Frequency of complications of DKA in children admitted in PICU Authors: Meisam Sharifzadeh, Masoud Mohammadpour, Bahareh Yaghmaee, Seyyed Abbas Hassani, Zeinab Najafi, Farzaneh Beirami, Roxana Pazouki, Mina Baroughi Published: 2026 ABSTRACT AbstractObjectiveDiabetic ketoacidosis (DKA) is a serious and potentially life-threatening complication of diabetes mellitus in children, frequently requiring admission to the pediatric intensive care unit (PICU). The treatment of diabetic ketoacidosis has long been problematic, divisive, and fraught with uncertainty. Understanding the frequency of associated complications is essential for improving clinical management and outcomes. The aim of this study was to determine the frequency and severity of DKA in pediatric patients admitted to the PICU, to identify the pattern of electrolyte abnormalities, and to evaluate the association between DKA severity, electrolyte disturbances, and the risk of cerebral injury and other acute complications.MethodsThis cross-sectional retrospective study was conducted on pediatric patients aged 1 month to 16 years admitted with DKA to the PICU of a tertiary referral center between December 2018 and December 2022. DKA was defined according to ISPAD criteria. Data on demographic characteristics, laboratory findings, and clinical outcomes were extracted from medical records. The frequency of complications, including electrolyte disturbances, cerebral injury, and acute kidney injury (AKI), was evaluated.ResultsA total of 161 patients were included, with a mean age of 8.22 ± 4.08 years. The most common electrolyte abnormality was hypophosphatemia (60.9%), followed by hypokalemia (50.9%) and hyponatremia (46.6%). Hypernatremia and hyperkalemia were observed in 15.5% and 19.3% of patients, respectively. Cerebral injury occurred in 8.1% of patients, while AKI was observed in 3.7%. Patients with severe DKA had a significantly higher risk of cerebral injury compared to those with moderate DKA. Hypernatremia was also associated with an increased risk of cerebral injury.ConclusionComplications of DKA, particularly electrolyte disturbances and cerebral injury, are relatively common among children admitted to the PICU. Severe DKA and hypernatremia are associated with a higher risk of cerebral injury. Early recognition and careful management of these complications are crucial to improving patient outcomes.Keywords: Diabetic ketoacidosis, Pediatric intensive care unit, Cerebral edema, Electrolyte disturbances Introduction Diabetic ketoacidosis (DKA) is the most serious acute complication of diabetes mellitus in children and remains a leading cause of morbidity and mortality in this population [1]. It occurs predominantly in patients with type 1 diabetes mellitus (T1DM), which is the most common form of diabetes in childhood [2]. DKA is characterized by the triad of hyperglycemia, metabolic acidosis, and ketonemia or ketonuria, resulting from insulin deficiency and an increase in counterregulatory hormones. DKA is common (30%) of children with T1DM, and it occurs more frequently in those with poor metabolic control and a history of past episodes of ketoacidosis (1–10% of known cases of diabetes per year) [3]. The occurrence of DKA is higher among adolescents, particularly females, and is also associated with psychosocial and behavioral disorders, as well as socioeconomic disadvantage and reduced access to healthcare. Diabetes-related deaths are far more common in underdeveloped and emerging nations, where patients may even pass away before receiving any medical attention [4]. DKA accounts for approximately 2–3% of diabetes-related mortality. The risk of death in young children is three times higher than in diabetes adults. Unusual processes include extravascular fluid transfer, dehydration, reduced perfusion, pH alterations, and electrolyte disturbances that impair numerous functions occur during DKA episode [5]. The brain is the organ most frequently impacted by DKA, which results in cerebral edema and is linked to the highest morbidity and fatality rates. Headache, bradycardia, frequent vomiting, changes in nervous status (restlessness, excitability, sleepiness, incontinence), localized neurological symptoms (cerebral nerve palsy and pupillary response alterations), elevated blood pressure, and irregular breathing are some of the signs of cerebral edema [6]. The occurrence of cerebral edema is less than 1%, and is due to multiple factors like severity of DKA, age under 5, severe reduction (PCo2 21), BUN > 20, and severe acidosis [7]. In addition to cerebral edema other complications like cerebral injury (thrombosis, bleeding, or infarction of cerebral vessels), electrolyte disorders (hyperkalemia, hypokalemia, hyponatremia, and hypernatremia, hypophosphatemia), cardiovascular changes and shock, acute kidney injury, and cardiopulmonary complications (arrhythmia, myocarditis, hypertension) [8]. A retrospective study conducted in Croatia reported that 24.7% of children with newly diagnosed type 1 diabetes presented with DKA, and a significant proportion had moderate to severe dehydration (76%) at admission. Cerebral edema occurred in 5.2% of cases and was associated with mortality, highlighting the clinical severity of DKA in pediatric patients [9]. In another retrospective cohort study of children admitted to the PICU with DKA, AKI was observed in 30% of patients, the majority being mild (stage 1). Importantly, AKI was not associated with prolonged hospitalization or adverse outcomes, suggesting a generally reversible condition in this population [10]. A monocentric study from Italy demonstrated that 51.5% of children with newly diagnosed type 1 diabetes presented with DKA, and approximately 10% required PICU admission due to severe clinical presentation. Reported complications included acute kidney injury, cerebral edema, and thrombotic events, with rare cases progressing to multi-organ failure and death [11]. A retrospective analysis from a tertiary care center reported a wide spectrum of complications in children with severe DKA admitted to the PICU, including electrolyte disturbances such as hyperchloremia (35.9%), hypokalemia (30.8%), and hyponatremia (26.7%). Additionally, cerebral edema (16.4%), acute kidney injury (10.2%), and shock (13.3%) were observed, with 30% requiring mechanical ventilation and a mortality rate of 5.4% [12]. Despite the existing literature, there is considerable variability in the reported frequency and spectrum of DKA-related complications across different regions and healthcare settings. In particular, data focusing on critically ill pediatric patients admitted to the PICU remain limited, especially in developing countries. A clearer understanding of the burden and pattern of these complications is essential for improving early recognition, optimizing management strategies, and reducing morbidity and mortality. Therefore, this study aimed to determine the frequency of complications, including electrolyte disturbances, cerebral injury, and acute kidney injury, in pediatric patients with DKA admitted to the PICU, and to assess their associated clinical outcomes. Method & materials Study design This cross-sectional, retrospective study was conducted at the PICU of Children’s Medical Center, a tertiary referral hospital affiliated with Tehran University of Medical Sciences, Tehran, Iran. The study period spanned four years, from December 2018 to December 2022. This center serves as a major referral hospital for critically ill pediatric patients from across the country, ensuring a diverse and representative patient population. All data were obtained through review of archived medical records and hospital databases. Participants and inclusion criteria All pediatric patients aged between 1 month and 16 years who were admitted to the PICU during the study period with a diagnosis of DKA were considered eligible for inclusion. The diagnosis of DKA was established based on the International Society for Pediatric and Adolescent Diabetes (ISPAD) criteria, including hyperglycemia (blood glucose ≥ 200 mg/dL), metabolic acidosis (venous pH < 7.3 or serum bicarbonate < 15 mmol/L), and the presence of ketonemia or ketonuria. Patients were included if sufficient clinical and laboratory data were available in their medical records to allow assessment of study variables and outcomes. Patients with incomplete or missing key data, including laboratory values or clinical outcomes, were excluded from the analysis. In cases of multiple admissions for the same patient, only the first episode was considered to avoid duplication of data. The study was approved by the Regional Research Ethics Committee of Tabriz University of Medical Sciences, Tabriz, Iran (Approval No: IR.TBZMED.VCR.REC.1401.2189). Definitions DKA severity was classified according to the International Society for Pediatric and Adolescent Diabetes (ISPAD) guidelines based on the degree of acidosis: mild DKA was defined as venous pH 7.2–7.3, moderate DKA as pH 7.1–7.2, and severe DKA as pH < 7.1. Electrolyte abnormalities were defined using standard laboratory thresholds as follows: hyponatremia (serum sodium < 135 mmol/L), hypernatremia (serum sodium > 145 mmol/L), hypokalemia (serum potassium < 3.5 mmol/L), hyperkalemia (serum potassium > 5.0 mmol/L), and hypophosphatemia (serum phosphate < 3 mg/dL). Cerebral injury was defined as the presence of neurological manifestations including decreased level of consciousness, abnormal neurological examination findings, or clinical suspicion of cerebral edema, with or without radiologic confirmation. Data collection Clinical data included level of consciousness at admission, need for mechanical ventilation, duration of PICU stay, and time to resolution of DKA. Laboratory data at admission included serum glucose, sodium, potassium, phosphate, and creatinine levels. In addition, outcome variables were recorded, including the occurrence of cerebral injury and AKI. All laboratory measurements were obtained as part of routine clinical care at the time of admission and during hospitalization. Data extraction was performed by trained researchers, and records were reviewed to ensure accuracy and completeness. To minimize bias, only cases with sufficient clinical and laboratory information were included, and standardized definitions were applied consistently across all patients. Statistical analysis Data were analyzed using SPSS version 23. Quantitative variables were reported as mean ± SD, and qualitative variables as frequency and percentage. Chi-square test was used for categorical variables and ANOVA for continuous variables. Odds ratios (OR) with 95% confidence intervals (CI) were calculated. A p-value ≤ 0.05 was considered statistically significant. Results Characterization of patients A total of 161 pediatric patients with DKA admitted to the PICU during the study period were included in the analysis. The mean age was 8.22 ± 4.08 years, and a female predominance was observed (56.9% females vs. 43.1% males). The average duration of PICU hospitalization was 2.13 ± 2.15 days, while the mean time to resolution of DKA was 1.62 ± 1.14 days, indicating relatively rapid metabolic stabilization in most patients. These demographic and hospitalization characteristics are summarized in Table 1. Table 1Baseline characteristics of pediatric DKA patients (n = 161)VariableResultAge (years), mean ± SD8.22 ± 4.08Sex, n (%) Female91 (56.9%) Male70 (43.1%)PICU length of stay (days), mean ± SD2.13 ± 2.15Time to DKA resolution (days), mean ± SD1.62 ± 1.14 Baseline characteristics of pediatric DKA patients (n = 161) Electrolyte disturbances Electrolyte disturbances were common among patients with DKA. Hyponatremia (sodium lower than 135 mmol/L) was observed in 75 patients (46.6%), while hypernatremia (serum sodium > 145 mmol/L) occurred in 25 patients (15.5%). Hypokalemia (serum potassium < 3.5 mmol/L) and hyperkalemia (serum potassium > 5.0 mmol/L) were present in 82 (50.9%) and 31 (19.3%) of patients, respectively. The most frequent electrolyte abnormality was hypophosphatemia (serum phosphate < 3 mg/dL), detected in 98 (60.9%) of cases. As well as, only 13 (8.1%) of patients had hypoglycemia, defined as blood sugar below 60 mg/dl (Table 2). Table 2Laboratory findings and electrolyte disturbances in DKA patientsVariableFrequency n (%)Hyponatremia (Na⁺ < 135 mmol/L)75 (46.6%)Hypernatremia (Na⁺ > 145 mmol/L)25 (15.5%)Hypokalemia (K⁺ < 3.5 mmol/L)82 (50.9%)Hyperkalemia (K⁺ > 5.0 mmol/L)31 (19.3%)Hypophosphatemia (P < 3 mg/dL)98 (60.9%)Hypoglycemia (Glucose < 60 mg/dl)13 (8.1%) Laboratory findings and electrolyte disturbances in DKA patients Prevalence of cerebral edema Another finding that was observed in 8.1% of patients was cerebral edema (13 patients). No cases of seizure-related complications were recorded. Visual impairment was rare and documented in only one patient (0.6%). The complication of decreased state of consciousness occurred in 13 patients (8.1%). Acute kidney injury occurred in 6 patients (3.7%), and mechanical ventilation was required in 6 patients (3.7%). Also, in the statistical hypothesis of examining the relationship between the severity of the disease and the need for intubation and mechanical ventilation in patients, there is no significant relationship (Table 3). Table 3Clinical complications and outcomes in pediatric DKA patientsOutcomeFrequency n (%)Cerebral edema13 (8.1%)Decreased consciousness13 (8.1%)Visual impairment1 (0.6%)Acute kidney injury (AKI)6 (3.7%)Mechanical ventilation6 (3.7%) Clinical complications and outcomes in pediatric DKA patients Associations between severity and brain injury In a separate statistical analysis, the relationship between disease severity and the occurrence of brain injury was evaluated. The results demonstrated a significant association between severe disease and brain injury, such that patients with severe disease had 9.1 times higher odds of developing brain injury compared to those with moderate severity (OR = 9.1, p = 0.012). However, no significant association was observed between other severity categories and brain injury. Additionally, a significant relationship was found between the length of PICU stay and the occurrence of brain injury (p < 0.001), indicating that patients with brain injury experienced longer hospital stays (2.21 days) (Table 4). Table 4Association between disease severity, laboratory findings, and outcomesComparisonOutcomeResultStatistical testSeverity vs. Brain injurySevere DKA ↑ riskOR = 9.1p = 0.012Hypernatremia vs. Brain injuryIncreased riskOR = 3.53p = 0.029Length of stay vs. Brain injuryLonger hospitalization+ 2.21 daysp < 0.001AKI vs. SeverityNo significant association-p > 0.05Severity vs. Mechanical ventilationsignificant association-p = 0.034 Association between disease severity, laboratory findings, and outcomes Investigating the relationship between hypernatremia and brain damage It was also important to investigate the link between hypernatremia and brain injury. Our analysis revealed that hypernatremia (Na⁺ > 145 mmol/L) was significantly associated with cerebral injury. Patients with hypernatremia had a 3.53-fold higher risk of brain injury compared to normo-natremic patients (OR = 3.53, p = 0.029) (Table 4). The extent of acute kidney injury in DKA patients Acute kidney injury occurred in 6 patients (3.7%) (Table 3). No statistically significant association was found between disease severity and the development of AKI (p > 0.05). However, a statistically significant relationship was observed between disease severity and the requirement for mechanical ventilation (p = 0.034), indicating that the need for mechanical ventilation varied across different severity levels (Table 4). Discussion This was a cross-sectional study that examined demographic data, electrolyte derangements, neurological and renal complications in children who were admitted in a tertiary care center PICU with DKA over a period of four years. The results gave insights into both the typical presentation of DKA and the relationships between disease severity, electrolyte imbalance, and clinical outcomes. The main findings were the high prevalence of electrolyte disturbances, particularly hypophosphatemia (60.9%) and hypokalemia (50.9%), a significant association between severe DKA and cerebral injury (OR = 9.1), and the identification of hypernatremia as a risk factor for brain injury (OR = 3.53). In contrast, AKI was relatively uncommon (3.7%) and was not associated with disease severity. The increasing prevalence of DKA as a presenting feature of T1D has been reported across multiple regions, including China, Australia, Saudi Arabia, and Serbia, where it is 50.1%, 48.1%, 37.7%, and 35.1%, respectively [13, 14]. These findings are consistent with the global trend toward delayed diagnosis of T1D in some populations. In a 15-year study (1971–1985) conducted in Birmingham with 746 patients (representing all age groups), (Krentz and colleagues) found that the lowest age of prevalence was over 70 years old and the highest age of prevalence was between 20 and 29 years [15]. The prevalence of diabetic ketoacidosis was recently found to be higher in type 1 diabetes, while type 2 diabetes also had a significant number of instances. In Balasubramanayam’s study (in 1999, in Houston, America), which was conducted on 120 patients with diabetic ketoacidosis, 39% of the patients had type 2 diabetes and 53% of them had type 1 diabetes. In 8% of the cases, an unknown type of diabetes was reported. Moreover, according to this study, diabetic ketoacidosis was common in non-white patients with type 2 diabetes [16]. Yan reported a five-year retrospective study conducted in Taiwan in January 2000 that included 141 patients with diabetic ketoacidosis. Based on this, it was determined that 32% of these patients had type 2 diabetes and 22.7% had type 1 diabetes. The type of diabetes was uncertain in 22.7% of the cases [17]. Numerous type 2 diabetes patients who developed diabetic ketoacidosis were included in the latest study. This study differs from earlier studies in that the majority of diabetic ketoacidosis patients had type 1 diabetes, with a 4.3 to 1 type 1 diabetes to type 2 diabetes ratio. Electrolyte abnormalities were highly prevalent in our cohort. Hyponatremia (46.6%) was more common than hypernatremia (15.5%), consistent with previous studies attributing this to osmotic shifts caused by hyperglycemia [18]. Hypokalemia and hyperkalemia were also frequently observed, reflecting total body potassium depletion despite normal or elevated serum levels at presentation. Notably, hypophosphatemia was the most common abnormality (60.9%), emphasizing the importance of close electrolyte monitoring during DKA management to prevent complications such as cardiac arrhythmias and neurological dysfunction. One of the most potentially lethal complications to be faced is cerebral edema, which observed in 8.1% of the children. While the overall rate is not high, the potential for severe morbidity and mortality renders this a serious consideration in the management of DKA in children. Importantly, severe DKA was strongly associated with cerebral injury, with a 9.1-fold increased risk compared to moderate cases. This makes it all the more important to exercise increased caution and early treatment in patients presenting with severe DKA. Additionally, hypernatremia was identified as a significant risk factor for cerebral injury, supporting the hypothesis that rapid shifts in serum sodium and osmolality contribute to neurological complications. Another significant finding was between cerebral trauma and PICU stay duration. The length of stay for cerebral trauma patients was much higher by 2.21 days, because of the extended care needed and longer recovery phase in such complex conditions. This puts focus on cost of neurological complications to health resources and the need for early prevention and detection strategies. AKI was observed in 3.7% of patients and was not significantly associated with disease severity. This finding is consistent with previous studies suggesting that AKI in DKA is often mild and reversible, particularly with early fluid resuscitation. The low incidence in our study may reflect effective early management protocols in the PICU setting. While potassium levels are frequently normal or slightly elevated during the initial evaluation of individuals with DKA, even while the total body potassium is inadequate, hypokalemia is a rare finding. Normal or elevated serum potassium levels reflect a shift of potassium from the intracellular to the extracellular space, caused by concurrent acidosis and hyperglycemia-induced extracellular hypertonicity [19]. These findings highlight the need for careful and gradual correction of electrolyte imbalances to prevent serious complications such as cardiac arrhythmias and neurological dysfunction [20]. Moreover, based on our results, hyponatremia was observed in nearly half of the patients, which is consistent with the well-established dilutional effect of hyperglycemia. Patients with DKA are more likely to have hyponatremia than hypernatremia [21]. High plasma glucose concentration raises plasma tonicity, which moves water out of the cell and into the extracellular environment, diluting the extracellular sodium. Additionally, the kidneys’ ability to retain water is limited by vasopressin secretion. Taken together, these mechanisms result in hyponatremia [22]. A typical electrolyte outcome of DKA, which manifests as a result of the dilutional action of hyperglycemia, is hyponatremia. However, hypernatremia in DKA is an uncommon presentation that tends to have higher rates of morbidity and mortality in children. Renal complications were relatively infrequent in this study. AKI was identified in a small proportion of patients and was not significantly associated with the severity of DKA. This contrasts with some previous reports that have suggested a higher incidence of renal impairment, particularly in the setting of severe dehydration. Renal acid excretion becomes insufficient as the glomerular filtration rate declines, which may lead to metabolic acidosis [23]. The existence of AKI may impede a patient’s recovery from metabolic acidosis because one of the objectives of DKA treatment is to normalise acidosis [24]. Additionally, AKI is a separate risk linked to a longer hospital stay and a higher rate of child mortality [25, 26]. Neurological complications, particularly cerebral edema, represent one of the most serious concerns in pediatric DKA. However, as children with hyperglycemia and DKA are susceptible to thrombosis, it is also important to consider more subtle cerebral damage, such as strokes. A case involving a 2-month-old patient who had cerebral edema and stroke as side effects of DKA is documented in a study [27]. In our study, cerebral edema was observed in a minority of patients but showed a strong association with disease severity. Also, the apparent diffusion coefficients (ADC) rise with DKA therapy, perhaps suggesting extracellular space growth relative to intracellular space. There may also be an increase in cerebral perfusion. These findings reveal a vasogenic mechanism of edema production in contrast to what would be predicted if cerebral edema following DKA treatment were brought on by osmotic cell swelling [28]. The low rate of AKI in our cohort may reflect early recognition and prompt fluid resuscitation in the PICU setting. This study contributes to the existing literature by demonstrating a notably high prevalence of hypophosphatemia and identifying hypernatremia as a significant risk factor for cerebral injury in pediatric DKA. These findings emphasize the importance of careful sodium and electrolyte management, particularly in severe cases, to reduce the risk of neurological complications. However, this study has several limitations. First, its retrospective design may be associated with incomplete or missing data and limits the ability to establish causal relationships. Second, the study was conducted in a single tertiary referral center, which may limit the generalizability of the findings to other settings or populations. Third, the relatively small sample size, particularly for less frequent complications such as acute kidney injury and cerebral injury, may have reduced the statistical power to detect certain associations. Finally, variations in clinical management and treatment protocols over the study period may have influenced the observed outcomes. Conclusion This study demonstrated that complications of DKA are common among children admitted to the PICU, with electrolyte disturbances being the most frequent findings, particularly hypophosphatemia (60.9%), hypokalemia (50.9%), and hyponatremia (46.6%). Cerebral injury occurred in 8.1% of patients and was significantly associated with disease severity, with patients presenting with severe DKA having a 9.1-fold higher risk compared to those with moderate DKA. In addition, hypernatremia was identified as an independent risk factor for cerebral injury, increasing the risk by 3.53 times. In contrast, acute kidney injury was relatively uncommon (3.7%) and was not associated with disease severity or adverse clinical outcomes. Patients with cerebral injury had a significantly longer PICU stay, highlighting the clinical and resource burden of neurological complications. Overall, these findings emphasize the importance of early recognition of severe DKA, careful monitoring of electrolyte imbalances, particularly sodium levels, and prompt management to reduce the risk of cerebral complications and improve clinical outcomes in pediatric patients. Acknowledgements None. Authors' contributions Mina Baroughi, Meisam Sharifzadeh, Masoud Mohammadpour, Bahareh Yaghmaee, Seyyed Abbas Hassani, and Zeinab Najafi designed and performed the experiments, and wrote the manuscript; Farzaneh Beirami and Roxana Pazouki analyzed the data and revised the manuscript; Mina Baroughi revised and supervised the manuscript. All authors read and approved the final manuscript. Funding None. Data availability The datasets generated and/or analysed during the current study are not publicly available due to patient confidentiality but are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This study was conducted in accordance with the principles of the Declaration of Helsinki. All experiments and procedures were conducted in compliance with the ethical principles of Tabriz University of Medical Science, Tabriz, Iran, and approved by the regional ethical committee for medical research (Ethical code: (IR.TBZMED.VCR.REC.1401.2189)). The requirement for informed consent was waived by the Ethics Committee due to the retrospective nature of the study. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets generated and/or analysed during the current study are not publicly available due to patient confidentiality but are available from the corresponding author on reasonable request.