INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The measurement of respiratory muscle strength is important in children with neuromuscular or skeletal disorders. In the presence of a neuromuscular disease, respiratory muscle strength can be reduced when lung volumes are still in the normal range. Conversely, in case of scoliosis respiratory muscle strength can be normal in spite of reduced lung volumes (1). Furthermore, interventions such as inspiratory muscle training in neuromuscular patients may improve inspiratory muscle strength, but not lung volumes (2).

The classic tests of respiratory muscle strength are maximal inspiratory pressure (PI_max) and maximal expiratory pressure (PE_max) developed volitionally against a near complete occlusion (3). In children, these pressures were found to be relatively close to those of adults (4). However, as in adults, the inferior limits of normal values are low for PI_max and PE_max, in particular in young children, probably reflecting the difficulty of these maneuvers for some subjects.

To obviate this problem, several tests of inspiratory muscle strength have been developed based on the sniff, which is a natural and easy maneuver (8, 9). The sniff nasal inspiratory pressure (Pn_sn) is a new noninvasive test. It consists of measuring nasal pressure in an occluded nostril during a maximal sniff performed through the contralateral nostril, and has been validated in adults (10). Normal values were established for Pn_sn in healthy adults, and were most often higher than PI_max (11). The aims of this study were to assess the feasability of Pn_sn, to establish reference values, and to compare them with PI_max in a large group of unselected healthy children.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

A total of 203 children of a school in the Lausanne area, located in a middle-class suburb, participated in this study. For each grade, one class was included and all children were asked to participate. Consent was obtained from parents who also completed a short medical questionnaire. The study was approved by the ethics committee of the Faculty of Medicine, University of Lausanne.

Twenty-three children were excluded from analysis for the following reasons. Seven were of non-European descent. Thirteen had an abnormal spirometry: forced vital capacity (FVC) < 80% of predicted value or forced expiratory volume in one second to forced vital capacity ratio (FEV₁/FVC) < 85% of predicted value. Two children were currently treated for asthma and one had a metabolic disease.

Finally, 180 children were included in the study. All were Caucasians, and none was suffering from a known metabolic, neuromuscular, or cardiac disease. Twelve children with known asthma were included because they had a normal spirometry and were not currently receiving antiasthmatic therapy. All children were studied at a time when they were free of upper airway infection, because Pn_sn loses its validity in cases of marked nasal congestion (10).

Experimental Protocol

All measurements were performed by the same investigator in a single session for each subject. Height and weight were measured and body mass index (BMI) was calculated as weight/height². After a brief instruction on the different tests, the investigator measured Pn_sn, PI_max, PE_max, and spirometry. All measures were taken in the sitting position.

Pn_sn was measured in an occluded nostril during a maximal sniff performed by the contralateral nostril (10). The plug was made of waxed ear plugs (Calmor, Neuhausen am Rheinfall, Switzerland) hand-fastened around the tip of a catheter (internal diameter, 1 mm; length, 100 cm). The catheter was connected to a hand-held pressure meter displaying peak pressure (Pmax Mouth Pressure Monitor; P. K. Morgan, Rainham-Gillingham, Kent, UK). Pn_sn was measured during 10 maximal sniffs performed from FRC, each separated by 30 s. All maneuvers were recorded and the highest pressure was considered.

The PI_max was measured using a standard flanged mouthpiece connected to a hand-held pressure meter computing average pressure sustained over 1 s (Mouth Pressure Meter; P. K. Morgan). The subjects were studied with their nose occluded with a noseclip. They were asked to perform five maximal inspiratory efforts from FRC, each separated by 30 to 60 s. PI_max was then measured from residual volume (RV) according to the same technique. Similarily, PE_max was measured during five maximal expiratory efforts from both TLC and FRC. Care was taken to eliminate any air leak around the mouthpiece. For each test, all trials were recorded and the highest pressure was considered.

Spirometry was measured with a portable device (Multispiro SA/ 100; Medical Equipment Designs, Laguna Hills, CA), which was calibrated before each session with a 3-L syringe. The subjects had their nose occluded by a noseclip. Three to eight forced expiratory maneuvers were performed until the difference between the best two trials (sum of FEV₁ and FVC) was inferior to 5%. The predicted equations of Polgar were used (12).

Data Analysis

For each sex, data were expressed as means, SD, and range. When pressures were related to age, they were also presented in two age groups, from 6 to 12 yr and from 13 to 17 yr. Linear regression analysis was used to assess the relationships between respiratory pressures (Pn_sn, PI_max, PE_max) and age, height, and weight. Pn_sn was compared between boys and girls, and between boys 6 to 12 yr and 13 to 17 yr using two-tailed unpaired t tests. Pn_sn and PI_max were compared using two-tailed paired t tests. The agreement between Pn_sn and PI_max was assessed by the method of differences against the means according to Bland and Altman (13). The coefficient of variation was used to express the within-session reproducibility of Pn_sn, PI_max, and PE_max.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Anthropometry

All boys were within the normal growth curves for height and weight for Swiss children (14), except one above the 97th percentile for weight. All girls were within the normal growth curves, except two above the 97th percentile for weight, two above the 97th percentile for height, and one above the 97th percentile for height and weight. In boys, the BMI was 17.3 ± 2.5 kg/m² (range, 12.6 to 23.8 kg/m²). In girls, the BMI was 17.8 ± 2.8 kg/m² (range, 13.1 to 24.5 kg/m²).

Pn_sn

The Pn_sn maneuver was performed by all children without difficulty. Considering the entire groups, Pn_sn was 104 ± 26 cm H₂O in boys and 93 ± 23 cm H₂O in girls (p < 0.005). In boys, Pn_sn correlated with age, height, and weight (Table 1). The prediction equation in boys is: Pn_sn (cm H₂O) = 3.3 age + 70; residual standard deviation = 24.3. Subtracting 1.64 residual standard deviation from the predicted value will provide the lower limit above which lie 95% of normal boys. Pn_sn was higher in boys 13 to 17 yr than in boys 6 to 12 yr (p < 0.005; Table 2). In girls, Pn_sn did not correlate with age, height, or weight (Table 1). The relationship between Pn_sn and age in boys and girls is presented in Figure 1. The mean within-session coefficient of variation of Pn_sn was 16.2% in boys and 17.0% in girls (Table 3).

View larger version (13K): [in this window] [in a new window]   Figure 1.   Relationship between Pnsn and age in boys and girls.

PI_max and PE_max

In boys 6 to 17 yr, the mean values were 88 ± 24 cm H₂O for PI_maxFRC, 94 ± 23 cm H₂O for PI_maxRV, 87 ± 26 cm H₂O for PE_maxFRC, and 103 ± 27 cm H₂O for PE_maxTLC. All pressures correlated with age, height, and weight (Table 1). In girls 6 to 16 yr, the mean values were 72 ± 19 cm H₂O for PI_maxFRC, 79 ± 20 cm H₂O for PI_maxRV, 68 ± 23 cm H₂O for PE_maxFRC, and 83 ± 23 cm H₂O for PE_maxTLC. All pressures correlated with age, height, and weight, except PE_maxFRC, which did not correlate with age or height (Table 1). For boys and girls, the pressures are presented according to age groups in Table 2. The mean within-session coefficients of variation of PI_max and PE_max are presented in Table 3.

In both sexes, Pn_sn was higher than PI_max measured at the same lung volume (FRC) (p < 0.0001). The value of Pn_sn was higher or equal to PI_maxFRC in 73 of 93 boys and 79 of 87 girls (Figure 2). In boys, the mean difference Pn_sn  PI_maxFRC was 16.6 ± 20.4 cm H₂O, and the limits of agreement were 57.4 cm H₂O and 24.2 cm H₂O. In girls, the mean difference Pn_sn  PI_maxFRC was 21.6 ± 18.6 cm H₂O, and the limits of agreement were 58.8 cm H₂O and 15.6 cm H₂O (Figure 3).

View larger version (14K): [in this window] [in a new window]   Figure 2.   Relationship between Pnsn and PImax measured at functional residual capacity (PImaxFRC) in boys and girls. The line represents the line of identity.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Difference between Pnsn and PImax measured at functional residual capacity (PImaxFRC) against the mean of these two variables in boys and girls.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main findings of this study performed in healthy children were: (1) without exception, Pn_sn could be measured easily in a large group of unselected children ranging in age from 6 to 17 yr; (2) Pn_sn correlated positively with age, height, and weight in boys, but not in girls; (3) in both sexes, Pn_sn was higher than PI_max; and (4) Pn_sn values were similar to those previously measured in normal adults (11).

Recently, several tests of inspiratory muscle function have been developed on the basis of the sniff maneuver, which is both easy to perform and reproducible (8, 9). In particular, the Pn_sn has the advantage of being entirely noninvasive and has been validated in normal adults and in patients (10, 11, 15). This study shows that Pn_sn can be used in children as well. Indeed, they all performed the maneuver easily and nasal pressure could be measured without difficulty in spite of the small size of their nostrils. The within-session coefficients of variation were higher than those previously reported in adults (16), but this was true both for Pn_sn and for PI_max.

We found that Pn_sn correlated with age, height, and weight in boys, but not in girls. This observation may reflect the greater increase of muscle mass in boys, in particular after puberty. The relationship between inspiratory muscle strength, as measured by PI_max, and anthropometric data in children varies in different studies. In children age 7 to 13 yr, Gaultier and Zinman (4) reported that PI_max correlated with age and height both in boys and girls, whereas Smyth and coworkers (5) found no correlation between PI_max and age, height, or weight in adolescents of both sexes. In children age 7 to 17 yr, Wilson and coworkers (7) reported that PI_max was related to weight in boys and girls. Our data of Pn_sn are similar to those of PI_max reported by Wagener and coworkers (6). In a group of children of similar age, these investigators found that PI_max correlated with age and height in boys, but not in girls. In both sexes, PI_max correlated with arm muscle area, as derived from arm circumference and triceps skinfold thickness.

In accordance with a previous study in adults (11), we found that Pn_sn was higher than PI_max measured at the same lung volume, i.e., FRC. This difference appeared even more clearly as Pn_sn equaled or exceeded PI_max in 152 of 180 children (84%), compared with 107 of 160 adults (67%) (11). As in adults, this difference is likely explained by the ease of the sniff maneuver when compared with the PI_max maneuver. The natural and painless character of the sniff probably allows the subjects to reach a maximal activation of inspiratory muscles more easily. Such higher activation would compensate for the force loss resulting from the dynamic character of the sniff. Another difference between the two maneuvers lies in the degree of recruitment of the different muscles. In adults, at least, the diaphragm is activated more during a sniff than during a PI_max maneuver (17). Finally, the difference observed between Pn_sn and PI_max in the present study cannot be ascribed to low values of PI_max. Indeed, PI_max values were slightly higher than those previously reported in children of similar age (4).

The values of Pn_sn in children were similar to those that we previously measured in healthy adults (11). Thus, mean Pn_sn was 99 and 117 cm H₂O in boys age 6 to 12 yr and 13 to 17 yr, respectively, and 111 cm H₂O in men age 20 to 65 yr. Similarly, mean Pn_sn was 93 cm H₂O in girls age 6 to 16 yr and 87 cm H₂O in women aged 20 to 65 yr. This similarity could relate to properties of maximal respiratory pressures in general, and/or to specific characteristics of Pn_sn. In our study, PI_max increased with age in both sexes. In the adolescent groups, PI_max values were similar to those that we previously measured in adults (11). Thus, mean PI_maxFRC was 107 cm H₂O in boys age 13 to 17 yr and 106 cm H₂O in men age 20 to 65 yr. Similarly, mean PI_maxFRC was 81 cm H₂O in girls age 13 to 16 yr and 83 cm H₂O in women age 20 to 65 yr. In contrast, PE_max was slightly lower in adolescents, in particular in boys, than in previously studied adults (11). This different evolution of PI_max and PE_max according to age has been previously reported by Wagener and coworkers (6). In general, PI_max and PE_max in children have been found either similar to or only slightly lower than in adults (4- 7). This is in contrast with other indices of muscle strength and is explained by the fact that a pressure is the ratio between a force and the surface to which it is applied. If respiratory muscle mass increases in proportion with the surface of the thorax, high respiratory pressures are expected to be generated by children in spite of a lower muscle mass (18). Recent autopsy data confirm that diaphragm mass is linearly related to age and height in childhood (19).

The similarity of Pn_sn between children and adults may also be related to specific characteristics of this method. Pn_sn probably reflects diaphragm strength predominantly, because this muscle is activated more than other inspiratory muscles during the sniff maneuver (17). Infants are able to develop high transdiaphragmatic pressures (Pdi) during inspiratory efforts while crying against an occluded airway. During this maneuver, Pdi increases with age and reaches a plateau of about 85 cm H₂O by the age of 6 mo (20). This value is close to the average Pdi measured during pure inspiratory effort in normal young adults (116 cm H₂O) (21). The diaphragm appears therefore particularly suited to generate high pressures independent of its actual muscle mass. McCool and coworkers (22, 23) recently explored the relationships between diaphragm structure and its pressure-generating capacity. They considered the diaphragm as a piston acting axially in the thoracoabdominal cavity. In this model, the transdiaphragmatic pressure is determined by the following variables: Pdi =  · CSA_di/ A_thor, where  is the tensile stress developed by the contractile elements, CSA_di is the cross-sectional area of the diaphragm, and A_thor is the axially projected area of the diaphragm. McCool and coworkers used ultrasound to measure diaphragm thickness at the upper level of the zone of apposition while the subjects were at FRC. The diaphragm cross-sectional area was calculated as the product of diaphragm thickness and circumference. In adults, either untrained or trained weightlifters, Pdi_max correlated with diaphragm thickness and cross-sectional area, as well as with the ratio CSA_di/A_thor. Studying children age 6 to 12 yr and untrained adults, they found that diaphragm thickness, circumference, and cross-sectional area increased with height and weight. Similarly, the axially projected area of the diaphragm increased with height and weight. As a result, the ratio CSA_di/A_thor remained almost stable among individuals of different sizes. Based on these data, it was calculated that for a given tensile stress, Pdi would increase by only 27% from the smallest child to the heaviest adult in this study (22, 23).

The Pn_sn method is based on the fact that the nasal flow- limiting segment collapses during a sharp sniff (24, 25). As a result, only a small pressure gradient exists between the upper airways beyond the point of collapse and the intrathoracic cavity. Nasal resistance is higher in children and decreases with age (26). It can be hypothesized that the nasal collapse during sniffs is more complete in children, and that Pn_sn reflects intrathoracic pressure even more closely than in adults. However, were this to occur, the possible pressure gain would be small because Pn_sn represents on average 92% of sniff esophageal pressure in adults (10). Thus the higher nasal resistance of children cannot explain by itself the similarity of Pn_sn between children and adults.

Caucasian children only were included in this study because of potential ethnic differences in muscle strength and nasal configuration. The pressure values measured in six children of Asian origin and one of African origin were within the range of values of Caucasian children. The Pn_sn was close to PI_max in the four children of Laotian descent and in the two of Indian descent. However, Pn_sn was markedly lower than PI_max in the only child of African descent. This could be related to a different nasal configuration, as the critical transmural pressure at which the nasal flow-limiting segment collapses has been reported to be higher in subjects of African descent (24).

We conclude that Pn_sn can be easily used to assess inspiratory muscle strength in children age 6 yr and older, and provides higher values than PI_max. Normal values are independent of age in girls, and can be predicted from age by a first-degree equation in boys. Being easy and noninvasive, Pn_sn may prove useful to assess inspiratory muscle strength in children with neuromuscular disorders.