Background <p>Generalized seizures in heatstroke have traditionally been interpreted as a clinical sign of severe central nervous system involvement. However, it remains uncertain whether seizures documented early in the disease course uniformly reflect irreversible injury or may be observed in patients with preserved short-interval physiologic responsiveness. We aimed to characterize early physiologic trajectories associated with early seizures and to explore their association with in-hospital mortality.</p> Methods <p>We conducted a retrospective nationwide cohort study of adults hospitalized with heat-related illness in the Japanese Association for Acute Medicine Heatstroke and Hypothermia Surveillance (JAAM-HsS) registry from 2017 to 2024. Early seizures were defined as generalized seizures observed or documented during prehospital transport or the initial emergency department assessment. The primary objective was to compare Day 1–Day 2 short-interval trajectories of key metabolic, electrolyte, and fluid-balance markers by early seizure status using linear mixed-effects models, focusing on seizure-by-time interactions. In-hospital mortality was evaluated as a secondary observational outcome using prespecified multivariable logistic regression (with vs. without lactate) and 1:1 propensity score matching, with attention to negative confounding and survivorship-related bias.</p> Results <p>Of 3,859 patients, 408 (10.6%) had early seizures. Patients with early seizures exhibited greater initial derangement (higher body temperature and lactate and lower base excess) yet showed more rapid short-interval normalization of base excess, lactate, glucose, and serum sodium by Day 2 (all interaction <i>p</i> &lt; 0.05). In-hospital mortality was 180/3,451 (5.2%) without early seizures and 27/408 (6.6%) with early seizures. In multivariable models adjusting for measured covariates, early seizures were inversely associated with mortality (lactate-included: aOR 0.48, 95% CI 0.27–0.86; lactate-excluded: aOR 0.57, 95% CI 0.33–0.99). This inverse association was directionally consistent after propensity score matching.</p> Conclusions <p>Early seizures in heat-related illness were associated with a distinct short-interval trajectory phenotype that may help calibrate early prognostication. An inverse association with in-hospital mortality was observed in adjusted and covariate-balanced analyses; however, this should be interpreted as a secondary marker–outcome association and may reflect residual or negative confounding, including differential seizure ascertainment among the most moribund patients. These findings are hypothesis-generating and warrant prospective validation with higher-granularity neurologic and treatment data.</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

Early seizures and heterogeneity of physiologic recovery in heat-related illness: a nationwide registry study

  • Koichi Inukai,
  • Koki Terakawa,
  • Takuya Matsui,
  • Takayuki Hattori,
  • Miyuu Uchida,
  • Shiho Tane,
  • Yu Hashimoto,
  • Masanori Kawamoto,
  • Fumitaka Kato,
  • Koji Amano,
  • Hiroyuki Kayata,
  • Hideaki Yakushiji,
  • Nobutaka Mukai,
  • Naoki Shinyama,
  • Akihiro Usui,
  • Masanori Morita

摘要

Background

Generalized seizures in heatstroke have traditionally been interpreted as a clinical sign of severe central nervous system involvement. However, it remains uncertain whether seizures documented early in the disease course uniformly reflect irreversible injury or may be observed in patients with preserved short-interval physiologic responsiveness. We aimed to characterize early physiologic trajectories associated with early seizures and to explore their association with in-hospital mortality.

Methods

We conducted a retrospective nationwide cohort study of adults hospitalized with heat-related illness in the Japanese Association for Acute Medicine Heatstroke and Hypothermia Surveillance (JAAM-HsS) registry from 2017 to 2024. Early seizures were defined as generalized seizures observed or documented during prehospital transport or the initial emergency department assessment. The primary objective was to compare Day 1–Day 2 short-interval trajectories of key metabolic, electrolyte, and fluid-balance markers by early seizure status using linear mixed-effects models, focusing on seizure-by-time interactions. In-hospital mortality was evaluated as a secondary observational outcome using prespecified multivariable logistic regression (with vs. without lactate) and 1:1 propensity score matching, with attention to negative confounding and survivorship-related bias.

Results

Of 3,859 patients, 408 (10.6%) had early seizures. Patients with early seizures exhibited greater initial derangement (higher body temperature and lactate and lower base excess) yet showed more rapid short-interval normalization of base excess, lactate, glucose, and serum sodium by Day 2 (all interaction p < 0.05). In-hospital mortality was 180/3,451 (5.2%) without early seizures and 27/408 (6.6%) with early seizures. In multivariable models adjusting for measured covariates, early seizures were inversely associated with mortality (lactate-included: aOR 0.48, 95% CI 0.27–0.86; lactate-excluded: aOR 0.57, 95% CI 0.33–0.99). This inverse association was directionally consistent after propensity score matching.

Conclusions

Early seizures in heat-related illness were associated with a distinct short-interval trajectory phenotype that may help calibrate early prognostication. An inverse association with in-hospital mortality was observed in adjusted and covariate-balanced analyses; however, this should be interpreted as a secondary marker–outcome association and may reflect residual or negative confounding, including differential seizure ascertainment among the most moribund patients. These findings are hypothesis-generating and warrant prospective validation with higher-granularity neurologic and treatment data.