Severe hypertriglyceridemia and factors associated with acute pancreatitis in an integrated health care system.

OBJECTIVE

To evaluate patient characteristics, treatment patterns, comorbidities, and risk factors associated with the development of acute pancreatitis (AP) in patients with severe hypertriglyceridemia (HTG) in an integrated health care delivery system.

METHODS

We identified a retrospective cohort of severe HTG patients with a fasting triglyceride level ≥ 1000 mg/dL during January 1, 2007 to June 30, 2013 (index date) in an integrated health care delivery system. Patients were aged ≥18 years on index date and had 12 months of continuous membership and drug eligibility before the index date and during postindex including index date. Baseline patient characteristics, comorbidities, and risk factors were evaluated during 12-month preindex. Outcomes such as development of AP, treatment patterns, adherence to index therapy, and change in triglyceride (TG) laboratory levels were evaluated during postindex. Descriptive statistics were used to identify differences between patients developing AP vs no development of AP. A stepwise multivariate logistic regression and backward elimination method were used to assess statistically significant predictive factors associated with development of AP vs no AP.

RESULTS

We identified 5550 patients with severe HTG, and 5.4% of these patients developed AP during postindex. Patients were mostly male (≥70%) in both groups; however, younger in the AP group (45 years ± 10.6) vs no AP group (50 years ± 11.4) with P value < .0001. The AP group had higher baseline Charslon Comorbidity Index score, alcohol abuse history (42.2%), any pancreatitis history (51.5%), diabetes (47%), and hypertension (55%), vs the no AP group (P values < .05). Patients in the AP group had higher baseline mean TG levels (2148, SD ± 1578) vs the no AP group (1559, SD ± 861), P value < .0001. Over 50% of the patients were prescribed their index therapy by a primary care provider. Predictive factors associated with the development of AP included younger age, alcohol use, and prior history of any pancreatitis, hypertension, renal disease stage 4, and other prescriber specialty. From parameters estimates, for each 100 mg/dL unit of increase in the TG level above 1000 mg/dL, there was a 3 percent increase in risk of developing AP.

CONCLUSIONS

Patients with severe HTG are at a higher risk of developing AP. A number of comorbidities, risk factors, and baseline TG levels are associated with an increased incidence of AP. Patients with severe HTG are underdiagnosed, undertreated and are nonadherent to their index lipid therapy. There is a need to better define optimal approaches to treating severe HTG so as to reduce the incidence of AP. Economic studies are also needed to evaluate the burden of AP on various health care systems.