Association of hyponatremia and risk of in-hospital mortality in patients with acute stroke: a systematic review and meta-analysis Authors: Prachi Sharma, Anuradha Pawar, Manabesh Nath, Deepti Vibha, Pradeep Kumar Published: 2026 ABSTRACT ObjectiveHyponatremia is among the most common electrolyte disturbances in acute stroke and has been increasingly recognized as a potential predictor of early complications. We aimed to evaluate whether hyponatremia on admission is associated with increased in-hospital mortality in patients with acute ischemic stroke (IS) and hemorrhagic stroke (HS).MethodsWe conducted a comprehensive systematic search of PubMed, Scopus, and Embase to identify relevant studies published between 1st January 2000 and 30th October 2025. Eligible studies included patients with acute stroke, reported admission serum sodium levels, and compared outcomes between hyponatremic and normonatremic groups. Pooled risk ratios (RRs) with 95% confidence intervals (CIs) were estimated using a random-effects model. Subgroup analyses were performed based on stroke type.ResultsSeventeen studies involving patients assessed within 24–72 h of stroke onset were included. Hyponatremia on admission was significantly associated with higher in-hospital mortality (RR 1.41, 95% CI 1.14–1.74). The association remained significant in ischemic stroke (RR 1.29, 95% CI 1.16–1.42) and in studies including mixed IS and HS populations (RR 1.59, 95% CI 1.19–2.12). No significant association was observed in hemorrhagic stroke alone (RR 1.47, 95% CI 0.70–3.10). Between-study heterogeneity was moderate (I2 = 58.9%, p < 0.001). Funnel plot assessment suggested no significant publication bias.ConclusionsHyponatremia on admission is associated with increased in-hospital mortality in acute stroke, particularly in ischemic stroke. Early identification and management of hyponatremia may have prognostic and therapeutic relevance in acute stroke care.Systematic Review RegistrationPROSPERO: CRD420251110001. Introduction Acute stroke disrupts cerebral homeostasis through ischemic or hemorrhagic injury, leading to rapid cellular and microvascular dysfunction (Salaudeen et al., 2024). Early alterations in sodium-dependent mechanisms contribute to cerebral edema, impaired perfusion, and neurological deterioration, thereby linking systemic sodium balance with brain water regulation and neuronal survival (Gankam Kengne, 2023). Serum sodium, a key determinant of extracellular osmolality, plays a central role in maintaining cerebral cell volume (Gankam Kengne and Decaux, 2017). Hyponatremia in acute stroke may arise from multiple mechanisms, including syndrome of inappropriate antidiuresis, cerebral salt wasting, and treatment-related factors, and may both reflect neuroendocrine stress and exacerbate ongoing cerebral injury, suggesting a bidirectional relationship between dysnatremia and brain injury (Cui et al., 2019; Liamis et al., 2019). Importantly, the timing of serum sodium assessment is a critical methodological consideration. Sodium levels measured beyond the initial 24–72 h may be influenced by iatrogenic factors such as fluid therapy or underlying comorbidities, rather than the direct effects of the acute cerebrovascular event (Yuen et al., 2022). Variability in the timing of sodium measurement across studies may therefore contribute to heterogeneity and inconsistent findings regarding its prognostic significance. Despite strong biological plausibility, the association between hyponatremia on admission and clinical outcomes in acute stroke remains uncertain. Several observational studies have reported links with adverse outcomes, including early neurological deterioration and mortality; however, results have been inconsistent, likely due to differences in study populations, definitions, and methodological approaches (Huang et al., 2012; Ha and Ryu, 2021; Qian et al., 2025). Therefore, we conducted a systematic review and meta-analysis to evaluate the association between hyponatremia on admission and in-hospital mortality in patients with acute stroke, with the aim of clarifying its prognostic relevance and informing early risk stratification. Methodology Protocol registration This systematic review and meta-analysis was designed and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations (Moher et al., 2015). The study protocol was registered prospectively with the International Prospective Register of Systematic Reviews (PROSPERO; Registration number: CRD420251110001) before commencement of the review, to enhance transparency and ensure methodological rigor. Eligibility criteria Study framework: study selection was based on the Population, Intervention/Exposure, Comparator, Outcomes, Study design (PICOS) framework. Study framework: study selection was based on the Population, Intervention/Exposure, Comparator, Outcomes, Study design (PICOS) framework. Population: adult patients aged ≥18 years presenting with: Population: adult patients aged ≥18 years presenting with: Acute ischemic stroke Acute ischemic stroke Hemorrhagic stroke Hemorrhagic stroke Mixed stroke cohorts Mixed stroke cohorts Exposure: hyponatremia at hospital admission, defined according to the prespecified serum sodium thresholds reported in individual studies. Exposure: hyponatremia at hospital admission, defined according to the prespecified serum sodium thresholds reported in individual studies. Comparator: patients with normonatremia at hospital admission. Comparator: patients with normonatremia at hospital admission. Outcome: in-hospital mortality. Outcome: in-hospital mortality. Study design: observational studies, including: Study design: observational studies, including: Cross-sectional studies (CSS) Cross-sectional studies (CSS) Prospective cohort studies (PCS) Prospective cohort studies (PCS) Retrospective cohort studies (RCS) Retrospective cohort studies (RCS) Case–control studies Case–control studies Exclusion criteria Studies without extractable data on admission serum sodium levels. Studies without extractable data on admission serum sodium levels. Studies that did not report in-hospital mortality or provided insufficient data to calculate effect estimates. Studies that did not report in-hospital mortality or provided insufficient data to calculate effect estimates. For overlapping datasets, only the most comprehensive or methodologically robust study was included. For overlapping datasets, only the most comprehensive or methodologically robust study was included. Search strategy A comprehensive literature search was conducted in PubMed, Embase, and Scopus to identify relevant studies published between 1st January 2000 and 30th October 2025. The search strategy combined free-text keywords and controlled vocabulary terms related to hyponatremia and acute stroke, including “hyponatremia,” “sodium,” “dysnatremia,” “ischemic stroke,” “Intracerebral hemorrhage,” and “acute stroke.” Boolean operators and database-specific search filters were applied as appropriate. Additionally, the reference lists of included studies and relevant review articles were manually screened to identify further eligible publications. The detailed search strategy is provided in Supplementary Table S1. No language restrictions were applied. Data extraction Two reviewers (PS and AP) independently screened titles, abstracts, and full-text articles to determine study eligibility. Data were extracted using a predefined, standardized extraction form and included information on study characteristics, patient demographics, serum sodium definitions, stroke subtype, timing of sodium assessment, and in-hospital mortality outcomes. Discrepancies between reviewers were resolved through discussion and consensus, with adjudication by the corresponding author (PK) when required. Statistical analysis Risk ratios (RR) with corresponding 95% confidence intervals (CIs) were calculated to evaluate the association between hyponatremia at hospital admission and in-hospital mortality. Adjusted effect estimates were preferentially used when available. Pooled estimates were derived using a random-effects model using the DerSimonian–Laird method to account for anticipated clinical and methodological heterogeneity. Statistical heterogeneity was assessed using the Cochran Q test and quantified using the I2 statistic (Higgins et al., 2003). Pre-specified subgroup analyses were conducted based on stroke subtype (ischemic, hemorrhagic, or mixed), timing of serum sodium assessment (at admission vs. within 24–48 h), and ethnicity of the study population. Publication bias was evaluated through visual inspection of funnel plots and quantitatively using Egger's regression test. All statistical analyses were performed using STATA version 16.1 (StataCorp, College Station, TX, USA), with a two-sided p value <0.05 considered statistically significant. Results Study selection The systematic literature search yielded 2,309 records, of which 1,880 unique citations remained after duplicate removal. After screening titles and abstracts, 104 articles were selected for full-text assessment. Ultimately, 17 studies fulfilled all predefined eligibility criteria and were included in the final analysis. The primary reasons for exclusion at the full-text stage included the absence of extractable data on in-hospital mortality, lack of an appropriate normonatremic comparator group, and outcomes that were not relevant to the objectives of the review. The study selection process is depicted in the PRISMA flow diagram (Figure 1). PRISMA 2020 flow diagram for the selection of studies. Study characteristics The 17 included studies were published between 2014 and 2025 and collectively evaluated patients with acute stroke across diverse geographic regions. Most studies were conducted in Asia (India, Pakistan, Bangladesh, Sri Lanka, Nepal and Iran), with additional contributions from Europe, North America, and Australia. Study designs included retrospective cohort (RCS), prospective cohort (PCS), cross-sectional (CSS), reflecting a mix of observational methodologies (Table 1). All studies focused on the acute stage of stroke, with diagnosis generally confirmed by CT and/or MRI, and in some cases supported by clinical criteria or standardized classifications such as AHA/ASA or TOAST. Admission timing varied, although many studies measured serum sodium within the first 24–48 h of hospitalization. Table 1Characteristics of included studies for investigating for the association of hyponatremia and risk of in-hospital mortality in patients with acute stroke.S. NoAuthor yearCountryEthnicityStudy designStudy periodStage ofstrokeDiagnosticcriteriaTime ofadmissionTotal samplesizeM/FAge (Mean ± SD)Cut-off value of Na (mmol/L)Outcome investigated1 Kuramatsu et al. (2014) GermanyCaucasianRCS2006–2010AcuteCT/MRIWithin 24 h422NR69.7 ± 12.0<135In hospital, 90 day, and 1 year mortality; FO2 Rodrigues et al. (2014) USACaucasianRCS2004–2014AcuteClinical + CT/MRIWithin 24–72 h3,5411,725/1,81671<135In hospital, mortality at 90 d, mortality at 1 year3 Gray et al. (2014) USACaucasianRCSJan 2006–Jul 2009AcuteClinical + CT/MRIWithin 48 h9961/3858.8<135Hospital LOS, ICU LOS, In hospital mortality4 Adekunle-Olarinde et al. (2017) UKCaucasianRCSJan 2003–May 2015AcuteClinical + SOAR ScoreWithin 24 h8,4934,025/4,46877.7 ± 11.6<135Inpatient and 7 d mortality5 Kalita et al. (2017) IndiaAsianPCSJan 2014–Jan 2016AcuteCT/MRIWithin 48 h10069/3162 (18–90)<135Disabilty, In-hospital mortality6 Shah et al. (2019) KarachiAsianPCSJul 2018–Dec 2018AcuteClinical + CT/MRIWithin 24–72 h234182/5259.14 ± 9.05<135Recovery, disability, In hospital mortality7 Hakim et al. (2019) BangladeshAsianCSSFeb 2017–Nov 2017AcuteClinical + CT/MRIWithin 24–72 h205124/8160 ± 13.1<135Hospital LOS, In hospital mortality8 Quinn et al. (2020) AustraliaCaucasianPCSJan 2015–May 2019AcuteClinical + CTWithin 24 h20056/14457 ± 12<135ICU mortality, Hospital mortality, FU9 Saadat et al. (2022) IranAsianCSSJan 2021–Jun 2021AcuteAHA/ASA +TOAST classification, CT/MRIWithin 24 h12555/7064.8 ± 8.3<135Mortality, disability, severity10 Padhy et al. (2023) IndiaAsianCSSMar 2021–May 2021AcuteClinical + CT/MRIWithin 24 h10046/5464.2<135In hospital mortality, 60 d mortality11 Khan et al. (2023) PakistanAsianCSSSep 2022–Feb 2023AcuteAHA/ASAWithin 24 h289162/12761 ± 7.93<135Severity, In-hospital mortality12 Ranasinghe et al. (2023) Sri LankaAsianPCSJune 2019–Feb 2021AcuteClinical + CTWithin 24 h246120/12668.14 ± 12.71<131In hospital, 70 d mortality13 Ezhumalai et al. (2023) IndiaAsianPCSOct 2021–Dec 2021AcuteClinical + CT/MRIWithin 24–72 h7756/2156<135Mortality14 Pandya et al. (2024) IndiaAsianCSSJan 2023–June 2023AcuteCT/MRIWithin 48 h210155/5558.5 ± 11.0<135Mortality15 Malhotra et al. (2024) IndiaAsianPCSMar 2021–Jul 2022AcuteClinicalWithin 24–72 h11070/40>60<135Severity; Mortality16 Sah et al. (2024) NepalAsianCSSJan 2023–Jan 2024AcuteCT/MRIWithin 24 h10063/3764<135Severity; Mortality17 Naidu and Kumari (2025) IndiaAsianRCSDec 2023–Dec 2024AcuteCT/MRIWithin 24–72 h6040/20NR<135MortalitySOAR Score, Stroke subtype Oxfordshire Community Stroke Project Classification Age Rankin(pre-stroke modified) scale; AHA/ASA, American Heart Association/American Stroke Association; CSS, cross-sectional study; PCS, prospective cohort study; RCS, retrospective cohort study; CT, computed tomography; MRI, magnetic resonance imaging; Na, sodium; ICU, intensive care unit; LOS, length of stay; FO, functional outcome; FU, follow-up; NR, not reported; M/F, Male/Female; SD, standard deviation; d, days; h, hour. Characteristics of included studies for investigating for the association of hyponatremia and risk of in-hospital mortality in patients with acute stroke. SOAR Score, Stroke subtype Oxfordshire Community Stroke Project Classification Age Rankin(pre-stroke modified) scale; AHA/ASA, American Heart Association/American Stroke Association; CSS, cross-sectional study; PCS, prospective cohort study; RCS, retrospective cohort study; CT, computed tomography; MRI, magnetic resonance imaging; Na, sodium; ICU, intensive care unit; LOS, length of stay; FO, functional outcome; FU, follow-up; NR, not reported; M/F, Male/Female; SD, standard deviation; d, days; h, hour. The total sample sizes ranged widely, from 60 to 8,493 participants, with most cohorts including both men and women. Hyponatremia was most commonly defined using a serum sodium cutoff of <135 mmol/L. Outcomes assessed across studies primarily included in-hospital mortality, with several studies also reporting short-term mortality (7–90 days), functional outcomes, disability, stroke severity, or length of hospital stay. Despite methodological and clinical heterogeneity, the included studies were broadly comparable in their focus on the prognostic role of hyponatremia on admission in acute stroke. Main results Overall, hyponatremia on admission was significantly associated with an increased risk of in-hospital mortality in patients with acute stroke (RR = 1.41; 95% CI, 1.14–1.74), with moderate heterogeneity across studies (I2 = 58.9%; Figure 2). Forest plot demonstrating the association between hyponatremia and in-hospital mortality stratified by stroke subtype. In subgroup analyses by stroke subtype, hyponatremia was associated with higher mortality in ischemic stroke (RR = 1.29; 95% CI, 1.16–1.42; I2 = 0.0%) and in mixed ischemic and hemorrhagic stroke cohorts (RR = 1.59; 95% CI, 1.19–2.12; I2 = 0.0%). In hemorrhagic stroke, the association was directionally similar but highly heterogeneous and did not reach statistical significance (RR = 1.47; 95% CI, 0.70–3.0; I2 = 86.7%). Ethnicity-based subgroup analysis demonstrated a significantly higher mortality risk associated with hyponatremia among Asian populations (RR = 1.78; 95% CI, 1.34–2.38; I2 = 26.7%). In contrast, no significant association was observed in predominantly Caucasian populations (RR = 1.11; 95% CI, 0.84–1.48), with considerable heterogeneity (I2 = 75.3%). Subgroup analysis based on the timing of serum sodium assessment showed an increased mortality risk when sodium was measured within 48 h (RR = 1.33; 95% CI, 0.74–2.41; I2 = 45.3%), although these associations were not statistically significant. Studies reporting sodium assessment within 24 h demonstrated a modest, non-significant increase in mortality (RR = 1.34; 95% CI, 0.96–1.86; I2 = 68.3%). A significant association was observed in studies reporting sodium assessment within 24–72 h (RR = 1.68; 95% CI, 1.00–2.81; I2 = 59.9%; Supplementary Table S2). When studies were stratified by the timing of sodium measurement (within 24, 48, and 24–72 h), the direction and magnitude of the association between hyponatremia and in-hospital mortality remained consistent across subgroups. There was no significant interaction between timing subgroups (p = 0.750), indicating that the prognostic impact of hyponatremia was not materially influenced by when serum sodium was measured during hospitalization. Publication bias Visual inspection of the funnel plot showed no substantial asymmetry, indicating a low risk of publication bias (p = 0.40; Figure 3). The majority of studies were symmetrically distributed around the pooled effect estimate with a consistent direction of association, supporting the robustness of the findings. Funnel plot for the association between hyponatremia and in-hospital mortality. Discussion This systematic review and meta-analysis demonstrates that hyponatremia on admission is associated with a significantly increased risk of in-hospital mortality in patients with acute stroke. Across 17 studies with diverse geographic representation and study designs, hyponatremia conferred an approximately 40% higher mortality risk, with no evidence of publication bias, supporting the robustness of the association. The observed association is biologically plausible. In acute stroke, hyponatremia often reflects neuroendocrine dysregulation, including syndrome of inappropriate antidiuretic hormone secretion and cerebral salt wasting, both of which are linked to greater cerebral injury and raised intracranial pressure (Pandya et al., 2024). Reduced serum sodium promotes osmotic cerebral edema, impairs neuronal function, and disrupts cerebral perfusion, thereby increasing the risk of early neurological deterioration and death (John et al., 2025). Even mild hyponatremia has been shown to adversely affect neuronal excitability and cerebral metabolism, compounding ischemic injury (Gankam Kengne and Decaux, 2017). Existing literature consistently supports the prognostic relevance of hyponatremia in acute neurological illnesses, including stroke. Previous observational studies have shown that hyponatremia is independently associated with increased mortality, longer hospital stay, and poorer functional outcomes in stroke patients (Eldhose et al., 2025; Soiza et al., 2015). Mechanistically, hyponatremia reflects central nervous system–mediated dysregulation of water and sodium homeostasis, commonly driven by syndrome of inappropriate antidiuretic hormone secretion or cerebral salt wasting following acute brain injury (Cui et al., 2019). These disturbances exacerbate cerebral edema, compromise cerebral perfusion, and increase intracranial pressure, thereby amplifying secondary brain injury (Gankam Kengne and Decaux, 2017). Moreover, hyponatremia has been linked to heightened inflammatory responses, autonomic dysfunction, and cardiovascular instability, all of which contribute to early clinical deterioration (Park and Shin, 2013). The consistent association observed across diverse populations and study designs in the present review reinforces prior evidence and underscores hyponatremia as not merely a biochemical abnormality, but a pathophysiological meaningful marker of disease severity and early physiological decompensating in acute stroke. Subgroup analyses showed a consistent association in ischemic and mixed stroke populations, with low heterogeneity, whereas findings in hemorrhagic stroke were less consistent, likely reflecting differences in underlying pathophysiology, fluid management practices, and smaller sample sizes. The stronger association observed in Asian populations may relate to population-specific factors such as baseline sodium levels, dietary patterns, stroke characteristics, or healthcare delivery systems. Variability in diagnostic criteria across the included studies represents an important source of heterogeneity. Differences in the definition of stroke, methods of confirmation (clinical vs. imaging-based), and thresholds used for defining hyponatremia may have influenced patient classification and outcome assessment. Notably, the prognostic impact of hyponatremia was not materially influenced by the timing of sodium measurement or study quality, indicating that hyponatremia at any point during early hospitalization carries prognostic significance. Beyond direct neurological effects, hyponatremia may also serve as an integrative marker of systemic illness severity, capturing inflammatory, metabolic, and cardiovascular stress responses that contribute to adverse outcomes (Podestà et al., 2015). Collectively, these findings support serum sodium as a simple, readily available, and clinically meaningful prognostic marker in acute stroke care. Limitations Several limitations merit consideration: (1) most included studies were observational, limiting causal inference and allowing for residual confounding; (2) heterogeneity in the definition and timing of hyponatremia assessment and adjustment for stroke severity may have influenced effect estimates; (3) limited data on the etiology, severity, duration, and correction of hyponatremia precluded dose–response and treatment-effect analyses; (4) outcomes were largely restricted to in-hospital mortality, with insufficient data on long-term functional and survival outcomes; (5) an important limitation of the present analysis is the lack of detailed and standardized reporting on the management of hyponatremia across included studies. Most studies did not provide sufficient information regarding specific therapeutic interventions (e.g., use of hypertonic saline), timing or protocols for sodium correction, or the direct impact of such treatments on clinical outcomes, including in-hospital mortality. Consequently, a quantitative synthesis evaluating the effect of hyponatremia correction on outcomes was not feasible, and (6) the predominance of studies from Asian populations may limit the generalizability of the findings to other regions. Conclusion In patients with acute stroke, hyponatremia on admission is independently associated with a significantly increased risk of in-hospital mortality. This association is consistent across stroke subtypes and timing of serum sodium assessment, and is not driven by any single study. These findings highlight hyponatremia as an important early and readily available prognostic marker in acute stroke care, underscoring the need for prompt recognition and appropriate management to potentially improve short-term outcomes. Further prospective studies are warranted to elucidate causal mechanisms and to determine whether targeted correction of hyponatremia can favorably influence clinical outcomes in stroke patients.