PEDIATRIC PUZZLER

GEORGE K. SIBERRY, MD, MPH, SECTION EDITOR

A 9-year-old boy with a first seizure:
Time to excel at diagnosis!

By Jay Berry, MD, and Paul Young, MD

You've been paged tonight by the emergency department (ED) while working the general medical ward of a children's hospital. They are admitting a 9-year-old boy who was transferred from a community hospital after a seizure. The ED fellow tells you that it was the child's first seizure and that the referring ED detected a low blood glucose level.

"Do we know why it was low?"

"No idea," is the response you get.

You introduce yourself to the parents in the examining room and obtain the following history: The family was having a "slumber party" in the den of their home when they awoke to find the boy in the midst of an apparent convulsion. He was rhythmically shaking both arms and legs, they report, had left facial twitching and left eye deviation, and was turning blue around the mouth. The father called 911. The parents estimate a total seizure time of approximately two minutes. The boy did not experience bowel or bladder incontinence.

Upon further review of systems, you discover that he had not had a fever at home before the seizure or complained of headache, photophobia, stiff neck, or problems with vision, including double or blurred vision or field defects. His parents and siblings were in contact with him throughout the day and night before the seizure, and no change had been observed in his mental status before the seizure. No head trauma had preceded the seizure. His activity level and appetite had not been diminished and he had not lost weight recently.

The boy's medical history provides additional information. His parents are quite certain that he has never had a seizure before. The same is true, they tell you, for head injury. Nor has he ever been given a diagnosis of a neurologic disorder. He takes several medications, including oxybutynin (Ditropan XL) one hour before bed for nocturnal enuresis, and Adderall, a formulation of mixed amphetamine salts, each morning for attention deficit hyperactivity disorder. In addition, he has been followed closely by a pediatric cardiologist for an asymptomatic junctional rhythm that was found incidentally during a physical examination at school. He does well in school, his parents report.

You determine that there is no family history of seizures, neurologic disorders, early infant death, or metabolic disease. The patient lives with his parents and siblings. He has had no sick contacts at home.

Hypoglycemia— but is that all?

Before you examine this patient, you take a minute to review the records from the referring hospital. On arrival by ambulance to their ED, he was described as "sleepy but arousable." The physical exam did not reveal any abnormalities. On a basic metabolic profile, electrolytes were normal except for a blood glucose level of 30 mg/dL. He was immediately given a bolus of intravenous dextrose, which raised the blood glucose level to 61 mg/dL. The physicians there noted that he remained quite sleepy and difficult to arouse.

You record a temperature of 36.5° C; pulse, 87/min; respirations 16/min; and blood pressure, 100/59 mm Hg. Oxygen saturation is >95% breathing room air. The patient is sleepy but arousable. Pupils are equal and reactive to light; extraocular movements are intact; and there is no papilledema. His head exhibits no signs of trauma. There is no nuchal rigidity. The oropharynx is clear and mucous membranes are moist. There is no lymphadenopathy. The cardiovascular exam reveals an irregularly irregular rhythm; 2+ pulses throughout; and a capillary refill time of less than two seconds.

Lungs are clear to auscultation bilaterally, without retractions. He is not jaundiced and does not have hepatosplenomegaly. The abdomen is soft and nontender. Cranial nerves are intact. Muscle tone is normal; deep tendon reflexes are 2+. His strength is 4/5 throughout, and tests of sensation show no focal deficit.

The next step? You decide to monitor the boy's blood glucose level often; to obtain a more detailed history from him and his family; and to decide what, if any, additional laboratory tests will help clarify the cause of his seizure and hypoglycemia. You order hourly measurement of blood glucose and continue the IV maintenance fluids with dextrose begun in the ED.

Thinking things through

What caused this boy's seizure? Nothing in the history suggests meningitis or any process that can raise intracranial pressure, and neither your physical exam nor the neurologic exam offers clues. The single abnormality is the blood glucose level of 30 mg/dL. Hypoglycemia can cause a seizure, but why was he hypoglycemic? That differential diagnosis is very extensive. You opt for consultation from the endocrinology service. Awaiting their input, you begin to pursue the common causes of hypoglycemia.

Could this be hypoglycemia secondary to fasting, if he has a condition in which fasting is poorly tolerated? You review what he ate before the seizure. His parents report that he has a "normal" appetite and ate a typical-sized dinner before falling asleep the night of the seizure. He has never had a seizure or any other symptoms suggestive of hypoglycemia after fasting between supper and breakfast, they assure you. Reviewing the record from the referring ED, you note that the urinalysis did not show ketones, which would be expected if the hypoglycemia was associated with fasting.

Could this be the result of an ingestion? Certain drugs and substances can lead to hypoglycemia. Lacking a better explanation, you reason that ingestion should be strongly considered in the differential. Accidental ingestion seems unlikely, given the patient's age—although in a younger child this would be a strong consideration. What about intentional ingestion? You review with his parents, in detail, all medications in the house, looking for any that could produce hypoglycemia. The patient and the family deny that he ingested or was exposed to any potential hypoglycemic agent, including alcohol and aspirin. An older sibling, interviewed separately, denies that the patient uses alcohol or drugs. There is no insulin or oral hypoglycemic medication in the house. On the strength of the family's report, you elect not to order toxicology tests.

Could this have an endocrine or metabolic basis? What about adrenal or pituitary insufficiency? Looking for clues in the physical exam, you note normal genitalia and no hyperpigmentation of skin creases. Height and weight are appropriate for age. Blood pressure is normal. Blood drawn in the ED was sent for a cortisol level: The result was within normal limits.

During this time, results of hourly tests of the blood glucose level have all been >70 mg/dL. The boy is still sleepy but easily aroused. His neurologic exam continues to show no focal deficits and he exhibits no further seizure activity. Given that he did not have ketonuria, you pursue causes of nonketotic hypoglycemia. Hyperinsulinemia caused by nesidioblastosis or insulinoma can present with hypoglycemia; these disorders manifest as an elevated serum insulin level. You order a test of the insulin level: The result is within normal limits. You plan to test the insulin level again if the blood glucose level drops below 70 mg/dL—but that does not happen. The boy becomes fully awake and alert.

A regular diet is initiated, which the boy tolerates without difficulty. A repeat physical exam is unchanged. You begin to wonder about the significance of the presenting hypoglycemia: Did the cart come before the horse? Could the seizure have caused the hypoglycemia? Proceeding with the routine work-up of a seizure, an electroencephalogram is ordered. You decide to involve genetics along with endocrinology, to get any other recommendations on how to pursue the hypoglycemia.

Both specialists concur: The patient should be challenged with a fast to see if the hypoglycemia can be reproduced; if it can, appropriate studies will then be ordered. The boy is given nothing by mouth, dextrose-containing IV fluids are stopped, and the blood glucose level is measured hourly. After 18 hours of fasting, the level remains above 70 mg/dL! The test is discontinued and the patient is allowed to eat. A metabolic cause of the hypoglycemia is unlikely.

A decision is made at this point to keep the patient in the hospital until the EEG is performed the next day. Frustrated, the boy's parents continue to replay the events of the night before for clues to the cause of the seizure. In doing this, his mother recalls something that jump-starts your diagnosis.

A pill's not always just what the doctor ordered

Two or three days before the seizure, the boy's mother relates, his Ditropan XL prescription was refilled. The pills from the new prescription, she says, were not the same size as the pills from the previous Ditropan XL prescription. She asked about the difference at the pharmacy, where "the person at the counter" told her that the new prescription was "generic." The boy ingested one pill from the new prescription an hour before bed on the night of the seizure.

You obtain the pills from the new prescription from the family and have them examined by a hospital pharmacist. They are Glucotrol XL, a formulation of the oral hypoglycemic drug glipizide. Without food during the night, and having taken the glipizide, the boy's blood glucose level dropped, causing the seizure. The pieces of the puzzle fit!

To err is . . . possible

You've heard on the news and read in the literature that medical errors are common, but you haven't had time to read an Institute of Medicine report in 2000, alleging that adverse events due to medical error occur in 3% of all hospitalizations and that more people die yearly in the US from medical error than from injuries sustained in motor vehicle crashes!¹ A brief literature search reveals that hypoglycemia secondary to a pharmacy dispensing error has been reported before; you find four case reports that describe hypoglycemia secondary to ingestion of acetohexamide that was erroneously substituted for acetazolamide in adults.^2–5 Either a physician's illegible handwriting or a pharmacy error could have led to improper dispensing of those two drugs.

But what could have happened in the case of your patient? Neither the generic names (oxybutynin, glipizide) nor the primary word in the brand names (Ditropan, Glucotrol) are remotely similar to each other. The two pills have dissimilar colors and shapes, and they are manufactured by different pharmaceutical companies and carry glaringly different FDA-approved indications. The only similarity between the two drugs? They are "extended-length" formulations, a fact that is signaled by the terminal letters "XL" in both brand names. And there, in that minor similarity, you must assume, is the origin of the mistaken substitution.

What can be learned from this?

Pharmacy dispensing errors in pediatrics are rare, but they do occur. In 2001, Kaushal reviewed more than 10,000 medication orders for pediatric inpatients and found six pharmacy dispensing errors.⁶ Data on dispensing errors among pediatric outpatients have not been reported, but you now know of one case that resulted in hypoglycemia and a seizure. The history in this boy's case would have revealed the diagnosis had someone posed the right questions to the parents: "Have any of your son's prescriptions been filled recently?" and "Are the pills from the new prescription different from the old pills—in shape, size, or color?"Resolved, then: To "excel" at your efforts, consider whether it is pertinent to ask those kinds of questions the next time a diagnosis proves so elusive.

REFERENCES

1. Kohn LT, Corrigan JM, Donaldson MS (eds): To Err Is Human: Building a Safer Health System. Washington, D.C., Institute of Medicine/National Academy Press, 2000

2. Hargett NA, Ritch R, Mardirossian J, et al: Inadvertent substitution of acetohexamide for acetazolamide. Am J Ophthalmol 1997;84:580

3. Raab EL: Substitution of acetohexamide for acetazolamide. Am J Ophthalmol 1979;87:848 (letter)

4. Olson LA, Miller DA, Goswami A, et al: Inadvertent administration of acetohexamide instead of acetazolamide. DICP 1991;25:100 (letter)

5. Sledge ED, Broadstone VL: Hypoglycemia due to a pharmacy dispensing error. South Med J 1993;86:1272

6. Kaushal R, Bates D, Landrigan C, et al: Medication error and adverse drug events in pediatric inpatients. JAMA 2001;285:2114

DR. BERRY is a pediatric resident at the University of Utah School of Medicine, Salt Lake City.

DR. YOUNG is professor of pediatrics and medical director of the pediatric clinic at University Hospital, University of Utah School of Medicine, Salt Lake City.

DR. SIBERRY is a fellow in pediatric infectious disease at The Johns Hopkins Hospital, Baltimore, Md.

George Siberry, ed. Jay Berry, Paul Young. Pediatric Puzzler: First seizure and hypoglycemia.

Contemporary Pediatrics

June 2003;20:22.