Recurrent Acute Pancreatitis Associated with Propionic Acidemia

Propionic acidemia (PA) is a severe metabolic disorder of infants and children (1), resulting from a defect in the mitochondrial enzyme propionyl-coenzyme A carboxylase (PCC). PA is inherited in an autosomal recessive manner. Patients experience metabolic acidosis, ketosis, seizures, and mental retardation. Acute pancreatitis is a potentially life-threatening inflammatory disorder that has many known inciting factors, including metabolic derangements (2,3). However, although acute pancreatitis has been associated with PA (4,5), recurrent acute pancreatitis is rare. We report a case of an infant with PA in whom recurrent acute pancreatitis developed. CASE REPORT The patient was an 18-month-old infant girl who was brought to Yale-New Haven Hospital in October 2000 with abdominal pain and vomiting. Her medical history was significant for PA. At 3 days of life, she had tachypnea, poor feeding, hypotonia, and metabolic acidosis. At that time, a urine organic acid profile revealed the excretion of methylcitrate and 3-hydroxy-propionate. Propionic acidemia was confirmed by the demonstration of reduced PCC activity in cultured fibroblasts. Her condition stabilized after resuscitation with intravenous fluids and glucose. Under treatment with Propimex, Pro-free, and low-fat milk, she achieved good metabolic control. At 18 months of age, she was brought to the hospital with a 1-day history of vomiting and abdominal pain. She was not taking any medications, and there was no history of trauma. Her vital signs were normal, and she was in no respiratory distress. She was anicteric. Her abdomen was diffusely tender without guarding or rebound. There was no rash or edema. She had normal capillary refill. Her electrolytes were unremarkable (Na 137 mmol/L, K 4.4 mmol/L, HCO3 27 mmol/L, Cl 97 mmol/L, serum urea nitrogen 5 mg/dL, Cr 0.4 mg/dL, serum glucose 67 mg/dL.). Her white blood cell count was 15 × 103/μL, and her hemoglobin was 11 mg/dL. However, her serum amylase (551 U/L, normal 40–150) and lipase (7.9 U/ml, normal <1.5) were greater than 3 times the upper limit of normal. Abdominal ultrasound showed a diffusely edematous pancreas. No biliary stones were noted, and the common bile duct was of normal size. The gallbladder was normal. The serum Ca2+ was normal (9.9 mg/dL; normal, 8.8–10.2), and the triglyceride level was mildly elevated (mean, 309 mg/mL; normal, 30–150). The patient's episode of acute pancreatitis improved with bowel rest, intravenous fluids, and meperidine. However, acute pancreatitis recurred 6 months later. Her abdominal pain and serum pancreatic enzyme levels resolved after 3 weeks with bowel rest, parenteral nutrition, and meperidine. At 24 months of age, acute pancreatitis recurred, with similar clinical and biochemical presentations. The pain episode lasted 4 weeks, despite supportive treatment. She went on to have 2 additional episodes at ages 52 and 64 months, each lasting between 2 and 5 weeks (Table 1). Her last episode was at 5 years of age, and she has had no further episodes at the time of this report, when she was 7 years old. None of her episodes of pancreatitis was associated with metabolic acidosis. She had no evidence of chronic pancreatitis; she had normal growth, no steatorrhea, and no pancreatic calcifications on ultrasound. The result of a sweat test was within normal limits. There was no family history of pancreatitis or pancreatic cancer.TABLE 1: Age during episodes of pancreatitis, supportive serum pancreatic enzyme elevations, and duration of each episodeDISCUSSION We report a case of PA with acute recurrent pancreatitis. To our knowledge, there are only 3 previously reported patients (4,5). Marsden et al (4) reported a case of a 7-year-old girl with PA who experienced 2 episodes of pancreatitis and died during her second episode as a result of basal ganglia infarction. Notably, she had no metabolic instability preceding the acute terminal period. Burlina et al (5) reported 2 cases of PA. A 3-year-old boy with PA experienced 2 episodes of acute pancreatitis 1 month apart. At least 1 episode was associated with ketonuria and lasted 2 weeks. The second case involved a 6-year-old girl who had diffuse pancreatitis at autopsy. She died of shock, and it is unclear whether the pancreatitis was a result of ischemia or multiorgan dysfunction because she had normal serum pancreatic enzyme levels and no radiographic evidence of pancreatitis at presentation. Our patient with PA and acute pancreatitis had an earlier age at onset than the patients previously reported. She had no metabolic derangements preceding her pancreatitis episodes. She had 2 additional episodes at ages 52 and 64 months, a course similar to that in the 3-year-old patient reported by Burlina et al (5). Finally, the severity of our patient's pancreatic episodes was mild to moderate. On the basis of these limited trends, acute pancreatitis in the setting of PA may have a longer clinical and biochemical duration than pancreatitis of other causes and can occur without metabolic acidosis. What may have caused pancreatitis in our patient? We excluded common causes such as common bile duct stones, ductal defects, medications, and trauma. Thus, we speculate that the metabolic defect in PA may be a primary contributor. Patients with PA have both cellular and systemic metabolic derangements that are a direct result of a defect in the mitochondrial enzyme PCC (1). Substrates such as odd chain fatty acids have been shown to accumulate (6). It is well known that hypertriglyceridemia is a risk factor in clinical pancreatitis (7), and both animal and cellular models of pancreatitis suggest that elevations in free fatty acids sensitize the pancreas to pancreatitis (8,9). Thus, a possible explanation for pancreatitis in this child could be the fatty acid upregulation observed in PA. It is notable that our patient had mild serum triglyceride elevations (mean 309 mg/mL). By itself, the triglyceride level does not seem to explain her pancreatitis episodes because in most reports linking hypertriglyceridemia to pancreatitis, patients had levels >1000 mg/dL (7). In the absence of PCC, propionyl-coenzyme A is shunted to pathways that produce acidic byproducts that can accumulate within cells and lead to systemic ketoacidosis. Recently it was shown that cellular acidosis is a sensitizing factor for the pathological activation of pancreatic protease proenzymes, or zymogens, within the pancreatic acinar cell, which is an early and critical event in pancreatitis (10). Even though our patient's episodes of pancreatitis did not correlate with systemic acidosis, the cellular pH within the pancreas could have been low, despite treatment for PA. Thus, acidosis within the pancreatic acinar cell or its cellular compartments may have predisposed our patient to pancreatitis. Mitochondrial dysfunction may provide yet another explanation (11). It has recently been appreciated that mitochondria play a critical role not only in generating adenosine triphosphate but also in buffering cytosolic calcium elevations within physiological levels in the pancreatic acinar cell (12). Mitochondrial dysfunction predisposes acinar cells to high globalized levels of calcium that are associated with pancreatitis. Thus, patients with PA may be predisposed to pancreatitis as a result of mitochondrial dysfunction. In summary, we report a case of a patient with PA who experienced acute pancreatitis, and we suggest possible explanations for the association. Prevention of pancreatitis in these patients requires the development of new insights into the pathophysiology involved. This discussion offers new possibilities for study of the development of acute pancreatitis in organic acid disorders. Finally, a diagnosis of acute pancreatitis should be considered in patients with PA who have abdominal pain.