Gupta, Arjun

69 Background: Due to improved survival, an increasing number of patients with advanced cancer are eligible for aggressive ICU care, including intubation. The American Society of Clinical Oncology recommends early integration of palliative care in patients with advanced cancer. We analyzed 10-year trends in the utilization of palliative care, and predictors of palliative care use in patients with metastatic cancer undergoing intubation using the National Hospital Discharge Survey database from 2001-2010. Methods: Diagnosis and procedure codes were utilized to identify adult patients with metastatic cancer who underwent intubation, and received palliative care. Demographics, diagnoses, length of stay (LOS), and discharge information were extracted. T-tests and χ^2 tests were used to compare medians and proportions respectively in patients that did or did not receive palliative care. Multivariate regression models with weighted analysis were conducted to study predictors of palliative care use. Results: Over the 10-year study period, 200,350 patients with metastatic cancer and intubation were identified, mean age 65.3 years, 46.2% male. Of these, 6690 (3.3%) received palliative care. There was a significant increase in the use of palliative care from 2001 (0%) to 2010 (17.4%), p for trend < 0.001. In multivariate analysis, patients who received palliative care were more likely to have age > 65 years, be white vs non-white, be hospitalized in 2009-2010 vs 2000-2001, and be admitted to hospitals in the Western part of the US. The overall inpatient mortality rate was 57.3%, discharge to a care facility (DTCF) rate was 15.2%, and mean LOS was 11.1 days. The LOS and DTCF rates were similar in patients with and without palliative care. Patients with palliative care had higher rates of inpatient mortality (adjusted OR 9.6, 95% CI 2.2- 42.0). Conclusions: There was a dramatic increase in the use of palliative care in intubated patients with metastatic cancer. Studies to address the racial and geographic discrepancies in palliative care use are warranted. The higher mortality in palliative care group suggests it is being used for the sickest patients. Early (vs end of life) palliative care may be more beneficial.

704 Background: Inpatients with gastrointestinal (GI) malignancies are at a high risk for Clostridium difficile infection (CDI) but the impact of CDI on outcomes in these patients needs elucidation. We analyzed the incidence of CDI and its impact on outcomes in GI cancer patients using the National Hospital Discharge Survey (NHDS) database from 2001 - 2010. Methods: NHDS collects clinical information on patients dismissed from non-Federal short-stay United States hospitals. Demographics, diagnoses (GI malignancies, CDI and comorbidities), length of stay (LOS), and dismissal information were abstracted using ICD-9 diagnosis and procedure codes. Weighted analyses were performed using SAS version 9.4. Results: Of an estimated 317.9 million unique hospitalizations; 4.6 million had a diagnosis of a GI malignancy (1.4%); median age 68 years, 46.1% female. CDI was more common in patients with GI malignancies compared to patients with no GI malignancy (1.05% vs 0.69%, aOR 1.16, 95% CI 1.15- 1.17, p < 0.0001). There was a significant increase in CDI incidence in GI cancer patients over the 10-year study period (72.7 in 2001-2002 to 109.1 in 2009-2010, per 10,000 discharges, p < 0001). In multivariable analysis, compared to GI cancer patients without CDI, GI cancer patients with CDI had a longer mean LOS (3.94 more days, 95% CI 3.31-4.56) and dismissal to a care facility (adjusted OR, 1.75; 95% CI, 1.71-1.79), but lower all-cause in-hospital mortality (adjusted OR, 0.76; 95% CI, 0.74-0.79), all p < 0.0001 Conclusions: In this national database of hospitalized patients, an increasing incidence of CDI in patients with GI malignancies was noted over the study period. CDI prolonged hospitalization, and was associated with increased dismissal to a care-facility. Lower rates of in-hospital mortality may represent early diagnosis due to vigilance or a milder form of CDI. Despite increased attention over the last few decades, CDI remained a serious infection in GI cancer patients, and merits appropriate prevention, management and follow-up.

686 Background: Preclinical evidence suggests that statins have anti-neoplastic properties by inhibiting cell proliferation, and inducing apoptosis. Statins may be associated with reduced risk of colorectal cancer (CRC), but their effect on CRC related mortality is unclear. We performed a meta-analysis of published observational studies evaluating statin use in patients with CRC and its association with overall and CRC-specific mortality. Methods: Through a systematic search upto August 2014, we identified 7 observational studies (64,773 patients with CRC;19.5% statin users) reporting the association between statin use and mortality in patients with CRC. Summary hazard ratio (HR) with 95% confidence intervals (CI) was estimated using the random effects model, and heterogeneity was measured using the inconsistency index (I2). Results: On meta-analysis, statin use was associated with a 26% reduction in all-cause mortality in CRC patients as compared to non-use (6 studies; adjusted HR, 0.74; 95% CI, 0.67-0.83) with minimal heterogeneity (I2=18%), after adjusting for age, sex and CRC stage. Only one study adjusted for concomitant metformin use. Five studies evaluated CRC-specific mortality; on meta-analysis, use of statins (vs non-use) was also associated with decreased CRC-specific mortality (adjusted HR, 0.74; 95% CI, 0.70-0.83; minimal heterogeneity, I2=17%). On limiting analysis to statin use after diagnosis of CRC, use of statins remained significantly associated with reduced CRC-specific mortality (3 studies; adjusted HR, 0.70; 95% CI, 0.60-0.81), but not overall mortality (4 studies; adjusted HR, 0.82; 95% CI, 0.64-1.05). On restricting analysis to studies, which adjusted for concomitant use of aspirin/NSAID, the inverse association between statin use and overall- (5 studies; adjusted HR, 0.74; 95% CI, 0.66-0.84) and CRC-specific mortality (4 studies; adjusted HR, 0.76; 95% CI, 0.69-0.84) remained significant. Conclusions: Based on this meta-analysis, statin use in patients with CRC may be associated with reduced all-cause and CRC-specific mortality. This observation merits prospective investigation of statins’ effect on outcome in CRC patients.