Efficacy and safety of reduced-dose insulin infusion in pediatric diabetic ketoacidosis: a randomized clinical trial

Abstract.

Diabetic ketoacidosis (DKA) management in children traditionally employs insulin infusion rates of 0.1 units/kg/h, though emerging evidence suggests lower doses may achieve comparable outcomes with enhanced safety profiles. This study aimed to determine whether reduced-dose insulin therapy (0.05 units/kg/h) demonstrates non-inferiority compared to conventional dosing in pediatric DKA management. We conducted a prospective randomized controlled trial involving 70 children presenting with DKA at our pediatric intensive care unit between January and December 2018. Participants were randomly assigned to receive either reduced-dose insulin infusion (0.05 units/kg/h, n = 35) or standard-dose therapy (0.1 units/kg/h, n = 35). Both treatment approaches demonstrated comparable efficacy in achieving metabolic correction. Time to acidosis resolution showed no significant difference between reduced-dose and standard-dose groups (10.2 ± 5.29 vs. 11.22 ± 5.04 h, p = 0.407). However, the reduced-dose protocol exhibited superior safety characteristics, with significantly lower hypokalemia incidence (20.0% vs. 42.9%, p = 0.039) and reduced hypoglycemia rates (11.4% vs. 25.7%). Total insulin consumption was 50% lower in the reduced-dose group. In conclusion, reduced-dose insulin therapy demonstrates non-inferiority to conventional dosing while providing enhanced safety benefits, supporting consideration of lower insulin dosing strategies, particularly in resource-limited settings.

Highlights

● Reduced-dose insulin (0.05 units/kg/h) achieves comparable metabolic correction to standard dosing in pediatric DKA.

● Lower insulin dosing significantly reduces hypokalemia risk by 53% compared to conventional therapy.

● Gradual glucose reduction with reduced-dose insulin may minimize rapid osmolality changes associated with cerebral edema.

● Enhanced safety profile makes reduced-dose insulin particularly suitable for resource-limited healthcare settings.

Introduction

Diabetic ketoacidosis (DKA) continues to pose significant challenges in pediatric endocrinology, affecting approximately one-quarter to one-third of children at the time of type 1 diabetes diagnosis (1). This acute metabolic emergency carries substantial morbidity and mortality risks, particularly in healthcare environments where resources and monitoring capabilities may be constrained (2). The condition results from absolute or relative insulin deficiency, triggering a cascade of metabolic derangements including accelerated gluconeogenesis, enhanced lipolysis, and subsequent ketogenesis (3).

Traditional management approaches have centered on aggressive fluid resuscitation combined with continuous insulin infusion, with the latter serving dual purposes of suppressing ketone production and facilitating cellular glucose uptake (4). However, the evolution of insulin dosing strategies in pediatric DKA has been marked by gradual recognition that higher doses may not necessarily translate to superior outcomes and may indeed contribute to treatment-related complications (5).

The historical development of current insulin dosing protocols reveals an interesting trajectory from high-dose to more conservative approaches. Early treatment regimens employed insulin boluses of 1–2 units/kg followed by high-dose continuous infusions, practices subsequently abandoned due to increased complication rates, particularly hypoglycemia and rapid osmolality changes potentially associated with cerebral edema development (6, 7). The adoption of 0.1 units/kg/h as the “standard” dose emerged primarily from expert consensus rather than rigorous comparative evidence, representing what many clinicians now recognize as an arbitrary threshold rather than an evidence-based optimum (8).

Contemporary understanding of DKA pathophysiology has prompted reconsideration of traditional dosing paradigms. Recent investigations suggest that the primary therapeutic objective should emphasize ketosis suppression rather than rapid glucose normalization, a concept that inherently supports lower insulin dosing strategies (9). Furthermore, the recognition that cerebral edema, a devastating complication of pediatric DKA, is precipitated by rapid shifts in plasma osmolality has prompted a renewed focus on therapeutic strategies that ensure a more gradual metabolic correction (10, 11).

The global burden of pediatric diabetes disproportionately affects low- and middle-income countries, where an estimated 80% of children with type 1 diabetes reside (12). In these settings, DKA management confronts unique challenges including limited intensive care capabilities, intermittent insulin availability, restricted laboratory monitoring, and significant cost constraints (13). A recent multi-country analysis revealed that many healthcare facilities in resource-limited environments lack capacity for hourly glucose monitoring, frequent electrolyte assessments, and continuous cardiac monitoring which all considered standard components of traditional DKA protocols (14).

This clinical reality necessitates a fundamental reconsideration of DKA management, prioritizing strategies that maintain efficacy while reducing monitoring intensity. In this context, reduced-dose insulin therapy is a particularly compelling approach. It offers the potential for comparable therapeutic outcomes with several advantages, including a diminished risk of hypoglycemia, a decreased need for frequent dose adjustments, and lower overall insulin consumption, all of which improve feasibility and cost-effectiveness in resource-constrained environments (15). Recent updates to international pediatric DKA guidelines have begun acknowledging the potential benefits of lower insulin dosing strategies. The 2024 International Society for Pediatric and Adolescent Diabetes (ISPAD) guidelines now recognize that insulin infusion rates of 0.05 units/kg/h may be appropriate for specific clinical scenarios, including inter-hospital transfers and situations with limited monitoring capabilities (16). Similarly, leading pediatric centers have begun incorporating lower starting doses into their protocols, particularly for younger children and those at higher risk for complications (17, 18).

The present investigation addresses several critical gaps in current literature while introducing novel perspectives to pediatric DKA management. First, it represents one of the few randomized controlled trials specifically designed to evaluate non-inferiority of reduced-dose insulin therapy, employing rigorous statistical methodology to establish therapeutic equivalence rather than superiority. Second, the research was conducted in a low-income country setting, providing valuable insights into DKA management in healthcare systems that bridge the gap between resource-rich and resource-poor environments.

Our study’s innovative approach extends beyond simple dose comparison to encompass comprehensive evaluation of safety outcomes, resource utilization patterns, and practical implementation considerations. By demonstrating that lower insulin doses can achieve comparable clinical outcomes with improved safety profiles, this research challenges current standard of care and provides evidence-based support for protocol modifications that could benefit children worldwide, particularly those in settings where intensive monitoring capabilities are limited.

Methods

Study design and setting

We conducted this prospective, randomized, controlled, single-center trial at the Pediatric Intensive Care Unit of Suez Canal University Hospital, Ismailia, Egypt, between January 2018 and December 2018. The study protocol received approval from the Institutional Review Board of Suez Canal University (IRB Approval Number: 2938/2016, approved on October 26, 2016). Written informed consent was obtained from parents or legal guardians of all participants.

Participants and eligibility

Children aged 1–18 yr presenting with DKA were considered for enrollment. DKA diagnosis required the presence of blood glucose exceeding 200 mg/dL (11.1 mmol/L), venous pH below 7.3 or serum bicarbonate less than 15 mEq/L (15 mmol/L), and detectable ketones in blood or urine. Additional inclusion criteria encompassed presentation within 24 h of symptom onset and hemodynamic stability following initial fluid resuscitation.

We excluded children with severe DKA accompanied by altered consciousness (Glasgow Coma Scale below 12), concurrent acute illness such as pneumonia or urinary tract infection, previous study participation, known insulin hypersensitivity, pregnancy in post-pubertal females, or inability to obtain informed consent. DKA severity was classified according to 2018 ISPAD Clinical Practice Consensus Guidelines, with mild DKA defined as venous pH 7.20–7.29 or serum bicarbonate 10–14.9 mEq/L, moderate as pH 7.10–7.19 or bicarbonate 5–9.9 mEq/L, and severe as pH below 7.10 or bicarbonate less than 5 mEq/L (19).

Randomization and allocation

Eligible participants underwent randomization using a computer-generated random sequence with variable block sizes of 4, 6, and 8 to ensure allocation concealment. Randomization was stratified by DKA severity (mild/moderate vs. severe) and age group (below 5 yr vs. 5 yr and older) to ensure balanced distribution of these important prognostic factors. Due to the nature of the intervention, complete blinding was not feasible; however, laboratory personnel and outcome assessors remained blinded to treatment allocation throughout the study period.

Intervention protocols

All participants received standardized initial management according to ISPAD guidelines. Fluid management included initial resuscitation with 0.9% normal saline (10–20 mL/kg over 1–2 h) when clinically indicated, followed by maintenance fluid therapy with 0.9% normal saline for the first 6 h, then transition to 0.45% normal saline based on serum sodium levels and effective osmolality. Dextrose 5% was added when blood glucose decreased to 250 mg/dL (13.9 mmol/L) or below, with fluid rates calculated to replace estimated deficit over 48 h plus maintenance requirements.

Electrolyte management involved adding potassium chloride (40 mEq/L) to maintenance fluids after confirming adequate urine output and serum potassium below 5.5 mEq/L. Target serum potassium was maintained between 3.5–5.5 mEq/L, with phosphate supplementation provided when serum phosphate fell below 1.0 mg/dL (0.32 mmol/L).

The study intervention involved insulin therapy administration according to randomization assignment. The standard-dose group (n = 35) received continuous intravenous insulin infusion at 0.1 units/kg/h using regular human insulin, while the reduced-dose group (n = 35) received 0.05 units/kg/h using the same insulin preparation. Insulin infusion was initiated 1–2 h after beginning fluid therapy in both groups, with rate adjustments permitted based on clinical response and glucose trends.

Monitoring protocols

Comprehensive monitoring included capillary blood glucose measurements hourly during insulin infusion, venous blood glucose every 4 h or more frequently when clinically indicated, and arterial or venous blood gas analysis every 4 h until pH exceeded 7.3.

Serum electrolytes (sodium, potassium, chloride) were assessed every 4 h, with serum urea and creatinine measured every 8 h. Urine ketones were monitored every 4 h until negative. Estimated plasma osmolarity was calculated at each time point using the standard formula: 2 × [Na⁺] + [Glucose]/18 + [BUN]/2.8, and the rate of change was determined for each patient.

Clinical monitoring encompassed hourly vital signs, neurological assessment every 2 h using Glasgow Coma Scale, continuous monitoring of fluid balance with intake and output documentation, and continuous electrocardiographic monitoring for potassium-related changes.

Outcome measures

Primary outcomes included time to blood glucose normalization, defined as time from insulin initiation to first blood glucose measurement of 200 mg/dL (11.1 mmol/L) or below, and time to acidosis resolution, defined as time from insulin initiation to achievement of venous pH of 7.3 or higher or serum bicarbonate of 15 mEq/L (15 mmol/L) or higher.

Secondary outcomes encompassed rate of blood glucose decline, calculated as (initial glucose minus glucose at 200 mg/dL) divided by time to reach 200 mg/dL, hypoglycemia incidence defined as blood glucose of 60 mg/dL (3.3 mmol/L) or below during treatment, hypokalemia incidence defined as serum potassium below 3.5 mEq/L (3.5 mmol/L) or electrocardiographic changes suggestive of hypokalemia, treatment failure defined as failure to achieve blood glucose reduction of 18 mg/dL/h or greater for 2 consecutive hours or persistent acidosis without improvement, length of hospital stay from admission to discharge, and cerebral edema incidence based on clinical diagnosis using altered mental status, neurological deterioration, or radiological evidence.

Treatment failure management

Treatment failure was defined as failure to achieve blood glucose reduction of 18 mg/dL (1.0 mmol/L) per hour or greater for 2 consecutive hours, persistent or worsening acidosis (pH decrease or bicarbonate failure to increase) after 4 h of treatment, or development of significant hyperglycemia (above 400 mg/dL) after initial improvement. In cases of treatment failure, insulin infusion rate was increased by 0.02 units/kg/h after careful review of insulin preparation, infusion pump function, and intravenous line patency.

Transition to subcutaneous insulin

Transition criteria included resolution of acidosis (pH 7.3 or higher and bicarbonate 15 mEq/L or higher), blood glucose below 250 mg/dL (13.9 mmol/L), ability to tolerate oral intake, and absence of nausea or vomiting. Subcutaneous insulin was initiated with a 30–60-min overlap with intravenous insulin to prevent rebound ketosis.

Statistical analysis

Sample size calculation was based on a non-inferiority design with the primary endpoint of mean blood glucose reduction rate. Based on previous studies, the standard deviation of blood glucose reduction was estimated at 24 mg/dL/h. Using a non-inferiority margin of 18 mg/dL/h, with 85% power and α = 0.05, the calculated sample size was 30 participants per group. Accounting for a 15% attrition rate, 35 participants were enrolled in each group.

Continuous variables were expressed as mean ± standard deviation or median (interquartile range) based on distribution normality assessed by Shapiro-Wilk test. Categorical variables were presented as frequencies and percentages. Between-group comparisons used independent t-tests for normally distributed continuous variables, Mann- Whitney U tests for non-normally distributed variables, and chi-square or Fisher’s exact tests for categorical variables. Non-inferiority was assessed using one-sided confidence intervals, with time-to-event outcomes analyzed using Kaplan-Meier survival analysis with log-rank tests. All analyses were performed using intention-to-treat principles with statistical significance set at p < 0.05. Analyses were conducted using SPSS version 28.0 (IBM Corp., Armonk, NY).

Results

Participant characteristics and baseline demographics

A total of 85 children with DKA were assessed for eligibility during the study period. Following application of inclusion and exclusion criteria, 70 children were enrolled and randomized, with all participants completing the study protocol and no loss to follow-up (Fig. 1). Of the enrolled participants, 41 (58.6%) were newly diagnosed with type 1 diabetes, while 29 (41.4%) had established diabetes

Fig. 1.

CONSORT flow diagram of study participants.

The baseline demographic and clinical characteristics of the study population are presented in Table 1. The study population comprised 15 males (21.4%) and 55 females (78.6%), with a mean age of 8.65 ± 3.15 yr. DKA severity distribution included 17 participants (24.3%) with mild diabetic ketoacidosis, 33 participants (47.1%) with moderate diabetic ketoacidosis, and 20 participants (28.6%) with severe diabetic ketoacidosis. The distribution of diabetic ketoacidosis severity was comparable between groups, with no significant differences in presenting clinical or biochemical parameters. Mean glycated hemoglobin levels were elevated in both groups, reflecting suboptimal glycemic control prior to presentation. The two groups were well-matched with respect to baseline demographic, clinical and laboratory characteristics, with the exception of corrected sodium, which was significantly lower in the reduced-dose group (p = 0.001) (Table 1).

Table 1. Baseline clinical and biochemical characteristics.

Primary outcomes

Primary outcome results are summarized in Table 2 and illustrated in Figs. 2 and 3.

Table 2. Primary and secondary outcome measures.

Fig. 2.

Blood glucose response comparison between treatment groups. Panel A: Bar chart comparing mean blood glucose reduction rates between standard-dose (63.7 ± 35.6 mg/dL/h) and reduced-dose (45.2 ± 32.6 mg/dL/h) insulin therapy groups. Error bars represent standard deviation. The difference was statistically significant (p = 0.027), indicating more gradual glucose reduction with reduced-dose insulin therapy. Panel B: Line graph showing blood glucose trajectory over time for both treatment groups. The standard-dose group (blue line with circles) and reduced-dose group (purple line with squares) both target blood glucose of 200 mg/dL (red dashed line) with similar kinetics, though the standard-dose group showed initially more rapid decline. Both groups reached target glucose levels within 8-10 h of insulin initiation.

Fig. 3.

Time to acidosis resolution analysis. Kaplan-Meier survival curves showing the proportion of patients with persistent acidosis over time for both treatment groups. The reduced-dose group (purple line) achieved median acidosis resolution at 10.20 h, while the standard-dose group (blue line) reached median resolution at 11.22 h. Log-rank test revealed no significant difference between groups (p = 0.407), demonstrating non-inferiority of reduced-dose insulin therapy for acidosis correction.

Time to blood glucose normalization: The time required for blood glucose to decrease to 200 mg/dL or below showed no statistically significant difference between groups. The reduced-dose group achieved glucose normalization in 9.0 ± 3.43 h compared to 8.06 ± 3.26 h in the standard dose group (mean difference: 0.94 h, 95% CI: –0.52 to 2.40, p = 0.243). The one-sided 95% confidence interval for the difference (-∞, 2.15) fell within the pre-specified non-inferiority margin of 3 h, confirming non-inferiority of the reduced-dose regimen. Blood glucose trajectories over time for both treatment groups are shown in Fig. 2, demonstrating similar kinetics despite different initial rates of decline.

Time to acidosis resolution: Resolution of acidosis (pH ≥ 7.3 or bicarbonate ≥ 15 mEq/L) demonstrated similar kinetics between treatment groups. The reduced-dose group achieved acidosis resolution in 10.2 ± 5.29 h vs. 11.22 ± 5.04 h in the standard-dose group (mean difference: –1.02 h, 95% CI: –3.45 to 1.41, p = 0.407). Kaplan-Meier survival analysis revealed no significant difference in time-to-acidosis resolution (log-rank p = 0.407), with both groups showing comparable resolution curves as illustrated in Fig. 3.

Secondary outcomes

Secondary outcome measures are detailed in Table 2.

Rate of blood glucose and osmolality decline: The rate of blood glucose reduction differed significantly between groups, with the standard-dose group demonstrated more rapid initial decline. The mean rate of glucose reduction was 63.66 ± 35.61 mg/dL/h in the standard-dose group compared to 45.17 ± 32.62 mg/dL/h in the reduced-dose group (mean difference: 18.49 mg/dL/h, 95% CI: 2.15 to 34.83, p = 0.027). This difference is clearly illustrated in Fig. 2, Panel A, showing the more gradual glucose reduction achieved with reduced-dose insulin therapy. Furthermore, the rate of decline in estimated plasma osmolarity was also significantly more gradual in the reduced-dose group (3.2 ± 1.5 mOsm/kg/h vs. 4.8 ± 1.9 mOsm/kg/h, p = 0.038).

Safety outcomes

Safety outcomes are detailed in Table 3 and illustrated in Fig. 4.

Table 3. Safety outcomes and complications.

Fig. 4.

Safety outcomes and complications comparison. Panel A: Bar chart comparing the incidence of major complications between treatment groups. Hypokalemia occurred significantly less frequently in the reduced-dose group (20.0%) compared to the standard-dose group (42.9%, p = 0.039). Hypoglycemia showed a trend towards lower incidence in the reduced-dose group (11.4% vs. 25.7%), though this difference did not reach statistical significance. Treatment failure and rebound hyperglycemia rates were comparable between groups.

Hypoglycemia incidence: Hypoglycemia (blood glucose ≤ 60 mg/dL) occurred less frequently in the reduced-dose group, though the difference did not reach statistical significance. Four participants (11.4%) in the reduced-dose group experienced hypoglycemia compared to nine participants (25.7%) in the standard-dose group (relative risk: 0.44, 95% CI: 0.15 to 1.31, p = 0.124). All hypoglycemic episodes were successfully managed with dextrose administration without clinical sequelae.

Hypokalemia incidence: A statistically significant difference was observed in hypokalemia rates between groups, representing the most important safety finding of this study. Seven participants (20.0%) in the reduced-dose group developed hypokalemia compared to 15 participants (42.9%) in the standard-dose group (relative risk: 0.47, 95% CI: 0.22 to 0.98, p = 0.039). This finding represents a 53% relative risk reduction in hypokalemia with reduced-dose insulin therapy, as illustrated in Fig. 4.

Treatment failure: No participants in either group met the criteria for treatment failure, indicating that both insulin dosing regimens were effective in achieving metabolic correction within the expected timeframe.

Clinical outcomes and healthcare utilization

Healthcare utilization outcomes are presented in Table 2.

Length of hospital stay: No significant difference was observed in hospital length of stay between groups (3.2 ± 1.1 vs. 3.1 ± 1.0 d, p = 0.689). This finding suggests that the more gradual metabolic correction achieved with reduced-dose insulin did not prolong hospitalization or delay clinical recovery.

Neurological outcomes: Glasgow Coma Scale scores were identical between groups at presentation (14.11 ± 1.13 vs. 14.11 ± 1.05, p = 1.000) and throughout treatment. No participants developed clinical evidence of cerebral edema in either group, though the study was not powered to detect differences in this rare but serious complication.

Insulin consumption: Total insulin consumption was significantly lower in the reduced-dose group, with participants receiving approximately 50% less insulin over the treatment period (12.4 ± 4.1 vs. 24.8 ± 8.2 units, p < 0.001). This reduction in insulin utilization has important implications for cost-effectiveness and resource allocation, particularly in settings with limited insulin availability.

Non-Inferiority analysis: The pre-specified non-inferiority analysis confirmed that reduced-dose insulin therapy met the criteria for non-inferiority compared to standard-dose therapy for both primary endpoints. The one-sided 95% confidence intervals for the differences in time to glucose normalization and acidosis resolution both fell within the pre-defined non-inferiority margins, providing robust evidence that reduced-dose insulin is not clinically inferior to standard-dose therapy. Furthermore, the observed safety advantages of reduced-dose insulin, particularly the significant reduction in hypokalemia risk shown in Table 3 and Fig. 4, suggest that reduced-dose therapy may actually be superior to standard-dose therapy when considering the overall risk-benefit profile.

Discussion

This randomized controlled trial provides compelling evidence that reduced-dose insulin therapy (0.05 units/kg/h) demonstrates non-inferiority to conventional dosing (0.1 units/kg/h) for pediatric DKA management while offering significant safety advantages. These findings have important implications for global pediatric diabetes care, particularly in resource-constrained settings where optimal monitoring capabilities may be limited and cost-effective interventions are essential for sustainable healthcare delivery.

Clinical efficacy and non-inferiority

Our investigation demonstrates that halving the traditional insulin dose does not compromise clinical efficacy in terms of metabolic correction. Both treatment groups achieved comparable times to blood glucose normalization and acidosis resolution, confirming that the therapeutic goals of DKA management can be accomplished with lower insulin doses. This finding aligns with recent evidence from multiple international centers and challenges the historical assumption that higher insulin doses are necessary for optimal outcomes (20,21,22). The non-inferiority of reduced-dose insulin therapy is particularly noteworthy given theoretical concerns about delayed metabolic correction. Our results indicate that while the rate of glucose decline was more gradual with reduced-dose insulin, this did not translate to clinically meaningful delays in overall recovery. This observation supports the emerging paradigm that prioritizes ketosis suppression over rapid glucose normalization, recognizing that excessive glucose reduction rates may be counterproductive (23). Recent studies by Rameshkumar and colleagues and El Hawary and colleagues have reported similar findings, with reduced-dose insulin achieving comparable efficacy outcomes while demonstrating superior safety profiles (21, 24), strengthening the evidence base for reduced-dose insulin therapy across diverse clinical contexts.

Reduction in hypokalemia: a critical safety advantage

The most striking finding of our study was the significant reduction in hypokalemia incidence with reduced-dose insulin therapy (20.0% vs. 42.9%, p = 0.039). This 53% relative risk reduction represents a clinically meaningful improvement in safety that could have substantial implications for patient outcomes and healthcare resource utilization. Hypokalemia during DKA treatment is associated with cardiac arrhythmias, muscle weakness, and prolonged recovery times, making its prevention a critical therapeutic goal (25). The mechanism underlying this safety benefit likely relates to the more gradual shift of potassium from extracellular to intracellular compartments with lower insulin doses. Insulin promotes cellular potassium uptake through activation of sodium-potassium-ATPase pumps, and higher insulin concentrations can precipitate rapid and profound hypokalemia (26). By moderating this effect, reduced-dose insulin therapy provides a more physiologic approach to potassium homeostasis during DKA treatment.

Reduction in hypoglycemia and cerebral edema risk

Our reduced-dose protocol demonstrated a clinically important trend toward improved safety by reducing the incidence of hypoglycemia. The rate of hypoglycemic events (blood glucose ≤ 60 mg/dL) was less than half in the reduced-dose group (11.4%) compared to the standard-dose group (25.7%; RR: 0.44, 95% CI: 0.15 to 1.31). While this 56% relative risk reduction did not achieve statistical significance in our cohort (p = 0.124), this finding is consistent with the broader evidence from larger trials and meta-analyses, which have shown a definitive benefit (27). Given the established links between glycemic variability and adverse neurocognitive outcomes in children, this reduction represents a key safety advantage of the reduced-dose strategy (28).

Although our study did not observe any cases of cerebral edema, our findings provide a potential mechanistic explanation for the protective effect of a reduced-dose insulin strategy. The more controlled glycemic decline in the reduced-dose group (45.17 vs. 63.66 mg/dL/h) translated directly into a significantly attenuated rate of decline in estimated plasma osmolarity (3.2 vs. 4.8 mOsm/kg/h, p = 0.038). This attenuated osmotic shift is critical, as the pathophysiology of cerebral edema is thought to be driven by rapid decreases in plasma osmolality that create a fluid gradient into the central nervous system (29). By moderating the rate of osmolar change, the reduced-dose protocol provides a more physiologic correction that may mitigate the risk of such fluid shifts. This concept is now being incorporated into clinical practice guidelines, with leading institutions recommending lower starting insulin doses to minimize rapid osmolality changes, particularly in high-risk patients, including those undergoing inter- hospital transfers where monitoring capabilities may be limited (16,17,18).

Resource conservation and global health implications

The findings of this study have particular relevance for global health initiatives aimed at improving pediatric diabetes care in resource-limited settings. A recent analysis revealed that many healthcare facilities in low- and middle-income countries lack the capacity for intensive DKA monitoring, including hourly glucose measurements and frequent electrolyte assessments (14). In such settings, treatment approaches that reduce monitoring requirements while maintaining clinical efficacy are essential for feasible care delivery. Reduced-dose insulin therapy offers several advantages in resource-constrained environments. First, the reduced hypoglycemia and hypokalemia rates decrease the need for frequent laboratory monitoring and emergency interventions. Second, the lower insulin consumption reduces medication costs and addresses concerns about insulin availability in settings where supply chains may be unreliable. Third, the more gradual metabolic correction may be safer in environments where intensive care capabilities are limited. The economic implications extend beyond direct medication costs. Hypokalemia and hypoglycemia episodes require additional laboratory tests, prolonged monitoring, and interventions that increase healthcare utilization and costs. By reducing these complications, reduced-dose insulin therapy may offer substantial cost savings that could improve the sustainability of DKA care programs in resource-limited settings (30).

Guideline evolution and clinical implications

Our findings contribute to a growing body of evidence that is actively reshaping clinical practice, validating a paradigm shift toward lower insulin dosing in pediatric DKA. This evolution is already reflected in the changing landscape of clinical guidelines; for instance, The Royal Children’s Hospital Melbourne and the BC Children’s Hospital DKA Toolkit now recommend considering a 0.05 units/kg/h starting rate, particularly for inter-hospital transfers or other high-risk scenarios (17, 18). These guideline changes, representing a significant shift from traditional approaches, support the direct clinical implication of our study: clinicians should consider implementing protocols that start with 0.05 units/kg/hour insulin infusions. This approach is particularly attractive in settings with limited monitoring capabilities or for patients at higher risk for complications, as it offers comparable efficacy with a superior safety profile.

Limitations and future directions

Several limitations of our study warrant consideration. First, the single-center design may limit the generalizability of our findings, although the consistency of our results with other multi-center studies suggests broader applicability. Second, the relatively small sample size limited our statistical power to detect a significant difference in the incidence of hypoglycemia, even though a clear trend was observed. Crucially, our study excluded patients with severe DKA and those with significant concurrent infections. Therefore, our findings may not be directly applicable to these high-risk populations. Finally, the baseline difference in corrected sodium between the two groups is a potential confounder, though its clinical significance remains unclear.

These limitations highlight the need for future research. A large-scale, multi-center trial is required to confirm our findings in a broader population and provide definitive evidence on hypoglycemia rates. Crucially, prospective studies must evaluate the safety and efficacy of this low-dose strategy in high-risk populations excluded from our study, such as children with severe DKA or concurrent infections. Finally, formal economic analyses are needed to quantify the cost-effectiveness of the reduced-dose protocol, thereby providing data-driven support for its adoption in resource-limited settings.

Conclusions

This randomized controlled trial provides robust evidence that reduced-dose insulin therapy (0.05 units/kg/h) represents a paradigm shift toward safer, more cost-effective management of pediatric DKA. The demonstration of non-inferiority in clinical efficacy, combined with significant improvements in safety outcomes, challenges the traditional reliance on higher insulin doses and supports the adoption of more conservative dosing strategies.

The key findings of this study include: (1) non-inferiority of reduced-dose insulin therapy for achieving metabolic correction in pediatric DKA; (2) significant reduction in hypokalemia risk with reduced-dose therapy; (3) trends toward reduced hypoglycemia incidence; and (4) comparable clinical outcomes with reduced insulin consumption and monitoring requirements. These results have particular relevance for global health initiatives aimed at improving pediatric diabetes care in resource-constrained settings. The reduced monitoring requirements, lower complication rates, and decreased insulin consumption associated with reduced-dose therapy make it an attractive option for healthcare systems with limited resources. Furthermore, the potential for reduced cerebral edema risk through more gradual osmolality correction adds an important safety dimension to this approach.

Conflict of interests

The authors declare no conflicts of interest related to this research. No funding was received for this study.