Scholarlink

<p>Diagnosing leptospirosis and assessing test accuracy are challenging because no single method can reliably identify true infection across all stages, thereby preventing the establishment of a definitive gold standard. This study analyzes the performance of ELISA-IgM, DPP<sup>®</sup> Leptospirosis rapid test, and MAT using Latent Class Analysis (LCA) to infer infection status in the absence of an absolute reference standard. Samples from 430 individuals were included, whose diagnoses had been performed in reference laboratories in four states of Brazil. In the first stage of the analysis, 215 samples comprised the dataset used to estimate the two-latent-class model (“infected” and “not infected”). The latent status obtained was then used to estimate the accuracy of the tests in the second group of 215 individuals. ELISA-IgM, although highly sensitive (100%), showed low specificity (48.62%). MAT, in contrast, had high specificity (97.25%) but reduced sensitivity (69.81%). The DPP<sup>®</sup> Leptospirosis rapid test demonstrated a more balanced performance between sensitivity (82.08%) and specificity (90.83%). The results reinforce the need for combined diagnostic strategies to optimize leptospirosis detection, highlight the potential contribution of rapid testing in settings with limited laboratory infrastructure, and underscore the importance of improving the accuracy of currently available assays.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Diagnostic validity of ELISA-IgM, DPP® rapid test, and MAT for leptospirosis diagnosis in Brazilian reference laboratories: a latent class analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月23日

查看全文
  • Renata Aparecida Nascimento Ribeiro,
  • Tiago Silveira Gontijo,
  • Fabiano Borges Figueiredo,
  • Heber Paulino Pena,
  • Leila Denise Alves Ferreira Amorim,
  • Michelle Pereira Vale dos Passos,
  • Alexsandro Sobreira Galdino,
  • Ana Alice Maia Gonçalves,
  • Carolina Alves Petit Couto,
  • Isadora Braga Gandra,
  • Kátia Eliane Santos Avelar,
  • Ilana Teruszkin Balassiano,
  • Gilcelia Correia Santos Bernardes,
  • Lidsy Ximenes Fonseca,
  • Roberta Morozetti Blanco,
  • Max Assunção Correia,
  • Cintya de Oliveira Souza,
  • Elaine Santos Lima,
  • Vinícius Silva Belo

摘要

Diagnosing leptospirosis and assessing test accuracy are challenging because no single method can reliably identify true infection across all stages, thereby preventing the establishment of a definitive gold standard. This study analyzes the performance of ELISA-IgM, DPP® Leptospirosis rapid test, and MAT using Latent Class Analysis (LCA) to infer infection status in the absence of an absolute reference standard. Samples from 430 individuals were included, whose diagnoses had been performed in reference laboratories in four states of Brazil. In the first stage of the analysis, 215 samples comprised the dataset used to estimate the two-latent-class model (“infected” and “not infected”). The latent status obtained was then used to estimate the accuracy of the tests in the second group of 215 individuals. ELISA-IgM, although highly sensitive (100%), showed low specificity (48.62%). MAT, in contrast, had high specificity (97.25%) but reduced sensitivity (69.81%). The DPP® Leptospirosis rapid test demonstrated a more balanced performance between sensitivity (82.08%) and specificity (90.83%). The results reinforce the need for combined diagnostic strategies to optimize leptospirosis detection, highlight the potential contribution of rapid testing in settings with limited laboratory infrastructure, and underscore the importance of improving the accuracy of currently available assays.