The heritability of doctor-diagnosed traumatic and degenerative meniscus tears

  • K. Magnusson
    Correspondence
    Address correspondence and reprint requests to: K. Magnusson, Clinical Epidemiology Unit, Orthopaedics Department of Clinical Sciences Lund, Lund University Remissgatan 4, Wigerthuset, 221 85, Lund, Sweden. Tel.: 47-991-16-463.
    Affiliations
    Lund University, Faculty of Medicine, Department of Clinical Sciences Lund, Orthopaedics, Clinical Epidemiology Unit, Lund, Sweden

    National Advisory Unit on Rehabilitation in Rheumatology, Department of Rheumatology, Diakonhjemmet Hospital, Oslo, Norway
    Search for articles by this author
  • A. Turkiewicz
    Affiliations
    Lund University, Faculty of Medicine, Department of Clinical Sciences Lund, Orthopaedics, Clinical Epidemiology Unit, Lund, Sweden
    Search for articles by this author
  • B. Snoeker
    Affiliations
    Lund University, Faculty of Medicine, Department of Clinical Sciences Lund, Orthopaedics, Clinical Epidemiology Unit, Lund, Sweden
    Search for articles by this author
  • V. Hughes
    Affiliations
    Lund University, Faculty of Medicine, Department of Clinical Sciences Lund, Orthopaedics, Clinical Epidemiology Unit, Lund, Sweden
    Search for articles by this author
  • M. Englund
    Affiliations
    Lund University, Faculty of Medicine, Department of Clinical Sciences Lund, Orthopaedics, Clinical Epidemiology Unit, Lund, Sweden

    Clinical Epidemiology Research and Training Unit, Boston University School of Medicine, Boston, MA, USA
    Search for articles by this author
Open AccessPublished:March 16, 2021DOI:https://doi.org/10.1016/j.joca.2021.03.005

      Summary

      Objective

      To estimate the genetic contribution to traumatic and degenerative meniscus tears for men and women across the lifespan.

      Methods

      We linked the Swedish Twin Register with individual-level national healthcare data to form a 30-year, population-wide, longitudinal twin cohort. To study genetic contribution to meniscus tears, we estimated the heritability and familial risk using incident traumatic and degenerative tear diagnostic codes in a cohort of 88,414 monozygotic and dizygotic twin-pairs, aged ≥17 years.

      Results

      During follow-up, 3,372 (3.8%) of 88,414 twins were diagnosed with a traumatic or degenerative meniscus tear. The heritability was 0.39 (95% CI = 0.32–0.47) for men and 0.43 (95% CI = 0.36–0.50) for women, and did not vary by age. Environmental factors that were unique to each twin in a pair explained a greater proportion of the variance than genetic factors, both for men (0.61, 95% CI = 0.53–0.68) and women (0.57, 95% CI = 0.50–0.64). Separate analyses of traumatic vs degenerative meniscus tears yielded similar results.

      Conclusion

      For the first time, we have estimated the genetic contribution to doctor-diagnosed meniscus tears using a twin study design. We found a relatively low to modest heritability for meniscus tears (∼40%). The heritability was also fairly stable over the lifespan, and equal in both men and women. Our findings suggest that environmental risk factors are a more important contributor to both traumatic and degenerative doctor-diagnosed meniscus tears than genetic factors.

      Keywords

      Background

      Meniscus tears are among the most common intra-articular knee pathologies. Traumatic meniscus tears most typically occur as a consequence of acute trauma to the knee in situations related to sports participation, usually in adolescents or young adults
      • Snoeker B.A.
      • Bakker E.W.
      • Kegel C.A.
      • Lucas C.
      Risk factors for meniscal tears: a systematic review including meta-analysis.
      . Risk factors for traumatic meniscus tears are believed to be similar to those for anterior cruciate ligament (ACL) injury, e.g., playing ground surface, malalignment, and hypermobility of ligaments
      • Smith H.C.
      • Vacek P.
      • Johnson R.J.
      • Slauterbeck J.R.
      • Hashemi J.
      • Shultz S.
      • et al.
      Risk factors for anterior cruciate ligament injury: a review of the literature-part 2: hormonal, genetic, cognitive function, previous injury, and extrinsic risk factors.
      . In contrast, degenerative meniscus tears often occur at middle-age or older, and are typically not associated with traumatic knee injury. Instead, they develop gradually, and are frequently observed in both symptomatic and asymptomatic knees
      • Kumm J.
      • Roemer F.W.
      • Guermazi A.
      • Turkiewicz A.
      • Englund M.
      Natural history of intrameniscal signal intensity on knee MR images: six years of data from the osteoarthritis initiative.
      ,
      • Englund M.
      • Guermazi A.
      • Gale D.
      • Hunter D.J.
      • Aliabadi P.
      • Clancy M.
      • et al.
      Incidental meniscal findings on knee MRI in middle-aged and elderly persons.
      . Traumatic and degenerative meniscus tears have been strongly linked with subsequent development of radiographic and symptomatic knee osteoarthritis (OA)
      • Kumm J.
      • Roemer F.W.
      • Guermazi A.
      • Turkiewicz A.
      • Englund M.
      Natural history of intrameniscal signal intensity on knee MR images: six years of data from the osteoarthritis initiative.
      , a trait with high familial clustering due to shared genetics
      • Zhai G.
      • Hart D.J.
      • Kato B.S.
      • MacGregor A.
      • Spector T.D.
      Genetic influence on the progression of radiographic knee osteoarthritis: a longitudinal twin study.
      ,
      • Magnusson K.T.A.
      • Englund M.
      Nature vs. nurture in knee osteoarthritis – the importance of age, sex and body mass index.
      .
      Mapping the genetics responsible for traumatic and degenerative tears in the meniscus is important for preventing these meniscus tears, which are an important risk factor for knee OA development. However, there is not much known on the topic, and a systematic review and meta-analysis did not find any studies that specifically investigated familial risk factors for meniscus tears
      • Snoeker B.A.
      • Bakker E.W.
      • Kegel C.A.
      • Lucas C.
      Risk factors for meniscal tears: a systematic review including meta-analysis.
      . A previous cohort study has reported increased risk for meniscal extrusion and tears in the offspring of persons with end-stage knee OA
      • Pan F.
      • Khan H.
      • Ding C.
      • Winzenberg T.
      • Martel-Pelletier J.
      • Pelletier J.P.
      • et al.
      Familial effects on structural changes relevant to knee osteoarthritis: a prospective cohort study.
      . However, the genetic contribution to meniscus tears has not been specifically quantified before. Twin studies are exceptionally well suited to explore the genetic contribution to a trait, because they allow for the quantification of the total genetic vs environmental contribution to variance in that trait
      • Rijsdijk F.V.
      • Sham P.C.
      Analytic approaches to twin data using structural equation models.
      . Thus, using the world's largest twin registry, we aimed to estimate the genetic contribution to traumatic and degenerative meniscus tears, for men and women across the lifespan.

      Methods

       Data sources

      The Swedish Twin Registry (STR) is a national twin registry covering all twins born in Sweden (95–100%)
      • Magnusson P.K.
      • Almqvist C.
      • Rahman I.
      • Ganna A.
      • Viktorin A.
      • Walum H.
      • et al.
      The Swedish Twin Registry: establishment of a biobank and other recent developments.
      . The STR was linked to the National Patient Register (NPR) in 2017, forming a population-based longitudinal cohort including 88,414 monozygotic (MZ) and dizygotic (DZ) same-sex and opposite-sex twins born 1911–1999. The NPR contains diagnostic codes of musculoskeletal disease obtained in specialist care from 1987 and onwards, including data on every hospitalization (from 1987 to 2016), 1-day surgery (from 1997 to 2016) and specialist outpatient care visit (from 2001 to 2016). These data were additionally linked to individual-level data on death and education level, obtained from the National Board of Health and Welfare and the longitudinal integration database for health insurance and labor market studies (LISA) from Statistics Sweden. Information on body mass index (BMI, i.e., weight/height2 in units of kg/m2) was self-reported on postal questionnaires at up to six occasions during 1961–2002. The study was approved by the Ethical Review Board at Lund University, Sweden.

       Study design

      We included twins aged 17 years or older with a minimum follow-up time of 1 year in the NPR. Both twins in the pair were required to be alive for at least the first year of follow-up. Follow-up began on the date the twins turned 17 years, or in 1987 for twins who had already turned 17 years before coverage began in the NPR. This design ensures that left truncation was always similar for both twins in a pair.

       Outcome –meniscus tear

      Meniscus tears were reported in the NPR according to the Swedish translations of ICD-9 and ICD-10 from 1987 to 2016. We defined a person as having a traumatic meniscus tear if diagnosed at least once with an ICD-9 code 717A-717E or ICD-10 codes S83.2 (traumatic meniscal tear) or S83.7 (injury to multiple structures of knee (including the meniscus)). Similarly, we defined a person as having a degenerative meniscus tear if diagnosed at least once with an ICD-9 code 836A-836C or ICD-10 codes M23.2 (old meniscal tear) or M23.3 (other changes to the meniscus like degeneration, instability and loose meniscus body). For all diagnoses, we included codes for both the medial and lateral compartments, as well as codes for unspecified compartment. We did not study diagnoses involving specific joint diseases, damage to the cartilage, or other specific diagnoses like cystic or discoid meniscus. Our primary outcome of interest was any meniscus tear (classified as either traumatic or degenerative). We also studied the codes for traumatic vs degenerative meniscus tears in separate analyses of the two traits, independent of the age at which the diagnoses were obtained.
      A twin was registered as having the outcome at the date of first diagnosis of a meniscus tear. Twins who did not have a meniscus tear prior to study end were censored at December 31st 2016. Twins who died during follow-up were censored at the date of death.

       Statistical analyses

      Because MZ twins share 100% of their genes, and DZ twins share 50%, any differences in within-pair co-occurrence in a trait between MZ and DZ twins can be utilized to inform on the likely contribution of genetic factors to the trait.
      In estimating genetic heritability, we applied the classical twin model, a so-called ACE/ADE-model using liability thresholds (accounting for censoring using inverse probability weighting as previously recommended)
      • Rijsdijk F.V.
      • Sham P.C.
      Analytic approaches to twin data using structural equation models.
      ,
      • Scheike T.H.
      • Holst K.K.
      • Hjelmborg J.B.
      Estimating twin concordance for bivariate competing risks twin data.
      ,
      • McGue M.
      When assessing twin concordance, use the probandwise not the pairwise rate.
      . “A” represents additive genetic factors, “C” represents shared environmental factors, “D” represents dominant genetic factors and “E” represents environmental factors unique to each twin in a pair. The sum of these components (A + C + D + E) represents the total phenotypic variance. Within this model, intra-pair correlations within MZ and DZ twin pairs are utilized to calculate the heritability which is the percentage variance in disease risk that can be explained by genetic factors. We compared different models (i.e., ACE vs ADE vs AE) using rMZ:rDZ intra-pair correlation ratios (2:1 indicates mainly additive genetics, 3:1 indicates mainly dominant genetics) as well as Akaike Information Criteria (AIC) to find the model that could most optimally quantify the contribution of each of these factors and provide the best model fit. Thus, the type of model with the best fit also indicates which variance components were of relevance. For example, if an ACE model provides the best fit, this would imply that additive genetic factors, shared environmental factors, and unique environmental factors are all important contributors to meniscus tear. Because heritability estimates may be inflated by common risk factors to the trait, that to a large extent are shared by twins, we adjusted the heritability analyses for available covariates, i.e., the latest achieved education level and the earliest available BMI
      • Snoeker B.A.
      • Bakker E.W.
      • Kegel C.A.
      • Lucas C.
      Risk factors for meniscal tears: a systematic review including meta-analysis.
      . In a separate analysis, we also stratified our outcome based on traumatic or degenerative meniscus tears.
      Furthermore, we calculated the casewise concordance, familial risk (relative recurrence risk ratio), and tetrachoric correlations for the MZ and DZ twin pairs as previously recommended for administratively censored twin data, i.e., applying a flexible model without assumptions of the underlying ACDE structure or probability weighting
      • Scheike T.H.
      • Holst K.K.
      • Hjelmborg J.B.
      Estimating twin concordance for bivariate competing risks twin data.
      ,
      • McGue M.
      When assessing twin concordance, use the probandwise not the pairwise rate.
      . The cumulative casewise concordance is the probability of a co-twin having a meniscus injury given that their twin has already had a meniscus injury. Similarly, the familial risk is the excess risk of meniscus injury for a twin given that their co-twin has already been injured, relative to the marginal (population) risk (i.e., the overall risk for the injury in the study sample, treating it as a population of unrelated individuals). The tetrachoric correlation is the intra-pair correlation between two theorized normally distributed continuous latent variables, from two observed binary variables (twin 1's meniscus tear and twin 2's meniscus tear).
      All analyses were performed separately for men and women, and across the lifespan to inform on potential sex and age differences in heritability. Thus, opposite-sex DZ twins were not included in the main heritability analyses. We also assessed potential quantitative and qualitative sex differences by comparing the described dependency measures between opposite-sex and same-sex male and female dizygotic twin pairs.
      All analyses were performed using the METS package in R (The R Project for Statistical Computing), v. 3.4.1.

      Results

       Descriptive data

      The study sample consisted of 88,414 twins, with a mean age of 32 at the start of follow-up (Table I) and has been described elsewhere
      • Magnusson K.T.A.
      • Englund M.
      Nature vs. nurture in knee osteoarthritis – the importance of age, sex and body mass index.
      ,
      • Magnusson K.
      • Turkiewicz A.
      • Hughes V.
      • Frobell R.
      • Englund M.
      High genetic contriution to anterior cruciate ligament rupture: heritability ∼69%.
      . During follow-up, 3,372 (3.8%) of the twins were diagnosed with a meniscus tear, whereas 17,382 (19.7%) died without a diagnosis of meniscus tear. The incidence rate of doctor-diagnosed meniscus tear per 100,000 person-years was 68 (95% CI = 66–71, Table II). The majority of incident diagnoses were degenerative meniscus tears (n = 2062, 61.2%). Also, 384 (11.4%) persons first had a traumatic meniscus tear diagnosis, followed by a later diagnosis of a degenerative meniscus tear.
      Table IParticipants' characteristics at the start of follow-up
      AllMenWomen
      No meniscus tearMeniscus tearNo meniscus tearMeniscus tear
      Twins, individual leveln=88,414n=39,431n=1,79545,611n=1,577
      Age at start of follow-up, median (IQR)32 (17–47)32 (17–47)30 (17–38)33 (17–49)30 (17–40)
      Primary school (≤9 years), n (%)
      Indicates highest level of education achieved.
      26,612 (30.1)12,422 (31.5)347 (19.3)13,608 (29.8)235 (14.9)
      Upper secondary school (1–4 years), n (%)
      Indicates highest level of education achieved.
      36,335 (41.1)16,669 (42.3)844 (47.0)18,115 (39.7)707 (44.8)
      College/university (>1 years), n (%)
      Indicates highest level of education achieved.
      25,467 (28.8)10,340 (26.2)604 (33.7)13,888 (30.5)635 (40.3)
      BMI, kg/m2, mean (SD)23.3 (3.5)23.9 (3.3)24.2 (3.3)22.7 (3.7)22.9 (3.7)
      Twins, pair levelN=44,207
      MZ pairs, N14,5315,9055137,605508
      DZ pairs, same-sex, N17,4087,4625998,789558
      DZ pairs, opposite-sex, N
      Twin pairs consisting of one male and one female twin, i.e., the same N is presented for men and women.
      12,26811,2681,00011,2681,000
      MZ, monozygotic; DZ, dizygotic; BMI, body mass index. Lower case n represents individuals, upper case N represents pairs. Pairs with meniscus tears are pairs with one or more twins having a meniscus tear.
      Indicates highest level of education achieved.
      Twin pairs consisting of one male and one female twin, i.e., the same N is presented for men and women.
      Table IIIncidence rates of doctor-diagnosed meniscus tear per 100,000 person-years, across the lifespan for men and women in the Swedish Twin Register
      MenWomen
      Person-yearsNRate (95% CI)Person-yearsNRate (95% CI)
      18–20 years60,180154256 (219–300)65,891173263 (226–305)
      20–40 years238,800481201 (184–220)276,400373135 (122–150)
      40–60 years334,300850254 (238–272)376,100698186 (172–200)
      60–80 years234,700296126 (113–141)290,300322111 (99–124)
      > 80 years31,2791445 (27–76)58,0861119 (10–34)
      N, number of incident cases of meniscus tears.

       Heritability

      When applying the classical twin model, the AE models including only additive genetics showed the best model fit (lowest AIC) with the most precise estimates. From the start of follow-up at age 17 years, the heritability was relatively constant across the lifespan [Fig. 1(A)]. Overall, the genetic heritability adjusted for BMI and education level was modest, at 0.39 (95% CI = 0.32–0.47) for men and 0.43 (95% CI = 0.36–0.50) for women. Unique environmental factors explained a greater proportion of the variance both for men (heritability = 0.61, 95% CI = 0.53–0.68) and women (heritability = 0.57, 95% CI = 0.50–0.64).
      Fig. 1
      Fig. 1A) The heritability of meniscal tears by age at diagnosis, with 95% confidence intervals. B) The familial risk (the relative recurrence risk ratio) of meniscal tear by age at diagnosis, for monozygotic (solid line) and dizygotic twins (dashed line, includes both same-sex and opposite-sex pairs), with 95% confidence intervals. The dotted line represents the reference population risk assuming genetically unrelated individuals.

       Traumatic vs degenerative meniscus tears

      A similar pattern was observed when meniscus tears, classified as traumatic and degenerative, were studied as separate traits (Table III, Table IV). Consistently, the intra-pair correlations suggested additive rather than dominant genetics, if any. For meniscus tears diagnosed as traumatic, the intra-pair correlation ratios, i.e., rMZ:rDZ were 0.40 (95% CI = 0.24–0.51): 0.19 (95% CI = −0.01–0.37) in male pairs, and 0.37 (95% CI = 0.22–0.50): 0.23 (95% CI = 0.05–0.40) in female pairs. For meniscus tears classified as degenerative, the rMZ:rDZ ratio was 0.36 (95% CI = 0.25–0.46): 0.26 (95% CI = 0.15–0.37) in men, and 0.42 (95% CI = 0.32–0.51): 0.30 (95% CI = 0.18–0.41) in women. There were only minor differences in the estimated genetic vs environmental components in the trait-specific analysis, when compared to studying the two traits combined. Environmental factors unique to each twin explained the largest part of the variance both in traumatic and degenerative meniscus tears, with education- and BMI-adjusted variance component estimates of 0.56–0.62 (Table IV).
      Table IIINumber of twin pairs discordant and concordant for traumatic and degenerative meniscus injury
      MenWomen
      DZMZDZMZ
      N = 8,062N = 6,418N = 9,347N = 8,113
      Traumatic meniscus tear
       Discordant pairs, N (%)209 (0.03)202 (0.03)219 (0.02)217 (0.03)
       Concordant pairs, N (%)4 (0.00)12 (0.00)5 (0.00)11 (0.00)
      Degenerative meniscus tear
       Discordant pairs, N (%)237 (0.03)355 (0.06)380 (0.04)332 (0.04)
       Concordant pairs, N (%)21 (0.00)27 (0.00)18 (0.00)27 (0.00)
      N; number of twin pairs in sex strata, DZ; dizygotic twins, MZ; monozygotic twins.
      Table IVModel fit and variance component estimates for the classical twin model for having a diagnostic code of traumatic vs degenerative meniscus tear in the Swedish Twin Registry
      ModelVariance components estimatesAIC
      A (95% CI)C (95% CI)D (95% CI)E (95% CI)
      Men - traumatic meniscus tear
       ACE0.37 (0.00–0.85)0.01 (0.00–0.42)0.62 (0.49–0.76)6,753.42
       ADE0.38 (0.25–0.50)0.00 (0.00–0.00)0.62 (0.50–0.75)6,753.42
       AE0.38 (0.250.50)0.62 (0.500.75)6,751.42
      Men - degenerative meniscus tear
       ACE0.17 (0.00–0.47)0.16 (0.00–0.40)0.66 (0.56–0.77)11,606.95
       ADE0.37 (0.28–0.46)0.00 (0.00–0.00)0.63 (0.54–0.72)11,608.69
       AE0.37 (0.280.46)0.63 (0.540.72)11,606.69
      Women - traumatic meniscus tear
       ACE0.25 (0.00–0.69)0.11 (0.00–0.49)0.64 (0.50–0.77)6,824.83
       ADE0.38 (0.25–0.50)0.00 (0.00–0.00)0.62 (0.50–0.75)6,825.16
       AE0.37 (0.250.50)0.62 (0.500.75)6,823.16
      Women - degenerative meniscus tear
       ACE0.21 (0.00–0.51)0.20 (0.00–0.44)0.59 (0.49–0.69)10,342.29
       ADE0.44 (0.36–0.53)0.00 (0.00–0.00)0.56 (0.47–0.64)10,344.73
       AE0.43 (0.350.53)0.56 (0.470.64)10,342.73
      A: Additive genetic factors, C: Common environmental factors, D: Dominant genetic factors, E: Unique environmental factors. AIC: Akaike Information Criteria. Adjusted for education level and body mass index. Bold represents estimates for the best fitting model.

       Age and sex differences in familial dependency

      Applying a flexible model without assumptions of the underlying ACDE structure, we found only minor differences in familial risk of any meniscus injury in MZ vs DZ twin pairs [Fig. 1(B)]. Further, we found that casewise concordances, familial risks and tetrachoric correlations overlapped for MZ and DZ male and female twins, again indicating that mainly environmental rather than genetic factors are involved in the etiology of doctor-diagnosed meniscus tears for both men and women (Table V). The pattern with similar tetrachoric correlations in same-sex male and same-sex female DZ pairs indicate no quantitative sex differences (i.e., the genetic and environmental influences are likely to account for proportionally the same amount of variance in meniscus tear for males and females). Also, we observed no sex-differences when comparing opposite-sex DZ twins with same-sex male and female DZ twins (Table V), further implying no qualitative sex differences (i.e., the genetic or shared environmental factors influencing meniscus tear did not differ for males and females)The correlational patterns consistently suggested additive genetics rather than dominant genetics, if any (i.e., rMZ:rDZ ratio = 2:1 rather than 3:1) (Table V).
      Table VMarginal risk and estimates of dependency and heritability of doctor-diagnosed meniscal tears for MZ and DZ men and women
      MenWomenOpposite sex
      DZ (N = 8,061)MZ (N = 6,418)DZ (N = 9,347)MZ (N = 8,113)DZ (N = 12,268)
      Pairs concordant for mensicus tear, N (%)31/599 (5.2)48/513 (9.4)29/558 (5.2)45/508 (8.9)41/1,000 (4.1)
      Marginal risk, 95% CI0.04 (0.04–0.04)0.04 (0.04–0.04)0.03 (0.03–0.03)0.03 (0.03–0.03)0.04 (0.04–0.04)
      Casewise concordance, probability, 95% CI0.10 (0.07–0.14)0.17 (0.13–0.22)0.10 (0.07–0.14)0.16 (0.13–0.21)0.08 (0.06–0.10)
      Familial risk (relative recurrence risk ratio), 95% CI2.52 (1.72–3.32)3.92 (2.98–4.85)3.15 (2.11–4.18)4.78 (3.59–5.96)1.86 (1.33–2.38)
      Tetrachoric correlation, r, 95% CI0.23 (0.14–0.32)0.39 (0.30–0.47)0.27 (0.18–0.37)0.41 (0.33–0.49)0.15 (0.07–0.23)
      Heritability, 95% CI0.39 (0.31–0.47)0.42 (0.35–0.50)N/A
      The marginal risk is the population risk, i.e., the risk of being diagnosed with a meniscus tear without taking familial dependency into account. The casewise concordance is the risk of a twin being diagnosed with a cruciate ligament injury given that the co-twin has already been diagnosed with a tear. The familial risk is the excess risk of meniscus tear for a twin given that the co-twin has had a tear. The tetrachoric correlation is the intra-pair correlation between two theorized normally distributed continuous latent variables, from two observed binary variables (meniscus tear in twin one and meniscus tear in twin 2) and is utilized for calculating the heritability. Heritability estimates are from an AE-model.

      Discussion

      Our findings suggest that environmental risk factors are a more important contributor to both traumatic and degenerative doctor-diagnosed meniscus tears than genetic factors. We found a relatively low to modest heritability for such meniscus tears, which was fairly stable over the lifespan, and similar in both men and women.
      To our knowledge, this is the first study to estimate the genetic contribution to doctor-diagnosed meniscus tears using a twin study design. Consequently, we are unable to offer a direct comparison of our findings with previous twin studies on this topic. However, the results can be compared with our recent twin study that reported a relatively high genetic contribution to end-stage knee OA, including important sex differences – women showed a higher and more consistent genetic contribution to end-stage knee OA than men
      • Magnusson K.T.A.
      • Englund M.
      Nature vs. nurture in knee osteoarthritis – the importance of age, sex and body mass index.
      . In the current study, we found rather low genetic contribution and no sex differences in the risk for meniscus tears. One possible explanation for the difference in genetic contribution to end-stage knee OA vs meniscus tears, which are a risk factor for the former, may be that genetic factors and sex differences do not play a role in the initial development of meniscal tears, but may be involved in the later development of OA
      • Englund M.
      • Guermazi A.
      • Roemer F.W.
      • Aliabadi P.
      • Yang M.
      • Lewis C.E.
      • et al.
      Meniscal tear in knees without surgery and the development of radiographic osteoarthritis among middle-aged and elderly persons: the Multicenter Osteoarthritis Study.
      . Another explanation for this difference between the genetic contribution to end-stage knee OA and meniscus tears may be that the doctor-diagnosed meniscal tears in the current study are a selection of meniscus tears, i.e., primarily meniscal tears in patients with knee symptoms. Asymptomatic meniscal tears, which often precede knee OA, may be largely undiagnosed in the general population
      • Englund M.
      • Guermazi A.
      • Gale D.
      • Hunter D.J.
      • Aliabadi P.
      • Clancy M.
      • et al.
      Incidental meniscal findings on knee MRI in middle-aged and elderly persons.
      . Therefore, it is possible we may find a greater genetic contribution to meniscus tears if asymptomatic meniscus tears were better captured in the study sample. Indeed, a previous study has reported that the offspring of persons with total knee replacement for end-stage knee OA had a higher risk of meniscal extrusion and meniscal tears as visualized on MRI
      • Pan F.
      • Khan H.
      • Ding C.
      • Winzenberg T.
      • Martel-Pelletier J.
      • Pelletier J.P.
      • et al.
      Familial effects on structural changes relevant to knee osteoarthritis: a prospective cohort study.
      . Thus, it would have been desirable to use meniscal tears defined by MRI screening as an alternative outcome in this study. However, this is unfortunately not feasible to accomplish in a large nation-wide study sample as in this study.
      Still, our results convincingly suggest that environmental factors are the major contributor to risk for doctor-diagnosed meniscal tears, for both men and women across the lifespan, adjusted for the frequently shared factors of education level and BMI. In addition to our analysis of all meniscus tears, we also studied traumatic and degenerative meniscus tears separately, as the two traits may have different risk factors. We found no significant differences in the genetic contribution to the two different tear types in our trait-specific analysis. Yet, the low genetic contribution to traumatic tears is in contrast to our recent findings of a high genetic contribution to ACL rupture in the same sample of Swedish twins
      • Magnusson K.
      • Turkiewicz A.
      • Hughes V.
      • Frobell R.
      • Englund M.
      High genetic contriution to anterior cruciate ligament rupture: heritability ∼69%.
      . Whereas the study of ACL rupture may have clear clinical consequences, i.e., implying that clinicans may ask athletes whether there is a history of ACL ruptures in the nearest family in attempts to prevent future ACL rupture, the present results suggest that knowledge of the family history of meniscus tears may not be equally important for the prevention of these tears. Here, lifestyle factors such as obesity and occupational exposures may be more important
      • Rytter S.
      • Jensen L.K.
      • Bonde J.P.
      • Jurik A.G.
      • Egund N.
      Occupational kneeling and meniscal tears: a magnetic resonance imaging study in floor layers.
      . However, considering the unclear distinction between normal aging, degenerative meniscus tears, and osteoarthritis
      • Englund M.
      • Guermazi A.
      • Gale D.
      • Hunter D.J.
      • Aliabadi P.
      • Clancy M.
      • et al.
      Incidental meniscal findings on knee MRI in middle-aged and elderly persons.
      ,
      • Magnusson K.T.A.
      • Englund M.
      Nature vs. nurture in knee osteoarthritis – the importance of age, sex and body mass index.
      ,
      • Magnusson K.
      • Turkiewicz A.
      • Hughes V.
      • Frobell R.
      • Englund M.
      High genetic contriution to anterior cruciate ligament rupture: heritability ∼69%.
      , we suggest that the reasons for the differing findings for ACL rupture, meniscus tears and knee OA need further attention, i.e., in order to improve the total understanding of the genetics of knee health throughout the lifespan. As an example, certain gene expressions have been suggested to be linked with increased risk for progression to OA, which may also explain part of the heritability observed for development of the tear itself.
      • Lamplot J.D.
      • Rai M.F.
      • Tompkins W.P.
      • Friedman M.V.
      • Schmidt E.J.
      • Sandell L.J.
      • et al.
      Gene expression in meniscal tears at the time of arthroscopic partial meniscectomy predicts the progression of osteoarthritis within 6 Years of surgery.
      We should state some limitations of our study. Firstly, the twin design may be vulnerable to violations of the equal environment assumption. We cannot exclude that MZ twins share their environment to a greater extent than DZ twins, or are treated more similarly by their parents. Along this line, we had no available data on several risk factors to meniscus tears that may be shared for twins in a pair, such as work-related activity like kneeling, squatting and climbing stairs
      • Snoeker B.A.
      • Bakker E.W.
      • Kegel C.A.
      • Lucas C.
      Risk factors for meniscal tears: a systematic review including meta-analysis.
      . This may have inflated our heritability estimates, and we suggest future studies should include adopted twins or use more sophisticated models that can help disentangle any potential gene–gene and gene–environment interactions
      • Mayhew A.J.
      • Meyre D.
      Assessing the heritability of complex traits in humans: methodologial challenges and opportunities.
      ,
      • Buil A.
      • Brown A.A.
      • Lappalainen T.
      • et al.
      Gene-gene and gene-environment interactions detected by transcriptome sequence analysis in twins.
      . Secondly, we may have underestimated the concordance of meniscus tears for twin pairs that were not covered by the NPR at young ages. As an example, those who were included from a relatively old age may have experienced meniscus tears that were not captured in our data and time frame. However, we expect that any potential underestimation in this group will be similar for MZ and DZ twins, and thus not substantially impact our heritability estimates. Thirdly, in our trait-specific analysis of traumatic vs degenerative meniscus tears, results should be interpreted with care due to the low number of pairs concordant for traumatic meniscus tears among younger individuals, i.e., those who would be expected to have true traumatic meniscal tears. Misclassification of diagnostic codes may also have affected the results for degenerative tears, as any diagnostic code of a traumatic meniscus tear occurring after age 35–40 may in fact have been a degenerative tear and vice versa
      • Bergkvist D.
      • Dahlberg L.E.
      • Neuman P.
      • Englund M.
      Knee arthroscopies: who gets them, what does the radiologist report, and what does the surgeon find? An evaluation from southern Sweden.
      . Lastly, as already mentioned, we only capture meniscus tears that have been diagnosed in health care. Ascertainment of meniscus status by e.g., magnetic resonance imaging would have been preferable, but unfortunately unfeasible in this large dataset. An important strength of our study is that it covers 95–100% of the twin population of Sweden
      • Magnusson P.K.
      • Almqvist C.
      • Rahman I.
      • Ganna A.
      • Viktorin A.
      • Walum H.
      • et al.
      The Swedish Twin Registry: establishment of a biobank and other recent developments.
      and is thus most likely representative for industrialized countries with a health care system giving equal access to all inhabitants.
      In conclusion, we found that genetic factors only had a limited impact on the risk of having a doctor-diagnosed traumatic or degenerative meniscus tear. Overall, the genetic heritability adjusted for BMI and education level was modest, at 0.39 (95% CI = 0.32–0.47) for men and 0.43 (95% CI = 0.36–0.50) for women. Environmental factors that are unique to each twin appear to be a more important contributor to risk for both of these types of tears.

      Author contribution

      Karin Magnusson had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Karin Magnusson performed statistical analyses and drafted the manuscript. Aleksandra Turkiewicz, Barbara Snoeker, Velocity Hughes and Martin Englund contributed with acquisition of data, conceptual design, and analysis and interpretation of data. All authors contributed in drafting the article or critically revising it for important intellectual content. All authors gave final approval for the version to be submitted.

      Competing interest statements

      We declare no conflicts of interest, except for Dr. Englund receiving grants from the Swedish Research Council , the Österlund Foundation, Governmental Funding of Clinical Research within National Health Service (ALF), the Greta and Johan Kock Foundation, and the Swedish Rheumatism Association , during the conduct of the study.

      Role of the funding source

      The funding sources had no influence on the design or conduct of the study, the collection, management, analysis, or interpretation of the data, the preparation, review, or approval of the manuscript, or the decision to submit the manuscript for publication.

      Acknowledgements

      We would like to acknowledge The Swedish Twin Registry and Barbro Sandin and Patrik Magnusson for access to the data and data management. The Swedish Twin Registry is managed by Karolinska Institute and receives funding through the Swedish Research Council under grant no. 2017-00641 . The study was funded by the Swedish Research Council ( E0234801 ), the Greta and Johan Kock Foundation , the Swedish Rheumatism Association , the Österlund Foundation, Governmental Funding of Clinical Research within National Health Service (ALF) and the Faculty of Medicine, Lund University, Sweden.

      Reference

        • Snoeker B.A.
        • Bakker E.W.
        • Kegel C.A.
        • Lucas C.
        Risk factors for meniscal tears: a systematic review including meta-analysis.
        J Orthop Sports Phys Ther. 2013; 43: 352-367
        • Smith H.C.
        • Vacek P.
        • Johnson R.J.
        • Slauterbeck J.R.
        • Hashemi J.
        • Shultz S.
        • et al.
        Risk factors for anterior cruciate ligament injury: a review of the literature-part 2: hormonal, genetic, cognitive function, previous injury, and extrinsic risk factors.
        Sport Health. 2012; 4: 155-161
        • Kumm J.
        • Roemer F.W.
        • Guermazi A.
        • Turkiewicz A.
        • Englund M.
        Natural history of intrameniscal signal intensity on knee MR images: six years of data from the osteoarthritis initiative.
        Radiology. 2016; 278: 164-171
        • Englund M.
        • Guermazi A.
        • Gale D.
        • Hunter D.J.
        • Aliabadi P.
        • Clancy M.
        • et al.
        Incidental meniscal findings on knee MRI in middle-aged and elderly persons.
        N Engl J Med. 2008; 359: 1108-1115
        • Zhai G.
        • Hart D.J.
        • Kato B.S.
        • MacGregor A.
        • Spector T.D.
        Genetic influence on the progression of radiographic knee osteoarthritis: a longitudinal twin study.
        Osteoarthr Cartil. 2007; 15: 222-225
        • Magnusson K.T.A.
        • Englund M.
        Nature vs. nurture in knee osteoarthritis – the importance of age, sex and body mass index.
        Osteoarthr Cartil. 2019; 27: 586-592
        • Pan F.
        • Khan H.
        • Ding C.
        • Winzenberg T.
        • Martel-Pelletier J.
        • Pelletier J.P.
        • et al.
        Familial effects on structural changes relevant to knee osteoarthritis: a prospective cohort study.
        Osteoarthr Cartil. 2015; 23: 559-564
        • Rijsdijk F.V.
        • Sham P.C.
        Analytic approaches to twin data using structural equation models.
        Briefings Bioinf. 2002; 3: 119-133
        • Magnusson P.K.
        • Almqvist C.
        • Rahman I.
        • Ganna A.
        • Viktorin A.
        • Walum H.
        • et al.
        The Swedish Twin Registry: establishment of a biobank and other recent developments.
        Twin Res Hum Genet. 2013; 16: 317-329
        • Scheike T.H.
        • Holst K.K.
        • Hjelmborg J.B.
        Estimating twin concordance for bivariate competing risks twin data.
        Stat Med. 2014; 33: 1193-1204
        • McGue M.
        When assessing twin concordance, use the probandwise not the pairwise rate.
        Schizophr Bull. 1992; 18: 171-176
        • Magnusson K.
        • Turkiewicz A.
        • Hughes V.
        • Frobell R.
        • Englund M.
        High genetic contriution to anterior cruciate ligament rupture: heritability ∼69%.
        Br J Sports Med. 2021; 55: 385-389
        • Englund M.
        • Guermazi A.
        • Roemer F.W.
        • Aliabadi P.
        • Yang M.
        • Lewis C.E.
        • et al.
        Meniscal tear in knees without surgery and the development of radiographic osteoarthritis among middle-aged and elderly persons: the Multicenter Osteoarthritis Study.
        Arthritis Rheum. 2009; 60: 831-839
        • Rytter S.
        • Jensen L.K.
        • Bonde J.P.
        • Jurik A.G.
        • Egund N.
        Occupational kneeling and meniscal tears: a magnetic resonance imaging study in floor layers.
        J Rheumatol. 2009; 36: 1512-1519
        • Lamplot J.D.
        • Rai M.F.
        • Tompkins W.P.
        • Friedman M.V.
        • Schmidt E.J.
        • Sandell L.J.
        • et al.
        Gene expression in meniscal tears at the time of arthroscopic partial meniscectomy predicts the progression of osteoarthritis within 6 Years of surgery.
        Orthop J Sport Med. 2020; https://doi.org/10.1177/2325967120936275
        • Mayhew A.J.
        • Meyre D.
        Assessing the heritability of complex traits in humans: methodologial challenges and opportunities.
        Curr Genom. 2017; 18: 332-340
        • Buil A.
        • Brown A.A.
        • Lappalainen T.
        • et al.
        Gene-gene and gene-environment interactions detected by transcriptome sequence analysis in twins.
        Nat Genet. 2015; 47: 88-91
        • Bergkvist D.
        • Dahlberg L.E.
        • Neuman P.
        • Englund M.
        Knee arthroscopies: who gets them, what does the radiologist report, and what does the surgeon find? An evaluation from southern Sweden.
        Acta Orthop. 2016; 87: 12-16