Heart & Lung: The Journal of Acute and Critical Care
Volume 32, Issue 1 , Pages 10-22, January 2003

Cardiovascular autonomic function in healthy adolescents☆☆☆★

Chicago, Illinois, and Memphis, Tennessee

aFrom the College of Nursing, University of Illinois, Chicago, and the bColleges of Nursing and cMedicine, University of Tennessee Health Science Center, Memphis

Article Outline

Abstract 

Objective: The purpose of this study was to determine effects of age, sex, race, body mass index, and Tanner's stage on short-term evoked cardiovascular autonomic tests (ie, Valsalva ratio and change in heart rate with deep breathing) and 24-hour heart rate variability (HRV) in a sample of healthy adolescents, as well as to identify normative indices of both short-term evoked and 24-hour HRV in this age group. Design: A descriptive, correlational design was used. Setting: Study took place in a university hospital in a health science center located in the mid-South. Subjects: Participants included 75 healthy adolescents: mean age was 15.0 ± 1.6 years, 14 were African American, 61 were white, 49 were girls, and 26 were boys. Outcome measures: Study measures included the Valsalva ratio, change in heart rate with deep breathing, and 24-hour HRV with power spectral analysis with Holter monitoring. Results: Major significant findings included lower values of 24-hour HRV measures for girls and African American adolescents (P <.05). Indices for normal ranges of both the short-term evoked and 24-hour HRV measures were computed with 95% confidence intervals. Conclusions: Few published studies address cardiac autonomic function, including 24-hour HRV, in adolescents. Most studies reporting actual normative control values of HRV for youth typically have not addressed sex or racial differences. Our study included the largest number of adolescents to date in the reported literature and demonstrated the importance of considering sex and race variation in interpreting test results. The availability of state-of-the-art technology for obtaining HRV data allows for the early identification of subclinical cardiac autonomic changes in youth who have predispositions for cardiac complications, such as those with diabetes, congential heart disease, or obesity. (Heart Lung® 2003;32:10-22.)

 

Correction

In the article entitled “Cardiovascular autonomic function in healthy adolescents” by Melissa Spezia Faulkner, RN, DSN, Donna Hathaway, RN, PhD, and Betsy Tolley, PhD, in the January/February issue (Heart Lung 2003;32:10-22), there was an error in Table VII on page 19. For DBPM, the Lower 95% CI for the total sample should indicate a value of 35.46, not 5.46.

Autonomic neural regulation of cardiovascular function is a reflection of the balance between sympathetic and parasympathetic innervation resulting in periodic fluctuation in heart rate and rhythm.1, 2, 3 Since the 1970s, measures of cardiovascular autonomic function have been used to evaluate clinical status in adults at risk for myocardial morbidity and mortality.4, 5, 6 For example, cardiovascular dysautonomia is typically present in people with diabetic neuropathy,7, 8, 9, 10, 11 alcoholism,12, 13 postmyocardial infarction,14, 15, 16, 17 and congestive heart failure.18, 19 Common methods for evaluating the beat-to-beat variation in cycle length (ie, R-R interval) of each heart period include measures such as changes in beats per minute with deep breathing or the Valsalva maneuver,20, 21 and the use of Holter monitoring for determining heart rate variability (HRV). Holter monitoring is used not only to compute the measurement of standard deviations of heart periods on the basis of sinus R-R intervals over time, but also to quantify and discriminate between sympathetic and parasympathetic autonomic function during a 24-hour period by recording the frequency (Hz) of R-R variation, also referred to as power spectral analysis.1, 22, 23

Measures of cardiovascular autonomic regulation with HRV are being used increasingly in studies of children and adolescents with diabetes, congenital heart anomalies, and obesity to detect early subclinical autonomic changes.7, 24, 25, 26, 27, 28, 29 As more research is conducted to identify cardiovascular autonomic changes in youth who may be at risk for developing dysautonomia and subsequent arrhythmias or sudden cardiac death later in life, additional confirmation of potential age, sex, or racial differences in measures of HRV for a healthy population are needed for comparison. Although several studies have explored normal variation in cardiovascular autonomic maturation in infants, children, and adolescents with no known diseases or risk factors,30, 31, 32, 33, 34, 35, 36 even fewer indicate normal ranges of measurement indices for age.30, 31, 32, 34, 35

Another issue in interpreting current research on cardiovascular autonomic function in children and adolescents is the lack of consistency in the methods used and the ways in which measures of HRV are reported in these studies. Values include the natural logarithm of beats per minute squared,35 beats per minute squared,34 the natural logarithm of milliseconds squared,7, 37 and milliseconds of the length of the R-R interval of the cardiac cycle or heart period.31, 32, 38 Such discrepancies in presenting HRV data add confusion to the interpretation of a complex autoregulatory, neurologic phenomenon that reflects parasympathetic-sympathetic balance. These issues are consistent with problems inherent in comparing findings of HRV research recognized by the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology.23 In 1996, the Task Force published standards of nomenclature, definitions of terms, and methods of measurement of HRV to facilitate appropriate comparisons and to lead to additional progress in the field. Recognizing the more advanced capabilities of currently available high-frequency, digital, 24-hour, multi-channel electrocardiogram (ECG) recorders, the international members of the Task Force identified the importance of studies involving various age and sex subsets to establish normal HRV standards.

Although research exploring HRV in infants, children, and adolescents who may be at risk for autonomic dysfunction has included youth with congenital heart disease and obesity,25, 26, 27, 28 the majority of studies in the pediatric age group included those with type 1 (insulin-dependent) diabetes mellitus.7, 29, 38, 39, 40, 41, 42, 43, 44, 45 The importance of exploring HRV in younger subjects who may have a predisposition for later cardiac disease is indicated by studies in adult populations that show a decrease in HRV, particularly vagally-mediated modulation that was associated with myocardial infarction, cardiac arrhythmias, and sudden death.46, 47, 48, 49 In reviewing the pediatric literature, HRV was found to be lower in children with atrial septal defects versus controls.28 In obese school-age children, higher cardiac parasympathetic activity was associated with lower levels of fat mass, subcutaneous abdominal adipose tissue, resting heart rate, and resting systolic blood pressure.26 Vagal tone influencing HRV was increased in obese youth who received physical training,25 thus supporting positive effects of exercise as an intervention for children and adolescents who may have a higher predisposition for developing coronary artery disease earlier in life than similarly aged, non-affected peers.

Studies of cardiovascular autonomic function in youth with type 1 diabetes are limited and reveal contradictory findings regarding the presence or absence of functional disturbances.24, 38, 40, 43 Investigations focus on measures that include variation in R-R interval of the ECG with deep breathing, heart rate response to the Valsalva maneuver, and systolic blood pressure response to standing.38, 40, 43 Cardiovascular reflex abnormality recorded as heart rate responses to deep breathing and standing were evaluated in children with and without insulin-dependent diabetes mellitus from 6 to 19 years of age and were found to be significantly lower in those with diabetes.50 In another study, control subjects similar in age, sex, and body mass index were compared with adolescents with diabetes aged 11 to 19 years to determine reference ranges for 4 autonomic measures: (1) heart rate response to deep breathing, (2) Valsalva ratio (ratio of the longest R-R interval during the reflex bradycardia after Valsalva maneuver divided by the shortest R-R during the maneuver), (3) heart rate response to standing, and (4) postural systolic blood pressure response.40 Although no significant differences in the 4 cardiovascular tests were noted for the diabetes and control groups, significantly more teenagers with diabetes (28%, P ≤.001)) had at least 1 abnormal autonomic test result.

In addition to the standard evoked cardiovascular reactivity tests, power spectral analysis of HRV offers a rich new source of data related to autonomic function. This highly sensitive technique quantifies and discriminates between sympathetic and parasympathetic autonomic function during a 24-hour period by recording the frequency (Hz) of R-R variation. Abnormalities detected by this test are more sensitive indicators of autonomic dysfunction than conventional tests, particularly in patients with diabetes who have diminished 24-hour R-R interval variability.51 Earlier studies completed with power spectral analysis of HRV in children and adolescents with diabetes found no abnormalities in autonomic function or differences in resting heart rates as compared with healthy controls.38, 45 However, more recent studies have found that measures of HRV are significantly lower in youth with type 1 diabetes who are not exhibiting clinical symptoms or overt complications of diabetes versus similarly aged controls.7, 29, 44 In studies measuring short-term evoked measures of cardiovascular autonomic function and 24-hour measures of HRV, adolescents with poor glycemic control had significantly lower values than youth with better glycemic control and healthy controls.7, 38, 40, 44, 45, 52 Duration of diabetes also was found to be negatively associated with HRV in youth with diabetes.44, 45

Some comparison studies of cardiovascular reactivity testing and HRV in children and adolescents publish values for the control groups.7, 25, 29, 38, 39, 41 However, the specific indices reported vary considerably, and values for age categories are rarely indicated. One study investigating autonomic function in adolescents with and without diabetes provided average values for the evoked test of deep breathing heart rate variation for each year of age from 11 to 19 years and for sex differences in Valsalva ratio.40 Heart rate variation to deep breathing declined with increasing age, and boys had slightly higher values for Valsalva ratio.

Studies of HRV in healthy infants, children, and adolescents indicated a progressive maturation of the autonomic nervous system during childhood with a trend toward increasing variability up to approximately age 6, followed by a decline toward adolescence.30, 31, 32, 33, 34 One explanation for this maturational process was the association with increased organization of sleep-wake cycles attributed to sympathetic withdrawal.33 Finley and Nugent34 reported that age dependence with HRV was indicated by an increase in low frequency (ie, predominantly sympathetic but some parasympathetic innervation), high frequency (parasympathetic innervation), and total power from 1 month to 6 years, followed by a decrease to 24 years. This study confirmed earlier findings of the same investigators that suggested there is a significant decrease in sympathetic activity between 5 and 10 years of age.30 The significant correlation of declining HRV with age was evident during both awake and sleep states.34, 35

Massin32 concurred with the relevance of age-related developmental changes of HRV, but emphasized a stabilization of sympathovagal balance during childhood, noted as increased cholinergic and decreased adrenergic modulation of HRV. Although Goto et al31 analyzed HRV from midnight to 5 am rather than the 24-hour recordings that other investigators have used,7, 30, 32, 33, 34 high-frequency and low-requency measures increased with age from 3 to 6 years and decreased from 6 to 15 years. Tanaka et al36 used short-term HRV to compare responses to standing in preadolescents with adolescents, and this revealed significantly higher levels of diastolic arterial pressure and higher, low-frequency power (ie, sympathetic innervation) in the adolescents. Thus, although evidence supports the inverse relationship between HRV and age from school-age to adulthood, the steadying of autonomic balance occurring in youth 6 to 12 years may be altered with the onset of puberty. These potential alterations resulting from adrenergic influence could pose an increased risk for hypertension in adolescents.

Sex variations were not reported in these published studies of 24-hour HRV, except for a trend toward higher HRV values in boys versus girls.7, 31 Previous studies indicated that HRV measures were also higher in adult men versus adult women, with differences narrowing as each age.7, 53, 54 Therefore, the tendency for sex variation may be linked to autonomic neural maturation as adolescents transition to young adulthood.

Two studies7, 55 reported racial differences in HRV for African American versus white adolescents. Both studies included relatively small sample sizes. In adolescent boys, the Bogalusa Heart Study55 found that during cardiovascular reactivity testing (ie, responses to standing, Valsalva maneuver, maximal hand grip, and cold pressor stimulation) healthy whites exhibited increased sympathetic tone compared with African Americans. Although data were not collected during cardiovascular reactivity testing, Faulkner et al7 revealed a similar finding with 24-hour HRV measures with power spectral analysis. Therefore, more research is needed to explore variation in autonomic function for differing racial groups to better understand physiologic mechanisms occurring in the development of cardiovascular disease in youth populations.

Back to Article Outline

Purpose 

Although studies of cardiovascular autonomic function in healthy adolescents are beginning to appear in the research literature, the samples include wide age variation and small numbers of adolescents (Table I). The purpose of this preliminary study was to identify age-related norms for measures of cardiovascular autonomic function in healthy adolescents. This work was part of a larger investigation comparing measures of autonomic function in adolescents with type 1 diabetes with age-matched peers with no chronic illness. The specific research questions were the following:

1.What is the effect of age, sex, race, body mass index (BMI), or Tanner's stage on measures of cardiovascular autonomic function in healthy adolescents?

2.What are normal ranges for cardiovascular autonomic function tests (ie, Valsalva ratio, change in heart rate with deep breathing, and 24-hour HRV with power spectral analysis) in healthy adolescents?

Table I. Studies of cardiovascular autonomic function in healthy youth
Author/YearTotal nAge RangeNumbers of Pre-Adolescents/Adolescents
Finley et al., 1987295-24 years9, 11-12 year olds
Finley & Nugent, 1995611 mo-24 years10, 11-12 year olds
Goto et al., 1997603-15 years12, 15 year olds
Massin & von Bernuth, 1997210neonate-14 years20, 13-14 year olds
Massin et al., 2000572 mo-15 years9, 13-15 year olds
Tanaka et al., 20001276-16 years56, 13-16 year olds
Yeragani et al., 1997339 years-adults11, 11-13 years olds

Back to Article Outline

Design and procedures 

Study population 

A descriptive correlational design was used to explore the possible effects of age, sex, race, BMI, and Tanner's stage on cardiovascular autonomic measures in healthy adolescents. Indices for normal ranges of both short-term evoked cardiovascular autonomic functions tests (ie, Valsalva ratio and change in heart rate with deep breathing) and 24-hour ambulatory ECG recordings for HRV were computed with 95% confidence intervals (CI). Healthy youths were obtained by voluntary means, such as recruitment flyers, communication with adolescents with diabetes in the larger study, or personal contact with the researchers. Adolescents who had any acute or chronic illness were excluded from the study. Exclusion criteria included youth who were receiving pharmacologic maintenance therapy for asthma resulting from the potential adrenergic effects of bronchodilators. The study sample consisted of 75 healthy adolescents from 13 to 18 years of age. The mean age for the total sample was 15.0 ± 1.6 years. There were 14 African American (mean age, 15.65 ± 1.66 years) and 61 white youth (mean age, 14.82 ± 1.52 years) in the study, including 49 girls and 26 boys (Table II). Mean ages for the early, middle, and late adolescence groups are listed in Table III.

Table III. Mean ages for early, middle, and late adolescence (n = 75)
nMeanStandard Deviation
Early Adolescence4313.830.57
Middle Adolescence1915.710.56
Late Adolescence1317.640.61

Period of Adolescence

Early Adolescence: ≥13 < 15 years

Middle Adolescence: ≥15 < 17 years

Late Adolescence: ≥17 < 19 years

The mean BMI was 21.5 ± 3.8, which is typical for the ages studied. Usually BMI gradually increases during childhood and through adolescence, reaching an approximate mean of 22.0 at age 17.56

Table II. Frequency distribution for gender, race, and period of adolescence (n = 75)
Early AdolescenceMiddle AdolescenceLate Adolescence
Females
Caucasian2296
African-American264
Males
Caucasian1743
African-American200
Total 431913

Period of Adolescence

Early Adolescence: ≥13 < 15 years

Middle Adolescence: ≥15 < 17 years

Late Adolescence: ≥17 < 19 years

The data were collected at the Physiologic Function Laboratory located in the university hospital of a metropolitan health sciences center in the mid-South. Approval to conduct the study was given by the institutional review board at the university. Individual consent for participation was provided by parental permission and adolescent assent. Demographic data were collected during each participant's laboratory appointment. As part of the demographic information, self-report of Tanner's stage for the subjects was completed by showing pictures of the stages of adolescent development for pubic hair distribution for appropriate male or female genitalia to the subjects and asking them to report the most accurate stage of development for them. This method is reported to have correlations of above 0.70 with physician ratings.57 As a result of recent findings that support variation in cardiac autonomic modulation occurring after puberty,36 an estimation of Tanner's staging was deemed necessary as a means of verifying pubertal development.

There was no food or beverage consumption, nor smoking for at least 30 minutes before the testing. The amount of caffeine intake or smoking for the 24 hours before testing was not restricted. Because both caffeine and nicotine can produce sympathomimetic effects,58, 59 participant use could be a limitation of the study. However, it is noteworthy that previous studies of cardiovascular autonomic function in healthy youth have not reported data on nicotine or caffeine use,30, 31, 34, 35, 36 with the exception of the work by Massin and colleagues who excluded any tobacco use.32, 33

Back to Article Outline

Methods 

Short-term evoked cardiovascular measures 

Cardiovascular autonomic function typically is measured by a battery of short-term evoked cardiovascular reflex tests originally developed by Clarke and Ewing.21 These reflex tests evaluate heart rate and blood pressure changes to postural and respiratory maneuvers and include beat-to-beat (R-R) variation with the expiration-to-inspiration ratio, change in heart rate with deep breathing and with the Valsalva maneuver, and alteration in blood pressure response to postural change.20, 21 The process of evaluating an individual's responses to the reflex tests allows for the detection of the “beat-to-beat” balance of sympathetic and parasympathetic modulation of the cardiac cycle, which can be altered by physical or psychologic stress.8

The reflex tests were conducted in a temperature-controlled laboratory, maintained at 25°C to 27°C, and minimized environmental noise. Data collection of ECG recordings was completed with a Power Macintosh 9500 computer with the AcqKnowledge III BIOPAC data acquisition and analysis system (Biopac Systems, Goleta, CA). A lengthy and rigorous set of reliability and validity studies were previously conducted with a sample of 21 healthy controls, including 11 patients with diabetes and no autonomic symptomatology and 10 patients who were symptomatic.60

The short-term R-R variability measures included in this study were change in heart rate with deep breathing and with the Valsalva maneuver. Change in heart rate with deep breathing was obtained during a 1-minute period of slow, deep breathing (6 breaths per minute) that followed a 2-minute period of regular breathing. Each subject was coached by the technician and breathing was timed for the 1-minute period of deep breathing. The Valsalva ratio was obtained by dividing the highest heart rate during a forced expiration of 40 mmHg for 15 seconds by the lowest heart rate immediately after completion of the forced expiration. In adult controls, values for a normal change in heart rate is normally ≥15 beats per minute and the Valsalva ratio is ≥1.21.60

24-hour HRV 

In addition to evoked cardiovascular autonomic measures, 24-hour ambulatory heart rate monitoring with power spectral analysis was obtained on subjects. Power spectral analysis of HRV quantifies and discriminates between sympathetic and parasympathetic autonomic modulation of heart rate during a 24-hour period by recording the frequency oscillations of R-R variation.22 Abnormalities detected by this test are more sensitive indicators of autonomic dysfunction than conventional tests, particularly in patients with diabetes who have less 24-hour R-R interval variability.51 Present literature published by the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology suggests greater stability of HRV measures derived from 24-hour monitoring.23

Heart rate monitoring was completed with Marquette Series 8500 Holter recording system (Marquette Electronics, Milwaukee, WI). The analog tapes were digitized at 128 samples per second. Tapes were analyzed with Marquette Electronics Laser SXP Ambulatory Electrocardiographic Analysis and Editing Systems, version 5.8 software program (Marquette Electronics). Each QRS complex was identified and labeled. The analyzed data file was scanned and manually edited to locate and correct possible errors in QRS labeling that would negatively affect measurement of HRV. The software program allowed for relabeling of QRS complexes that may occasionally be incorrectly identified as ectopy or artifact. For each data file, only R-R intervals derived from normal sinus rhythm were used to calculate HRV with power spectral analysis.

Measurement and analysis of HRV are classified into time and frequency domain analysis. Time domain analysis is computed on altered versions of the measurement of the standard deviation of heart period on the basis of sinus R-R intervals over time.1 The time domain analysis of HRV can be additionally divided into 2 categories. One category is derived from the R-R intervals with the use of means and standard deviations in milliseconds of the interval. Measures in this category include the standard deviation of all R-R intervals during a 24-hour period (SDNN). Values for SDNN that are less than 50 milliseconds have been associated with sudden cardiac death.61 Another measure is the standard deviation of the means of R-R intervals found in successive 5-minute time periods during 24 hours (SDANN). The calculation for SDANN makes it the most resistant HRV measure for QRS labeling errors and the best measure for circadian fluctuation in heart rate. The final measure is the mean of the standard deviations of the R-R intervals of each 5-minute segment (SD), and it is sensitive to fluctuation within the 5-minute segment. The second category of time domain variables is derived from differences between adjacent R-R intervals and includes indices that are independent of circadian rhythms. The proportion of the total R-R intervals that have differences of successive R-R intervals greater than 50 milliseconds is known as the pNN50. The square root of the mean squared differences of successive R-R intervals is the rMSSD. Representing alterations in autonomic function that are primarily vagally mediated, the pNN50 and rMSSD correlate highly with high-frequency power that is reflective of parasympathetic modulation.1

Power spectral analysis of the frequency domain category of HRV uses fast Fourier transformation to examine variations in R-R intervals. Fluctuations in R-R interval widths are transformed into a frequency waveform that depicts periodic oscillations in sympathetic and parasympathetic function. The spectrum of frequencies (Hz) during the 24-hour periods is provided and indicates the relative amount of total (0.01-1.00 Hz), low (0.04-0.15 Hz), and high (0.15-0.40 Hz) frequency power. Quality assurance data provided from the manufacturer illustrate mathematically correct results after submission of known electronically generated cardiac signals.62 HRV measures of both groups and individuals are reported to be highly reliable for short periods of time (3-65 days).63

Back to Article Outline

Analysis 

The data analysis consisted of statistical procedures to detect any significant differences in cardiovascular autonomic measures for the early, middle, and late adolescence groups, as well as estimating Pearson product moment and Spearman rank correlations with age, sex, race, BMI, or Tanner's stage for the sample of healthy adolescents. An α level of 0.05 was used for determining significance in statistical analyses. Because there were missing data on 4 subjects for the evoked measures and on 5 different subjects on the 24-hour HRV measures, the sample size for each analysis is provided.

As part of the larger investigation that compared the healthy adolescents with those with type 1 diabetes, two 3-way analyses of variances with preplanned contrasts and least square means were used to test for differences in autonomic measures on the basis of age, sex, and diabetes status (ie, with and without diabetes) and on the basis of age, race, and diabetic status. The sum of squares for the 3-way interaction was pooled with the residual sum of squares. Only contrasts for age and sex, or age and race subgroups of healthy adolescents are reported in this paper. Independent t tests were used to evaluate sex and race differences in autonomic measures for the sample of healthy adolescents. Although the overall sample size is quite large (n = 75), the power of tests involving specific main and interaction effects was highly variable. For the 3 main effects, class sizes were sufficiently large to detect differences between means ranging from 0.80 to 1.25 standard deviations at power of 0.80, and α of 0.05. Because sex and race were partially confounded, separate analyses were required to fit these main effects and 2-way interactions. In addition, lack of power likely produced type II errors when subclass means of interaction effects were compared. Results are reported as means and standard deviations.

Back to Article Outline

Results 

Boys in the early adolescence group had significantly higher values of SDNN, SDANN, and SD than similarly aged girls (P <.05) (Table IV).

Table IV. Means and standard deviations of cardiovascular autonomic function for healthy youth in early, middle, and late adolescence
nEarlynMiddlenLate
TotHz
Male198.29 ± .4048.17 ± .3838.39 ± .54
Female247.92 ± .60117.79 ± .5697.87 ± .54
LowHz
Male197.40 ± .3947.32 ± .3537.46 ± .42
Female247.05 ± .64116.89 ± .6297.06 ± .54
HiHz
Male196.67 ± .5946.56 ± .5436.70 ± 1.01
Female246.34 ± .77116.24 ± .6596.20 ± .62
pNN50
Male1928.83 ± 9.79426.35 ± 10.01336.47 ± 12.34
Female2422.44 ± 12.371120.65 ± 6.38922.58 ± 9.47
rMSSD
Male1960.68 ± 15.35457.00 ± 16.57375.00 ± 25.12
Female2450.08 ± 18.291147.45 ± 11.67948.78 ± 13.33
SDNN
Male19178.16 ± 39.59*4163.75 ± 26.453188.66 ± 29.96
Female24145.79 ± 34.83*11145.64 ± 31.609140.00 ± 39.74
SDANN
Male19149.52 ± 38.59*4140.50 ± 29.963159.33 ± 31.94
Female24120.33 ± 34.90*11130.09 ± 28.649115.67 ± 38.54
SD
Male1992.68 ± 18.76*485.00 ± 18.133103.33 ± 19.14
Female2476.29 ± 21.11*1173.36 ± 17.93974.22 ± 17.13
DBPM
Male1840.50 ± 8.45430.50 ± 9.75235.50 ± 2.12
Female2339.47 ± 7.291436.86 ± 7.651031.60 ± 9.97
VR
Male182.00 ± .5841.86 ± .3221.52 ± .59
Female231.95 ± .33141.97 ± .57101.54 ± .42
*P < .05

Early adolescence: ≥13 < 15 years

Middle adolescence: ≥15 < 17 years

Late adolescence: ≥17 < 19 years

TotHz - total frequency power (hertz) of 24 hr heart rate variability in natural log ms2

LowHz - low frequency power (hertz) of 24 hr heart rate variability in natural log ms2

HiHz - high frequency power (hertz) of 24 hr heart rate variability in natural log ms2

pNN50 - percent of adjacent R-R intervals differing >50 milliseconds

rMSSD - square root of mean squared differences of successive R-R intervals

SDNN - standard deviation of all R-R intervals during a 24-hour period; measure of 24 hour circadian rhythmicity

SDANN - measure of variability within 5 minute blocks of time; measure of 24 hour circadian rhythmicity

SD - mean of the standard deviations of the R-R intervals of each 5-minute block

DBPM - difference in beats per minute with deep breathing

VR - Valsalva ratio

Additionally, whites in the late adolescence group had significantly higher values of SDANN than African Americans in the same group (mean, 144 ± 29 vs 74 ± 14; P <.05). Although not significant, values for SDNN were also higher in whites in this group, (mean, 170 ± 30 vs 98 ± 20; P <.06).

For the sample of healthy controls, some significant differences were found. Girls and African American youths had significantly lower values for measures of HRV (Tables V and VI).

Table V. Means and standard deviations of 24-hour heart rate variability in healthy male and female adolescents
NTotHz†LowHz†HiHz*pNN50†rMSSD†SDNN‡SDANN†SD‡
Male268.28 ± 407.39 ± .376.65 ± .6129.3 ± 10.161.8 ± 16.7177.2 ± 36.5149.3 ± 35.992.62 ± 18.51
Female447.88 ± .577.01 ± 606.29 ± .7022.0 ± 10.449.2 ± 15.7144.6 ± 34.4121.8 ± 33.875.14 ± 19.22
*P < .05 †P < .01 ‡P < .001
Table VI. Means and standard deviations of 24-hour heart rate variability in healthy African American and Caucasian adolescents
NTotHz‡LowHz‡HiHz†pNN50rMSSD*SDNN‡SDANN‡SD‡
AA117.47 ± .456.53 ± .435.93 ± .5119.6 ± 10.343.8 ± 12.6118.5 ± 21.895.5 ± 19.363.0 ± 13.7
C598.13 ± .497.27 ± .506.51 ± .6725.6 ± 10.755.7 ± 17.22163.8 ± 36.6138.8 ± 35.485.1 ± 19.9
*P < .05 †P < .01 ‡P < .001

Note:

AA = African-American

C = Caucasian

TotHz - total frequency power (hertz) of 24 hr heart rate variability in natural log ms2

LowHz - low frequency power (hertz) of 24 hr heart rate variability in natural log ms2

HiHz - high frequency power (hertz) of 24 hr heart rate variability in natural log ms2

pNN50 - percent of adjacent R-R intervals differing >50 milliseconds

rMSSD - square root of mean squared differences of successive R-R intervals

SDNN - standard deviation of all R-R intervals during a 24-hour period; measure of 24 hour circadian rhythmicity

SDANN - measure of variability within 5 minute blocks of time; measure of 24 hour circadian rhythmicity

SD - mean of the standard deviations of the R-R intervals of each 5-minute block

No other significant differences were indicated except for Valsalva ratio, which was lower for African Americans than whites (mean, 1.64 ± 0.41 vs mean, 1.95 ± 0.49; t = −2.08; df = 69; P = .04).

Pearson product moment correlations were also estimated for age and BMI with all autonomic measures in healthy controls. There were no significant associations between BMI with any of the autonomic measures. The only significant correlations with age were the difference in beats per minute with deep breathing (r = −0.44, P = .0001) and Valsalva ratio (r = −0.31, P = .0096).

With the use of Spearman rank coefficients, no significant associations between Tanner's stage of sexual maturity and autonomic measures were found, except for one index of time domain of HRV that reflects circadian variation, SDNN (rs = 0.29, P = .036). A closely related measure of HRV is the SDANN, which was not significantly related to Tanner stage (rs = 0.27, P = .053).

Normal ranges for autonomic function tests were estimated by computing 95% CI for a 2-tailed test of significance for the total sample of adolescents and for male and female subclasses (Table VII).

Table VII. Estimated normal values for evoked and 24-hour HRV in healthy adolescents
NMean (SD)Lower 95% CIUpper 95% CI
TotHz
Total708.03 ± .557.908.16
Male268.28 ± .408.128.44
Female447.88 ± .577.718.05
LowHz
Total707.15 ± .567.027.29
Male267.39 ± .377.247.54
Female447.01 ± .606.076.50
HiHz
Total706.42 ± .696.266.59
Male266.65 ± .616.416.90
Female446.29 ± .706.076.50
pNN50
Total7024.74 ± 10.822.1527.32
Male2629.33 ± 10.125.2733.39
Female4422.02 ± 10.418.8525.20
rMSSD
Total7053.84 ± 17.149.7657.92
Male2661.77 ± 16.755.0268.51
Female4449.16 ± 15.744.3953.93
SDNN
Total7056.67 ± 38.3147.51165.83
Male26177.15 ± 36.5162.42191.89
Female44144.57 ± 34.4134.10155.04
SDANN
Total70132.01 ± 36.8123.20140.82
Male26149.27 ± 35.9134.78163.75
Female44121.82 ± 33.8111.51132.13
SD
Total7081.63 ± 20.676.6986.57
Male2692.62 ± 18.585.14100.09
Female4475.14 ± 19.269.2880.99
DBPM
Total7137.49 ± 8.575.4639.53
Male2438.42 ± 8.9734.6342.20
Female4737.02 ± 8.4134.5439.50
VR
Total711.89 ± .491.782.01
Male241.94 ± .551.712.17
Female471.87 ± .461.732.00

TotHz - total frequency power (hertz) of 24 hr heart rate variability in natural logarithm of ms2

LowHz - low frequency power (hertz) of 24 hr heart rate variability in natural logarithm of ms2

HiHz - high frequency power (hertz) of 24 hr heart rate variability in natural logarithm of ms2

pNN50 - percent of adjacent R-R intervals differing >50 milliseconds

rMSSD - square root of mean squared differences of successive R-R intervals

SDNN - standard deviation of all R-R intervals during a 24-hour period; measure of 24 hour circadian rhythmicity

SDANN - measure of variability within 5 minute blocks of time; measure of 24 hour circadian rhythmicity

SD - mean of the standard deviations of the R-R intervals of each 5-minute block

DBPM - difference in beats per minute with deep breathing

VR - Valsalva ratio

CI were not presented for the male and female subgroups for each 2-year age interval for youths 13 through 18 years because the majority were clustered in the early group and there were smaller samples in the middle and older groups.

Back to Article Outline

Discussion 

This investigation sought to identify normative values for HRV with the use of both short-term evoked measures and 24-hour ECG measures of HRV in healthy adolescents. As can be noted by reviewing Table I, our study had the largest reported sample of adolescents who have been included in studies of cardiovascular autonomic function to date. We explored the possibility of age-related variation in these measures during early, middle, and late adolescence. Although distinct age differences among the groups were not revealed, evoked measures of the difference in beats per minute with deep breathing and the Valsalva ratio declined as ages increased. These negative correlations suggest that parasympathetic modulation may decline as adolescents mature. Donaghue et al40 also found deep breathing HRV, but not Valsalva ratio, to be age-dependent. The fact that we found age to be related to the evoked tests also may have been related to greater variation in these measures as compared with the 24-hour HRV measures for our sample.

As a measure of sexual maturation, Tanner's stage generally correlates with age during adolescence. Although we did not find Tanner's staging to be associated with the evoked measures of Valsalva ratio or difference in beats per minute with deep breathing, the small but significant link with SDNN may be an indication of an increase in circadian variation with maturity. Additionally, as adolescents mature there is an increase in their BMI, similar to Tanner's staging. We did not find any linkage between BMI and HRV in our sample. One possible explanation is the affiliation of the healthy controls for comparison with adolescents with type 1 diabetes in the larger study, thus limiting the variation or spread of weight distribution in the sample. Teenagers with type 1 diabetes tend not to be overweight and have an average BMI for age.7

Major findings of our investigation include the significantly lower values for HRV in girls and African Americans. Most studies that report actual normative control values of HRV for youth typically have not addressed sex or racial differences.29, 31, 38, 39, 41 Donaghue et al40 found that the Valsalva ratio was significantly higher in boys versus girls (mean, 2.32 vs 1.88, respectively). Although our values for the Valsalva ratio were not significantly different (P = .56), values for girls (mean, 1.87) were similar to Donaghue et al40 and lower than those for boys (mean, 1.94). The only sex variation associated with age subgroup was in the early adolescence period for the time domain variables of SDNN, SDANN, and SD, which may have occurred as a result of the larger numbers for comparison in this category and the possible initiation of sex variation in circadian patterns.

One possible explanation for the significant racial effect of higher SDANN in whites in the late adolescence group may be related to confounding of African Americans with sex and age effects. Only 2 African American boys in early adolescence were assessed (Table II). Overall, girls tend to have lower HRV values than boys. Therefore, the differences detected were more likely attributable to sex rather than racial effects. However, investigators planning future studies should consider the possibility that racial as well as sex effects may affect HRV values.

When comparing our normative values of both the evoked and 24-hour HRV measures for the total sample with published data from other studies, we found some similarities for the evoked tests. Investigators used standard techniques for collecting data for the Valsalva ratio. Techniques varied somewhat for difference in beats per minute with deep breathing with study methods including instructing participants to take 6 deep breaths per minute versus 4 deep breaths for 20 seconds.29, 39, 40 Mean values for the difference in beats per minute ranged from 31.8 to 33.7,29, 39, 40, 41 slightly lower than our findings. One explanation for a higher value in our sample may be that the majority of youth were less than 14 years of age. Results published by Javorka et al39 were consistent with our results for the Valsalva ratio, falling within the 95% CI of 1.88 to 2.13 provided by Donaghue et al.40 We were not able to compare normative values for the HRV measures from the Holter recordings with other investigations that did report actual data for these variables because the recordings were either less than 24 hours or the measurement units included milliseconds, milliseconds squared, and logarithm (base 10) versus natural logarithm values for power spectral analysis obtained with the Marquette system. The variation of measurement units used for reporting results of HRV precludes synthesis of findings across investigations. As stated previously in this article, there is greater stability of HRV measures derived from 24-hour monitoring versus shorter recordings.23 Therefore, reports from shorter recordings cannot be interpreted with the same degree of accuracy as 24-hour recordings.

Jung et al64 investigated the reliability of time and frequency domain results with 4 different commercially available systems for HRV analysis. They found that HRV analysis of the same Holter recording by different systems is statistically uncomparable, additionally supporting the rationale for not synthesizing results of multiple studies with a variety of HRV software editing programs.

Although their sample consisted of controls with a mean age of 23 ± 4 years, Hoffman and Kienzle37 reported values for power spectral analysis with the same Holter series 8500 monitors and editing system (Marquette Electronics, Milwaukee, WI) used in our investigation. Mean and standard error values for total, low, and high frequency were 8.03 (± 0.13), 7.15 (± 0.14), and 6.12 (± 0.18), respectively. These values were similar to our findings except for the increased high-frequency power in our sample that is reflective of younger participants.

Back to Article Outline

Limitations and implications 

Given the limited studies of cardiovascular autonomic function in youth, this study clearly emphasizes the importance of sex differences in interpreting results for youth who may be at risk for dysautonomia. Previous investigations report results for the entire study sample and do not attempt to isolate findings on the basis of sex. Comparisons should be made to either male or female values as appropriate. One limitation of this study was the predominance of female to male participation and the concentration of youth in the early adolescent group (13 to 15 years of age). Additional studies with larger samples and more equally distributed age and sex subgroups are needed to support or refute findings that can be used later for normative comparisons.

This study, along with results of the Bogalusa Heart Study,55 suggests that differences in cardiovascular autonomic modulation between white and African American youth may be present. Determining such racial variations can offer valuable insight into preventative strategies for averting cardiac complications in target populations. Intentional efforts for recruiting minority youth to participate in cardiovascular research are particularly relevant because related morbidity and mortality rates are highest among African Americans.65

Researchers need to be aware of the difficulty in comparing results of HRV studies when various software editing and analysis programs are implemented. These programs may use system-dependent options for the removal of ectopic beats and artifact before computing the HRV analysis.64 The availability of this state-of-the-art technology for obtaining HRV data allows for the early identification of subclinical cardiac autonomic changes in youth who have predispositions for cardiac complications, such as those with diabetes, congential heart disease, or obesity. The remaining challenge is to validate findings that can then be useful for clinical interpretation and to develop interventions to prevent premature cardiac dysautonomia in youth at risk.

Back to Article Outline

References 

  1. Cowan MJ. Measurement of heart rate variability. West J Nurs Res. 1995;17:32–48
  2. Low PA, Pfeifer M. Standardization of autonomic function. In:  Low PA editors. Clinical autonomic disorders. Philadelphia: Lippincott-Raven; 1997;p. 287–295
  3. Stein PK, Bosner MS, Kleiger RE, Conger BM. Heart rate variability: a measure of cardiac autonomic tone. Am Heart J. 1994;127:1376–1381
  4. Ewing DJ, Martyn CN, Young RJ, Clarke BF. The value of cardiovascular autonomic function tests: 10 years experience in diabetes. Diabetes Care. 1985;8:491–498
  5. Ewing DJ, Campbell IW, Clarke BF. Heart-rate response to standing as a test for automatic neuropathy. Br Med J. 1978;1:1700
  6. Ewing DJ, Campbell IW, Murray A, Neilson JM, Clarke BF. Immediate heart-rate response to standing: simple test for autonomic neuropathy in diabetes. Br Med J. 1978;1:145–147
  7. Faulkner MS, Hathaway DK, Milstead EJ, Burghen GA. Heart rate variability in adolescents and adults with type 1 diabetes. Nurs Res. 2001;50:95–104
  8. Hathaway DK, Cashion AK, Wicks MN, Milstead EJ, Gaber AO. Cardiovascular dysautonomia of patients with end-stage renal disease and type I or type II diabetes. Nurs Res. 1998;47:171–179
  9. Makimattila S, Schlenzka A, Mantysaari M, et al.  Predictors of abnormal cardiovascular autonomic function measured by frequence domain analysis of heart rate variability and conventional tests in patients with type 1 diabetes. Diabetes Care. 2000;23:1686–1693
  10. Osterhues HH, Grossmann G, Kochs M, Hombach V. Heart-rate variability for discrimination of different types of neuropathy in patients with insulin-dependent diabetes mellitus. J Endocrinol Invest. 1998;21:24–30
  11. Wheeler T, Watkins PJ. Cardiac denervation in diabetes. Br Med J. 1973;4:584–586
  12. Johnson RH, Robinson BJ. Mortality in alcoholics with autonomic neuropathy. J Neurol Neurosurg Psych. 1988;51:476–480
  13. Malpas SC, Whiteside EA, Maling TJ. Heart rate variability and cardiac autonomic function in men with chronic alcohol dependence. Br Heart J. 1991;65:84–88
  14. Bigger JT, Kleiger RE, Fleiss JL, Rolnitzky LM, Steinman RC, Miller JP. Components of heart rate variability measured during healing of acute myocardial infarction. Am J Cardiol. 1990;65:408–411
  15. Kleiger RE, Miller JP, Krone RJ, Bigger JT. The independence of cycle length variability and exercise testing on predicting mortality of patients surviving acute myocardial infarction. The Multicenter Postinfarction Research Group Am J Cardiol. 1990;65:408–411
  16. La Rovere MTMD, Pinna GDMS, Hohnloser SHMD, et al.  Baroreflex sensitivity and heart rate variability in the identification of patients at risk for life-threatening arrhythmias: implications for clinical trials. Circulation. 2001;103:2072–2077
  17. Rothschild M, Rothschild A, Pfeifer M. Temporary decrease in cardiac parasympathetic tone after acute myocardial infarction. Am J Cardiol. 1988;62:637–639
  18. Casolo G, Balli E, Taddei T, Amuhasi J, Gori C. Decreased spontaneous heart rate variability in congestive heart failure. Am J Cardiol. 1989;64:1162–1167
  19. Casolo GC, Stroder P, Sulla A, Chelucci A, Freni A, Zerauschek M. Heart rate variability and functional severity of congestive heart failure secondary to coronary artery disease. Eur Heart J. 1995;16:360–367
  20. Levitt NS, Stansberry KB, Wynchank S, Vinik AI. The natural progression of autonomic neuropathy and autonomic function tests in a cohort of people with IDDM. Diabetes Care. 1996;19:751–754
  21. Clarke BF, Ewing DJ. Cardiovascular reflex tests; in the natural history of diabetic autonomic neuropathy. NY State J Med. 1982;82:903–908
  22. Akselrod S, Gordon D, Ubel FA, Shannon DC, Berger AC, Cohen RJ. Power spectrum analysis of heart rate fluctuation: a quantitative probe of beat-to-beat cardiovascular control. Science. 1981;213:220–222
  23. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology . Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation. 1996;93:1043–1065
  24. Tanaka H, Hyllienmark L, Thulesius O, et al.  Autonomic function in children with type 1 diabetes mellitus. Diabetic Med. 1998;15:402–411
  25. Gutin B, Owens S, Slavens G, Riggs S, Treiber F. Effect of physical training on heart-period variability in obese children. J Pediatr. 1997;130:938–943
  26. Gutin B, Barbeau P, Litaker MS, Ferguson M, Owens S. Heart rate variability in obese children: relations to total body and visceral adiposity, and changes with physical training and detraining. Obes Res. 2000;8:12–19
  27. Massin MM, Derkenne B, von Bernuth G. Correlations between indices of heart rate variability in healthy children and children with congenital heart disease. Cardiol. 1999;91:109–113
  28. Massin MM, Derkenne B, von Bernuth G. Heart rate behavior in children with atrial septal defect. Cardiol. 1998;90:269–273
  29. Wawryk AM, Bates DJ, Couper JJ. Power spectral analysis of heart rate variability in children and adolescents with IDDM. Diabetes Care. 1997;20:1416–1421
  30. Finley JP, Nugent ST, Hellenbrand W. Heart-rate variability in children. Spectral analysis of developmental changes between 5 and 24 years. Can J Physiol Pharmacol. 1987;65:2048–2052
  31. Goto M, Nagashima M, Baba R, et al.  Analysis of heart rate variability demonstrates effects of development on vagal modulation of heart rate in healthy children [see comments] [published erratum appears in J Pediatr 1997 Sep;131(3):502]. J Pediatr. 1997;130:725–729
  32. Massin M, von Bernuth G. Normal ranges of heart rate variability during infancy and childhood. Pediatr Cardiol. 1997;18:297–302
  33. Massin MM, Maeyns K, Withofs N, Ravet F, Gerard P. Circadian rhythm of heart rate and heart rate variability. Arch Dis Child. 2000;83:179–182
  34. Finley JP, Nugent ST. Heart rate variability in infants, children and young adults. J Auton Nerv Syst. 1995;51:103–108
  35. Yeragani VK, Sobolewski E, Kay J, Jampala VC, Igel G. Effect of age on long-term heart rate variability. Cardiovasc Res. 1997;35:35–42
  36. Tanaka H, Borres M, Thulesius O, Tamai H, Ericson MO, Lindblad LE. Blood pressure and cardiovascular autonomic function in healthy children and adolescents [see comments]. J Pediatr 200;137:63-7.
  37. Hoffman RP, Kienzle MG. Circadian control of heart rate in young insulin-dependent diabetes mellitus patients. J Diabetes Complications. 1996;10:220–222
  38. Akinci A, Celiker A, Baykal E, Tezic T. Heart rate variability in diabetic children: sensitivity of the time- and frequency-domain methods. Pediatr Cardiol. 1993;14:140–146
  39. Javorka K, Javorkova J, Petraskova M, Tonhajzerova I, Buchanec J, Chroma O. Heart rate variability and cardiovascular tests in young patients with diabetes mellitus type 1. J Pediatr Endocrinol Metabol. 1999;12:423–431
  40. Donaghue KC, Bonney M, Simpson JM, et al.  Autonomic and peripheral nerve function in adolescents with and without diabetes. Diabetic Med. 1993;10:664–671
  41. Barkai L, Madacsy L. Cardiovascular autonomic dysfunction in diabetes mellitus. Arch Dis Child. 1995;73:515–518
  42. Barkai L, Madacsy L, Vamosi I. Autonomic dysfunction and severe hypoglycaemia in insulin dependent diabetes mellitus. Arch Dis Child. 1991;66:1438–1441
  43. Barkai L, Madacsy L, Kassay L. Investigation of subclinical signs of autonomic neuropathy in the early stage of childhood diabetes. Horm Res. 1990;34:54–59
  44. Massin MM, Derkenne B, Tallsund M, et al.  Cardiac autonomic dysfunction in diabetic children. Diabetes Care. 1999;22:1845–1850
  45. Rollins MD, Jenkins JG, Carson DJ, McClure BG, Mitchell RH, Imam SZ. Power spectral analysis of the electrocardiogram in diabetic children. Diabetologia. 1992;35:452–455
  46. Burger AJ, Charlamb M, Sherman HB. Circadian patterns of heart rate variability in normals, chronic stable angina and diabetes mellitus. Int J Cardiol. 1999;71:41–48
  47. Buchanan LM, Cowan M, Burr R, Waldron C, Kogan H. Measurement of recovery from myocardial infarction using heart rate variability and psychological outcomes. Nurs Res. 1993;42:74–78
  48. Ewing DJ. Heart rate variability: an important new risk factor in patients following myocardial infarction. Clin Cardiol. 1991;14:683–685
  49. Thomson BJ, McAreavey D, Neilson JM, Winney RJ, Ewing DJ. Heart rate variability and cardiac arrhythmias in patients with chronic renal failure. Clin Auton Res. 1991;1:131–133
  50. Ringel RE, Chalew SA, Armour KA, McLaughlin J, McCarter RJ, Kramer WE. Cardiovascular reflex abnormalities in children and adolescents with diabetes mellitus. Diabetes Care. 1993;16:734–741
  51. Ewing DJ, Neilson JM, Travis P. New method for assessing cardiac parasympathetic activity using 24 hour electrocardiograms. Br Heart J. 1984;52:396–402
  52. Barkai L, Peja M, Vamosi I. Physical work capacity in diabetic children and adolescents with and without cardiovascular autonomic dysfunction. Diabetic Med. 1996;13:254–258
  53. Stein PK, Kleiger RE, Rottman JN. Differing effects of age on heart rate variability in men and women. Am J Cardiol. 1997;80:302–305
  54. Umetani K, Singer DH, McCraty R, Atkinson M. Twenty-four hour time domain heart rate variability and heart rate: relations to age and gender over nine decades [see comments]. J Am Coll Cardiol. 1998;31:593–601
  55. Urbina EM, Bao W, Pickoff AS, Berenson GS. Ethnic (black-white) contrasts in heart rate variability during cardiovascular reactivity testing in male adolescents with high and low blood pressure: the Bogalusa Heart Study. Am J Hypertens. 1998;11:196–202
  56. Siervogel R, Roche AF, Guo SM, Mukherjee D, Chumlea WC. Patterns of change in weight/stature2 from 2 to 18 years: Findings from long-term serial data for children in the Fels longitudinal growth study. Int J Obes. 1991;15:479–485
  57. Morris NM, Udry JR. Validation of a self-administered instrument to assess stage of adolescent development. J Youth Adolesc. 1980;9:271–280
  58. Spallone V, Menzinger G. Diagnosis of cardiovascular autonomic neuropathy in diabetes. Diabetes. 1997;46(Suppl 2):S67–S76
  59. Wilson BA, Shannon MT, Strang CL. Nurses' drug guide. Stamford (CT): Appleton & Lange; 1999;
  60. Hartwig MS, Cardoso SS, Hathaway DK, Gaber AO. Reliability and validity of cardiovascular and vasomotor autonomic function tests. Diabetes Care. 1994;17:1433–1440
  61. Kleiger RE, Miller JP, Bigger JT, Moss AJ. Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol. 1987;59:256–262
  62. Keehn M. Heart rate variability physician's guide. Milwaukee (WI): Marquette Electronics; 1992;
  63. Kleiger RE, Bigger JT, Bosner MS, et al.  Stability over time of variables measuring heart rate variability in normal subjects. Am J Cardiol. 1991;68:626–630
  64. Jung J, Heisel A, Tscholl D, Fries R, Schieffer H, Ozbek C. Assessment of heart rate variability by using different commercially available systems. Am J Cardiol. 1996;78:118–120
  65. US Department of Health and Human Services . Healthy people 2010: understanding and improving health and objectives for improving health. 2 vol. Washington, DC: US Government Printing Office; 2000;

☆ Funded by PHS Grant No. K08 NR00075, National Institute of Nursing Research.

☆☆ Reprint requests: Melissa S. Faulkner, RN, DSN, University of Illinois at Chicago, Department of Maternal-Child Nursing (MC 802), College of Nursing, 845 South Damen Ave, Chicago, IL 60612.

★ 0147-9563/2003/$30.00 + 0

PII: S0147-9563(02)70206-9

doi:10.1067/mhl.2003.6

Heart & Lung: The Journal of Acute and Critical Care
Volume 32, Issue 1 , Pages 10-22, January 2003

Article Tools