Abstract
Objective
To determine whether severity of lung disease at age 6 years is associated with changes in nutritional status before age 6 within individual children with cystic fibrosis (CF).
Study Design
Children with CF born between 1994 and 2005 and followed in the CF Foundation Patient Registry from age ≤2 through 7 years were assessed according to changes in annualized weight-for-length (WFL) percentiles between ages 0 and 2 and body mass index (BMI) percentiles between ages 2 and 6. The association between growth trajectories before age 6 and forced expiratory volume in one second (FEV1) % predicted at age 6-7 years was evaluated using multivariable linear regression.
Results
A total of 6,805 subjects met inclusion criteria. Children with annualized WFL-BMI always >50th percentile [N=1,323 (19%)] had the highest adjusted mean [95% Confidence Interval (CI)] FEV1 at 6-7 years [101.8 (100.1, 103.5)]. FEV1 at 6-7 years for children whose WFL-BMI increased >10 percentile points by age 6 years was 98.3 (96.6, 100.0). This was statistically significantly higher than FEV1 for children whose WFL-BMI was stable [94.4 (92.6, 96.2)] or decreased >10 percentile points [92.9 (91.1, 94.8)]. Among children whose WFL-BMI increased >10 percentile points, achieving and maintaining WFL-BMI >50th percentile at younger ages was associated with significantly higher FEV1 at 6-7 years.
Conclusions
Within-patient changes in nutritional status in the first 6 years of life are significantly associated with FEV1 at age 6-7 years, suggesting that interventions that improve nutrition in early life may lead to improvements in later lung function.
Keywords: CF, FEV1, Forced expiratory volume in 1 second, Weight, BMI, Newborn screening
INTRODUCTION
Cystic fibrosis (CF) is a multisystem genetic condition that leads to deficits in growth and nutrition as well as progressive lung disease.(1) Strong cross-sectional correlations between growth indices and forced expiratory volume in 1 second (FEV1) have led the CF Foundation (CFF) to recommend that all children maintain at least the 50th percentile for weight-for-length (WFL) or body mass index (BMI).(2) With the widespread adoption of newborn screening (NBS) for CF, clinicians have the opportunity to encourage early weight gain, but the evidence to support an association between longitudinal changes in growth early in life and later lung health for individual patients is limited. Konstan et al. described an association between growth indices at age 3 and FEV1 at age 6, even after adjusting for the severity of lung disease at age 3.(3) Yen et al. extended this analysis and described an association between patients with higher weight-for-age percentile at age 4 years and better FEV1 from 6 to 18 years of age, fewer pulmonary exacerbations, and better survival.(4) Together these two studies suggest the need to optimize nutrition in early in life. However, their assessments of growth were taken at only one time point and measures of lung disease are limited in early childhood. Therefore, it is not known if improvements in growth in early life can minimize later lung disease, or whether the timing of these changes is important. Lai et al. (5) demonstrated that children with CF whose WFL decreased in the first 2 years of life had worse lung disease at age 6 than children whose WFL increased or remained stable in the first 2 years of life, but small numbers precluded the ability to assess changes in growth at other time points.
To maximize the potential benefits of NBS, it is necessary to understand if efforts to achieve and maintain the CFF goal of WFL and BMI >50th percentile in early life are associated with later lung health. The CFF Patient Registry (CFFPR) is a unique database with data on over 45,000 people with CF in the US, enabling longitudinal assessments of growth trajectories in individual patients. We hypothesized that increases in growth indices in early life in individual patients would be associated with higher FEV1 at age 6 than growth indices that were stable or decreasing. Our objectives for this analysis were to: characterize early life growth trajectories in children with CF, determine if these trajectories are associated with FEV1 at age 6-7 years, and determine if the timing of changes affects the relationship between growth trajectories and FEV1 at age 6-7 years.
METHODS
The CFFPR contains data on demographics, cystic fibrosis transmembrane conductance regulator (CFTR) genotype, growth, FEV1, microbiology, therapies, hospitalizations, and complications.(6) Data were entered into the CFFPR quarterly from 1994 to 2002 and at every encounter beginning in 2003. For the current analysis, we obtained data for children with CF born between 1994 and 2005, diagnosed before age 2 years, and followed in the CFFPR through at least age 7 years. Subjects were excluded if they had a solid organ transplantation prior to age 7. Age was defined as the age on December 31st. We compared mean FEV1 % predicted at 6-7 years of age with changes in growth indices (WFL for ages 0-≤2 years and BMI for ages 2-6 years). FEV1 % predicted was calculated as the average of the maximum FEV1 value for each quarter in the year that each child was age 7 on December 31st using the Global Lung Initiative (GLI) reference equations.(7) Quarters that did not have data were not included in the average.
Annualized values for WFL and BMI were calculated for each year between ages 0 and 6 years from the average of the maximum WFL and BMI percentiles from each quarter. Changes in annualized WFL and BMI percentiles were calculated based on the Center for Disease Control (CDC) growth charts (8) between ages 0 and 6 years. The median within-patient difference between WFL percentiles at age 2 years and BMI percentiles at age 2 years was only 0.2 percentile, so WFL and BMI percentiles were considered to be a continuous variable (WFL-BMI). Growth before age 6 years was classified into the following mutually exclusive categories: annualized WFL-BMI always above the 50th percentile, i.e., always meeting the CFF goal; annualized WFL-BMI that increased >10 percentile points from the first year in the study to age 6 years; annualized WFL-BMI that was stable, i.e., <10 percentile increase or decrease from the first year in the study to age 6 years; or annualized WFL-BMI that decreased >10 percentile points from the first year in the study to age 6 years. For subjects in the last three categories, at least one annualized measurement of WFL-BMI fell below the 50th percentile. Subjects with annualized WFL-BMI always >50th percentile could have increases or decreases in WFL-BMI, as long as the annualized measurements of WFL-BMI always remained above the 50th percentile. In contrast, a subject would be included in the increased >10 percentile points category if their initial WFL was above the 50th percentile, followed by at least one year with WFL-BMI below the 50th percentile, as long as the final BMI was at least 10 percentile points higher than the initial WFL percentile. A missing category was also created for those with only one year with a recorded WFL or BMI. These categories were chosen to represent growth patterns that clinicians may seek to encourage (WFL-BMI always >50th percentile, increasing, or stable) or avoid (decreasing WFL-BMI).
Chi squared and t-tests were used to compare proportions and means. Bivariate analyses were conducted using t-tests and analysis of variance to assess the statistical significance of the observed differences in means. Multivariable linear regression was used to test for an association between the categories of growth trajectories and FEV1 % predicted at age 6-7 years. The following potential confounders were included a priori based on a review of the literature for factors associated with growth and/or FEV1: gender, race (white as compared to non-white), insurance status (ever a Medicaid recipient as compared to never), pancreatic enzyme replacement therapy as a surrogate for pancreatic status (patients were categorized as pancreatic insufficient if they were prescribed pancreatic enzyme replacement therapy before age 6 years), and age of first reported Pseudomonas aeruginosa infection (0-2 years, 3-6 years, and never infected).(9-14)
To test the robustness of our results, we performed several sensitivity analyses. We tested the effect of adding the following variables individually to our final multivariable linear regression model: (1) height percentile at age 6 years, (2) CF center (i.e., clustering by site), and (3) the mean annual frequency of pulmonary exacerbations between study entry and age 6 years. In addition, we tested replacing Medicaid insurance with maternal education and White/non-White with Hispanic/non-Hispanic. To address changes in the pattern of diagnosis after the gradual introduction of NBS during this time period (which may identify more subjects with pancreatic sufficiency), we (1) repeated our analysis, restricted to subjects with class I-III cystic fibrosis transmembrane conductance regulator (CFTR) mutations (i.e., those associated with pancreatic insufficiency) or those who entered the study during the first year of life, (2) compared the categories of growth trajectories according to mode of diagnosis (NBS as compared to other modes), and (3) evaluated for effect modification by mode of diagnosis (NBS as compared to other modes). To examine the effect of obesity, we added a category for WFL-BMI always >85th percentile. For subjects whose WFL-BMI increased >10 percentile points, we examined FEV1 % predicted at age 6-7 years according to the age at which they first reached and maintained a WFL or BMI >50th percentile. For subjects whose WFL-BMI decreased >10 percentile points, we examined the FEV1 % predicted at age 6-7 years according to the lowest annual WFL or BMI percentile. Using the same growth trajectory categories, we tested for an association with mean FEV1 % predicted at ages 6-7 years with alternative predictors, including (1) annualized weight for age percentile, (2) annualized length/height for age percentile, (3) annualized BMI for age percentile only (i.e., excluding ages 0-1), and (4) annualized WFL for age only (i.e., excluding ages 3-6 years) Finally, we tested for an association between the categories of growth trajectories and alternative outcomes, including: (1) mean forced vital capacity (FVC) % predicted at ages 6-7 years and (2) best FEV1 % predicted at ages 6-7 years. Statistical analyses were performed using SAS 9.3 (SAS Institute, Inc, Cary, North Carolina). P values of <0.05 were considered significant. No adjustments were made for multiple comparisons. The Seattle Children’s Hospital Institutional Review Board approved the study.
RESULTS
Cohort characteristics
There were 10,916 children in the CFFPR born between 1994 and 2005 that were diagnosed with CF (Figure 1). Of these, there were 2,765 who were diagnosed after their second birthday, 593 not tracked in the CFFPR at age 7 years, 139 who died before age 7 years, 3 who received an organ transplant prior to age 7 years, and 2,479 who did not have measurements of FEV1 by age 7 years. Thus, there were 6,805 total eligible subjects for this study. Baseline demographic data and clinical status for each category of growth trajectory are shown in Table 1. There were 120 (1.8%) children who met inclusion criteria but only had WFL or BMI available in one year and thus did not have a defined growth trajectory. NBS was not widespread in the US until after 2004. Because we restricted our analysis to children diagnosed before age 2 years, the percent of patients with meconium ileus is higher than in the general CF population. In comparison to the other growth trajectory categories, the group with annualized WFL-BMI always >50th percentile had more children who were male, diagnosed after the first 3 months of life, born in 2002-2005, diagnosed after “other” presenting symptoms, and no infection with Pseudomonas aeruginosa before age 6 years. This group also had fewer children born in 1994-1997 or diagnosed after meconium ileus or failure to thrive. Figure 2 (available at www.jpeds.com) illustrates the mean WFL-BMI percentile for each category of growth trajectory between 0 and 6 years of age. The mean (SD) FEV1 % predicted at age 6-7 years for the entire study population was diminished at 92.2 (17.8).
Figure 1.
Diagram of study cohort
Table 1.
Demographics and clinical characteristics of study cohort.
| WFL-BMI growth trajectory | ||||||
|---|---|---|---|---|---|---|
| Characteristic | Total | Always >50th percentile |
Increased >10 percentile points |
Stable | Decreased >10 percentile points |
Missing |
| N (%) | N (%) | N (%) | N (%) | N (%) | N (%) | |
| 6805 | 1323 (19.4) | 3205 (47.1) | 1184 (17.4) | 973 (14.3) | 120 (1.8) | |
| Male | 3424 (50.3) | 744 (56.2) | 1559 (48.6) | 578 (48.8) | 476 (48.9) | 67 (55.8) |
| Age at Diagnosis | ||||||
| <3 months | 3831 (56.3) | 650 (49.1) | 1863 (58.1) | 664 (56.1) | 595 (61.2) | 59 (49.2) |
| 3 to <6 months | 1286 (18.9) | 250 (18.9) | 598 (18.7) | 227 (19.2) | 179 (18.4) | 32 (26.7) |
| 6 months to <1 year | 939 (13.8) | 208 (15.7) | 446 (13.9) | 156 (13.2) | 113 (11.6) | 16 (13.3) |
| 1 to <2 years | 749 (11.0) | 215 (16.2) | 298 (9.3) | 137 (11.6) | 86 (8.8) | 13 (10.8) |
| Birth cohort | ||||||
| 1994-1997 | 2256 (33.2) | 362 (27.4) | 1118 (34.9) | 413 (34.9) | 313 (32.2) | 50 (41.7) |
| 1998-2001 | 2261 (33.2) | 419 (31.7) | 1148 (35.8) | 374 (31.6) | 291 (29.9) | 29 (24.2) |
| 2002-2005 | 2288 (33.6) | 542 (41.0) | 939 (29.3) | 397 (33.5) | 369 (37.9) | 41 (34.2) |
| Race/ethnicity* | ||||||
| Caucasian | 6412 (94.2) | 1253 (94.7) | 3037 (94.8) | 1108 (93.6) | 905 (93.0) | 109 (90.8) |
| Black | 321 (4.7) | 45 (3.4) | 157 (4.9) | 59 (5.0) | 56 (5.8) | 4 (3.3) |
| Hispanic | 587 (8.9) | 112 (8.8) | 281 (9.0) | 105 (9.2) | 65 (7.0) | 24 (20.0) |
| Mode of diagnosis | ||||||
| Meconium Ileus | 1867 (27.4) | 271 (20.5) | 970 (30.3) | 328 (27.7) | 275 (28.3) | 23 (19.2) |
| Newborn screening | 1035 (15.2) | 216 (16.3) | 444 (13.8) | 190 (16.0) | 174 (17.9) | 11 (9.2) |
| Failure to Thrive | 2253 (33.1) | 359 (27.1) | 1140 (35.6) | 422 (35.6) | 292 (30.0) | 40 (33.3) |
| Other | 1650 (24.2) | 477 (36.0) | 651 (20.3) | 244 (20.6) | 232 (23.8) | 46 (38.3) |
| CFTR genotype | ||||||
| Classes I – III | 5273 (77.5) | 965 (72.9) | 2514 (78.4) | 955 (80.6) | 761 (78.2) | 78 (65.0) |
| Classes IV, V | 325 (4.8) | 98 (7.4) | 125 (3.9) | 50 (4.2) | 40 (4.1) | 12 (10.0) |
| Other/unknown | 1048 (15.4) | 235 (17.8) | 481 (15.0) | 156 (13.2) | 154 (15.8) | 22 (18.3) |
| Not genotyped | 159 (2.3) | 25 (1.9) | 85 (2.7) | 23 (1.9) | 18 (1.8) | 8 (6.7) |
| Insurance status | ||||||
| Medicaid/state program** | 4449 (65.4) | 831 (62.8) | 2174 (67.8) | 762 (64.4) | 642 (66.0) | 40 (33.3) |
| Age at first infection with Pseudomonas aeruginosa |
||||||
| 0-2 years | 3166 (46.5) | 526 (39.8) | 1630 (50.9) | 536 (45.3) | 469 (48.2) | 5 (4.2) |
| 3-6 years | 1690 (24.8) | 343 (25.9) | 760 (23.7) | 318 (26.9) | 250 (25.7) | 19 (15.8) |
| No Infection | 1949 (28.6) | 454 (34.3) | 815 (25.4) | 330 (27.9) | 254 (26.1) | 96 (80.0) |
| Chronic CF medications** | ||||||
| Dornase alfa | 4650 (68.3) | 924 (69.8) | 2219 (69.2) | 786 (66.4) | 678 (69.7) | 43 (35.8) |
| Inhaled tobramycin | 4384 (64.4) | 804 (60.8) | 2132 (66.5) | 762 (64.4) | 656 (67.4) | 30 (25.0) |
| Azithromycin | 652 (9.6) | 141 (10.7) | 299 (9.3) | 107 (9.0) | 101 (10.4) | 4 (3.3) |
| Pancreatic enzyme replacement therapy |
6589 (96.8) | 1255 (94.9) | 3158 (98.5) | 1158 (97.8) | 952 (97.8) | 66 (55.0) |
| Nutritional Interventions** | ||||||
| H2 blockers | 1798 (26.4) | 357 (27.0) | 839 (26.2) | 302 (25.5) | 295 (30.2) | 5 (4.2) |
| Proton pump inhibitors | 2221 (32.6) | 431 (32.6) | 1064 (33.2) | 368 (31.1) | 351 (36.1) | 6 (5.0) |
| Supplemental oral feedings | 3951 (58.1) | 780 (59.0) | 1838 (57.4) | 701 (59.2) | 604 (62.1) | 28 (23.3) |
| Nasogastric or gastrostomy tube supplemental feedings |
1354 (19.9) | 111 (8.4) | 731 (22.8) | 282 (23.8) | 224 (23.0) | 6 (5.0) |
not mutually exclusive
Ever from study entry to age 6 years
Figure 2.
Mean WFL-BMI percentile by age for each category of growth trajectories. There were 1,323 children with annualized WFL-BMI always >50th percentile through 6 years. Among children with at least one annualized WFL-BMI <50th percentile, WFL-BMI increased >10 percentile points for 3,205, remained stable for 1,184, and decreased >10 percentile points for 973.
Association of growth trajectories with lung function at age 6-7 years
In a multivariable linear regression model, subjects with WFL-BMI always >50th percentile had the highest FEV1 % predicted at age 6-7 years (Table 2). Among subjects with at least one annualized WFL-BMI percentile <50th percentile, FEV1 % predicted for subjects whose WFL-BMI increased >10 percentile points was significantly lower than for subjects with annualized WFL-BMI always >50th percentile, but significantly higher than for subjects with WFL-BMI that was stable or decreased >10 percentile points. Subjects with WFL-BMI that decreased >10 percentile points had the lowest FEV1 % predicted; this difference was statistically significant in comparison to subjects with WFL-BMI always >50th percentile and WFL-BMI that increased >10 percentile points. The statistical significance and effect size of the differences between categories of growth trajectories were similar in the unadjusted linear regression model.
Table 2.
Multivariable linear regression model for FEV1 % predicted at ages 6-7 years associated with growth trajectory from study entry to age 6 years. Total N = 6,805 subjects.
| Mean (95% CI) Estimate for FEV1 % predicted at age 6-7 years* |
Mean (95% CI) Difference in FEV1 % predicted at age 6-7 years* |
|
|---|---|---|
| Growth trajectory | ||
| WFL-BMI always >50th percentile | 101.8 (100.1, 103.5) | Reference |
| WFL-BMI increased >10 percentile points | 98.3 (96.6, 100.0) | −3.5 (−4.6, −2.4) |
| WFL-BMI stable | 94.4 (92.6, 96.2) | −7.4 (−8.8, −6.0) |
| WFL-BMI decreased >10 percentile points | 92.9 (91.1, 94.8) | −8.9 (−10.3, −7.4) |
| Male | 94.6 (93.1, 96.0) | Reference |
| Female | 94.3 (92.8, 95.8) | −0.4 (−1.2, 0.4) |
| White | 93.2 (92.0, 94.4) | Reference |
| Non-White | 95.6 (93.6, 97.7) | 2.2 (0.5, 4.0) |
| Ever Medicaid** | 91.7 (90.2, 93.2) | Reference |
| Never Medicaid** | 97.1 (95.6, 98.6) | 5.5 (4.6, 6.4) |
| Ever on pancreatic enzyme replacement therapy** | 92.1 (91.1, 93.2) | Reference |
| Never on pancreatic enzyme replacement therapy** | 96.7 (94.2, 99.2) | 6.6 (3.9, 9.3) |
| Never infected with P. aeruginosa | 96.6 (95.1, 98.1) | Reference |
| Acquired P. aeruginosa ages 0-2 years | 93.1 (91.5, 94.6) | −3.5 (−4.5, −2.5) |
| Acquired P. aeruginosa ages 3-6 years | 93.6 (91.9, 95.2) | −3.2 (−4.3, −2.0) |
Adjusted for all covariates listed
Ever from study entry to age 6 years
Sensitivity analyses
The magnitude and direction of the association between WFL-BMI and FEV1 % predicted at age 6-7 years were similar when we (1) individually added height percentile at age 6 years, CF Center, and the mean annual frequency of pulmonary exacerbations to our final multivariable regression model (Table 3A; available at www.jpeds.com); (2) replaced White/non-White race with Hispanic/non-Hispanic ethnicity or replaced Medicaid insurance status with maternal education; or (3) restricted our analysis to only patients with CFTR class I-III mutations or those who entered the study during the first year of life. FEV1 % predicted at age 6-7 years was higher for children diagnosed via NBS, 100.6 (98.8, 102.3)% predicted than for children diagnosed with meconium ileus, 96.6 (94.8, 98.3)% predicted, or FTT, 94.3 (92.6, 96.1)% predicted in a model with mode of diagnosis added to the final multivariable regression model. However, the interaction between mode of diagnosis and categories of growth trajectories was not significant (p value=0.36).
Table 3.
Estimates of mean (95% CI) pulmonary function at ages 6-7 years from multivariable linear regression models including sensitivity analyses with additional variables added individually to the final multivariable regression model, alternative predictors used in the final multivariable regression model, and alternative outcomes.
| Growth trajectory | ||||
|---|---|---|---|---|
| Always >50th percentile | Increased >10 percentile points |
Stable | Decreased >10 percentile points |
|
| Final regression model | 101.8 (100.1, 103.5) | 98.3 (96.6, 100.0) | 94.4 (92.6, 96.2) | 92.9 (91.1, 94.8) |
| + height at age 6 years | 99.7 (97.9, 101.4) | 96.6 (94.9, 98.2) | 92.8 (91.0, 94.6) | 91.4 (89.5, 93.2) |
| + CF center | 98.6 (96.7, 100.4) | 95.3 (93.5, 97.0) | 91.9 (90.0, 93.8) | 90.3 (88.3, 92.3) |
| + mean annual pulmonary exacerbation frequency |
97.0 (95.4, 98.5) | 94.6 (93.1, 96.1) | 91.4 (89.7, 93.0) | 89.8 (88.1, 91.5) |
| Replace White/non-White with Hispanic/non- Hispanic ethnicity |
99.3 (97.6, 101.0) | 95.8 (94.2, 97.4) | 91.9 (90.2, 93.7) | 90.3 (88.5, 92.1) |
| Replace Medicaid with maternal education | 99.9 (97.8, 102.0) | 96.7 (94.7, 98.7) | 92.7 (90.5, 94.9) | 92.0 (89.7, 94.3) |
| Restricted to CFTR mutation classes I-III only | 98.8 (97.3, 100.3) | 94.9 (93.6, 96.2) | 91.1 (89.6, 92.7) | 89.4 (87.8, 91.0) |
| Restricted to children entering during first year | 98.7 (97.4, 100.1) | 95.1 (94.0, 96.2) | 91.3 (89.9, 92.7) | 89.9 (88.4, 91.3) |
| Alternative predictors | ||||
| Weight for age percentile | 102.8 (100.8, 104.7) | 99.1 (97.4, 100.7) | 93.9 (92.2, 95.6) | 94.7 (92.6, 96.7) |
| Length/height for age percentile | 101.7 (99.8, 103.6) | 98.4 (96.7, 100.1) | 95.7 (94.0, 97.4) | 96.5 (94.7, 98.3) |
| BMI percentile for age only (ages 2-6 years) | 101.3 (99.7, 102.9) | 97.2 (95.5, 98.8) | 95.1 (93.3, 96.8) | 92.5 (90.8, 94.) |
| WFL percentile for age only (ages 0-1 years) | 100.8 (98.3, 103.3) | 97.2 (94.7, 99.7) | 95.4 (92.8, 98.1) | 94.7 (91.9, 97.4) |
| Alternative outcomes | ||||
| Mean FVC % predicted at ages 6-7 years | 106.1 (104.5, 107.7) | 102.5 (100.9, 104.1) | 98.6 (96.9, 100.2) | 97.7 (96.0, 99.5) |
| Best FEV1 % predicted at ages 6-7 years | 111.6 (109.9, 113.3) | 108.4 (106.8, 110.0) | 104.5 (102.8, 106.3) | 103.0 (101.2, 104.8) |
Of the 1,323 subjects with WFL-BMI always >50th percentile, 98 (7.4%) maintained a WFL-BMI >85th percentile. In the final multivariable regression model, children with WFL-BMI always >85th percentile had an adjusted mean (95% CI) FEV1 % predicted at age 6-7 of 103.6 (99.9, 107.2). This was similar to children who maintained WFL-BMI between the 50th and 85th percentiles [101.7 (99.8, 103.4)].
Among the subjects whose WFL-BMI increased >10 percentile points, we compared FEV1 at age 6-7 years according to the age at which they first achieved and maintained WFL-BMI >50th percentile (Figure 3A). The FEV1 at age 6-7 years was higher for subjects who achieved this goal at younger ages. This difference was statistically significant between subjects who maintained WFL-BMI >50th percentile from age 1 year, as compared to subjects who intermittently or never reached this goal, or reached it only at age 6 years. In subjects whose WFL-BMI decreased >10 percentile points, FEV1 % predicted at age 6-7 years was associated with the lowest annual WFL-BMI percentile reached before age 6 years (Figure 3B). Among subjects whose lowest WFL-BMI was between the 0 and 9th percentiles, the mean age was 3.6 years at this nadir; for subjects who had their lowest WFL-BMI between the 40th and 49th percentiles, the mean age was 4.5 years.
Figure 3.
Mean FEV1 % predicted at ages 6-7 years associated with (A) age at which subjects whose WFL-BMI increased >10 percentile points first reached and maintained WFL-BMI >50th percentile and (B) the lowest annualized WFL-BMI percentile for subjects whose WFL-BMI decreased >10 percentile points. The vertical bars represent the 95% confidence intervals.
Using the same growth trajectory categories, the magnitude and direction of the association with mean FEV1 % predicted at ages 6-7 years was similar when we used annualized weight for age percentile, annualized length/height for age percentile, annualized BMI for age percentile only (i.e., excluding ages 0-1), or annualized WFL for age only (i.e., excluding ages 3-6 years) as predictors (Table 3B). Finally, results were similar between the categories of growth trajectories and mean forced vital capacity (FVC) % predicted at ages 6-7 years and the best FEV1 % predicted at ages 6-7 years (Table 3C).
DISCUSSION
This study demonstrates that individual growth trajectories in the first few years of life in children with CF are associated with clinically meaningful differences in FEV1 % predicted at age 6 years. Our results support the CFF goal of maintaining WFL-BMI >50th percentile throughout childhood, as children who met this goal through age 6 had the best FEV1 % predicted at age 6-7. For children who did not maintain the CFF goal throughout the first 6 years, it is encouraging that those who had increases in WFL-BMI had higher FEV1 % predicted at age 6-7 than children who had stable or worsening WFL-BMI. Furthermore, among children who had increases in WFL-BMI, FEV1 % predicted at age 6-7 was higher the younger the children were when they reached and maintained the CFF goal, and among children who had worsening WFL-BMI, FEV1 % predicted at age 6-7 was lower according to the lowest WFL-BMI reached before age 6. These results suggest that interventions that result in improved nutrition in the first few years of life may have important effects on severity of lung disease later in life.
Previous studies have identified a correlation between nutrition measures at ages 2, 3, and 4 years with better pulmonary function test results later in life, and less morbidity and mortality.(3-5, 15) A recent study that analyzed secular trends in growth and pulmonary function in children with CF born between 1994 and 2012 and tracked in the CFFPR demonstrated increases in mean height-for-age percentiles and FEV1 z-scores in children born more recently.(16) FEV1/FVC ratios did not increase over this time period, leading the authors to speculate that the improvements in FEV1 may be a result of improved lung growth due to increased attention to nutrition in early life. Our study supports this hypothesis by demonstrating that improvements in nutrition within a given child are associated with higher FEV1 % predicted at age 6-7 years, even when adjusting for gender, race, socioeconomic status, pancreatic status, and infection with P. aeruginosa. Our sensitivity analysis also demonstrated an association between improvements in nutrition and FVC % predicted at age 6-7 years and that within a given growth trajectory category, taller children had better FEV1 % predicted. Differences in FEV1 % predicted at age 6 years are likely to affect life-long outcomes, as higher FEV1 % predicted at age 6 years is associated with later pulmonary outcomes.(17) These results should encourage providers and families that efforts to achieve improvements in nutrition early in life are associated with later improvements in lung health.
There are several important aspects to consider in nutritional interventions in young children with CF. Problematic feeding behaviors, which can occur more frequently in children with CF and may make gaining weight difficult, may need to be addressed.(18-20) A recent study that made use of CFFPR data demonstrated that CF Centers in the highest quartile for mean BMI percentiles prescribed significantly higher enzyme doses, and had more patients using acid blockers and nasogastric or gastrostomy tube feeds than CF Centers in the lowest quartiles for mean BMI percentiles.(21) These results indicate that more aggressive nutritional interventions may be important in improving nutritional outcomes. More broadly, systematic approaches to improving nutritional status have been demonstrated to be sustainable over time in different CF Centers.(22) This is especially pertinent as nearly half of US CF Centers have a median WFL-BMI in the first 6 years of life below the 50th percentile.(6)
Our study confirmed and extends prior findings by demonstrating that subjects whose WFL or BMI was >50th percentile had the best FEV1 at age 6-7 years.(3, 4) These findings support the CFF guidelines that all children maintain BMI >50th percentile.(2, 23) The importance of avoiding growth faltering early in life is highlighted by the comparison of subjects who entered the study with a WFL <50th percentile, but who eventually achieved a WFL-BMI >50th percentile: children who reached this goal before age 2 years had the highest FEV1 at age 6 years. These results suggest the importance of early growth in establishing a life course of progression of lung disease. Lai et al. showed that subjects who had early growth faltering and malnutrition but who recovered birth weight z-scores by age 2 years had better FEV1 at age 6 years, regardless of growth patterns at ages 2-6 years.(5) A more recent analysis of CFFPR data showed that low weight and length at age 4 months was the best predictor of poor growth at 24 months.(24) The CF Foundation recommends frequent evaluation in CF clinics for all infants with CF. This frequent monitoring provides a significant opportunity to intervene and avoid malnutrition and growth faltering that may be present as early as the first 2 weeks of life for children with CF.(25) In a comparison of infant care practices between CF clinics with the highest and lowest mean FEV1 in 1994-1999, centers who had the highest mean FEV1 evaluated infants more often, both for routine and sick visits, obtained more respiratory cultures, and used more oral and IV therapies.(26)
Another key finding of our study that should be emphasized is that children whose WFL-BMI was stable but below the 50th percentile had lower FEV1 % predicted at age 6 years than children who had increasing WFL-BMI. Even though not all children with CF can achieve WFL and BMI >50th percentile due to genetic [e.g., familial growth patterns or insulin-like growth factor (IGF)-1 deficiency (27)] and/or environmental factors [e.g., in utero or early life tobacco smoke exposure (28, 29)] that may limit potential growth, our findings suggest that there may be risks (i.e., lower future lung function) associated with settling for stable nutrition rather than increasing WFL-BMI and/or achieving WFL-BMI >50th percentile.
There are several limitations to our study. First, we would point out that spirometry is the only measure of lung disease included in the CFFPR. Recent studies have made clear that even the children with normal FEV1 in this analysis are likely to show evidence of lung disease if evaluated by more sensitive measures such as the lung clearance index (LCI) and chest CT, even in the presence of a normal FEV1.(30, 31) Second, as with all observational studies, we must be careful to point out that the association we found between early childhood growth trajectories and lung function does not prove causation. It is possible that early silent underlying lung disease may impact nutritional parameters. Furthermore, our study does not demonstrate whether different growth trajectories occur because of direct nutritional or pulmonary interventions, or due to underlying factors contributing to both nutritional status and lung health, such as gene modifiers. Ongoing prospective studies that closely monitor growth, nutrition, and pulmonary status may provide stronger evidence of causation, as would clinical intervention trials.
The subjects in the current study were born before widespread NBS, and before many current standard CF therapies were available, limiting its generalizability to more recent birth cohorts. We should also point out that our study does not address the appropriateness of alternative nutrition goals, i.e., other than a WFL or BMI >50th percentile, although these have been chosen because of previous research showing that patients who maintain these levels of nutrition have the best pulmonary outcomes.(2) In our sensitivity analysis, we compared mean FEV1 % predicted at age 6-7 years by height at age 6 and found that increased height was significantly associated with higher FEV1 % predicted. Additionally, even within each growth trajectory category, taller children had higher mean FEV1 % predicted than shorter children. Finally, the CFFPR has limited data to address the genetic potential of subjects, such as birth weights and parental data, so we could not consider this in our analysis.
In conclusion, children with WFL-BMI always >50th percentile between the diagnosis of CF and 6 years, in keeping with current CFF guidelines, had the highest mean FEV1 % predicted at age 6-7 years. Among those with WFL-BMI <50th percentile in at least one year before age 6, those children whose WFL-BMI increased >10 percentile points by age 6 years had higher FEV1 at age 6-7 years in comparison to subjects whose WFL-BMI was stable or decreased >10 percentile points. These results suggest that interventions to improve growth in the first few years of life within an individual patient with CF may result in improved lung health. These early life interventions may have life-long effects and should encourage providers to promote optimal nutrition frequently and consistently.
ACKNOWLEDGMENTS
The authors would like to thank Bruce Marshall, Emily Knapp, and the Cystic Fibrosis Foundation for the use of CF Foundation Patient Registry data to conduct this study. Additionally, we would like to thank the patients, care providers, and clinic coordinators at CF Centers throughout the United States for their contributions to the CF Foundation Patient Registry.
D.B.S. wrote the first draft and is supported by the Cystic Fibrosis Foundation (SANDERS11A0) and the Institute for Clinical and Translational Research (ICTR) through the NIH National Center for Advancing Translational Sciences (NCATS) grants UL1 TR000427 and KL2TR000428. P.A.F. is supported by the South Carolina Clinical & Translational Research (SCTR) Institute, with an academic home at the Medical University of South Carolina, through NIH Grant Number UL1 TR000062. The sponsors had no involvement in any aspect of this report.
D.B.S., N.M.H., M.S.S., G.S.S., M.R., P.A.F., and W.J.M have received honoraria from the CF Foundation for serving as a member of the CF Foundation Patient Registry Committee; no compensation was provided in exchange for the creation of this manuscript.
Abbreviations
- BMI
Body Mass Index
- CF
Cystic Fibrosis
- CFF
Cystic Fibrosis Foundation
- CFFPR
Cystic Fibrosis Foundation Patient Registry
- CI
Confident Interval
- FEV1
Forced Expiratory Volume in 1 Second
- FVC
Forced Vital Capacity
- NBS
Newborn Screening
- WFL
Weight For Length
REFERENCES
- 1.Gibson R, Burns J, Ramsey B. Pathophysiology and management of pulmonary infections in cystic fibrosis. Am J Respir Crit Care Med. 2003;168:918–51. doi: 10.1164/rccm.200304-505SO. [DOI] [PubMed] [Google Scholar]
- 2.Stallings VA, Stark LJ, Robinson KA, Feranchak AP, Quinton H, Clinical Practice Guidelines on Growth and Nutrition Subcommittee Ad Hoc Working Group et al. Evidence-based practice recommendations for nutrition-related management of children and adults with cystic fibrosis and pancreatic insufficiency: results of a systematic review. J Am Diet Assoc. 2008;108:832–9. doi: 10.1016/j.jada.2008.02.020. [DOI] [PubMed] [Google Scholar]
- 3.Konstan M, Butler S, Wohl M, Stoddard M, Matousek R, Wagener J, et al. Growth and nutritional indexes in early life predict pulmonary function in cystic fibrosis. J Pediatr. 2003;142:624–30. doi: 10.1067/mpd.2003.152. [DOI] [PubMed] [Google Scholar]
- 4.Yen EH, Quinton H, Borowitz D. Better nutritional status in early childhood is associated with improved clinical outcomes and survival in patients with cystic fibrosis. J Pediatr. 2013;162:530–5.e1. doi: 10.1016/j.jpeds.2012.08.040. [DOI] [PubMed] [Google Scholar]
- 5.Lai H, Shoff S, Farrell P, Wisconsin Cystic Fibrosis Neonatal Screening Group Recovery of birth weight z score within 2 years of diagnosis is positively associated with pulmonary status at 6 years of age in children with cystic fibrosis. Pediatrics. 2009;123:714–22. doi: 10.1542/peds.2007-3089. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.CF Foundation 2012 Patient Registry Annual Data Report. 2013.
- 7.Quanjer PH, Stanojevic S, Cole TJ, Baur X, Hall GL, Culver BH, et al. Multi-ethnic reference values for spirometry for the 3-95-yr age range: the global lung function 2012 equations. Eur Respir J. 2012;40:1324–43. doi: 10.1183/09031936.00080312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kuczmarski R, Ogden C, Grummer-Strawn L, Flegal K, Guo S, Wei R, et al. CDC growth charts: United States. Adv Data. 2000:1–27. [PubMed] [Google Scholar]
- 9.Corey M, Edwards L, Levison H, Knowles M. Longitudinal analysis of pulmonary function decline in patients with cystic fibrosis. J Pediatr. 1997;131:809–14. doi: 10.1016/s0022-3476(97)70025-8. [DOI] [PubMed] [Google Scholar]
- 10.Emerson J, Rosenfeld M, McNamara S, Ramsey B, Gibson R. Pseudomonas aeruginosa and other predictors of mortality and morbidity in young children with cystic fibrosis. Pediatr Pulmonol. 2002;34:91–100. doi: 10.1002/ppul.10127. [DOI] [PubMed] [Google Scholar]
- 11.Braun AT, Farrell PM, Ferec C, Audrezet MP, Laxova A, Li Z, et al. Cystic fibrosis mutations and genotype-pulmonary phenotype analysis. Journal of Cystic Fibrosis. 2006;5:33–41. doi: 10.1016/j.jcf.2005.09.008. [DOI] [PubMed] [Google Scholar]
- 12.Quittner AL, Schechter MS, Rasouliyan L, Haselkorn T, Pasta DJ, Wagener JS. Impact of socioeconomic status, race, and ethnicity on quality of life in patients with cystic fibrosis in the United States. Chest. 2010;137:642–50. doi: 10.1378/chest.09-0345. [DOI] [PubMed] [Google Scholar]
- 13.Hamosh A, FitzSimmons SC, Macek M, Knowles MR, Rosenstein BJ, Cutting GR. Comparison of the clinical manifestations of cystic fibrosis in black and white patients. J Pediatr. 1998;132:255–9. doi: 10.1016/s0022-3476(98)70441-x. [DOI] [PubMed] [Google Scholar]
- 14.Schechter M, Shelton B, Margolis P, Fitzsimmons S. The association of socioeconomic status with outcomes in cystic fibrosis patients in the United States. Am J Respir Crit Care Med. 2001;163:1331–7. doi: 10.1164/ajrccm.163.6.9912100. [DOI] [PubMed] [Google Scholar]
- 15.VanDevanter DR, Wagener JS, Pasta DJ, Elkin E, Jacobs JR, Morgan WJ, et al. Pulmonary outcome prediction (POP) tools for cystic fibrosis patients. Pediatr Pulmonol. 2010;45:1156–66. doi: 10.1002/ppul.21311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.VanDevanter DR, Pasta DJ, Konstan MW. Improvements in Lung Function and Height among Cohorts of 6-Year-Olds with Cystic Fibrosis from 1994 to 2012. J Pediatr. 2014;165:1091–7.e2. doi: 10.1016/j.jpeds.2014.06.061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Taylor-Robinson D, Whitehead M, Diderichsen F, Olesen HV, Pressler T, Smyth RL, et al. Understanding the natural progression in %FEV1 decline in patients with cystic fibrosis: a longitudinal study. Thorax. 2012;67:860–6. doi: 10.1136/thoraxjnl-2011-200953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Powers SW, Patton SR, Byars KC, Mitchell MJ, Jelalian E, Mulvihill MM, et al. Caloric intake and eating behavior in infants and toddlers with cystic fibrosis. Pediatrics. 2002;109:E75–5. doi: 10.1542/peds.109.5.e75. [DOI] [PubMed] [Google Scholar]
- 19.Powers SW, Mitchell MJ, Patton SR, Byars KC, Jelalian E, Mulvihill MM, et al. Mealtime behaviors in families of infants and toddlers with cystic fibrosis. J Cyst Fibros. 2005;4:175–82. doi: 10.1016/j.jcf.2005.05.015. [DOI] [PubMed] [Google Scholar]
- 20.Powers SW, Jones JS, Ferguson KS, Piazza-Waggoner C, Daines C, Acton JD. Randomized clinical trial of behavioral and nutrition treatment to improve energy intake and growth in toddlers and preschoolers with cystic fibrosis. Pediatrics. 2005;116:1442–50. doi: 10.1542/peds.2004-2823. [DOI] [PubMed] [Google Scholar]
- 21.Haupt ME, Kwasny MJ, Schechter MS, McColley SA. Pancreatic enzyme replacement therapy dosing and nutritional outcomes in children with cystic fibrosis. J Pediatr. 2014;164:1110–5.e1. doi: 10.1016/j.jpeds.2014.01.022. [DOI] [PubMed] [Google Scholar]
- 22.Savant AP, Britton LJ, Petren K, McColley SA, Gutierrez HH. Sustained improvement in nutritional outcomes at two paediatric cystic fibrosis centres after quality improvement collaboratives. BMJ quality & safety. 2014;23(Suppl 1):i81–9. doi: 10.1136/bmjqs-2013-002314. [DOI] [PubMed] [Google Scholar]
- 23.Borowitz D, Robinson KA, Rosenfeld M, Davis SD, Sabadosa KA, Spear SL, et al. Cystic Fibrosis Foundation evidence-based guidelines for management of infants with cystic fibrosis. J Pediatr. 2009;155:S73–93. doi: 10.1016/j.jpeds.2009.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Heltshe SL, Borowitz DS, Leung DH, Ramsey B, Mayer-Hamblett N. Early attained weight and length predict growth faltering better than velocity measures in infants with CF. J Cyst Fibros. 2014;13:723–9. doi: 10.1016/j.jcf.2014.05.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Marcus M, Sondel S, Farrell P, Laxova A, Carey P, Langhough R, et al. Nutritional status of infants with cystic fibrosis associated with early diagnosis and intervention. Am J Clin Nutr. 1991;54:578–85. doi: 10.1093/ajcn/54.3.578. [DOI] [PubMed] [Google Scholar]
- 26.Padman R, McColley SA, Miller DP, Konstan MW, Morgan WJ, Schechter MS, et al. Infant care patterns at epidemiologic study of cystic fibrosis sites that achieve superior childhood lung function. Pediatrics. 2007;119:e531–7. doi: 10.1542/peds.2006-1414. [DOI] [PubMed] [Google Scholar]
- 27.Sermet-Gaudelus I, Souberbielle JC, Azhar I, Ruiz JC, Magnine P, Colomb V, et al. Insulin-like growth factor I correlates with lean body mass in cystic fibrosis patients. Arch Dis Child. 2003;88:956–61. doi: 10.1136/adc.88.11.956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Rosenfeld M, Emerson J, McNamara S, Joubran K, Retsch-Bogart G, Graff GR, et al. Baseline characteristics and factors associated with nutritional and pulmonary status at enrollment in the cystic fibrosis EPIC observational cohort. Pediatr Pulmonol. 2010;45:934–44. doi: 10.1002/ppul.21279. [DOI] [PubMed] [Google Scholar]
- 29.Rubin BK. Exposure of children with cystic fibrosis to environmental tobacco smoke. N Engl J Med. 1990;323:782–8. doi: 10.1056/NEJM199009203231203. [DOI] [PubMed] [Google Scholar]
- 30.Gustafsson P, De Jong P, Tiddens H, Lindblad A. Multiple-breath inert gas washout and spirometry versus structural lung disease in cystic fibrosis. Thorax. 2008;63:129–34. doi: 10.1136/thx.2007.077784. [DOI] [PubMed] [Google Scholar]
- 31.de Jong P, Lindblad A, Rubin L, Hop W, de Jongste J, Brink M, et al. Progression of lung disease on computed tomography and pulmonary function tests in children and adults with cystic fibrosis. Thorax. 2006;61:80–5. doi: 10.1136/thx.2005.045146. [DOI] [PMC free article] [PubMed] [Google Scholar]