SPC365

Prognostic role of neutrophil-lymphocyte ratio in infective endocarditis: a simple predictor for a complex disease?

Session: CO 06 - Valvulopatias
Speaker: João Gameiro

Congress: CPC 2021
Topic: F. Valvular, Myocardial, Pericardial, Pulmonary, Congenital Heart Disease
Theme: 16. Infective Endocarditis
Subtheme: 16.2 Infective Endocarditis – Epidemiology, Prognosis, Outcome
Session Type: Comunicações Orais
FP Number: ---
Authors: João Gameiro; André Freitas; Diana Campos; Carolina Saleiro; José Sousa; Ana Rita Gomes; Luís Puga; Eric Monteiro; Gonçalo Costa; Joana Silva; Lino Gonçalves

<p><span style="font-size:16px"><span style="font-family:Arial,Helvetica,sans-serif"><strong>Background</strong></span></span></p>

<p style="text-align:justify"><span style="font-size:14px"><span style="font-family:Arial,Helvetica,sans-serif">Infective endocarditis (IE) is a infectious disease with high morbidity and mortality. Because of its complex and heterogeneous nature, identifying high risk patients is both challenging and crucial. The neutrophil-to-lymphocyte ratio (NLR), as an inexpensive and easily accessible inflammatory marker, is gaining interest as an independent predictor of worse prognosis in some infectious and cardiovascular diseases. &nbsp;Whether NLR can have a prognostic role in IE is still under investigation.</span></span></p>

<p style="text-align:justify"><span style="font-size:16px"><span style="font-family:Arial,Helvetica,sans-serif"><strong>Purpose</strong></span></span></p>

<p style="text-align:justify"><span style="font-size:14px"><span style="font-family:Arial,Helvetica,sans-serif">The purpose of this study is to assess and compare, in patients (P) with IE, the prognostic value of 3 variables: NLR at hospital admission, total number of leukocytes at hospital admission and the highest total number of leukocytes during hospital stay.</span></span></p>

<p style="text-align:justify"><span style="font-size:16px"><span style="font-family:Arial,Helvetica,sans-serif"><strong>Methods</strong></span></span></p>

<p style="text-align:justify"><span style="font-size:14px"><span style="font-family:Arial,Helvetica,sans-serif">A retrospective cohort study from consecutive P&nbsp;diagnosed with definite IE (Duke criteria), admitted to our cardiology ward between January 2010 to December 2020. &nbsp;Baseline clinical data and in-hospital mortality were determined.</span></span></p>

<p style="text-align:justify"><span style="font-size:14px"><span style="font-family:Arial,Helvetica,sans-serif">Receiver operating characteristic (ROC) curves and area under curve (AUC) were calculated for the 3 variables and used for comparison. The cut-off value for the NLR was derived from the Youden index.&nbsp; Predictors of in-hospital mortality and time to the first event were analysed using logistic regression and survival analysis with multivariate Cox regression model.</span></span></p>

<p style="text-align:justify"><span style="font-size:16px"><span style="font-family:Arial,Helvetica,sans-serif"><strong>Results</strong></span></span></p>

<p style="text-align:justify"><span style="font-size:14px"><span style="font-family:Arial,Helvetica,sans-serif">A total of 262 P were included (70.6% male sex, mean age of 63.8 &plusmn; 15). In this cohort, the mean length of stay was 38 &plusmn; 27 days. A prosthetic valve was present in 30 % of P and an implanted device in 26% of P. The aortic valve was the most affected valve (43.5%). In 50.8 % of P,&nbsp;blood cultures were positive. The most common organism was <em>Staphyloccocus aureus</em> (19.1%).&nbsp;P were referred to cardiac surgery in 29% of cases. The mean level of NLR in this cohort was 10.67&plusmn; 8. In-hospital mortality in our study was 30.5%. &nbsp;</span></span></p>

<p style="text-align:justify"><span style="font-size:14px"><span style="font-family:Arial,Helvetica,sans-serif">The NLR at admission yielded an acceptable prognostic performance in predicting in-hospital death using ROC analysis (AUC: 0.705, 95% CI: 0.621-0.789, p &lt; 0.001) and performed better than the other variables in predicting death (total number of leukocytes at hospital admission: AUC: 0.665, p = 0.001; highest total number of leukocytes during hospital stay: AUC: 0.684, p &lt;0.001). A NLR of 5 was suggested as a predictive cut-off by the Youden index calculated with this analysis.</span></span></p>

<p style="text-align:justify"><span style="font-size:14px"><span style="font-family:Arial,Helvetica,sans-serif">After dividing our cohort in two groups (NLR &le; 5 and NLR &gt; 5), we used a multivariate Cox regression analysis adjusted to confounding factors (age, gender, multiple cardiovascular risk factors and other typical IE prognostic factors) that demonstrated a significant statistical impact of NLR &gt; 5 on hospital mortality (HR adjusted: 5.257; p = 0.001).</span></span></p>

<p style="text-align:justify"><span style="font-size:16px"><span style="font-family:Arial,Helvetica,sans-serif"><strong>Conclusion</strong></span></span></p>

<p><span style="font-size:14px"><span style="font-family:Arial,Helvetica,sans-serif">NLR at admission is an easy to calculate marker&nbsp;with good capacity to predict in-hospital mortality. A NLR level &gt; 5 was significantly associated with higher in-hospital mortality.</span></span></p>