears had low pain scores, high RTS rates, high hip functionality scores, and low revision rates. Despite differences in anatomical structure, men and women showed statically and clinically similar results regarding all subjective patient‐reported outcome scores, RTS, and revision rate. The reasons for the revisions were not divided by sex; however, Lebus et al., Menge et al., and Philippon et al. reported several causes for revision with the most being because of another labral tear, lysis or release of adhesions, capsular laxity, psoas entrapment, and cam and pincer deformities. 16 , 21 , 23 Maldonado et al. did not report the causes of reinjury. 22 The reasoning for this may be of clinical significance. Perhaps women are more prone to reinjury, require more extensive rehabilitation before RTS causing labral fragility, or have an anatomic/physiologic difference not recognized. 6 , 7 Women were found to have lower alpha angles and higher levels of hip dysplasia which may lead to more frequent hip subluxations and reinjury of the repaired labrum. 3 The anatomical acetabular defect that led to the labral tear was not fixed in these studies, potentially accounting for reinjury and lower RTS rates in women. Additionally, RTS rates may differ based on the sport to which the athlete is returning, necessitating further research into the RTS rates based on sport after arthroscopic treatment for hip labral tears. Several studies included the rate of conversion to THA; however, all but 1 (Dean et al.) had a less than 6‐year follow‐up, possibly making these rates misrepresentative on the long‐term effects of having a previous hip labral repair or reconstruction. 24 For VAS Pain, the statistically significant finding that women reported higher pain levels after 2 years postoperatively may not be clinically significant because the SMD is very small. Additionally, both male and female patients reported higher mHHS than HOS‐SS. Although these are 2 separate scales, both are out of 100, with 100 being the best possible score. Therefore, these findings could suggest that arthroscopic hip labral repair or reconstruction yields better results in general hip functionality than sports‐specific hip functionality. This score difference also likely reflects that HOS‐SS requires higher demands to be met to achieve a high score than mHHS given its focus on the hip's function in athletics rather than daily activities. The fact that both genders have higher mHHS scores suggests that the anatomical differences did not impact functionality; rather, the nature of the hip use (sports or general) is of great influence on subjective outcomes. The 5‐year follow‐up cohort from Maldonado et al. reported slightly higher mean scores for mHHS and HOS‐SS compared with those of the 2‐year postoperative meta‐analysis, suggesting that hip function may slightly improve after 2 years postoperation. 22 Both hip labral repairs and reconstructions were included in this systematic review. These are different procedures and thus may have different recovery processes and outcomes. Although no difference was found when a subdivision meta‐analysis was performed for RTS, the difference in treatment might influence other measures of outcome. Future studies should focus on 1 surgical treatment or divide their outcomes by surgical treatment to eliminate the potential confounding variable of different surgeries. Limitations A limitation to this review was the relatively short follow‐up. This review analyzed data after 2‐year follow‐up, aiming to provide adequate time for recovery and rehabilitation to get a clear picture of the long‐term effects of hip labral repairs and reconstructions. The impact of hip labral repair and reconstruction of joint longevity and rate of THA was not able to be systematically reviewed because of the lack of data. Additionally, other confounding variables include individual variations in cartilage damage, acetabular shapes or dysplasia, and surgical techniques such as capsular management. These variations were rarely reported, so inclusion of these variables was not possible. CONCLUSIONS The equality of outcomes between sexes and the low rates of revision, high rates of RTS, low pain scores, and high functional scores suggest that arthroscopic treatments are beneficial to both sexes. There was no significant difference in functional scores, revision rates, or RTS rates between men and women undergoing arthroscopic treatment for labral tears. The practical relevance of the statistically significant difference in VAS Pain scores is negligible, as evidenced by the small mean difference. DISCLOSURES The authors (D.M., L.V., B.G.V., J.P.S.) declare the following financial interests/personal relationships which may be considered as potential competing interests: D.M. reports a relationship with Agada Medical that includes consulting or advisory and equity or stocks and Texas Back Institute that includes consulting or advisory. L.V. reports a relationship with American Orthopaedic Foot and Ankle Society that includes board membership, Stryker that includes consulting or advisory, Altior that includes equity or stocks, Carbon 22 that includes equity or stocks, Ortho Bullets that includes equity or stocks, and Spinal Simplicity that includes equity or stocks. B.G.V. reports a relationship with American Orthopaedic Foot and Ankle Society that includes board membership, Stryker that includes consulting or advisory, Altior that includes equity or stocks, Carbon 22 that includes equity or stocks, and Spinal Simplicity that includes equity or stocks. J.P.S. reports a relationship with Arthrex that includes consulting or advisory and Smith & Nephew that includes consulting or advisory. The other authors (A.G., J.A., R.C., G.T., R.L., S.V.D.A.) declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article. REFERENCES