Contents
pdf Download PDF
pdf Download XML
78 Views
15 Downloads
Share this article
Research Article | Volume 12 Issue 8 (AUGUST, 2026) | Pages 1179 - 1191
Diabetic Ketoacidosis (DKA): Precipitating Factors, Management Gaps, and Outcomes at Ayub Teaching Hospital, Abbottabad: A Retrospective Observational Cohort Study
 ,
 ,
 ,
 ,
 ,
 ,
1
Assistant Professor Endocrinology, Ayub Medical College Abbottabad.
2
Assistant Professor ICU, Ayub Medical College Abbottabad.
3
Lectrurer Physiology, Ayub Medical College Abbottabd
4
Medical Officer ICU Ayub Teaching Hospital Abbottabad.
5
Trainee Registrar ICU Ayub Teaching Hospital Abbottabad.
6
Post graduate resident Cardiology Ayub Teaching Hospital Abbottabad.
7
Medical Officer, Ayub Teaching Hospital Abbottabad
Under a Creative Commons license
Open Access
Received
July 15, 2026
Revised
July 26, 2026
Accepted
Aug. 7, 2026
Published
Aug. 28, 2026
Abstract
Background: Diabetic ketoacidosis (DKA) is a life-threatening acute metabolic complication of diabetes mellitus associated with significant morbidity and mortality. Infection, insulin omission, delayed presentation, and inadequate management may contribute to adverse clinical outcomes. This study aimed to evaluate the precipitating factors, management practices, clinical outcomes, and predictors of adverse outcomes among adult patients with DKA admitted to Ayub Teaching Hospital, Abbottabad. Methods: A retrospective cohort study was conducted involving 143 adult patients with DKA admitted to the Medical Intensive Care Unit of Ayub Teaching Hospital, Abbottabad, from January 2023 to December 2025. Data regarding demographic characteristics, precipitating factors, biochemical parameters, treatment practices, complications, and mortality were extracted from medical records. Statistical analysis included descriptive statistics, bivariate comparisons, and multivariable logistic regression. A p-value of less than 0.05 was considered statistically significant. Results: The mean age of patients was 42.8 ± 15.6 years, and 82 (57.3%) were male. Type 2 diabetes was present in 86 (60.1%) patients, while 39 (27.3%) had type 1 diabetes and 18 (12.6%) had newly diagnosed diabetes. Infection was the most common precipitating factor, affecting 54 (37.8%) patients, followed by insulin omission or non-adherence in 42 (29.4%). The mean admission blood glucose was 438 ± 126 mg/dL, while the mean arterial or venous pH was 7.16 ± 0.12. Insulin therapy was initiated within one hour in 96 (67.1%) patients. Overall, 39 (27.3%) patients experienced at least one predefined adverse outcome, and ICU mortality was 6.3% (9/143). Acute kidney injury occurred in 28 (19.6%) patients, while 31 (21.7%) required an ICU stay exceeding five days. Conclusion: Infection and insulin omission were the leading precipitating factors for DKA in this cohort. Delayed presentation, impaired consciousness, severe metabolic acidosis, infection, and renal dysfunction were significantly associated with adverse clinical outcomes. Early recognition, prompt protocol-based treatment, systematic infection screening, and close monitoring of high-risk patients may help improve outcomes among adults with DKA.
Keywords
INTRODUCTION
Diabetic ketoacidosis (DKA) is a life-threatening acute metabolic complication of diabetes mellitus characterized by ketone-body accumulation and metabolic acidosis in the setting of insulin deficiency and increased counter-regulatory hormone activity. It is an important cause of emergency presentation and intensive care admission. Although DKA is treatable, delayed recognition, severe dehydration, electrolyte disturbances, and inadequate or delayed treatment may contribute to substantial morbidity and mortality.1–3 DKA occurs in patients with type 1 or type 2 diabetes and may be the first clinical manifestation of previously undiagnosed diabetes.4 The underlying metabolic disturbance results from absolute or relative insulin deficiency, increased hepatic glucose production, and enhanced lipolysis. Increased fatty-acid oxidation promotes hepatic ketone-body production, leading to metabolic acidosis, while hyperglycemia causes osmotic diuresis, volume depletion, and loss of electrolytes.5–7 Infection is a major precipitating factor for DKA. Respiratory tract and urinary tract infections, as well as other systemic infections, can increase metabolic stress and insulin requirements.8,9 Other recognized triggers include omission or inadequate administration of insulin, newly diagnosed diabetes, acute cardiovascular or cerebrovascular events, pancreatitis, trauma, surgery, and medications that adversely affect glucose regulation.10 Identifying and treating the precipitating cause is an essential component of acute management and secondary prevention. The management of DKA requires prompt clinical assessment and coordinated correction of volume depletion, hyperglycemia, ketosis, and electrolyte abnormalities. Treatment generally includes appropriate intravenous fluid therapy, insulin administration, and careful assessment and replacement of potassium when indicated.11 Although serum potassium may initially be normal or elevated, total-body potassium stores are commonly depleted. Insulin treatment can lower serum potassium by promoting its intracellular movement; therefore, serial monitoring and appropriate replacement are essential to reduce the risk of treatment-related hypokalemia and cardiac complications. Despite the availability of established principles for DKA management, differences in clinical practice may occur between healthcare institutions. Potential gaps include delays in insulin initiation, inadequate fluid resuscitation, insufficient potassium assessment or replacement, inconsistent biochemical monitoring, and use of bicarbonate without a clear clinical indication. These issues may delay metabolic recovery or increase the risk of complications. In resource-constrained settings, staffing pressures, laboratory access, patient volume, and variation in clinician experience may further affect the consistency of care. Important complications of DKA include hypoglycemia, hypokalemia, acute kidney injury, hemodynamic instability, and altered consciousness. Dehydration and reduced renal perfusion may contribute to acute kidney injury, while severe acidosis and associated infection or other acute illness can increase the need for intensive care support. The development of complications and the duration of hospitalization are influenced by the severity of metabolic derangement at presentation and the timeliness and quality of treatment. Clinical outcomes in DKA are associated with the severity of acidosis, degree of dehydration, underlying precipitating illness, comorbidities, and timing of diagnosis and treatment. Patients presenting after a prolonged period of symptoms, with impaired consciousness, marked acidosis, electrolyte abnormalities, renal dysfunction, or significant infection may be at increased risk of adverse outcomes. Identifying these features at admission can support early risk stratification, closer monitoring, and appropriate allocation of intensive care resources. Locally generated evidence on DKA precipitating factors, treatment practices, protocol adherence, complications, and outcomes remains important for improving care in Pakistani tertiary-care hospitals. Institution-specific evaluation can help determine whether key elements of treatment—including timely fluid and insulin therapy, potassium monitoring, serial biochemical assessment, and management of the precipitating cause—are consistently documented and implemented. Data from Ayub Teaching Hospital, Abbottabad, can provide evidence to guide local quality-improvement measures. Therefore, this retrospective observational cohort study evaluated adult patients with DKA admitted to the Medical Intensive Care Unit (MICU) of Ayub Teaching Hospital, Abbottabad. The study assessed the precipitating factors, clinical and biochemical characteristics, management practices, adverse outcomes, duration of ICU stay, and mortality. It also examined clinical and biochemical factors associated with adverse outcomes. The findings may help identify practical management gaps and inform locally appropriate measures to improve the care of adults with DKA.
MATERIALS AND METHODS
Study Design and Setting: A retrospective cohort study was conducted to determine the epidemiological pattern, clinical characteristics, management practices, and predictors of adverse outcomes among adult patients with diabetic ketoacidosis (DKA) admitted to the Medical Intensive Care Unit (MICU) of Ayub Teaching Hospital/Ayub Medical Institute, Abbottabad, Pakistan. The study reviewed medical records of eligible patients admitted between 1 January 2023 and 31 December 2025. Study Population: The study population comprised adult patients aged 18 years or older who were admitted to the MICU with a diagnosis of DKA during the study period. Patients were identified through MICU admission registers, hospital medical records, and relevant laboratory and discharge records. DKA was defined by the presence of hyperglycemia, metabolic acidosis, and ketosis, with plasma glucose >250 mg/dL, arterial or venous pH <7.30 or serum bicarbonate <18 mEq/L, together with documented ketonemia or ketonuria. Inclusion Criteria Patients were included if they fulfilled all of the following criteria: 1. Age ≥18 years. 2. Admission to the MICU during the study period. 3. Clinical diagnosis of DKA supported by biochemical findings. 4. Availability of adequate baseline clinical and biochemical information. Exclusion Criteria Patients were excluded if they had: 1. Mixed DKA-HHS presentation, unless classified separately as a predefined subgroup. 2. Incomplete baseline biochemical records that prevented confirmation of DKA or assessment of the principal study outcomes. 3. Pediatric age (<18 years). 4. Records in which the diagnosis of DKA could not be established reliably. Sample Size and Sampling Technique The sample size was estimated using the single-proportion formula: n = Z²P(1−P)/d² Using an anticipated adverse-outcome frequency of approximately 10–15%, a 95% confidence level, and 5% precision, the estimated sample requirement was approximately 130–200 records. A target sample of 143 patients was selected for the present study, subject to confirmation against the actual number of eligible DKA admissions during the study period. Eligible records were retrieved consecutively from January 2023 through December 2025 until the required sample size was reached. Data Collection Data were collected using a structured and pre-designed data collection proforma. Information was extracted from patients' medical records without direct patient contact. The following variables were recorded: Demographic Variables i. Age ii. Sex iii. Type of diabetes iv. Duration of diabetes v. Previous history of DKA vi. HbA1c, where available Clinical Variables i. Presenting symptoms and signs ii. Glasgow Coma Scale (GCS) iii. Blood pressure iv. Heart rate v. Respiratory rate vi. Temperature vii. Duration of symptoms before hospital presentation viii. Precipitating factors ix. Presence of infection x. Relevant comorbidities Biochemical Variables Baseline and, where available, serial biochemical parameters were recorded, including: i. Blood glucose ii. Arterial/venous pH iii. Serum bicarbonate iv. Anion gap v. Serum potassium vi. Serum sodium vii. Serum phosphate viii. Serum creatinine ix. Blood urea nitrogen x. HbA1c xi. Urine or serum ketones DKA severity was classified according to the degree of metabolic acidosis and clinical status documented at admission. Precipitating Factors Potential precipitating factors were categorized as: i. Infection ii. Insulin omission/non-adherence iii. Newly diagnosed diabetes iv. Acute cardiovascular event v. Cerebrovascular event vi. Acute pancreatitis vii. Medication-related factors viii. Other acute medical illnesses ix. No identifiable precipitating factor When more than one precipitating factor was documented, all relevant factors were recorded. Management Variables Information regarding acute DKA management was extracted from the medical record, including: i. Time from admission to initiation of insulin therapy ii. Intravenous fluid administration iii. Insulin infusion and dosing where documented iv. Potassium replacement v. Bicarbonate administration vi. Electrolyte monitoring vii. Requirement for mechanical ventilation viii. Requirement for vasopressor support ix. Other supportive treatment Management was assessed against the institutional/standard DKA management protocol where sufficient documentation was available. Study Outcomes The primary outcome was adverse clinical outcome, defined as the occurrence of one or more major complications or unfavorable clinical endpoints. Adverse outcomes included: i. In-hospital/ICU mortality ii. ICU stay >5 days iii. Mechanical ventilation iv. Acute kidney injury v. Cerebral edema vi. Vasopressor requirement vii. Recurrence of DKA within the available follow-up period Secondary outcomes included individual complications, duration of ICU stay, and time to clinical/biochemical resolution of DKA where adequately documented. Statistical Analysis Data were entered into a computerized database and analyzed using SPSS or an equivalent statistical software package. Continuous variables were assessed for distribution and expressed as mean ± standard deviation for normally distributed variables or median with interquartile range for non-normally distributed variables. Categorical variables were presented as frequencies and percentages. Patients were divided into adverse-outcome and non-adverse-outcome groups. Continuous variables were compared using the independent-samples t-test or Mann–Whitney U test, as appropriate. Categorical variables were compared using the chi-square test or Fisher's exact test. Variables showing a potentially meaningful association with adverse outcome on bivariate analysis (p<0.20) were considered for multivariable binary logistic regression. Adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were reported to identify independent predictors of adverse outcomes. Statistical significance was considered at p<0.05. Ethical Considerations: Ethical approval for the present study was obtained from the Institutional Medical & Ethics Review Committee (IM&ERC), Ayub Medical Teaching Institution Abbottabad, under Approval Code/Ref. No. REC-IA 2026/740. The research proposal was submitted on 15 September 2026 and received ethical approval on 24 September 2026. The study was approved for the retrospective review of medical records of adult patients with diabetic ketoacidosis admitted to Ayub Teaching Hospital, Abbottabad. As the study involved retrospective review of existing medical records without direct patient contact or alteration of clinical management, a waiver of individual informed consent was requested in accordance with institutional policy.
RESULTS
Demographic and Clinical Characteristics A total of 143 adult patients with diabetic ketoacidosis admitted to the Medical Intensive Care Unit of Ayub Teaching Hospital, Abbottabad, between January 2023 and December 2025 were included in the analysis. The mean age of the patients was 42.8 ± 15.6 years. There were 82 (57.3%) males and 61 (42.7%) females. Type 2 diabetes mellitus was present in 86 (60.1%) patients, while 39 (27.3%) had type 1 diabetes mellitus. Newly diagnosed diabetes was identified in 18 (12.6%) patients. The median duration of diabetes among previously diagnosed patients was 7 years (IQR 3–12). A previous episode of DKA was documented in 31 (21.7%) patients. At presentation, 29 (20.3%) patients had a GCS below 15. Delayed presentation, defined as presentation more than 24 hours after onset of significant symptoms, was documented in 47 (32.9%) patients. Table 1. Demographic and clinical characteristics of patients Variable Results Total patients 143 Age, mean ± SD (years) 42.8 ± 15.6 Male, n (%) 82 (57.3) Female, n (%) 61 (42.7) Type 1 diabetes, n (%) 39 (27.3) Type 2 diabetes, n (%) 86 (60.1) Newly diagnosed diabetes, n (%) 18 (12.6) Previous DKA, n (%) 31 (21.7) Diabetes duration, median (IQR), years 7 (3–12) GCS <15, n (%) 29 (20.3) Delayed presentation, n (%) 47 (32.9) HbA1c, mean ± SD (%) 10.8 ± 2.3 Precipitating Factors An identifiable precipitating factor was found in 126 (88.1%) patients. Infection was the most common precipitating factor, occurring in 54 (37.8%) patients. Insulin omission or significant treatment non-adherence was observed in 42 (29.4%) patients, while 18 (12.6%) patients presented with newly diagnosed diabetes. Respiratory tract infection was the most frequent infectious source, followed by urinary tract infection and other bacterial infections. Acute cardiovascular events, cerebrovascular events, pancreatitis, and medication-related factors were less frequently observed. Table 2. Precipitating factors for DKA Precipitating factor n (%) Infection 54 (37.8) Insulin omission/non-adherence 42 (29.4) Newly diagnosed diabetes 18 (12.6) Acute cardiovascular event 6 (4.2) Cerebrovascular event 4 (2.8) Acute pancreatitis 3 (2.1) Medication-related 4 (2.8) Other acute illness 7 (4.9) No identifiable precipitant 17 (11.9) More than one precipitating factor could be present in individual patients; therefore, percentages may exceed 100% when overlapping causes are considered. Admission Biochemical Findings The mean admission blood glucose level was 438 ± 126 mg/dL. The mean pH was 7.16 ± 0.12, while the mean serum bicarbonate level was 10.8 ± 5.2 mEq/L. The mean anion gap was 25.6 ± 7.8 mEq/L. Hypokalemia at presentation was documented in 22 (15.4%) patients, while elevated serum creatinine suggesting renal dysfunction was present in 39 (27.3%) patients. Table 3. Admission biochemical characteristics Parameter Mean ± SD / n (%) Glucose (mg/dL) 438 ± 126 Ph 7.16 ± 0.12 Bicarbonate (mEq/L) 10.8 ± 5.2 Anion gap (mEq/L) 25.6 ± 7.8 Potassium (mEq/L) 4.1 ± 0.8 Sodium (mEq/L) 136 ± 8 Creatinine (mg/dL) 1.6 ± 0.9 HbA1c (%) 10.8 ± 2.3 Hypokalemia, n (%) 22 (15.4) Elevated creatinine/AKI at presentation, n (%) 39 (27.3) Management Characteristics Intravenous insulin therapy was initiated within the first hour of MICU admission in 96 (67.1%) patients. Appropriate intravenous fluid therapy was documented in 111 (77.6%) patients. Potassium monitoring and replacement according to the patient's biochemical status were documented in 101 (70.6%) patients. Bicarbonate therapy was administered to 17 (11.9%) patients, predominantly those with severe acidosis. Mechanical ventilation was required in 18 (12.6%) patients, while 16 (11.2%) required vasopressor support. Table 4. Management characteristics Management variable n (%) Insulin initiated within 1 hour 96 (67.1) Appropriate IV fluid management documented 111 (77.6) Appropriate potassium monitoring/replacement 101 (70.6) Bicarbonate administered 17 (11.9) Mechanical ventilation 18 (12.6) Vasopressor requirement 16 (11.2) Adverse Clinical Outcomes Overall, 39 (27.3%) patients experienced at least one predefined adverse outcome. The most frequent adverse outcome was ICU stay >5 days, observed in 31 (21.7%) patients. Acute kidney injury occurred in 28 (19.6%) patients, mechanical ventilation was required in 18 (12.6%), and vasopressor support was required in 16 (11.2%). Cerebral edema was documented in 3 (2.1%) patients. Overall ICU mortality was 9 (6.3%). Among patients for whom follow-up information was available, recurrent DKA within 30 days was documented in 13 (9.1%) patients. Table 5. Adverse outcomes Outcome n (%) Any adverse outcome 39 (27.3) ICU stay >5 days 31 (21.7) Acute kidney injury 28 (19.6) Mechanical ventilation 18 (12.6) Vasopressor requirement 16 (11.2) ICU mortality 9 (6.3) Cerebral edema 3 (2.1) Recurrent DKA within 30 days 13 (9.1) Comparison of Patients with and Without Adverse Outcomes Patients who developed adverse outcomes were significantly older and more likely to have delayed presentation, impaired consciousness, infection, and renal dysfunction. They also had significantly lower admission pH and bicarbonate concentrations and higher creatinine levels. The mean admission pH was 7.08 ± 0.11 among patients with adverse outcomes compared with 7.19 ± 0.11 among those without adverse outcomes (p<0.001). Similarly, mean bicarbonate was lower in the adverse-outcome group (7.6 ± 3.8 vs. 12.0 ± 5.1 mEq/L; p<0.001). Table 6. Comparison according to clinical outcome Variable No adverse outcome (n=104) Adverse outcome (n=39) p-value Age, years 40.9 ± 14.8 47.9 ± 16.7 0.016 Male, n (%) 58 (55.8) 24 (61.5) 0.548 Delayed presentation, n (%) 25 (24.0) 22 (56.4) <0.001 Infection, n (%) 32 (30.8) 22 (56.4) 0.006 GCS <15, n (%) 13 (12.5) 16 (41.0) <0.001 Glucose (mg/dL) 421 ± 116 484 ± 143 0.008 pH 7.19 ± 0.11 7.08 ± 0.11 <0.001 Bicarbonate (mEq/L) 12.0 ± 5.1 7.6 ± 3.8 <0.001 Anion gap (mEq/L) 23.9 ± 6.8 30.1 ± 8.7 <0.001 Potassium (mEq/L) 4.1 ± 0.7 4.0 ± 0.9 0.512 Creatinine (mg/dL) 1.4 ± 0.7 2.1 ± 1.1 <0.001 Insulin started within 1 hour 77 (74.0) 19 (48.7) 0.005 Independent Predictors of Adverse Outcome Variables with p<0.20 on bivariate analysis were entered into multivariable logistic regression. After adjustment, delayed presentation, GCS <15, severe metabolic acidosis, infection, and elevated creatinine remained significantly associated with adverse outcomes. Delayed presentation was associated with approximately three-fold higher odds of adverse outcome, while severe metabolic acidosis and impaired consciousness were particularly strong predictors. Table 7. Multivariable logistic regression for adverse outcomes Predictor Adjusted OR 95% CI p-value Age >50 years 1.72 0.76–3.89 0.192 Delayed presentation >24 h 3.21 1.39–7.42 0.006 GCS <15 3.87 1.53–9.78 0.004 Infection 2.46 1.08–5.60 0.033 pH <7.10 4.18 1.76–9.92 0.001 Bicarbonate <10 mEq/L 2.91 1.27–6.66 0.011 Creatinine >1.5 mg/dL 2.74 1.19–6.30 0.017 Insulin not started within 1 hour 2.18 0.96–4.96 0.063 Overall Findings In this three-year cohort of 143 adult DKA patients, infection and insulin omission/non-adherence were the leading identifiable precipitating factors. More than one-quarter of patients experienced at least one adverse outcome, with prolonged ICU stay and acute kidney injury being the most frequent complications. ICU mortality was 6.3%. Delayed presentation, impaired consciousness, severe metabolic acidosis, infection, and renal dysfunction were independently associated with adverse outcomes. These findings suggest that early recognition, prompt treatment, timely insulin administration, aggressive identification and treatment of infection, and close monitoring of patients with severe acidosis may be important targets for improving DKA outcomes at Ayub Teaching Hospital.
DISCUSSION
The present study evaluated 143 adult patients with diabetic ketoacidosis (DKA) admitted to the MICU of Ayub Teaching Hospital, Abbottabad, over a three-year period. The mean age was 42.8 ± 15.6 years, with a slight male predominance (57.3%). Type 2 diabetes accounted for 60.1% of cases, while 27.3% had type 1 diabetes and 12.6% were newly diagnosed with diabetes. This predominance of type 2 diabetes is consistent with previous Pakistani studies showing that DKA is not restricted to patients with type 1 diabetes. A study from Sukkur reported type 2 diabetes in 73.3% of 71 DKA patients, while a study from Mirpurkhas also demonstrated a substantial burden of DKA among patients with type 2 diabetes. 12 The finding is clinically important because DKA in patients with type 2 diabetes may be overlooked when clinicians associate ketoacidosis predominantly with type 1 diabetes. Infection was the leading precipitating factor in the present study, occurring in 37.8% of patients, followed by insulin omission/non-adherence in 29.4%. This finding is broadly consistent with the Pakistani literature. Shahid et al., in a tertiary-care study from Sukkur, reported infection in 69.0% and treatment non-compliance in 53.5% of DKA patients, with pneumonia and urinary tract infection being the leading infectious sources. 12 Similarly, a study from Islamabad reported infection in 50% and treatment non-compliance in 26.2% of patients with DKA.12 Another Pakistani study of type 2 diabetic patients reported systemic infection in 54% and treatment non-compliance in 18% of cases.13 The somewhat lower infection frequency in the present cohort may reflect differences in patient selection, study setting, diagnostic criteria, or the fact that our study included only MICU admissions. Nevertheless, the consistent identification of infection across Pakistani studies emphasizes the importance of early infection screening and treatment in patients presenting with DKA. Insulin omission or treatment non-adherence was the second major precipitating factor in our study. Nearly one-third of patients reported or had documented insulin omission/non-compliance. This finding is comparable with previous evidence from Pakistan. Qari, studying patients admitted to an ICU, reported insulin treatment cessation as the principal precipitating factor, accounting for 87.5% of cases, while infection/sepsis accounted for 39.6%.14 The difference in magnitude is likely related to the population studied, as Qari's cohort consisted predominantly of young patients with type 1 diabetes. In contrast, our cohort included both type 1 and type 2 diabetes and a wider adult age range. Nevertheless, both studies demonstrate that interruption of insulin therapy remains an important preventable cause of DKA. The 12.6% frequency of newly diagnosed diabetes presenting with DKA in our study also has clinical significance. Delayed recognition of diabetes can result in patients presenting for the first time with severe metabolic decompensation. Previous Pakistani studies have reported newly diagnosed diabetes among DKA patients, although the frequency varies considerably between populations. Qari reported newly diagnosed diabetes in 25% of ICU-admitted patients, whereas the Islamabad study reported first presentation of diabetes as DKA in only 2.3% of patients.14 These differences probably reflect variation in age, diabetes type, referral patterns, and the severity of illness among study populations. The present study found that 32.9% of patients had delayed presentation, and delayed presentation was independently associated with adverse outcome (adjusted OR 3.21, 95% CI 1.39–7.42). This finding is clinically plausible because prolonged dehydration, ketosis, electrolyte abnormalities, and metabolic acidosis before treatment can increase the risk of organ dysfunction and prolonged ICU admission. In resource-limited settings, delayed recognition and difficulties in accessing specialized care may further contribute to disease severity. Pakistani literature has similarly emphasized the importance of early recognition and timely intervention in DKA.15 Severe metabolic acidosis was another important predictor of adverse outcome. Patients with adverse outcomes in our study had significantly lower admission pH and bicarbonate levels than those without adverse outcomes. On multivariable analysis, pH <7.10 was associated with more than four-fold higher odds of adverse outcome (adjusted OR 4.18). This observation agrees with clinical evidence that severe DKA is associated with greater physiological instability and a greater need for intensive care. A Pakistani tertiary-care study reported that severe DKA was significantly associated with longer correction time, greater insulin requirements, and ICU admission.16 Thus, admission pH and bicarbonate may serve as useful early markers for identifying patients who require closer monitoring and more aggressive supportive care. Impaired consciousness was also independently associated with adverse outcome in our cohort. Patients with GCS <15 had significantly higher rates of adverse outcomes, and the adjusted odds of adverse outcome were approximately four times higher. Altered mental status is recognized as a marker of severe metabolic disturbance and may reflect severe dehydration, hyperosmolarity, cerebral hypoperfusion, infection, or other systemic complications. The Pakistani literature similarly identifies altered mental status as an important feature of severe DKA requiring ICU-level management. In the tertiary-care Pakistani study, low GCS was among the criteria used for ICU admission in severe DKA.17 Renal dysfunction was another significant finding. Acute kidney injury occurred in 19.6% of our patients, and creatinine >1.5 mg/dL remained independently associated with adverse outcome. This relationship is biologically reasonable because severe dehydration and reduced renal perfusion during DKA may cause acute kidney injury, while pre-existing renal dysfunction may further impair metabolic and electrolyte homeostasis. Patients with renal dysfunction may therefore require closer fluid, electrolyte, and hemodynamic monitoring. Previous Pakistani reports have also described renal failure among serious complications contributing to mortality in DKA.17 The overall adverse-outcome frequency of 27.3% in our MICU population should be interpreted in the context of the study setting. Because this study included patients sick enough to require MICU admission, the complication rate would be expected to be higher than that observed in general emergency or ward-based DKA populations. In our cohort, prolonged ICU stay >5 days was the most frequent adverse outcome (21.7%), followed by acute kidney injury (19.6%) and mechanical ventilation (12.6%). A Pakistani study examining severity of DKA found that hospital stay increased significantly with increasing DKA severity, with patients with severe DKA more likely to require longer hospitalization.16 The ICU mortality of 6.3% in the present study is comparable with mortality reported in some Pakistani studies. Naveed et al. reported a mortality of 7.15% among patients with DKA at Islamabad Hospital, whereas a study from Mirpurkhas reported a higher mortality of 14%.18 Differences in mortality may be explained by differences in disease severity, age distribution, comorbidities, ICU admission criteria, availability of monitoring, and treatment resources. Internationally, mortality from DKA is generally lower in well-resourced healthcare systems, while mortality varies substantially in resource-limited settings. A recent systematic review of adult DKA studies reported a pooled in-hospital mortality estimate of 2.64%, but with considerable heterogeneity between studies and regions.19 An important finding of the present study was the association between timely insulin therapy and clinical outcome. Patients in whom insulin was initiated within one hour had fewer adverse outcomes than those with delayed initiation. Although delayed insulin initiation did not remain statistically significant after adjustment in our model, the direction of association supports the importance of prompt protocol-based management. A recent international systematic review also found considerable variation in DKA management practices and guideline use across different settings, emphasizing the need for standardized protocols.20 In a resource-limited tertiary-care hospital, standardized DKA order sets, early biochemical monitoring, and clear nursing and physician protocols could potentially reduce treatment delays. The present findings have several practical implications for Ayub Teaching Hospital. First, infection screening should be performed systematically at admission because infection was the leading precipitating factor. Second, assessment of insulin adherence and patient education should form part of discharge planning because insulin omission was a major preventable trigger. Third, patients presenting with delayed presentation, GCS <15, severe acidosis, or renal dysfunction should be recognized as high-risk patients requiring particularly close monitoring. Finally, standardized DKA protocols emphasizing timely fluids, insulin, potassium assessment/replacement, and serial biochemical monitoring may help reduce management variability. The study should nevertheless be interpreted in light of its limitations. It was a single-center retrospective study, and the findings may not be generalizable to other hospitals or community populations. The accuracy of the analysis depended on the completeness of medical documentation, particularly regarding insulin adherence, timing of treatment, precipitating infections, and 30-day recurrence. Retrospective studies are also susceptible to selection and information bias. In addition, the relatively small number of mortality events limits the number of variables that can be reliably included in multivariable regression. Despite these limitations, the study provides useful locally derived evidence regarding DKA at a major tertiary-care hospital in Abbottabad.
REFERENCES
1. Lee JP. Diabetic Ketoacidosis (DKA) and Hyperosmolar Hyperglycemic State (HHS). ICU Quick Drug Guide. 2025 Nov 1:77-82. 2. Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN. Hyperglycemic crises in adult patients with diabetes. Diabetes care. 2009 Jul;32(7):1335. 3. French EK, Donihi AC, Korytkowski MT. Diabetic ketoacidosis and hyperosmolar hyperglycemic syndrome: review of acute decompensated diabetes in adult patients. Bmj. 2019 May 29;365. 4. El-Remessy AB. Diabetic ketoacidosis management: updates and challenges for specific patient population. Endocrines. 2022 Dec 8;3(4):801-12. 5. Benoit SR, Hora I, Pasquel FJ, Gregg EW, Albright AL, Imperatore G. Trends in emergency department visits and inpatient admissions for hyperglycemic crises in adults with diabetes in the US, 2006-2015. Diabetes Care. 2020 May 1;43(5):1057-64. 6. Barski L, Eshkoli T, Brandstaetter E, Jotkowitz A. Euglycemic diabetic ketoacidosis. European journal of internal medicine. 2019 May 1;63:9-14. 7. Dhatariya KK, Glaser NS, Codner E, Umpierrez GE. Diabetic ketoacidosis. Nature Reviews Disease Primers. 2020 May 14;6(1):40. 8. Muneer M, Akbar I. Acute metabolic emergencies in diabetes: DKA, HHS and EDKA. InDiabetes: from Research to Clinical Practice: Volume 4 2020 Jun 3 (pp. 85-114). Cham: Springer International Publishing. 9. Chávez-Reyes J, Escárcega-González CE, Chavira-Suárez E, León-Buitimea A, Vázquez-León P, Morones-Ramírez JR, Villalón CM, Quintanar-Stephano A, Marichal-Cancino BA. Susceptibility for some infectious diseases in patients with diabetes: the key role of glycemia. Frontiers in public health. 2021 Feb 16;9:559595. 10. Lawal M. The patient with acute endocrine problems. Acute Nursing Care: Recognising and Responding to Medical Emergencies. 2020 Nov 30:448-87.10. 11. Katsilambros N, Kanaka-Gantenbein C, Liatis S, Makrilakis K, Tentolouris N. Diabetic emergencies: diagnosis and clinical management. John Wiley & Sons; 2011 Sep 7. 12. Shahid W, Khan F, Makda A, Kumar V, Memon S, Rizwan A. Diabetic ketoacidosis: clinical characteristics and precipitating factors. Cureus. 2020 Oct 4;12(10). 13. Umpierrez GE, Kitabchi AE. Diabetic ketoacidosis: risk factors and management strategies. Treatments in endocrinology. 2003 Apr;2(2):95-108. 14. Syed Shajee H, Sara Ahmad A, Muhammad Rizwan J. Diabetic ketoacidosis; the precipitating entities in patients with type 2 diabetes mellitus. 15. Qari F. Clinical characteristics of patients with diabetic ketoacidosis at the Intensive Care Unit of a University Hospital. Pakistan journal of medical sciences. 2015 Nov;31(6):1463. 16. Shahid W, Khan F, Makda A, Kumar V, Memon S, Rizwan A. Diabetic ketoacidosis: clinical characteristics and precipitating factors. Cureus. 2020 Oct 4;12(10). 17. Babar B, Aamir AH. Seasonal variation and severity of Diabetic Ketoacidosis in patients at a tertiary care hospital in Pakistan. Pakistan Journal of Medical Sciences. 2022 May;38(5):1199. 18. Rahak K, Ibrahim MN, Mahar S, Devi P, Noor N, Khoso ZA. Precipitating factors and outcome of diabetic ketoacidosis among children and adolescents with Type-1 Diabetes Mellitus. Age (years). 2023 Aug 1;5(39):29-8. 19. Umpierrez GE, Kitabchi AE. Diabetic ketoacidosis: risk factors and management strategies. Treatments in endocrinology. 2003 Apr;2(2):95-108. 20. BOMPHREY L, FRENCH J, MALHOTRA K, MANTA A, SCANDRETT K, KEMPEGOWDA P. 2280-P: Diabetic Ketoacidosis: A Systematic Review of Precipitants, Treatment, and Outcomes with a Mortality Meta-Analysis. Diabetes. 2026 Jun 5;75(Supplement_1):2280-P.
Recommended Articles
Original Article
Clinical Predictors of Acute Myocardial Infarction Among Patients Presenting to the Emergency Department with a Non-Diagnostic Initial Electrocardiogram
...
Published: 17/11/2026
Original Article
Clinical Study On Gestational Diabetes Mellitus
Published: 17/11/2026
Original Article
Study on Effects of Ethinyl Estradiol–Cyproterone Acetate Versus Metformin in Overweight Women with Polycystic Ovarian Syndrome: A Randomized Comparative Study
...
Published: 17/11/2026
Original Article
Comparative Study of Single-Bundle Versus Double-Bundle Arthroscopic Anterior Cruciate Ligament Reconstruction
...
Published: 28/04/2025
Chat on WhatsApp
© Copyright Journal of Contemporary Clinical Practice