Unsupervised clustering of serum lipase activity in cats: a data-driven approach to correlate clinical, laboratory, and ultrasonographic findings.

Elevated feline serum lipase lacks clinical specificity; ultrasonography needed for pancreatitis confirmation.

This retrospective study analyzed serum lipase activity in 563 hospitalized cats using unsupervised k-means clustering to establish clinically relevant lipase ranges and correlate findings with ultrasonographic and clinical presentations. Three distinct clusters emerged: mild elevation (53 U/L), moderate elevation (229 U/L), and marked elevation (778 U/L). Ultrasonographic pancreatitis diagnosis was significantly more common in the moderate cluster (64%) compared to mild (47%), with pancreatic enlargement, hypoechogenicity, and hyperechoic mesentery being more prevalent in moderate elevations. However, 40-50% of moderately elevated and 25-66% of markedly elevated lipase cats lacked ultrasonographic pancreatic abnormalities. Clinical signs were minimal across clusters, with only lethargy significantly associated with moderate lipase elevation. Laboratory abnormalities included elevated urea in the highest cluster, elevated triglycerides in the moderate cluster, and decreased protein in the highest cluster. The findings suggest that serum lipase elevation does not reliably predict pancreatic pathology or clinical signs in cats, and approximately half of cats with hyperlipasemia lack ultrasonographic confirmation of pancreatitis.

This summary was distilled by AI and may occasionally misinterpret data. Confirm critical details with the primary literature before clinical application.