Baseline Ultrasound and Clinical Correlates in Children with Cystic Fibrosis

on behalf of the; Cystic Fibrosis Liver Disease Network (CFLD NET); Cystic Fibrosis Liver Disease Network (CFLD NET)

doi:10.1016/j.jpeds.2015.06.062

Baseline Ultrasound and Clinical Correlates in Children with Cystic Fibrosis

on behalf of the, Cystic Fibrosis Liver Disease Network (CFLD NET), Cystic Fibrosis Liver Disease Network (CFLD NET)

School of Medicine

Research output: Contribution to journal › Article › peer-review

Abstract

Objective To investigate the relationship between abdominal ultrasound findings and demographic, historical, and clinical features in children with cystic fibrosis (CF). Study design Children age 3-12 years with CF without known cirrhosis, were enrolled in a prospective, multicenter study of ultrasound to predict hepatic fibrosis. Consensus ultrasound patterns were assigned by 3 radiologists as normal, heterogeneous, homogeneous, or cirrhosis. Data were derived from direct collection and US or Toronto CF registries. χ² or ANOVA were used to compare variables among ultrasound groups and between normal and abnormal. Logistic regression was used to study risk factors for having abnormal ultrasound. Results Findings in 719 subjects were normal (n = 590, 82.1%), heterogeneous (64, 8.9%), homogeneous (41, 5.7%), and cirrhosis (24, 3.3%). Cirrhosis (P = .0004), homogeneous (P < .0001), and heterogeneous (P = .03) were older than normal. More males were heterogeneous (P = .001). More heterogeneous (15.0%, P = .009) and cirrhosis (25.0%, P = .005) had CF-related diabetes or impaired glucose tolerance vs normal (5.4%). Early infection with Pseudomonas aeruginosa (<2 years old) was associated with a lower risk (OR 0.42, P = .0007) of abnormal. Ursodeoxycholic acid use (OR 3.69, P < .0001) and CF-related diabetes (OR 2.21, P = .019) were associated with increased risk of abnormal. Conclusions Unsuspected cirrhosis is seen in 3.3% of young patients with CF, heterogeneous in 8.9%. Abnormal ultrasound is associated with CF-related diabetes, and early P aeruginosa is associated with normal ultrasound. Prospective assessment of these risk factors may identify potential interventional targets. Trial registration ClinicalTrials.gov: NCT01144507.

Original language	English (US)
Pages (from-to)	862-868.e2
Journal	Journal of Pediatrics
Volume	167
Issue number	4
DOIs	https://doi.org/10.1016/j.jpeds.2015.06.062
State	Published - Oct 1 2015

ASJC Scopus subject areas

Pediatrics, Perinatology, and Child Health

Access to Document

10.1016/j.jpeds.2015.06.062

Cite this

@article{1436e2bb52824fb49fb07cec02dbf71c,

title = "Baseline Ultrasound and Clinical Correlates in Children with Cystic Fibrosis",

abstract = "Objective To investigate the relationship between abdominal ultrasound findings and demographic, historical, and clinical features in children with cystic fibrosis (CF). Study design Children age 3-12 years with CF without known cirrhosis, were enrolled in a prospective, multicenter study of ultrasound to predict hepatic fibrosis. Consensus ultrasound patterns were assigned by 3 radiologists as normal, heterogeneous, homogeneous, or cirrhosis. Data were derived from direct collection and US or Toronto CF registries. χ2 or ANOVA were used to compare variables among ultrasound groups and between normal and abnormal. Logistic regression was used to study risk factors for having abnormal ultrasound. Results Findings in 719 subjects were normal (n = 590, 82.1%), heterogeneous (64, 8.9%), homogeneous (41, 5.7%), and cirrhosis (24, 3.3%). Cirrhosis (P = .0004), homogeneous (P < .0001), and heterogeneous (P = .03) were older than normal. More males were heterogeneous (P = .001). More heterogeneous (15.0%, P = .009) and cirrhosis (25.0%, P = .005) had CF-related diabetes or impaired glucose tolerance vs normal (5.4%). Early infection with Pseudomonas aeruginosa (<2 years old) was associated with a lower risk (OR 0.42, P = .0007) of abnormal. Ursodeoxycholic acid use (OR 3.69, P < .0001) and CF-related diabetes (OR 2.21, P = .019) were associated with increased risk of abnormal. Conclusions Unsuspected cirrhosis is seen in 3.3% of young patients with CF, heterogeneous in 8.9%. Abnormal ultrasound is associated with CF-related diabetes, and early P aeruginosa is associated with normal ultrasound. Prospective assessment of these risk factors may identify potential interventional targets. Trial registration ClinicalTrials.gov: NCT01144507.",

author = "{on behalf of the} and {Cystic Fibrosis Liver Disease Network (CFLD NET)} and {Cystic Fibrosis Liver Disease Network (CFLD NET)} and Leung, {Daniel H.} and Wen Ye and Molleston, {Jean P.} and Alexander Weymann and Simon Ling and Paranjape, {Shruti M.} and Rene Romero and Schwarzenberg, {Sara Jane} and Joseph Palermo and Alonso, {Estella M.} and Murray, {Karen F.} and Marshall, {Bruce C.} and Sherker, {Averell H.} and Siegel, {Marilyn J.} and Rajesh Krishnamurthy and Roger Harned and Boaz Karmazyn and Magee, {John C.} and Narkewicz, {Michael R.} and Alonso, {Estella M.} and Nicholas, {Jennifer L.} and Elizabeth Kaurs and Narkewicz, {Michael R.} and Sokol, {Ronald J.} and Roger Harned and Susanna Burr and Rene Romero and Jay Freeman and Adina Alazraki and Ellen Patrick and Eric Hunter and Ling, {Simon C.} and Oscar Navarro and Ling, {Julie P.} and Palermo, {Joe J.} and Alex Towbin and Andrea Ferris and Julie Denlinger and Molleston, {Jean P.} and Bozic, {Molly A.} and Girish Subbarao and Boaz Karmazyn and Ann Klipsch and Paranjape, {Shruti M.} and Wikrom Karnsakul and Benson, {Jane E.} and Callahan, {Karen A.} and Kim Kafka and Murray, {Karen F.} and Ron Gibson",

note = "Funding Information: In this prospective study, abnormalities in ultrasound patterns are present in 18% of children with pancreatic insufficient CF, which is in agreement with data from Patriquin et al 8 obtained >15 years ago. Although the distribution of ultrasound abnormalities in our study is similar, the overall prevalence of heterogeneous ultrasound findings in our cohort is higher than that previously reported. 7,10 Unexpectedly, an ultrasound pattern of cirrhosis was found in 3.3% of subjects who had no clinical evidence of liver disease such as portal hypertension (ie, splenomegaly, ascites) or thrombocytopenia. This is consistent with prior reports and suggests that cirrhosis may present early in life. The possibility of detecting CF cirrhosis in its earliest stages, prior to the development of clinical signs or symptoms, has implications for establishing important clinical endpoints that could be tracked when therapeutic drugs become available. The PUSH Study illustrates that sonographic echotexture abnormalities are common in patients with CF, but their ability to predict clinical outcomes such as the risk for the development of cirrhosis, will only be determined through long-term follow-up. At study entry, a heterogeneous pattern was demonstrated in 8.9%, and 5.7% had a homogeneous pattern. It is also possible that the ultrasound findings will vary over time, with intermittent hepatic injury and repair, or because of as yet to be determined factors. Notably, subjects with an abnormal ultrasound were older and more likely to be male in the heterogeneous group than those with a normal pattern. This is consistent with a higher frequency of multilobular cirrhosis in males 11 and further suggests that there indeed may be a progression from heterogeneous to cirrhosis. A homogeneous echogenic pattern is felt to reflect hepatic steatosis. 12 In the non-CF pediatric population, a homogeneous pattern is most frequently associated with nonalcoholic fatty liver disease (NAFLD) related to obesity. 13 In our subjects, even though there was a higher BMI z-score in those with a homogeneous pattern, only 15% of those in the homogeneous group were overweight (BMI z-score >1.5). Thus, in CF, the homogeneous pattern is not exclusive to overweight children. CF-specific dietary interventions, medications, and nutritional deficiencies such as protein-calorie malnutrition or essential fatty acid and fat soluble vitamin deficiency may influence the development of the homogeneous pattern. NAFLD disproportionally affects Hispanic persons, so it was not surprising that nearly 20% of our homogeneous subjects were Hispanic. 14,15 However, there were no Hispanic subjects in our reported heterogeneous group, despite representation among the other ultrasound grades. In light of these findings, it is still not clear if homogeneous is part of the progression of heterogeneous to cirrhosis. Correlation of homogeneous ultrasound findings and liver biopsy in patients with CF has not yet been undertaken. This study allowed us to investigate potential clinical and historical factors associated with an abnormal ultrasound. Previously suggested factors for liver involvement in CF including meconium ileus, undernutrition, and worse FEV 1 were not found to be associated with abnormal ultrasound patterns in our study. 3,6,7 The age-controlled univariate analysis demonstrated that ursodeoxycholic acid use and CF-related diabetes or impaired glucose tolerance were associated with a higher frequency of an abnormal ultrasound and that early Pseudomonas infection (<2 years of age) was associated with a higher frequency of normal ultrasound. Our data show an association of abnormal ultrasound with ursodeoxycholic acid even after controlling for age. More subjects with an abnormal ultrasound were identified as having noncirrhotic liver disease in the CF registry a priori, likely secondary to documentation of elevated liver transaminases. As such, this association could be indicative of therapy for clinically suspected liver disease in the absence of detectable portal hypertension and not imply causality. Although high-dose ursodeoxycholic acid has been associated with increased morbidity, including mortality and cancer in adults with primary sclerosing cholangitis, 16,17 we are not aware of prior studies that suggest that ursodeoxycholic acid may induce ultrasound abnormalities in children with CF. Our data, however, neither confirm nor exclude this possibility, and we believe this deserves further study. Our finding of an association between CF-related diabetes or impaired glucose tolerance and an abnormal ultrasound that is independent of age is noteworthy. An association with CF-related diabetes and progressive liver disease has been previously reported. 18 Our data did not reveal any increased incidence of CF-related diabetes or impaired glucose tolerance in the homogeneous group. Thus, even though CF-related diabetes may be a marker of more severe CFTR dysfunction, it may also be a marker of other insulin associated mechanisms, distinct from that of NAFLD, that impact liver disease. Although this association could in part be related to the older age of the subjects and more frequent glycemic testing performed in this group, this trend remained after controlling for age. Most of the subjects tested for glucose metabolism were over 10 years of age, consistent with clinical guidelines. 19 Thus, even though this association should be regarded with caution as the number of subjects diagnosed with either CF-related diabetes or impaired glucose tolerance is small, there may be a higher overall prevalence of CF-related diabetes and impaired glucose tolerance than reported in this study. Meconium ileus has historically been considered a risk factor for CF liver disease (CFLD). However, a recent large, single-center analysis of 401 infants with CF over a 25-year period and other studies have found no difference in the percentage of patients with meconium ileus who developed CFLD compared with those without liver disease. 20-22 Our results suggest that a history of meconium ileus increases risk for a homogeneous ultrasound pattern. The significant association of early infection and colonization with Pseudomonas after controlling for age is the most provocative finding of this study. Many more subjects in the normal group were infected with Pseudomonas compared with the abnormal group. This may indicate a modifier gene(s) effect similar to the recent work by Li et al 23 demonstrating an association between the meconium ileus modifier polymorphism SLC6A14 and early Pseudomonas infection. This protective association could also be a marker of alteration in the gut or respiratory microbiome related to antibiotic exposure or reduction of systemic inflammation due to the early treatment of Pseudomonas infection. Changes in both the microbiome and systemic inflammation have been associated with NAFLD and other liver diseases. 24,25 It could also mean that this set of patients may have a less vigorous immune response leading to early Pseudomonas colonization but decreased immune mediated liver inflammatory injury. This phenomenon may also be secondary to early identification of Pseudomonas infection leading to closer follow-up or other unrecognized intervention that leads to a protective effect. Other imaging technologies, notably transient elastography, which measures liver stiffness, has been shown to correlate with advanced fibrosis in pediatric liver diseases, including CFLD, 26-28 but its utility in early fibrosis is unproven. Our findings demonstrate that the frequency of heterogeneous ultrasound abnormalities is higher than previously described, with 3% having unanticipated sonographic cirrhosis. The association between CF-related diabetes and impaired glucose tolerance with abnormal ultrasound patterns and the protective effect of early Pseudomonas infection suggest potential mechanisms predisposing to liver disease that merit further study. Although the ultrasound grading system for this study is rigorous, it is descriptive and may not directly correlate with severity of liver disease in CF. We recognize that no histologic correlation was available for the imaging patterns that are used in the grading system as liver biopsy was not part of this study. The findings in this study do not yet support routine monitoring with abdominal ultrasound, but ongoing clinical follow-up of this large study group will help define the utility of screening abdominal ultrasound and the clinical significance of the spectrum of findings in young children with CF as well as the potential role of ultrasound in tracking responses to newer therapies. We thank the Childhood Liver Disease Research Network (ChiLDReN) investigators, coordinators, directors, members, and families of the CF Centers at each participating site who participated and made the present work possible. We recognize Drucy Borowitz, MD, for serving as a research design and clinical consultant for this study (active member of the CF Foundation, from which she receives funding [BOROWI03CS0]). Appendix Members of CFLD NET include (principal investigator [PI], radiologist [R], site research coordinator [SC]): Ann and Robert Lurie Children's Hospital, Chicago, IL: Estella M. Alonso, MD (PI), Jennifer L. Nicholas, MD (R), Elizabeth Kaurs (SC); Children's Hospital Colorado, Aurora, CO: Michael R. Narkewicz, MD (PI), Ronald J. Sokol, MD (co-PI), Roger Harned, MD (R), Susanna Burr (SC); Emory University, Children's Heathcare of Atlanta, Atlanta, GA: Rene Romero, MD (PI), Jay Freeman, MD, Adina Alazraki, MD (R), Ellen Patrick, MD (R), Eric Hunter (SC); Hospital for Sick Children, Toronto, Canada: Simon C. Ling, MD (PI), Oscar Navarro, MD (R), Julie P. Ling (SC); Cincinnati Children's Hospital Medical Center, Cincinnati, OH; Joe J. Palermo, MD (PI), Alex Towbin, MD (R), Andrea Ferris (SC), Julie Denlinger (SC); Indiana University/Riley Hospital for Children, Indianapolis, IN: Jean P. Molleston, MD (PI), Molly A. Bozic, MD (Co-PI), Girish Subbarao, MD (Co-PI), Boaz Karmazyn, MD (R), Ann Klipsch (SC); Johns Hopkins University School of Medicine, Baltimore, MD: Shruti M. Paranjape, MD (PI), Wikrom Karnsakul, MD (Co-PI), Jane E. Benson, MD (R), Karen A. Callahan (SC), Kim Kafka (SC); Seattle Children's Hospital, Seattle, WA: Karen F. Murray (PI), Ron Gibson (Co-PI), Randolph Otto (R), Alan Genatossio (SC), Melissa Young (SC); St. Louis Children's Hospital, St. Louis, MO: Alexander Weymann (PI), Marilyn J. Siegel (R), Kathy Harris (SC); Texas Children's Hospital, Houston, TX: Daniel H. Leung (PI), Rajesh Krishnamurthy (R), Jameisha Brown (SC); University of Minnesota Masonic Children's Hospital, Minneapolis, MN: Sara Jane Schwarzenberg (PI), Denise Stacklie (SC), F. Glenn Seidel (R—now at Lucille Salter Packard Children's Hospital, Stanford, CA). Data Safety Monitoring Board oversight and program support (National Institute of Diabetes and Digestive and Kidney Diseases, Division of Digestive Diseases & Nutrition, and National Institutes of Health): Edward Doo, MD (Program Official), Averell H. Sherker, MD, FRCP(C) (Scientific Advisor for Viral Hepatitis and Liver Diseases), Sherry R. Hall, MS (Health Program Analyst), and Rebecca Torrance, RN, MS (Executive Secretary; Clinical Trials Specialist). Publisher Copyright: {\textcopyright} 2015 Elsevier Inc.",

year = "2015",

month = oct,

day = "1",

doi = "10.1016/j.jpeds.2015.06.062",

language = "English (US)",

volume = "167",

pages = "862--868.e2",

journal = "Journal of Pediatrics",

issn = "0022-3476",

publisher = "Mosby Inc.",

number = "4",

}

TY - JOUR

T1 - Baseline Ultrasound and Clinical Correlates in Children with Cystic Fibrosis

AU - on behalf of the

AU - Cystic Fibrosis Liver Disease Network (CFLD NET)

AU - Leung, Daniel H.

AU - Ye, Wen

AU - Molleston, Jean P.

AU - Weymann, Alexander

AU - Ling, Simon

AU - Paranjape, Shruti M.

AU - Romero, Rene

AU - Schwarzenberg, Sara Jane

AU - Palermo, Joseph

AU - Alonso, Estella M.

AU - Murray, Karen F.

AU - Marshall, Bruce C.

AU - Sherker, Averell H.

AU - Siegel, Marilyn J.

AU - Krishnamurthy, Rajesh

AU - Harned, Roger

AU - Karmazyn, Boaz

AU - Magee, John C.

AU - Narkewicz, Michael R.

AU - Alonso, Estella M.

AU - Nicholas, Jennifer L.

AU - Kaurs, Elizabeth

AU - Narkewicz, Michael R.

AU - Sokol, Ronald J.

AU - Harned, Roger

AU - Burr, Susanna

AU - Romero, Rene

AU - Freeman, Jay

AU - Alazraki, Adina

AU - Patrick, Ellen

AU - Hunter, Eric

AU - Ling, Simon C.

AU - Navarro, Oscar

AU - Ling, Julie P.

AU - Palermo, Joe J.

AU - Towbin, Alex

AU - Ferris, Andrea

AU - Denlinger, Julie

AU - Molleston, Jean P.

AU - Bozic, Molly A.

AU - Subbarao, Girish

AU - Karmazyn, Boaz

AU - Klipsch, Ann

AU - Paranjape, Shruti M.

AU - Karnsakul, Wikrom

AU - Benson, Jane E.

AU - Callahan, Karen A.

AU - Kafka, Kim

AU - Murray, Karen F.

AU - Gibson, Ron

N1 - Funding Information: In this prospective study, abnormalities in ultrasound patterns are present in 18% of children with pancreatic insufficient CF, which is in agreement with data from Patriquin et al 8 obtained >15 years ago. Although the distribution of ultrasound abnormalities in our study is similar, the overall prevalence of heterogeneous ultrasound findings in our cohort is higher than that previously reported. 7,10 Unexpectedly, an ultrasound pattern of cirrhosis was found in 3.3% of subjects who had no clinical evidence of liver disease such as portal hypertension (ie, splenomegaly, ascites) or thrombocytopenia. This is consistent with prior reports and suggests that cirrhosis may present early in life. The possibility of detecting CF cirrhosis in its earliest stages, prior to the development of clinical signs or symptoms, has implications for establishing important clinical endpoints that could be tracked when therapeutic drugs become available. The PUSH Study illustrates that sonographic echotexture abnormalities are common in patients with CF, but their ability to predict clinical outcomes such as the risk for the development of cirrhosis, will only be determined through long-term follow-up. At study entry, a heterogeneous pattern was demonstrated in 8.9%, and 5.7% had a homogeneous pattern. It is also possible that the ultrasound findings will vary over time, with intermittent hepatic injury and repair, or because of as yet to be determined factors. Notably, subjects with an abnormal ultrasound were older and more likely to be male in the heterogeneous group than those with a normal pattern. This is consistent with a higher frequency of multilobular cirrhosis in males 11 and further suggests that there indeed may be a progression from heterogeneous to cirrhosis. A homogeneous echogenic pattern is felt to reflect hepatic steatosis. 12 In the non-CF pediatric population, a homogeneous pattern is most frequently associated with nonalcoholic fatty liver disease (NAFLD) related to obesity. 13 In our subjects, even though there was a higher BMI z-score in those with a homogeneous pattern, only 15% of those in the homogeneous group were overweight (BMI z-score >1.5). Thus, in CF, the homogeneous pattern is not exclusive to overweight children. CF-specific dietary interventions, medications, and nutritional deficiencies such as protein-calorie malnutrition or essential fatty acid and fat soluble vitamin deficiency may influence the development of the homogeneous pattern. NAFLD disproportionally affects Hispanic persons, so it was not surprising that nearly 20% of our homogeneous subjects were Hispanic. 14,15 However, there were no Hispanic subjects in our reported heterogeneous group, despite representation among the other ultrasound grades. In light of these findings, it is still not clear if homogeneous is part of the progression of heterogeneous to cirrhosis. Correlation of homogeneous ultrasound findings and liver biopsy in patients with CF has not yet been undertaken. This study allowed us to investigate potential clinical and historical factors associated with an abnormal ultrasound. Previously suggested factors for liver involvement in CF including meconium ileus, undernutrition, and worse FEV 1 were not found to be associated with abnormal ultrasound patterns in our study. 3,6,7 The age-controlled univariate analysis demonstrated that ursodeoxycholic acid use and CF-related diabetes or impaired glucose tolerance were associated with a higher frequency of an abnormal ultrasound and that early Pseudomonas infection (<2 years of age) was associated with a higher frequency of normal ultrasound. Our data show an association of abnormal ultrasound with ursodeoxycholic acid even after controlling for age. More subjects with an abnormal ultrasound were identified as having noncirrhotic liver disease in the CF registry a priori, likely secondary to documentation of elevated liver transaminases. As such, this association could be indicative of therapy for clinically suspected liver disease in the absence of detectable portal hypertension and not imply causality. Although high-dose ursodeoxycholic acid has been associated with increased morbidity, including mortality and cancer in adults with primary sclerosing cholangitis, 16,17 we are not aware of prior studies that suggest that ursodeoxycholic acid may induce ultrasound abnormalities in children with CF. Our data, however, neither confirm nor exclude this possibility, and we believe this deserves further study. Our finding of an association between CF-related diabetes or impaired glucose tolerance and an abnormal ultrasound that is independent of age is noteworthy. An association with CF-related diabetes and progressive liver disease has been previously reported. 18 Our data did not reveal any increased incidence of CF-related diabetes or impaired glucose tolerance in the homogeneous group. Thus, even though CF-related diabetes may be a marker of more severe CFTR dysfunction, it may also be a marker of other insulin associated mechanisms, distinct from that of NAFLD, that impact liver disease. Although this association could in part be related to the older age of the subjects and more frequent glycemic testing performed in this group, this trend remained after controlling for age. Most of the subjects tested for glucose metabolism were over 10 years of age, consistent with clinical guidelines. 19 Thus, even though this association should be regarded with caution as the number of subjects diagnosed with either CF-related diabetes or impaired glucose tolerance is small, there may be a higher overall prevalence of CF-related diabetes and impaired glucose tolerance than reported in this study. Meconium ileus has historically been considered a risk factor for CF liver disease (CFLD). However, a recent large, single-center analysis of 401 infants with CF over a 25-year period and other studies have found no difference in the percentage of patients with meconium ileus who developed CFLD compared with those without liver disease. 20-22 Our results suggest that a history of meconium ileus increases risk for a homogeneous ultrasound pattern. The significant association of early infection and colonization with Pseudomonas after controlling for age is the most provocative finding of this study. Many more subjects in the normal group were infected with Pseudomonas compared with the abnormal group. This may indicate a modifier gene(s) effect similar to the recent work by Li et al 23 demonstrating an association between the meconium ileus modifier polymorphism SLC6A14 and early Pseudomonas infection. This protective association could also be a marker of alteration in the gut or respiratory microbiome related to antibiotic exposure or reduction of systemic inflammation due to the early treatment of Pseudomonas infection. Changes in both the microbiome and systemic inflammation have been associated with NAFLD and other liver diseases. 24,25 It could also mean that this set of patients may have a less vigorous immune response leading to early Pseudomonas colonization but decreased immune mediated liver inflammatory injury. This phenomenon may also be secondary to early identification of Pseudomonas infection leading to closer follow-up or other unrecognized intervention that leads to a protective effect. Other imaging technologies, notably transient elastography, which measures liver stiffness, has been shown to correlate with advanced fibrosis in pediatric liver diseases, including CFLD, 26-28 but its utility in early fibrosis is unproven. Our findings demonstrate that the frequency of heterogeneous ultrasound abnormalities is higher than previously described, with 3% having unanticipated sonographic cirrhosis. The association between CF-related diabetes and impaired glucose tolerance with abnormal ultrasound patterns and the protective effect of early Pseudomonas infection suggest potential mechanisms predisposing to liver disease that merit further study. Although the ultrasound grading system for this study is rigorous, it is descriptive and may not directly correlate with severity of liver disease in CF. We recognize that no histologic correlation was available for the imaging patterns that are used in the grading system as liver biopsy was not part of this study. The findings in this study do not yet support routine monitoring with abdominal ultrasound, but ongoing clinical follow-up of this large study group will help define the utility of screening abdominal ultrasound and the clinical significance of the spectrum of findings in young children with CF as well as the potential role of ultrasound in tracking responses to newer therapies. We thank the Childhood Liver Disease Research Network (ChiLDReN) investigators, coordinators, directors, members, and families of the CF Centers at each participating site who participated and made the present work possible. We recognize Drucy Borowitz, MD, for serving as a research design and clinical consultant for this study (active member of the CF Foundation, from which she receives funding [BOROWI03CS0]). Appendix Members of CFLD NET include (principal investigator [PI], radiologist [R], site research coordinator [SC]): Ann and Robert Lurie Children's Hospital, Chicago, IL: Estella M. Alonso, MD (PI), Jennifer L. Nicholas, MD (R), Elizabeth Kaurs (SC); Children's Hospital Colorado, Aurora, CO: Michael R. Narkewicz, MD (PI), Ronald J. Sokol, MD (co-PI), Roger Harned, MD (R), Susanna Burr (SC); Emory University, Children's Heathcare of Atlanta, Atlanta, GA: Rene Romero, MD (PI), Jay Freeman, MD, Adina Alazraki, MD (R), Ellen Patrick, MD (R), Eric Hunter (SC); Hospital for Sick Children, Toronto, Canada: Simon C. Ling, MD (PI), Oscar Navarro, MD (R), Julie P. Ling (SC); Cincinnati Children's Hospital Medical Center, Cincinnati, OH; Joe J. Palermo, MD (PI), Alex Towbin, MD (R), Andrea Ferris (SC), Julie Denlinger (SC); Indiana University/Riley Hospital for Children, Indianapolis, IN: Jean P. Molleston, MD (PI), Molly A. Bozic, MD (Co-PI), Girish Subbarao, MD (Co-PI), Boaz Karmazyn, MD (R), Ann Klipsch (SC); Johns Hopkins University School of Medicine, Baltimore, MD: Shruti M. Paranjape, MD (PI), Wikrom Karnsakul, MD (Co-PI), Jane E. Benson, MD (R), Karen A. Callahan (SC), Kim Kafka (SC); Seattle Children's Hospital, Seattle, WA: Karen F. Murray (PI), Ron Gibson (Co-PI), Randolph Otto (R), Alan Genatossio (SC), Melissa Young (SC); St. Louis Children's Hospital, St. Louis, MO: Alexander Weymann (PI), Marilyn J. Siegel (R), Kathy Harris (SC); Texas Children's Hospital, Houston, TX: Daniel H. Leung (PI), Rajesh Krishnamurthy (R), Jameisha Brown (SC); University of Minnesota Masonic Children's Hospital, Minneapolis, MN: Sara Jane Schwarzenberg (PI), Denise Stacklie (SC), F. Glenn Seidel (R—now at Lucille Salter Packard Children's Hospital, Stanford, CA). Data Safety Monitoring Board oversight and program support (National Institute of Diabetes and Digestive and Kidney Diseases, Division of Digestive Diseases & Nutrition, and National Institutes of Health): Edward Doo, MD (Program Official), Averell H. Sherker, MD, FRCP(C) (Scientific Advisor for Viral Hepatitis and Liver Diseases), Sherry R. Hall, MS (Health Program Analyst), and Rebecca Torrance, RN, MS (Executive Secretary; Clinical Trials Specialist). Publisher Copyright: © 2015 Elsevier Inc.

PY - 2015/10/1

Y1 - 2015/10/1

N2 - Objective To investigate the relationship between abdominal ultrasound findings and demographic, historical, and clinical features in children with cystic fibrosis (CF). Study design Children age 3-12 years with CF without known cirrhosis, were enrolled in a prospective, multicenter study of ultrasound to predict hepatic fibrosis. Consensus ultrasound patterns were assigned by 3 radiologists as normal, heterogeneous, homogeneous, or cirrhosis. Data were derived from direct collection and US or Toronto CF registries. χ2 or ANOVA were used to compare variables among ultrasound groups and between normal and abnormal. Logistic regression was used to study risk factors for having abnormal ultrasound. Results Findings in 719 subjects were normal (n = 590, 82.1%), heterogeneous (64, 8.9%), homogeneous (41, 5.7%), and cirrhosis (24, 3.3%). Cirrhosis (P = .0004), homogeneous (P < .0001), and heterogeneous (P = .03) were older than normal. More males were heterogeneous (P = .001). More heterogeneous (15.0%, P = .009) and cirrhosis (25.0%, P = .005) had CF-related diabetes or impaired glucose tolerance vs normal (5.4%). Early infection with Pseudomonas aeruginosa (<2 years old) was associated with a lower risk (OR 0.42, P = .0007) of abnormal. Ursodeoxycholic acid use (OR 3.69, P < .0001) and CF-related diabetes (OR 2.21, P = .019) were associated with increased risk of abnormal. Conclusions Unsuspected cirrhosis is seen in 3.3% of young patients with CF, heterogeneous in 8.9%. Abnormal ultrasound is associated with CF-related diabetes, and early P aeruginosa is associated with normal ultrasound. Prospective assessment of these risk factors may identify potential interventional targets. Trial registration ClinicalTrials.gov: NCT01144507.

AB - Objective To investigate the relationship between abdominal ultrasound findings and demographic, historical, and clinical features in children with cystic fibrosis (CF). Study design Children age 3-12 years with CF without known cirrhosis, were enrolled in a prospective, multicenter study of ultrasound to predict hepatic fibrosis. Consensus ultrasound patterns were assigned by 3 radiologists as normal, heterogeneous, homogeneous, or cirrhosis. Data were derived from direct collection and US or Toronto CF registries. χ2 or ANOVA were used to compare variables among ultrasound groups and between normal and abnormal. Logistic regression was used to study risk factors for having abnormal ultrasound. Results Findings in 719 subjects were normal (n = 590, 82.1%), heterogeneous (64, 8.9%), homogeneous (41, 5.7%), and cirrhosis (24, 3.3%). Cirrhosis (P = .0004), homogeneous (P < .0001), and heterogeneous (P = .03) were older than normal. More males were heterogeneous (P = .001). More heterogeneous (15.0%, P = .009) and cirrhosis (25.0%, P = .005) had CF-related diabetes or impaired glucose tolerance vs normal (5.4%). Early infection with Pseudomonas aeruginosa (<2 years old) was associated with a lower risk (OR 0.42, P = .0007) of abnormal. Ursodeoxycholic acid use (OR 3.69, P < .0001) and CF-related diabetes (OR 2.21, P = .019) were associated with increased risk of abnormal. Conclusions Unsuspected cirrhosis is seen in 3.3% of young patients with CF, heterogeneous in 8.9%. Abnormal ultrasound is associated with CF-related diabetes, and early P aeruginosa is associated with normal ultrasound. Prospective assessment of these risk factors may identify potential interventional targets. Trial registration ClinicalTrials.gov: NCT01144507.

UR - http://www.scopus.com/inward/record.url?scp=84938634331&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84938634331&partnerID=8YFLogxK

U2 - 10.1016/j.jpeds.2015.06.062

DO - 10.1016/j.jpeds.2015.06.062

M3 - Article

C2 - 26254836

AN - SCOPUS:84938634331

VL - 167

SP - 862-868.e2

JO - Journal of Pediatrics

JF - Journal of Pediatrics

SN - 0022-3476

IS - 4

ER -

Baseline Ultrasound and Clinical Correlates in Children with Cystic Fibrosis

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this