LAUSR

I read with interest a recent article published online in your journal by Lee et al (1). The authors evaluated the value of multiphasic computed tomography (CT) in patients of gastrointestinal bleeding (GIB) with hemodynamic instability and concluded that multiphasic CT is useful in patients with lower GIB before embolization. I would like to share my experience of interventional management of GIB at a tertiary care center, particularly related to pancreatitis. GIB, both upper and lower, are wellknown complications of pancreatitis with pseudoaneurysm being the cause in the majority of patients (2). Over the past eight years, I have performed angiography for pancreatitisrelated acute GIB in 125 patients (109 men, 16 women; mean age = 35.7 years) (Table 1). Of them, 45 patients were hemodynamically unstable at presentation (systolic blood pressure <100 mmHg) and 80 patients were stable. CT angiograms (CTA) were performed in 42 (93.3%) and 74 (92.5%) patients, respectively. In patients with hemodynamic instability, CTA and digital subtraction angiogram (DSA) localized the lesion in 25 (55.6%) and 32 (71.1%) patients, respectively. Similarly, in the stable group, CTA and DSA were positive in 60 (75%) and 65 (81.3%) patients, respectively. Thus, the CTA localization of the bleeding site was significantly lower in patients with hemodynamic instability when compared to stable patients (P= 0.015). In my short experience, CTA seemed to delay the DSA procedure although it did not affect the outcome of the patients. I feel that the reason for lower rates of localization on CTA in unstable patients is mainly the attenuated arteries leading to negative findings (3). On the other hand, DSA is performed with direct injection of contrast into the arteries which better identifies the lesion. Lee et al. also suggested that in the presence of a known cause of GIB, CTA may be avoided to reduce the time to angiography (1). However, we still continue to perform CTA before DSA even in hemodynamically unstable patients as it shortens the time to DSA (when CTA is positive) and the DSA procedure itself by identifying the source in many patients. Nonetheless, larger studies are needed for a definitive answer. Table 1. Comparison of parameters between hemodynamically stable and unstable patients of gastrointestinal bleeding.