Author + information
- Received November 16, 2011
- Revision received April 16, 2012
- Accepted May 11, 2012
- Published online October 23, 2012.
- Costan G. Magnussen, PhD⁎,†,⁎ (, )
- Juha Koskinen, MD, PhD⁎,
- Markus Juonala, MD, PhD⁎,‡,
- Wei Chen, MD, PhD§,
- Sathanur R. Srinivasan, PhD§,
- Matthew A. Sabin, MD, PhD∥,
- Russell Thomson, PhD†,
- Michael D. Schmidt, PhD¶,
- Quoc Manh Nguyen, MD, MPH§,
- Ji-Hua Xu, MD, PhD§,
- Michael R. Skilton, PhD#,
- Mika Kähönen, MD, PhD⁎⁎,
- Tomi Laitinen, MD, PhD††,
- Leena Taittonen, MD, PhD‡‡,§§,
- Terho Lehtimäki, MD, PhD∥∥,
- Tapani Rönnemaa, MD, PhD‡,
- Jorma S.A. Viikari, MD, PhD‡,
- Gerald S. Berenson, MD§ and
- Olli T. Raitakari, MD, PhD⁎,¶¶
Reprint requests and correspondence:
Dr. Costan Magnussen, Menzies Research Institute Tasmania, Medical Science 1, 17 Liverpool Street, Hobart 7000, Australia
Objectives The aim of this study was to examine the effect of resolution from metabolic syndrome (MetS) between youth and adulthood on carotid artery intima-media thickness (IMT) and type 2 diabetes mellitus (T2DM).
Background Published findings demonstrate that youth with MetS are at increased risk of cardio-metabolic outcomes in adulthood. It is not known whether this risk is attenuated in those who resolve their MetS status.
Methods Participants (n = 1,757) from 2 prospective cohort studies were examined as youth (when 9 to 18 years of age) and re-examined 14 to 27 years later. The presence of any 3 components (low high-density lipoprotein cholesterol, high triglycerides, high glucose, high blood pressure, or high body mass index) previously shown to predict adult outcomes defined youth MetS; the harmonized MetS criteria defined adulthood MetS. Participants were classified according to their MetS status at baseline and follow-up and examined for risk of high IMT and T2DM.
Results Those with MetS in youth and adulthood were at 3.4 times the risk (95% confidence interval: 2.4 to 4.9) of high IMT and 12.2 times the risk (95% confidence interval: 6.3 to 23.9) of T2DM in adulthood compared with those that did not have MetS at either time-point, whereas those that had resolved their youth MetS status by adulthood showed similar risk to those that did not have MetS at either time-point (p > 0.20 for all comparisons).
Conclusions Although youth with MetS are at increased risk of adult high IMT and T2DM, these data indicate that the resolution of youth MetS by adulthood can go some way to normalize this risk to levels seen in those who have never had MetS.
The metabolic syndrome (MetS) is considered as a clustering of multiple interrelated metabolic irregularities often including obesity (particularly central), insulin resistance, dyslipidemia, hypertension, and hyperglycemia (1). The importance of the diagnosis above and beyond identification and treatment of its component parts, however, is controversial in children and adolescents (herein termed youth) (2,3) as well as adults (4,5). A recent consensus statement (2) has called for more research, particularly in longitudinal studies from youth to adulthood, on a number of areas relating to pediatric MetS before a definition for the clinical diagnosis of MetS among youth is considered. Some of the unwillingness to issue a definite definition is due to concerns over the demonstrated short-term instability of a categorical diagnosis of MetS in the pediatric setting (6,7). Further to this, however, is that—irrespective of instability—the ability to predict future disease status is critically important. We and others have shown that a number of pediatric MetS definitions predict important outcomes later in life, including carotid intima-media thickness (IMT), type 2 diabetes mellitus (T2DM), and cardiovascular morbidity (3,8–11). We have also shown that 6-year spontaneous resolution from MetS in young to middle adulthood has a beneficial impact on structural and functional markers of preclinical atherosclerosis, compared with those who had persistent MetS (12). A necessary extension of this work that has not previously been reported would be to examine the impact of resolution of the MetS during the life-course on adult disease outcomes. With the current absence of data from long-term clinical trials that span youth and adulthood on this issue (13), these types of analyses would significantly contribute to the published data and provide insight into whether or not interventions to tackle pediatric MetS might be clinically relevant. Therefore we sought to determine whether resolution of the MetS between youth and adult life is associated with changes in carotid IMT and T2DM risk.
The analysis sample included participants from the Bogalusa Heart Study (14) and the Cardiovascular Risk in Young Finns Study (15) where MetS risk factor variables were measured in youth (baseline) and again in adulthood (follow-up). For the Bogalusa study, data related to 376 youth 9 to 18 years of age who had participated in either the 1984 to 1985 or 1987 to 1988 youth surveys and attended either the 2001 to 2002 or 2003 to 2007 adult surveys (then 25 to 41 years of age). For the Young Finns study, this represented 1,381 youth 9 to 18 years of age who attended the 1986 youth survey and either the 2001 or 2007 adult follow-ups (then 24 to 39 years of age). These baseline and follow-up samples were selected because: 1) glucose screening first commenced in the Young Finns study in the 1986 survey and in the Bogalusa study from the 1984–85 survey; 2) the youngest participant in the 1986 Young Finns study was 9 years old, so for consistency, we limited the baseline Bogalusa sample to those 9 years of age or older; and 3) in the case of the adult follow-ups, they were the most consistent between studies and minimized differences in length of follow-up. For individuals that participated in multiple baseline (Bogalusa) or follow-up surveys, we used those measures that provided the longest period between baseline and follow-up. All analyses were restricted to those that did not have type 1 diabetes and female subjects who were not pregnant at follow-up. Both studies received ethical approval, and written informed consent was obtained from participants. Study details have been previously described in detail (14,15). We encourage readers to view the Online Appendix for more comprehensive Methods.
Definition of youth MetS
In the absence of a consensus pediatric MetS definition (6), we used the definition that we have previously shown to predict adult outcomes (3). Body mass index (BMI) was used as the measure of adiposity, because waist circumference was not measured at baseline in either cohort. Briefly, we generated age-, sex-, race- (Bogalusa only), cohort-, and study-year-specific z-scores of BMI, systolic and diastolic blood pressures, high-density lipoprotein (HDL) cholesterol, triglycerides, and glucose from each of the complete cohorts. A participant was categorized as having MetS if they had any 3 of the following 5 components: BMI ≥75th percentile, systolic or diastolic blood pressure ≥75th percentile, HDL-cholesterol ≤25th percentile, triglycerides ≥75th percentile, or glucose ≥75th percentile. We note that use of other pediatric MetS definitions that we have previously found to predict adult outcomes (3) did not modify the conclusions drawn.
Definition of adulthood MetS
Adult MetS was classified according to the “harmonized” definition proposed in a joint statement from the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; the American Heart Association; the World Heart Federation; the International Atherosclerosis Society; and the International Association for the Study of Obesity (1). MetS was identified when 3 or more of the following 5 criteria were present: waist circumference ≥102 cm in men or ≥88 cm in women, triglycerides ≥1.695 mmol/l (≥150 mg/dl or specific drug treatment for elevated triglycerides), HDL-cholesterol <1.036 mmol/l (<40 mg/dl) in men or <1.295 mmol/l (<50 mg/dl) in women (or specific drug treatment for reduced HDL-cholesterol), blood pressure ≥130/≥85 mm Hg (or antihypertensive drug treatment in those with a history of hypertension), fasting plasma glucose ≥5.6 mmol/l (100 mg/dl or specific drug treatment of elevated glucose). The joint statement acknowledges that the definition of central adiposity is yet to be finalized but recommends that either the former (lower threshold) International Diabetes Federation or (higher threshold) American Heart Association/National Heart, Lung, and Blood Institute cut-points be used. We chose the higher cut-point for waist circumference in our data, because these coincide with definitions of abdominal obesity used in the United States and Europe (16).
Derivation of continuous MetS risk score in youth and adulthood
A continuous metabolic syndrome (cMetS) risk score was created for baseline and follow-up with the methods described by Wijndaele et al. (17,18), which we have previously detailed (3). A higher cMetS score indicates a less favorable MetS profile (17).
Classification of high carotid IMT in adulthood
As we have previously reported (3,19), the most consistent IMT measurement available across both study centers incorporated the maximum measurement at the far wall of the left common carotid artery. We defined high IMT in adulthood as a maximum IMT ≥90th percentile for age, sex, race (Bogalusa), cohort, and study year specific values (3,19).
Classification of type 2 diabetes in adulthood
Type 2 diabetes mellitus was classified if participants: 1) had a fasting plasma glucose ≥7.0 mmol/l (126 mg/dl); 2) reported treatment with oral hypoglycemic agents and/or insulin injections without a diagnosis of type 1 diabetes; or 3) reported a history of physician-diagnosed T2DM. Women who reported having physician-diagnosed diabetes only during the term of their pregnancy were considered as having had gestational diabetes and were classified as currently not having T2DM (20).
Participants were classified into 4 groups according to their MetS status at baseline and follow-up: “control group” (no MetS at baseline or at follow-up); “resolution group” (MetS at baseline but not at follow-up); “incident group” (MetS at follow-up but not at baseline); and “persistent group” (MetS both at baseline and follow-up). This approach has been adopted in our previous work (12). All analyses were performed with STATA (version 10, StataCorp, College Station, Texas).
Baseline and Follow-Up Characteristics Among Participants in Each MetS Group
A series of linear regression models was used to estimate differences in continuous baseline and follow-up risk factor levels according to MetS group with the control group as the reference category. These models allow comparisons at both time-points on differences in risk factor levels between participants in the control MetS group and those in the other MetS groups. All linear regression models were adjusted for age, sex, and race (Bogalusa). Because of right-skewed distribution, triglyceride and insulin levels were log-transformed before analysis. Comparisons for sex and race (Bogalusa) across MetS groups were performed with chi-square analyses.
Change in Risk Factor Levels Between Youth and Adulthood Among Participants in Each MetS Group
Change in continuous metabolic risk components and lifestyle variables was defined as the difference between adult and youth levels. For smoking, a 4-level variable that indicated smoking status at baseline and follow-up was developed. For change in socioeconomic position, a social mobility variable was created (21), with highest level of parental education at baseline and highest level of own education at follow-up. Multinomial logistic regression, which estimates effects for outcomes with multiple attributes, was used to examine the effect of changes in risk factor variables between youth and adulthood on MetS group. Age-, sex-, and race (Bogalusa)-adjusted models were fitted for each predictor variable with the control group as the reference outcome category. Because the baseline level of the risk factor might have an effect on the magnitude of change, we also included the baseline variable as a covariate and examined for interactions between the baseline variable and change (no significant interactions were found).
Effect of Baseline and Follow-Up Mets Status on Carotid IMT and T2DM in Adulthood
Relative risks and 95% confidence intervals were estimated with Poisson regression with robust standard errors and were used to examine associations between MetS groups and outcomes of adult high IMT and adult T2DM. Analyses were performed for cohort-stratified and cohort-pooled data. All estimates were adjusted for age, sex, race (Bogalusa), and length of follow-up (19). For pooled estimates, we additionally adjusted for cohort. Cohort × MetS group, race × MetS group, age × MetS group, and sex × MetS group interaction effects in the pooled models were tested for all outcomes (no significant interactions were found). In addition to the pooled estimates, random-effects meta-analysis (22) techniques using the “metan” module in STATA was used. To complement the categorical analyses, we also examined the effect of decreasing cMetS score between youth and adulthood (derived as youth cMetS score minus adult cMetS score) on high IMT and T2DM in adulthood. These models included the same covariates as outlined in the preceding text but also included the baseline cMetS score to account for any effect of baseline level on the magnitude of change.
The length of follow-up between baseline and follow-up was 24.4 ± 3.7 years, ranging from 14 to 27 years. Type 2 diabetes mellitus was present at follow-up in 32 Bogalusa participants (4 black male subjects, 11 black female subjects, 4 white male subjects, 13 white female subjects; prevalence in blacks = 11.7%, whites = 6.7%, overall = 8.5%) and 11 Young Finns participants (4 male and 7 female, prevalence = 0.8%).
Baseline and follow-up characteristics among participants in each MetS group
Table 1 displays baseline and follow-up characteristics of participants according to MetS groups for each study. Eighty-two (22%) Bogalusa and 231 (17%) Young Finns participants had MetS at baseline. The MetS had resolved in 54 (66%) Bogalusa and 145 (61%) Young Finns participants by adulthood. Across both studies, participants in the resolution group tended to have a substantially worse metabolic profile at baseline compared with those in the control group, but this between-group difference became nonsignificant by follow-up. Although there were some baseline risk factor differences between the resolution and persistent groups—in these instances the persistent group had poorer metabolic profiles—this difference was exacerbated at follow-up.
Change in risk factor levels between youth and adulthood among participants in each MetS group
Table 2 shows change in risk factors levels between youth and adulthood across the 4 MetS groups for each study. Compared with the control group, participants with incident and persistent MetS tended to experience significant increased gains in weight, BMI, systolic and diastolic blood pressure, triglycerides, and glucose, whereas HDL cholesterol levels had reduced overall. The resolution groups in both studies experienced a net reduction in cMetS score that exceeded the control group and was in contrast to the substantial gains in cMetS score experienced by participants in both incident and persistent groups. Taken together, the resolution groups tended to show improvement in risk factor levels at least equivalent to (and in some instances greater than) the control groups, whereas the incident and persistent groups experienced a shift toward increased risk factors compared with their counterparts. Of the lifestyle factors examined, most notable in the Young Finns study was a net gain in physical activity in the resolution group relative to their peers in the incident and persistent groups (both of which tended to decrease physical activity levels) and an increased fruit consumption relative to their peers in the incident group.
The effect of baseline and follow-up MetS status on carotid IMT and T2DM in adulthood
For high IMT, those who had resolved from youth MetS were at similar risk as those that did not have MetS at both time-points in pooled analyses (Table 3) and meta-analyses (Fig. 1). However, those who had incident or persistent MetS were at substantially increased risk of developing high IMT in adulthood compared with those in the control group in pooled (Table 3) and meta (Fig. 1) analyses. A 1SD decrease in cMetS score between youth and adulthood showed a consistent 20% reduction in risk for high adult IMT (Table 3).
Youth in the resolution group were not at increased risk of T2DM when compared with the control group in a pooled analysis (Table 4) and meta analysis (Fig. 2) analyses. However, those who had incident or persistent MetS were at substantially increased risk of developing T2DM in adulthood, compared with those in the control group in pooled analyses (Table 4) and meta-analyses (Fig. 2). In pooled analyses, a 1-SD decrease in cMetS score between youth and adulthood was associated with a 50% reduced risk for T2DM in adulthood. The results were essentially similar when using an alternate outcome that combined impaired fasting glucose (fasting glucose ≥6.1 mmol/l [≥110 mg/dl]) and T2DM (Online Fig. 1, Online Table 1).
We repeated the aforementioned pooled analyses after separately adjusting for change in weight or BMI z-score or an adiposity resolution variable we have shown to predict these outcomes (23). Adjustment for change in weight or BMI z-score did not appreciably modify the effect estimates shown in Tables 3 and 4 (data not shown). Adjustment for adiposity resolution somewhat attenuated the effect estimates comparing persistent with control groups for high IMT and T2DM, but both remained statistically significant (data not shown).
A recent scientific statement from the American Heart Association called for additional work on delineating the clinical utility of identifying and intervening in youth with MetS (2). Our data highlighted 3 key aspects concerning the putative utility of identifying youth MetS. First, the prevalence of high carotid IMT and T2DM in adulthood among those that resolved from youth MetS was not significantly different from those that never had MetS, whereas those with persistent MetS were at substantially increased risk. Second, our data showed a high-rate of resolution from MetS, approximately two-thirds of individuals, highlighting the fluidity over time that is seen after the diagnosis of MetS in youth. Third, individuals that resolved from MetS tended to lose weight and increase physical activity levels relative to their peers that had persistent MetS. Although we acknowledge that the observational nature of our study only allows comparison of those where the MetS spontaneously resolved with the other groups, the findings could have clinical relevance in that efforts to prevent persistence of pediatric MetS into adulthood and the acquisition of MetS in adulthood might translate to reductions in later cardio-metabolic risk.
We and others have shown that youth afflicted with MetS are at increased risk of cardio-metabolic outcomes in adulthood (3,8–11). Taken together, these new data suggest that, although youth with MetS are at increased risk of meaningful cardio-metabolic outcomes in adulthood, they are not destined for a lifetime of increased risk if they resolve their MetS status by adulthood.
We have shown a relative high rate of resolution, which approximates two-thirds, among those who had MetS in youth. These data are consistent with findings from other studies that demonstrate the relative short- and longer-term instability in the diagnosis of youth MetS (3,6,7). Several factors might contribute to this. First, it is possible that misclassification owing to different definitions of youth and adulthood MetS might be responsible. Previously, we observed a high rate—approximately one-third—of MetS resolution in young adults over a much shorter follow-up period of 6-years and with the same definition of MetS at both time-points, suggesting that, although misclassification might be possible, it is not likely the principle reason behind the high rate of MetS resolution observed in our population. Second, pubertal stage has a known influence on a number of MetS components (24) that might have contributed to a high rate of resolution. Third, and perhaps most promising from a public health perspective, because our study was observational, the resolution of MetS was likely accomplished individually in some way. For example, we showed that those in the resolution group were more likely to limit weight gain and partake in more physical activity in the period between youth and adulthood relative to their peers who had MetS at both time-points or developed MetS in adulthood. These data provide insight on potential modes of intervention that might benefit youth with MetS but are also consistent with public health interventions aimed at reducing the obesity epidemic.
In our previous study of the vascular effects of MetS resolution in adulthood, we observed that, after adjustment for change in adiposity indexes, differences in the outcomes between the resolution and persistent groups were diluted or attenuated, thus suggesting that the favorable changes in vascular measures was principally mediated by weight loss (12). Interestingly, the inclusion of change in adiposity terms in our youth to adulthood models did not appreciably attenuate the observed differences between the resolution and persistent groups. These data are of interest, because our previous report suggested that youth BMI alone was as good a predictor of high IMT and T2DM and in some instances a better predictor than youth MetS (3). In concert, we interpret these new findings to suggest that, although measurement of youth BMI provides a simpler and more accurate means than MetS to identify youth at risk of later cardio-metabolic outcomes, other risk factors, beyond adiposity, seem to become important in the transitional period between youth and adulthood. These data emphasize that, to combat against permanent or incident MetS in the period between youth and adulthood, public health or intervention programs should be targeted toward all MetS risk components in addition to limiting weight and BMI gain.
Study strengths and limitations
The main strength of this study is the opportunity to combine data from 2 well-phenotyped and independent cohorts followed since youth. Nevertheless, limitations need to be acknowledged. First, bias due to differential loss to follow-up is possible. Nevertheless, we have shown previously that—although nonparticipants at follow-up were more likely to be younger, male, and black (Bogalusa)—baseline risk factor levels were similar between those who did and those who did not attend follow-up, suggesting that a major bias is unlikely (19,25). Second, we used BMI to define adiposity in our MetS definition rather than waist circumference, because neither study collected these data in youth. Although BMI might be considered a less sensitive index of metabolically relevant adiposity in the context of defining the MetS, recent syntheses of the area (5,26) suggest no evidence that waist circumference improves the diagnosis of obesity or the cardio-metabolic comorbidities of obesity in children and adolescents (26), which is consistent with most findings in the adult setting (5). Third, the low numbers with T2DM in both cohorts and use of fasting glucose levels and self-report data to indicate adult T2DM mean that associations with T2DM should be interpreted cautiously. However, in sensitivity analyses, we observed a similar trend when examining impaired fasting glucose or T2DM as an alternate outcome. Fourth, because carotid IMT data were not available from youth, we are not able to discount that differences in IMT already existed between the resolution and persistent groups. Fifth, our data are observational with no attempt made to prevent MetS or intervene in established MetS. Sixth, the acquisition of data differed between the 2 studies. Reassuringly, where power allowed, the results were consistent across study. Seventh, the lack of association between change in fruit or vegetable consumption observed for Young Finns in this study might have been due to the nonstandardized measurement of or inconsistency in the diet variables collected at baseline and follow-up. The resultant measurement error from comparing related but different indicators of fruit and vegetable consumption would likely shift any true effect toward the null. Finally, because we only had data from 2 time-points separated by 14 to 27 years, we were unable to account for multiple changes in MetS status that might have occurred in the intervening period or the timing of these changes; we were also unable to examine the concomitant changes in lifestyle-related risk factors.
Youth diagnosed with MetS are at 2- to 3-fold increased risk of meaningful cardio-metabolic derangements in adulthood, but to date it has not been clear whether these individuals are destined to maintain this risk into adulthood. Data presented in this manuscript suggest that the potential deleterious effects of MetS can be resolved and that this resolution can go some way to normalize the risk of later cardio-metabolic disorders, at least those that we examined, to levels associated with never having had MetS. Although we could only speculate on the means of resolution, weight loss and increased physical activity relative to peers might be central. These data contribute to the ongoing debate concerning the clinical utility of identifying and intervening in youth with MetS.
The authors thank Pronabesh Das Mahapatra, MD, MPH, Ville Aalto, MSSc, and Irina Lisinen, MSSc for assistance in compiling these data.
For supplementary text, figures, and tables, please see the online version of this article.
A Diagnosis of the Metabolic Syndrome in Youth That Resolves by Adult Life Is Associated With a Normalization of High Carotid Intima-Media Thickness and Type 2 Diabetes Mellitus Risk: The Bogalusa Heart and Cardiovascular Risk in Young Finns Studie
The Bogalusa Heart Study was supported by Grants HD-061437 and HD-062783 from the National Institute of Child Health and Human Development and AG-16592 from the National Institute on Aging. The Cardiovascular Risk in Young Finns study was financially supported by the Academy of Finland (Grants 117797, 126925, and 121584), the Social Insurance Institution of Finland, the Turku University Foundation, Special Federal Grants for the Turku, Tampere, and Kuopio University Central Hospital, the Juho Vainio Foundation, the Finnish Foundation of Cardiovascular Research, the Finnish Cultural Foundation, and the Orion Farmos Research Foundation. The contribution of Dr. Magnussen to this paper was supported in part by The Finnish Foundation for Cardiovascular Research; and holds a National Health and Medical Research Council Early Career Fellowship (Public Health Fellowship, APP1037559). Dr. Lehtimäki is supported in part by the Emil Aaltonen Foundation. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- body mass index
- continuous metabolic syndrome score
- high-density lipoprotein
- intima-media thickness
- metabolic syndrome
- type 2 diabetes mellitus
- Received November 16, 2011.
- Revision received April 16, 2012.
- Accepted May 11, 2012.
- American College of Cardiology Foundation
- Alberti K.G.,
- Eckel R.H.,
- Grundy S.M.,
- et al.
- Steinberger J.,
- Daniels S.R.,
- Eckel R.H.,
- et al.
- Magnussen C.G.,
- Koskinen J.,
- Chen W.,
- et al.
- Goodman E.,
- Daniels S.R.,
- Meigs J.B.,
- Dolan L.M.
- Chen W.,
- Srinivasan S.R.,
- Li S.,
- Xu J.,
- Berenson G.S.
- Morrison J.A.,
- Friedman L.A.,
- Gray-McGuire C.
- Koskinen J.,
- Magnussen C.G.,
- Taittonen L.,
- et al.
- Simell O.,
- Niinikoski H.,
- Rönnemaa T.,
- et al.
- Raitakari O.T.,
- Juonala M.,
- Rönnemaa T.,
- et al.
- ↵(2000) Obesity: Preventing and Managing the Global Epidemic: Report of a World Health Organization Technical Report Series (World Health Organization, Geneva, Switzerland) i–xii, pp 1–253.
- Wijndaele K.,
- Beunen G.,
- Duvigneaud N.,
- et al.
- Magnussen C.G.,
- Venn A.,
- Thomson R.,
- et al.
- World Health Organization
- Cleland V.J.,
- Ball K.,
- Magnussen C.,
- Dwyer T.,
- Venn A.
- Frontini M.G.,
- Srinivasan S.R.,
- Berenson G.S.
- Magnussen C.G.,
- Raitakari O.T.,
- Thomson R.,
- et al.