SUMMARY
Background:
Failure to Thrive (FTT) describes the development of an inappropriate pattern of growth, generally secondary to inadequate nutritional intake, and is associated with several negative outcomes. We describe key features among neonates with FTT as well as the variables that predicted their growth after birth at a Neonatal Intensive Care Unit.
Methods:
A retrospective single center study of 340 patients grouped into FTT (n = 100) and non-FTT (n = 240) was conducted. FTT was defined as having a weight <10th percentile on the Fenton 2013 curve at the time of discharge. For analyzing growth velocity, 204 patients were grouped into 4 quartiles based on their calculated growth velocity (grams/kilograms/day [g/kg/day]; 4th quartile had the highest velocity). Multivariate regression models were used to identify predictors of growth velocity.
Results:
When comparing FTT vs. non-FTT patients, lower birth weights (1897.9 ± 561.4 vs. 2445.9 ± 783.0 g, t(255.1) = −7.2, p < 0.001) and higher growth velocities (9.2 ± 3.9 vs. 8.0 ± 4.1 g/kg/day, t(153.6) = 2.2, p = 0.03) were noted. Among patients with higher growth velocities, birth weights were lower (1st to 4th quartiles: 2474.0 ± 677.0, 2000.0 ± 297.0, 1715.0 ± 285.0, 1533.0 ± 332.0 g, F(3, 200), = 46.5, p < 0.001, adjusted R2 = 0.4). Days to regain birth weight was the most consistent predictor of growth velocity in our overall patient sample (β [SE] = −0.3 [0.03], p < 0.001) and in the lowest growth velocity quartile subgroup (β [SE] = −0.3 [0.04], p < 0.001).
Conclusions:
Days to regain birth weight was consistently the strongest predictor of neonatal growth velocity along with difference in gender positive predicting growth velocity in the total sample. This highlights the importance of the first week of life in growth pattern establishment.
Keywords: Growth velocity, Neonatal growth, Time to regain birth weight, Failure to thrive
Introduction
Failure to thrive (FTT) or postnatal growth failure, is the development of an inappropriate growth pattern most commonly due to insufficient nutritional intake and is generally documented in children by the measurement of weight or height for age [1,2]. Without any definitive criteria, below 10th percentile for weight at post-menstrual age is commonly used for FTT identification in the neonatal population with Fenton Curve being the most commonly used chart [3,4]. FTT is thought to be the final result of multiple factors that contribute to poor weight gain which include preterm birth and low birth weight (<2500 g [g]) [5-7]. Preterm birth rates being on the rise, affecting 1 of every 10 infants born in the United States in 2018 with approximately 17% of the infant deaths in 2017, highlight the need for heightened awareness of early childhood nutrition [7,8].
FTT results in poor outcomes that may include impairment in normal growth and development, cognition, and immunological response [2,9]. These dire consequences highlight the need to study and assess the risk factors that can predict positive outcomes for these patients [10]. Reportedly, there is a significant variation in FTT severity among different Neonatal Intensive Care Units (NICUs) in the US with postnatal growth failure ranging from 50.3% to 64.5% in the past decade [11-14]. This variation could be due to data paucity and the lack of a standardized approach towards nutritional management and growth monitoring in the NICU [15].
Several advances to improve nutritional care including standardized feeding protocols have been introduced with the goal of improving growth outcomes among very low birth weight infants [15-17]. Despite these efforts the implementation of standardized protocols has been difficult due to specific requirements for each infant due to which growth management remains a challenge among preterm and low birth weight infants [15]. Factors that predict neonatal growth velocity need to be understood in order to improve management and outcomes among NICU patients at risk of developing FTT. Studies have shown that growth outcomes in preterm infants can be improved by identifying the severity of the problem and assessment of current practices at the level of an individual NICU including monitoring of feeding protocols with individualized modifications, development of a nutritional support team within the NICU, and regular follow-ups after feeding protocol introduction and post-discharge [15]. Several nutritional protocols exist in our institution's NICU, including a nutritional hospital course specific to each infant based on gestational age or weight and the presence of a registered dietician during to help tailor each patient's nutritional management.
Previous studies have focused on the nutritional and growth management of low birth weight and very low birth weight populations, which often require the greatest acuity of care [6,10,12,13,18,19]. Our study is aimed at better understanding the factors that affect neonatal growth (including postnatal growth velocity) during the admission period, without any particular emphasis on low birth weight populations. This will allow true assessment of neonatal growth without low birth weight being a contributing factor. The trends identified in the management and outcomes of susceptible neonates will lead to better quality control measures and an improvement in the management of this population.
Methods
Study population
A retrospective chart review was performed of 340 neonates admitted to our institution's NICU between 2015 and 2017. Patients were included if they met the following criteria: 1) ≥30 weeks gestational age at the time of birth; and, 2) ≥1250 g weight at the time of birth. Exclusion factors were: 1) receiving care at a different hospital during newborn admission which included being transferred to and from our institution; 2) being eventually discharged from the newborn nursery; 3) deceased at the time of chart review; and, 4) diagnosed with a cyanotic congenital heart lesion or required cardiothoracic surgery during newborn admission. The reason behind exclusion criteria 2 was to exclude medically stable patients who were admitted to the NICU solely for a brief observation period.
We further categorized our patient sample into “Failure to Thrive” (FTT) or a “Non-Failure to Thrive” (non-FTT) groups. FTT was defined as weight less than the 10th percentile for gestational age at the time of discharge based on the Fenton 2013 curve [6]. For neonatal growth velocity assessment, only patients admitted for >14 days were included (n = 204). This additional exclusion criterion was based on the expected weight loss in the early neonatal period prior to regaining birth weight within the first two weeks of life regardless of the baseline FTT status [20].
Data collection
500 neonates were randomly selected, 340 of whom met our inclusion criteria. 100 were identified as FTT and 240 were identified as non-FTT based on discharge weight percentile. Out of these, 204 patients with a hospital length of stay >14 days (as noted under study population above) were included for the analyses involving growth velocity assessment. For further robust assessment of growth velocity, these 204 patients were subdivided into four growth velocity quartiles (n = 51 each; explained below). FTT patients had discharge weight <10th percentile for corrected gestational age. Demographic variables (e.g. gestational age, sex) and descriptive characteristics (e.g. length of stay, days needed to regain birth weight, age at initiation of enteral nutrition) were collected through the electronic medical records. Prenatal diagnosis of intrauterine growth restriction (IUGR) was assessed via documentation of IUGR or if the patient was noted to be small for gestational age at time of delivery. We calculated Z-Scores for weights at time of birth and discharge using the Fenton 2013 Growth Curves.
Growth velocity was calculated in grams/kilogram/day (g/kg/ day) and was calculated as change in weight (from birth to discharge) divided by average weight (averaging weight at birth and discharge) divided by length of stay [21]. Patients were further split into four growth velocity quartiles (n = 51 per group), labeled Q1, Q2, Q3, and Q4 (with Q1 having the lowest and Q4 having the highest growth velocity).
All statistical analyses were performed using R 3.5.0. Continuous variables were expressed as mean ± standard deviation (SD) or median (interquartile range [IQR]) and categorical variables were expressed as percentages (%). Variables included in statistical models were chosen based on clinical relevance. Differences between FTT and non-FTT groups were assessed via either unpaired two-sample t-tests or Wilcoxon Mann—Whitney U tests (depending on normality). For differences between growth velocity quartiles, univariate ANOVA or Kruskal—Wallis tests were performed. ANOVA was followed by Tukey's HSD test and Kruskal—Wallis was followed by pairwise Wilcoxon Mann—Whitney U tests. Normality was assessed via QQ plots.
Multivariate linear regression models were built to explain neonatal growth velocity (g/kg/day) and included variables based on clinical relevance. The initial model included the entire sample with a hospital length of stay >14 days (n = 204). Further sub-group analyses performed for growth velocity were among: 1) patients identified with FTT (n = 67); and, 2) lowest growth velocity quartile (Q1) patients (n = 45). Variables with missing predictor values were removed before performing linear regression. These were missing if the neonate was discharged before regaining birth weight or reaching 110 ml/kg/day of enteral feeds. 110 ml/kg/day is typically the volume considered to be full enteral feeding at our institute. Notably, birth weight and hospital length of stay were included as co-variates in the model rather than as predictors as these are directly included in the growth velocity calculation.
Ethics statement
This study was approved by University of Virginia Institutional Review Board: exemption was granted (#20116).
Results
FTT and non-FTT groups had an equal gender distribution (50% and 52% male, respectively) and a roughly similar method of delivery (66% and 57% via cesarean section, respectively) as shown in Table 1. These groups were significantly different for gestational age (35.0 ± 2.6 vs. 34.3 ± 3.0 weeks, t(210.0) = 2.1, p = 0.03), birth weight (1897.9 ± 561.4 vs. 2445.9 ± 783.0 g, t(255.1) = −7.2, p < 0.001), and growth velocity (9.2 ± 3.9 vs. 8.0 ± 4.1 g/kg/day, t(153.6) = 2.2, p = 0.03). Additionally, the FTT group had lower discharge weights (median [IQR]: 2170.0 [475.5] vs. 2560.0 [760.0], p < 0.001) and longer hospital lengths of stay (median [IQR]: 19.5 [16.3] vs. 16.0 [16.0], p = 0.01) compared to their non-FTT counterparts (Table 1).
Table 1.
Descriptive characteristics for patients identified with Failure to Thrive (FTT) and non-Failure to Thrive (non-FTT).
| FTT | Non-FTT | t | W | p-value | |
|---|---|---|---|---|---|
| Patient Characteristics | |||||
| Gestational age in weeks; mean (SD) | 35.0 (2.6) | 34.3 (3.0) | 2.1 | 0.03 | |
| Birth weight in g; mean (SD) | 1897.9 (561.4) | 2445.9 (783.0) | −7.2 | <0.001 | |
| Birth length in cm; mean (SD) | 43.1 (4.3) | 45.4 (4.9) | −4.2 | <0.001 | |
| Discharge gestational age in weeks; median [IQR] | 34.6 [3.14] | 34.0 [3.75] | 16,234.0 | <0.001 | |
| Discharge weight in g; median [IQR] | 2170.0 [475.5] | 2560.0 [760.0] | 5643.5 | <0.001 | |
| Growth Characteristics | |||||
| Days to regain birth weight; median [IQR] (nFTT = 88, nNon-FTT = 188) | 5.0 [6.0] | 8.0 [7.0] | 15,574.0 | 0.1 | |
| Age at start of enteral feeds in days; mean (SD) | 1.2 (2.7) | 1.1 (2.0) | 0.03 | 0.1 | |
| Days to achieve 110 ml/kg/day enteral feeds; median [IQR](nFTT = 92, nNon-FTT = 214) | 6.0 [4.0] | 7.0 [2.0] | 9524.0 | 0.6 | |
| Days with parenteral feeds; median [IQR] | 5.0 [5.5] | 5.0 [4.0] | 12,234.0 | 0.8 | |
| Average daily weight gain in g/day; median [IQR] | 1.6 [12.5] | 13.1 [22.0] | 13,545.0 | 0.06 | |
| Growth velocity in g/kg/day; mean (SD) | 9.2 (3.9) | 8.0 (4.1) | 2.2 | 0.03 | |
| Miscellaneous Characteristics | |||||
| Days with respiratory support; median [IQR] | 0.0 [3.0 | 1.0 [4.0] | 10,486.0 | 0.05 | |
| Hospital length of stay in days; median [IQR] | 19.5 [16.3] | 16.0 [16.0] | 14,041.0 | 0.01 |
FTT = Failure to Thrive. Non-FTT = Non Failure to Thrive. nFTT = numbers of patients with Failure to Thrive. nNon-FTT = numbers of patients with Non Failure to Thrive. SD = standard deviation. t = t-test statistic. W = Wilcoxon test statistic. IQR = Interquartile Range. FTT was defined as having a weight <10th percentile on the Fenton 2013 curve at the time of discharge. Variables analyzed with a t-test are reported as mean (SD). Variables analyzed with a Wilcoxon Mann—Whitney test are reported as median [IQR]. The sample size was n = 100 FTT and n = 240 non-FTT patients unless stated otherwise.
Median [IQR] growth velocities for Q1 to Q4 were 4.2 [2.6], 7.7 [1.2], 10.0 [1.1], and 12.0 [1.5] g/kg/day, respectively. In terms of patient characteristics, Q1 showed greater birth weights, greater time to regain birth weight, and decreased average weight gain compared to other quartiles (Table 2 and Fig. 1). Based on hospital characteristics, Q1 showed decreased number of days with parenteral feeds and decreased hospital length of stay (Table 3). Number of FTT patients within each growth velocity quartile were as follows: n = 14 Q1, n = 19 Q2, n = 17 Q3, and n = 22 Q4. Results from the post-hoc analyses highlighting interquartile differences are shown in Supplemental Tables 1 and 2
Table 2.
Comparison of patient variables between Growth Velocity Quartiles (Q1, Q2, Q3, Q4).
| Q1 Mean (SD) | Q2 Mean (SD) | Q3 Mean (SD) | Q4 Mean (SD) | F | DF(num) | DF(den) | p-value | Adjusted R2 | |
|---|---|---|---|---|---|---|---|---|---|
| Birth weight (g) | 2474.0 (677.0) | 2000.0 (297.0) | 1715.0 (285.0) | 1533.0 (332.0) | 46.5 | 3 | 200 | <0.001 | 0.4 |
| Gestational age (weeks) | 34.5 (2.4) | 33.5 (1.9) | 32.1 (1.7) | 32.0 (2.2) | 16.6 | 3 | 200 | <0.001 | 0.2 |
| Birth length (cm) | 45.9 (3.5) | 44.0 (2.6) | 41.7 (2.9) | 40.4 (3.4) | 32.1 | 3 | 200 | <0.001 | 0.3 |
| Time to regain birth weight (days) | 11.0 (7.0) | 8.0 (5.0) | 7.0 (4.0) | 6.0 (4.0) | 7.2 | 3 | 193 | <0.001 | 0.1 |
| (nQ1 = 46, nQ2 = 51, nQ3 = 50, nQ4 = 51) | |||||||||
| Discharge weight (g) | 2635.0 (617.0) | 2458.0 (379.0) | 2374.0 (436.0) | 2507.0 (545.0) | 2.4 | 3 | 200 | 0.1 | 0.02 |
| Average daily weight gain (g/day) | 7.0 (11.9) | 17.0 (2.8) | 20.6 (3.0) | 25.4 (3.8) | 72.8 | 3 | 200 | <0.001 | 0.5 |
Q1, Q2, Q3, Q4 = 1st, 2nd, 3rd, 4th Growth Velocity Quartiles. nQ1, nQ2, nQ3, nQ4 = number of patients with 1st, 2nd, 3rd, 4th Growth Velocity Quartiles. DF = degrees of freedom. SD = standard deviation. Num = numerator. Dem = denominator. Comparisons were performed using univariate ANOVA (R2: coefficient of determination). Variables were selected based on clinical relevance. Sample size for each growth velocity quartile was 51 unless otherwise stated.
Fig. 1.
Strip chart and box plots showing mean and standard deviation for comparison of patient variables between Growth Velocity Quartiles. Key: Q1, Q2, Q3, and, Q4: first (lowest), second, third, and fourth growth velocity quartiles.
Table 3.
Comparison of hospital variables between Growth Velocity Quartiles (Q1, Q2, Q3, Q4).
| Q1 Median [IQR] | Q2 Median [IQR] | Q3 Median [IQR] | Q4 Median [IQR] | χ2 | DF | p-value | |
|---|---|---|---|---|---|---|---|
| Hospital length of stay (days) | 19.0 [5.0] | 23.0 [11.0] | 28.0 [16.0] | 32.0 [29.0] | 42.4 | 3 | <0.001 |
| Discharge gestational age (weeks) | 36.6 [1.9] | 36.6 [2.2] | 36.3 [1.6] | 37.4 [2.3] | 7.3 | 3 | 0.1 |
| Days with parenteral feeds | 5.0 [2.5] | 6.0 [2.5] | 6.0 [2.0] | 6.0 [3.5] | 8.7 | 3 | 0.03 |
| Age at start of enteral feeds (days) | 1.0 [1.0] | 1.0 [1.0] | 1.0 [2.0] | 1.0 [1.0] | 4.1 | 3 | 0.3 |
| Days of respiratory support | 1.0 [4.5] | 2.0 [5.0] | 2.0 [7.0] | 4.0 [15.5] | 6.6 | 3 | 0.1 |
| Days to reach 110 ml/kg/day of enteral feedsnQ3 = 51, nQ4 = 51) | 7.0 [2.0] | 8.0 [3.0] | 7.0 [2.5] | 7.0 [4.0] | 4.9 | 3 | 0.2 |
Q1, Q2, Q3, Q4 = 1st, 2nd, 3rd, 4th Growth Velocity Quartiles. nQ1, nQ2, nQ3, nQ4 = number of patients with 1st, 2nd, 3rd, 4th Growth Velocity Quartiles. DF = degrees of freedom. IQR = interquartile range. χ2: chi-square. Comparisons were performed using Kruskal—Wallis. Variables were selected based on clinical relevance. Sample size for each growth velocity quartiles was 51 unless otherwise stated.
Multivariable linear regression analysis was conducted to identify predictors of growth velocity among patients with length of stay >14 days within the total sample, FTT patients, and Q1 (Table 4). In the total sample, the predictors of growth velocity were male sex, time to regain birth weight, and days of respiratory support. Within FTT patients, the significant predictors were time to regain birth weight and days with respiratory support. Among the Q1 group, predictors were gestational age, time to regain birth weight, age at start of enteral feeds, days with parenteral feeds, and prenatal diagnosis of IUGR. Time to regain birth weight was the only variable that was consistently identified across all the groups as a predictor of growth velocity during the neonatal admission period (Fig. 2).
Table 4.
Multivariate associations with growth velocity (g/kg/day) for neonates with hospital length of stay > 14 days.
| β | SE | t | p-value | |
|---|---|---|---|---|
| Total Sample | ||||
| Male sex | 0.8 | 0.3 | 2.5 | 0.01 |
| Birth weight (g) | −0.003 | 0.001 | −6.0 | <0.001 |
| Gestational age (weeks) | −0.2 | 0.1 | −1.3 | 0.1 |
| Days to regain birth weight | −0.3 | 0.03 | −7.7 | <0.001 |
| Age at start of enteral feeds (days) | 0.1 | 0.1 | 0.5 | 0.5 |
| Days to reach 110 ml/kg of enteral feeds | −0.2 | 0.1 | −1.9 | 0.1 |
| Days with parenteral feeds | 0.04 | 0.1 | 0.8 | 0.7 |
| Days with respiratory support | −0.07 | 0.02 | −3.7 | <0.001 |
| Hospital length of stay (days) | 0.1 | 0.02 | 4.5 | <0.001 |
| Oxygen resuscitation at birth * | 0.5 | 0.4 | 1.3 | 0.2 |
| Diagnosis of respiratory distress | −0.3 | 0.5 | −0.9 | 0.5 |
| Prenatal diagnosis of IUGR ** | 0.8 | 0.5 | 1.6 | 0.1 |
| Failure to Thrive | ||||
| Male sex | 1.2 | 0.6 | 1.8 | 0.07 |
| Birth weight (g) | −0.007 | 0.001 | −5.2 | <0.001 |
| Gestational age (weeks) | 0.5 | 0.3 | 2.0 | 0.2 |
| Days to regain birth weight | −0.2 | 0.08 | −2.3 | 0.01 |
| Age at start of enteral feeds (days) | −0.02 | 0.2 | −0.4 | 0.9 |
| Days to reach 110 ml/kg of enteral feeds | −0.3 | 0.2 | −1.6 | 0.1 |
| Days with parenteral feeds | 0.2 | 0.1 | 1.8 | 0.2 |
| Days with respiratory support | −0.06 | 0.04 | −2.1 | 0.1 |
| Hospital length of stay (days) | 0.03 | 0.04 | 1.2 | 0.4 |
| Oxygen resuscitation at birth* | −0.6 | 1.0 | −0.6 | 0.5 |
| Diagnosis of respiratory distress | 1.1 | 1.0 | 1.0 | 0.3 |
| Prenatal diagnosis of IUGR** | 0.7 | 0.7 | 0.8 | 0.4 |
| Growth Velocity Quartile 1 | ||||
| Male sex | 0.5 | 0.6 | 0.8 | 0.4 |
| Birth weight (g) | −0.001 | 0.001 | −1.0 | 0.4 |
| Gestational age (weeks) | −1.0 | 0.3 | −3.2 | <0.001 |
| Days to regain birth weight (days) | −0.3 | 0.04 | −6.7 | <0.001 |
| Age at start of enteral feeds (days) | 0.7 | 0.3 | 2.3 | 0.02 |
| Days to reach 110 ml/kg of enteral feeds (days) | 0.1 | 0.1 | 0.5 | 0.4 |
| Days with parenteral feeds | −0.4 | 0.1 | −2.5 | 0.01 |
| Days with respiratory support | 0.02 | 0.07 | 0.2 | 0.8 |
| Hospital length of stay (days) | 0.2 | 0.07 | 2.8 | 0.01 |
| Oxygen resuscitation at birth* | 1.0 | 0.6 | 1.9 | 0.1 |
| Diagnosis of respiratory distress | −0.8 | 0.6 | −1.5 | 0.2 |
| Prenatal diagnosis of IUGR** | 2.1 | 1.0 | 2.1 | 0.1 |
SE = standard error. IUGR = intrauterine growth restriction. Models constructed as follows: Y = β0 + β1X1 + β2X2 … βnXn where Y represents growth velocity, and βn represents the marginal effect of the nth predictor, Xn. Predictors were selected based on clinical relevance and low co-linearity with other predictors. Sample size was 196 for Total Population, 67 for Failure to Thrive, and 45 for Growth Velocity Quartile 1.
Recorded as whether required or not required.
Prenatal diagnosis of intrauterine growth restriction (IUGR) was assessed via documentation of IUGR or if the patient was noted to be small for gestational age at time of delivery.
Fig. 2.
Scatterplot showing time to regain birth weight (in days) versus growth velocity (g/kg/day) in the total sample with green representing males and purple representing females.
Discussion
This retrospective study assessed the outcomes of neonates admitted to our NICU within a two-year study period in which assessed growth velocity of patients with FTT versus non-FTT and identified predictors of neonatal growth. We aimed to investigate neonatal growth velocity in our total sample, infants with FTT, and those with the lowest growth velocities (Q1 groups). Notable findings include: 1) decreased growth velocity among non-FTT when compared to FTT patients; 2) male gender as a positive predictor of growth velocity in the total sample; and, 3) time to regain birth weight as the most consistent predictor of neonatal growth.
As stated earlier, much of the literature regarding growth failure in neonatal settings focuses on low birth weight patients [6,10,12,13,18,19]. In contrast, our sample provides a potential avenue for stratifying patients at risk for poor growth who may otherwise have received less attention secondary to less concerning birth weights and gestational ages. In our total sample, growth velocity decreased by approximately 0.3 g/kg/day for every additional day a patient took to regain birth weight. Higher growth velocities have been associated with better neurologic outcomes in previous studies [22,23]. Despite this, infants with weight appropriate for gestational age as compared to those with FTT have been reported to have slower growth velocities which can often be explained by catch-up growth, i.e. those with lower neonatal weight at birth gain significant strides in terms of growth velocity when provided optimal conditions for growth at the NICU [24]. It is also generally accepted that healthy newborns regain birth weight within 14 days although it is not uncommon for them to continue to be below birth weight after 14 days [20]. Cautious interpretation of birth weight is advised since intravenous fluid usage in mothers during parturition may initially result in falsely elevated birth weights with subsequent weight loss (within the first 24 h) representing neonatal diuresis [25]. Notably, at our institution, healthy patients born at > 35 weeks gestational age are generally admitted to the newborn nursery instead of the NICU. While it was interesting to note that in our sample, the mean gestational age in patients with FTT was higher than in patients without FTT (35.0 vs 34.3 weeks, respectively), this could also be reflective of the fact that NICU patients born > 35 weeks have more complex medical needs at baseline than the standard 34 week NICU patient.
Our results also displayed male gender as a positive predictor of growth velocity. This significance was lost when the infants were grouped into FTT and those within the lowest growth velocity quartile (Q1). Even though there is no gender or race predisposition for FTT, studies have shown an association of gender with growth velocities and postnatal growth failure [26]. A study by Bertino et al. assessing the role of gender in growth velocity found that males have higher growth velocities compared to females among infants with very low birthweights from the 2nd to the 6th month of postnatal life [27]. Another study by Marks et al. analyzed data from Israel Neonatal Network database on very low birth weight patients between 1995 and 2001 and found an association of gender with severe postnatal growth failure among very low birthweight infants [28].
Negative predictors of growth velocity identified within the total sample were: time to regain birth weight, and days of respiratory support (Table 4). Days on respiratory support can be a negative predictor since it represents the acuity of care required by critically ill patients; critically ill neonates frequently require prolonged respiratory support. For the FTT group, time to regain birth weight was the only identified negative predictor of growth velocity (Table 4). For the Q1 group, the positive predictors were: age at the start of enteral feeds and the presence of a prenatal diagnosis of IUGR. Negative predictors were: gestational age, days of parenteral feeds, and time to regain birth weight (Table 4). These findings also suggest that in this group of patients with lowest growth velocities, the age at the time enteral feeds were started played an important role as it was significantly a positive predictor of growth velocity.
The time it took for a patient to regain their birth weight was the most consistently significant predictor of growth velocity. Prior studies have reported that an increased time required to regain birth weight is associated with a greater drop in weight for age Z-scores between birth and discharge in extremely low birth weight patients [29], which would correlate with a decreased postnatal growth velocity. It is interesting that the specific time period that a neonate takes to regain birth weight was around a week (mean ± standard deviation: 8 ± 5 days) for our patients. This finding is of particular interest from a quality improvement perspective: understanding that the length of time needed for a neonate to regain birth weight is a predictor of eventual growth velocity (and potentially long-term outcomes) can prove useful in stratifying patients by risk during the neonatal admission period.
Generally, postnatal weight loss is understood to primarily represent fluid loss [30], but it may also be due to loss of fat stores during establishment of breastfeeding within the first few days of life [31]. Research concerning the epigenetics and establishment of early nutrition is also ongoing [32]. Intrauterine inflammation exposure increases newborn susceptibility to epigenetic modifications which also coincides with the inflammatory elements associated with various diseases of prematurity (e.g. necrotizing enterocolitis, chronic lung disease of prematurity etc.) [32]. While outside the initial scope of this study, the significance of the timeline associated with regaining birth weight and the corresponding epigenetic component in the neonate's development offers much to consider when studying the factors that contribute to early neonatal nutrition and growth.
Strengths of this study include: 1) patient groupings designed to reflect standard NICU patients while removing those admitted for observation and those with morbidities associated with severe prematurity and/or low birth weights; 2) relatively large sample size; 3) use of multivariate regression to characterize variables influencing a specific measurable outcome. Limitations were: 1) retrospective data collection which resulted in missing variables and limited data, especially in terms of IUGR identification which was more evident if the mother had received prenatal care elsewhere; 2) limitations on ascertaining specific nutritional strategies as such interventions are frequently adjusted day-by-day with variable degrees of detail when documenting the specific nuances of nutritional changes; 3) referral bias given our institution's status as an academic tertiary referral center; 4) lack of an established definition for neonatal FTT with varying reference charts used to for the generally acceptable definition of weight below 10th percentile for post-menstrual age — Fenton curve was selected for our study based on its usage in our institution's electronic medical records and its prevalence in NICUs across the country [21]; and, 5) confounding factors for assessment of growth in our study population as patients with complex medical histories (apart from cardiac conditions) were not excluded and fluid shifts were not accounted for.
Potential future directions include exploration of the factors that influence the critical period of postnatal growth including in-depth nutritional assessment (daily amounts of protein, carbohydrate, and calories received) where a neonate regains their birth weight as well as studying the outcomes associated with growth velocity. We have shown that the time required to regain neonatal birth weight is a predictor of growth velocity for patients with FTT. Areas to look into during this early period of growth could include further examining the role of epigenetic changes in the context of the early microbiome and studying the differences in metabolomics between neonatal patients with differing growth velocities. It would be of great interest to determine clinical outcomes associated with NICU patients as a function of the time it takes for them to regain their birth weight and to identify differences in their respective backgrounds that contribute to their presentations using prospectively designed studies.
Conclusion
We assessed the outcomes of neonates admitted to our institution's NICU in the setting of FTT. We identified factors that were associated with neonatal growth, specifically growth velocity. The primary predictor of growth velocity consistently identified across our population was the time it took for a neonate to regain birth weight, which was about a week for our population (mean ± standard deviation: 8 ± 5 days). This study encourages further investigation to ascertain the specific mechanisms occurring in the early postnatal period that could predict neonatal growth, which may include newborn metabolomics or early onset epigenetic modifications.
Supplementary Material
Acknowledgments
Grants and funding
No funding was obtained for this study.
Abbreviations
- FTT
Failure to Thrive
- Non-FTT
Non Failure to Thrive
- NICU
Neonatal Intensive Care Unit
- IUGR
Intrauterine Growth Restriction
- IQR
Interquartile Range
- SD
Standard Deviation
Q1, Q2, Q3, Q4 1st, 2nd, 3rd, 4th Growth Velocity Quartiles grams/ kilograms/ day g/kg/day
Footnotes
Declaration of Competing Interest
None.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.clnesp.2020.05.010.
References
- [1].Goh LH, How CH, Ng KH. Failure to thrive in babies and toddlers. Singap Med J 2016;57(6):287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Homan GJ Failure to thrive: a practical guide. Am Fam Physician 2016;94(4):295–9. [PubMed] [Google Scholar]
- [3].Fenton TR, Kim JH. A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants. BMC Pediatr 2013;13(1):59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Lee SM, Kim N, Namgung R, et al. Prediction of postnatal growth failure among very low birth weight infants. Sci Rep 2018;8(1):1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Gahagan S. Failure to thrive: a consequence of. Pediatr Rev 2006;27(1):e1–11. [DOI] [PubMed] [Google Scholar]
- [6].Dusick AM, Poindexter BB, Ehrenkranz RA, et al. Growth failure in the preterm infant: can we catch up?. Semin perinatol. Elsevier; 2003. p. 302–10. [DOI] [PubMed] [Google Scholar]
- [7].De Curtis M. Rigo J Extrauterine growth restriction in very-low-birthweight infants. Acta Paediatr 2004;93(12):1563–8. [DOI] [PubMed] [Google Scholar]
- [8].Preterm birth. 2019. Accessed, https://www.cdc.gov/reproductivehealth/maternalinfanthealth/pretermbirth.htm.
- [9].Blenner S, Hironaka LK, Vanderbilt DL, Frank DA. Prematurity and failure to thrive: the interplay of medical conditions and development. In: Handbook of developmental psychopathology. Springer; 2014. p. 389–408. [Google Scholar]
- [10].Ehrenkranz R, Dusick A, Vohr B, Wright LL, Wrage LA, Poole WK. The National Institutes of Child Health and Human Development Neonatal Research Network. Growth in the neonatal intensive care unit influences neurodevelopmental and growth outcomes of extremely low birth weight infants. Pediatrics 2006;117(4):1253–61. [DOI] [PubMed] [Google Scholar]
- [11].Cooke R, Ainsworth S. Fenton A Postnatal growth retardation: a universal problem in preterm infants. Arch Dis Child Fetal Neonatal Ed 2004;89(5):F428–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Griffin IJ, Tancredi DJ, Bertino E, Lee HC, Profit J. Postnatal growth failure in very low birthweight infants born between 2005 and 2012. Arch Dis Child Fetal Neonatal Ed 2016;101(1):50–5. [DOI] [PubMed] [Google Scholar]
- [13].Horbar JD, Ehrenkranz RA, Badger GJ, Edwards EM, Morrow KA, Soll RF, et al. Weight growth velocity and postnatal growth failure in infants 501 to 1500 grams: 2000–2013. Pediatrics 2015;136(1):e84–92. [DOI] [PubMed] [Google Scholar]
- [14].Olsen IE, Richardson DK, Schmid CH, Ausman LM, Dwyer JT. Intersite differences in weight growth velocity of extremely premature infants. Pediatrics 2002;110(6):1125–32. [DOI] [PubMed] [Google Scholar]
- [15].Cooke RJ. Improving growth in preterm infants during initial hospital stay: principles into practice. Arch Dis Child Fetal Neonatal Ed 2016;101(4):F366–70. [DOI] [PubMed] [Google Scholar]
- [16].Ziegler EE, Thureen PJ. Carlson SJ Aggressive nutrition of the very low birthweight infant. Clin Perinatol 2002;29(2):225–44. [DOI] [PubMed] [Google Scholar]
- [17].McCallie K, Lee H, Mayer O, Cohen RS, Hintz SR, Rhine WD. Improved outcomes with a standardized feeding protocol for very low birth weight infants. J Perinatol 2011;31(1):S61–7. [DOI] [PubMed] [Google Scholar]
- [18].Henderson G, Fahey T, McGuire W Nutrient-enriched formula milk versus human breast milk for preterm infants following hospital discharge. Cochrane Database Syst Rev 20074). [DOI] [PubMed] [Google Scholar]
- [19].Maruyama H, Yonemoto N, Kono Y, Kusuda S, Fujimara M, Neonatal Research Network of Japan. Weight growth velocity and neurodevelopmental outcomes in extremely low birth weight infants. PloS One 2015;10(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Paul IM, Schaefer EW, Miller JR, Kuzniewicz MW, Li SX, Walsh EM, et al. Weight change nomograms for the first month after birth. Pediatrics 2016;138(6):e20162625. [DOI] [PubMed] [Google Scholar]
- [21].Fenton TR, Chan HT, Madhu A, Griffin IJ, Hoyos A, Ziegler EE, et al. Preterm infant growth velocity calculations: a systematic review. Pediatrics 2017;139(3):e20162045. [DOI] [PubMed] [Google Scholar]
- [22].Simon L, Hanf M, Frondas-Chauty A, Darmaun D, Rouger V, Gascoin G, et al. Neonatal growth velocity of preterm infants: the weight Z-score change versus Patel exponential model. PloS One 2019;14(6):e0218746. e46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Ehrenkranz RA, Dusick AM, Vohr BR, Wright LL, Wrage LA, Poole WK. Growth in the neonatal intensive care unit influences neurodevelopmental and growth outcomes of extremely low birth weight infants. Pediatrics 2006;117(4):1253–61. [DOI] [PubMed] [Google Scholar]
- [24].Claris O, Beltrand J, Levy-Marchal C. Consequences of intrauterine growth and early neonatal catch-up growth. Semin Perinatol 2010;34(3):207–10. [DOI] [PubMed] [Google Scholar]
- [25].Noel-Weiss J, Woodend AK, Peterson WE, Gibb W, Groll DL. An observational study of associations among maternal fluids during parturition, neonatal output, and breastfed newborn weight loss. Int Breastfeed J 2011;6(1):9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Smith AEBM. Failure to thrive. 2019. https://www.ncbi.nlm.nih.gov/books/NBK459287/. Accessed.
- [27].Bertino E, Coscia A, Boni L, Rossi C, Martano C, Giuliani F, et al. Weight growth velocity of very low birth weight infants: role of gender, gestational age and major morbidities. Early Hum Dev 2009;85(6):339–47. [DOI] [PubMed] [Google Scholar]
- [28].Marks KA, Reichman B, Lusky A, Zmora E, Israel Neonatal Network. Fetal growth and postnatal growth failure in very-low-birthweight infants. Acta Paediatr 2006;95(2):236–42. [DOI] [PubMed] [Google Scholar]
- [29].Steward DK. Pridham KF Growth patterns of extremely low-birth-weight hospitalized preterm infants. J Obstet Gynecol Neonatal Nurs 2002;31(1):57–65. [DOI] [PubMed] [Google Scholar]
- [30].Polin RA, Fox WW. Abman SH fetal and neonatal physiology. 3rd ed., 1; 2003. p. 1–2008. [Google Scholar]
- [31].Wright C, Parkinson K. Postnatal weight loss in term infants: what is “normal” and do growth charts allow for it? Arch Dis Child Fetal Neonatal Ed 2004;89(3):F254–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Lu L Claud EC Intrauterine inflammation, epigenetics, and microbiome influences on preterm infant health. Current pathobiology reports 2018;6(1):15–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.