A 30-month-old boy was referred to the Children's Memorial Hospital for evaluation of loss of balance and epistaxis. His motor and cognitive development had been normal until the past 4 months. Over this time, he experienced increasing difficulty walking steadily. His expressive language had regressed; he had been speaking in five-word sentences, but currently only uses two-word phrases. In addition, his parents noted that he is experiencing increasingly prolonged episodes of epistaxis. Since his mother has a history of a "bleeding disorder" treated with vitamin K drops, she started giving these to her son without effect. There was no history of fever, weight loss, vomiting, or other symptoms.
His medical history is remarkable in that he was a planned delivery at home without any health care personnel in attendance. There were no perinatal problems. He did not receive vitamin K at birth and no newborn screening tests were performed. He was diagnosed with an allergy to wheat at age 15 months. Since then, he had been receiving a gluten-free diet which consisted of rice, eggs, green beans, carrots, and an occasional cheeseburger. He was provided health care from birth by a non-MD practitioner; no previous growth parameters were available.
The family history was remarkable in that he had six siblings, all unimmunized. Three had been diagnosed as having celiac disease.
On physical examination, he was an irritable boy who didn't like being examined. Vital signs were unremarkable. His weight and height were in the 25th percentile, and his head circumference was 56 cm, markedly greater than the 95th percentile. HEENT exam was unremarkable save for macrocephaly. His lungs were clear. Cardiac exam was unremarkable. The abdomen was soft and non-tender. The liver was palpable 4 cm below the right costal margin, and the spleen was palpable 4 cm below the left costal margin. On neurologic examination, he was ataxic; he could not stand alone without wobbling and falling. There was generalized hyperreflexia; Babinski responses were abnormal bilaterally. Strength was normal in both the upper and lower extremities. Cranial nerve examination was normal.
Initial laboratory examination revealed: hemoglobin 9.5 g/dl; white blood cell count 4,500 per mm3 with a normal differential; platelet count 58,000 per mm3; erythrocyte sedimentation rate 47; SGPT 29 IU; total bilirubin 3.5 mg/dL; direct bilirubin 1.5 mg/dL; and alkaline phosphatase 369 IU. The prothrombin time and partial thromboplastin time were both significantly elevated, 17.6 and 53 seconds, respectively.
Robert Listernick, MD, moderator: There's so much information to digest here. We can either use a deductive approach, proceeding logically from one fact to another, or use a more inductive line of reasoning. The disadvantage to the latter approach is that, in haste, one may not use all the available information and may bypass reasonable alternate diagnoses. First, let's dispose of some of the possible red herrings. What about this incredible history of celiac disease?
Timothy Sentongo, MD, pediatric gastroenterologist: Celiac disease has become a fashionable diagnosis, particularly in persons who have alternate health care beliefs. It has become quite common to label children as being allergic to wheat without any sound scientific basis. I have seen innumerable children on poorly conceived elimination diets who clearly didn't have celiac disease. Celiac disease is a lifelong condition; therefore, a gluten-free diet is very restrictive and should not be prescribed lightly. The diagnosis of celiac disease should be based on the results of serologic testing, anti-gliadin IgG and IgA antibodies and tissue transglutaminase antibodies, and confirmed by small bowel biopsy. Thus, I would be very suspect of this family's diagnoses. With that said, individuals with celiac disease may have neurological manifestations that include a seizure disorder, intracranial calcifications, and greywhite changes on neuroimaging. Developmental regression in association with celiac disease is usually the result of malnutrition.
Kent Kelley, MD, pediatric neurologist: This child has a progressive encephalopathy leading to developmental regression. From the history alone, it would have been conceivable that his regression was a result of weakness due to a systemic illness. However, he clearly has an abnormal neurologic examination.
Dr. Listernick: How much secondary developmental regression might we expect to see in children who have systemic illnesses without direct central nervous system involvement?
Karan Emerick, MD, pediatric hepatologist: As an example, children who have biliary atresia and end-stage liver disease may become progressively weaker and lose muscle mass, but their cognitive development remains normal. I would not expect to see loss of expressive language, as this child expressed.
Dr. Listernick: How should we proceed with the evaluation?
Dr. Kelley: The temptation is to try to tie together developmental regression, macrocephaly, and hepatosplenomegaly. With the information at hand, it's difficult to put it all together. There are several diseases in which one sees macrocephaly and developmental regression, such as Alexander and Canavan diseases, but those children don't develop hepatosplenomegaly. Other conditions that might produce developmental regression and hyperreflexia, such as the leukoencephalopathies, which are demyelinating diseases, aren't associated with hepatosplenomegaly. Nothing appears to hang together.
Brad Angle, MD, geneticist: In considering the possibility of storage disease, it's helpful to consider the different categories of abnormally stored materials, such as carbohydrates, lipids, and mucopolysaccharides, and determine into which category the signs and symptoms fit. For instance, the absence of hypoglycemia and the presence of neurologic symptoms speak against a disorder of carbohydrate metabolism, such as glycogen storage disease. The absence of dysmorphic features, in particular a coarse facies, makes the mucopolysaccharidoses less likely. Therefore, initially I would focus on the sphingolipidoses. I would test this child for GMl gangliosidosis, Niemann-Pick disease, and Gaucher disease. Each of these disorders may present with hepatosplenomegaly and neurologic involvement. However, with that said, these disorders generally present earlier in life and none are associated with macrocephaly.
Dr. Listernick: I should give you a little more information. The day after admission, he had a formal ophthalmologic examination and was found to have bilateral papilledema. Maybe I should have told you sooner!
Dr. Kelley: You've been holding out on me! Obviously this substantially changes our approach. Now, we should be considering why he might have increased intracranial pressure (ICP). His macrocephaly very well may be due to long-standing increased ICP. This, in turn, could be caused by either increased cerebrospinal fluid (CSF) production (eg, choroid plexus papilloma), decreased CSF reabsorption (eg, non-communicating hydrocephalus), or obstructive hydrocephalus. Chronically elevated ICP could have led to his neurologic findings as well as his developmental regression. Neuroimaging would be the next step.
Delilah Burrowes, MD, pediatric neuroradiologist: The magnetic resonance imaging scan showed moderately dilated ventricles and prominent cortical sulci, indicating volume loss. There were no mass lesions or signs of obstruction. These are all findings of increased ICP. Magnetic resonance spectroscopy showed an elevated glutamine/glutamate peak with absence of the myoinositol peak, which is suggestive of liver disease, specifically hepatic encephalopathy.
Dr. Listernick: How should we approach his liver disease? He has normal transaminases and marked abnormalities in synthetic function.
Dr. Emerick: It makes no sense to go through an extensive differential diagnosis of chronic liver disease. This child has increased ICP without a mass lesion or obstruction; these are the criteria for pseudotumor cerebri. Thus, the appropriate question is what causes liver disease and pseudotumor cerebri. The differential diagnosis for this combination is very short.
Ruba Azzam, MD, pediatric gastroenterology fellow: When I reviewed the history with the family, I specifically asked what types of vitamins the child was being given. Initially, they weren't sure but thought that he might have been taking vitamins A, B, K, and echinacea. Further research determined that he had been receiving daily doses of Vitamin A between 75,000 and 150,000 international units (IU), or 22,000 to 44,000 ug per day, for the past year. As a frame of reference, the recommended daily allowance of vitamin A for children 1 to 3 years of age is approximately 400 µg/day. In fact, the health care provider was writing a prescription for the vitamin A and selling it to the family himself.
Dr. Listernick: So, this child had severe vitamin A toxicity. As we've said many times in this conference, history tells all. It may take many physicians asking the same questions multiple times before the key piece of information is elicited. Of course, it also helps to be smart and ask the right question, as Dr. Azzam did. The clues were there, particularly given the family's alternate health care beliefs.
Do we understand why vitamin A is toxic?
Dr. Sentongo: Vitamin A is fat soluble and is stored in the body for long periods of time. Unfortunately, serum levels do not reliably reflect total body stores. Vitamin A influences cell differentiation, membrane function, and response to steroid, thyroid, and other hormones. In so doing, it influences growth, differentiation, and function of target cells. Vitamin A is preferentially stored in Ito cells, also known as stellate cells, which form 1 % of the liver cell mass and participate in tissue healing and fibrogenesis. Excess storage appears to stimulate the Ito cells, leading to perisinusoidal fibrosis, cirrhosis, and portal hypertension.
Hector Melin-Aldana, MD, pediatric pathologist: The liver biopsy showed mild portal fibrosis without any evidence of cirrhosis. There is a moderate degree of hepatocyte damage. However, the most remarkable finding is extensive hyperplasia of the Ito cells, which are swollen with abnormal fat stores. Normally, these cells store both fat and vitamin A. However, once stellate cells are activated, they transform into fibroblasts and induce fibrosis. This ultimately may lead to portal hypertension. In the context of an appropriate history, these findings are classic for vitamin A hepatotoxicity.
Dr. Listernick: What's the difference between hepatic fibrosis and cirrhosis?
Dr. Melin-Aldana: Fibrosis is an irreversible result of hepatic damage. Cirrhosis occurs when the fibrosis becomes severe and nodules of regenerating hepatocytes become surrounded by scar tissue.
Dr. Listernick: Will he eventually require liver transplantation?
Dr. Emerick: Generally, vitamin A hepatotoxicity is reversible. There are rare cases of cirrhosis that ultimately required transplantation. In addition, there are case reports of late onset cirrhosis after withdrawal of vitamin A from the diet. Presumably, these individuals experienced ongoing hepatotoxicity as vitamin A was slowly released from fat stored in other body tissues. Fortunately, there was no evidence of cirrhosis on his liver biopsy.
Dr. Listernick: He'll be followed for the next several years to make sure that there is no ongoing fibrosis or the development of cirrhosis.
Let's shift gears. If a parent asks if mistakes have been made in a child's care, how should one respond?
Joel Frader, MD, pediatric ethicist: You can't ignore the facts, as you know them, keeping in mind that referral center physicians may not have all the possible information. The physician should provide the family with the information about what has happened (ie, that the liver damage came from vitamin-induced insult) without placing blame. People will draw their own conclusions based on the facts and their affective response.
A Todd Davis, MD, general academic pediatrician: Parents often ask if a particular diagnosis should have been made sooner. Often, the truth is that it's much easier with hindsight when all of the facts are in front of you to aid in the correct diagnosis. However, in egregious examples such as this, the only possible response is that it should never have happened.
The way the message is heard by the primary care providers is also crucial. While they need to hear the facts, it does no good to alienate them. One of the worst things that can happen is that they stop referring patients who critically need to be in tertiary care pediatric centers.
Dr. Listernick: What is our responsibility in reporting gross negligence or malpractice to the proper authorities? Tell me what you think of this statement: academic physicians often shirk our responsibility to report health care providers who act this irresponsibly.
Robert Tanz, MD, general academic pediatrician: There are several different scenarios. Physicians may make clinical errors, sometimes serious, that are often correctable with education. In those cases, it's appropriate to try to educate the physician. However, if serious errors persist or if there's a potentially catastrophic mistake made, we have an obligation to the children, the public, and our profession to report health care providers who are doing harm. I agree that we don't do it often enough.
Dr. Davis: If we don't police ourselves as professionals, someone else will police us who may not understand all the issues involved.
Dr. Listernick: This practitioner was reported to the Department of Professional Regulation and an investigation is in progress. Thank you, everybody.