Clinical Characteristics and Prognostic Factors of Posterior Segment Intraocular Foreign Body: Canadian Experience from a Tertiary University Hospital in Quebec

Purpose. To identify predictive factors for visual outcomes of patients presenting with a posterior segment intraocular foreign body (IOFB). Methods. A retrospective chart review was performed for all consecutive patients operated for posterior segment IOFB removal between January 2009 and December 2018. Data were collected for patient demographics, clinical characteristics at presentation, IOFB characteristics, surgical procedures, and postoperative outcomes. A multiple logistic regression model was built for poor final visual acuity (VA) as an outcome (defined as final VA 50 letters or worse [Snellen equivalent: 20/100]). Results. Fifty-four patients were included in our study. Ninety-three percent of patients were men, with a mean age of 40.4 ± 12.6 years. Metallic IOFB comprised 88% of cases with a mean ± standard deviation (SD) size of 5.31 ± 4.62 mm. VA improved in 70% of patients after IOFB removal. Predictive factors for poor VA outcome included poor baseline VA, larger IOFB size, high number of additional diagnoses, an anterior chamber extraction, a second intervention, the use of C3F8 or silicone tamponade, and the presence of vitreous hemorrhage, hyphema, and iris damage. Predictive factors for a better visual outcome included first intention intraocular lens (IOL) implantation and the use of air tamponade. In the multiple logistic regression model, both baseline VA ( p = 0.009) and number of additional complications ( p = 0.01) were independent risk factors for a poor final VA. Conclusions. A high number of concomitant complications and poor baseline VA following posterior segment IOFB were significant predictive factors of poor visual outcome.

Development and validation of a Brazilian nomogram to assess sentinel node biopsy positivity in melanoma

Background: Although well-established, sentinel node biopsy (SNB) for melanoma is not free from controversies and sometimes it can be questionable if SNB should be considered even for patients who meet the criteria for the procedure. Mathematical tools such as nomograms can be helpful and give more precise answers for both clinicians and patients. We present a nomogram for SNB positivity that has been internally validated. Methods: Retrospective analysis of patients who underwent SNB from 2000 to 2015 in a single institution. Single logistic regressions were used to identify variables that were associated to SNB positivity. All variables with a p value < 0.05 were included in the final model. Overall performance, calibration, and discriminatory power of the final multiple logistic regression model were all assessed. Internal validation of the multiple logistic regression model was performed via bootstrap analysis based on 1000 replications. Results: Site of primary lesion, Breslow thickness, mitotic rate, histologic regression, lymphatic invasion, and Clark level were statistically related to SNB positivity. After internal validation, a good performance was observed as well as an adequate power of discrimination (area under the curve 0.751). Conclusions: We have presented a nomogram that can be helpful and easily used in daily practice for assessing SNB positivity.

“Big data” and Gram-negative Resistance: A Multiple Logistic Regression Model Using EMR Data to Predict Carbapenem Resistance in Patients with Klebsiella pneumoniae Bloodstream Infection

Background The timely identification of carbapenem resistance is essential in the management of patients with Klebsiella pneumoniae bloodstream infection (BSI). An algorithm using electronic medical record (EMR) data to quickly predict resistance could potentially help guide therapy until more definitive resistance testing results are available. Methods All cases of K. pneumoniae BSI at Mount Sinai Hospital from September 2012 through September 2016 were identified. Cases of persistent BSI or recurrent BSI within 2 weeks were included only once. Patients with recurrent BSI after more than 2 weeks of negative blood cultures were considered distinct cases and included more than once. Carbapenem resistance was defined as an imipenem minimum inhibitory concentration of ≥2 μg/ml. Extensive EMR data for each patient were compiled into a relational database using SQLite. Possible risk factors for carbapenem resistance were queried from the database and analyzed via univariate methods. Significant factors were then entered into a multiple logistic regression model in a forward stepwise approach using SPSS. Results A total of 613 cases of K. pneumoniae BSI were identified in 540 unique patients. The overall incidence of imipenem resistance was 10% (61 cases). Significant markers of resistance included in the final model were (1) prior colonization with imipenem-resistant Klebsiella pneumoniae; (2) hospital unit (defined as high-risk unit, low-risk unit, and emergency department); (3) total inpatient days in the previous 5 years; (4) total days of oral or parenteral antibiotics in the past 2 years; and (5) age &gt;60 years old (Figure 1). The model generated a receiver operating characteristic curve with an area under the curve of 0.75 (Figure 2). At a cut point of 0.083, the model correctly predicted 72% of imipenem-resistant cases while incorrectly labeling 32% of susceptible cases as resistant (Sn = 72%, Sp = 63%, Figure 3). Conclusion A multiple logistic regression model using EMR data can generate immediate, clinically useful predictions of carbapenem resistance in patients with K. pneumoniae BSI. Larger data sets are needed to improve and validate these findings. Disclosures All authors: No reported disclosures.