Increased amylase and lipase occurs 16–25% of the time in diabetic ketoacidosis (DKA) (1). Acute pancreatitis can present or coexist with DKA and aggravate its severity (2). Nonspecific elevations of amylase in DKA have been reported (3), although increased serum lipase is assumed to indicate actual pancreatic involvement. Two cases of DKA are described in which elevated enzymes were seen without clinical or radiographic evidence of acute pancreatitis.
The first patient was a 74-year-old female with a 32-year history of type 1 diabetes who presented with DKA. Although she had no abdominal pain or tenderness, her amylase and lipase levels were elevated and peaked at values of 1,024 units/l (normal range 17–100 units/l) and 3,455 units/l (normal range <53 units/l), respectively. The enzymes declined precipitously to normal with rehydration and insulin administration. Calcium was 11.7 mg/dl (8.5–10.5) and parathyroid hormone was 232 pg/ml (10–65). It was felt that hypercalcemia due to primary hyperparathyroidism caused acute pancreatitis, precipitating DKA. However, computed tomography (CT) did not show any pancreatic inflammation. The patient had a sestamibi neck scan and underwent successful parathyroidectomy. Several months later she had another episode of DKA but was normocalcemic. The serum amylase was 781 units/l, and lipase was 972 units/l. Transaminases and triglyceride levels were normal. She did not exhibit any abdominal complaints or pancreatic inflammatory changes on CT. Endoscopic retrograde cholangiopancreatography revealed a normal hepatobiliary tree and pancreatic duct. The enzymes normalized with resolution of DKA.
The second patient was a 23-year-old white female with type 1 diabetes who presented with dehydration and DKA. She had an undetectable serum alcohol level and normal calcium, hepatic, and lipid profiles. The amylase level peaked at 406 units/l, and lipase peaked at 2,404 units/l. The patient denied abdominal pain and tenderness on physical examination. Ultrasound of the liver and gall bladder was unremarkable, and CT did not show any inflammation, enlargement, or other changes in the pancreatic area. At an office visit 10 days after discharge, her lipase and amylase levels were normal at 32 and 41 units/l, respectively.
The clinical diagnosis of acute pancreatitis rests on the presence of abdominal pain and associated increases of amylase and lipase, while CT scan is recognized as the gold standard for confirmation. The prevailing opinion is that absolute enzyme elevations greater than three times normal indicate pancreatic involvement (4). Although this may be true in many situations, the above cases and recent studies (1,2) support the notion that such elevations without actual pancreatic involvement may be nondiagnostic in the presence of DKA, as evidenced by absence of abdominal findings and CT scan abnormalities. However, it should not be overlooked that acute pancreatitis can sometimes accompany or precipitate DKA.
The source of elevated enzymes in DKA without acute pancreatitis remains unclear. Subtle injury to the pancreatic acinar cells may liberate them into the circulation. Another possibility is an extrapancreatic origin triggered by the dysmetabolic state, like release of salivary gland amylase, or its accumulation secondary to suboptimal excretion in the urine (3). Increase in lipase may be due to release of nonpancreatic lipolytic enzymes into the bloodstream from sources such as the stomach, liver, small bowel, tongue, esophagus, etc. (5). Some authors have suggested that hyperlipasemia may be related to assay inaccuracy in such cases (6).
In conclusion, significant but nonspecific elevations of amylase can be seen in DKA. Elevated lipase, traditionally thought to be more specific for pancreatitis, may also accompany DKA and does not necessarily denote concomitant pancreatic inflammation. Hyperlipasemia may therefore be less reliable for diagnosing acute pancreatitis in this setting.
