Search

Objective Describe the association of USMLE Step 1 scores and the institution of a dedicated board review curriculum with resident performance on the Otolaryngology training examination. Study design Retrospective cross sectional study. Methods We reviewed American Board of Otolaryngology Training Examination (OTE) scores for an otolaryngology residency program between 2005 and 2016. USMLE Step 1 scores were collected. In 2011 a resident-run OTE review curriculum was instituted with the goal of improving test preparation. Scores were compared before and after curriculum institution. Linear regression was performed to identify predictors of OTE scores. Results 47 residents were evaluated, 24 before and 23 after instituting the curriculum. There was a moderate correlation between USMLE step 1 scores and OTE scores for all years. For PGY-2 residents, mean OTE scores improved from 25th percentile to 41st percentile after institution of the review curriculum (p = 0.05). PGY 3–5 residents demonstrated no significant improvement. On multivariate linear regression, after controlling for USMLE step 1 scores, a dedicated board review curriculum predicted a 23-point percentile improvement in OTE scores for PGY-2 residents (p = 0.003). For other post-graduate years, the review curriculum did not predict score improvement. Conclusion USMLE step 1 scores are moderately correlated with OTE performance. A dedicated OTE review curriculum may improve OTE scores for PGY-2 residents, but such a curriculum may have less benefit for intermediate- and senior-level residents. Level of evidence 4.

Bcl-2 is a tumorigenic protein that is expressed in 25% to 50% of breast cancers. Although its expression has been widely accepted as a favorable prognostic marker, its protective mechanism of action remains unclear. "Triple-negative" tumors are an aggressive subgroup known to carry a poor prognosis. Studies documenting prognostic significance of Bcl-2 expression in triple-negative in comparison to non-triple-negative breast cancers are limited. Bcl-2 expression was correlated with tumor size, grade, histologic type, lymphovascular invasion, lymph node status, patients' overall survival, estrogen receptor, progesterone receptor, Her-2, p53, and epidermal growth factor receptor in 124 triple-negative and 458 non-triple-negative tumors. There were significant differences between triple-negative and non-triple-negative tumors in their relationship to Bcl-2 expression (81% versus 29%, respectively) and tumor aggression. As previously reported, in non-triple-negative tumors, Bcl-2 positivity correlated with less aggressive tumors (94% of grade I tumors were Bcl-2+ versus 62% of grade III tumors, P < .011) and overall survival (P = .008). However, the opposite was true in patients with triple-negative tumors, where Bcl-2 positivity was associated with poorer survival (P = .64). In triple-negative tumors, Bcl-2 positivity was not associated with any of the aforementioned parameters except for a lower incidence of lymph node metastasis. Moreover, by Cox regression analysis of all variables, in patients with triple-negative tumors, lymphovascular invasion (P = .009) and Bcl-2 expression (P = .028) were predictors of poor survival. In conclusion, there are major clinicopathologic differences between breast cancer phenotypes. Our results establish the value of using Bcl-2 in prognostic stratification of patients and its potential therapeutic implications in selecting patients for treatment.