Abstract
Aims
To examine the independent and combined effects of diabetes mellitus (DM) and hypertension (HTN) on acute kidney injury (AKI) and 90-day mortality in critically ill patients.
Patients and Methods
We retrospectively analyzed 3,282 adult intensive care unit (ICU) admissions to King Abdullah University Hospital, Jordan (2012–2022). Patients were stratified by DM and HTN status. Logistic regression identified predictors of AKI, and Cox proportional hazards models assessed 90-day mortality.
Results
AKI occurred in 44.7% of DM patients versus 40.8% without DM (p = 0.0423). Patients with both DM and HTN had the highest AKI incidence (48.6%) and the highest 90-day mortality (37.8%). In addition, the HTN × DM interaction was a significant predictor of AKI in multivariable analysis (OR = 1.18, 95% CI: 1.07–1.36, p = 0.0059) and increased the mortality hazard independently (HR = 1.43, 95% CI: 1.04–1.97, p = 0.0286).
Conclusions
DM is associated with increased AKI and mortality in critically ill patients, and concurrent HTN further amplifies these risks. These findings support integrated cardiometabolic risk assessment in ICU settings.
Keywords: Diabetes, hypertension, ICU, acute kidney injury, mortality
PLAIN LANGUAGE SUMMARY
Diabetes and high blood pressure are two of the most common chronic conditions seen in patients admitted to intensive care units (ICUs). While each condition on its own is known to affect health outcomes, less is known about what happens when a patient has both conditions at the same time.
In this study, we examined records from 3,282 patients admitted to the ICU at a major hospital in Jordan over a 10-year period (2012–2022). We looked at two key outcomes: whether patients developed acute kidney injury (AKI)—a sudden decline in kidney function—during their ICU stay and whether they survived to 90 days from admission.
We found that patients who had both diabetes and high blood pressure together faced a significantly higher risk of both kidney injury and death compared to patients who had either condition alone or neither. Specifically, the combination of the two conditions increased the risk of death beyond what would be expected from their individual effects. Other important predictors of worse outcomes included older age, abnormal blood electrolyte levels (potassium, sodium, and bicarbonate), and lower lymphocyte percentages, which reflect greater inflammation.
These findings suggest that ICU clinicians should pay particular attention to patients who present with both diabetes and high blood pressure, as this combination carries an added risk that goes beyond the sum of its individual parts. Early identification and targeted management of these patients may help improve outcomes in critical care settings.
ARTICLE HIGHLIGHTS
Diabetes and hypertension jointly increase AKI and mortality in ICU patients.
The coexistence of both conditions confers synergistic—not merely additive—risk for both AKI and mortality.
Hypertension, CKD, male sex, electrolyte disturbances, lymphopenia, and the HTN–DM interaction independently predicted AKI, while age, CKD, heart failure, electrolyte disturbances, lymphopenia, and the HTN–DM interaction independently predicted mortality.
Findings support integrated cardiometabolic risk management in critical care.
1. Introduction
Intensive care units (ICUs) care for the most critically ill patients and are often faced with high rates of complications and mortality [1]. Among the many adverse outcomes in ICU populations, acute kidney injury (AKI) is particularly common, with a strong association with short- and long-term mortality [2,3]. Beyond its immediate impact, AKI predisposes survivors to chronic kidney disease (CKD) [4] and cardiovascular events [5], making it one of the most consequential complications in critical care.
Among chronic comorbidities, diabetes mellitus (DM) stands out with a markedly higher prevalence in hospitalized adults compared with the general population. In general medical wards, DM accounts for approximately 11%–35% of admissions, depending on population and healthcare setting [6]. The burden is even more pronounced in critically ill populations; 12%–40% of ICU patients are reported to have preexisting DM [6], highlighting its clinical significance in critical care. This higher burden reflects not only the rising prevalence of DM in the community but also the increased susceptibility of these patients to severe complications requiring intensive care. This rising burden is of particular concern in the Middle East, a region with one of the highest global DM burdens, reflecting both lifestyle factors and epidemiological transitions [7].
DM contributes to poor outcomes through mechanisms including vascular complications, impaired immunity, and metabolic dysregulation, rendering patients especially vulnerable to acute organ dysfunction and death during critical illness [8]. Glycated hemoglobin (HbA1c) is a widely used marker of long-term glycemic control and has gained recognition as a valuable prognostic indicator in critical care research. Beyond its role in identifying patients with previously undiagnosed DM, elevated HbA1c levels on admission have been consistently linked to higher mortality and adverse outcomes in ICU populations [9,10].
The coexistence of DM and hypertension (HTN) is of particular concern, as these conditions frequently cluster and may act synergistically to worsen renal and cardiovascular vulnerability [11]. Evidence from large cohort studies demonstrates that their concurrence has a synergistic effect on poorer health outcomes, notably increased overall and cardiovascular-related mortality [12]. In the Middle East, the challenge is even more pronounced. Jordan, for instance, ranks among the countries with the highest global burden of DM, with an age-standardized prevalence of 23.7% in 2017 [13]. Hospital-based studies further demonstrate that HTN coexists in over 70% of patients with DM in Jordan [14,15], underscoring the interactive impact of these comorbidities on health outcomes. This coexistence is particularly critical in intensive care settings, where both DM and HTN substantially influence the risk of complications such as AKI and mortality. Understanding these interactions is essential for tailoring clinical management and improving patient outcomes in this regional context.
Building on this evidence, the present study aimed to examine the interplay between DM, HTN, and acute outcomes such as AKI and mortality and to evaluate HbA1c as a key variable in the assessment of DM and its impact on outcomes in critically ill patients. This study seeks to address an important gap in the regional critical care literature and contribute to strategies for improving patient management in high-burden settings.
2. Research design and methods
2.1. Study population
This study followed a retrospective observational design and took place at King Abdullah University Hospital (KAUH), the major tertiary care and teaching hospital in northern Jordan. From the hospital’s electronic inpatient records, 13,567 patients admitted to the intensive care unit (ICU) between July 2012 and July 2022 were initially identified. Patients younger than 18 years (n = 267) and those without at least one documented HbA1c measurement at ICU admission (n = 10,018) were excluded. The final analytic cohort therefore comprised 3,282 adult ICU patients.
2.2. Data collection and variables
Data were extracted from electronic health records and included demographic characteristics (age and gender), preexisting comorbidities, and primary causes of ICU admission. In addition to HbA1c, a set of laboratory parameters was retrieved to provide insights into patients’ physiological and metabolic status. These included:
Lymphocyte percentage—as an indicator of systemic inflammation.
Serum bicarbonate (HCO3-)—as a marker of acid–base balance.
Serum potassium and sodium concentrations—as indicators of electrolyte homeostasis.
Serum creatinine as indicator of AKI incidence.
AKI was defined in accordance with the Kidney Disease: Improving Global Outcomes (KDIGO) criteria as an increase in serum creatinine of ≥ 0.3 mg/dL within 48 hours, or an increase to ≥ 1.5 times the baseline value known or presumed to have occurred within the preceding 7 days.
2.3. Statistical analysis
Baseline differences were analyzed using the chi-square test for categorical variables, or Fisher’s exact test when any expected cell count was less than 5, and two-sample t-tests or one-way ANOVA for continuous variables.
For the AKI outcome, multivariable logistic regression models were applied to estimate odds ratios (ORs) with 95% confidence intervals (CIs). HTN and DM were modeled as separate binary variables, and their interaction term (HTN × DM) was included to formally assess the synergistic interaction effects. Baseline comorbidities and laboratory measures were included as covariates to account for potential confounding.
For the mortality outcome, Cox proportional hazards regression was employed, estimating hazard ratios (HRs) with 95% CIs. Binary covariates (HTN, DM, CKD, IHD, heart failure (HF), and gender) were entered as indicator variables (0 = absent, 1 = present), and hazard ratios (HRs) represent the effect of the presence versus absence of each condition. Time-to-event was defined as the time from ICU admission to death within 90 days. Patients who survived to 90 days were administratively censored at 90 days. Clinically relevant and statistically significant variables from univariable analyses were included in multivariable models. Statistical significance was defined as a two-tailed p ≤ 0.05, and results are reported with effect estimates, 95% confidence intervals, and p-values to reflect both statistical and clinical significance.
3. Results
3.1. Patient clinical and demographic characteristics
A total of 3282 ICU patients were included, comprising 1818 without DM and 1464 with DM. Patients with DM were significantly older compared to those without (66.6 ± 15.5 vs. 62.2 ± 17.7 years, p < 0.0001). Male sex was more common among non-diabetic patients (62.5 vs. 55.5%, p < 0.0001) (Table 1).
Table 1.
Baseline Characteristics of the study population stratified by DM status.
| Variable | No DM (n = 1818) | DM (n = 1464) | p-value |
|---|---|---|---|
| Age (years) | 62.2 ± 17.7 | 66.6 ± 15.5 | <0.0001*** |
| Gender (male) | 1137 (62.5%) | 812 (55.5%) | <0.0001*** |
| Comorbidities | |||
| HTN | 564 (31.0%) | 1133 (77.4%) | <0.0001*** |
| IHD | 235 (12.9%) | 407 (27.8%) | <0.0001*** |
| HF | 112 (6.2%) | 195 (13.3%) | <0.0001*** |
| CKD | 68 (3.7%) | 178 (12.2%) | <0.0001*** |
| Laboratory tests | |||
| HbA1c (%) | 6.36 ± 1.80 | 8.37 ± 2.39 | <0.0001*** |
| Sodium (mmol/L) | 140.40 ± 7.14 | 138.96 ± 7.04 | <0.0001*** |
| Potassium (mmol/L) | 4.20 ± 0.74 | 4.27 ± 0.74 | 0.0121* |
| HCO3 (mmol/L) | 23.04 ± 9.38 | 22.78 ± 6.46 | 0.4568 |
| Lymphocytes (%) | 16.61 ± 11.40 | 16.75 ± 10.66 | 0.7273 |
| Diagnoses | |||
| Stroke | 135 (7.4%) | 108 (7.4%) | 0.9578 |
| Septicemia | 118 (6.5%) | 87 (5.9%) | 0.5190 |
| Acute renal failure | 45 (2.5%) | 35 (2.4%) | 0.8759 |
| Chronic renal failure | 11 (0.6%) | 10 (0.7%) | 0.7806 |
| COPD | 14 (0.8%) | 10 (0.7%) | 0.7712 |
| HF (acute) | 28 (1.5%) | 55 (3.8%) | <0.0001*** |
| Ketoacidosis | 10 (0.6%) | 61 (4.2%) | <0.0001*** |
| Pneumonia | 17 (0.9%) | 22 (1.5%) | 0.1358 |
| Urinary tract infection | 6 (0.3%) | 17 (1.2%) | 0.0045** |
| COVID-19 | 66 (3.6%) | 62 (4.2%) | 0.3738 |
| Hypoglycemia | 18 (1.0%) | 39 (2.7%) | 0.0003*** |
| Hyperkalemia | 8 (0.4%) | 7 (0.5%) | 0.8722 |
| Hypokalemia | 7 (0.4%) | 2 (0.1%) | 0.3143 |
| Hyponatremia | 18 (1.0%) | 27 (1.8%) | 0.0365* |
| Hypernatremia | 0 (0.0%) | 1 (0.1%) | 0.4461 |
| GI hemorrhage | 47 (2.6%) | 28 (1.9%) | 0.1999 |
| Angioedema | 14 (0.8%) | 7 (0.5%) | 0.2971 |
Continuous variables are presented as mean ± standard deviation (SD), and categorical variables as number (percentage). Statistical significance is denoted as follows: * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001. CKD: Chronic kidney disease, COPD: Chronic obstructive pulmonary disease; GI: Gastrointestinal; HF: Heart failure; HTN: Hypertension; IHD: Ischemic heart disease.
Regarding comorbidities, diabetic patients had markedly higher prevalence of HTN (77.4 vs. 31.0%), ischemic heart disease (IHD) (27.8 vs. 12.9%), HF (13.3 vs. 6.2%), and CKD (12.2 vs. 3.7%) (all p < 0.0001) (Table 1).
Laboratory tests revealed that diabetic patients had significantly higher HbA1c levels (8.37 ± 2.39 vs. 6.36 ± 1.80%, p < 0.0001), lower sodium levels (138.96 ± 7.04 vs. 140.40 ± 7.14 mmol/L, p < 0.0001), and slightly higher potassium (4.27 ± 0.74 vs. 4.20 ± 0.74 mmol/L, p = 0.0121). HCO3 and lymphocyte percentages did not differ significantly.
In terms of admission diagnoses, diabetic patients had significantly (p < 0.05) higher rates of acute HF (3.8 vs. 1.5%), ketoacidosis (4.2 vs. 0.6%), urinary tract infection (1.2 vs. 0.3%), hypoglycemia (2.7 vs. 1.0%), and hyponatremia (1.8 vs. 1.0%) (all statistically significant). Other diagnoses, including stroke, sepsis, pneumonia, acute or chronic renal failure, COPD, COVID-19, hyperkalemia, hypokalemia, GI hemorrhage, and angioedema, were comparable between groups (Table 1).
3.2. Impact of diabetes on AKI incidence and survival
Following the baseline comparisons in Table 1, we next examined the incidence of AKI during ICU admission (Figure 1A) and 90-day mortality (Figure 1B) across DM strata. The difference in AKI incidence between groups was statistically significant (p = 0.0423), with AKI occurring in 44.7% of patients with DM compared to 40.8% of those without DM, indicating a modest but clinically meaningful excess risk in the diabetic group.
Figure 1.
Acute kidney injury (AKI) and survival by DM status and DM-HTN classification. (A) Proportion of ICU patients who developed AKI within their ICU stay stratified by DM status. (B) 90-day Kaplan–Meier survival curves by DM status. Median survival was not reached in either group. The 90-day mortality rate was 34.8% in DM patients versus 31.2% in non-DM patients (log-rank χ2 = 4.46, p = 0.0347). (C) Proportion of ICU patients who developed AKI across four classification groups: no HTN/DM, HTN only, DM only, and HTN+DM. AKI incidence was highest in patients with both HTN and DM (p < 0.0001). (D) Kaplan–Meier 90-day survival curves stratified by DM–HTN classification. Median survival was not reached in any group. The 90-day mortality rate was highest in the HTN+DM group (37.8%) and lowest in the DM-only group (24.5%) (log-rank χ2 = 26.2, p < 0.0001). Abbreviations: AKI = acute kidney injury; DM = diabetes mellitus; HTN = hypertension; ICU = intensive care unit.
To assess longer-term outcomes, Kaplan–Meier survival analysis was performed (Figure 1B). Median survival was not reached in either group (i.e., more than 50% of patients survived to 90 days). The 90-day mortality rate was 34.8% among DM patients compared with 31.2% among non-DM patients. The log-rank test confirmed a significant survival difference between groups (χ2 = 4.46, p = 0.0347).
3.3. Independent and combined effects of diabetes and hypertension on clinical characteristics
Because HTN was highly prevalent among patients with DM (77.4%, Table 1), additional analyses were performed to determine whether the observed associations reflected an independent effect of DM or a combined effect of DM and HTN (Table 2). A classification approach was adopted to disentangle the independent and combined effects of HTN and DM. Patients were stratified into four groups: no HTN/DM (n = 1,254), HTN only (n = 564), DM only (n = 331), and both HTN+ DM (n = 1,133).
Table 2.
Baseline Characteristics by classification (No DM/HTN, HTN only, DM only, HTN+DM).
| Variable | No HTN/DM (n = 1254) | HTN–only (n = 564) | DM-only (n = 331) | HTN+DM (n = 1133) | p-value |
|---|---|---|---|---|---|
| Age (years) | 59.4 ± 18.0 | 68.4 ± 15.3 | 56.4 ± 18.7 | 69.6 ± 13.0 | <0.0001*** |
| Gender (Male) | 789 (62.9%) | 348 (61.7%) | 203 (61.3%) | 609 (53.8%) | <0.0001*** |
| Comorbidities | |||||
| CKD | 19 (1.5%) | 49 (8.7%) | 14 (4.2%) | 164 (14.5%) | <0.0001*** |
| IHD | 105 (8.4%) | 130 (23.1%) | 47 (14.2%) | 360 (31.8%) | <0.0001*** |
| HF | 39 (3.1%) | 73 (12.9%) | 16 (4.8%) | 179 (15.8%) | <0.0001*** |
| Laboratory tests | |||||
| HbA1c (%) | 6.51 ± 1.99 | 6.04 ± 1.24 | 9.13 ± 2.77 | 8.15 ± 2.22 | <0.0001*** |
| Potassium (mmol/L) | 4.22 ± 0.75 | 4.18 ± 0.72 | 4.30 ± 0.72 | 4.26 ± 0.75 | 0.0512 |
| Sodium (mmol/L) | 140.4 ± 7.0 | 140.4 ± 7.4 | 139.0 ± 6.0 | 138.9 ± 7.3 | <0.0001*** |
| HCO3 (mmol/L) | 22.7 ± 10.4 | 23.7 ± 6.5 | 21.9 ± 6.3 | 23.1 ± 6.5 | 0.0391* |
| Lymphocytes (%) | 16.5 ± 11.4 | 16.9 ± 11.4 | 18.5 ± 11.5 | 16.2 ± 10.4 | 0.0158* |
| Diagnoses | |||||
| Stroke | 85 (6.8%) | 50 (8.9%) | 17 (5.1%) | 91 (8.0%) | 0.1323 |
| Septicemia | 85 (6.8%) | 33 (5.9%) | 14 (4.2%) | 73 (6.4%) | 0.3720 |
| Acute renal failure | 27 (2.2%) | 18 (3.2%) | 5 (1.5%) | 30 (2.7%) | 0.3645 |
| Chronic renal failure | 6 (0.5%) | 5 (0.9%) | 0 (0.0%) | 10 (0.9%) | 0.2277 |
| COPD | 9 (0.7%) | 5 (0.9%) | 5 (1.5%) | 5 (0.4%) | 0.2080 |
| HF (Acute) | 11 (0.9%) | 17 (3.0%) | 3 (0.9%) | 52 (4.6%) | <0.0001*** |
| Ketoacidosis | 9 (0.7%) | 1 (0.2%) | 47 (14.2%) | 14 (1.2%) | <0.0001*** |
| Pneumonia | 9 (0.7%) | 8 (1.4%) | 3 (0.9%) | 19 (1.7%) | 0.1538 |
| Urinary tract infection | 3 (0.2%) | 3 (0.5%) | 4 (1.2%) | 13 (1.2%) | 0.0212* |
| COVID | 46 (3.7%) | 20 (3.6%) | 16 (4.8%) | 46 (4.1%) | 0.7492 |
| Hypoglycemia | 13 (1.0%) | 5 (0.9%) | 7 (2.1%) | 32 (2.8%) | 0.0027** |
| Hyperkalemia | 2 (0.2%) | 6 (1.1%) | 1 (0.3%) | 6 (0.5%) | 0.0585 |
| Hypokalemia | 5 (0.4%) | 2 (0.4%) | 0 (0.0%) | 2 (0.2%) | 0.6670 |
| Hyponatremia | 9 (0.7%) | 9 (1.6%) | 1 (0.3%) | 26 (2.3%) | 0.0026** |
| Hypernatremia | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 1 (0.09%) | 0.6179 |
| GI hemorrhage | 34 (2.7%) | 13 (2.3%) | 7 (2.1%) | 21 (1.9%) | 0.5705 |
| Angioedema | 12 (1.0%) | 2 (0.4%) | 2 (0.6%) | 5 (0.4%) | 0.3743 |
Continuous variables are presented as mean ± standard deviation (SD), and categorical variables as number (percentage). Statistical significance is denoted as follows: * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001. CKD: Chronic kidney disease, COPD: Chronic obstructive pulmonary disease; HF: Heart failure; HTN: Hypertension; IHD: Ischemic heart disease.
Age differed significantly across groups (p < 0.0001), with patients in the HTN+DM group being the oldest (69.6 ± 13.0 years) and those with DM only the youngest (56.4 ± 18.7 years). Male predominance was highest in the no HTN/DM group (62.9%) and lowest in the HTN+DM group (53.8%, p < 0.0001). In addition, comorbidities of IHD, CKD, and HF were significantly more common in HTN+DM patients (31.8, 14.5, and 15.8%, respectively; all p < 0.0001).
HbA1c was markedly elevated in diabetic patients (9.13 ± 2.77% in DM-only and 8.15 ± 2.22% in HTN+DM) compared with non-diabetic groups (6.51 ± 1.99% and 6.04 ± 1.24%, p < 0.0001). Sodium levels were lower in DM and HTN+DM patients (139.0 ± 6.0 and 138.9 ± 7.3 mmol/L, p < 0.0001). Potassium was borderline statistically significant (p = 0.0512), while HCO3 and lymphocyte percentage showed small but statistically significant variations (p = 0.0391 and p = 0.0158, respectively).
Clinical diagnosis upon admission revealed that ketoacidosis was strongly confined to the DM-only group (14.2%, p < 0.0001), while hypoglycemia was more common in both DM and HTN+DM (2.1 and 2.8%, respectively, p = 0.0027). Hyponatremia and acute HF occurred more often in HTN+DM (2.3%, p = 0.0028; 4.6%, p < 0.0001, respectively). Urinary tract infection was also slightly higher in diabetic patients. Other complications, including stroke, sepsis, pneumonia, and COPD, were not significantly different across groups.
3.4. Combined effects of diabetes and hypertension on AKI incidence and survival
Subsequently, we assessed the combined HTN-DM classification effect on AKI incidence during the ICU admission period (Figure 1C) and the 90-day mortality (Figure 1D). The highest rates of AKI were found in the HTN+DM group (48.6%) and the HTN-only group (44.5%), while the DM-only group had the lowest incidence (31.4%) (p < 0.0001 across groups).
Survival analysis further highlighted these differences (Figure 1D). Median survival was not reached in any group, as more than 50% of patients survived to 90 days across all strata. However, 90-day mortality rates differed markedly: HTN+DM had the highest mortality (37.8%), followed by HTN-only (32.4%), No HTN/DM (30.6%), and DM-only had the lowest mortality (24.5%). Global comparison across groups revealed statistically significant survival differences, as confirmed by both the log-rank test (χ2 = 26.2, df = 3, p < 0.0001) and the Wilcoxon test (χ2 = 26.5, df = 3, p = 0.0001). These results indicate that the coexistence of HTN and DM confers the greatest risk of adverse outcomes in the ICU, both in terms of AKI development and survival duration.
3.5. Multivariable analysis of predictors of AKI and the interaction between diabetes and hypertension
To further explore predictors and the independent effects of HTN and DM on AKI, we performed a multivariable logistic regression analysis (Table 3). Admission diagnosis was not included in the model because it reflects the acute clinical presentation at ICU entry, which is often highly correlated with the outcome, and including it could therefore result in overadjustment and obscure the underlying impact of baseline comorbidities and laboratory parameters. Further, instead of using the four-level classification variable (no HTN/DM, HTN only, DM only, HTN+DM), we modeled HTN and DM as separate binary variables and added their interaction term (HTN × DM). This approach allowed us to estimate the independent contribution of each comorbidity while also formally testing whether their coexistence confers a synergistic risk for AKI. Moreover, treating HTN and DM in this way provided a more parsimonious model with greater statistical efficiency, while still capturing both the main and combined effects.
Table 3.
Logistic regression for risk factors of acute kidney injury (AKI).
| Factor | Odds ratio | 95% CI | p-value |
|---|---|---|---|
| Age (per year) | 1.00 | 0.99–1.01 | 0.9048 |
| Gender (Male vs. Female) | 1.34 | 1.11–1.62 | 0.0028** |
| HTN | 1.50 | 1.19–1.89 | 0.0007*** |
| DM | 1.01 | 0.80–1.28 | 0.9293 |
| HTN × DM interaction | 1.18 | 1.07–1.36 | 0.0059** |
| CKD | 2.09 | 1.46–2.98 | <0.0001*** |
| IHD | 0.92 | 0.71–1.20 | 0.5548 |
| HF | 1.15 | 0.83–1.60 | 0.3909 |
| HbA1c (per 1%) | 0.93 | 0.89–0.97 | 0.0019** |
| Potassium (per mmol/L) | 1.22 | 1.08–1.38 | 0.0010*** |
| Sodium (per mmol/L) | 1.05 | 1.03–1.06 | <0.0001*** |
| HCO₃ (per mmol/L) | 0.99 | 0.97–1.00+ | 0.0223* |
| Lymphocytes (per 1%) | 0.97 | 0.96–0.98 | <0.0001*** |
Statistical significance is denoted as follows: * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001.
: The actual upper bound of the CI is 0.999.
CI: Confidence interval. CKD: Chronic kidney disease. DM: Diabetes mellitus. HF: Heart failure. HTN: Hypertension. IHD: Ischemic heart disease.
Specifically, HTN (OR = 1.50, 95% CI: 1.19–1.89, p = 0.0007) and CKD (OR = 2.09, 95% CI: 1.46–2.98, p < 0.0001) markedly increased the odds of AKI. Importantly, the interaction between DM and HTN remained significant (OR = 1.18, 95% CI: 1.07–1.36, p = 0.0059), reinforcing that the coexistence of these two conditions confers a synergistic risk beyond either condition alone.
Among laboratory variables, higher HbA1c (OR = 0.93, 95% CI: 0.89–0.97, p = 0.0019), higher HCO3 (OR = 0.99, 95% CI: 0.97–1.00, p = 0.0223), and higher lymphocyte percent (OR = 0.97, 95% CI: 0.96–0.98, p < 0.0001) were associated with reduced AKI risk, while higher serum potassium (OR = 1.22, 95% CI: 1.08–1.38, p = 0.0010) and sodium (OR = 1.05, 95% CI: 1.03–1.06, p < 0.0001) were significant predictors of increased risk. Male sex was also associated with higher odds of AKI (OR = 1.34, 95% CI: 1.11–1.62, p = 0.0028).
In contrast, age, DM alone, IHD, and HF were not independently associated with AKI after adjustment.
3.6. Cox proportional hazards model identifying mortality predictors
To identify independent predictors of mortality, we constructed a multivariable Cox proportional hazards model incorporating baseline comorbidities, laboratory values, and lymphocyte count as a marker of systemic inflammation (Table 4). Older age was a strong predictor (HR = 1.01, 95% CI: 1.01–1.02, p < 0.0001), with each additional year associated with a 1% increase in the hazard of death. Electrolyte abnormalities were also important: elevated potassium (HR = 1.31, 95% CI: 1.22–1.41, p < 0.0001) and sodium (HR = 1.03, 95% CI: 1.02–1.04, p < 0.0001) increased mortality risk, while higher bicarbonate was protective (HR = 0.97, 95% CI: 0.96–0.98, p < 0.0001). Reduced lymphocyte percentage, reflecting systemic inflammation, was strongly associated with increased mortality (HR = 0.96, 95% CI: 0.95–0.97, p < 0.0001). Among comorbidities, HF was associated with increased mortality (HR = 1.27, 95% CI: 1.02–1.58, p = 0.0299), while CKD appeared paradoxically protective after adjustment (HR = 0.76, 95% CI: 0.60–0.97, p = 0.0284), possibly reflecting overadjustment due to the inclusion of electrolyte variables that mediate CKD’s effect on mortality. Notably, the interaction between HTN and DM was significant (HR = 1.43, 95% CI: 1.04–1.97, p = 0.0286), indicating that the coexistence of both conditions conferred additional risk of mortality beyond their individual effects. Higher HbA1c was associated with reduced mortality (HR = 0.96, 95% CI: 0.93–0.99, p = 0.0131), possibly reflecting closer glycemic monitoring in diabetic patients. Other variables, including IHD, HTN, DM, and gender, were not independently associated with survival.
Table 4.
Cox proportional hazards models for mortality (90-day survival).
| Variable | HR | 95% CI | p-value |
|---|---|---|---|
| Age (per year) | 1.01 | 1.01–1.02 | <0.0001*** |
| Gender (Male vs. Female) | 0.95 | 0.83–1.08 | 0.4398 |
| HTN | 0.92 | 0.76–1.13 | 0.4291 |
| DM | 0.88 | 0.67–1.15 | 0.3304 |
| HTN × DM interaction | 1.43 | 1.04–1.97 | 0.0286* |
| CKD | 0.76 | 0.60–0.97 | 0.0284* |
| IHD | 0.92 | 0.76–1.11 | 0.3711 |
| HF | 1.27 | 1.02–1.58 | 0.0299* |
| HbA1c (per 1%) | 0.96 | 0.93–0.99 | 0.0131* |
| Potassium (mmol/L) | 1.31 | 1.22–1.41 | <0.0001*** |
| Sodium (mmol/L) | 1.03 | 1.02–1.04 | <0.0001*** |
| HCO₃ (mmol/L) | 0.97 | 0.96–0.98 | <0.0001*** |
| Lymphocytes (per 1%) | 0.96 | 0.95–0.97 | <0.0001*** |
Statistical significance is denoted as follows: * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001. CKD: Chronic kidney disease. CI: Confidence interval. DM: Diabetes mellitus; HF: Heart failure. HTN: Hypertension. HR: Hazard ratio. IHD: Ischemic heart disease.
4. Discussion
In this large cohort of ICU patients, we examined the role of DM, HTN, and their interaction in shaping the risk of AKI and mortality, while accounting for baseline comorbidities, laboratory markers, and lymphocyte percentage as an index of systemic inflammation. Our findings highlight that both AKI and mortality in the ICU are strongly influenced by the combined burden of HTN and DM. Consistent with prior large-scale work, such as the retrospective analysis by Falciglia and colleagues, where 30% of critically ill patients had DM and hyperglycemia predicted excess mortality across ICU types [16], our results confirm the adverse prognostic impact of DM in critical illness, reinforcing the heightened vulnerability of this subgroup.
HTN is among the most common chronic conditions encountered in hospitalized and critically ill patients. In general inpatient populations, elevated blood pressure may be present in up to 72% of hospital admissions [17]. Within ICU settings, hypertensive emergencies alone account for nearly one-third of cardiovascular-related admissions in a tertiary care center in Nigeria [18]. Because most of the enrolled DM patients were also hypertensive, classification by both HTN and DM status revealed that the HTN+DM and HTN-only groups had the highest AKI incidence and the lowest 90-day survival rates. This extends the evidence that patients with HTN and DM are at increased risk of chronic renal and cardiovascular diseases [19] and highlights the statistically significant interaction conferred by their coexistence. In the context of diabetic kidney disease (DKD), recent advances in pharmacotherapy have demonstrated significant renoprotective potential. The so-called four pillars of DKD management—renin–angiotensin system inhibitors (RASi), sodium–glucose cotransporter 2 inhibitors (SGLT2i), glucagon-like peptide-1 receptor agonists (GLP-1 RAs), and mineralocorticoid receptor antagonists (MRAs)—have each shown benefits in slowing the progression of kidney disease and reducing cardiovascular risk in patients with DM [20–22]. Notably, our study did not capture data on the administration of these agents, which represents an important limitation, as their use could have modulated the risk of AKI and influenced outcomes in the diabetic subgroup.
With respect to AKI predictors, logistic regression analysis identified HTN and CKD as strong independent predictors of AKI. In line with previous reports [23], CKD itself emerged as a major risk factor for AKI, reinforcing the concept of cumulative renal vulnerability in critical illness. Importantly, the interaction between HTN and DM remained significant, reinforcing that the coexistence of these two conditions confers a statistically significant interaction risk beyond either condition alone. Male sex and elevated serum sodium and potassium levels were also associated with higher odds of AKI, whereas HbA1c, lymphocyte percentage, and HCO3 were inversely associated with the risk of AKI (OR < 1). The association of low HbA1c with AKI may reflect that low HbA1c in the ICU often represents frailty, malnutrition, or impaired physiologic reserve, such as in patients with end-stage renal disease [24]. Conversely, other studies have reported that elevated HbA1c levels are associated with increased AKI risk in adults with type 2 DM and underlying CKD [25,26].
Lymphopenia also emerged as an important predictor of AKI. This finding is biologically plausible, as lymphopenia represents immune dysregulation and, in the ICU setting, reflects a maladaptive host response marked by excessive inflammation and impaired immunity, predisposing patients to secondary infections, organ dysfunction, and poor outcomes [27]. Persistent lymphopenia and an elevated neutrophil-to-lymphocyte ratio at admission have been identified as predictors of AKI following respiratory viral infections, with the greatest risk observed in influenza [28]. Similarly, reduced serum bicarbonate, a marker of metabolic acidosis, has been reported as a risk factor for AKI [29]. Electrolyte abnormalities, including hyperkalemia [30] and hypernatremia [31], have also been shown to accelerate AKI progression and worsen prognosis. Taken together, our findings support the view that AKI risk in the ICU is amplified by the convergence of cardiometabolic comorbidities, electrolyte disturbances, acid–base imbalance, and immune dysfunction.
In terms of mortality, the Cox regression model identified several independent predictors of 90-day mortality. Age was among the strongest risk factors, with each additional year increasing the hazard of death by 1%. Disturbances in serum electrolytes were also powerful predictors: elevated potassium and sodium were associated with higher mortality, while lower lymphocyte percentages and lower serum HCO3 were each associated with increased mortality risk. These results align with previous ICU studies showing that age is a consistent predictor of mortality [32,33] and that hyperkalemia increases death risk [34]. Rising serum sodium within the first 48 hours of ICU admission has also been linked to higher mortality, both in patients admitted with normonatremia and hypernatremia [35]. In a large multicenter cohort of general ICU patients (n ≈ 11,125), dysnatremia at admission—including mild to severe deviations—was independently associated with increased 30-day mortality [36]. Moreover, studies in critically ill patients with acute respiratory distress syndrome (ARDS) [37] or infective endocarditis [38] found that lower admission serum bicarbonate was independently associated with increased 28-day mortality. Consistent with a prior study, lower lymphocyte count was a robust marker of poor survival [39], although one study suggested a U-shaped association with mortality in sepsis and septic shock, with increased risk at both low and high levels [40]. This highlights the central role of systemic inflammation in driving adverse ICU outcomes.
HF was associated with increased mortality risk (HR = 1.27, p = 0.0299), consistent with prior evidence of cardiac dysfunction worsening ICU outcomes. CKD, however, appeared paradoxically protective after adjustment (HR = 0.76, p = 0.0284), likely reflecting overadjustment, as the model includes potassium, sodium, and bicarbonate, which may mediate CKD’s harmful effects on mortality. Importantly, the coexistence of HTN and DM significantly increased risk beyond their individual effects, supporting prior studies that emphasize the interaction impact of these common cardiometabolic disorders [41]. Higher HbA1c was associated with reduced mortality (HR = 0.96, p = 0.0131), which may reflect closer glycemic monitoring in recognized diabetic patients. By contrast, IHD comorbidity and DM or HTN in isolation did not independently predict death after adjustment, suggesting that their effects are mediated through pathways involving renal dysfunction, electrolyte imbalance, and inflammation. Nevertheless, in a large general ICU cohort of over 3,000 patients without known DM, admission HbA1c exhibited a U-shaped association with post-ICU mortality. Both low (< 5.0%) and high (≥ 6.5%) HbA1c levels were independently associated with increased risk of death [42].
The study has several strengths and limitations that should be acknowledged. The study utilized a large cohort spanning 10 years of ICU real-world data, providing robust statistical power. Furthermore, the use of electronic health records has eliminated recall bias; however, as with all observational studies, residual confounding from unmeasured (e.g., lifestyle) or incomplete variables cannot be ruled out. Misclassification is possible; however, evidence from epidemiological studies suggests that associations are unlikely to be greatly affected where misclassification is non-differential [43]. Furthermore, the primary endpoint (90-day mortality) is objectively and routinely recorded; thus, it is unlikely to be affected. Additionally, data on the use of specific medications—including renin–angiotensin system inhibitors (RASi), sodium–glucose cotransporter 2 inhibitors (SGLT2i), glucagon-like peptide-1 receptor agonists (GLP-1 RAs), and mineralocorticoid receptor antagonists (MRAs)—were not available in the electronic health records for the study period, and some of these drug classes were infrequently prescribed during the earlier years of data collection. Given the established renoprotective effects of these agents [20–22], their absence as covariates represents a potential source of residual confounding. Future studies should incorporate detailed pharmacotherapy data to better delineate the contribution of these medications to AKI risk and survival in critically ill patients with DM.
5. Conclusion
Taken together, these results underscore that ICU outcomes are determined by a complex interplay between chronic comorbidities and acute physiologic derangements. The consistent role of lymphopenia across models highlights inflammation as a central mediator linking comorbidities to both AKI and mortality. Notably, the interaction between HTN and DM exerted a synergistic effect across outcomes: patients with both conditions had a significantly higher risk of developing AKI during their ICU stay and a greater hazard of death to 90 days compared with those with either condition alone. This indicates that the coexistence of HTN and DM does not simply additively increase risk but acts in a synergistic manner, amplifying renal vulnerability and mortality.
Funding Statement
This research received financial support provided by the Deanship of Research at Jordan University of Science and Technology (Grant number: 20230430). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Author contributions
NA: Conceptualization, Methodology, Formal analysis, Data Curation, Writing - Original Draft, Visualization, Supervision, Project administration; SA: Methodology, Investigation, Data Curation, Visualization, Supervision, Funding acquisition, Project administration; RK: Investigation, Data Curation, Visualization. All authors reviewed and edited the manuscript and approved the final version.
Disclosure statement
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
Ethical statement
Ethical approval for this study was obtained from the Institutional Review Board (IRB) of Jordan University of Science and Technology (IRB#: 2023/451), Irbid, Jordan. All patient data were de-identified prior to analysis to ensure confidentiality, and no personal identifiers were accessed or stored. The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki and followed institutional and national guidelines for research involving human subjects.
Data availability statement
The datasets generated and analyzed during the current study are not publicly available due to patient confidentiality and institutional data protection policies but are available from the corresponding author on reasonable request and with permission from King Abdullah University Hospital, Jordan.
References: Papers of special note have been highlighted as either of interest (*) or of considerable interest (**) to readers.”
- 1.Adhikari NK, Fowler RA, Bhagwanjee S, et al. Critical care and the global burden of critical illness in adults. Lancet. 2010;376(9749):1339–1346. doi: 10.1016/S0140-6736(10)60446-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hoste EAJ, Kellum JA, Selby NM, et al. Global epidemiology and outcomes of acute kidney injury. Nat Rev Nephrol. 2018;14(10):607–625. doi: 10.1038/s41581-018-0052-0 [DOI] [PubMed] [Google Scholar]; **Authoritative global review establishing AKI epidemiology benchmarks across healthcare settings and its association with short- and long-term mortality
- 3.Kellum JA, Romagnani P, Ashuntantang G, et al. Acute kidney injury. Nat Rev Dis Primers. 2021;7(1):52. doi: 10.1038/s41572-021-00284-z [DOI] [PubMed] [Google Scholar]
- 4.Neyra JA, Chawla LS.. Acute kidney disease to chronic kidney disease. Crit Care Clin. 2021;37(2):453–474. doi: 10.1016/j.ccc.2020.11.013 [DOI] [PubMed] [Google Scholar]
- 5.Odutayo A, Wong CX, Farkouh M, et al. AKI and long-term risk for cardiovascular events and mortality. J Am Soc Nephrol. 2017;28(1):377–387. doi: 10.1681/ASN.2016010105 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kar P, Jones KL, Horowitz M, et al. Management of critically ill patients with type 2 diabetes: The need for personalised therapy. World J Diabetes. 2015;6(5):693–706. doi: 10.4239/wjd.v6.i5.693 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Namazi N, Moghaddam SS, Esmaeili S, et al. Burden of type 2 diabetes mellitus and its risk factors in North Africa and the Middle East, 1990-2019: findings from the Global Burden of Disease study 2019. BMC Public Health. 2024;24(1):98. doi: 10.1186/s12889-023-16540-8 [DOI] [PMC free article] [PubMed] [Google Scholar]; * Comprehensive Global Burden of Disease analysis quantifying the rising prevalence of type 2 diabetes in the Middle East and North Africa region, providing essential epidemiological context for the present study.
- 8.Forbes JM, Cooper ME.. Mechanisms of diabetic complications. Physiol Rev. 2013;93(1):137–188. doi: 10.1152/physrev.00045.2011 [DOI] [PubMed] [Google Scholar]
- 9.Kompoti M, Michalia M, Salma V, et al. Glycated hemoglobin at admission in the intensive care unit: clinical implications and prognostic relevance. J Crit Care. 2015; 2015/02/01/30(1):150–155. doi: 10.1016/j.jcrc.2014.08.014 [DOI] [PubMed] [Google Scholar]
- 10.Mahmoodpoor A, Hamishehkar H, Shadvar K, et al. Relationship between glycated hemoglobin, Intensive Care Unit admission blood sugar and glucose control with ICU mortality in critically ill patients. Indian J Crit Care Med. 2016;20(2):67–71. doi: 10.4103/0972-5229.175938 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bretzel RG. Comorbidity of diabetes mellitus and hypertension in the clinical setting: a review of prevalence, pathophysiology, and treatment perspectives. Clin Ther. 2007;29:S35–S43. doi: 10.1016/j.clinthera.2007.07.010 [DOI] [Google Scholar]
- 12.Yuan Y, Isasi CR, Al-Rousan T, et al. Associations of concurrent hypertension and type 2 diabetes with mortality outcomes: a prospective study of US adults. Diabetes Care. 2025;48(7):1241–1250. doi: 10.2337/dca24-0118 [DOI] [PMC free article] [PubMed] [Google Scholar]; ** Large prospective cohort study demonstrating that concurrent hypertension and type 2 diabetes are synergistically associated with increased all-cause and cardiovascular mortality, directly supporting the interaction findings of the present study.
- 13.Ajlouni K, Batieha A, Jaddou H, et al. Time trends in diabetes mellitus in Jordan between 1994 and 2017. Diabet Med. 2019;36(9):1176–1182. doi: 10.1111/dme.13894 [DOI] [PubMed] [Google Scholar]
- 14.Salameh AB, Hyassat D, Suhail A, et al. The prevalence of hypertension and its progression among patients with type 2 diabetes in Jordan. Ann Med Surg (Lond). 2022;73:103162. doi: 10.1016/j.amsu.2021.103162 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Al-Azzam N, Azzam A, Elsalem S, et al. Hypertension prevalence and associated factors among patients with diabetes: A retrospective cross-sectional study from Jordan. Ann Med Surg (Lond). 2021;61:126–131. doi: 10.1016/j.amsu.2020.12.038 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Falciglia M, Freyberg RW, Almenoff PL, et al. Hyperglycemia-related mortality in critically ill patients varies with admission diagnosis. Crit Care Med. 2009;37(12):3001–3009. doi: 10.1097/CCM.0b013e3181b083f7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Bress AP, Anderson TS, Flack JM, et al. The management of elevated blood pressure in the acute care setting: a scientific statement from the American Heart Association. Hypertension. 2024;81(8):e94–e106. doi: 10.1161/HYP.0000000000000238 [DOI] [PubMed] [Google Scholar]
- 18.Oyan B, Jumbo MF, Abere S.. Burden and outcomes of cardiovascular diseases in the ICU of a Nigerian Tertiary Hospital. Cureus. 2025;17(8):e90886. /;17(8):e90886. doi: 10.7759/cureus.90886 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lastra G, Syed S, Kurukulasuriya LR, et al. Type 2 diabetes mellitus and hypertension: an update. Endocrinol Metab Clin North Am. 2014;43(1):103–122. doi: 10.1016/j.ecl.2013.09.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Mima A, Nomura A, Yasuzawa T.. Update on the pathophysiology and treatment of diabetic kidney disease: a narrative review. Expert Rev Clin Immunol. 2025;21(7):921–928. doi: 10.1080/1744666X.2025.2521086 [DOI] [PubMed] [Google Scholar]; * Up-to-date narrative review summarizing the four pillars of diabetic kidney disease management, including RASi, SGLT2i, GLP-1 RAs, and MRAs, and their renoprotective mechanisms.
- 21.Mima A. A narrative review of diabetic kidney disease: previous and current evidence-based therapeutic approaches. Adv Ther. 2022;39(8):3488–3500. doi: 10.1007/s12325-022-02223-0 [DOI] [PubMed] [Google Scholar]
- 22.Mima A. Renal protection by sodium-glucose cotransporter 2 inhibitors and its underlying mechanisms in diabetic kidney disease. J Diabetes Complications. 2018;32(7):720–725. doi: 10.1016/j.jdiacomp.2018.04.011 [DOI] [PubMed] [Google Scholar]
- 23.See EJ, Jayasinghe K, Glassford N, et al. Long-term risk of adverse outcomes after acute kidney injury: a systematic review and meta-analysis of cohort studies using consensus definitions of exposure. Kidney Int. 2019;95(1):160–172. doi: 10.1016/j.kint.2018.08.036 [DOI] [PubMed] [Google Scholar]
- 24.Abdelhafiz AH, Sinclair AJ.. Low HbA1c and increased mortality risk-is frailty a confounding factor? Aging Dis. 2015;6(4):262–270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Xu Y, Surapaneni A, Alkas J, et al. Glycemic control and the risk of acute kidney injury in patients with type 2 diabetes and chronic kidney disease: parallel population-based cohort studies in U.S. and Swedish routine care. Diabetes Care. 2020;43(12):2975–2982. doi: 10.2337/dc20-1588 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Lin Y, Zhang S, Chen Z, et al. Stress hyperglycemia ratio as a predictor of acute kidney injury and its outcomes in critically ill patients. Ren Fail. 2025;47(1):2499228. doi: 10.1080/0886022X.2025.2499228 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wang Z, Zhang W, Chen L, et al. Lymphopenia in sepsis: a narrative review. Crit Care. 2024;28(1):315. doi: 10.1186/s13054-024-05099-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Fan G, Liu X, Xu F, et al. Associations between lymphopenia and the development/recovery of acute kidney injury among patients hospitalized with respiratory virus infections. Kidney Med. 2025;7(9):101063. doi: 10.1016/j.xkme.2025.101063 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hu J, Wang Y, Geng X, et al. Metabolic acidosis as a risk factor for the development of acute kidney injury and hospital mortality. Exp Ther Med. 2017;13(5):2362–2374. doi: 10.3892/etm.2017.4292 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wang W, Yang H, Perry Wilson F, et al. Association between hyperkalemia and outcomes in hospitalized patients with acute kidney injury. Intensive Crit Care Nurs. 2025;90:104118. doi: 10.1016/j.iccn.2025.104118 [DOI] [PubMed] [Google Scholar]
- 31.Jansch C, Matyukhin I, Marahrens M, et al. Hypernatremia: epidemiology and predictive role in emerging and established acute kidney injury. J Clin Med Res. 2023;15(8-9):399–405. doi: 10.14740/jocmr4990 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ismail AJ, Hassan W, Nor MBM, et al. The impact of age on mortality in the intensive care unit: a retrospective cohort study in Malaysia. Acute Crit Care. 2024;39(3):390–399. doi: 10.4266/acc.2024.00640 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Gonçalves-Pereira J, Oliveira A, Vieira T, et al. Critically ill patient mortality by age: long-term follow-up (CIMbA-LT). Ann Intensive Care. 2023;13(1):7. doi: 10.1186/s13613-023-01102-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Xu C, Luo Y, Chen X, et al. Hyperkalemia is associated with short- and mid-term mortalities in critically ill patients in the MIMIC IV database. Sci Rep. 2025;15(1):6539. doi: 10.1038/s41598-025-91194-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Grim CCA, Termorshuizen F, Bosman RJ, et al. Association between an increase in serum sodium and in-hospital mortality in critically ill patients. Crit Care Med. 2021;49(12):2070–2079. doi: 10.1097/CCM.0000000000005173 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Breen T, Brueske B, Sidhu MS, et al. Abnormal serum sodium is associated with increased mortality among unselected cardiac intensive care unit patients. J Am Heart Assoc. 2020;9(2):e014140. doi: 10.1161/JAHA.119.014140 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Han J, Wang L, Jin L, et al. Association between bicarbonate levels and mortality among acute respiratory distress syndrome patients: an analysis based on Medical Information Mart for Intensive Care database. PLoS One. 2025;20(6):e0325498. doi: 10.1371/journal.pone.0325498 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Huang Y, Ao T, Zhen P, et al. Association between serum bicarbonate and 28-day mortality in critically ill patients with infective endocarditis: a cohort study from MIMIC-IV. Sci Rep. 2025;15(1):312. doi: 10.1038/s41598-024-84385-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Gutierrez-Rodriguez R, Aguilar-Alonso E, Arias-Verdu MD, et al. Lymphopenia assessment in ICU patients and relationship with mortality. ICMx. 2015;3(S1):A340. doi: 10.1186/2197-425x-3-s1-a340 [DOI] [Google Scholar]
- 40.Zhang G, Wang T, An L, et al. U-shaped correlation of lymphocyte count with all-cause hospital mortality in sepsis and septic shock patients: a MIMIC-IV and eICU-CRD database study. Int J Emerg Med. 2024;17(1):101. doi: 10.1186/s12245-024-00682-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Strain WD, Paldánius PM.. Diabetes, cardiovascular disease and the microcirculation. Cardiovasc Diabetol. 2018;17(1):57. doi: 10.1186/s12933-018-0703-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Liu C, Pang K, Tong J, et al. The association between hemoglobin A1c and all-cause mortality in the ICU: a cross-section study based on MIMIC-IV 2.0. Front Endocrinol (Lausanne). 2023;14:1124342. doi: 10.3389/fendo.2023.1124342 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Wacholder S, Hartge P, Lubin JH, et al. Non-differential misclassification and bias towards the null: a clarification. Occup Environ Med. 1995;52(8):557–558. doi: 10.1136/oem.52.8.557 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and analyzed during the current study are not publicly available due to patient confidentiality and institutional data protection policies but are available from the corresponding author on reasonable request and with permission from King Abdullah University Hospital, Jordan.