YECKEL, CATHERINE W.1,2; GULANSKI, BARBARA3,4; ZGORSKI, MELINDA L.4; DZIURA, JAMES3,4; PARISH, REBECCA3; SHERWIN, ROBERT S.3,4
Disruption in autonomic nervous system (ANS) balance is observed for both cardiovascular disease and type 2 diabetes (12,13,19). Whether ANS dysfunction serves to link these disease states is still unknown. The ANS has two arms that work primarily in opposition. The parasympathetic arm is thought to work in favor of conserving energy and helping digestion and absorption. Reducing HR and blood pressure are important functions. In contrast, the sympathetic arm allows the body to respond to stress, i.e., "fight or flight." Fundamental to this action is the increase in HR and blood pressure. Early exercise HR recovery (HRR) is a strong predictor of all-cause mortality, cardiovascular events, and diabetes (7,9-11,32). Recovery from intense exercise, however, requires the coordinated action of both arms of the ANS. After cessation of exercise, there is strong and rapid parasympathetic reactivation, to reestablish a stable basal control of the heart, followed by a slower withdrawal of the sympathetic activation. The sympathetic arm permits the high HR achieved during exercise (for review, see ). This complex complementary ANS pattern during exercise recovery acts to reestablish resting sympathovagal balance, thereby making it an appealing physiological challenge for efficient functional assessment of ANS dysfunction.
Whereas the early exercise recovery HRR index is known to reflect the degree of parasympathetic ANS tone (25), there is no analogous index to assess the ability of the sympathetic system to disengage. Importantly, the longitudinal component of the Coronary Artery Revascularization in Diabetes (CARDIA) trial has provided evidence that suggests poor HRR occurs secondary to the degree of baseline metabolic syndrome risk profile (26), whereas insulin resistance is believed to be a primary factor underlying the development of metabolic syndrome (42) (even at an early age ). Parasympathetic function is generally strong in adolescents and young adults (1). No metabolic risk differences were observed as a function of early HRR in the children and adolescents in the recently released National Health and Nutrition Examination Survey (NHANES; 1999-2002) data (26). Furthermore, it is sympathetic overactivity that is commonly observed in diverse insulin-resistant states, both from muscle sympathetic nerve activity recording experiments (18,23,49) and HR variability (36,38). There is increasing evidence of sympathetic overactivity as a function of metabolic syndrome (22,29,46,47).
If recovery from exercise can be used to quantify the degree to which sympathovagal balance is shifting toward sympathetic overactivity, it could serve as a simple minimally invasive assessment technique to track early health risk in young individuals (especially those who are overweight or obese). Importantly, the degree of elevated HR, after the initial HRR, in part reflects the absolute V˙O2 achieved during exercise-higher peak V˙O2 will also lead to higher recovery V˙O2 values and thus the necessity for elevated HR, especially in untrained individuals. This basic exercise intensity principle was demonstrated previously by Pierpont and Voth (40). We hypothesized that after the immediate rapid HRR (parasympathetic) response, a sustained elevation in HR, normalized for the prevailing rate of oxygen consumption (HR/V˙O2plat), would provide a simple functional index of sympathovagal balance shifted toward sympathetic overactivity in young healthy individuals. The prediction was that this simple novel index (HR/V˙O2plat), not HRR, would be associated with the magnitude of insulin resistance in otherwise healthy young individuals. We chose to explore insulin resistance both from basal fasting values of control and the stimulated insulin sensitivity condition, using a standard clinical oral glucose tolerance test because this latter testing condition often dramatizes dysfunction in young people (52).
We further checked the reliance of the simple exercise recovery index to independently reflect an overall state of functional sympathetic overactivity by examining the association between the HR/V˙O2plat and high-carbohydrate meal-induced thermogenesis. Sympathetic overactivity is recognized in its connection to thermogenesis (27). Obese compared to lean individuals (31,44), including young boys (35), reveal a blunted thermogenic response to high carbohydrate ingestion. Blunted thermogenesis is believed to reflect basal sympathetic overactivity. The prediction was if the HR/V˙O2plat ratio provides a functional index for sympathetic overactivity, it would be inversely associated with high-carbohydrate meal thermogenesis, whereas there would be no relationship to HRR as a marker of parasympathetic function.
Healthy young individuals (N = 20; aged 14-23 yr) participated in this study. Participants were recruited by flyers around Yale University campus and the Hospital Research Unit for the Yale Center for Clinical Investigation (YCCI) to participate in a larger study examining the effects of mild cold exposure on metabolic risk factors. All subjects (parents of participants <18 yr) were fully explained the protocol and provided written informed consent/permission. Minors provided written assent. The protocol was reviewed and approved by the Yale Human Investigation Committee and the Advisory Committee for the YCCI.
General Experimental Protocol
Subjects were health screened before participation. Any subjects with known cardiovascular disease, asthma, or endocrinopathies were excluded from participation, as well as any individuals on medications known to influence cardiovascular or endocrine systems, including weight loss medications. Oral contraception was permitted; all females were studied during the early follicular phase of their menstrual cycle. Females were pregnancy tested before participation; a positive test excluded participation. A wide range in body fat and level of inactivity were permitted to accomplish a range in insulin resistance. Subjects were excluded if performing >1 h of aerobic exercise per wk.
Participants were scheduled within 1 wk for a standard 75-g oral glucose tolerance test (OGTT) and maximal (peak) fitness test on a standard cycle ergometer (Monarch 839E, Monark Exercise AB, Varberg, Sweden). The order of these testing procedures was random. Blood pressure and fasting blood samples were taken for metabolic risk status before the OGTT. The evening before the OGTT, subjects completed a high-carbohydrate meal challenge to examine meal-induced thermogenesis, glucose, and insulin responses. Body composition was determined after the OGTT by dual-energy x-ray absorptiometry (DEXA) whole-body scanning using a Hologic Scanner; Discovery W Model (Hologic Inc., Boston, MA). Data for lean body mass (LBM) are in kilograms minus both fat and bone mineral contributions. Reported precision (coefficient of variation) for percent body fat is 1.8% and for lean mass is 0.6% with a strong association to results on the basis of underwater weighing: r = 0.916, P < 0.0001 (41).
Oral glucose tolerance test (OGTT)
Subjects were studied in the Hospital Research Unit of the YCCI at 8 a.m. after a 10- to 12-h overnight fast. A standard 75-g OGTT was performed in all adolescents and young adults to establish glucose tolerance status and insulin levels. An antecubital intravenous catheter was inserted for blood sampling and maintained patent by a normal saline drip. Baseline measurements included blood pressure taken by an automated analyzer (Welch Allyn, Skaneateles Falls, NY) and blood samples for lipid profile, glucose, and insulin. Thereafter, the flavored glucose (Orangedex; Custom Laboratories, Baltimore, MD) was given orally, and blood samples were obtained every 30 min for 180 min for the measurements of plasma glucose and insulin.
Biochemical analyses and calculations
Plasma glucose was determined using a glucose analyzer by the glucose oxidase method (YSI; Yellow Springs Instrument, Yellow Springs, OH). Plasma insulin was measured by the Linco RIA assay, which has less than 1% cross-reactivity with C-peptide and proinsulin. Serum lipids (total cholesterol, HDL and LDL cholesterol fractions, and triglycerides) were determined using the Alfa-Wasserman Ace Analyzer (Alfa Wasserman Inc, West Caldwell, NJ), using the ACE cholesterol, HDL-C, LDL-C, and triglyceride reagents, respectively. The serum LDL and HDL cholesterol classes are thus acted on using selective detergents to liberate either the HDL or LDL cholesterol to react with the cholesterol esterase and oxidase to produce H2O2 for quantification. The assays for determining insulin and lipid profile were performed by the YCCI Analytical Core Laboratory.
Glucose and insulin values were used to calculate physiological indices of insulin resistance and sensitivity: The Homeostasis Assessment Model for Insulin Resistance (HOMA-IR) was calculated from the fasting samples (51): HOMA-IR = (FI × FG) / 22.5, where FI = fasting insulin (μU·mL−1) and FG = fasting glucose (mmol·L−1). Lower HOMA-IR values indicate less insulin resistance, whereas higher HOMA-IR values indicate greater insulin resistance. The composite Whole Body Insulin Sensitivity Index (WBISI) is based on values of insulin (μU·mL−1) and glucose (mg·dL−1) obtained from the OGTT and the corresponding fasting values, as originally described by Matsuda and DeFronzo (30). We subsequently validated this index for use in younger obese individuals (54). This is considered a composite index because it provides combined information for liver and peripheral insulin sensitivity. Higher values indicate greater stimulated insulin sensitivity. Consistent with our validation in the obese youth, here we used the entire 180 min of collection for the mean glucose and insulin values.
Equation (Uncited)Image Tools
High-carbohydrate meal challenge
All subjects (N = 20) also consumed a dinner pasta meal consisting of 65% CHO, 20% fat, and 15% protein, equal to 40% of daily energy expenditure. The YCCI Metabolic Kitchen prepared the high-carbohydrate meal for each subject. Peak thermogenesis was assessed by use of the canopy system for indirect calorimetry (DeltaTracsII; Sensormedics, Yorba Linda, CA). After a baseline measurement was obtained, subjects consumed the meal, and indirect calorimetry measurements were collected for 10-15 min at each 30-min time point for 2 h. We operationally defined peak thermogenesis as the largest deviation in energy expenditure (kcal·d−1) from baseline. Likewise, before and at each 30-min time point after the meal, blood samples for glucose and insulin were collected and processed as for the OGTT.
Fitness test with recovery
For the cycle ergometer peak fitness test subjects arrived in the afternoon. Subjects were in the 3-h postabsorptive state. A Polar Heart Rate Monitor (Polar Electro, Oy, Finland) was used to capture HR. The Oxycon Mobile portable system (Viasys Healthcare, Yorba Linda, CA) was used to assess indirect calorimetry simultaneously with HR. Continuous data for indirect calorimetry and HR were sent together via telemetry to the Oxycon Mobile collection software. The subjects pedaled at 60 rpm (electronically braked; Monarch 839E). The initial power was set at 50 W for 3 min, after which power was increased each 2 min by 25-W increments until peak oxygen consumption was achieved. We verified V˙O2peak by a plateau in V˙O2 with increasing workload, by the achievement of age-predicted HRmax, RER of greater than 1.15, and exhaustion (the inability to pedal at 60 rpm). This protocol was chosen given its success with studying obese subjects (17). After V˙O2peak was reached, subjects were instructed to stop pedaling and relax for 3 min, still seated on the ergometer. During this 3-min recovery period, indirect calorimetry and HR were collected continuously. After the recovery period, the ergometer was unloaded, and the subjects cycled easily to cool down. This primary (N = 20) population was subdivided by median level of fitness (49.5 mL·kg(LBM)−1·min−1). These two groups, N = 10 each, are represented as "low" and "moderate" fitness groups. This fitness value is also consistent with recently published NHANES data for estimated V˙O2max in adolescents 12-19 yr (37).
Exercise Recovery-Assessing Sympathovagal Balance
Early period of exercise recovery
HRR = HR2 min − HRmax was used to signify the clinical parasympathetic functional index (7,11) (Fig. 1A). Reference data in young people are not yet published; however, the CARDIA trial data (18-30 yr) report HRR values of 45 and 42 for men and women, respectively (8).
Later period of exercise recovery (beyond the HRR time frame)
This is believed to represent the degree of residual sympathetic activation before reestablishing resting balance (39,40). This is further evidenced by one study examining muscle sympathetic nerve activity (24). Although recordings were only possible in a few of their subjects, the data revealed a slow decay in sympathetic nerve activity during recovery after a high-intensity cycling exercise. We therefore proposed that the period during recovery where the ratio of HR/V˙O2 initially reached a plateau (approximately 3 min), would reflect sympathovagal balance-as a continuum of functional sympathetic overactivity (Figs. 1A, B, for region of interest). Our decision to use the initial plateau region for the HR/V˙O2 ratio was based on its relative stability during this time frame. Also important was retaining a close temporal connection to when the parasympathetic activation would be strongest. Although HR and V˙O2 values continue to fall at a slower rate after 3 min of recovery, pilot studies revealed progressive and significant increases in the SD of HR/V˙O2 (bpm/(mL·kg−1 min−1)): 0.88 ± 0.25 (2-3 min) to 1.88 ± 0.52 (4-5 min), P = 0.01. The HR/V˙O2plat therefore represents a simple index that may provide useful immediate, on-site information regarding a young individual's health status because no invasive blood draws or modeling (e.g., HR variability) procedures are required.
Reference group of endurance-trained athletes for fitness comparison
The reference group of young athletes (N = 7) was aged 17-29 yr. These individuals were competitive collegiate cross-country runners. Exercise recovery data were available from performing a maximal graded treadmill test using a modified Astrand protocol on a Vision Fitness Model T9700HRT Treadmill (Vision Fitness, Lake Mills, WI). Speed was variable, selected to elicit an HR of 140 bpm at zero percent grade, followed by increments in grade of 2% every 2 min until V˙O2max was achieved (same termination criteria as for cycle ergometer test). These individuals had fasting glucose and insulin values available for the calculation of HOMA-IR and lipid, body composition, and blood pressure data for metabolic syndrome risk. These data were intended to provide a fitness reference only because individuals performed a treadmill versus cycle test. These individuals represent the "high" fitness group.
Analyses were performed using SAS V9.2 (SAS Institute, Cary, NC). Data are expressed as frequencies or means ± SE. One-way ANOVA was used to evaluate differences between groups by fitness. Simple associations were evaluated using Pearson correlation coefficients. Partial correlations adjusting for percent body fat and parasympathetic function and further by adding in V˙O2peak (fitness) were obtained using multiple linear regression modeling. Log transformations of the insulin sensitivity indices were used to satisfy assumptions of the statistical methods.
Demographic and metabolic syndrome risk factors for participants are shown in Table 1. Importantly, although participants ranged form normal weight to obese, none had evidence of metabolic syndrome as defined for young individuals (52). A unified definition of reference ranges for risk factors in young people is not yet available. The group of N = 7 young endurance athletes (six men and one woman) had an average age of 23 ± 5 yr; BMI, 23 ± 1 kg·m−2; percent body fat, 9.6% ± 3.5%; and V˙O2max, 60 ± 5.6 mL·kg(LBM)−1·min−1. Lipid levels: total cholesterol, 167 ± 14 mg·dL−1; HDL, 53 ± 3 mg·dL−1; LDL, 93 ± 12 mg·dL−1; and triglycerides, 108 ± 15 mg·dL−1; and blood pressure: systolic, 120 ± 3 mm Hg and diastolic, 69 ± 1 mm Hg.
Figure 1A shows the average HR and V˙O2 decay values during exercise recovery. As anticipated, we found that V˙O2 plateau values reflected the degree of peak oxygen consumption (r = 0.74, P < 0.001). However, we also observed a strong inverse association between fractional V˙O2plat to V˙O2peak and percent body fat (r = −0.72, P < 0.001), although no association existed for either lean body mass or magnitude of resting oxygen consumption.
Exercise recovery by fitness and basal insulin resistance
Figure 2 provides the initial plateau values for both HR and V˙O2 as a function of dividing the primary study population into low (N = 10) and moderate (N = 10) fitness levels by the median value of 49.5 mL·kg(LBM)−1·min−1. Importantly, this division did not lead to group-specific differences in lean body mass or percent body fat. The reference group of athletes was brought in as the high (N = 7) fitness comparison. All subjects achieved similar HRmax values (ANOVA: F = 0.73, P = 0.99). Shown in Figure 2A, the HRplat values did not adequately resolve differences between groups for these healthy young people (ANOVA: F = 1.5, P = 0.24). In contrast, the V˙O2plat values (Fig. 2B) were significantly different for all groups (ANOVA: F = 36.3 P < 0.0001). These data demonstrate the striking physiological incongruence between a persistently elevated HR despite an already reduced level of oxygen consumption, especially in the low fitness group. The high fitness group, in contrast, had a tightly coupled HR to oxygen consumption.
Figure 3 provides the comparative data for the same fitness groups: both HRR (parasympathetic function) and HR/V˙O2plat (proposed to reflect functional sympathetic overactivity) variables are presented. Given that sympathetic overactivity is commonly linked to insulin resistance, we examined both HRR and HR/V˙O2plat in association with basal insulin resistance (HOMA-IR) for all subjects. No differences were found in HRR between groups (ANOVA: F = 0.92, P = 0.41; Fig. 3A). Although the V˙O2plat data were significant, the data for HRplat and V˙O2plat trend in opposite directions as a function of fitness; therefore, taken together, the HR/V˙O2plat ratio provides a simple, effective means for distinguishing the aberrant physiological recovery pattern in these young participants (ANOVA: F = 17.5, P < 0.0001; Fig. 3B). Figure 3C provides the HRR data as these relate to basal insulin resistance. There was no significant association between HRR and HOMA-IRlog in the primary study population; however, when the athletes were added into the regression, there was a modest association: r = 0.39, P < 0.02. As shown in Figure 3D, the HR/V˙O2plat index was observed to have a strong association with basal insulin resistance in this young healthy cohort: r = 0.76, P < 0.001; this association improved further with the addition of the athletes: r = 0.83, P < 0.0001. Table 2 provides a comparison of both indices to basic characteristics of percent body fat, fitness, systolic blood pressure, and insulin resistance. Although to a lesser magnitude compared to the HR/V˙O2plat index, percent body fat and fitness were correlated to insulin resistance as would be expected: r = 0.56, P = 0.01 and r = −0.53, P = 0.02, respectively. HRR was not associated with the magnitude of HR/V˙O2plat.
Exercise recovery and stimulated insulin sensitivity
Table 3 provides correlation coefficients for exercise recovery indices and OGTT-derived insulin sensitivity (primary study population only). There was a distinct association between insulin sensitivity and HR/V˙O2plat, r = −0.74, P < 0.001. No relationship existed for HRR. Using multiple regression modeling, the association between HR/V˙O2plat index and insulin sensitivity (WBISIlog) was determined to be independent while controlling for both percent body fat and parasympathetic function (HRR; partial correlation was r = −0.48, P = 0.04). The model lost significance, however, if fitness was included.
The two main component variables of the OGTT (glucose and insulin) were examined in association with both the HR/V˙O2plat index, proposed for sympathetic overactivity, and the index for the traditional parasympathetic function (HRR; Table 3, Fig. 4). Mean glucose (or 2-h glucose) during the OGTT (Fig. 4A) was not related to HR/V˙O2plat or HRR in these young individuals. In contrast, mean insulin during the OGTT (Fig. 4B) was associated with HR/V˙O2plat (r = 0.71, P < 0.001) but not parasympathetic function (HRR). As with insulin sensitivity, a multiple regression model focused more specifically on the mean insulin response to the OGTT and revealed the same independence of the novel index in relation to mean insulin or fasting insulin (used to compute HOMA-IR; P = 0.05 and P = 0.02, respectively), independent from percent body fat and parasympathetic function. However, the association to mean and fasting insulin was not independent of fitness.
Exercise recovery and meal-induced thermogenesis
Thermogenesis provides an independent noninvasive assessment of sympathetic nervous system function, which is distinct from parasympathetic function. As a continuous variable, the HRR for parasympathetic function was not related to any variable. In contrast, the HR/V˙O2plat index was inversely associated with peak thermogenesis and directly with meal insulin responses (all variables, P < 0.01). Table 3 provides the correlation coefficients for the high-carbohydrate meal data. Figures 4A, B, show the glucose and insulin responses for the meal on the same graphs as the OGTT responses. As with the glucose level in the OGTT, HRR and HR/V˙O2plat were not related to the glucose response during the meal. Note, however, that although meal thermogenic data were available for all participants, glucose and insulin data were only available in N = 17.
Peak thermogenesis from the high-carbohydrate meal (all primary study population subjects) was significantly associated with lean body mass (r = 0.89, P < 0.0001), but only for those individuals with less than 28% body fat (median value). We therefore divided the group by the median value for body fat (28% body fat) creating two groups of N = 10 each. These groups had similar lean body mass and aerobic fitness characteristics; however, there was a blunted peak thermogenic response 220 ± 20 vs 350 ± 50 kcal·d−1 in the high body fat group compared to the low body fat group (P = 0.035). Consistent with the thermogenic data, the HR/V˙O2plat was significantly greater in the high body fat group (15.9 ± 1.5 and 22.4 ± 2.2 bpm/(mL·kg−1·min−1), P = 0.024, for low and high body fat groups, respectively. The HRR response was found to be similar −56 ± 4 (low body fat) and −59 ± 4 (high body fat) between groups, P = 0.47.
ANS dysfunction associated with cardiovascular disease and diabetes risk is complex and not fully understood but may contribute to the integration of both pathophysiologies. Although parasympathetic dysfunction such as HRR is a predictor of all-cause mortality, cardiovascular events, and diabetes (7,9-11,32), sympathetic dysfunction may occur at an earlier point in the pathogenesis of disease (24). Here, we observed a strong early exercise HRR (parasympathetic function) response. In contrast, during the initial plateau phase of exercise recovery, HR remained elevated, especially in the low fitness group, despite already low rates of oxygen consumption. Combining these variables, the simple exercise recovery HR/V˙O2plat index proposed here represents functional sympathetic overactivity. This novel index was distributed continuously with fitness and with both basal and stimulated insulin sensitivity, independent of percent body fat and parasympathetic function. High-carbohydrate meal-induced thermogenesis was found to be inversely associated with HR/V˙O2plat but was not related to HRR. Therefore, this simple exercise recovery index is linked to one of the primary cardiovascular and diabetes risk factors expressed in youth, namely, insulin resistance. From a practical standpoint, extending exercise testing for only several minutes into exercise recovery provides a unique, minimally invasive, means to indirectly assess health status with respect to: (i) fitness, (ii) ANS control, both as estimates of functional parasympathetic and sympathetic overactivity, and (iii) level of insulin resistance, in the growing population of at-risk young people.
Early exercise HRR (indexing parasympathetic function ) is an effective assessment technique predictive of mortality, cardiovascular events, and diabetes (7,9-11,32). Perhaps not surprising, because of our young study population, we found that HRR was similar across all subjects; therefore, this index did little to resolve differences, even in fitness. The average HRR was consistent with the healthiest quartile of results reported by CARDIA trial (26) and was no different than the reference athletes also examined. No associations were determined for the primary study group when considering the OGTT or high-carbohydrate meal responses. Although we did not achieve statistical significance for the HRR data, all measurements are taken as a result of the same exercise stimulus and in the same individual; therefore, it is unlikely the HRR would gain strong associations compared to those already achieved with the HR/V˙O2plat index. It can be appreciated from recent data reported from the CARDIA trial that parasympathetic dysfunction may occur secondary to increasing metabolic syndrome risk status (26) and, as such, is inconsistent with early developing insulin resistance. In addition, evidence suggests that not only is metabolic syndrome associated with sympathetic overactivity (22,29,46,47) but also combining multiple metabolic syndrome factors (e.g., obesity and hypertension) further increases sympathetic overactivity as directly determined by microneurography of skeletal muscle (29). However, to date, there is no comparable simple test to the HRR for parasympathetic function to index sympathetic overactivity. We were therefore interested in quantifying sympathovagal balance from an exercise recovery model when the parasympathetic system was apparently normal and strongly engaged.
Previous work has carefully modeled the complex interaction between parasympathetic and sympathetic activities during exercise recovery (40). These data emphasize the fact that high exercise intensities result in higher HR recovery plateaus (40). This latter observation is critical because, as expected, the magnitude of the V˙O2plat was found to reflect the peak V˙O2 attained; thus, there was, by necessity, a higher HR to sustain cardiac output in the more cardiovascularly fit of our untrained participants. Normalizing the HR for the prevailing V˙O2plat, as we do with the recovery index, allows for the individual adjustment of absolute intensity achieved during the test and the requirement for early excess postexercise oxygen consumption as a function of duration and intensity of exercise achieved in the test (5). Here, we found the V˙O2plat values to be reduced in the low fitness group very early after peak exercise, which were physiologically inconsistent with their high HR. Of note, Fu et al. (16) recently demonstrated a strong inverse association between muscle sympathetic nerve activity and stroke volume and positive association with peripheral resistance under tilt conditions, lending further support for the use of the HR/V˙O2plat index (inverse of the O2 pulse). Individuals with better stroke volume, e.g., individuals who perform regular aerobic exercise, do not require a high HR to sustain V˙O2. These individuals should theoretically be better equipped to respond to stress.
It is well understood that, at least from a cardiovascular risk standpoint, decreased sympathetic and enhanced parasympathetic function are beneficial to disease outcome (see (43) for review). It is the aberrant shift in sympathovagal tone toward sympathetic overactivity that is associated with insulin resistance, as evidenced by muscle sympathetic nerve activity studies (18,23,49). For example, a study of normal weight offspring of parents with type 2 diabetes has revealed a strong positive connection between insulin levels and muscle sympathetic nerve activity (23). Likewise, obese individuals have elevated basal muscle sympathetic nerve activity (49). Consistent with these findings, we found that increased insulin resistance was associated with sympathovagal dysfunction, as defined by exercise recovery kinetics. However, these results revealed a continuum of functional sympathetic overactivity in the healthy young people studied. The association of the HR/V˙O2plat index to insulin resistance was closely tied to fitness but was independent of percent body fat and parasympathetic function.
To further elucidate whether the novel exercise recovery index was indeed providing information of sympathetic overactivity, we examined the participants for high-carbohydrate meal-induced thermogenic responsiveness. It is generally recognized that dietary thermogenesis is mediated by the sympathetic nervous system activation (27). Obese adults (44,48) and boys (35) are reported to have a poor thermogenic response to a high-carbohydrate load compared to normal weight controls as a result of basal sympathetic overactivity. Our results were in accord; we observed a blunted thermogenic response in this study when considering the individuals on the basis of a median level of body fat (28%). Furthermore, these individuals were found to have an elevated HR/V˙O2plat in keeping with sympathetic overactivity inhibiting the plasticity of the thermogenic response despite equivalent levels of lean body mass.
A euglycemic-hyperinsulinemic clamp infusion study revealed that insulin is likely the main modulator for sympathetic activation (4). Importantly, insulin clamp studies have demonstrated that insulin resistance can negatively affect sympathetic nervous system responsiveness, independent of BMI (2). We previously reported that no increase in energy expenditure occurred in response to an insulin infusion in obese-resistant compared to obese-sensitive adolescents who were matched for body composition (53). Thermogenesis associated with food ingestion studies combined with measurements of muscle sympathetic nerve activity support both insulin and glucose effects (3,14). However, the more direct effect of insulin is further supported by studies of glucose versus fructose ingestion. Both compounds cause similar carbohydrate oxidation, but unlike glucose, fructose elicits neither a major insulin response nor a muscle sympathetic nerve response (50).
The high-carbohydrate meal data revealed that the exercise recovery index was consistent with sympathetic overactivity: there was not only a significant association between the magnitude of peak thermogenic response, such that higher index values were associated with blunted thermogenesis, but also the index remained associated with meal insulin (as with the OGTT). The HRR response (parasympathetic) was not associated with any variables related to the meal. These data therefore provide an independent verification that the exercise recovery index is providing information regarding the extent of functional sympathetic overactivity.
Our results support the use of the exercise recovery HR/V˙O2plat ratio as an index of sympathetic overactivity in young individuals associated with postprandial (OGTT)-stimulated insulin sensitivity as well as fasting insulin resistance independent from percent body fat and parasympathetic function. More specifically, the association of HR/V˙O2plat was to insulin levels per se, not glucose, in accord with euglycemic-hyperinsulinemic clamp studies (4). The relationship between this novel exercise recovery ratio for functional sympathetic overactivity and insulin resistance was not, however, independent from fitness in this cross-sectional study. Given that insulin sensitivity can change even with a single bout of exercise or short-term exercise training model that do not alter overall fitness, the possibility exists for a more rapid alteration in ANS control within this fitness pathway (6). However, to date, the majority of acute exercise-related insulin sensitivity changes have focused predominantly on muscle-specific alterations (20,21). This simple exercise recovery index may provide a useful tool to better understand one aspect of how fitness affects health risk and improvement.
The obvious target population for this type of functional assessment is overweight/obese adolescents and young adults that generally present with strong insulin resistance, but the majority of whom continue to have normal glucose tolerance (54). A high level of metabolic syndrome risk factors is reported in pediatric obesity (52). Recently, Morrison et al. (33) reported a dramatically high rate of incident cardiovascular disease in adults from youth expressing metabolic syndrome, as well as type 2 diabetes (34). Importantly, the insulin-resistant obese adolescents (14-17 yr) studied here had HR/V˙O2plat values in the highest 25th percentile of those studied. Determining simple measures of autonomic function related to insulin resistance may be a key feature to a better understanding of their health risk profile. Microneurography work has revealed that hypertension can further add to sympathetic overactivity associated with insulin resistance (21). We only found a weak association of blood pressure with HR/V˙O2plat; however, our study population expressed a narrow range of blood pressure.
There are several limitations to our study: the results from this study demonstrate a strong continuum of functional sympathetic overactivity across a wide range of body composition. However, we are supporting our claim with independent but indirect measures of sympathetic activity. Further work is needed to demonstrate the correspondence of the HR/V˙O2plat to a direct measure, such as muscle sympathetic nerve activity. One study examining the response to incremental cycling exercise has now been published, which indicates high residual total muscle sympathetic nerve activity (435% of rest at 3-5 min of recovery) after termination of high-intensity exercise (24). Unfortunately, recordings were only possible throughout their testing in 5 of the 15 subjects (mean body weight of all subjects was 63 kg). These data lend further support to the slower withdrawal of the sympathetic nervous system as observed with muscle sympathetic nerve activity. No data for oxygen consumption were reported.
The performance of our novel exercise recovery index has not been tested under conditions when there is potential confounding due to poor parasympathetic reactivation. With slow parasympathetic system reactivation, the initial plateau phase for HR/V˙O2plat will occur at a later time interval. Our data should not be interpreted as suggesting that parasympathetic function is not important in young people, especially given the growing number of people with family history of cardiovascular- and diabetes-related diseases. These individuals, even children may be at further risk owing to overweight or obese status (45). We made use of our healthy population with good parasympathetic function to maximally separate effects of the parasympathetic and sympathetic nervous systems. It should be emphasized that the HRR is still measured; we are only advocating extending the data collection by a couple of minutes and including oxygen consumption so that both arms of the ANS can be evaluated. Further work is needed in young individuals with compromised parasympathetic function to better understand how the novel recovery index will perform. It is important that the data released from NHANES (1999-2002) for children and adolescents did not demonstrate differences in metabolic risk factors across quartiles for early HRR (28). Metabolic risk factors did, however, contribute to the results across quartiles for later HRR (3-min data). The 3-min HRplat values collected here did not achieve significance across fitness groups, but the trend was visible, and served to further physiologically uncouple from the V˙O2plat when used together as the HR/V˙O2plat index. Combining these cardiorespiratory factors may provide improved resolution for metabolic risk profiling in young people.
In conclusion, we propose that recovery from peak exercise can be used to quantify not only early HRR for parasympathetic function, but also the later HR/V˙O2plat that holds promise for quantifying functional sympathetic overactivity. This novel minimally invasive recovery index was found to be strongly associated with insulin resistance in a diverse group of young healthy people, including obese adolescents. Therefore, recombining the classic exercise testing measurements may help to elucidate the role of sympathetic overactivity in cardiovascular disease and diabetes pathophysiology under conditions related to insulin resistance.
The authors thank the adolescents and young adults who participated in the study, to the research nurses for the excellent care given to our subjects, to Mary Savoye, R.D., C.D.-N., C.D.E., and Donna Caseria, R.D., for their expertise in preparing the high-carbohydrate meal, to Aida Groszmann, Andrea Belous, Codruta Todeasa for their superb technical assistance, and Michelle Gosselin for her help with article preparation. This study was supported in part by the American Diabetes Association No. 7-05-JF-54 awarded to Dr. Yeckel and by the Yale Center for Clinical Investigation (YCCI), Grant No. UL1 RR024139 from the National Center of Research Resources (NCRR), a component of the National Institutes of Health (NIH), and the NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NCRR or NIH. Before the YCCI grant award, the study was supported in part by the GCRC Program Grant No. M01 RR000125 from the National Center for Research Resources, National Institutes of Health. The results of this study do not constitute endorsement by the American College of Sports Medicine.
None of the authors have a conflict of interest with work completed for this article.
1. Benatar A, Decraene T. Comparison of formulae for heart rate correction of QT interval in exercise ECGs from healthy children. Heart
2. Bergholm R, Westerbacka J, Vehkavaara S, Seppala-Lindroos A, Goto T, Yki-Jarvinen H. Insulin sensitivity regulates autonomic control of heart rate variation independent of body weight in normal subjects. J Clin Endocrinol Metab
3. Berne C, Fagius J, Niklasson F. Sympathetic response to oral carbohydrate administration. Evidence from microelectrode nerve recordings. J Clin Invest
4. Berne C, Fagius J, Pollare T, Hjemdahl P. The sympathetic response to euglycaemic hyperinsulinaemia. Evidence from microelectrode nerve recordings in healthy subjects. Diabetologia
5. Borsheim E, Bahr R. Effect of exercise intensity, duration and mode on post-exercise oxygen consumption. Sports Med
6. Brooks GA, Mercier J. Balance of carbohydrate and lipid utilization during exercise-the crossover concept. J Appl Physiol
7. Carnethon MR, Jacobs DR Jr, Sidney S, Liu K. Influence of autonomic nervous system dysfunction on the development of type 2 diabetes: the CARDIA study. Diabetes Care
8. Carnethon MR, Prineas RJ, Temprosa M, Zhang ZM, Uwaifo G, Molitch ME. The association among autonomic nervous system function, incident diabetes, and intervention arm in the Diabetes Prevention Program. Diabetes Care
9. Cheng YJ, Lauer MS, Earnest CP, et al. Heart rate recovery following maximal exercise testing as a predictor of cardiovascular disease and all-cause mortality in men with diabetes. Diabetes Care
10. Cole CR, Blackstone EH, Pashkow FJ, Snader CE, Lauer MS. Heart-rate recovery immediately after exercise as a predictor of mortality. N Engl J Med
11. Cole CR, Foody JM, Blackstone EH, Lauer MS. Heart rate recovery after submaximal exercise testing as a predictor of mortality in a cardiovascularly healthy cohort. Ann Intern Med
12. Doux JD, Yun AJ. The link between carotid artery disease and ischemic stroke may be partially attributable to autonomic dysfunction and failure of cerebrovascular autoregulation triggered by Darwinian maladaptation of the carotid baroreceptors and chemoreceptors. Med Hypotheses
13. Esler M, Rumantir M, Wiesner G, Kaye D, Hastings J, Lambert G. Sympathetic nervous system and insulin resistance: from obesity to diabetes. Am J Hypertens
. 2001;14(11 Pt 2):304S-309S.
14. Fagius J, Berne C. Increase in muscle nerve sympathetic activity in humans after food intake. Clin Sci (Lond)
15. Freeman JV, Dewey FE, Hadley DM, Myers J, Froelicher VF. Autonomic nervous system interaction with the cardiovascular system during exercise. Prog Cardiovasc Dis
16. Fu Q, Shook RP, Okazaki K, et al. Vasomotor sympathetic neural control is maintained during sustained upright posture in humans. J Physiol
. 2006;577(Pt 2):679-87.
17. Goodpaster BH, Wolfe RR, Kelley DE. Effects of obesity on substrate utilization during exercise. Obes Res
18. Grassi G, Dell'Oro R, Facchini A, Quarti Trevano F, Bolla GB, Mancia G. Effect of central and peripheral body fat distribution on sympathetic and baroreflex function in obese normotensives. J Hypertens
19. Hellstrom HR. The altered homeostatic theory: a hypothesis proposed to be useful in understanding and preventing ischemic heart disease, hypertension, and diabetes-including reducing the risk of age and atherosclerosis. Med Hypotheses
20. Henriksen EJ. Invited review: effects of acute exercise and exercise training on insulin resistance. J Appl Physiol
21. Holloszy JO. Exercise-induced increase in muscle insulin sensitivity. J Appl Physiol
22. Huggett RJ, Burns J, Mackintosh AF, Mary DA. Sympathetic neural activation in nondiabetic metabolic syndrome and its further augmentation by hypertension. Hypertension
23. Huggett RJ, Hogarth AJ, Mackintosh AF, Mary DA. Sympathetic nerve hyperactivity in non-diabetic offspring of patients with type 2 diabetes mellitus. Diabetologia
24. Ichinose M, Saito M, Fujii N, et al. Modulation of the control of muscle sympathetic nerve activity during incremental leg cycling. J Physiol
. 2008;586(Pt 11):2753-66.
25. Imai K, Sato H, Hori M, et al. Vagally mediated heart rate recovery after exercise is accelerated in athletes but blunted in patients with chronic heart failure. J Am Coll Cardiol
26. Kizilbash MA, Carnethon MR, Chan C, Jacobs DR, Sidney S, Liu K. The temporal relationship between heart rate recovery immediately after exercise and the metabolic syndrome: the CARDIA study. Eur Heart J
27. Landsberg L, Saville ME, Young JB. Sympathoadrenal system and regulation of thermogenesis. Am J Physiol
. 1984;247(2 Pt 1):E181-9.
28. Lin LY, Kuo HK, Lai LP, Lin JL, Tseng CD, Hwang JJ. Inverse correlation between heart rate recovery and metabolic risks in healthy children and adolescents-insight from the National Health and Nutrition Examination Survey 1999-2002. Diabetes Care
29. Mancia G, Bousquet P, Elghozi JL, et al. The sympathetic nervous system and the metabolic syndrome. J Hypertens
30. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care
31. Matsumoto T, Miyawaki C, Ue H, Kanda T, Yoshitake Y, Moritani T. Comparison of thermogenic sympathetic response to food intake between obese and non-obese young women. Obes Res
32. Mora S, Redberg RF, Cui Y, et al. Ability of exercise testing to predict cardiovascular and all-cause death in asymptomatic women: a 20-year follow-up of the lipid research clinics prevalence study. JAMA
33. Morrison JA, Friedman LA, Gray-McGuire C. Metabolic syndrome in childhood predicts adult cardiovascular disease 25 years later: the Princeton Lipid Research Clinics Follow-up Study. Pediatrics
34. Morrison JA, Friedman LA, Wang P, Glueck CJ. Metabolic syndrome in childhood predicts adult metabolic syndrome and type 2 diabetes mellitus 25 to 30 years later. J Pediatr
35. Nagai N, Sakane N, Hamada T, Kimura T, Moritani T. The effect of a high-carbohydrate meal on postprandial thermogenesis and sympathetic nervous system activity in boys with a recent onset of obesity. Metabolism
36. Paolisso G, Manzella D, Rizzo MR, et al. Effects of insulin on the cardiac autonomic nervous system in insulin-resistant states. Clin Sci (Lond)
37. Pate RR, Wang CY, Dowda M, Farrell SW, O'Neill JR. Cardiorespiratory fitness levels among US youth 12 to 19 years of age: findings from the 1999-2002 National Health and Nutrition Examination Survey. Arch Pediatr Adolesc Med
38. Perciaccante A, Fiorentini A, Paris A, Serra P, Tubani L. Circadian rhythm of the autonomic nervous system in insulin resistant subjects with normoglycemia, impaired fasting glycemia, impaired glucose tolerance, type 2 diabetes mellitus. BMC Cardiovasc Disord
39. Perini R, Orizio C, Comande A, Castellano M, Beschi M, Veicsteinas A. Plasma norepinephrine and heart rate dynamics during recovery from submaximal exercise in man. Eur J Appl Physiol Occup Physiol
40. Pierpont GL, Voth EJ. Assessing autonomic function by analysis of heart rate recovery from exercise in healthy subjects. Am J Cardiol
41. Pritchard JE, Nowson CA, Strauss BJ, Carlson JS, Kaymakci B, Wark JD. Evaluation of dual energy x-ray absorptiometry as a method of measurement of body fat. Eur J Clin Nutr
42. Reaven GM. Role of insulin resistance in human disease (syndrome X): an expanded definition. Annu Rev Med
43. Rosenwinkel ET, Bloomfield DM, Arwady MA, Goldsmith RL. Exercise and autonomic function in health and cardiovascular disease. Cardiol Clin
44. Segal KR, Chun A, Coronel P, Valdez V. Effects of exercise mode and intensity on postprandial thermogenesis in lean and obese men. J Appl Physiol
45. Singh TP, Rhodes J, Gauvreau K. Determinants of heart rate recovery following exercise in children. Med Sci Sports Exerc
46. Straznicky NE, Lambert EA, Lambert GW, Masuo K, Esler MD, Nestel PJ. Effects of dietary weight loss on sympathetic activity and cardiac risk factors associated with the metabolic syndrome. J Clin Endocrinol Metab
47. Sverrisdottir YB, Mogren T, Kataoka J, Janson PO, Stener-Victorin E. Is polycystic ovary syndrome associated with high sympathetic nerve activity and size at birth? Am J Physiol Endocrinol Metab
48. Tentolouris N, Tsigos C, Perea D, et al. Differential effects of high-fat and high-carbohydrate isoenergetic meals on cardiac autonomic nervous system activity in lean and obese women. Metabolism
49. Vollenweider P, Randin D, Tappy L, Jequier E, Nicod P, Scherrer U. Impaired insulin-induced sympathetic neural activation and vasodilation in skeletal muscle in obese humans. J Clin Invest
50. Vollenweider P, Tappy L, Randin D, et al. Differential effects of hyperinsulinemia and carbohydrate metabolism on sympathetic nerve activity and muscle blood flow in humans. J Clin Invest
51. Wallace TM, Matthews DR. The assessment of insulin resistance in man. Diabet Med
52. Weiss R, Dziura J, Burgert TS, et al. Obesity and the metabolic syndrome in children and adolescents. N Engl J Med
53. Weiss R, Taksali SE, Dufour S, et al. The "obese insulin-sensitive" adolescent: importance of adiponectin and lipid partitioning. J Clin Endocrinol Metab
54. Yeckel CW, Weiss R, Dziura J, et al. Validation of insulin sensitivity indices from oral glucose tolerance test parameters in obese children and adolescents. J Clin Endocrinol Metab