Symposium
Chair: Josephine V. Brown
Discussant: Janet A. DiúPietro
This symposium is proposed in celebration of the scientific career ofAnneliese Korner who has contributed to our knowledge of the newborn formore than 35 years. Since the early 1960s, Korner has increased ourunderstanding in a number of important areas, such as stable individualdifferences, the importance of vestibular-proprioceptive stimulation,and the assessment of the relative maturity of the preterm. Herfindings in these areas have both theoretical importance and practicalimplications. Although a number of topics could have been chosen incelebration of her contribution to the field, we chose to focus onrecent research conducted with the NAPI (The Neurobehavioral Assessmentof the Preterm Infant) for a variety of reasons. First, the NAPI isunique because of its approach to infant assessment, its conceptualframework, and its psychometric properties. Further, because of itsrecent development, few published studies using the NAPI exist (Espy etal., 1997; Brown et al., 1998; Anand et al., 1999), yet a growing numberof investigators in the US and abroad are beginning to use the NAPI intheir studies. Finally, the NAPI is the culmination of Korner'sscientific career and will be used to answer potentially importantquestions for years to come. The impetus for the development of the NAPI in 1977 was the need tomeasure the differential maturity of preterm infants who participated ina randomized control study of the effectiveness of vestibularstimulation. The NAPI was published in 1990 (Korner & Thom, 1990)following more than ten years of extensive work to establish theconceptual framework to guide the selection of items, the statisticalcriteria for the inclusion of individual items and clusters, and thetest-retest reliability and developmental validity of selected items andclusters. Validity and reliability of items and clusters wereestablished with two independent samples, consisting of 179 and 290preterm infants who received a total of 907 examinations (Korner et al.,1987). Although the NAPI was designed to assess the maturity of preterms from32 weeks post-conceptional age to term, Korner and her colleagues weresubsequently able to establish its clinical validity (Korner et al.,1994). Finally, using the longitudinal data of infants who comprisedthe original NAPI samples, Korner showed that preterms exhibit stabledifferences in their response to stimulation provided by the examination(1996). This study originated from Korner's long-term interest in thebiological antecedents of individual differences (e.g. Korner, 1964,1971). The proposed symposium focuses on four recent investigations using theNAPI. The first paper examines the neurobehavioral developmental ofVLBW infants with low and high medical risk and compares thedevelopmental status of these two groups of infants to that offullterms; the second and third papers examine the extent to which NAPIscores in samples of VLBW infants can be used to predictneurodevelopmental problems or CP during the first year of life. Afinal paper extends the use of the NAPI to preterm infants at very lowmedical risk but at high socio-economic risk as well as to fullterms.
Details of individual items:
paper
Measurement of neurobehavioral characteristics of preterm infants hasbeen of interest for several decades. Several neurodevelopmental examshave been developed for that purpose. Korner's NeurodevelopmentalAssessment of the Preterm Infant (NAPI) has items that were selectedonly if they demonstrated developmental change and was designed tomeasure the relative maturity of functioning in preterm infants aged 32conceptional age to term. This presentation concerns145 very low birth weight (VLBW < 1500g) and38 fullterm (FT) infants involved in a longitudinal study for 24months. This study involved responses to early interaction andneurobehavioral exams were included in order to compare responses of 66high-risk with 79 low-risk VLBW infants from 32 to 42 postconceptionalage (high risk (HR) > 5 and low risk (LR) < 5 on the NeurobehavioralRisk Score). In addition, the VLBW infants were compared with FTinfants at 42 weeks. Questions: 1) do VLBW infants show reliablechanges in neurobehavioral characteristics over time; 2) are thesechanges affected by degree of illness (risk); and 3) do high and lowrisk VLBW infants differ from each other and FT infants at 42 weekspostconceptional age? Repeated Measures Analyses of variance were used to compare HR and LRinfants at all ages and with FT's at 42 weeks. Only cluster and summaryscores were used for analysis. Clusters were Motor Development andVigor (KMD), Alertness and Orientation (KAO) and Irritability (KIRR).Summary scores were Reactions to Handling (SRH), Physiological Responsesto Handling (SPR) and Spontaneous Movement (SSM). There were significant risk effects for SSM, KMD,SPR and KAO;significant exam (time) effects for KMD,SRH, KAO, and KIRR. There wereno significant interaction effects. HR infants showed moretremulousness and jerkiness and less smooth movements (SSM), less motordevelopment (KMD), more physiologic responses to handling (SPR) and lessalertness and orientation (KAO) than LR infants. Motor development,alertness and orientation, irritability and reactions to handling allincreased over time. At 42 weeks, there were risk effects for allmeasures except for KIRR. Duncan post-hoc analyses revealed that HRinfants showed significantly more tremulousness, jerkiness and lesssmooth movements (SSM) than LR and FT. LR infants showed more of thesebehaviors than FT's. HR infants demonstrated less motor developmentthan LR and FT infants, which did not differ. HR infants showedsignificantly less alertness and orientation than LR infants and LRinfants showed significantly less than FT infants. Both HR and LRinfants showed significantly more crying and increased tone to handling(SRH) than FT infants but did not differ from each other. HR infantsshowed significantly more physiological unstableness (SPR) than LR andFT infants, which did not differ. These findings support the use of the NAPI to track developmentalchange and demonstrate reliable risk differences in the development ofVLBW infants and differences in the neurodevelopmental characteristicsat 42 weeks. Most means for our cluster scores were within .5 S.D. andall were within 1 S.D. of the means published in the NAPI manual.
paper
The purpose of this study was to examine the predictive power of the Neurobehavioral Assessment of the Preterm Infant[1] (NAPI) at 36 weeks post conceptual age (PCA) to the Bayley Infant Neurodevelopmental Screener[2] (BINS) at 4 and 12 months corrected age (CA).Forty seven infants with birthweight <1250 grams and <30 weeks gestational age were recruited sequentially while in the NICU. There were 26 females and 19 males with a mean birth weight of 918.64 grams (range 502-1241). Mothers were a mean age of 30.4 years (range 15-42 years) with a mean of 15 years of education (range 0Ð21 years) and from a racially diverse population of Asian (4), Black (9), Caucasian (15), Hispanic (15), Other (4). Infants were assessed at 36 weeks PCA by an examiner masked to the infants prior medical course. At 4 and 12 months infants were examined by a different examiner from the one making the initial assessment. Spearman rank correlations were calculated between the total BINS score and each of the seven NAPI cluster scores: scarf sign, motor development, popliteal angle, alertness and orientation, irritability, cry quality and percent asleep rating. There were no significant correlations between any of the NAPI clusters and the BINS total score at 4 months CA. There were significant correlations between the total BINS score and the NAPI motor cluster (rho 0.40, p 0.005), the alertness and orientation cluster (rho 0.30, p 0.04) and the scarf sign (rho 0.30. p 0.04) at 12 months CA. The mean total BINS score at 4 months was 8.63 ± 2.06 which is in the moderate risk range for developmental delay and at 12 months the mean score was 5.73 ± 2.96 which is high risk for developmental delay with a significant difference between the two scores (p 0.0001).The NAPI assesses reflexes and early automated responses which are subcortical in origin. By 4 months the early reflexes are being suppressed by higher cortical function and by 12 months voluntary motor behavior is operational. For this reason short-term prediction may at times be less reliable than long-term prediction, contrary to what has been shown at later ages. These results support the value of early assessment before discharge as a predictor of one year outcome. At 4 months emerging developmental delays may be masked, as indicated by the significantly higher BINS scores at 4 than at 12 months CA. These results also suggest discontinuity of development, supporting the theory that development takes a non-linear course.1. Korner, A.F. and V.A. Thom, Neurobehavioral Assessment of the Preterm Infant. 1990, San Antonio: Psychological Corporation, Harcourt, Brace & Jovanovich.2. Aylward, G.P., Bayley Infant Neurodevelopmental Screener. 1995, San Antonio: Psychological Corporation, Harcourt, Brace & Co.
paper
We have found clinically that extremely premature infants diagnosedwith cerebral palsy at one year of age often have extreme tone withunderlying weakness. Through this study, we wanted to explore whetherthis motor risk pattern (MRP) is present in infants during their initialhospital stay and how motor movement tested in the NICU relates tooutcome at eight months corrected age. The modified NeurobehavioralAssessment of the Preterm Infant (NAPI; Korner and Thom, 1990) was usedto examine motor patterns in hospitalized infants. More specifically,we hypothesized that infants experiencing increased tone in thePopliteal Angle or Scarf Sign and weak Motor Development and Vigor(Motor) would have increased risk of motor difficulties. All infants in the study were born at 28 weeks or less gestation andwere free of established risks (intraventicular hemorrhage,perventicular leukomalacia, anoxic ischemic encephalopathy or geneticconditions). Infants (n25) had a mean gestational age of 26 weeks (SD1 week); mean birth weight of 878 grams (SD 180 grams); and were testedwith the NAPI at a mean conceptional age of 35 weeks (SD 1 week). Infants were tested at 8 months corrected age using the MovementAssessment of Infants ( Chandler, Andrews, and Swanson, 1980), totalscores were used in data analysis. NAPI standard scores for Scarf Sign,Motor Development and Vigor, and Popliteal Angle were entered into ANOVAas main effects. Additionally, interaction effects were also entered forPopliteal Angle and Motor; as well as Scarf Sign and Motor. Whole modelanalysis (main and interaction effects) suggests NAPI clusters weresignificantly related to the 8 month MAI scores (F (5,19) 7.42, p <0.0005). In looking at just main effects, the NAPI clusters of Motor(p<.001) and Popliteal Angle (p<.003) reached significance. Also, theinteraction between Motor and Popliteal Angle demonstrated significance(p<.01). Main effects accounted for 52% (p<003) of the variance andadding the interaction effects increased accounting to 66%. The NAPIcluster of Scarf Sign (p<.06) and the interaction effects with Motor(p.38) did not reach significance. However, due to the limited numberof subjects, there may not be enough power to reject the null hypotheses(power details available). Contribution of the Scarf Sign to the modelmust be explored with further research. However, Popliteal Angle andMotor shows promise in the early identification of movement problems.Analysis suggests that the interaction effects of Popliteal Angle andMotor may enhance prediction of motor outcome, as compared to the use ofsingle item prediction or use of the use of just main effects.Application of research to practice in a clinical setting will bediscussed. This study is ongoing and continues to enroll infants. Additionally,plans are to follow up these high risk infants through 2 years correctedage.
paper
The NAPI (Neurobehavioral Assessment of the Preterm Infant) was designedto assess the relative maturity of preterm infants (Korner & Thom,1990). We examined whether the NAPI is a useful instrument to examinethe neurobehavior of preterms at low medical risk but at highsocio-economic risk and whether the NAPI can be extended to assess theneurobehavior of fullterms. We asked three questions: (1) Are the NAPIcluster scores of the preterms in our sample similar to those ofpreterms in the normative sample? (2) Are the NAPI cluster scorespositively related to post-conceptional age (age at test) for pretermsas well as for fullterms? (3) Do NAPI cluster scores obtained during thenewborn period predict Bayley scores at 6, 12, and 24 month of age? The sample consisted of 216 newborns born in an urban hospital: 44cocaine exposed and 35 non-exposed preterms, all without major medicalcomplications, and 75 cocaine exposed and 62 non-exposed fullterms. In the hospital, preterms were examined with the NAPI close to 36 weekspost-conceptional age (M 36.05, SD 1.75) and fullterms, between 24and 72 hr postpartum. At 6, 12, and 24 months corrected age, infantswere examined in the laboratory with the Bayley Scales of InfantDevelopment (Bayley, 1969; n123 at 6 and 12 months, n113 at 24months). Mean NAPI cluster scores of the preterms fell within one standarddeviation of the means reported for the normative sample. However, meanscores of the preterms in our sample fell below the normative meanscores, except for the irritability scores, which fell above thenormative means. The Percent asleep cluster was excluded from furtheranalyses. The six remaining NAPI scores were associated with theinfants' post-conceptional age, but differently for preterms andfullterms (interaction F(6,190) 4.30, p<.01). NAPI scores accountedfor 32% and 16% of the variance for preterms and fulltermsrespectively. Scarf sign, motor development and vigor, and irritabilityincreased significantly (p<.05) as a function of post-conceptional agefor preterms, whereas popliteal angle and cry quality increased forfullterms. Finally, each of the six Bayley scores (MDIs and PDIs atthree ages) were regressed first on the infants' NAPI test age and codedvariables representing the infants' birth status (pre- or fullterm) andcocaine exposure (yes or no) and then on the six NAPI cluster scores.The NAPI scores accounted for an increase of 10% (p .07) and 11%(p<.05) in the variance of 6-month MDI and PDI scores respectively. Ofthe six NAPI scores, only the regresssion coefficients associated withirritability were significant (p< .05 for both): higher scores wereassociated with higher 6-month Bayley scores. NAPI scores did notaccount for significant increases in variance accounting for the 12- and24-month Bayley scores. Although the NAPI was developed to assess preterm infants with varyingmedical risk conditions, our data suggest that it may be possible toextend the use of the NAPI to preterms at low medical but highsocio-economic risk as well as to fullterm infants.