EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Key predictors of mortality in profound hyponatremia beyond the correction rate

Koya Nagase iD, Takahiro Imaizumi, Atsushi Yamamori, Fumika N Nagase, Toshikazu Ozeki, Nobuhiro Nishibori iD, Takaya Ozeki iD, Hideaki Shimizu, Yoshiro Fujita iD, Kazuhiro Furuhashi, Tsuyoshi Watanabe iD, Shoichi Maruyama iD

Clinical Kidney Journal · 31.07.2026 · 19 (8)

DOI: 10.1093/ckj/sfag219 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Clinical Kidney Journal, online erschienen am 30.06.2026. Verfasst von Koya Nagase und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Nephrology, Nagoya University Graduate School of Medicine, Tsurumai-cho, Showa-ku, Nagoya, Aichi, Japan. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Education, Culture, Sports, Science and Technology, Japan Science and Technology Agency, JST BOOST.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Current guidelines for hyponatremia recommend slow sodium correction; however, recent large-scale observational studies have associated slow correction with increased mortality. The association between correction rate and mortality can be influenced by numerous factors, including comorbidities, overall condition, and treatment interventions, but this complex interplay remains unclear. We aimed to clarify this association by developing interpretable machine-learning models using detailed clinical features.

Methods

We included 546 patients with serum sodium ≤120 mEq/l, collected clinical features through chart review, and developed four machine-learning models to predict in-hospital mortality. The best-performing model, selected by the area under the receiver operating characteristic curve (ROC-AUC), was interpreted using SHapley Additive exPlanations (SHAP) to quantify each feature’s contribution to mortality prediction.

Results

In-hospital mortality was 18%. The random forest model demonstrated the best predictive performance (ROC-AUC = 0.907; 95% confidence interval, 0.832–0.965). SHAP analysis revealed that the most influential predictors were baseline characteristics reflecting underlying illness severity: serum albumin, C-reactive protein, Charlson comorbidity index, and metastatic malignancy; their predictive effects were consistent across correction rates. Among treatment-related features, intravenous fluid choice and sodium monitoring frequency had a greater predictive impact than correction rate. Although slower correction was associated with higher mortality, the correction rate ranked 14th among all 59 features in predictive importance.

Conclusions

In profound hyponatremia, multiple key predictors of mortality exist beyond the correction rate. These findings suggest that the observed association between slow correction and higher mortality may not be causal but rather an epiphenomenon driven by underlying illness severity and treatment intensity.

Schlagwörter

Stichwörter der Autor:innen

correction ratehyponatremiamachine learningmortalityrandom forest

Autor:innen und Institutionen

  • Department of Nephrology, Nagoya University Graduate School of Medicine, Tsurumai-cho, Showa-ku, Nagoya, Aichi, Japan.
  • Department of Clinical Research Education, Nagoya University Graduate School of Medicine, Tsurumai-cho, Showa-ku, Nagoya, Aichi, Japan.
  • Department of Nephrology, Chubu Rosai Hospital, Komei-cho, Minato-ku, Nagoya, Aichi, Japan.
  • Department of Rheumatology, Chubu Rosai Hospital, Komei-cho, Minato-ku, Nagoya, Aichi, Japan.
  • Department of Nephrology and Renal Replacement, Daido Hospital, Hakusui-cho, Minami-ku, Nagoya, Aichi, Japan.
  • Department of Internal Medicine, Kou Hospital, Hazenji, Kubo-cho, Toyokawa, Aichi, Japan.

ORCID-Profile: Koya Nagase, Nobuhiro Nishibori, Takaya Ozeki, Yoshiro Fujita, Tsuyoshi Watanabe, Shoichi Maruyama

Förderung

  • Ministry of Education, Culture, Sports, Science and Technology
  • Japan Science and Technology Agency
  • JST BOOST

Zitieren

APA

Nagase, K., Imaizumi, T., Yamamori, A., Nagase, F. N., Ozeki, T., Nishibori, N., Ozeki, T., Shimizu, H., Fujita, Y., Furuhashi, K., Watanabe, T., & Maruyama, S. (2026). Key predictors of mortality in profound hyponatremia beyond the correction rate. Clinical Kidney Journal, 19(8). https://doi.org/10.1093/ckj/sfag219

Vancouver

Nagase K, Imaizumi T, Yamamori A, Nagase FN, Ozeki T, Nishibori N, et al. Key predictors of mortality in profound hyponatremia beyond the correction rate. Clinical Kidney Journal. 2026;19(8). doi:10.1093/ckj/sfag219

BibTeX
@article{Nagase2026,
  author = {Nagase, Koya and Imaizumi, Takahiro and Yamamori, Atsushi and Nagase, Fumika N and Ozeki, Toshikazu and Nishibori, Nobuhiro and Ozeki, Takaya and Shimizu, Hideaki and Fujita, Yoshiro and Furuhashi, Kazuhiro and Watanabe, Tsuyoshi and Maruyama, Shoichi},
  title = {Key predictors of mortality in profound hyponatremia beyond the correction rate},
  journal = {Clinical Kidney Journal},
  year = {2026},
  volume = {19},
  number = {8},
  doi = {10.1093/ckj/sfag219}
}
Kostenlos starten →