P2130 - Disparities in Referral and Surveillance Patterns for Incidental Pancreatic Cysts

Komal Khoja, BA¹, Stacy O'Connor, MD, MPH, MMS², Susan Tsai, MD, MHS, FACS², Zachary Smith, DO^2
1Medical College of Wisconsin, Morton Grove, IL; ²Medical College of Wisconsin, Milwaukee, WI

Introduction: Pancreatic cystic lesions (PCLs) are often found incidentally on cross-sectional imaging. PCLs that have malignant potential, should undergo appropriate guideline-based surveillance. The rate and pattern of surveillance and referral of newly identified incidental PCLs on imaging obtained in the outpatient and emergency department (ED) setting is unknown. This study aimed to analyze surveillance and referral patterns of new incidental PCLs.

Methods: All cross-sectional imaging radiology reports obtained in the outpatient and ED settings during 2016-2017 in a large academic health system were queried for variations of "pancreas" and "cyst". Each chart was manually reviewed, and patients were included if this was the first-ever identification of a PCL. Patient and cyst characteristics were recorded. Radiology reports were reviewed for language pertaining to follow-up imaging or referral recommendations and whether a differential diagnosis was rendered. Univariate analyses were performed on all covariates, and a logistic regression model was fashioned retaining covariates with p< 0.25. The primary outcome was a dichotomous “appropriate/inappropriate” referral defined as seeing a PCL specialist or undergoing surveillance cross-sectional imaging within 12 months of the index scan.

Results: A total of 240 patients were identified. Patient characteristics and univariate results are present in Table 1. Over the entire cohort, 36.7% of PCLs were not appropriately referred for expert consultation or surveillance. Univariate analyses demonstrated significantly higher rates of appropriate referral/surveillance for outpatient vs ED scans, reports containing follow-up recommendations from the dictating radiologist, private insurance status, and dilated pancreatic duct (PD) caliber (Table 1). Multivariable logistic regression showed that outpatient versus ED scans (OR 3.0, 95% CI 1.5 – 6.4, P = 0.003), and dilated PD (OR 5.1, 95% CI 1.3 – 20.9, P = 0.023) were independently associated with appropriate PCL referral.

Discussion: In the ED and outpatient setting, incidental new PCLs on cross-sectional imaging studies are referred for surveillance or expert consultation at suboptimal rates. Because the prevalence of PCLs is high, these insufficiencies could result in missed opportunities to prevent malignant transformation of PCLs, or to resect early cancers.



Disclosures:


Komal Khoja, BA¹, Stacy O'Connor, MD, MPH, MMS², Susan Tsai, MD, MHS, FACS², Zachary Smith, DO². P2130 - Disparities in Referral and Surveillance Patterns for Incidental Pancreatic Cysts, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.