Concurrent Diabetic Ketoacidosis and Acute Coronary Syndrome: A Systematic Review of Case Reports

Introduction

Diabetic ketoacidosis (DKA) is one of the most serious complications of diabetes mellitus (DM), with approximately 220,000 patients with a primary diagnosis of DKA in the United States, representing nearly 60 cases per 100,000 hospital admissions [¹]. This high prevalence contributes significantly to the financial burden of DKA, with hospitalization costs for DKA reaching nearly 6.76 billion dollars, with an average length of stay of only three days [²]. DKA results from insulin depletion that often happens in cases of undiagnosed DM, with a triad of hyperglycemia, ketosis, and high anion gap [³,⁴]. On the other hand, nearly 805,000 acute myocardial infarctions (AMI) occur each year, with approximately 600,00 of them being first-time myocardial infarction (MI) events [⁵]. Acute coronary syndrome (ACS) refers to conditions caused by a sudden reduction or blockage of blood supply to the heart. It includes unstable angina, non-ST-elevation myocardial infarction (NSTEMI), and ST-elevation myocardial infarction (STEMI) [⁶].

DKA and ACS are medical emergencies requiring immediate intervention to prevent potentially fatal outcomes. However, the relationship between the two conditions appears bidirectional and clinically significant, as DM is one of the modifiable risk factors for ACS. At the same time, ACS can precipitate DKA by inducing physiological stress responses [⁷]. Additionally, there is a strong association between DM and cardiovascular diseases, with approximately one-third of diabetes patients reporting serious cardiovascular events [⁸]. The precise pathophysiological mechanisms connecting DKA and ACS remain incompletely understood, despite the clear epidemiological association between the two conditions, as MI and heart failure account for 28% mortality in DKA cases [⁹]. The diagnostic challenge is further complicated by euglycemic diabetic ketoacidosis (DKA), particularly in patients using sodium-glucose cotransporter-2 (SGLT2) inhibitors, where normal or mildly elevated glucose levels may delay recognition of ketoacidosis in those presenting with cardiac symptoms. To better understand this relationship, we aim to systematically investigate reported cases of DKA and ACS, describe the clinical presentation and laboratory patterns when both conditions co-occur, and evaluate the effect of this comorbidity on patient outcomes and hospitalization course.

Methods

In this systematic review of case reports, we followed the updated version of the Preferred Reporting Items for Systematic Reviews (PRISMA) [¹⁰].

Results

In our systematic review, 1,743 records were initially identified from the searched databases, of which only 1,278 were retrieved after removing 465 duplicate articles. A total of 1212 records were excluded following title and abstract screening, resulting in 66 records for full-text screening. Of these, 45 records did not meet our eligibility criteria. Finally, we included 21 case reports [¹²,¹³,¹⁴,¹⁵,¹⁶,¹⁷,¹⁸,¹⁹,²⁰,²¹,²²,²³,²⁴,²⁵,²⁶,²⁷,²⁸,²⁹,³⁰,³¹,³²], as shown in Figure 1.

Patients’ Characteristics

Out of review included 21 patients, with a male predominance of 11 males (52.3%) and 9 females (42.8%). The patients’ ages ranged from 18 to 77, with a mean age of 51. The majority of patients were between 50 and 60 years old. Out of these 21 patients, 12 (57.1%) were diagnosed with type 2 diabetes mellitus (T2DM) and 9 (42.8%) with type 1 diabetes mellitus (T1DM). Additionally,

Table 1

Table 1: Summary of Patient Demographics and ACS Presentation

Author	Age, years	Sex	Diabetes type	Presentation	Troponin, ng/ml	Type of ACS
Baral 2021 [12]	77	M	T2DM	SOB, cough, fever for 3 days, oxygen saturation 80% on 10 liters nasal cannula (improved to 95% on a non-rebreather mask with 15L/min).	Initial: 0.923, Peaked at 23.684	STEMI
Baytu0̆11fan 2023 [13]	61	M	T2DM	CP, generalized weakness, polyuria, nausea & vomiting for several days.	161.7	STEMI
Bouzhir 2024 [14]	76	F	T2DM	Severe CP at rest, palpitations post-cataract surgery, persistent symptoms after 30 hours, & a dry cough for 10 days.	3,859	STEMI
Briggs 2022 [15]	56	F	T2DM	48 hours of nausea, vomiting, generalized abdominal pain, and diarrhea. No CP, SOB, or fever.	212	STEMI
Çakır 2012 [16]	18	F	T1DM	Fatigue, weakness, nausea, vomiting for two days, and abdominal pain. No CP.	WNL	Non-STEMI
Doherty 2022 [17]	24	M	T2DM	CP radiating to the left arm, sweating, tachycardia, 24-h history of vomiting & diarrhoea after alcohol consumption.	1212	STEMI
Dorcely 2021 [18]	61	M	T2DM	Nausea and CP	Negative	Inferior Q waves, unchanged from prior ECG.
Eliades 2014 [19]	71	F	T2DM	Abdominal pain, vomiting, confusion, & altered mental status.	5.67	STEMI
Fronczyk 2016 [20]	20	M	T1DM (diagnosed 10 years prior)	Weakness, low-grade fever, epigastric and abdominal pain, nausea, vomiting, burning retrosternal pain, and high blood glucose for 2 days.	WNL	STEMI
Gerede 2016 [21]	58	F	T1DM (diagnosed 33 years prior)	Confusion, nausea, vomiting, cough, sore throat, fever, dehydration, and an inclination to sleep.	5.43	Non-STEMI
Goto 2021 [22]	52	F	T2DM	Tachypnea, vomiting, decreased blood pressure, metabolic acidosis (euglycemic DKA), and myocardial ischemia.	324.5	Non-STEMI
Kaefer 2019 [23]	49	M	T1DM	DKA, nausea, and vomiting after inappropriate insulin therapy, troponin elevated to 142ng/l after 14 h post-admission and peaked at 142 ng/l at 38 h post-admission, no CP but ST elevation less than 2mm.	33, 14 h post-admission, peaked at 142, 38 h post-admission	STEMI
Landa 2021 [24]	48	M	T1DM	Diffuse pain and fatigue. He began feeling tired three days prior to the presentation with associated nausea and vomiting, but denied any fever, chills, or CP.	0.04	STEMI
Lee 2014 [25]	60	M	T2DM (New diagnosis)	Intermittent central CP radiated to left shoulder & neck, not affected by posture or exertion, with polydipsia, polyuria, weight loss, fever, sore throat.	0.77 μg/L	STEMI
Mhanna 2020 [26]	50	F	T1DM	Lethargy and low Blood pressure.	On admission: 0.21; second day: 20	STEMI
Odubanjo 2017 [27]	Middle age	NA	T1DM	Nausea & vomiting for 1 day.	On admission: 0.012, day 2: 7.3	STEMI
Oriot 2023 [28]	77	F	T2DM	CP, later respiratory distress & atrial fibrillation.	143	STEMI
Petersen 2023 [29]	28	M	T2DM	CP, diagnosed as NSTEMI after PCI.	NA	Non-STEMI
Umadat 2022 [30]	61	M	T1DM	Epigastric abdominal pain, nausea, & vomiting.	336	Nonspecific T-wave abnormalities
Wray 2020 [31]	19	M	T1DM	One day of nausea, vomiting, diffuse back & abdominal pain. Tender abdomen & back without trauma.	Undetectable	STEMI
Zughaib 2023 [32]	54	F	T2DM	Several days of typical CP, intractable nausea & vomiting. Repeated ED visits.	0.13	STEMI

ACS, Acute coronary syndrome; AFib, Atrial fibrillation; CP, Chest pain; DKA, Diabetic ketoacidosis; ED, Emergency department; F, Female; M, Male; NA, Not available; NSTEMI, Non-ST-elevation myocardial infarction; PCI, Percutaneous coronary intervention; SOB, Shortness of breath; STEMI, ST-elevation myocardial infarction; T1DM, Type 1 diabetes mellitus; T2DM, Type 2 diabetes mellitus; WNL, Within normal limits; ng/mL, Nanograms per milliliter; g/L, micrograms per liter.

Table 2

Table 2: Summary of Patient Echocardiographic, Angiographic, Laboratory, and Outcome Data

Author	EF%	Coronary Angiography	Home Medication	RBS (mg/dL)	pH	K (mmol/L)	HCO 3 (mEq/L)	HbA1c	Outcome
Baral 2021 [12]	15–20%	95% occlusion RCA at ostium, hazy filling defect in stent, patent LAD stent	Aspirin, clopidogrel, atorvastatin, metoprolol	541	7.36	5.8	16	NA	DES in RCA. Discharged after 32 days with EF 25–30%.
Baytu0̆11fan 2023 [13]	50%	Total occlusion of proximal LAD	Amlodipine/perindopril, metformin, dapagliflozin	171	7.03	5.03	5.6	9.60%	Successful PCI, insulin and bicarbonate, discharged day 6.
Bouzhir 2024 [14]	43%	No significant stenosis	NA	NA	NA	NA	NA	NA	Full LV recovery, EF 62% post-event.
Briggs 2022 [15]	40–45%	100% occlusion mid-circumflex, LAD 50%, RCA 60%	Rivaroxaban	392	7.2	NA	NA	NA	Successful PCI with stent, discharged post-op day 2.
Çakır 2012 [16]	NA	Not performed	None	476	7.07	2.1	NA	11.07%	Resolved with insulin, saline. Discharged stable.
Doherty 2022 [17]	49%	Occlusion of distal PLV branch, RCA thrombus	Olanzapine, fluoxetine	20.4 mmol/L	metabolic acidosis	3.2	15	64 mmol/mol	PCI and insulin therapy. Discharged day 4.
Dorcely 2021 [18]	Stress-induced abnormalities	Total occlusion mid-RCA, non-obstructive elsewhere	Empagliflozin, metformin, liraglutide, rosuvastatin, ezetimibe, omeprazole	84	metabolic acidosis	NA	17	8.30%	CP and DKA resolved. SGLT2 discontinued.
Eliades 2014 [19]	30%	Normal coronaries	Insulin, metformin	NA	NA	NA	NA	NA	EF improved to 45–50% in 10 days.
Fronczyk 2016 [20]	No contractility abnormalities	Normal coronaries	Insulin pump	33.1 mmol/L	7.18	Normal	5.6	10.37%	Pain and ECG normalized. Discharged day 11.
Gerede 2016 [21]	No wall motion abnormality	Non-obstructive lesions	Insulin, ramipril	522	7.159	6.3–6.4	5.8	NA	ECG/biochem normalized. Discharged day 8.
Goto 2021 [22]	NA	Coronary stenting	Empagliflozin, sitagliptin, ezetimibe, rosuvastatin, clopidogrel	178	6.84	-	2.1	NA	ICU, VA-ECMO, CRRT. Discharged ICU day 15.
Kaefer 2019 [23]	Apical dyskinesia	No occlusion	-	38.9 mmol/L	6.93	-	2.7	10.4	Angioplasty, stenting.
Landa 2021 [24]	Fair LV function	NA	Insulin only	952	6.94	7.6	<5	10.4	ICU insulin drip.

Table 3

Table 2 (continued): Summary of Patient Echocardiographic, Angiographic, Laboratory, and Outcome Data

Author	EF%	Coronary Angiography	Home Medication	RBS (mg/dL)	pH	K (mmol/L)	HCO 3 (mEq/L)	HbA1c	Outcome
Lee 2014 [25]	Normal	50% LCX obstruction	NA	35.6 mmol/L	7.25	NA	13	17%	ICU admission, abscess drainage, 10-week stay.
Mhanna 2020 [26]	60–65% to 25%	Normal coronaries	NA	1637	6.8	7.6	2.7	NA	Extubated, Life Vest, outpatient insulin.
Odubanjo 2017 [27]	Normal	Non-obstructive	NA	565	6.99	5.8	<5	NA	DKA resolved. Treated myocarditis. Discharged day 4.
Oriot 2023 [28]	41%	Triple-vessel disease, LAD stented	Metformin, nebivolol, statin, aspirin, spironolactone, citalopram	247	7.2	NA	6.1	7.80%	Discharged day 22, basal insulin.
Petersen 2023 [29]	NA	LAD NSTEMI	Full diabetes and cardiac regimen	154	7.048	5.8	<7	11.1	ICU stay, DKA resolved in 5 days, outpatient follow-up.
Umadat 2022 [30]	NA	99% LAD, 90% diagonal	Aspirin, statin, insulin pump	326	7.39	5	14	8%	Post-DKA PCI with stenting. Discharged on pump.
Wray 2020[31]	NA	Normal coronaries	NA	700	7.09	5.7	9.07	NA	PCI showed normal coronaries. Discharged day 3.
Zughaib 2023 [32]	NA	Ulcerated RCA plaque, mild LAD/LCX disease	Metformin, Empagliflozin	135	NA	3.2	NA	NA	PCI and stent. Resolved CP and DKA. Discharged.

DES, drug-eluting stent; EF, ejection fraction; RBS, random blood sugar; RCA, right coronary artery; LAD, left anterior descending artery; LCX, left circumflex artery; PLV, posterior left ventricular branch; NSTEMI, non–ST-elevation myocardial

infarction; PCI, percutaneous coronary intervention; DKA, diabetic ketoacidosis; VA-ECMO, veno-arterial extracorporeal membrane oxygenation; SGLT2, sodium–glucose cotransporter-2 inhibitor; T1DM, type 1 diabetes mellitus; HbA1c,

hemoglobin A1c; NA, not available; mg/dL, milligrams per deciliter; mmol/L, millimoles per liter; mEq/L, milliequivalents per liter; pH, potential of hydrogen; HCO3, bicarbonate; SC, subcutaneous; BID, twice daily; TIMI, Thrombolysis In Myocardial Infarction flow grade; GRACE, Global Registry of Acute Coronary Events; CP, chest pain; AFib, atrial fibrillation; ICU, intensive care unit.

six patients (28.5%) were reported to be on SGLT2 inhibitors, as shown in Table 1.

Discussion

The relationship between DKA and ACS is bidirectional, with each condition capable of precipitating and influencing the pathogenesis of the other. ACS can precipitate DKA through a stress response, that occurs due to the release of significant amount of counter-regulatory hormones, such as epinephrine, glucagon and cortisol [²³]. This hormonal surge raises blood glucose level, antagonizes the action of insulin, promotes fat breakdown, facilitates the production of ketones and eventually triggers the development of DKA [³³]. On the other hand, DKA can lead to the occurrence of myocardial ischemic damage by a multitude of mechanisms, including free fatty acids accumulation, metabolic acidosis and counter-regulatory hormones [³⁴].

When acidosis of the blood occurs, it leads to the acceleration of intracellular ionic movement of calcium $\text{Ca}^{2+}$, resulting in the accumulation of $\text{Ca}^{2+}$ inside the myocardial cells. This excess intracellular $\text{Ca}^{2+}$ facilitates the process of protein breakdown, known as proteolysis [³⁵]. This process is accompanied by a reduced ability of calcium to bind to the contractile proteins in the heart muscle—an effect caused by acidosis—which ultimately leads to impaired pump function, a phenomenon known as myocardial stunning. Ultimately, the combined effect of proteolysis and myocardial stunning results in myocardial injury leading to an increase in the serum cardiac troponin level [³⁶]. Moreover, acute decompensation of diabetes is associated with a sudden increase in the levels of counter-regulating hormones, such as epinephrine, cortisol, and glucagon, to try to increase the blood glucose level and provide energy during exacerbation [³⁷,³⁶]. However, these hormones can significantly increase the workload and metabolic demand of the cardiac muscle. In patients who already have coronary artery disease (CAD), this increase in the myocardial oxygen demand can induce the occurrence of ACS [³⁷]. Additionally, acute diabetic decompensation such as DKA is associated with a significant increase in the level of circulating free fatty acids (FFA). The excess FFA can be incorporated into the myocytes of the heart muscle, forming micelle-like formations that weaken and destabilize the cell membrane [³⁸]. Additionally, the severe lack of insulin, associated with increased ketone bodies and FFA, results in dysfunctional glucose uptake by the cardiac muscles and promotes the alternative uptake of FFA as the main energy source [³⁸,³⁷].

Generally, the interpretation of cardiac biomarkers, such as troponin, in DKA patients can be challenging, as ACS may either initiate the onset of DKA or arise as a consequence of it [³¹]. Additionally, the increase in troponin associated with DKA patients could be due to either the presence of underlying CAD disease, which is exacerbated by metabolic stress and acidosis, or to the infiltration of myocytes by the excess FFA associated with the severe insulin deficiency [³⁹,⁴⁰]. This could explain the absence of coronary abnormalities on some patients despite a rise in troponin. Moreover, our analysis revealed that both type 1 and type 2 diabetes mellitus (T1DM and T2DM) were present almost equally (42.8% vs. 57.1%), underscoring that neither diabetes phenotype is exempt from the risk of ACS during DKA. This demographic pattern aligns with established epidemiological data for acute coronary syndrome, where middle-aged males represent the highest-risk population, suggesting that traditional ACS risk factors may contribute to the concurrent presentation. T1DM patients are at risk of acquiring DKA due to their absolute insulin deficiency. In contrast, T2DM patients suffer from relative insulin deficiency, which is exacerbated by the rise in counter-regulatory hormones in cases of severe events, such as infection or stress [⁴¹,⁴²].

Clinical presentation of patients included a wide variety of symptoms and signs, such as chest pain, abdominal pain, nausea, vomiting, fatigue, altered mental status, hypertension, hypotension and tachypnea. The most common symptom presented was chest pain, however it should be noted that it was absent in nearly half of the patients. Our analysis demonstrated that some patients-mainly old patients- did not report chest pain, despite having high troponin and coronary artery abnormalities. This could be attributed to the sensory neuropathy associated with DM, which can decrease the pain perception of these patients, preventing the sensation of chest pain during the anginal attack [⁴³]. The absence of chest pain in these patients poses a critical diagnostic challenge, potentially leading to delayed recognition of ACS. This atypical presentation, combined with the metabolic focus required for DKA management, creates a scenario where cardiac evaluation may be deprioritized. Healthcare providers managing DKA patients should maintain heightened awareness for silent myocardial ischemia, particularly in patients with cardiovascular risk factors, unexplained hemodynamic instability, or ECG changes that cannot be attributed solely to electrolyte abnormalities [⁴⁴].

This fact complicates the process of ischemic heart disease diagnosis in DM patients, as in these patients the diagnosis is mainly based on the presence of non-specific angina-equivalent symptoms such as dyspnea and dizziness along with abnormal glucose level [⁴⁵].

In the majority of cases, DKA was associated with an increased blood glucose level; however, in six cases of individuals receiving SGLT2i drugs at home, euglycemic DKA with a blood glucose level below 250 mg/dL was a prominent feature. Notably, euglycemic DKA occurred in 6 patients (28.6%), which contrasts sharply with the general DKA population, where euglycemic presentations account for only 2.6-3.2% of cases [48). This atypical presentation of DKA can delay the diagnosis and obscure the coexisting myocardial damage [⁴⁶]. SGLT2i drugs can cause euglycemic DKA by enhancing urinary glucose excretion, thereby lowering blood glucose levels, which in turn can lead to a reduction in insulin secretion or dosing. Additionally, SGLT2i can also stimulate glucagon release, which, when accompanied by a diminished insulin level, promotes a shift in metabolism towards lipolysis, leading to increased ketone production [⁴⁷]. Thus, the reliance on high blood glucose level alone to diagnose DKA is not a proper clinical practice, especially in patients who are receiving oral anti-diabetic drugs [⁴⁸]. Additionally, this also underscores the importance of heightened monitoring of DKA patients who are on SGLT2i drugs and the need for immediate discontinuation of the drug in cases of DKA suspicion [⁴⁹]. The association between SGLT2 inhibitors and euglycemic DKA in our cohort raises important ethical considerations regarding informed consent and risk-benefit communication. Clinicians prescribing SGLT2 inhibitors must ensure that patients understand the risk of DKA, particularly those with additional risk factors, such as insulin deficiency, acute illness, or planned procedures. The cardiovascular benefits of SGLT2 inhibitors are well-established, but the potential for life-threatening ketoacidosis, especially with atypical presentations, necessitates careful patient selection, education, and monitoring protocols [⁵⁰].

Regarding the analysis of the treatments provided, all of the included cases received the standard DKA therapy, which included aggressive fluid resuscitation, electrolyte correction and IV insulin infusion [⁵¹]. Only one patient who was associated with severe metabolic acidosis (PH>7, HCO3=5.6) was indicated to receive intravenous bicarbonate administration. This aligns with current recommendations which deter from routine use of bicarbonate in the treatment of DKA, due to the risk of cerebral edema [⁵²]. Hence, the use of bicarbonate should be solely exclusive in cases with refractory metabolic acidosis that failed to be corrected with standard DKA therapy [⁵³]. Furthermore, 18 cases underwent PCI for the treatment of ACS, of which five patients had normal coronaries suggesting either demand ischemia or coronary vasospasm.

In our review, despite the severity of the presentation of some cases, the majority of the patients survived the acute phase and no mortality was reported. However, some cases, particularly patients with severe metabolic acidosis and elevated cardiac biomarkers, required prolonged hospitalization after undergoing the PCI procedure or after admission to the ICU. Patients with euglycemic DKA on SGLT2i drugs suffered from delayed diagnosis which might have influenced their clinical course and prolonged their hospital stay. Furthermore, whereas no mortality was reported in the selected cohort, the lack of long-term follow up for potential complications limit our understanding of the true long-term prognosis of those patients.

Another important observation is the significantly high HbA1c level in the reported cases, indicating in the process significantly poorly controlled diabetic status. In literature, patients with HbA1c levels above 9% are significantly more likely to develop DKA compared to those with lower levels with one study reporting that this risk could be five times more likely compared to patients with HbA1c levels between 7% and 8% [⁵⁴]. Moreover, the level of HbA1c could correlate to the severity level of DKA as well as the severity level of the associated metabolic acidosis [⁵⁵]. Furthermore, elevated HbA1c level alone is an independent risk factor for ACS and is associated with significantly worse outcomes and prognosis [⁵⁶]. This highlights the importance of good monitoring of diabetic patients, especially patients who are at high risk for CAD [⁵⁷]. Proper control of blood glucose level over the long-term can be beneficial in preventing the exacerbating events of DKA and ACS [⁵⁸]. Our findings advance clinical understanding in several key areas: First, the prevalence of euglycemic DKA in concurrent presentations is substantially higher than in general DKA populations, suggesting unique pathophysiological or pharmacological mechanisms. Second, the high frequency of atypical presentations without chest pain necessitates revised diagnostic protocols that include routine cardiac evaluation in DKA patients with cardiovascular risk factors. Third, the prominence of SGLT2 inhibitor-associated cases establishes this drug class as a key risk factor requiring enhanced monitoring and patient education. These insights should inform updated clinical guidelines for both emergency department protocols and SGLT2 inhibitor prescribing practices.

This systematic review aggregated all the known case reports on this subject in the literature, in order to provide a comprehensive, up-to-date assessment of the relationship between DKA and ACS. However, this study still has some limitations that should be considered, including the inherent biases of the study design. A case report was the best available option to address this issue, given the rarity of concurrent DKA and ACS, higher-level evidence, such as cohort studies or RCTs, is not feasible. Additional limitations included heterogeneity in the diagnostic criteria and management protocols, publication bias, a small sample size, and a lack of control groups and long-term follow-up. Thus, larger-scale observational prospective or retrospective studies are required to address these limitations.

Conclusion

In conclusion, our review highlights the bidirectional association between DKA and ACS. Clinicians should maintain a low threshold for suspicion of concurrent DKA in patients presenting with ACS, particularly when traditional DKA symptoms (polyuria, polydipsia, nausea) are absent. Routine ECG, troponin assessment and echocardiography are essential to differentiate between true ischemia and metabolic mimicry. Ultimately, early recognition, standard treatment and use of coronary angiography are pivotal to optimizing outcomes.