Summary
Objectives:
Diabetic ketoacidosis (DKA) seasonal variations among children and adolescents with type 1 diabetes mellitus (T1DM) have been widely studied; however, findings remain inconsistent. Although several studies reported a higher incidence during colder months, others showed no consistent patterns. This study investigated whether DKA frequency and severity vary across seasons, particularly in humid cities.
Methods:
This retrospective cohort analyzed children and adolescents with T1DM admitted to a tertiary hospital with DKA (2015–2025). Data included demographics, seasonal admission trends, clinical severity, and outcomes. Statistical analyses involved the Chi-square and ANOVA tests.
Results:
A total of 369 DKA cases were analyzed, with mean age of 14.7 years and 62.6% were females. Most had established T1DM (88.6% with prior DKA episodes). Severe cases showed significantly higher heart rates, lower GCS scores (p < 0.001), higher HbA1c levels (p = 0.001), longer hospital (p = 0.007) and ICU (p = 0.003) stays. Diastolic blood pressure also correlated with severity (p = 0.036). Although DKA severity showed no significance seasonal association (p = 0.865), winter peak was observed, suggesting confounding factors or bias in seasonal patterns.
Conclusions:
The DKA admission peaked in winter, particularly in December; however, this trend was not clinically significant. Confounding factors, including infection or delayed care, may explain this variation. The findings emphasize focusing on modifiable patient factors, including year-round glycemic control, education, and timely interventions, to reduce DKA recurrence.
Keywords: Diabetic ketoacidosis, Type 1 diabetes mellitus, Seasonal variation, Pediatric endocrinology, Glycemic control
Introduction
Despite advances in insulin therapy and glucose monitoring technologies, type 1 diabetes mellitus (T1DM) remains an expanding global health concern, with an increasing incidence particularly among children and young adults [1]. This disease involves a complex interplay of genetic susceptibility, immune dysregulation, and environmental triggers, such as viral infections [2]. Among the acute metabolic complications of diabetes, diabetic ketoacidosis (DKA) is the most severe and life-threatening. DKA arises from absolute or relative insulin deficiency, causing unopposed lipolysis, ketone body accumulation, hyperglycemia, dehydration, and metabolic acidosis [3]. Individuals presenting with DKA are more susceptible to experience poor glycemic control and recurrent hospitalization. Studies confirm an association between DKA at diagnosis and adverse metabolic trajectories [4,5]. Although the immediate biochemical crisis can be reversed with insulin and fluid therapy, DKA carries a significant risk of cerebral edema, arrhythmia, and death, particularly in pediatric patients. Therefore, DKA remains a major cause of morbidity and mortality, particularly in resource-limited settings, where mortality can reach up to 2.5% [6].
The worldwide increase in DKA diagnoses has caught attention to environmental factors, including infection exposure and climate-related influences, which may shape its temporal and geographic patterns [7,8]. Several coexisting medical conditions, including cardiovascular diseases, dehydration, and poor adherence to therapy, frequently contribute to DKA crisis [9,10]. The frequency of DKA as a presenting feature of new-onset T1DM varies widely among countries, ranging from as low as 15% in regions with strong healthcare networks to nearly 80% in low-resource environments [11]. These differences reflect healthcare system accessibility, socioeconomic status, and possibly seasonal conditions that influence infection rates and patient behavior.
Seasonality is an established feature of many medical conditions, reflecting the influence of environmental, behavioral, and infectious factors that fluctuate throughout the year [12]. Diseases, such as influenza, respiratory syncytial virus (RSV) infections, and asthma, show well-defined winter peaks linked to viral transmission and climate-driven physiological responses [12,13,14]. Similarly, evidence is accumulating regarding seasonal variations in DKA incidence, particularly among children with T1DM [12,14]. Multiple international studies have reported higher DKA admissions during colder months, suggesting a correlation between temperature, infection prevalence, and changes in lifestyle behaviors [15,16,17,18,19]. Infections, such as influenza and other respiratory viruses, surge in winter, increasing insulin resistance and predisposing individuals with diabetes to ketoacidosis. Furthermore, decreased physical activity, altered diet, and potential delays in seeking medical care during colder months may exacerbate this risk [20]. Research from Northern Europe and Scandinavia provides additional context. Turtinen et al. [21] examined 4,993 Finnish children and identified a pronounced peak in DKA diagnosis during fall and winter. This suggested that environmental triggers, such as viral infections, may not only precipitate disease onset, but also influence its acute complications.
Climatic variations may modify these seasonal effects. Countries in arid or semi-arid regions, such as Saudi Arabia, have distinct climate profiles with prolonged hot summers and mild winters, potentially influencing both glycemic control and infection patterns. Moreover, religious and cultural practices, such as fasting during Ramadan, introduce additional variables that may affect DKA risk through changes in diet, hydration, and medication adherence [22,23]. Although several studies from Saudi Arabia and other Gulf countries have examined DKA in children with T1DM, most have either not assessed seasonal or temperature-related variations or have reported contradictory findings. This highlights a persistent gap in understanding regional patterns and climatic influences on DKA occurrence [18,24].
In Saudi Arabia, epidemiological data on DKA seasonality in children and young adults with T1DM are limited. The nation's long hot summers, mild winters, and regional climatic differences, combined with cultural factors, such as Ramadan and Hajj, may influence dehydration, infection patterns, and insulin adherence. This study aimed to determine whether seasonal variation exists in DKA incidence among children and young patients with T1DM in humid, warm climate of Jeddah City, Saudi Arabia. A further aim was to compare clinical and demographic patterns to guide preventive and season-specific interventions.
Method
This retrospective, cohort study was conducted at a single tertiary hospital. It included all children and young adults (n = 369) admitted to the pediatric emergency department between January 2015 and December 2025. Ethical approval was obtained from the Institutional Review Board's Biomedical Ethical Committee in December 2024 (Reference no. 534-25), and the study adhered to the principles of the Declaration of Helsinki. Due to the retrospective design and lack of direct patient contact, parental consent was waived.
Eligibility required participants to under 18 years of age with a diagnosis of T1DM and admission for DKA. We excluded patients with type 2 or secondary diabetes, those older than 18 years of age, those with incomplete medical records, and those with missing admissions or diagnostic details. Data for all included patients were retrieved from hospital records. Collected variables comprised: demographic data (age, sex, nationality), clinical and biochemical characteristics by season (DKA severity based on pH/HCO3, HbA1c, serum glucose, electrolytes, and pH), and hospital outcomes (length of stay, Intensive Care Unit (ICU) admission, and complications). All data were systematically entered and organized for statistical analyses.
Statistical analysis
Descriptive statistics summarized the data. Continuous variables were reported as means and standard deviations, and categorical variables as frequencies with percentages. We explored associations between the season of admission, DKA severity, and various clinical and demographic factors. For group comparisons, we conducted the Yates’ corrected chi-square test for categorical variables. For continuous variables, we used analysis of variance (ANOVA) when data were normally distributed and the Kruskal-Wallis H test when statistical assumptions were violated. Analyzed outcomes included age, vital signs, metabolic markers, and clinical measures. All analyses were performed using the Statistical Package for the Social Sciences version 28.0, with significance set at p < 0.05.
Results
This study included 369 children and adolescents with DKA. The patients had a mean age of 14.7 ± 4.4 years, with a nearly equal gender distribution (62.6% female, 37.4% male), and a mix of Saudi (47.7%) and non-Saudi (52.3%) nationalities (Table 1). Data spanned the full calendar year (January–December), revealing differences in admission frequencies between males and females (Fig. 1). Throughout the year, female patients consistently had higher admission rates than male patients. The peak admission month for females was December (25 cases), while the lowest was April (12 cases). For males, the highest number of admissions occurred in May (20 cases) and the lowest in August and October (6 cases) (Fig. 1). Overall, a clear sex disparity was observed in DKA admissions, with females showing a higher incidence in most months. Notably, both sexes experienced high admissions in the winter, particularly in December, suggesting potential seasonal influences on DKA episodes.
Table 1.
Baseline demographic and diabetes history of children and adolescents presenting with diabetic ketoacidosis (N = 369).
| Characteristic | Overall (N = 369) |
|---|---|
| Demographics | |
| Age, years | |
| Mean (SD) | 14.7 (4.4) |
| Range | 2.0 – < 18 |
| Gender | |
| Female | 231 (62.6%) |
| Male | 138 (37.4%) |
| Nationality | |
| Non-Saudi | 193 (52.3%) |
| Saudi | 176 (47.7%) |
| Diabetes history | |
| New diagnosis of T1DM | |
| No | 267 (72.4%) |
| Yes | 102 (27.6%) |
| Age at onset of T1DM, years (n = 358) | |
| Mean (SD) | 8.4 (4.2) |
| Range | 0.0 – 20.0 |
| Duration of T1DM, years (n = 356) | |
| Mean (SD) | 3.5 (2.9) |
| Range | 0.0 – 16.0 |
| Previous DKA episode | |
| No | 42 (11.4%) |
| Yes | 327 (88.6%) |
| Previous non-DM ICU admission | |
| No | 326 (88.3%) |
| Yes | 43 (11.7%) |
| Anthropometrics | |
| Weight, kg (n=363) | |
| Mean (SD) | 33.1 (17.2) |
| Range | 5.8 – 124.0 |
| Height, cm (n = 352) | |
| Mean (SD) | 130.8 (25.3) |
| Range | 24.0 – 175.0 |
| BMI, kg/m2 (n = 339) | |
| Mean (SD) | 18.6 (9.5) |
| Range | 6.8 – 160.0 |
BMI: Body mass index, DKA: Diabetic ketoacidosis, T1DM: Type 1 diabetes, SD: Standard deviation, ICU: Intensive care unit.
Fig. 1.
Monthly distribution of diabetic ketoacidosis admissions by gender in children and adolescents with T1DM. TIDM: Type 1 Diabetes meletus.
The majority of patients (72.4%) had a pre-existing diagnosis of T1DM, with a mean age at onset of 8.4 years, and a mean disease duration of 3.5 years. Notably, 88.6% had a history of at least one prior episode of DKA. DKA episodes were distributed relatively evenly across the seasons, with a slight predominance in winter (27.6%) and summer (26.8%) (Table 2). Upon admission, most patients were managed in the general ward (79.1%), while 19.5% required ICU admission. Among cases with documented severity, episodes were almost evenly classified as: mild (36.6%) moderate (36.3%), and severe (27%) (Table 2). Clinically, patients presented with tachycardia (mean heart rate, 118.6 bpm) and elevated HbA1c levels (mean, 12.1%), indicating poor glycemic control preceding the event. The mean hospital length of stay was 4.5 days. For the subset admitted to the ICU, the mean stay was 1.6 days.
Table 2.
Clinical presentation and outcomes of diabetic ketoacidosis episodes.
| Characteristic | Overall (n = 369) |
|---|---|
| Admission Details | |
| Season of presentation | |
| Winter | 102 (27.6%) |
| Summer | 99 (26.8%) |
| Spring | 85 (23.0%) |
| Autumn | 83 (22.5%) |
| Admission unit | |
| Ward | 292 (79.1%) |
| Intensive care unit (ICU) | 72 (19.5%) |
| Emergency | 5 (1.4%) |
| Clinical presentation | |
| DKA severity (n = 344) | |
| Mild | 126 (36.6%) |
| Moderate | 125 (36.3%) |
| Severe | 93 (27.0%) |
| Heart rate, bpm (n = 352) | |
| Mean (SD) | 118.6 (21.7) |
| Range | 70.0 – 184.0 |
| Systolic BP, mmHg (n=344) | |
| Mean (SD) | 114.7 (13.8) |
| Range | 80.0 – 163.0 |
| Diastolic BP, mmHg (n = 342) | |
| Mean (SD) | 69.4 (10.5) |
| Range | 37.0 – 104.0 |
| Temperature, °C (n = 363) | |
| Mean (SD) | 36.9 (0.6) |
| Range | 35.5 – 39.9 |
| Glasgow coma scale (n = 195) | |
| Mean (SD) | 14.4 (1.6) |
| Range | 3.0 – 23.0 |
| Laboratory | |
| Outcomes | |
| HbA1c at admission, % (n = 326) | |
| Mean (SD) | 12.1 (2.4) |
| Range | 6.6 – 19.4 |
| Length of hospital stay, days (n = 363) | |
| Mean (SD) | 4.5 (3.1) |
| Range | 0.0 – 22.0 |
| Length of ICU stay, days (n = 102) | |
| Mean (SD) | 1.6 (1.5) |
| Range | 0.0 – 8.0 |
BP: Blood pressure, DM: Diabetes mellitus, ICU: Intensive care unit, SD: Standard deviation, DKA: Diabetic Ketoacidosis, SD: Standard deviation.
Several significant differences emerged based on metabolic derangement severity. Patients with severe DKA exhibited markedly worse clinical indicators at presentation (Table 3). They presented with significantly higher heart rates (127.7 ± 20.8 bpm) compared to those with mild (115.5 ± 21.8 bpm) or moderate (116.1 ± 20.9 bpm) DKA (p < 0.001), reflecting greater volume depletion and acidosis. Neurological assessment also differed significantly; the severe DKA group had the lowest mean Glasgow Coma Scale (GCS) score (13.6 ± 2.4), indicating an altered mental status that worsened with increasing severity (p < 0.001) (Table 3). Metabolic control prior to admission was also worst in the severe group, which had the highest mean HbA1c level (12.9 ± 2.4%) than the mild and moderate groups (p = 0.001). The consequences of greater severity were reflected in significantly worse clinical outcomes. Patients with severe DKA experienced a longer total hospital stay (5.2 ± 3.2 days) and a substantially longer ICU stay (2.0 ± 1.4 days) than the other groups (p = 0.007 and p = 0.003, respectively). Among the vital signs assessed, diastolic blood pressure demonstrated a modest yet statistically significant correlation with DKA severity (p = 0.036). In contrast, no significant associations were found between DKA severity and other baseline characteristics (Table 3).
Table 3.
Association between clinical characteristics and severity of diabetic ketoacidosis.
| Severity | |||||
|---|---|---|---|---|---|
|
|
|||||
| Characteristics | Mild | Moderate | Severe | Total | P-value |
| Total n (%) | 126 (36.6) | 125 (36.3) | 93 (27.0) | 344 | |
| Age | |||||
| Mean (SD) | 15.5 (4.0) | 14.5 (4.5) | 14.5 (4.6) | 14.8 (4.4) | 0.129 |
| Gender | |||||
| Female | 78 (61.9) | 77 (61.6) | 65 (69.9) | 220 (64.0) | 0.377 |
| Male | 48 (38.1) | 48 (38.4) | 28 (30.1) | 124 (36.0) | |
| BP Systolic | |||||
| Mean (SD) | 112.8 (13.0) | 115.3 (13.6) | 116.7 (15.7) | 114.8 (14.0) | 0.125 |
| BP Diastolic | |||||
| Mean (SD) | 68.9 (10.5) | 68.2 (10.1) | 71.9 (11.1) | 69.5 (10.6) | 0.036 |
| Length of hospital stay days | |||||
| Mean (SD) | 4.2 (3.2) | 4.0 (2.3) | 5.2 (3.2) | 4.4 (3.0) | 0.007 |
| Length of stay in intensive care Days | |||||
| Mean (SD) | 0.7 (1.5) | 1.2 (1.7) | 2.0 (1.4) | 1.6 (1.6) | 0.003 |
| HR at presentation | |||||
| Mean (SD) | 115.5 (21.8) | 116.1 (20.9) | 127.7 (20.8) | 119.1 (21.8) | 0.003 |
| Temperature at presentation | |||||
| Mean (SD) | 36.7 (0.5) | 36.9 (0.5) | 37.0 (0.6) | 37.8 (18.0) | 0.001 |
| GCS | |||||
| Mean (SD) | 14.9 (0.4) | 14.8 (0.7) | 13.6 (2.4) | 14.4 (1.6) | 0.002 |
| A1C at admission | |||||
| Mean (SD) | 11.6 (2.5) | 12.0 (2.1) | 12.9 (2.4) | 12.1 (2.3) | 0.001 |
| BMI | |||||
| Mean (SD) | 17.8 (5.1) | 18.3 (5.9) | 19.9 (16.4) | 18.6 (9.7) | 0.299 |
| Season | |||||
| Autum | 28 (22.2) | 28 (22.4) | 21 (22.6) | 77 (22.4) | 0.865 |
| Spring | 25 (19.8) | 32 (25.6) | 20 (21.5) | 77 (22.4) | |
| Summer | 40 (31.7) | 31 (24.8) | 24 (25.8) | 95 (27.6) | |
| Winter | 33 (26.2) | 34 (27.2) | 28 (30.1) | 95 (27.6) | |
BP: Blood pressure, HR: Heart rate, GCS: Glascow Coma Scale, BMI: Body metabolic index.
The seasonal variation among children and adolescents with T1DM experiencing severe DKA and clinical disturbance
The analysis of variance indicated that DKA severity did not vary significantly across seasons for key indicators of illness severity and metabolic disturbance (p > 0.05, Table 4). The distribution of DKA severity was consistent across all four seasons (p = 0.865). Similarly, no significant differences in vital signs at presentation, including systolic blood pressure (p = 0.910), diastolic blood pressure (p = 0.460), heart rate (p = 0.462), and temperature (p = 0.528). However, the mean temperature in winter (40.1°C) was notably higher, potentially due to a higher prevalence of febrile illness (Table 4). Key metabolic parameters showed no seasonal variation, with no significant differences in HbA1c levels (p = 0.303) or BMI (p = 0.149) at admission. Furthermore, patient outcomes were consistent throughout the year. The length of the total hospital stays (p = 0.062) and length of stay in the ICU (p = 0.730) did not differ significantly by season.
Table 4.
Comparison of diabetic ketoacidosis severity and clinical outcomes by season of admission.
| Season | ||||||
|---|---|---|---|---|---|---|
|
|
||||||
| Variables | Autumn | Spring | Summer | Winter | Total | P-value |
| Total n (%) | 83 (22.5) | 85 (23.0) | 99 (26.8) | 102 (27.6) | 369 | |
| Severity | ||||||
| Mild | 28 (36.4) | 25 (32.5) | 40 (42.1) | 33 (34.7) | 126 (36.6) | 0.865 |
| Moderate | 28 (36.4) | 32 (41.6) | 31 (32.6) | 34 (35.8) | 125 (36.3) | |
| Severe | 21 (27.3) | 20 (26.0) | 24 (25.3) | 28 (29.5) | 93 (27.0) | |
| BP systolic | ||||||
| Mean (SD) | 115.5 (12.2) | 114.7 (14.2) | 114.8 (14.4) | 114.0 (14.4) | 114.7 (13.8) | 0.910 |
| BP diastolic | ||||||
| Mean (SD) | 71.0 (10.5) | 69.2 (10.1) | 69.2 (11.2) | 68.5 (10.2) | 69.4 (10.5) | 0.460 |
| Length of hospital stay Days | ||||||
| Mean (SD) | 4.5 (3.6) | 4.2 (3.0) | 4.0 (2.4) | 5.1 (3.4) | 4.5 (3.1) | 0.062 |
| Length of stay in intensive care (days) | ||||||
| Mean (SD) | 1.3 (1.3) | 1.6 (1.7) | 1.5 (1.6) | 1.8 (1.6) | 1.6 (1.5) | 0.730 |
| HR at presentation | ||||||
| Mean (SD) | 119.4 (22.2) | 115.2 (21.5) | 120.1 (22.8) | 119.4 (20.4) | 118.6 (21.7) | 0.462 |
| Temperature at presentation | ||||||
| Mean (SD) | 36.8 (0.5) | 36.8 (0.5) | 36.9 (0.6) | 36.8 (0.6) | 36.9 (0.6) | 0.528 |
| GCS | ||||||
| Mean (SD) | 14.6 (1.0) | 14.7 (1.9) | 14.1 (2.0) | 14.4 (1.3) | 14.4 (1.6) | 0.414 |
| A1C at admission | ||||||
| Mean (SD) | 12.1 (2.4) | 12.0 (2.3) | 12.4 (2.4) | 11.8 (2.4) | 12.1 (2.4) | 0.303 |
| BMI | ||||||
| Mean (SD) | 17.7 (4.8) | 20.7 (17.0) | 18.4 (4.7) | 17.7 (6.5) | 18.6 (9.5) | 0.149 |
HR: Heart rate, BP: Blood pressure, BMI: Body metabolic Index.
Seasonal variation in the mean length of hospital and ICU stay among children and adolescents with T1DM who experience DKA
For patients with no prior DKA episodes, the mean hospital length of stay was relatively consistent across seasons, ranging from approximately 7 days in winter to 4 days in spring (Fig. 2A). Conversely, patients with prior DKA episodes exhibited shorter hospital stays of approximately 3–4 days, with minimal seasonal variation. Similar to hospital stay, those with no previous DKA episodes showed a consistent pattern in the mean length of stay in ICU, averaging 0.5 to 2 days, with autumn having the highest duration (Fig. 2B). For patients with a history of DKA, ICU stays were slightly shorter, typically less than 1.5 days, and peaked in summer.
Fig. 2.
Seasonal variation in hospital and ICU Stays for diabetic ketoacidosis in children and adolescents with T1DM. DKA: Diabetic ketoacidosis.
Discussion
Seasonal variations in DKA among children with T1DM may reflect environmental, infectious, or behavioral triggers that affect glycemic control and access to care. However, the influence of seasonal patterns on disease severity and outcomes remains an area of uncertain evidence. This study investigated seasonal variations in DKA among children and adolescents with pre-existing T1DM and explored whether DKA severity correlates with specific times of the year.
The cohort predominantly included patients with established T1DM, most of whom had experienced at least one previous DKA episode. Although our data showed an overall trend of increased DKA admissions during winter, particularly in December, this seasonal variation was not statistically significant. Notably, the December peak was primarily driven by female patients, who accounted for 25 admissions that month, while male admissions were relatively low (n = 9). Across the entire year, females consistently had higher DKA admission rates than males in nearly all months, which influenced the observed seasonal pattern. This sex disparity may reflect underlying hormonal, behavioral, or psychosocial factors influencing disease management and glycemic control in adolescent females, rather than a uniform seasonal effect across both sexes.
Although patients with severe DKA had significantly higher clinical, metabolic, and mental alterations and longer ICU and hospital stays, we found no statistically significant association between seasonal variation and DKA severity (Table 3,Table 4). While seasonality did not significantly correlate with clinical outcomes, visual inspection showed longer hospital and ICU stays in autumn and winter (Fig. 2). Confounding factors, such as a higher incidence of respiratory infections, decreased physical activity, or delayed access to care during holidays, may contribute to a prolonged hospitalization. Our findings contrast those of Saadeh et al. [25] who identified a significant peak in DKA admissions during winter and spring among children in northern Jordan. In their study, infections, particularly sepsis, were the leading precipitating factors, supporting the hypothesis that seasonal illnesses contribute to metabolic decompensation in vulnerable children. Similarly, Lee et al. [26] reported clear seasonal variations in Seoul, Korea, with a peak in DKA cases in winter, particularly in December, and no reported cases in summer. This pattern strongly suggested that cold weather and seasonal infections precipitate DKA episodes in pediatric East Asian population. In Jeddah City, pronounced climatic variations may be absent due to minimal temperature difference between winter and summer. However, other factors, such as the Hajj or Umrah seasons, may increase infection rates without significantly affecting the seasonal pattern of DKA incidence. Nevertheless, our findings align with global studies that have not identified a strong seasonal impact on DKA severity. This discrepancy may also be influenced by geographical, climatic, and healthcare system differences. Additionally, in areas with well-established outpatient diabetes education and timely emergency care, seasonal triggers may have less influence on acute DKA presentation. Our study showed a relatively high rate of recurrent DKA, indicating that chronic disease mismanagement, rather than acute seasonal stressors, may have been the dominant driver of hospital admissions in this cohort. Additionally, Dong et al. [27] provided evidence that individual metabolic risk factors, such as elevated HbA1c levels, younger age of onset, and weight loss, are stronger predictors of ketosis than seasonal trends. These findings support the notion that intrinsic disease factors and patient-specific risk profiles outweigh environmental triggers in predicting DKA.
While our data showed minor seasonal fluctuations in DKA admissions and hospital stay duration, we did not identify a statistically significant association between season and DKA severity. This suggests that although winter may slightly increase the risk due to environmental confounding factors, the primary drivers of DKA in children with T1DM potentially lie in modifiable, patient-centered domains, such as glycemic control, adherence, and timely care-seeking behavior. Therefore, future studies should focus on continuous patient education, early symptom recognition, and consistent follow-ups to reduce recurrence and improve outcomes across all seasons.
Study limitation
This study has several limitations. Its retrospective, single-center design may limit generalizability and introduce selection and information bias. Reliance on medical records may have resulted in incomplete or inaccurately documented data, despite exclusion of missing cases. The study included only patients presenting with DKA, potentially overlooking milder or unreported cases managed elsewhere. Seasonal classification did not account for environmental variables such as humidity, infections, or socioeconomic factors that may influence DKA occurrence. Additionally, causal relationships cannot be established due to the observational design. Finally, unmeasured confounders, including adherence to treatment and access to care, may have influenced the findings.
Implication for future research
Future research should prioritize prospective, multicenter studies to clarify the role of environmental, infectious, and behavioral factors in DKA occurrence. Investigations should incorporate climate data, infection rates, and healthcare access variables to better understand seasonal influences. Additionally, studies focusing on sex-specific differences and patient-centered factors, such as adherence and glycemic control, are warranted. Interventional research evaluating education and early detection strategies across all seasons may further reduce recurrence and improve outcomes.
In conclusion, our study identified a higher frequency of DKA admissions during winter, particularly in December; however, this trend did not reach marked clinical significance. The absence of a strong seasonal association suggests that environmental factors alone are not primary drivers of DKA severity. Confounding influences, such as increased infections, reduced physical activity, and delayed access to care during holiday periods, could account for variations in hospital and ICU stays. These findings emphasize the importance of prioritizing modifiable, patient-specific factors, such as glycemic control, adherence, and timely intervention. Strengthening education and follow-up strategies remains essential to reduce DKA recurrence throughout the year.
Disclosure statement
No AI-assisted technology was used in the writing or analysis of this research project. The manuscript or essence of its contents was not previously published or submitted as pre-print in partial or full in the website or printed journal in other language than English. Informed consent is not required in this study owing to the retrospective design and lack of direct patient contact.
Acknowledgement
We would like to thank Editage (www.editage.com) for the English language editing.
Disclosure
The authors have no relevant conflict of interests to disclose. The work was not supported or funded by any drug company. The study was approved by the research biomedical ethical committee of King Abdulaziz University (Reference no. 534-25).
Contributor Information
Nour Gazzaz, Email: ngazzaz@kau.edu.sa.
Fajr A Saeedi, Email: fasaeedi@kau.edu.sa.
References
- [1].Gregory GA, Robinson TIG, Linklater SE, Wang F, Colagiuri S, de Beaufort C, et al. , International diabetes federation diabetes atlas type 1 diabetes in adults special interest group. https://diabetesatlas.org/. [Google Scholar]
- [2].Magliano DJ, Maniam J, Orchard TJ, Rai P, Ogle GD. Global incidence, prevalence, and mortality of type 1 diabetes in 2021 with projection to 2040: a modelling study. Lancet Diabetes Endocrinol. 2022;10(10):741–760. https://pubmed.ncbi.nlm.nih.gov/36113507/. 10.1016/S2213-8587(21)00327-2 [DOI] [PubMed] [Google Scholar]
- [3].Ilonen J, Lempainen J, Veijola R. The heterogeneous pathogenesis of type 1 diabetes mellitus. Nat Rev Endocrinol. 2019;15:635–650. https://pubmed.ncbi.nlm.nih.gov/31534209/. 10.1038/s41574-019-0254-y [DOI] [PubMed] [Google Scholar]
- [4].Dhatariya KK, Glaser NS, Codner E, Umpierrez GE. Diabetic ketoacidosis. Nat RevDis Prim 2020;6:1–20. https://pubmed.ncbi.nlm.nih.gov/32409703/. [DOI] [PubMed] [Google Scholar]
- [5].Duca LM, Wang B, Rewers M, Rewers A. Diabetic ketoacidosis at diagnosis of type1 diabetes predicts poor long-term glycemic control. Diabetes Care. 2017;40:1249–1255. https://pubmed.ncbi.nlm.nih.gov/28667128/. 10.2337/dc17-0558 [DOI] [PubMed] [Google Scholar]
- [6].Shalitin S, Fisher S, Yackbovitch-Gavan M, de Vries L, Lazar L, Lebenthal Y, Phillip M. Ketoacidosis at onset of type 1 diabetes is a predictor of long-term glycemic control. Pediatr Diabetes. 2018;19:320–328. https://pubmed.ncbi.nlm.nih.gov/28568379/. 10.1111/pedi.12546 [DOI] [PubMed] [Google Scholar]
- [7].Lizzo JM, Goyal A, Gupta V. Adult diabetic ketoacidosis. StatPearls. StatPearls Publishing. 2022. Accessed February 2, 2023. http://www.ncbi.nlm.nih.gov/books/NBK560723/. [PubMed] [Google Scholar]
- [8].Tzimenatos L, Nigrovic LE. Managing diabetic ketoacidosis in children. Ann Emerg Med. 2021;78(3):340–345. https://pubmed.ncbi.nlm.nih.gov/33966934/. 10.1016/j.annemergmed.2021.02.028 [DOI] [PubMed] [Google Scholar]
- [9].Alhamdani YF, Almadfaa LO, AlAgha AE. Clinical variables influencing the severity of diabetes ketoacidosis. Saudi Med J. 2024;45(5):502–509. https://pubmed.ncbi.nlm.nih.gov/38734437/ 10.15537/smj.2024.45.4.20240058 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Parviainen A, But A, Siljander H, Knip M; Finnish pediatric diabetes register. Decreased incidence of type 1 diabetes in young finnish children. Diabetes Care. 2020;43(12):2953–2958. https://pubmed.ncbi.nlm.nih.gov/32998988/. 10.2337/dc20-0604 [DOI] [PubMed] [Google Scholar]
- [11].Al Hayek AA, Al Dawish MA. Frequency of diabetic ketoacidosis in patients with type 1 diabetes using FreeStyle Libre: A retrospective chart review. Adv Ther. 2021;38(6):3314–3324. https://pubmed.ncbi.nlm.nih.gov/34009604/. 10.1007/s12325-021-01765-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Groe J, Hornstein H, Manuwald U, Kugler J, Glauche I, Rothe U. Incidence of diabetic ketoacidosis of new-onset type 1 diabetes in children and adolescents in different countries correlates with human development index (HDI): an updated systematic review, meta-analysis, and meta-regression. Horm Metab Res. 2018;50:209–222. https://pubmed.ncbi.nlm.nih.gov/29523007/. 10.1055/s-0044-102090 [DOI] [PubMed] [Google Scholar]
- [13].Sohal A, Bains K, Dhaliwal A, Chaudhry H, Sharma R, Singla P, et al. , Seasonal variations of hospital admissions for alcohol-related hepatitis in the United States. Gastroenterology Res. 2022. Apr;15(2):75–81. https://pmc.ncbi.nlm.nih.gov/articles/PMC9076155/. 10.14740/gr1506 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Obando-Pacheco P, Justicia-Grande AJ, Rivero-Calle I, Rodriǵuez- Tenreiro C, Sly P, Ramilo O, et al. , Respiratory syncytial virus seasonality: a global overview. J Infect Dis. 2018;217(9):1356–1364. https://pubmed.ncbi.nlm.nih.gov/29390105/. 10.1093/infdis/jiy056 [DOI] [PubMed] [Google Scholar]
- [15].Westergren T, Aagaard H, Hall EOC, Ludvigsen MS, Fegran L, Robstad N, et al. , Physical activity enforces well-being or shame in children and adolescents with asthma: A meta-ethnography. Inquiry. 2024;61:469580241290086. https://pubmed.ncbi.nlm.nih.gov/39497650/. 10.1177/00469580241290086 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Hou L, Li M, Huang X, et al. , Seasonal variation of hemoglobin A1c levels in patients with type 2 diabetes. Int J Diabetes Dev Ctries. 2017; 37, 432–436. https://www.google.com/url?esrc=s&q=&rct=j&sa=U&url=https://www.researchgate.net/publication/303484217_Seasonal_variation_of_hemoglobin_A1c_levels_in_patients_with_type_2_diabetes&ved=2ahUKEwip_b3LuPyUAxVGUqQEHeHfNq4QFnoECAcQAg&usg=AOvVaw3fNPQz9SMl_lPk82XxDaK8p. 10.1007/s13410-016-0500-y [DOI] [Google Scholar]
- [17].Butalia S, Johnson JA, Ghali WA, Southern DA, Rabi DM. Temporal variation of diabetic ketoacidosis and hypoglycemia in adults with Type-I diabetes: A nationwide cohort study. J Diabetes. 2016;8(4):552–558. https://pubmed.ncbi.nlm.nih.gov/26301804/. 10.1111/1753-0407.12336 [DOI] [PubMed] [Google Scholar]
- [18].Taieb A, Hela G, Salsabil A, Asma G, Koussay A. Factors associated with severe diabetic ketoacidosis in patients diagnosed with type 1 diabetes: a decade-long cross-sectional analysis. J Int Med Res. 2024;52(10):3000605241281654. https://pubmed.ncbi.nlm.nih.gov/39422057/. 10.1177/03000605241281654 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Ghouri N, Gatrad R, Sattar N, Dhami S, Sheikh A. Summer-winter switching of the Ramadan fasts in people with diabetes living in temperate regions. Diabet Med. 2012;29(6):696–697. https://pubmed.ncbi.nlm.nih.gov/22060294/. 10.1111/j.1464-5491.2011.03519.x [DOI] [PubMed] [Google Scholar]
- [20].Razavi Z, Hamidi F. Diabetic ketoacidosis: demographic data, clinical profile and outcome in a tertiary care hospital. Iran J Pediat. 2017;27(3). https://brieflands.com/journals/ijp/articles/7649. [Google Scholar]
- [21].Berglund L, Berne C, Svardsudd K, Garmo H, Melhus H, Zethelius B. Seasonal variations of insulin sensitivity from a euglycemic insulin clamp in elderly men. Ups J Med Sci. 2012;117(1):35–40. https://pmc.ncbi.nlm.nih.gov/articles/PMC3282240/. 10.3109/03009734.2011.628422 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Turtinen M, Härkönen T, Ilonen J, Parkkola A, Knip M; Finnish pediatric diabetes register. Seasonality in the manifestation of type 1 diabetes varies according to age at diagnosis in Finnish children. Acta Paediatr. 2022;111(5):1061–1069. https://pubmed.ncbi.nlm.nih.gov/35137452/. 10.1111/apa.16282 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Beshyah SA, Chowdhury TA, Ghouri N, Lakhdar AA. Risk of diabetic ketoacidosis during Ramadan fasting: a critical reappraisal. Diabetes Res Clin Pract. 2019;151:290–298. https://pubmed.ncbi.nlm.nih.gov/30836132/. 10.1016/j.diabres.2019.02.027 [DOI] [PubMed] [Google Scholar]
- [24].Sunni M, Brunzell C, Nathan B, Moran A. Management of diabetes during Ramadan: practical guidelines. Minn Med. 2014;97(6):36–38. https://pubmed.ncbi.nlm.nih.gov/25029798/. [PubMed] [Google Scholar]
- [25].Habib HS. Frequency and clinical characteristics of ketoacidosis at onset of childhood type 1 diabetes mellitus in Northwest Saudi Arabia. Saudi Med J. 2005;26(12):1936–1939. https://pubmed.ncbi.nlm.nih.gov/16380776/. 10.15537/1658-3175.3238 [DOI] [PubMed] [Google Scholar]
- [26].Saadeh NA, Hammouri HM, Zahran DJ. Diabetic ketoacidosis in Northern Jordan: Seasonal morbidity and characteristics of patients. Diabetes Metab Syndr Obes. 2023;16:3057–3064. https://pubmed.ncbi.nlm.nih.gov/37810572/. 10.2147/DMSO.S413405 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Song SO, Yun JS, Ko SH, Ahn YB, Kim BY, Kim CH, et al. , Type 1 diabetes study group of gyeonggi-incheon branch of the Korean diabetes association. Prevalence and clinical characteristics of fulminant type 1 diabetes mellitus in Korean adults: A multi-institutional joint research. J Diabetes Investig. 2022;13(1):47–53. https://pubmed.ncbi.nlm.nih.gov/34313011/. 10.1111/jdi.13638 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Dong W, Zhang S, Yan S, Zhao Z, Zhang Z, Gu W. Clinical characteristics of patients with early-onset diabetes mellitus: a single-center retrospective study. BMC Endocr Disord. 2023;23(1):216. https://pubmed.ncbi.nlm.nih.gov/37814295/. 10.1186/s12902-023-01468-2 [DOI] [PMC free article] [PubMed] [Google Scholar]