This 14-year-old girl was referred for evaluation of thyromegaly noted by her physician. Except for occasional fatigue, she had no complaints. Specifically, there was no history of skin complaints, constipation, cold intolerance, sleep disturbances, weakness, or neurological or psychiatric symptoms. Family history was remarkable for a maternal grandfather who had taken thyroid pills for two years as an adolescent and for "thyroid disease" in a paternal aunt.
Physical examination at presentation was normal except for a thyroid gland that was described by the consulting endocrinologist as being symmetrically and diffusely enlarged, 2.5 times normal, without nodules. Breast development was Tanner 2. Growth and weight gain were normal. On laboratory evaluation, she was found to have a free T4 of 0.88 (normal) and a thyroid stimulating hormone (TSH) of 18.6 (elevated).
She was started on levothyroxine and her fatigue improved. Several months later, it was found that antithyroid peroxidase antibodies and antithyroglobulin antibodies were both significantly elevated.
Ben Katz, MD, moderator: What are the causes of hypothyroidism in childhood?
Don Zimmerman, MD, pediatric endocrinologist: The most common cause of primary hypothyroidism and an enlarged thyroid gland in childhood in developed countries is autoimmune thyroiditis. This was confirmed in this child by the presence of antibodies against two molecules that are components of the thyroid gland. The most common cause of hypothyroidism in the world is iodine deficiency. When I encounter a child with thyroiditis, I always ask myself if there is evidence of other autoimmune diseases. For example, there is an association between thyroiditis and adrenal insufficiency. This is extremely important to consider when starting treatment for any child with hypothyroidism because if the patient also has adrenal insufficiency and you only treat the hypothyroidism, you may precipitate an adrenal crisis. As we follow these children over the years, we periodically perform a variety of screening tests for other autoimmune diseases.
Dr. Katz: Which screening tests do you perform on these patients?
Dr. Zimmerman: I usually will perform a serum calcium and phosphorus to screen for hypoparathyroidism. I usually obtain an early morning serum Cortisol before starting treatment for hypothyroidism. More recent information suggests that we should be looking for gastric parietal cell antibodies, which over time can lead to pernicious anemia.
Stanford Shulman, MD, pediatric infectious disease specialist: Liver function tests also should be monitored periodically. Patients who have the constellation of mucocutaneous candidiasis and a number of autoimmune endocrinopathies also may develop autoimmune hepatitis.
Dr. Katz: What's the significance of the two autoantibodies?
Dr. Zimmerman: We use these antibodies as diagnostic markers. Interestingly, if you infuse them into an experimental animal, there is no thyroid destruction. The pathophysiology of autoimmune thyroiditis is an attack on the thyroid gland by natural killer cells. However, there is a different autoantibody that stimulates the TSH receptor leading to Grave's disease, and yet another autoantibody that prevents TSH access to its receptor and causes hypothyroidism. If either is present in a pregnant woman, the fetus can be adversely affected with transient congenital hyperthyroidism or hypothyroidism, depending on the particular circulating autoantibody.
Dr. Katz: The story continues 1 year later when, after two long soccer games, she began to complain of generalized headache, left-sided weakness, and an inability to feel her tongue. She was taken to a local emergency department, where her physical examination was reportedly normal. They performed a computed tomography (CT) scan of her head, a complete blood count (CBC), and serum chemistries, all of which were normal. A free T4 and TSH were both normal. She received intravenous fluids and her symptoms completely resolved.
One week later following another soccer game, she complained about a headache and right-sided weakness. According to her parents, they noticed a right facial droop for 10 minutes. She was seen at the same emergency department again, where she was reportedly normal and was sent home. She was referred to a neurologist by her endocrinologist.
Given that these symptoms were all associated with exercise, could be cardiac in origin?
Wayne Franklin, MD, pediatric cardiologist: If she had generalized symptoms and signs such as loss of consciousness or dizziness, perhaps the events could be ascribed to an arrhythmia such as long QT syndrome. However, since she has migrating focal neurologic signs, you'd have to postulate that she has an atrial myxoma, intracardiac clot, or bacterial vegetation throwing off microemboli. Such episodes usually occur spontaneously rather than being provoked by exercise.
Dr. Katz: While awaiting her appointment with the neurologist, she returned to school. Several days later, she fell while running in gym class. During the next half-hour, she became increasingly confused and agitated and was taken to the local emergency department, where she was agitated, combative, and intermittently clenching her jaws. There were reportedly no other abnormalities on neurologic examination. Serum chemistries, CBC, ammonia, ethanol level, toxicology screen, chest x-ray, CT scan of the head, electroencephalogram (EEG), and urine pregnancy test were all normal. She was given two doses of lorazepam and a full loading dose of phosphophenytoin because the physicians thought that she was having seizures. She was then transferred to our hospital.
Additional history revealed that she was an athletic, straight-? student. She often camps and rides her bicycle in a wooded area. On examination, she was afebrile with normal vital signs. She was very sleepy but would occasionally open her eyes in response to calling her name and sought out comfortable positions while being examined. She localized and responded to pain normally. The remainder of her general physical examination was normal save for a mildly enlarged, nontender thyroid gland. The neurologic examination was normal save for her mental status.
Marc Wainwright, MD, pediatric neurologist: There are several ways to conceptualize this case. This degree of alteration of mental status must be localized to either the brain stem or both cerebral hemispheres. Coupling this episode with the two previous neurologic events, one would have to assume either a brainstem ischemic episode or multiple bilateral ischemic or embolic events. Alternately, one can conceptualize this case as that of a teenager who has had neurologic deficits separated in time and space; this raises the possibility of multiple sclerosis. Given that she appears to have a diffuse encephalopathy, one should also consider infectious or toxic causes, although the earlier focal events would be difficult to explain. The importance of a detailed social history in teenagers cannot be overemphasized. Conversion disorders can certainly mimic neurologic injuries. Finally, migraine headaches can be a great fooler, particularly basilar migraines, which can present with a variety of neurologic symptoms.
Robert Listernick, MD, general academic pediatrician: Being the eternal skeptic, given that the EEG was reportedly normal, I would think that a conversion disorder would be near the top of the differential diagnostic list. Also, how would multiple sclerosis produce such an acute encephalopathy?
Dr. Wainwright: The lesions would have to be in the brainstem.
Dr. Shulman: What about the possibility of a hypercoagulable state leading to multiple thromboses?
Dr. Wainwright: That's certainly a possibility. However, this may be difficult to prove during the acute event. The levels of many of these factors - protein C, protein S, antithrombin-3, and so on - are altered by the ischemic event itself. To prove their actual contribution to the genesis of the stroke, these factors need to be shown to be abnormal 2 to 4 months after recovery.
Dr. Katz: After being transferred here, she had a normal electrocardiogram and echocardiogram, and negative antibody titers for arboviruses, enteroviruses, and varicella zoster virus. The results of a lumbar puncture were normal save for a mildly elevated cerebrospinal (CSF) fluid protein of 64 mg/dL. Herpes simplex virus DNA was not detected. The erythrocyte sedimentation rate was 9 mm/hour. Antinuclear antibody, C3, C4, and total hemolytic complement were normal.
She underwent endotracheal intubation because of worsening mental status, and she was given intravenous acyclovir. She never displayed any seizure activity. Although she became febrile the following morning, her mental status improved, and she began following commands. An EEG showed diffuse slowing, but no epileptiform discharges. The following day we were notified that an enteroviral polymerase chain reaction (PCR) test for RNA was positive in the cerebrospinal fluid.
Dr. Shulman: I should say that when this child arrived, encephalitis due to an infectious agent was low on my diagnostic list mainly because fever had not been part of the story. In addition, the three separate neurologic events over time with intervals of being normal definitely didn't suggest infection. The modestly elevated CSF protein is very nonspecific. Having said that, we did recommend treatment with acyclovir initially because herpes simplex encephalitis is treatable. Although symptoms of herpes encephalitis can wax and wane, it would be extremely unusual for them to do so over a month as in this child.
Dr. Listernick: Why send all the viral diagnostic studies if they really weren't a consideration?
Dr. Shulman: I didn't recommend them. Looking for herpes simplex DNA is reasonable because one can have brain biopsyproven herpes encephalitis with few or no cells in the spinal fluid. The presence of enteroviral RNA is problematic. While enteroviral season in Chicago is the summer, it had been a very mild fall, so it's conceivable that we could see enteroviral disease this late, at the end of October. Certainly, enteroviral disease needs to be considered in any patient with an obscure central nervous system disorder in the summer.
Dr. Katz: What about the possibility of enteroviral encephalitis in this patient?
Dr. Shulman: Enteroviral central nervous system infections are predominantly aseptic meningitis with little or no brain involvement. She clearly didn't have meningitis, as there were no cells in the cerebrospinal fluid. When one sees a positive lab result that really doesn't fit the clinical setting, one has to think long and hard before accepting the result as fact. Even in an excellent laboratory such as ours, the possibility of a false positive PCR test is real, as the potential for contamination of specimens is great given the extreme sensitivity of PCR methodology.
Dr. Katz: The following day she became more cooperative but was still sleepy. The thyroid antimicrosomal antibody titer was found to be 1:100,000.
Gesina Keating, MD, pediatrie neurologist: I had the benefit of approaching this child for the first time after she had been in the hospital for several days and had been seen by many different consultants. Stepping back to look at the whole picture, one diagnosis to entertain once the others that have been discussed have been eliminated is Hashimoto's encephalopathy. Such children may be euthyroid with normal thyroid function tests.
Dr. Katz: How do you establish this diagnosis?
Dr. Keating: First, it's important to exclude the other treatable causes of encephalopathy that have been mentioned. Recurrent neurologic events are quite common in this syndrome, as is a high CSF protein. The presence of a high thyroid antimicrosomal titer in a child with a compatible clinical syndrome as described is sufficient to establish the diagnosis.
Dr. Katz: Is there effective treatment?
Dr. Keating: The literature suggests that most patients respond favorably to intravenous methylprednisolone. If this fails, plasmapheresis has been reported to be effective.
Dr. Katz: What's the prognosis?
Dr. Keating: Looking at the literature initially, one gets the impression that the patients do very well. However, on closer examination, there's as high as a 30% relapse rate; some pediatric patients may be left with cognitive defects and seizures. Many neurologists slowly taper the steroid medication in these patients to try to prevent a relapse. Some provide longer immunosuppression with a drug such as azathioprine.
Based on our collective experience, we decided to start plasmapheresis immediately. Her mental status continued to improve nicely during the next several weeks. At the last follow-up, she was very close to her premorbid neurologic and behavioral baseline.
Dr. Katz: Thank you, everybody.