Double-Blind Randomized SMOFLipid Study

A recent double-blind randomized study (A Repa et al. J Pediatr 2018; 194: 87-93) compared a mixed lipid emulsion (SMOFlipid) to a soybean-oil lipid in 223 extremely low birth weight infants. Median time on parenteral nutrition was ~23 days.

Key findings:

  • The primary outcome of parenteral nutrition associated cholestasis (PNAC) was NOT significantly different in the two groups: 10.1% for SMOF and 15.9% for control group (P=.20).
  • No other outcome measures were affected, including ROP, BPD and growth.

The authors note that even the control group had less cholestasis than previous cohorts and indicated that the use of probiotics and possibly more aggressive enteral feeds were at work.

My take (borrowed in part from authors): These results “cannot be generalized to infants with substantially longer time on PN.” However, this study shows that SMOFlipid alone will not prevent cholestasis, which is well-known to be multifactorial.

Sandy Springs

Favorable Fish Oil Outcomes in High Risk Preterm Infants

Briefly noted: M Sorrell et al. JPGN 2017; 64: 783-88. In this small study with 13 infants (mean gestational age of 28 weeks) who had short bowel syndrome or severe dysmotility and direct bilirubin ≥4 mg/dL (mean 9.8 at enrollment), patients received a fish oil-based lipid emulsion (1 g/kg/d). They were compared with 119 GA-matched controls.

Overall, the authors found the fish oil supplement to be safe.  All patients had resolution of cholestasis. They note the difficulty of proving effectiveness and performing studies in this population.  “Neonatologists…find themselves faced with …a growing body of uncontrolled data that suggests benefits of an unapproved treatment…An attempt to perform a randomized controlled comparison of a plant-based lipid emulsion to FishLE in preventing PNALD in infants at risk was terminated early after an interim analysis revealed much lower than expected incidence of PNALD…[making] trials ethically problematic.”

My take: The data remain incomplete and make it difficult to use a therapy like Omegaven that is quite expensive (not covered) and not FDA approved.  The availability of SMOFlipid is likely to result in less usage of plant-based soy products.

Related blog posts:

Rodin Museum

Nutrition Week (Day1) Downside on Lipid Reduction

Recently I gave a lecture on parenteral nutrition associated liver disease (PNALD), though the term intestinal failure associated liver disease (IFALD) is probably more popular at this time.  The day afterwards, I read an important study (L Beauport et al. J Pediatr 2017; 181: 29-36) reiterating one of the concerns in the lecture.

This study showed that higher lipid intake in a cohort of neonates born at <30 weeks during the first 2 weeks after birth was associated with a lower incidence of brain lesions and dysmaturation when examined by MRI at term equivalent age (TEA).

Details: This prospective cohort study examined energy/lipid intake in the first 2 weeks of life. Eligible patients were neonates ≤30 weeks.  Group 1 with 27 patients had birth weight median of 900 gm compared with Group 2 with 15 patients had median weight of 844 gm. During the first year of the study, participants received a soybean emulsion whereas in the last year of the study, the neonates received a mixture of soybean and olive oil (Clinoleic).

Key finding: After adjusting for clinical risk scores and sepsis, the authors found that the higher energy/lipid intakes resulted in improved brain MRIs in group 1. A “10 Kcal/kg/day increase in energy of 0.7 g/kg/day increase in lipids intake would reduce the risk of having more severely abnormal MRI at TEA by >60%.”

Here are some slides from my talk relevant to this topic and to parenteral nutrition associated liver disease (PNALD):

screenshot-102 screenshot-103

Explaining Differences in Disease Severity for Alagille Syndrome

A recent study (DOI: has shown a gene which may help explain the difference in disease severity in Alagille syndrome.

Here’s a link to full text: THBS2 is a Candidate Modifier of Liver Disease Severity in Alagille Syndrome

Here’s the abstract:

Background & Aims

Alagille syndrome is an autosomal-dominant, multisystem disorder caused primarily by mutations in JAG1, resulting in bile duct paucity, cholestasis, cardiac disease, and other features. Liver disease severity in Alagille syndrome is highly variable, however, factors influencing the hepatic phenotype are unknown. We hypothesized that genetic modifiers may contribute to the variable expressivity of this disorder.


We performed a genome-wide association study in a cohort of Caucasian subjects with known pathogenic JAG1 mutations, comparing patients with mild vs severe liver disease, followed by functional characterization of a candidate locus.


We identified a locus that reached suggestive genome-level significance upstream of the thrombospondin 2 (THBS2) gene. THBS2 codes for a secreted matricellular protein that regulates cell proliferation, apoptosis, and angiogenesis, and has been shown to affect Notch signaling. By using a reporter mouse line, we detected thrombospondin 2 expression in bile ducts and periportal regions of the mouse liver. Examination of Thbs2-null mouse livers showed increased microvessels in the portal regions of adult mice. We also showed that thrombospondin 2 interacts with NOTCH1 and NOTCH2 and can inhibit JAG1–NOTCH2 interactions.


Based on the genome-wide association study results, thrombospondin 2 localization within bile ducts, and demonstration of interactions of thrombospondin 2 with JAG1 and NOTCH2, we propose that changes in thrombospondin 2 expression may further perturb JAG1–NOTCH2 signaling in patients harboring a JAG1mutation and lead to a more severe liver phenotype. These results implicate THBS2 as a plausible candidate genetic modifier of liver disease severity in Alagille syndrome.

Related blog posts:


Sertraline and Liver Disease

A recent case report (MA Conrad, HC Lin. J Pediatr 2016; 169: 313-5) on sertraline-associated cholestasis provided a good reason to take a quick review on the NIH Liver Toxicity website:

Hepatotoxicity with sertraline (zoloft)

Liver test abnormalities have been reported to occur in up to 1% of patients on sertraline, but elevations are usually modest and infrequently require dose modification or discontinuation.  Rare instances of acute, clinically apparent episodes of liver injury with marked liver enzyme elevations with or without jaundice have been reported in patients on sertraline.  The onset of injury is usually within 2 to 24 weeks and the pattern of serum enzyme elevations has varied from hepatocellular to mixed and cholestatic.  Autoimmune (autoantibodies) and immunoallergic features (rash, fever, eosinophilia) are uncommon.  Actue liver failure due to sertraline has been described but is very rare.

The case report describes a 15 yo who developed jaundice (peak bilirubin 33.7 mg/dL with a direct fraction of 29.2 mg/dL) after 6 months of treatment with 75 mg per day.  After negative blood tests, he had a liver biopsy which was notable for rare bile ducts.  A jaundice chip was negative for underlying disorders like Alagille syndrome.  Urine bile acids were negative as well.  His laboratories normalized completely four months after cessation of sertraline.

It is interesting to note that sertraline has been used therapeutically for patients with pruritus due to cholestasis (Understanding Cholestatic Pruritus | gutsandgrowth)

My take: This case report describes bile duct paucity (vanishing bile ducts) as a result of sertraline therapy.  For practitioners, the bottom line is that SSRIs rarely cause liver toxicity; however, for patients with persistently-abnormal liver chemistries on SSRI therapy, discontinuation and identification of a safe alternative medication may be warranted.

Castillo San Felipe del Morro

Castillo San Felipe del Morro

Parenteral Lipids & Cholestasis –a Little More Data

A recent publication in JPGN indicates that resuming low dose soy-based parenteral lipid can be effective in patients (n=7) whose cholestasis had resolved on a fish oil-based parenteral lipid. It does not resolve the larger question of whether fish oil-based parenteral lipids are truly more effective than soy-based parenteral lipids (see previous blog links below).

Here’s the abstract:

Objectives: Intestinal failure associated liver disease (IFALD) contributes to significant morbidity in pediatric intestinal failure (IF) patients. However, the use of parenteral nutrition (PN) with a fish oil-based IV emulsion (FO) has been associated with biochemical reversal of cholestasis and improved outcomes. Unfortunately, FO increases the complexity of care: as it can only be administered under FDA compassionate use protocols requiring special monitoring, is not available as a 3-in-1 solution and is more expensive than comparable soy-based lipid formulation (SO). Due to these pragmatic constraints a series of patient families were switched to low-dose (1 g/kg/day) SO following biochemical resolution of cholestasis. This study examines if reversal of cholestasis and somatic growth are maintained following this transition.

Methods: Chart review of all children with IFALD who switched from FO to SO following resolution of cholestasis. Variables are presented as medians (interquartile ranges). Comparisons performed using Wilcoxon signed-rank test.

Results: 7 patients aged 25.9 (16.2,43.2) months were transitioned to SO following reversal of cholestasis using FO. At a median follow up 13.9 (4.3,50.1) months there were no significant differences between pre- and post-transition serum alanine and aspartate aminotransferases, direct bilirubin, and weight-for-age z-scores. Due to recurrence of cholestasis, one patient was restarted on FO after four months on SO.

Conclusions: Biochemical reversal of IFALD and growth were preserved after transition from FO to SO in 6/7 (86%) patients. Given the challenges associated with the use of FO, SO may be a viable alternative in select home PN patients.

Related blog posts:


“Genetic Testing and the Future of Pediatric Gastroenterology”

Last night, a symposium on “Genetic Testing and the Future of Pediatric Gastroenterology” sponsored by Children’s Healthcare of Atlanta took place.  The speakers included Dr. Ben Gold from our pediatric GI group (GI Care for Kids), Dr. Saul Karpen and Dr. Subra Kugasthasan (Emory), and Dr. Robert Heuckeroth (CHOP).

This blog entry has abbreviated/summarized the presentations. Though not intentional, some important material is likely to have been omitted; in addition, transcription errors are possible as well.  All of the speakers had terrific presentations.

GI Genetic Testing –Subra Kugathasan

Reasons for genetic testing:

  • Predicting prognosis: predicting stricturing/fibrosis in Crohn’s, predicting cancer in ulcerative colitis; BRCA1 in breast cancer
  • Choosing the right medicine: pharmcogenomics
  • Precision medicine: prevention of disease, slowing progression of disease.

Examples in current medicine:

  • Recurrent pancreatitis –novel mutations identified in SPINK1.  Also now hereditary pancreatitis may be due to mutations in CPA1, GGT1, CLDN2, MMP1, MTHFR in addition to CTRC, SPINK1, CFTR, and PRSS1.
  • Inflammatory bowel disease (IBD):IL10 Receptor mutation , TTC7A –>VEO IBD; IPEX gene (can worsen with immunosuppression). Panel testing now available for 40 genes –4 of 22 patients identified with IBD.  Identifying cause of VEO IBD may lead to treatment: bone marrow transplantation.
  • IBD: CLIA/CAP certified Emory Genetics panel ~50 genes ( Dr. Kugathasan indicated that this testing is likely to be a better 1st step then exome testing. Yield with exome sequencing (in highly selected populations) about 25% at this time but likely to increase. If negative, can proceed with whole exome sequence.  Numerous problems with exome sequencing; for example, exome sequencing may identify genes of unknown significance and identifying genetic problems unrelated to clinical issue.

Who/When to test?

  • Very early onset disease (<10 years), atypical presentation, perhaps treatment-refractory.

Take-home point: “All GI diseases have genetic testing in future.” Testing for highly selected patients for gene defects can be accomplished with gene panel and if negative, whole exome testing.

Related blog posts:

Liver: Cholestatic & Metabolic Diseases of Infants and Children —Saul Karpen

Potential areas for genetic testing:

  • Neonatal cholestasis: PFIC, Metaboic, Biliary Atresia
  • Transplant

“Why bother…they all get transplanted anyway…”  According to Dr. Karpen, this view needs to be reconsidered.

Neonatal cholestasis:

  • (Front Pediatr 2014; 2: 65)  41% with biliary atresia, 13% idiopathic, and a lot of others.  N=82. Other etiologies: Genetic disorders; Biliary disease (eg. Caroli), transporter defects (PFICs/BRICs), Metabolic (Niemann-Pick C, tyrosinemia, HFI, Peroxisomal, GSDs, Peroxisomal, Mitochondrial, A1AT). Thus, panels to identify these disorders can be very helpful.
  • Emory Cholestasis 56+ Gene Panel. Testing is cheaper than endoscopy

PFIC: Progressive Familial Intrahepatic Cholestasis

  • PFIC1: ATP8B1 (Byler) –besides cholestasis, patients often with diarrhea, hearing loss, very itchy; can have cirrhosis at 2 years of life
  • PFIC2: ABCB11 (BSEP deficiency) –can have cirrhosis at 6 mo, prone to HCC (as early as 13 mo), very itchy
  •  PFIC3 (high GGT) ABCB4 –can have cirrhosis at 5 mo, can cause problems at later ages as well (eg. intrahepatic cholestasis of pregnancy, gallstones); increase risk for HCC/cholangiocarcinoma.
  • Identifying PFIC (could mimic PSC) and BRIC (Benign Recurrent Intrahepatic Cholestasis)–is helpful in following patients for specific management when symptoms recur and to screen for complications (eg. HCC).

Biliary Atresia:

  • No clear genetics in most
  • Laterality defects in 5-10% -asplenia/polysplenia, cardiac defects
  • GPC1 gene is a susceptibility gene in zebrafish
  • ADD3 gene identified in Han Chinese OR 2.38 –may be a susceptibility gene. (30% of cases, 17% of controls)

NAFLD: Associated with increased mortality compared with matched controls. Patients develop thicker atherosclerotic plaques. PNPLA3 gene identified as a susceptibility gene for NAFLD and is highly prevalent in Hispanic populations.  Similarly, PNPLA3 has been associated with NASH in Italian populations.  If you have this genotype, this increases risk of liver fat in the face of increased sugar consumption.

Transplant medicine: Deoxyguanosine Kinase Deficiency (DGUOK) –rapid sequencing for this gene pretransplant –If positive, should not be transplanted. These individuals have systemic disease that cannot be cured with liver transplantation.

Who/When for genetic testing?: DGUOK in liver failure patients, and in infants without diagnosis after liver biopsy/exclusion of A1AT

Take-home message: Genetic testing has a role in pediatric liver disease and it is affordable.

Related blog posts:

GI –Single Microbes to the Microbiome and GI Disease —Ben Gold

  • Described why changes in our environment can trigger development of disease due to changes in microbiome (eg. immigrants/children with IBD in developed countries at much higher rate than at developing countries)
  • Discussed Helicobacter pylori –‘how a single microbe which may have been good turned bad’
  • Described pathogenesis. What you get exposed to early on may lead to an exaggerated response by T-cells/immune system.  Healthy microbiota is critical to train the immune system via GALT to protect host and decrease the chances for immune overexpression.

Key points:

  • 100 trillion bacteria that live in our GI tract. 10x number of human cells in our body and 100x as many genes as there are in the human genome.  Partnership between humans and their microbiome developed over thousands of years.
  • Vaginal delivery is NOT sterile. Are there consequences to C-section? Food allergy for infant –OR 2.5 if Mom with food allergy delivers vaginally vs OR 7.8 if Mom has food allergy and delivers via C-section. Also, some data indicates increased risk of EoE if born via C-section.  From DAY 1, microbiome can be influence by environmental factors.
  • Influencing microbiome happens mainly during first three years of life.

Why the microbiome is so important/more pointers:

  1. Since 1950, there has been a huge decline in infectious diseases like measles, mumps, hepatitis A, tuberculosis, etc
  2. Coincident with these decreases there has been increased multiple sclerosis, Crohn’s disease, asthma, food allergy, autoimmune diseases
  3. Sanitized food supply, decrease in naturally fermented foods, urban lifestyle, antibiotics, C-section all lead to lower microbial exposure and altered intestinal microbiota. This in turn may lead to an inadequate immune response.
  4. Elie Metchnikoff 1845-1916: suggested ingested bacteria could be healthy. Probiotics/prebiotics are not a new idea!
  5. Obese patients had very high levels of Firmicutes and low Bacteroidetes.
  6. Fecal microbial transplantation (FMT)–reseeding GI microbiome. FMT may be beneficial to many diseases and is being  studied.

Helicobacter pylori -evidence of H pylori as far back as 60,000 years ago and has evolved with humans. H pylori may have helped provided a positive immune response in children and adults.

Bottomline: Human genetic diseases may be heavily influenced by the 300 trillion bacteria and their genes; these bacteria are susceptible to environmental disease.

Related blog posts:


Genetic Basis of Motility —Robert Heuckeroth

  • Basic machinery controlling motility described –enteric neurons, muscles, pacemaker cells.
  • Very little clinical overlap between modern genetic testing and applicable motility disorders: achalasia, gastroparesis, pseudoobstruction, Hirschsprung’s or irritable bowel
  • Focused testing for suspected diagnosis is being displaced by broader testing in serious disease, especially since more extensive genetic testing may be more cost-effective. When to do exome sequencing?

Hirschsprung’s disease:

  • 1:5000 children.
  • 100X higher risk in Down Syndrome.
  • Prenatal testing not helpful at this time. There may be >360 genes that increase risk (variable degree of risk) of Hirschsprung’s disease; hence prenatal testing not that helpful at this time.
  • 30 associated genetic syndromes with Hirschsprung’s, >12 known gene defects.  Hirschsprung’s disease: 25% with RET haploinsufficiency.  RET haploinsufficiency –increases risk of Hirschsprung’s disease >2500-fold risk.
  • Gene environment interactions can increase risk of developing Hirschsprung’s disease –if vitamin A deficient, mice with increased risk.
  • RET gene –>too little RET increases risk of Hirschsprung’s
  • RET gene –>too much RET increases risk of MEN2B, MEN2A.  Though 7.5% of MEN2A have Hirschsprung’s –works out to be 1 in 100 kids with Hirschsprung’s have MEN2A mutations.  ??test for this??

Pseudoobstruction genetic basis– a number of genes identified, including ACTG2 (smooth muscle actin gene).  If you understand etiology, this may lead to prevention and treatment.

Take-home message: Currently biggest problem with genetic testing, especially with motility disorders, is identifying genetic defect of unknown significance.  Thus, testing needs to be done as part of research studies.

Related blog posts: