Acute triggers and long-term postpartum outcomes of diabetic ketoacidosis in pregnancy: a 7-case retrospective series

Abstract

Objective

To investigate the clinical characteristics, acute triggers, and long-term maternal and fetal outcomes of gestational diabetic ketoacidosis (DKA).

Methods

A retrospective analysis was performed on 7 pregnant women with DKA admitted to two tertiary hospitals from January 2018 to December 2025. Additionally, a long-term postpartum follow-up (range: 1.6–7.0 years; median: 5.3 years) was conducted to evaluate chronic complications and metabolic trajectories.

Results

Six cases occurred in the third trimester and one in the first. The primary iatrogenic trigger was dexamethasone administration for fetal lung maturation (n = 3), followed by acute pancreatitis (n = 3) and infection (n = 1). Acidosis was corrected within 12–24 h via standardized resuscitation. While no maternal deaths occurred, three cases suffered adverse fetal outcomes (two stillbirths, one miscarriage). Postpartum follow-up of six successfully traced patients revealed persistent metabolic derangements in five women (83.3%), including one severe microvascular progression requiring cataract surgery and one case of complete therapeutic non-compliance.

Conclusion

Gestational DKA is a critical obstetric emergency with high fetal mortality. Dexamethasone and lipid-induced pancreatitis are prominent triggers. Given the high rate of persistent postpartum metabolic derangements over long-term follow-up, prolonged multidisciplinary surveillance and stringent continuous glycemic education are mandatory.

Introduction

As a severe metabolic crisis, diabetic ketoacidosis (DKA) is defined by the classic triad of marked hyperglycemia, ketonemia, and metabolic acidosis. The physiological shifts characteristic of pregnancy, particularly the emergence of placental hormones that drive insulin resistance, notably elevate a woman’s vulnerability to this metabolic decompensation [1]. Current literature indicates that among pregnant individuals with established diabetes, the development of DKA occurs in approximately 0.5% to 3% of cases [2].

In obstetric practice, antenatal corticosteroids—most notably dexamethasone—are routinely employed to mature fetal lungs when preterm birth threatens. Nevertheless, dexamethasone acts as a potent antagonist to insulin action, inducing a rapid surge in maternal glucose levels by exacerbating peripheral resistance [3]. Consequently, in individuals with diagnosed or unrecognized glucose intolerance, such iatrogenic interventions can rapidly precipitate DKA if monitoring and insulin titration are suboptimal.

To date, comprehensive descriptions detailing pregnancy-specific DKA presentations, its immediate catalysts, and long-term consequences remain sparse, particularly within resource-limited clinical settings. To bridge this information gap, this study reviews a case series of seven pregnant individuals presenting with DKA, aiming to parse specific clinical profiles and highlight preventable risk considerations.

Methods

Study design and population

This retrospective institutional review tracked pregnant individuals diagnosed with DKA who received care at Jinhua Central Hospital or Jinhua Maternal and Child Health Care Hospital from January 2018 through December 2025. Our initial screening encompassed all pregnant patients admitted with known glucose metabolism disorders, including both pre-existing and gestational diabetes mellitus. From this larger pool, we selected individuals who strictly fulfilled the standardized diagnostic criteria for an acute DKA episode.

DKA was defined by: blood glucose > 13.9 mmol/L, arterial pH < 7.3, serum bicarbonate < 18 mmol/L, and presence of ketonuria (≥ ++) or serum β-hydroxybutyrate > 0.6 mmol/L [4].

Data collection

Demographic data, gestational age at onset, BMI, precipitating factors, clinical symptoms, laboratory findings, treatment protocols, and pregnancy outcomes were extracted from medical records. We also collected the time interval between the application of dexamethasone and the onset of DKA. In this study, an ‘acute trigger’ is defined as a specific medical intervention (such as antenatal dexamethasone administration) or an acute illness (e.g., acute pancreatitis or infection) that directly precipitates DKA within 24 to 72 h. For dexamethasone-induced cases, causality was established if the patient met three criteria: (1) no evidence of ketoacidosis prior to injection; (2) a sharp deterioration in glycemic control following the injection; and (3) DKA onset within 24 to 72 h, with no other identifiable metabolic or infectious causes.

Statistical analysis

Descriptive statistics were used. Continuous variables were presented as mean ± standard deviation, and categorical variables as frequencies and percentages. Analyses were performed using SPSS version 26.0.

Postpartum follow-up management

To assess long-term maternal metabolic trajectories, a structured postpartum follow-up was conducted in May 2026. The follow-up duration ranged from 1.6 to 7 years, with a median follow-up of 5.3 years. Data on postpartum diabetic classification, current lifestyle and pharmacological interventions, compliance with glucose monitoring, subsequent obstetric histories, and chronic diabetic complications were systematically gathered and updated.

Results

General characteristics

Among 69,412 deliveries during the study period, 7,056 women had gestational or pre-gestational diabetes, and 7 developed DKA (incidence: 0.101 per 1000 deliveries). Mean maternal age was 29.29 ± 4.11 years, and mean BMI was 29.12 ± 4.28 kg/m2. Five patients were multiparous. DKA occurred between 5^+ 5 and 32^+ 1 weeks of gestation; 85.7% (n = 6) in the third trimester. The detailed screening, exclusion, and participant enrollment process are systematically illustrated in Fig. 1.

Fig. 1.

Flowchart of participant screening and enrollment

Precipitating factors and symptoms

Triggers included: dexamethasone for fetal lung maturation (n = 3), acute pancreatitis (n = 3), and infectious fever (n = 1). The most common symptoms included abdominal pain (n = 4), dyspnea (n = 4), nausea and vomiting (n = 3), and altered mental status (n = 1, coma on admission). Cases 1, 3, and 7 developed ketoacidosis 24 h, 27 h, and 30 h after the administration of dexamethasone.

Laboratory findings

Mean blood glucose: 19.81 ± 4.53 mmol/L; pH: 7.19 ± 0.124; bicarbonate: 9.14 ± 4.26 mmol/L; HbA1c: 9.06 ± 2.27%. All patients had ketonuria of ≥ ++. Serum β-hydroxybutyrate (measured in 4 cases): 1.3–3.98 mmol/L. Leukocytosis and elevated CRP levels were common findings. The detailed clinical and laboratory data of all seven patients are presented in Table 1.

Table 1.

Clinical and Laboratory Data of 7 Pregnant Patients with DKA

No.	Age	BMI	Onset GW	Trigger	Symptoms	Past Hx	Prenatal care	BG	pH	β-HB	KET	HCO3-	HbA1c	Pregnancy Outcomes
1	25	35.2	30+ 4	DXM	Thirst	None	Regular	25.3	7.28	3.37	4+	12.2	6.6	Live birth via CS at 32+ 4 GW, 2,635 g, Apgar 10
2	33	24.6	29+ 6	Inf. Fever	N/V, Fever	T2DM (7y)	Irregular	20.9	7.00	1.30	4+	7.0	9.4	Intrauterine fetal death (IUFD) at 30+ 2 GW
3	34	26.2	31+ 6	DXM	Chest Distress	T2DM, IUFD	Regular	26.4	7.24	3.53	4+	9.5	6.7	Live birth via CS at 36+ 0 GW, 4,770 g, Apgar 10
4	24	30.5	30+ 4	Acute Pan.	Abd. Pain	Hyperlipidemia	Irregular	15.0	7.17	NA	4+	6.3	8.4	Live birth via CS at 37+ 4 GW, 4,715 g, Apgar 10
5	27	NA	5+ 5	Acute Pan.	Abd. Pain, Coma	None	None	18.2	7.29	3.98	4+	13.8	11.9	Spontaneous abortion (Miscarriage) at 5+ 5 GW
6	33	32.6	30+ 2	Acute Pan.	Low Abd. Pain	None	Irregular	16.4	7.03	NA	4+	2.1	11.4	Induced labor due to IUFD at 30+ 5 GW
7	29	25.6	32+ 1	DXM	Dizzy, Abd. Pain	Pulm. HTN	Irregular	16.5	7.29	NA	2+	13.1	NA	Live birth via CS at 38+ 0 GW, 2,940 g, Apgar 10

DXM dexamethasone, GW gestational week, CS cesarean section, IUFD intrauterine fetal death, BG blood glucose (mmol/L), β-HB β-hydroxybutyrate (mmol/L), KET urine ketone bodies, HCO3- bicarbonate (mmol/L), HbA1c glycated hemoglobin (%), T2DM type 2 diabetes mellitus, N/V nausea/vomiting, Pan. Pancreatitis, Inf. Infectious, Abd. Abdominal, Pulm. Pulmonary, HTN hypertension, Hx history, NA  not available

Management and outcomes

All patients received intravenous fluid resuscitation (3500–5000 mL/24 h), continuous insulin infusion (0.1 U/kg/h), and electrolyte replacement. Acidosis resolved within 12–24 h. No maternal deaths occurred. Perinatal outcomes: 3 fetal losses (2 IUFD, 1 miscarriage); 4 live births (2 term, 2 preterm). Notably, adverse fetal outcomes (including two cases of intrauterine fetal death and one case of spontaneous abortion) were predominately observed in patients with suboptimal long-term glycemic control, with HbA1c levels ranging from 9.4% to 11.9% (Table 1).

Regarding long-term maternal prognosis, successful follow-up was completed for six patients. The median follow-up duration for these successfully followed patients was 5.3 years (range: 1.6–7.0 years). Specifically, earlier cases had extended tracking durations ranging from 5.2 to 7.0 years, while more recent cases (admitted in 2024) had naturally shorter follow-up periods of 1.6 to 2.0 years. Case 6 was lost to follow-up (Table 2). The long-term metabolic outcomes varied substantially among individuals. Case 1 demonstrated rapid microvascular deterioration and underwent a postoperative cataract removal; she currently requires combined treatment consisting of oral antidiabetic drugs (OADs) and long-acting insulin glargine (Toujeo, 26 U/day). Case 2 showed a total lack of post-discharge compliance, failing to monitor blood glucose or take medication despite a documented follow-up fasting blood glucose level of 11.1 mmol/L. For Case 3, a multi-dose basal-bolus insulin regimen (Detemir 20 U/day plus Aspart 8 U per meal) was essential to manage persistent post-pregnancy insulin resistance. Regarding cases secondary to acute pancreatitis, Case 4 (initial triglycerides at 19.35 mmol/L) continues NPH insulin therapy (14 U/day), while Case 5 (who delivered a 4000 g infant in 2022 and a 4500 g infant in 2025) utilizes metformin monotherapy. Case 7, with baseline pulmonary arterial hypertension and a history of two spontaneous abortions, maintained glycemic targets via dietary and lifestyle modifications alone.

Table 2.

Postpartum Long-Term Follow-Up and Maternal Metabolic Outcomes

No.	Follow-Up Status	Follow-Up Duration	Maternal medical history	Current Postpartum Diabetic Management & Treatment	Long-Term Complications / Clinical Notes
1	Successful	7.0 years	1 index live birth*	Long-acting insulin glargine (Toujeo) 26 U/day + oral antidiabetic drugs (OADs)	Postoperative cataract removal; indicates severe microvascular progression.
2	Successful	2.0 years	1 index IUFD; Type 2 diabetes mellitus (T2DM) for 7 years. Registered fasting blood glucose was 11.1 mmol/L.	No medication or glucose monitoring	Complete non-compliance; extremely high risk of recurrent metabolic crisis.
3	Successful	1.6 years	1 historical IUFD; 1 index live birth*	Basal-bolus regimen: Detemir 20 U/day and Aspart 8 U/meal	Required intensive multi-dose insulin regimen due to severe insulin resistance.
4	Successful	5.9 years	1 index live birth*; Hyperlipidemia with triglycerides at 19.35 mmol/L	Neutral Protamine Hagedorn (NPH) 14 U/day	Managed with intermediate-acting insulin post-pancreatitis.
5	Successful	5.3 years	1 index miscarriage	Oral metformin monotherapy	Delivered 2 consecutive macrosomic infants (2022: 4000 g; 2025: 4500 g); persistent T2DM.
6	Lost	NA	1 index IUFD	Unknown	Contacted ex-spouse; patient lost to clinical follow-up due to divorce.
7	Successful	5.2 years	2 historical miscarriages; 1 index live birth*; Pulmonary arterial hypertension	Dietary and lifestyle modifications only	Managed with nutrition therapy; close surveillance required for future pregnancies.

IUFD intrauterine fetal death, T2DM type 2 diabetes mellitus, Toujeo insulin glargine 300 U/mL, OADs oral antidiabetic drugs, NPH Neutral Protamine Hagedorn insulin, NA  not available

* “Index live birth” refers to the infant delivered during the current DKA pregnancy detailed in Table 1

Discussion

The clinical manifestation of DKA during gestation typically clusters within the third trimester, a period mirroring the peak of placental hormone production and subsequent physiological insulin resistance [1]. Despite generally favorable maternal survival rates, gestational DKA remains a formidable obstetric hazard, carrying an exceptionally heavy toll on the fetus [2].

A critical observation in our cohort centers on the role of antenatal dexamethasone as a leading iatrogenic trigger. Corticosteroids markedly compound systemic insulin resistance while concurrently accelerating hepatic gluconeogenesis, producing rapid upward shifts in circulating blood glucose [3]. A recent large cohort study demonstrated that following antenatal corticosteroid administration, maternal blood glucose typically peaks at approximately 9 h and hyperglycemia persists for up to 72 h [5]. Furthermore, maternal hyperglycemia during this period significantly increases the risk of severe neonatal hypoglycemia [5]. For women with underlying or unmasked pancreatic insufficiency, this 72-hour metabolic shock can easily precipitate ketoacidosis. The observation that our patients developed clear DKA within 24–30 h of receiving dexamethasone strongly aligns with this timeline, validating the clinical need for aggressive glycemic surveillance and proactive insulin therapy whenever corticosteroids are scheduled [6, 7].

Concurrently, acute pancreatitis emerged as another dominant metabolic trigger, frequently linked to pregnancy-induced hypertriglyceridemia. Normal gestation naturally alters lipid profiles toward increased lipolysis; however, a severe surge in triglycerides can irritate pancreatic tissue, sparking an intense inflammatory cascade that fundamentally disrupts glucose homeostasis and precipitates DKA [8]. This dual pathology indicates that early lipid screening and strict metabolic surveillance during pregnancy are key preventative measures, particularly among patients presenting with maternal obesity or carbohydrate intolerance.

Acid-base disturbances and marked ketosis characterized our entire cohort [4]. Markers such as an arterial pH below 7.30, low bicarbonate levels, and an HbA1c exceeding 9.4% typically point to sustained poor glycemic control and suboptimal prenatal oversight. This elevated baseline HbA1c points to long-standing maternal hyperglycemia, which impairs placental function and fosters fetal hypoxia. Furthermore, circulating ketone bodies readily cross the placental barrier, exerting direct cytotoxic and teratogenic effects on the developing fetal myocardium and central nervous system [9]. When severe chronic hyperglycemia (as seen in Cases 5 and 6) occurs alongside profound maternal acidemia (as in Case 6, pH 7.03), the intrauterine environment becomes highly hostile, leading to poor fetal outcomes.

Diagnosing DKA in obstetric settings is frequently delayed because its initial symptoms often mimic standard gestational complaints, such as benign emesis, early labor pains, or acute gastroenteritis. Clinicians must maintain a low threshold for suspicion when managing pregnant patients with persistent hyperglycemia, moving quickly to assess blood gases, serum chemistry, and ketones to prevent diagnostic delays.

The cornerstone of treating this crisis hinges on prompt, aggressive fluid resuscitation, a continuous intravenous insulin drive, and vigilant management of potassium and other electrolytes. Continuous tracking of maternal hemodynamics, metabolic panels, and fetal heart patterns remains indispensable. Although maternal acid-base equilibrium typically normalizes within a 12-to-24-hour window under standard protocols, the ultimate survival and neurological intactness of the fetus depend heavily on the duration of the metabolic insult and baseline glycemic control.

The inclusion of an extended postpartum follow-up period ranging from 1.6 to 7 years (median: 5.3 years) offers an essential perspective on the chronic maternal impact of gestational DKA (Table 2). Although acute resuscitation during the index pregnancy succeeded in all patients, Case 1 experienced rapid microvascular progression leading to cataract surgery within the post-discharge window. This observation indicates that gestational DKA should not be viewed merely as a transient obstetric crisis; rather, it often marks the acceleration of underlying vascular complications, highlighting the necessity of long-term ophthalmic and metabolic tracking.

The clinical management challenge is further illustrated by Case 2, who completely ceased glucose monitoring and medication adherence after delivery. Despite a follow-up fasting blood glucose of 11.1 mmol/L and her history of an intrauterine fetal death (Table 1), her unmanaged metabolic status creates a high susceptibility to recurrent crises. This finding reveals a clear deficiency in continuous postpartum care, indicating that counseling regarding chronic risks must extend far beyond the acute hospitalization phase to ensure patient compliance.

Additionally, our long-term follow-up reflects how historical metabolic risks and past adverse obstetric events cluster in these patients. The severe hypertriglyceridemia (19.35 mmol/L) observed in Case 4 supports the established concept that lipid-driven acute pancreatitis dramatically worsens maternal glucose metabolism [8]. Similarly, the recurrent history of macrosomia in Case 5 (4000 g and 4500 g neonates) and multiple miscarriages combined with pulmonary hypertension in Case 7 reveal that these women harbor long-standing insulin resistance and vascular abnormalities that outlast placental separation. Consequently, pre-pregnancy counseling and persistent metabolic surveillance are vital to improve reproductive and maternal health over the long term.

Our study has several limitations. First, the sample size is extremely small (n = 7). This limits the statistical power to identify specific risk factors. Crucially, our study lacks a non-DKA diabetic control group. Artificially selecting a small, matched control group from our database would introduce significant selection bias and residual confounding. Any comparative analysis in such a small cohort would be severely underpowered. Therefore, we focused solely on descriptive clinical observations to maintain methodological integrity. Second, some specific laboratory data (e.g., arterial blood gas and serum β-hydroxybutyrate) were missing. These invasive or specialized tests are not routinely performed in general obstetric patients. Third, our current ethical approval strictly limits data collection to patients who experienced DKA. Future large-scale, prospective multicenter studies are needed to evaluate comparative risks.

Conclusion

In summary, DKA during pregnancy represents a dangerous obstetric and metabolic crisis linked to high fetal mortality. Corticosteroid administration for fetal lung maturation and lipid-induced acute pancreatitis stand out as major triggers that require intense clinical surveillance and timely insulin adjustment. Crucially, our follow-up data extending up to 7 years demonstrate that the systemic effects of gestational DKA persist long after delivery. This is evidenced by high rates of chronic glucose intolerance, rapid microvascular degradation such as cataracts, and poor post-discharge compliance. Managing these risks requires long-term metabolic tracking, structured preconception counseling, and enhanced patient education to close the gap in postpartum care and improve subsequent maternal outcomes.

Authors’ contributions

Ya-Li Yang: Data curation, formal analysis, writing—original draft and review.Qiao-Fang Yang: Data collection, writing—original draft and editing.Xu-Fei Fan: Conceptualization, methodology, supervision, project administration, writing—review and editing.

Funding

This study was supported by the Science and Technology Plan Project of Jinhua Municipal Bureau of Science and Technology (Grant No. 2025-4-145). The funder had no role in study design, data collection, analysis, or writing.

Data availability

The datasets generated and/or analysed during the current study are not publicly available due to the protection of patient privacy and clinical confidentiality, but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval was obtained from the Ethics Committee of Jinhua Maternal and Child Health Care Hospital (Approval No. 2025QT003). All procedures performed in this study involving human participants were conducted in accordance with the ethical standards of the institutional research committee and the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. The need for informed consent to participate was formally waived by the Ethics Committee of Jinhua Maternal and Child Health Care Hospital due to the retrospective nature of the study and the complete anonymization of all patient data.

Consent for publication

Not applicable. This manuscript does not contain any individual patient’s identifying images, videos, or personal details that could compromise anonymity.

Competing interests

The authors declare that they have no relevant financial or non-financial competing interests to disclose.