(Chest. 2005;128:1266-1273.)
© 2005
American College of Chest Physicians
Impulse Oscillometry*
Reference Values in Children 100 to 150 cm in Height and 3 to 10 Years of Age
Jennifer Frei;
Jagtar Jutla;
Gretchen Kramer;
George E. Hatzakis, PhD, MSc;
Francine M. Ducharme, MD, MSc and
G. Michael Davis, MB,ChB, MSc
* From the Department of Pediatrics (Ms. Frei, Ms. Jutla, Ms. Kramer, and Dr. Davis), Division of Respiratory Medicine, Montreal Childrens Hospital; Department of Medicine (Dr. Hatzakis), Divisions of Clinical Immunology & Allergy, Clinical Epidemiology Montreal General Hospital; and Department of Epidemiology and Community Health (Dr. Ducharme), McGill University Health Centre, Montreal, QC, Canada.
Correspondence to: George E. Hatzakis, PhD, MSc, Montreal General Hospital, Room A5-145, 1650 Cedar Ave, Montreal, QC, H3G 1A4, Canada; e-mail: Georges.Hatzakis{at}gmail.com
 |
Abstract
|
|---|
Objectives: To generate reference equations in North American children to be used for assessing respiratory function through the forced oscillation (Rfo) technique, and to determine the changes in oscillatory resistance, reactance, and resonant frequency (Fres) in relation to age, body height, and weight.
Design/setting: A prospective cross-sectional study performed on healthy children selected according to strict criteria of American Thoracic Society and European Respiratory Society recommendations.
Measurements: Triplicate measures were obtained of resistance and reactance at 5, 10, 15, 20, 25, and 35 Hz as well as Fres through the impulse oscillometer (MasterScreen IOS; Jaeger/Toennies; Höchberg, Germany). Two hundred twenty-two white childrennormally distributed within the 3- to 10-year age range and 100 to 150 cm in heightwere recruited in Montreal, Canada. We used regression analysis to generate multiple predictive equations separately per gender and frequency on age, height, and body weight.
Results: Stepwise multiple regression in both natural and logarithmic forms for height, weight, age, and gender showed that standing height was the only significant predictor for all variables. Minimal variability was noted in each subject among the triplicate measurements (p = 0.68 to 0.96). Coherence was > 0.9 at all oscillating frequencies except 5 Hz (< 0.72), with tendencies to lower values in young children.
Conclusions: Resistance and Fres decrease by height, but also by age; and reactance increases. As opposed to our past experience with spirometry in compatible age groups, the Rfo technique was well accepted by preschool children.
Key Words: children pulmonary function reference values regression
 |
Introduction
|
|---|
As a noninvasive and effort-independent technique, respiratory resistance obtained by the forced oscillation (Rfo) technique is well suited for lung function measurement in young children.12345 Previous publications1678 in adults have confirmed the strong correlation between airway resistance measured by body plethysmography and respiratory resistance measured by forced oscillation, thus attesting to the validity of this technique. We and others4789 have demonstrated the general applicability of this Rfo technique in previously untrained asthmatic children aged > 3 years and tested during an acute exacerbation but using a different device. Rfo measurements depend on reproducible tidal breathing. Children of preschool age can better adhere to this test in contrast to spirometry/pulmonary function testing, which, albeit established, requires the subject to perform a more difficult maneuver. Indeed, we have demonstrated that children at the age of 3 years can perform Rfo at a 50% success rate.49
Reference values have been published previously for children using random frequency10 and fixed frequency (6 to 10 Hz). Several commercial devices for the measurement of oscillatory resistance at multiple frequencies have become generally available (MasterScreen IOS; Jaeger/Toennies; Höchberg, Germany; and Vmax System R.O.S. Oscilink; SensorMedics; Yorba Linda, CA),91112 superseding previous equipment at fixed frequencies. Nevertheless, reference values specific for this measurement technique are required to interpret resistance measurements obtained in children for whom no prior best values are available. The only reference values that have previously been reported using an Rfo technique were limited to children from 3 to 6 years old.713 Furthermore, there are no published reference values obtained with the Rfo system for North American children.
The selection of a normal population to establish reference values has been carefully defined by the European Respiratory Society (ERS) and the American Thoracic Society (ATS).1415 Only healthy children free from conditions known to adversely affect ventilatory function should be considered, with adherence to these strict recommendations being documented. Furthermore, although the strongest determinant of lung function parameters is height, models have also occasionally included age and/or weight.16 It is also well recognized that gender and/or race frequently influence the relationship between these determinants and lung function parameters.17 Nevertheless, the purpose of this study was to elaborate specific reference values of Rfo and concomitant variables for North American children using the ATS and ERS recommendations.
 |
Materials and Methods
|
|---|
We conducted a cross-sectional study of patients and/or accompanying siblings, aged 3 to 10 years, who presented at the Ophthalmology or General Surgery Clinics of the Montreal Children Hospital. Patients were accrued over two consecutive summers, at a time when viral infections were minimal. The protocol was reviewed and approved by the Institutional Review Board, and informed consent for participation was obtained for each subject from parents or guardians.
Study Subjects
In accordance with the ATS and ERS recommendations for the selection of healthy subjects,1415 exclusions occurred for the following reasons: (1) personal or family history (including parents and siblings) of wheezing or asthma1819; (2) personal history of allergic rhinitis2021 or eczema2223; (3) low birth weight (< 1,500 g)2425; premature birth (< 37 weeks), neonatal mechanical ventilation, or bronchopulmonary dysplasia2627; (4) passive smoking in the house2829; (5) obesity (weight for height > 95% of predicted)30; (6) concurrent upper respiratory tract infection313233; (7) dyspnea, cough, wheezing, accessory muscle use, or an abnormal (< 95%) oxygen saturation level as measured by pulse oximetry (Nellcor N10; Nellcor; Hayward, CA); or (8) any other contraindication in obtaining respiratory resistance measurement (such as a significant facial or oral abnormality). Subjects could be enrolled only once in the study to ensure independence of participants and of the pertinent recorded data.
Procedures
Assessment of respiratory function through the Rfo technique is used on a daily basis in our clinic. In 2003 only, of a total of 3,997 children evaluated for pulmonary function, the Rfo technique was employed in 761 children, the majority of whom were of preschool age. However, for the purposes of the current study, potentially eligible subjects visiting the Ophthalmology or General Surgery Clinics were approached and initially screened by questionnaire. The 12-item questionnaire, extracted from the International Study of Asthma and Allergies in Childhood questionnaire,1934 screened for a personal and family history suggestive of asthma, eczema, and rhinitis. Moreover, we added three questions related to exposure to cigarette smoke (current smoking by the immediate family, regular exposure to secondhand smoke, and regular exposure to active or passive smoking in the preceding year).
A trained researcher (J.F. or J.J.) approached all subjects, and measurements were obtained independent of the results of the screening questionnaire. This study reports only the data for subjects who were retained after analysis of the questionnaire. Demographic data, including age, gender, race, and medical history, were recorded, along with height by stadiometer (Holtain Ltd; Crymych; Dyfed, UK) and weight by an electronic scale (Ancaster Scale; Brantford, ON, Canada). The absence of respiratory symptoms as well as the normality of oxygen saturation (> 95%) was documented by the interviewer.
For all assessing subjects, three replicate measurements of oscillatory resistance (Rfo) were obtained using the system software (MasterLab, Version 4.53; Jaeger/Toennies). Resistance measurements were retained for analysis if reproducible, that is, if the coefficient of variation between replicate measurements was < 0.15.353637 Furthermore, we aimed for a coherence measurement of > 0.9 at all frequencies in all children.
For measurement of respiratory resistance, the standing child was asked to breathe quietly for 15 to 20 s using a rigid oval mouthpiece with a tongue guard, with the head in a neutral position, nose clips in place, and while supporting both cheeks. This Rfo technique has been described in detail elsewhere.7
Briefly, small, square-wave pressure oscillations (< 0.1 kPa) were superimposed on the spontaneous breathing pattern of the child, with pressure and airflow variations measured by a pneumotachograph-transducer system at the mouthpiece. The magnitude and phase of the flow oscillations resulting from the pressure fluctuations are dependent on the total impedance of the respiratory system. Through the phase relationship between pressure and flow, the impedance is then partitioned into the resistance and the imaginary portion, the reactance. For the respiratory system, resistance represents the effective resistance of lungs and chest wall, whereas reactance is the net effect of the two opposite (a compliant and an inertial) components. Each recording on the MasterScreen IOS assessment yielded both the expiratory resistance and reactance at different oscillatory frequencies between 5 Hz and 35 Hz within the flow range of normal tidal breathing. In addition, the resonant frequency (Fres) [ie, the frequency at which the reactance was zero] and the area of the reactance curve less than zero (AX) were also computed. All computations occur automatically with the MasterLab software.
Statistical Analysis
Stepwise multiple regressions were carried out to identify the best predictors of respiratory resistance parameters using height, weight, gender, age, and ethnicity as potential determinants. Log-transformation of variables were also tested for the best-fitting function using the method of least squares. The fifth and ninety-fifth percentiles were calculated using parametric methods. Analysis of variance was used to test the difference of triplicate trials recorded in each subject. Pearson R correlation coefficients were computed between Rfo values and age, height, and weight.
 |
Results
|
|---|
During the study period, 1,127 children were approached for the study. Of these, 247 children (22%) were not involved in the study because of parental refusal, and another 65 children (5.5%) were excluded because parents were not available, or because of language barrier. Of the remaining 715 children, 72 children (6.3%) were not evaluated because of concurrent enrolment of another candidate or they were ready to be seen by the treating physician for their clinical disease. Thus, 643 children were screened using our previously described self-administered questionnaire. Of these, 404 children (62%) were excluded because of the following nonmutually exclusive reasons: (1) family history of wheezing or asthma (38% of sample, n = 247), (2) personal history of wheezing or asthma (26%, n = 170), (3) personal history of rhinitis or eczema (33%, n = 215), and (4) active or passive smoking (47%, n = 303). In addition, a further 6 children were excluded because of neonatal conditions (prematurity), and 11 children because of their inability to cooperate with resistance measurement without a leak at the mouth or obstruction of the mouthpiece. Thus, a total of 222 children aged 3 to 10 years are reported in this article (male gender, 54%; n = 121). All participants were white with both white parents.
The distributions of age and height for the sample are provided in Figure 1
. Age and height were strongly correlated (R = 0.81). Reproducible values were obtained for resistance, reactance, and Fres in 222 children and AX in 177 children. Stepwise multiple regression of the data points, using the lowest value for each patient, in both natural and logarithmic form (on height, weight, age, and gender), showed that standing height was the only significant predictor for all variables, except for reactance at an oscillating frequency of 20 Hz (height and age). With or without logarithmic transformation, neither gender nor race appeared as an important covariate. A summary of the linear regression equations appears in Table 1
.
There was minimal variability among measurement trials after selection for technically acceptable data. The triplicate trials for each frequency were compared by analysis of variance to show no significant differences between trials (p values from 0.68 to 0.96). The median trial was used in the regression modeling. However, the coherence, ie, the "goodness-of-fit" of the measured data to the mathematical model used, was low (< 0.65) at 5 Hz, but progressively increased (> 0.88) at 10 Hz and above (Fig 2
).
 |
Discussion
|
|---|
In this group of 222 healthy children, oscillatory resistance (Rfo) was closely related to height, with 56 to 66% of the variance explained solely by the measurement of stature. Log-transformation of height and resistance did not improve this relationship. Furthermore, standing height correlated at least as well as age with resistance (Table 2
). Moreover, consideration of gender or race as individual classification factors did not improve prediction in multiple regression analysis. This study presents reference values for respiratory resistance, between 5 Hz and 35 Hz, along with data on reactance, Fres, and AX with fifth and ninety-fifth percentiles of the data using height measurements. Moreover, the results obtained were nearly identical with the normal values published by Hellinckx,7 and similar to values of Klug and Bisgaard,13 although a minor technique difference may explain the variation in values (Fig 3
). Despite the similarities of these study findings, normative values for North American children have not been published. Although there was no indication that North American children would perform differently than European children, the argument in question had yet to be determined. Moreover so, pulmonary function testing through the Rfo technique is gaining ground for assessment of preschool children. Bridging the gap between Rfo and "traditional" pulmonary function measurements, studies3839 have correlated Rfo resistance and FEV1. Yet, it is not clear what reactance measurements represent, and it has been postulated that reactance is rather a surrogate for the viscoelastic properties of the lung.40
View this table:
[in this window]
[in a new window]
|
Table 2. Pearson Correlation Coefficient for Resistance, Reactance, and Fres Against Age, Body Height, and Weight Separately for Boys and Girls*
|
|

View larger version (12K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Our prediction of normative Rfo values matches those from Hellinckx,8 and it is similar to values from Klug and Bisgaard,13 although a minor technique difference may explain the variation in values.
|
|
We intentionally chose stringent eligibility criteria for our sample in accordance with current recommendations issued by the ATS and the ERS.1415 These children were recruited in the General Surgical or Ophthalmology Clinics when they or their siblings presented with minor surgical/eye problems. The use of these strict eligibility criteria ensured that only children with no known previous or present respiratory conditions that may possibly impair lung function were enrolled in this study. Active/passive smoking, personal or family asthma, and personal atopy resulted in the exclusion of 68% of the 743 children who were approached. Furthermore, we believe the accuracy of the reference value equations was improved by only including subjects who demonstrated reproducibility of the test (intertest variability of < 10%). In addition, to avoid multicolinearity-related interactions of the triplicate measurements within the data set, only one set of measurements from each patient were used in the regression analysis (median). The use of the ATS/ERS guidelines defining acceptable "normal" subjects for predictive values, while severely restricting the sample size (< 30% of subjects approached), has resulted in a homogeneous group comparable to other studies of normal values.
In concordance with other investigators, we also found that there was frequency dependence of resistance throughout the population examined, with resistance falling with increasing height, and with increasing oscillating frequency. Indeed, the sole significant predictor for Rfo was height, a finding consistent with other studies713 of normal children within this height range. Furthermore, in this age range (3 to 10 years), there was no relationship between Rfo and gender, a result that is consistent with the previous studies713 of oscillatory resistance in prepubescent children.
Despite our assumption that the coherence function for all measurements would be > 0.9, this was not true, and the greatest abnormality was found in the measurements of resistance at an oscillating frequency of 5 Hz, where the coherence factor was a mean of 0.72, and frequently < 0.60. There was a tendency for coherence at 5 Hz to be lower in the younger child, especially those < 7 years old, but this was not statistically significant. In agreement with the data of Klug and Bisgaard,13 the coherence rose to > 0.9 at other frequencies, but we did not find the age difference at higher frequencies. At each frequency, the fit of the regression equations was very similar when standing height was tested, as has previously been shown in pediatric studies4816 investigating the relationship between lung function and anthropometric indexes.
The results of our study demonstrate that not only is there a negative linear correlation of resistance with height but also a frequency dependence of resistance (Fig 4
). Conversely, a positive linear correlation of reactance with height was noted (Fig 5
). Our regression coefficients of resistance with height are strikingly similar to those published by Hellinckx et al7 (Fig 3). These results differ from those published by Klug and Bisgaard,13 but the difference is primarily that of an "offset" (intercept difference) rather than a difference in the slope of the resistance to height relationship. In part, the difference may be explained by a minor difference in measurement technique as a facemask, rather than a mouthpiece, was used.
Similarly, there is a difference in reactance, with these results being higher than those previously reported for the same equipment. In addition, we report Fres (the frequency at which reactance = 0) based on height with results that are similar to Klug and Bisgaard,13 but markedly higher (Fres > 15 Hz) than those reported by Duiverman et al.10
Goldman41 has proposed an alternative method of assessing the low-frequency capacitance of the system, termed AX. It is an integrated sum of the negative reactance values below the Fres, resulting in a better "signal-to-noise" ratio. These low-frequency signals most clearly reflect peripheral airway obstruction, and the use of an integrated result avoids the problem of low coherence in the 5-Hz signal. In small children, the Fres is higher, related to the airway size, and it is possible that the calculated AX will include inappropriate data points. The role of AX measurements is not clear, and further developmental work is needed, especially in small children, to explore the utility of this measurement when the upper limit of incorporation into AX is defined by the Fres. In conclusion, we present reference values for oscillatory resistance measurements for the MasterScreen IOS System generated from a group of North American children.
 |
Footnotes
|
|---|
Abbreviations: ATS = American Thoracic Society; AX = area of the reactance curve less than zero; ERS = European Respiratory Society; Fres = resonant frequency; Rfo = forced oscillation
Dr. Hatzakis was supported by a Chercheur Boursier Career Award from the Fonds de la Recherche en Santé du Québec.
Received for publication November 2, 2004.
Accepted for publication February 2, 2005.
 |
References
|
|---|
- Buhr, W, Jorres, R, Berdel, D, et al (1990) Correspondence between forced oscillation and body plethysmography during bronchoprovocation with carbachol in children. Pediatr Pulmonol 8,280-288[ISI][Medline]
- Bohadana, AB, Peslin, R, Megherbi, SE, et al Dose-response slope of forced oscillation and forced expiratory parameters in bronchial challenge testing. Eur Respir J 1999;13,295-300[Abstract]
- Delacourt, C, Lorino, H, Herve-Guillot, M, et al Use of the forced oscillation technique to assess airway obstruction and reversibility in children. Am J Respir Crit Care Med 2000;161,730-736[Abstract/Free Full Text]
- Ducharme, FM, Davis, GM, Ducharme, GR Pediatric reference values for respiratory resistance measured by forced oscillation. Chest 1998;113,1322-1328[Abstract/Free Full Text]
- Lebecque, P, Stanescu, D Respiratory resistance by the forced oscillation technique in asthmatic children and cystic fibrosis patients. Eur Respir J 1997;10,891-895[Abstract]
- Chinet, T, Pelle, G, Macquin-Mavier, I, et al Comparison of the dose-response curves obtained by forced oscillation and plethysmography during carbachol inhalation. Eur Respir J 1988;1,600-605[Abstract]
- Hellinckx, J, De Boeck, K, Bande-Knops, J, et al Bronchodilator response in 36.5 years old healthy and stable asthmatic children. Eur Respir J 1998;12,438-443[Abstract]
- Hellinckx, J, Cauberghs, M, De Boeck, K, et al Evaluation of impulse oscillation system: comparison with forced oscillation technique and body plethysmography. Eur Respir J 2001;18,564-570[Abstract/Free Full Text]
- Ducharme, FM, Davis, GM Measurement of respiratory resistance in the emergency department: feasibility in young children with acute asthma. Chest 1997;111,1519-1525[Abstract/Free Full Text]
- Duiverman, EJ, Clement, J, van de Woestijne, KP, et al Forced oscillation technique: reference values for resistance and reactance over a frequency spectrum of 226 Hz in healthy children aged 2.312.5 years. Bull Eur Physiopathol Respir 1985;21,171-178[ISI][Medline]
- Lebecque, P, Desmond, K, Swartebroeckx, Y, et al Measurement of respiratory system resistance by forced oscillation in normal children: a comparison with spirometric values. Pediatr Pulmonol 1991;10,117-122[ISI][Medline]
- Gimeno, F, van der Weele, LT, Koeter, GH, et al Forced oscillation technique: reference values for total respiratory resistance obtained with the Siemens Siregnost FD5. Ann Allergy 1992;68,155-158[ISI][Medline]
- Klug, B, Bisgaard, H Specific airway resistance, interrupter resistance, and respiratory impedance in healthy children aged 27 years. Pediatr Pulmonol 1998;25,322-331[CrossRef][ISI][Medline]
- American Thoracic Society.. Lung function testing: selection of reference values and interpretative strategies. Am Rev Respir Dis 1991;144,1202-1218[ISI][Medline]
- Stocks, J, Quanjer, PH Reference values for residual volume, functional residual capacity and total lung capacity: ATS Workshop on Lung Volume Measurements; Official Statement of The European Respiratory Society. Eur Respir J 1995;8,492-506[CrossRef][ISI][Medline]
- Quanjer, PH, Stocks, J, Polgar, G, et al Compilation of reference values for lung function measurements in children. Eur Respir J 1989;4(Suppl),184S-261S
- Yang, TS, Peat, J, Keena, V, et al A review of the racial differences in the lung function of normal Caucasian, Chinese and Indian subjects. Eur Respir J 1991;4,872-880[Abstract]
- Nicolai, T, Mutius, EV, Reitmeir, P, et al Reactivity to cold-air hyperventilation in normal and in asthmatic children in a survey of 5,697 schoolchildren in southern Bavaria. Am Rev Respir Dis 1993;147,565-572[ISI][Medline]
- Asher, MI, Keil, U, Anderson, HR, et al International Study of Asthma and Allergies in Childhood (ISAAC): rationale and methods. Eur Respir J 1995;8,483-491[Abstract]
- Kelly, WJ, Hudson, I, Phelan, PD, et al Atopy in subjects with asthma followed to the age of 28 years. J Allergy Clin Immunol 1990;85,548-557[CrossRef][ISI][Medline]
- Annesi, I, Oryszczyn, MP, Neukirch, F, et al Relationship of upper airways disorders to FEV1 and bronchial hyperresponsiveness in an epidemiological study. Eur Respir J 1992;5,1104-1110[Abstract]
- Salob, SP, Laverty, A, Atherton, DJ Bronchial hyperresponsiveness in children with atopic dermatitis. Pediatrics 1993;91,13-16[Abstract/Free Full Text]
- Jones, A Asymptomatic bronchial hyperreactivity and the development of asthma and other respiratory tract illnesses in children. Thorax 1994;49,757-761[Abstract]
- Riedel, F Long term effects of artificial ventilation in neonates. Acta Paediatr Scand 1987;76,24-29[ISI][Medline]
- Rona, RJ, Gulliford, MC, Chinn, S Effects of prematurity and intrauterine growth on respiratory health and lung function in childhood. BMJ 1993;306,817-820[ISI][Medline]
- Kitchen, WH, Olinsky, A, Doyle, LW, et al Respiratory health and lung function in 8-year-old children of very low birth weight: a cohort study. Pediatrics 1992;89,1151-1158[Abstract/Free Full Text]
- Doyle, LW, Ford, GW, Olinsky, A, et al Bronchopulmonary dysplasia and very low birth weight: lung function at 11 years of age. J Paediatr Child Health 1996;32,339-343[ISI][Medline]
- Sherrill, DL, Martinez, FD, Lebowitz, MD, et al Longitudinal effects of passive smoking on pulmonary function in New Zealand children. Am Rev Respir Dis 1992;145,1136-1141[ISI][Medline]
- Cook, DG, Strachan, DP Health effects of passive smoking-10: Summary of effects of parental smoking on the respiratory health of children and implications for research. Thorax 1999;54,357-366[Abstract/Free Full Text]
- Inselma, LS, Milanese, A, Deurloo, A Effect of obesity on pulmonary function in children. Pediatr Pulmonol 1993;16,130-137[ISI][Medline]
- Openshaw, PJ, Lemanske, RF Respiratory viruses and asthma: can the effects be prevented? Eur Respir J 1998;27(Suppl),35s-39s
- Johnston, SL, Pattemore, PK, Sanderson, G, et al The relationship between upper respiratory infections and hospital admissions for asthma: a time-trend analysis. Am J Respir Crit Care Med 1996;154,654-660[Abstract]
- Johnston, SL, Pattemore, PK, Sanderson, G, et al Community study of role of viral infections in exacerbations of asthma in 911 year old children. BMJ 1995;310,1225-1229[Abstract/Free Full Text]
- Kuehni, CE, Brooke, AM, Silverman, M Prevalence of wheeze during childhood: retrospective and prospective assessment. Eur Respir J 2000;16,81-85[Abstract]
- Solymar, L, Aronsson, PH, Bake, B, et al Respiratory resistance and impedance magnitude in healthy children aged 218 years. Pediatr Pulmonol 1985;1,134-140[Medline]
- Hordvik, NL, Konig, P, Morris, DA, et al Normal values for forced oscillatory respiratory resistance in children. Pediatr Pulmonol 1985;1,145-148[Medline]
- Goldman, MD, Carter, R, Klein, R, et al Within- and between-day variability of respiratory impedance, using impulse oscillometry in adolescent asthmatics. Pediatr Pulmonol 2002;34,312-319[CrossRef][ISI][Medline]
- Bisgaard, H, Klug, B Lung function measurement in awake young children. Eur Respir J 1995;8,2067-2075[Abstract]
- Wang, M, Niu, S, Li, Y, et al The diagnostic value of total respiratory impedance by impulse oscillometry in chronic obstructive lung disease. Chin Med J (Engl) 1999;112,982-984
- Vogel, J, Smidt, U Impulse oscillometry. 1994 pmi Verlagsgruppe GmbH. Frankfurt am Main, Germany:
- Goldman, MD Clinical application of forced oscillation. Pulm Pharmacol Ther 2001;14,341-350[CrossRef][ISI][Medline]