Infant Behavior & Development 34 (2011) 226234

Contents lists available at ScienceDirect

Infant Behavior and Development

Reliability of the M-FLEXTM : Equipment to measure palmar grasp strength in infants

Marcus Vinicius Marques Moraes a,b, , Eloisa Tudella b , Joyce Ribeiro c , Thas Silva Beltrame c , Ruy Jornada Krebs d
a b c d

Department of Physiotherapy Regional University of Blumenau, Brazil Department of Physiotherapy Neuropediatrics Sector So Carlos Federal University, Brazil Department of Health Sciences State University of Santa Catarina, Brazil Department of Physical Education State University of Santa Catarina, Brazil

a r t i c l e

i n f o

a b s t r a c t
The palmar grasp behavior investigation is frequent in evaluation routine of infants. The aim of this study was to determine the reliability of an instrument for measuring palmar grasp strength in infants from birth to four months of age. Thirty-nine healthy infants from both sexes from birth to four months of age were evaluated and divided into three groups (SC, FR, BL) according to the city they came from. The equipment under test is called M-FLEX , which provides maximum grasp strength (FMAX), mean grasp strength (FMEAN) and grasp time (GRT) values. The repeatability and reproducibility of measures produced by the M-FLEXTM were veried. Three measurements were obtained (M1, M2, M3) from both hands, with the infants face turned to the right. To calculate the repeatability, the three measures were compared to themselves in each group. For reproducibility, the difference between the averages of M1 of the three variables provided by M-FLEXTM was veried in the three sampling sites. The statistical tests used were ANOVA, Pearsons correlation and calculation of the intra-class correlation coefcient (ICC) with their respective condence intervals (CI). The M-FLEXTM proved to be an outt that provides measures capable of characterizing the palmar grasp phenomenon in infants. 2010 Elsevier Inc. All rights reserved.

Article history: Received 15 May 2010 Received in revised form 28 September 2010 Accepted 3 December 2010 PsycINFO classication: 2330 Keywords: Palmar grasp reex Motor development Neurological exam

1. Introduction The acquisition of motor skills occurs progressively. They are the product of the interaction between the organization of the nervous and musculoskeletal system and environmental experiences and the demands of tasks to which infants are exposed (Thelen, Kelso, & Skala, 1987; Thelen, Corbetta, & Spencer, 1996; Thelen, Fisher, & Ridley-Johnson, 2002). The abilities to reach and grasp objects present in the environment developed during the rst year of life from newborns, whose hands move towards the object, to the action of touch and grasp around the 4th month of life, and then by grasping small objects using the ngertip with precision (Carvalho, Tudella, & Savelsberg, 2007; Carvalho, Tudella, Caljouw, & Savelsberg, 2008; Von Hofsten & Rnnqvist, 1988; Von Hofsten, 1982). In infants, the rst manifestation of hand function is through the palmar grasp behavior, which is strongly present during the rst month of life and occurs with exion of the ngers in response to placing an object in his palm. Pathological

Corresponding author at: Department of Physical Therapy, Regional University of Blumenau, Room J-105, Antonio da Veiga Street, 89012-900 Blumenau, SC, Brazil. Tel.: +55 4732322952. E-mail address: moraes.mvm@gmail.com (M.V.M. Moraes). 0163-6383/$ see front matter 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.infbeh.2010.12.004

M.V.M. Moraes et al. / Infant Behavior & Development 34 (2011) 226234

227

conditions affect the development of voluntary catch and release and impair the tactile sensory input (Geerts, Einspieler, Dibiasi, Garzarolli, & Bos, 2003; Von Hofsten & Rnnqvist, 1988; Zafeiriou, 2004; Zafeiriou, Tikoulas, & Kremenopoulos, 1995). The palmar grasp behavior evaluation is part of several neurological examination routines of infants (Glick, 2005; Mercuri, Ricci, Pane, & Baranello, 2005; Romeo et al., 2007). Different authors have developed methodologies to investigate this behavior, among which the primitive reex prole (Capute et al., 1978; Capute et al., 1984), the ohmmeterammeter (Tan & Tan, 1999; Tan & Zor, 1994; Tan, rs, & Kutlu, 1992; Tan et al., 1993), ammeterpolygraph (Rochat, 1987) and pressure sensorvoltmeter (Molina & Jouen, 1998). The function of the primitive reex prole is to subjectively quantify the intensity of several primitive reexes, including the palmar grasp reex. In the evaluation of the palmar grasp reex, the examiner is to assign values from 0 to 4 to the response given by the infant before the stimulus of putting his nger on the base of the infants third, fourth and fth ngers. Value zero means absence of grasp, value 1 indicates minimal change of tonus, value 2 indicates grasp physically present and visible, value 3 indicates remarkable grasp strength and value 4 indicates a strong grasp that prevents the typical movement of the upper limb in the healthy infant (Capute, 1986; Capute et al., 1982; Payne & Isaacs, 2007). In the late 80s, ner Tan and his colleagues showed to the scientic community a series of studies on the behavior of the palmar grasp reex. The studies had in common the verication of the intensity of this reex, measured with an ohmmeter and an ammeter (Tan, 2002; Tan & Tan, 2001). The ohmmeter is a device that measures the resistance of an electrical conductor and the ammeter measures the electrical current running through the branch of an electrical circuit (Creder, 2002). Thus, Tan and his colleagues measured the palmar grasp reex with a device composed of a small balloon attached to a piston-ring, an injector connected to an ohmmeter and ammeter. This small balloon was placed in contact with the palmar surface of the infants hand. At the moment of the response of nger exion, the resistance measured by the ohmmeter decreased and the ammeter recorded an increase in the electric current. The palmar grasp reex strength was not expressed in units of physical quantities (Tan, 1994; Tan & Tan, 1999, 2001). Philippe Rochat conducted a study on the palmar grasp behavior and suction (Rochat, 1987, 1993). The author observed variations of positive pressure applied to two different objects connected to an air pressure transducer. Objects that touched the infants palmar surface had identical cylindrical shapes and were distinguished only by their consistency. One object was rigid (hard object) and the other elastic (soft object), both covered with a thin rubber lm that was sealed to the pressure transducer tubing. The external pressure exerted by the infant on the cylindrical object caused pressure variations on the air layer between the rubber lm and its hard or pliable core. The pressure variations were recorded by the polygraph (Rochat, 1987). Molina and Jouen (1998) concluded that the different textures of objects that stimulate the palmar surface of infants are able to modulate the palmar grasp strength. To achieve this result, they used a device to analyze the palmar grasp strength variation. This equipment consisted of objects of different textures linked to a pipette connected to a pressure sensor and a voltmeter. The different responses of infants induced by the texture of objects stimulated the pressure sensor that generated signals which were recorded and processed on a personal computer. The authors expressed the results of data obtained by this equipment in the unit of physical quantity of electric potential difference Volt. Thus, the authors observed different patterns in the curves drawn by the data recorded on the PC and associated them to the features of the texture of objects (Molina & Jouen, 2001, 2003, 2004). It was observed that from studies previously mentioned, there is a gradual increase in the technological complexity and specicity of equipments used to check the palmar grasp strength in human newborns. However, such equipments do not express the physical quantities or values that characterize the typical behavior (physical units that denote strength) of the population investigated, which makes reproducibility and repeatability studies difcult, since these studies do not present tables with values that characterize such behavior. Through the importance of studying the palmar grasp strength, both for the detection of developmental changes, as to study the tactile perception of infants to different objects; the present study aims to determine the reliability of an instrument to measure the palmar grasp strength in infants. This instrument, named M-FLEXTM , produces reliable measurements and details of the palmar grasp strength behavior as numerical and graphical representation of the maximum grasp strength, mean grasp strength and grasp time. The device was developed and manufactured by CSE Service Company in the city of Blumenau, SC, Brazil. It has the following characteristics: lightweight, portability, domestic technology, allowing capture of data in different contexts (such as in use at home and at basic health units) and assigning physical units (mmHg and g/cm2 ) to the phenomenon measured.

2. Methods 2.1. Study participants Thirty-nine full-term healthy infants from both sexes from birth to four months of age were evaluated and divided into three groups (SC, FR, BL) according to the city they came from.

228

M.V.M. Moraes et al. / Infant Behavior & Development 34 (2011) 226234

Fig. 1. (a) M-FLEXTM , equipment in details and (b) rubber balloon (cuff).

No signicant differences (one-way ANOVA) were found between groups for gestational age F(2,46) = 2.87 and (p = 0.078), weight at the time of assessment F(2,46) = 3.02 and (p = 0.09) and between age in days F(2,46) = 1.97 and (p = 0.150). 2.1.1. City SC This city belongs to the southeastern region of Brazil. It has a population of 220,425 inhabitants, per capita GDP of R $ 16,441.00 and HDI = 0.841. The climate is altitude temperate with mean annual temperature of 20 C (Brasil, 2007). The group of infants evaluated in the city SC was composed of 13 infants, six male and seven female, with gestational age of 38.8 (0.95) weeks, weighing 3058.77 g (268.7) at birth and measuring 48.78 cm (2.24). On the day of testing, the infants had a mean weight of 4953.09 g (823.49) and measured 56.82 cm ( 3.98). Their ages ranged from 9 to 120 days with mean age of 64.59 ( 21.4) days. 2.1.2. City FR This city belongs to the southern region of Brazil and has a population of 36,469 inhabitants, per capita GDP of R $ 13,249.00 and HDI = 0.779. The climate is altitude temperate with mean annual temperature of 16.1 C (Brasil, 2007). The group of infants evaluated in the city FR was composed of 13 infants, seven male and six female, with gestational age of 38.3 (0.94) weeks. Their birth weight was 3061.98 g (278.5) and mean length of 46.98 cm (2.21). On the day of testing, the infants had mean weight of 4551.2 g (1066.6) and 55.31 cm (3.41) in length. Their ages ranged from 10 to 120 days with mean age of 55.09 (31.47) days. 2.1.3. City BL This city belongs to the southern region of Brazil and has a population of 299,416 inhabitants, per capita GDP of R $ 22,809.00 and HDI = 0.855. The climate is warm temperate with mean annual temperature of 21 C (Brasil, 2007). The group of infants evaluated in the city BL was composed of 13 infants, seven males and six females, with gestational age of 39.75 (0.84) weeks. Their birth weight was 3452.83 g (468.8) and mean length of 49.09 cm (2.09). On the day of testing, the infants had mean weight of 4439.47 g (580.16) and length 56.82 cm (3.98). Their ages ranged from 5 to 120 days, with mean age of 47.40 (33.08) days. The sample size was obtained from the calculation of proposals submitted by Bussab and Moretin (2002). Standard deviation of 2.28 g/cm2 , expected sampling error of 2 g/cm2 and a signicance level of 5% were considered, resulting in a sample of 13 infants per group. The standard deviation value assumed was adopted from a pilot study. 2.2. The M-FLEXTM equipment The M-FLEXTM , see Fig. 1, is an instrument that records the palmar grasp strength applied on the rubber balloon (cuff). Both numerical and in the graphic form, they express the time that the cuff was pressed, the maximum pressure and the mean pressure. The cuff has an outer diameter of 11 mm and internal diameter of 8 mm. Its wall has hardness of 40 Shore. The thickness of the tube that connects the cuff to the device is 1.25 mm. The equipment has self-calibration system that res immediately when the examiner presses the start key. This selfcalibration system is based on the reading of the pressure in the system (cuff + tube that connects the cuff to the equipment)

M.V.M. Moraes et al. / Infant Behavior & Development 34 (2011) 226234

229

Fig. 2. Flowchart for the repeatability and reproducibility analysis.

at the time and interprets it as zero. From the rst pressure on the cuff, the data are expressed in mmHg (millimeters of mercury) or g/cm2 (g/squared cm). Such self-calibration procedure allows for a mean square error of 1%, according to the manufacturer. The device has the ability to store the infants names and to hold up to 32 readings. These readings can then be transferred to a computer (le in dat or txt format) to be treated by the statistical package. Additionally, an 8-bit micro controller, 128 64 points graphic liquid crystal display, an analog/digital converter (A/D) with resolution of 8 bits and a ash memory were used to store the readings. Also, a pressure sensor and an instrumentation amplier that condition the electrical signal of the sensor to the A/D converter were used. The equipment has been added to a keyboard with 5 keys and a USB interface that enables communication of the equipment with the personal computer (Moraes et al., 2004; Moraes, Krebs, Martins, Possamai, & Todorov, 2005). The equipment is supplied with mains electricity and/or charger and batteries. The physical characteristics of the equipment provide the completion of data collection in different environments. This enables the investigation of the infants interaction with the environment without losing the ecological validity. 2.3. Repeatability and reproducibility of the M-FLEXTM measurements The instrument reliability or testretest reliability assesses the ability of the equipment to measure or test phenomena and provide measures or results with the same consistency. This test is performed by taking measurements of the same phenomenon several times (Baker, Cook, & Redfern, 2009; Lau, Chiu, & Lam, 2009). The intra-observer reliability is given by the repeatability of measurements of a phenomenon at different times. This indicates the ability of observers to measure the same phenomenon at different times (Amiri, Jull, & Bullock-Saxton, 2003; Gadotti, Vieira, & Magee, 2006; McEwan, Herrington, & Thom, 2007). The intra-subject reliability refers to the performance of the subject examined. It compares the extent of it, taking two or more times by the same examiner (Gadotti et al., 2006; Wolf et al., 2004). Reproducibility (of measurement results) is the concordance level between the measurement results of the same measured thing (term used by Taylor and Kuyatt (1994) to describe an object or an event measured), performed under different measurement conditions. For a reproducibility expression to be valid, the conditions changed must be specied. Such conditions may include the measurement principle, measurement method, observer, measuring instrument, reference standard, location, conditions of use and time (Caruana, Bradbury, & Adam, 2005; Taylor & Kuyatt, 1994). In this study, the repeatability was veried from the analysis of three measurements (M1, M2 and M3) performed in the same infant within the same group. The reproducibility was veried through the comparison of the rst measures (M1) of each infant between groups. The analysis ow can be seen below, see Fig. 2. 2.4. Statistical planning The variables analyzed were: (a) Maximum grasp strength (FMAX): given by the highest value measured in each test originated by the pressure peak exerted by the infant on the M-FLEXTM cuff. (b) Mean grasp strength (FMEAN): given by the value of the average pressure on the M-FLEXTM cuff in a given time. (c) Grasp time (GRT), given by the value of the time interval, in milliseconds, elapsed from the rst reading of the pressure on the M-FLEXTM cuff until the pressure reaches zero. The measures and the instrument were considered reliable because no signicant differences (p > 0.05) between the three groups of measures were observed.

230

M.V.M. Moraes et al. / Infant Behavior & Development 34 (2011) 226234

(a) Removal of ve extreme outlier measures for each collection site, making the data more homogeneous. These measures were discrepant due to some technical problem in the collection, for example, a sudden pull on the tube or the silicone cuff of the M-FLEXTM . The removal of outliers excluded measures, but did not exclude the infants. (b) Logarithmic normalization (log 10) for variables FMAX, FMEAN and GRT. (c) Application of the Students t test for unpaired samples to compare the variables in relation to the right and left hand. (d) Analysis of variance (ANOVA) with two factors (hand and site), where factor 1 is the comparison between right and left hand and factor 2 is the site where the data were collected. This procedure was performed for FMAX, FMEAN and GRT. (e) Analysis of variance with repeated measures to compare groups of measurements made by the same observer. (f) Analysis of variance and test of medians with repeated measures to compare the M1 measures in the groups of infants. (g) Calculation of the two-way random type intra-class correlation coefcients (ICC) for consistency (rater tested are part of a population of raters) and their respective condence intervals (CI) (Barnhart, Song, & Haber, 2005; McGraw & Wong, 1996; Shrout & Fleiss, 1979; Tinsley & Weiss, 1975). The intra-class correlation coefcients, according to Weir (2005), can range from 0, when there is no reliability to 1, when the reliability is perfect. 2.5. General procedures This work was approved in Ethical in Human Research Committee of the Regional University of Blumenau on number 47/04. The teams responsible for the collection were trained so that the procedure was repeated with the same accuracy in all database collections. To facilitate the ecological insertion of researches, data were collected in environments that were familiar to mothers, for example, the basic health units and in some cases the actual residence of infants. It was emphasized that the test sites should be in suitable environmental conditions and with limited number of people, only the researcher, the researchers assistant and the infants mother. At the time of the evaluation, a data collection card was lled out with data such as anthropometric variables of the infant, pre, peri and post-natal information and maternal habits. The Free and Clear Consent Term was also presented. During the entire test with the M-FLEXTM equipment, the infants should remain in the inactive or active behavioral state of alert (Prechtl & Beintema, 1964). If the infant would cry, the test was suspended and only the measures already collected were used. 2.6. Test conditions For the present study, three measures taken for each hand were proposed, when the infants face was turned to the right. This position was dened because it was the one that had the largest number of measures, 171 (M1 = 66, M2 = 60 and M3 = 45). There was no containment of the infants head by the researcher. The infant was positioned in dorsal decubitus position, the upper limbs with elbows bent and the forearm in supine position. The examiner gently stimulated the infants hand opening and then put the rubber transducer at the base of the last 3 ngers, not allowing the sliding of the transducer over the described region (Tudella, 1996; Tudella, Oishi, & Bergamasco, 2000). The examiner allowed the infant to press the transducer to the maximum time of 30 s. The transducer would be considered loose when the M-FLEXTM display scored zero. The measurement would then be resumed. The time spent for this step would be approximately 5 min, 2.5 min for each hand.
3. Results 3.1. Maximum grasp strength (FMAX) The FMAX data are presented in Table 1. These data correspond to the mean values and standard deviations for each measure (M1, M2 and M3) for each city and their p values for comparisons within and between cities. For the FMAX, no signicant difference was observed between measures obtained for city SC F(2,44) = 2.46 and p = 0.10 with ICC = 0.78. There was no signicant difference F(2,56) = 1.06 and p = 0.90 with ICC = 0.79 between measures for city FR and no signicant difference between the measures for city BL F(2,50) = 1.09 and p = 0.91 with ICC = 0.68. The comparison made between M1 measures of the three cities was obtained with better representation when the median was used and the p-value comparing the three groups X2 (2.47) = 5.21 and p = 0.074 was obtained from the median test for three samples (Campos, 2001). The comparison between the three cities showed ICC = 0.9. The interaction factor between right hand/left hand with the collection sites for M1 measure of the FMAX showed no signicant difference F(2,49) = 0.412 and p = 066. Likewise, the interaction factor between male/female with the collection sites showed no signicant differences either F(2,49) = 3.01 and p = 0.59. 3.2. Grasp time (GRT) Table 2 shows the values for the GRT variable. No signicant differences between measures (M1, M2 and M3) for cities SC F(2,44) = 0.46 and p = 0.63, FR F(2,56) = 0.12 and p = 0.88 and BL F(2,50) = 0.14 and p = 0.87 were observed. The intra-class correlation coefcients were low for a condence interval of 95%, SC (ICC = 0.09), FR (ICC = 0.11) and BL (ICC = 0.04). A comparison of the M1 measures between infants from different cities also showed no signicant difference F(2,46) = 2.01 and p = 0.145. The interaction factor between right hand/left hand with the collection sites for M1 measure of the GRT showed no signicant difference F(2,49) = 0.93 and p = 0.4. Likewise, the interaction factor between male/female with the collection sites showed no signicant differences either F(2,49) = 1.63 and p = 0.20. 3.3. Mean grasp strength (FMEAN) Table 3 showed no statistically signicant differences for any of the three cities in the intra-group analysis of FMEAN (M1, M2 and M3), with SC F(2,44) = 0.48 and p = 0.62, FR F(2,56) = 0.28 and p = 0.75 and BL F(2,50) = 0.13 and p = 0.88. In the analysis between groups, no signicant differ-

M.V.M. Moraes et al. / Infant Behavior & Development 34 (2011) 226234 Table 1 Comparison of statistical values for FMAX in gram per squared centimeters (g/cm2 ) of measures taken intra group and between groups. City SC Measure M1 M2 M3 M1 M2 M3 M1 M2 M3 Mean (SD) in g/cm2 5.7 (5.8) 7.5 (5.3) 7.7 (3.5) 6.6 (5.8) 5.9 (4.1) 5.5 (5.2) 2.7 (1.7) 3.4 (1.7) 2.7 (1.1) p-Value 0.074 F p-Value ICC (95% CI) intra-observer

231

p-Value

F(2,44) = 2.46

0.10

0.78 (0.53 to 0.92)

0.00

FR

F(2,56) = 1.06

0.90

0.79 (0.56 to 0.92)

0.00

BL

F(2,50) = 1.09

0.91 ICC (95% CI) intra-observer

0.68 (0.80 to 0.97) p-Value 0.00

0.00

Difference between groups for M1 X2 = 5.210

0.9 (0.80 to 0.95)

Table 2 Comparison of the statistical values for GRT in milliseconds (ms) of measures taken intra-group and between groups. City SC Measure M1 M2 M3 M1 M2 M3 M1 M2 M3 Mean (SD) in ms 2600 (2159) 3039 (2450) 2872 (2502) 2831 (3124) 2100 (2707) 2409 (3054) 2884 (3133) 2583 (2751) 2282 (2168) p-Value 0.145 p-Value ICC (95% CI) intra-observer p-Value

0.63

0.09 (0.17 to 0.46)

0.26

FR

0.75

0.11 (0.18 to 0.51)

0.24

BL

0.87

0.44 (0.22 to 0.44) ICC (95% CI) intra-observer 0.14 (0.11 to 0.47)

0.38 p-Value 0.148

Difference between groups for M1

ences F(2,46) = 0.151 and p = 0.86 were found either. The ICC were low, SC (ICC = 0.23), FR (ICC = 0.18) and BL (ICC = 0.03) for a condence interval of 95%. The interaction factor between right hand/left hand with the collection sites for M1 measure of the GRT showed no signicant difference F(2,49) = 0.87 and p = 0.42. Likewise, the interaction factor between male/female with the collection sites showed no signicant differences either F(2,49) = 1.43 and p = 0.24.

4. Discussion This study aimed to determine the reliability of an instrument for measuring palmar grasp strength in infants from birth to four months of age. The maximum grasp strength (FMAX) values generated by M-FLEXTM showed no signicant differences in the intra-group analysis. This means that the equipment enabled repeatability with similar measures. The equipment has also proved to be
Table 3 Comparison of the statistical values for FMEAN in grams per squared centimeters (g/cm2 ) of measures taken intra-group and between groups. City SC Measure M1 M2 M3 M1 M2 M3 M1 M2 M3 Mean (SD) (g/cm2 ) 1.4 (1.5) 2.2 (2.4) 1.8 (1.3) 1.8 (1.7) 1.1 (1.1) 0.6 (2.2) 0.8 (0.6) 1.0 (1.3) 0.7 (0.5) p-Value 0.805 p-Value ICC (95% CI) intra-observer p-Value

0.62

0.23 (0.99 to 0.63)

0.091

FR

0.75

0.18 (0.13 to 0.57)

0.139

BL

0.88

0.33 (0.23 to 0.43) ICC (95% CI) intra-observer 0.05 (0.19 to 0.38)

0.405 p-Value 0.341

Difference between groups for M1

232

M.V.M. Moraes et al. / Infant Behavior & Development 34 (2011) 226234

reliable for reproducibility studies, since the measures compared between the three cities showed no difference. The FMAX, measured by M-FLEXTM represents the quantication of the palmar grasp intensity. The values presented in the results may serve as an auxiliary tool for the manufacture of clinical and functional diagnoses. The raw scores give typicity parameters of infants to professionals during their assessment or during the evaluation of the effectiveness of their interventions. The grasp time (GRT) can be understood by analogy as the time the infant takes to modulate the palmar grasp behavior. This modulation process depends on factors such as maturation (Herschkowitz, 2000), environment and task (Molina & Jouen, 1998; Rochat, 1987; Thelen et al., 1996, 2002). Previous studies have refereed it to the intensity of strength infants hold the object that touched their palmar face (Capute & Accardo, 1996; Capute et al., 1984; Edwards, Buckland, & Powlen, 2002), however, there is no mention as to the time the infant is still doing the grasping, which measure is provided by the M-FLEXTM . The GRT measurements showed to be reliable as to the repeatability of the test and its reproducibility with different samples, since the ANOVA for repeated measures showed no signicant differences between them. However, it was observed that the intra-class correlation coefcients for this variable were low. This occurs when there is a high intrasubject and a low inter-subject variability. According to Weir (2005), this is a situation that impacts negatively on the ICC values. The mean grasp strength (FMEAN) was judged reliable for repeatability and reproducibility of the test with the M-FLEXTM ; however, it showed the same phenomenon of low intra-class correlation coefcients because FMEAN is a mathematical function of FMAX in relation to the GRT. As time showed a high variability, the ICC values of FMEAN were also low. The neurological examination and the scales for the infants development assessment evaluate the palmar grasp behavior and test the symmetry of the grasping strength exerted on the examiners nger (Mercuri et al., 2005). Lack of grasp, poor grasp, exaggerated grasp or persistence beyond six months of age are undesirable and may indicate developmental disorder (Brazelton & Nugent, 1984). Studies have shown that some degree of asymmetry in motor behavior of infants is part of normal development (Grattan, De Vos, Levy, & McClintock, 1992; Rnnqvist, 1995; Rnnqvist & Hopkins, 1998). Other studies that describe test routines of infants have shown that responses symmetrical to tests are expected in a typical infant (Dubowitz, Mercuri, & Dubowitz, 1998; Herschkowitz, 2000). Thus, the M-FLEXTM is a tool that will aid studies that seek how much asymmetry is typical and how long it can remain without characterizing developmental disorders. The palmar grasp examination, in clinical practice and in child care evaluations, is often done using the examiners nger as stimulator of the infants palm. The stimulus is not controlled, there is variation in temperature, thickness and consistency. The M-FLEXTM standardizes the stimulus that is given in the infants palm, since studies have shown that the response to the stimulus in the palm of his hand can be modulated according to the characteristics of the object (Molina & Jouen, 2001, 2003, 2004). As the silicon transducer of the M-FLEXTM equipment is manufactured from a mold made in a precision ofce, all parts have the same size, strength and hardness. Researchers have always demonstrated the strength that the infant presses the stimulus object. This can be observed both in studies using the primitive reex prole (Capute et al., 1984; Capute et al., 1982) as instrument as in studies using ammeter, ohmmeter, with or without polygraph (Molina & Jouen, 2004; Rochat, 1987; Tan & Tan, 2001). A technological evolution reected in the quality of measures is observed; however, none of them expresses such measures in a scale of physical quantities. Thus, the M-FLEXTM follows the historical ow of researches on palmar grasp in infants, presenting consistent measures expressed in units of physical quantity of pressure (gf/cm2 ). A parameter never published in studies conducted to date, is the grasp time expressed in milliseconds (ms). The reliable measures obtained will allow future studies to perform analysis of continuous variables of the palmar grasp behavior, scope and laterality in infants. 5. Study limitations The inability to specify the pressure area of the palm on the M-FLEXTM cuff prevents the calculation of the strength in Newtons (N) held by the infant. 6. Conclusions The M-FLEXTM is a tool that standardizes the stimulus and records the infants motor response efciently and reliably, assigning module to the intensity and duration of the grasp. The M-FLEXTM proved to be equipment that produces reliable measures (especially the FMAX measures) of the grasp behavior in the palmar grasp study in infants aged between birth and four months of age. The consistency between maximum grasp strength, mean grasp strength and grasp time assigns the M-FLEXTM the condition of equipment that generates valid measures for the palmar grasp strength assessment. The small economic and socio-cultural differences, climate variations and maternal education aimed at care of infants were not factors strong enough to interfere with their palmar grasping strength. Due to being function of FMAX and GRT, the FMEAN showed great variability between individuals that, as the grasp time, can express a developmental characteristic of typical infants.

M.V.M. Moraes et al. / Infant Behavior & Development 34 (2011) 226234

233

References
Amiri, M., Jull, G., & Bullock-Saxton, J. (2003). Measuring range of active cervical rotation in a position of full head exion using the 3D Fastrak measurement system: An intra-tester reliability study. Manual Therapy, 8(3), 176179. Baker, N. A., Cook, J. R., & Redfern, M. S. (2009). Rater reliability and concurrent validity of the keyboard personal computer style instrument (K-PeCS). Applied Ergonomics, 40, 136144. Barnhart, H., Song, J., & Haber, M. J. (2005). Assessing intra, inter and total agreement with replicated readings. Statistics in Medicine, 24, 13711384. Brasil (2007). IBGE Cidades (Publication), retrieved 27/08/2010, from Ministrio do Planejamento, Orcamento e gesto: http://www.ibge.gov.br/ cidadesat/default2.php. Brazelton, T. B., & Nugent, J. K. (1984). Neonatal behavioral assessment scale (3rd ed.). London: Cambridge University Press. Bussab, O. W., & Moretin, P. A. (2002). Estatstica bsica. So Paulo: Saraiva. Campos, G. M. (Ed.). (2001). Estatstica prtica para docentes e ps-graduandos. So Paulo: USP. Capute, A. J. (1986). Early neuromotor reexes in infancy. Pediatrics Annals, 15(3), 217226. Capute, A. J., & Accardo, P. J. (1996). The infant neurodevelopmental assessment: A clinical interpretive manual for CAT-CLAMS in the rst two years of life, part 2. Current Problems in Pediatrics, 26(8), 279306. Capute, A. J., Accardo, P. J., Vinning, E. P. G., Rubenstein, J. E., Walcher, J. R., & Ross, A. (1978). Primitive reex prole: A pilot study. Physicaltherapy, 58(9), 10611065. Capute, A. J., Palmer, F. B., Shapiro, B. K., Wachtel, R. C., Ross, A., & Accardo, P. J. (1984). Primitive reex prole: A quantitation of primitive reexes in infancy. Developmental Medicine & Child Neurology, 26, 375383. Capute, A. J., Shapiro, B. K., Accardo, P. J., Wachtel, R. C., Ross, A., & Palmer, F. B. (1982). Motor function: Associated primitive reex proles. Developmental Medicine & Child Neurology, 24, 662669. Caruana, M. F., Bradbury, A. W., & Adam, D. J. (2005). The validity, reliability, reproducibility and extended utility of ankle to brachial pressure index in current vascular surgical practice. European Journal of Vascular and Endovascular Surgery, 29, 443451. Carvalho, R. P., Tudella, E., Caljouw, S. R., & Savelsberg, G. J. P. (2008). Early control of reaching: Effects of experience and body orientation. Infant Behavior & Development, 31, 2333. Carvalho, R. P., Tudella, E., & Savelsberg, G. J. P. (2007). Spatio-temporal parameters in infants reaching movements are inuenced by body orientation. Infant Behavior & Development, 30, 2635. Creder, H. (2002). Instalaces eltricas (14th ed.). Rio de Janeiro: Livros Tcnicos e Cientcos. Dubowitz, L., Mercuri, E., & Dubowitz, V. (1998). An optimality score for the neurologic examination of the term newborn. Journal of Pediatrics, 133(3), 406416. Edwards, S. J., Buckland, D. J., & Powlen, M. (2002). Development & functional hand grasps. Thorofare: SLACK Incorporated. Gadotti, I. C., Vieira, E. R., & Magee, D. J. (2006). Importance and clarication of measurement properties in rehabilitation. Revista Brasileira de Fisioterapia, 10(2), 137146. Geerts, W. K., Einspieler, C., Dibiasi, J., Garzarolli, B., & Bos, A. F. (2003). Development of manipulative hand movements during the second year of life. Early Human Development, 75, 91103. Glick, T. H. (2005). Toward a more efcient and effective neurologic examination for the 21st century. European Journal of Neurology, 12, 994997. Grattan, M. P., De Vos, E., Levy, J., & McClintock. (1992). Asymmetric action in the human newborn: Sex differences in patterns of organization. Child Development, 63, 273289. Herschkowitz, N. (2000). Neurobiological bases of behavioral development in infancy. Brain & Development, 22, 411416. Lau, H. M. C., Chiu, T. T. W., & Lam, T. H. (2009). A clinical measurement of craniovertebral angle by electronic head posture instrument: A test of reliability and validity. Manual Therapy, 14(4), 363368. McEwan, I., Herrington, L., & Thom, J. (2007). The validity of clinical measures of patella position. Manual Therapy, 12, 226230. McGraw, K. O., & Wong, S. P. (1996). Forming inferences about some intraclass correlation coefcients. Psychological Methods, 1(1), 3046. Mercuri, E., Ricci, D., Pane, M., & Baranello, G. (2005). The neurological examination of the newborn baby. Early Human Development, 81, 947956. Molina, M., & Jouen, F. (1998). Modulation of the palmar grasp behavior in neonates according to texture property. Infant behavior & Development, 21(4), 659667. Molina, M., & Jouen, F. (2001). Modulation of manual activity by vision in human newborns. Developmental Psychobiology, 38, 123132. Molina, M., & Jouen, F. (2003). Haptic intramodal comparison of texture in human neonates. Developmental Psychobiology, 42, 378385. Molina, M., & Jouen, F. (2004). Manual cyclical activity as an exploratory tool in neonates. Infant Behavior & Development, 27, 4253. Moraes, M. V. M., Krebs, R. J., Martins, S. M., Possamai, G. F., & Todorov, L. B. (2005). A relaco da forca do reexo de preenso palmar com as dimenses da mo II Cong. Intern. de Pedagogia do Esporte XVIII Semana da Educaco Fsica da Universidade Estadual de Maring-PR-ANAIS, (pp. 153161). Moraes, M. V. M., Krebs, R. J., & Todorov, L. B. (2004). A medida do reexo de preenso palmar como varivel do desenvolvimento motor. In Paper presented at the III Congresso Latino-Americano de Educaco Fsica; III Congresso Cientco-Latino Americano de Educaco Fsica, Piracicaba-Brasil Payne, V. G., & Isaacs, L. D. (2007). Desenvolvimento motor humano: Uma abordagem vitalcia (6th ed.). Rio de Janeiro: Guanabara Koogan. Prechtl, H. F. R., & Beintema, D. J. (1964). The neurological examination of the full-term newborn infant. Clinics in Developmental Medicine, 12, 173. Rochat, P. (1987). Mouthing and grasping in neonates: Evidence for early detection or soft substances afford for action. Infant Behavior & Development, 10, 435449. Rochat, P. (1993). Hand-mouth coordination in the newborn: Morphology, determinants, and early development of a basic act. In G. J. P. Savelsbergh (Ed.), The development of coordination in Infancy (pp. 265289). Elsevier Science Publisher. Romeo, D. M. M., Cioni, M., Guzzetta, A., Scoto, M., Conversano, M., Palermo, F., et al. (2007). Application of a scorable neurological examination to near-term Infants: Longitudinal data. Neuropediatrics, 38, 233238. Rnnqvist, L. (1995). A critical examination of the moro response in newborn infants-symmetry, state relation, underlying mechanisms. Neuropsychologia, 33(6), 713726. Rnnqvist, L., & Hopkins, B. (1998). Head position preference in humans newborn: A new look. Child Development, 69(1), 1323. Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin, 86(2), 420428. Tan, U. (1994). The grasp reex from the right and left hand in human neonates indicates that the development of both cerebral hemisfheres in males, but only the right hemisphere in females, is favoured by testosterone. International Journal of Neuroscience, 16, 3947. Tan, U. (2002). Grasp-reex in human neonates: Distribution, sex difference, familial sinistrality, and testosterone. In M. K. Mandal, M. B. Bulman-Fleming, & G. Tiwari (Eds.), Side bias: A neuropsychological perspective (p. 350). Dordrecht: Kluwer Academic Publishers. Tan, U., rs, R., & Kutlu, N. (1992). Lateralization of the grasp reex in male and female human newborns. International Journal of Neuroscience, 62, 155163. Tan, U., & Tan, M. (1999). Assymmetries of the palmar grasp reex in neonates and hand preferences in adults. NeuroReport, 10, 32533256. Tan, U., & Tan, M. (2001). Testosterone and grasp-reex differences in human neonates. Laterality, 6(2), 181192. Tan, U., & Zor, N. (1994). Grasp-reex strength from right and left hand inrelation to serum cortisol level and fetal position in human neonates. International Journal of Neuroscience, 74, 2732. Tan, U., Zor, N., Kckzkan, T., Ackay, F., Yigitoglu, R., Bakan, E., et al. (1993). Grasp-reex strength from right and left hands associated with pH stressor from the umbilical arterial blood in human newborns: Handednes and sex-related differences. International Journal of Neuroscience, 72, 149156. Taylor, B. N., & Kuyatt, C. E. (1994). NIST Tech. Note 1297 (Washington, DC: US GPO). Thelen, E., Corbetta, D., & Spencer, J. P. (1996). Development of reaching during the rst year: Role of movement speed. Journal of Experimental Psychology Human Perception and Performance, 22, 10591076.

234

M.V.M. Moraes et al. / Infant Behavior & Development 34 (2011) 226234

Thelen, E., Fisher, D., & Ridley-Johnson, J. (2002). The relationship between physical growth and a newborn reex. Infant Behavior and Development, 25(1), 7285. Thelen, E., Kelso, J. A. S., & Skala, K. D. (1987). The dynamic nature of early coordination: Evidence from bilateral leg movements in young infants. Developmental Psychology, 23(2), 178186. Tinsley, H. E. A., & Weiss, D. J. (1975). Research methodology. Journal of Counseling Psychology, 22(4), 358376. Tudella, E. (1996). Contato das Mos com as Regies Oral e Perioral em Recm-Nascidos: O Papel da Estimulaco Tato-Bucal, Tato-Manual e Oro-Gustativa. Unpublished Doutorado. So Paulo: Universidade de So Paulo-USP. Tudella, E., Oishi, J., & Bergamasco, N. H. P. (2000). The effect of oral-gustatory, tactile-bucal and tactile-manual stimulation on the behavior of the hands in newborns. Developmental Psychobiology, 37(2), 8289. Von Hofsten, C. (1982). Eye-hand coordination in the newborn. Developmental Psychobiology, 18, 450461. Von Hofsten, C., & Rnnqvist, L. (1988). Preparation for grasping an object: A developmental study. Journal of Experimental Psychology Human Perception and Performance, 14(4), 610621. Weir, J. P. (2005). Quantifying testretest reliability using the intraclass correlation coefcient and the SEM. Journal of Strength and Conditioning Research, 19(1), 231240. Wolf, S. L., Butler, A. J., Campana, G. I., Parris, T. A., Struys, D. M., Weinstein, S. R., et al. (2004). Intra-subject reliability of parameters contributing to maps generated by transcranial magnetic stimulation in able-bodied adults. Clinical Neurophysiology, 115, 17401747. Zafeiriou, D. I. (2004). Primitive reexes and postural reactions in the neurodevelopmental examination. Pediatrics Neurology, 31, 18. Zafeiriou, D. I., Tikoulas, I. G., & Kremenopoulos, I. (1995). Prospective follow-up of primitive reex proles in high-risk infants: Clues to an early diagnosis of cerebral palsy. Pediatric Neurology, 13(2), 148152.