Great Issue: We Need More Negative Studies (Published)

A recent ACG “Negative Issue” had some terrific articles –thanks to Ben Gold for sharing his issue.

Here are a few of the studies:

  1. Buspirone had similar efficacy as placebo in a randomized clinical trial for childhood functional abdominal pain, (n=117)  Full text: Comparison of the Efficacy of Buspirone and Placebo in Childhood Functional Abdominal Pain Key finding: Treatment response rates for buspirone and placebo were 58.3% and 59.6% at week 4 (P = 0.902) and 68.1% and 71.1% at week 12 (P = 0.753), respectively.
  2. IBS does not increase mortality in a nationwide cohort study (>300,000 in study)  Full text: Mortality Risk in Irritable Bowel Syndrome Key finding: After adjustment for confounders, IBS was not linked to mortality (HR = 0.96; 95% CI = 0.92–1.00) …and patients with IBS not undergoing a colorectal biopsy were at no increased risk of death (HR = 1.02; 95% CI = 0.99–1.06).
  3. Mongerson was not effective for active Crohn’s disease in a large phase 3 study, n=701 Full text: Mongersen (GED-0301) for Active Crohn’s Disease Key finding: The primary endpoint, clinical remission achievement (CD Activity Index score <150) at week 12, was attained in 22.8% of patients on GED-0301 vs 25% on placebo (P = 0.6210). At study termination, proportions of patients achieving clinical remission at week 52 were similar among individual GED-0301 groups and placebo.
  4. Treatment of H pylori did not increase the risk of C difficile infection (retrospective study)  Full text: Treatment of Helicobacter pylori & Clostridium difficile  Key finding: Of these 38,535 patients with H pylori based on endoscopic pathology, urea breath testing, or stool antigen, 284 (0.74%) had subsequent CDI. Those who developed CDI were less likely to have received treatment for HP within the VHA (66.2% vs 74.8%, P < 0.001)
  5. Percutaneous liver biospy was not safer when done by experienced clinician compared to a fellow, n=212 biopsies  Full text: Major Complications of Pediatric Percutaneous Liver Biopsy Do Not Differ Among Physicians With Different Degrees of Training  Key finding: No significant differences were found between groups (fellows vs staff) regarding number of punctures (median of 1.7 for both), nonrepresentative biopsies (4.2% vs 2.6%), and hemoglobin drop (median of 0.7 vs 0.5 g/L).  Interestingly, in the discussion, the authors assert: “previous studies do not support the conclusion that ultrasound-guided biopsies are superior in terms of safety or adequacy when compared with the use of ultrasound to mark the puncture” (this is based on a study referenced from 2007:  J Gastroenterol Hepatol 2007;22(9):1490–3.)  However, given that severe complications from liver biopsy are infrequent, this current study may be underpowered to detect a small difference between experienced clinicians and fellows.

Related blog posts:

It’s come to this:  Link: YouTube: Dirty Dancing Remake -Safest with a Lamp (this link is for Bernsie). 4 minute video.

#NASPGHAN 17 More Abstracts

This link for the NASPGHAN abstracts :NASPGHAN 2017 Scientific Abstracts

The following slides are from some of the abstract posters. This first poster (next 5 pics) showed that symptom association with meals is not predictive of aspiration among a selected group of children who underwent swallow study evaluations. In the figures, the blue bars are children who passed the swallow study whereas the red bars indicate the children who failed the swallow study.

This next slide demonstrated that a six food diet for EoE could be administered blenderized via a gastrostomy tube.

The next slide showed that irritable bowel syndrome was more frequent (overall hazard ratio of 1.52) following a urinary tract infection in the first year of life.

The next pictures are from a poster discussing high rates of recurrent C difficile infection following fecal microbial transplantation in pediatric patients with inflammatory bowel disease (mainly ulcerative colitis).  An inference from this study would be that many cases of C difficile that were attributed as causing symptoms could in fact have been from a flare up of their IBD.  More details about the diagnosis of C difficile (based on PCR or ELISA) would be helpful

The next poster provides data from CHOP experience with Ustekinumab.  Overall, in this highly-selected (refrcactory) population the long term improvement was low; while one-third had steroid-free remission at week 8, this was not maintained at week 16 and week 24.  In addition, among the 22 patients, one developed transverse myelitis.

This study that follows (next two pics) documented the relative safety of liver biopsies (mainly percutaneous without interventional radiology) in the post-transplant period.  The two most serious adverse events, cholangitis and bile leak, helped identify biliary strictures.

The following collaborative study examined the neurocognitive status of children with Alagille syndrome.  Overall, this study shows that children with Alagille syndrome are at increased risk of low IQ compared to children with other cholestatic diseases.



Moving Away from Liver Biopsies

A recent review (EB Tapper, AS-F Lok. NEJM 2017; 377: 756-68) provides a good review of liver biopsy and liver imaging. My take of this review is that it highlights the emergence of noninvasive tools (imaging & fibrosis markers) which may supplant liver biopsy.  This article does not delve into how more widespread genetic testing may obviate a liver biopsy in many cases as well. The article notes that about 8% of persons in the U.S. have elevated liver enzymes.

Liver biopsy:

  • “A typical liver biopsy samples one fifty-thousandth of the liver.”
  • Limitations of liver biopsy: sampling error is common, biopsy interpretation is subjective, and biopsies can cause complications.  Pain is noted in 30-50% of patients, serious bleeding in 0.6%, injury to other organs (0.08%), and in rare cases, death (up  to 0.1%).
  • Cost: “the average direct cost of a percutaneous liver biopsy is $1448 (in 2016 U.S. dollars).” Transjugular biopsies are much more expensive.  In addition, there are unmeasured indirect costs, due to missing work.

Some prior blogs on liver biopsy

Blood tests:

  • The article details the formulas for biomarker measurements that predict the risk of fibrosis, inlcuding FIB-4, Lok Index, and NAFLD Fibrosis Score.
  • In most liver diseases, aspartate aminotransferase levels “exceed alanine aminotransferase levels when cirrhosis develops.”
  • Thrombocytopenia “is the earliest indicator of cirrhosis among routine blood tests…[due to] diminished liver function (throbopoietin underproduction) and portal hypertension (splenic sequestration).”
  • Proprietary algorithms to assess fibrosis have variable sensitivity, specificity –include FibroTest (aka FibroSure [LabCorp]), FibroMeter, HepaScore (Quest), FIBROSpect, and the Enhanced Liver Fibrosis Score.


  • Elastography with vibration-controlled transient elastography (VCTE) OR magnetic resonance elastography
  • “Elastography offers excellent negative likelihood ratios for advanced fibrosis but much poorer positive likelihood ratios.”
  • Patients with severe obesity are less likely to obtain adequate study with VCTE and could need magnetic resonance elastography to assess fibrosis.

My take: Noninvasive tests have already sharply reduced the need for liver biopsy.

Related posts:

Briefly Noted: Outpatient Liver Biopsy

A small retrospective study (R Bolia et al. JPGN 2017; 65: 86-88) with 497 patients (626 biopsies) found that all complications were identified within 8 hours.  Thirty (48%) had complications, with a subcapsular hematoma being most common (n=14).  Less common adverse events included fever (n=5), skin site ooze (n=3), intraperitoneal bleeding (n=3), hemobilia (n=2), anaphylaxis to gelfoam (n=2), and sepsis (n=1). In this study, the majority of biopsies were performed by interventional radiology (n=492); though, the complication rate was similar in both groups.

The authors conclude that their data support the outpatient liver biopsies in children.

My take: I disagree with the authors’ conclusion to some extent.  Their population is too small to detect rare but severe complications.  Our empiric practice is watch children older than 6 years of age for 6 hours and watch younger children (or others deemed at increased risk) for 24 hours.

Related blog posts:


Concise Review: Fatty Liver in Pediatrics

A recent review (J Schwimmer. Hepatology 2016; 1718-25) provides a succinct up-to-date approach to the common problem of Nonalcoholic Fatty Liver Disease.

As this was a review, much of the material has been covered by this blog and previous publications.  The review discusses the upper limit of normal for alanine aminotransferase and its utility.  Liver imaging is discussed: “MRI is well suited for use in clinical research” whereas “ultrasound does not meet the standard clinical threshold required to be used to diagnose fatty liver…or used as an outcome measure.”

Dr. Schwimmer reviews a prospective study of 347 overweight or obese children with suspected NAFLD (blog review of this study: Screening for NAFLD).  He notes that 24% (n=61) of those who underwent liver biopsy ultimately had other diagnoses, especially autoimmune liver disease (n=11) and celiac disease (n=4). “The clinical challenge is to determine who needs how much of a workup. The greater potential for hepatotoxicity and the more advanced the disease is believed to be, the greater the need to be certain of the diagnosis and to properly grade and stage the disease.”  Currently, “no other diagnostic modality has shown sufficient accuracy to be appropriate for clinical use in the place of biopsy.”

He reviews associated health conditions with NAFLD including obesity, dyslipidemia, hypertension, cardiac dysfunction, and obstructive sleep apnea (~60% of NAFLD patients).

What about treatment? “There is not an available, proven, safe, and effective [pharmacologic] treatment for NAFLD in children…Current treatment is …focused on optimizing lifestyle, including nutrition, physical activity, and mental well-being.”

My take: Despite 20 years of clinical practice, the workup for NAFLD remains a vexing problem.  It is not practical to offer a liver biopsy to 10% of the pediatric population.  So determining who (besides those with more severe presentations) will benefit from an exhaustive workup remains unclear.  In the meanwhile, at a minimum, we need to keep looking for treatable liver conditions (eg. autoimmune hepatitis, celiac disease, Wilson’s disease, and viral hepatitis).

An article with a similar focus (Dr. Schwimmer is the corresponding author): J Pediatric 2016; 172: 9-13.  This report and Dr. Schwimmer’s review both tout the safety of liver biopsy.  Neither report presents much data on costs of either liver biopsies or MRI.

Related blog posts:

Zoo Atlanta 2016

Zoo Atlanta 2016


NASPGHAN Postgraduate Course 2014 -Liver Module

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.  I’ve attached the course syllabus as well:

PG Course Syllabus – FINAL

Primary Sclerosing Cholangitis –Dennis Black (Le Bonheur Children’s Hospital)

  • Up-to-date review provided
  • GWAS (genome-wide association study) identified 16 significant risk loci which account for only 7.3% of overall risk; environmental influences need to be worked out
  • Pediatric studies –total of 328 patients reported to date

Is pediatric disease the same disease as in adults?

  • Incidence in pediatrics: 0.23/100,000 incidence vs 1.1/100,000 in adults
  • Mean age at diagnosis 13 years in pediatrics.
  • 30% of pediatric patients have overlap with autoimmune hepatitis which is higher than in adult patients.

Other pointers:

  •  Discussed “Autoimmune cholangitis.” Imaging needed in autoimmune hepatitis to look for primary sclerosing cholangitis.
  • IBD Association with PSC: IBD occurs in about 55% of PSC patients. If PSC diagnosed first, usually with right-sided colitis.  If IBD diagnosed first, than pancolitis is more commonly noted.
  • Add IgG4 as part of workup to rule out IgG4 cholangiopathy (sensitive to immunosuppression).


  • Supportive care for cholestasis (vitamins, pruritus management, etc
  • Monitoring for complications (rare cases of cholangiocarcinoma in pediatric population).  14 drugs tested to date –mainly in adults.  “All without proven positive impact on long-term outcome.”
  • Ursodeoxycholic acid –widely used but controversial because higher doses associated with worsened outcomes in adult study (Lindor et al).  Ongoing study in pediatric population with ursodeoxycholic acid.
  • Vancomycin (Aliment Pharm 37: 2013; 604.  Adults n=35). Both Flagyl and Vanc seemed to be helpful. Uncontrolled pediatric studies with vancomycin reviewed. Vancomycin study in the works for pediatric/adults.
  • No prospective randomized controlled trials in children and very little data in adults. Hard endpoints –very difficult in children/not practical in children (eg. portal hypertension, transplant, death).

PSC and Transplantation: PSC 2.6% of total transplants –long-term outcome is similar.

Related Blog Posts:

The Jaundiced Infant –Saul Karpen (Emory)

  •  “We don’t estimate jaundice very well… Our eyes do an awful job.”
  • Breastmilk Jaundice: Archives of Disease in Childhood 1978; 53: 506-16.  Only 12 of 853 had jaundice beyond 3 weeks of life.
  • Cholestasis. One of the best studies looking at etiology was recently published:  Hoerning A, et al Front Pediatr. 2014; 2: 65. N=82.  Only 1 patient had CMV.  41% had biliary atresia.

Biliary atresia (BA):

  • Reviewed study indicating that liver biopsy was most accurate means of making diagnosis of biliary atresia (blog comment: this study result may not be accurate in all settings as the interpretation relies on the ability/reliability of pathologist).  High utility of stool pigment & ultrasound (including flow).
  • In retrospective study (Pediatrics 2011; 128 e1428-33), all the BA patients had elevated direct bilirubin by 24-48 hrs of life.
  • Genetic panels and whole exome sequencing (~$4-7K) are happening now. Cost-effective.

Take-home message: Molecular understanding possible for conjugated/unconjugated hyperbilirubinemias. Direct bilirubin >1 is abnormal

Related blog posts:

Acute Liver Failure –Estella Alonso (Children’s Hospital of Chicago) (pg 43)


  • Few patients receive a full diagnostic workup (J Pediatr 2009;155:801‐6)–especially with regard to metabolic and autoimmune disorders.
  • Reviewed etiologies –most frequently “indeterminant” especially in younger patients.  Acetaminophen is most frequent etiology in teenagers and adults.
  • Systemic inflammation is common in acute liver failure (Bucuvalas, J JPGN 2013;56: 311–315). Soluble IL2 receptor alpha –significantly higher in patients that died.  Immune regulation important aspect regarding survival. Should steroids be used in cases with high inflammation?

Prognosis: Squires et al. J Pediatr 2006;148:652-8, Lee et al. JPGN 2005;40:575-81, Baliga et al. Liver Transpl 2004;10:1364-71

  • 33% ‐53% survival with native liver
  • 61% survival including LT
  • 70%‐80% after LT
  • Multiorgan failure is most common etiology of death. Bleeding is “a rare cause of mortality.”


  • Reviewed including coagulopathy/bleeding, cardiovascular collapse, hepatic encephalopathy/cerebral edema
  • Pediatric N-acetylcysteine Trial Squires, et al Hepatology 2013;57:1542‐9 N=182.  Patients with NAC seemed to do worse, but not statistically proven.  This study has stopped the widespread use of NAC in acute liver failure.
  • Discussed approach to neurological complications in ALF. Hussain et al, JPGN 2014;58:449‐56. Retrospective study (n=18). Early EEGs obtained. Hypertonic saline may be more effective than mannitol.  Hypothermia may be helpful adjunct.
  • Timing of Transplantation discussed (pg 54 in syllabus). Difficult to predict spontaneous survival.

Related blog posts:

Disclaimer: These blog posts are for educational purposes only. Specific dosing of medications/diets (along with potential adverse effects) should be confirmed by prescribing physician/nutritionist.  This content is not a substitute for medical advice, diagnosis or treatment provided by a qualified healthcare provider. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a condition.

This Year's Pumpkin

This Year’s Pumpkin

Liver Biopsy -Risks and Benefits

Even in the ‘old USA,’ there is a mortality risk from liver biopsy in the pediatric population.  A recent study from Los Angeles confirms this (JPGN 2013; 57: 644-48).

This retrospective review of all children (n=213 children & 328 biopsies) who underwent a percutaneous liver biopsy between 2008-2011 were examined.  These biopsies were completed by radiology with ultrasound or CT.  Gel foam was injected in cases of multiple biopsies.


  • 9 (4.2%) dropped hemoglobin > 2 /dL.
  • 7 (3.3%) needed a transfusion.
  • 1 (0.5%) died.  This was a 2.6 kg infant seen for transplant evaluation.
  • 63 (19%) had insufficient samples for definitive histologic evaluation.
  • In 81% of initial biopsies, “a definitive pathologic diagnosis was obtained.”
  • Biopsies for unexplained elevation of liver function tests were nondiagnostic in 34.9%.

The authors take: “our data demonstrate that percutaneous liver biopsy is generally safe; yet, finite risk remains, with bleeding-related complications occurring 5.2% of children.”

Bottomline: make sure you need the information from the liver biopsy enough to justify the risk, particularly in small children and in those at increased risk for bleeding.

Related blog post (with annotated references):

Liver biopsy risk in children | gutsandgrowth

Liver biopsy risk in children

A recent retrospective study from Oslo looks at risk factors and safety of liver biopsy (JPGN 55: 82-87).

Among 190 patients who underwent 275 ultrasound-guided liver biopsies (interventional radiology), there were four major complications –two were due to post-biopsy bleeding, one was due to variceal bleeding within 12 hours, and one was due to the development of pain/acidosis, and tachycardia.  28 patients had minor bleeding.   There were no mortalities, though one patient dropped hemoglobin in half (11.6 to 5.3).

In their patient population, the following had increased risk for major complications:

  • Focal space-occupying lesion/tumor (n=25) OR 2.84  for all bleeding risk; patients with these lesions typically had more biopsy passes (average 4.9)
  • Acute liver failure (n=12) OR 26.1 for major complication risk
  • Low-molecular weight heparin therapy (n=18)  OR 2.43 for all bleeding risk

Not identified as risk factors for complications in this cohort:

  • Low platelet count (<70 [n=14])or coagulopathy (INR>1.7 [n=18]),  –all received blood products before biopsy
  • Aspirin treatment (often used in transplant population to prevent hepatic artery thrombosis) (n=55)  OR 0.96
  • Liver transplant patients (n=97) -odds ratio was lower than entire cohort (OR 0.52)

Additional references:

  • -JPGN 2011; 53: 202. Good safety results with IR liver biopsy, n=249. 2/249 had drop in Hgb of 2g -no transfusions needed, no mortalities.
  • -Clin Gastro & Hep 2010; 8: 877. 0.5% complication rate, n=2740. No deaths. Bleeding was most common risk –increased with advanced liver disease.
  • -Hepatology 2009; 49: 1017. AASLD Position Paper
  • -Gastroenterol 1978; 74: 101 & 103. Early article discussing safety of LBx and that most pts could be d/c’d w/in 6 hours.
  • -JPGN 2005; 41: 639. Safety of liver biopsy in infants less than 3 months (w/o U/S). complication rate was 18% including sedation-related, 1 bile leak, and 3 needed PRBCs.
  • -JPGN 2003; 36: 364. Ultrasound useful.
  • -Can J Gastroenterol 2000; 14: 543-548.  Mortality rate 3/10,000
  • -Clin Perspectives Gastroenterol 2002; 5: 117. Rec ultrasound, plts >75K, PT c/in 3 secs of normal; observation for 3-6hrs in adults.
  • -NEJM 2001; 344: 495. Liver bx review.
  • -Hepatol 27: 1220-26, 1998. U/S marking reduces complications
  • -Fox VL, Cohen MB, Whitington PF, Colletti RB. Outpatient liver biopsy in children (n=450). J Pediatr Gastroenterol Nutr 1996;23:213-6.  High mortality rate reported, primarily in bone marrow transplant patients
  • -JPGN 2000; 31: 536-39. Safety of liver biopsy in children, n=249.

Pediatric NAFLD Position Paper

A previous blog post (NAFLD Guidelines 2012) described comprehensive, up-to-date NAFLD guidelines from AASLD, AGA, and ACG.   Another group of experts from ESPGHAN (European Society for Pediatric Gastroenterology, Hepatology, and Nutrition) has also published a position paper on the diagnosis of NAFLD in children; coincidentally, these were published recently as well (JPGN 2012; 54: 700-13).

While there is some overlap in the information between the two guidelines, there are some notable differences.  The JPGN manuscript does include a nice differential diagnosis list  which can cause fatty liver disease (Table 2), including some rare entities like Dorfman-Chanarin syndrome, Cantu syndrome, Madelung lipomatosis, and numerous medications.  This review has more emphasis on etiology.

Table 3 lists a recommended workup in children with suspected NAFLD:

  • Standard liver function tests/blood counts/coagulation studies
  • Fasting glucose & insulin
  • Lipid profile
  • Glucose tolerance test & glycosylated hemoglobin
  • Calculation of HOMA-IR, markers of insulin resistance

AND Tests to exclude other liver diseases: 

  • Lactate, uric acid, iron, ferritin, pyruvate
  • Copper, ceruloplasmin, 24-hour urinary copper
  • Sweat test
  • Celiac serology (TTG IgA and serum IgA)
  • α-1-antitrypsin levels and phenotype when indicated
  • Amino and organic acids
  • Plasma free fatty acids and acyl carnitine profile
  • Urinary steroid metabolites
  • Other specific tests as suggested by evaluation (eg. viral hepatitis panel, serum immunoglobulins, liver autoantibodies)

When one looks at the recommended diagnostic algorithm (Figure 1) and tests outlined, these guidelines are not nearly as practical as the NAFLD guidelines from AASLD, AGA, and ACG and often contradictory between the tables/figures and the text.  How much would it cost for the recommended testing if/when extrapolated to the vast numbers of individuals with these disorders?  In addition, a much more limited diagnostic approach is suggested in the final section than outlined in Table 3 and Figure 1.

Imaging: these authors advocate LFTs and ultrasonography in all obese children (> 3 years) and adolescents.  If normal LFTS and sonography, the algorithm suggests the use of MRI if clinical signs of insulin resistance.  Later, the authors conclude “MRI is not cost-effective.”

Liver Biopsy: while the authors state that there is “no present consensus or evidence base to formulate guidelines” for liver biopsy, this is not well-reflected in their diagnostic algorithm in which arrows point to liver biopsy in almost everyone –either early liver biopsy or eventual biopsy in patients with persistent disease.  Accepted liver biopsy indications, according to the executive summary, include the following:

  • Exclude other treatable disease
  • Suspected advanced disease
  • Before pharmaceutical/surgical treatment
  • Research purposes

My conclusion about this position paper is it is less helpful than the AASLD/AGA/ACG guidelines.  In fact, when extensive diagnostic testing is recommended by experts, it is fortunate that other expert guidelines are available that support a more cost-effective approach.  In NAFLD cases that seem atypical and especially in the very young patient, this reference may still be helpful.