Clinical findings of acute necrotizing esophagitis complicated by diabetic ketoacidosis

Abstract

Acute necrotizing esophagitis (ANE) is a rare and potentially life-threatening complication of diabetic ketoacidosis (DKA). While its association with DKA is established, specific clinical characteristics that predict ANE in DKA patients remain less understood. This study aimed to identify these characteristics by analyzing data from 30 DKA patients admitted from January 2018 to September 2022. Seven patients in this study presented with ANE, forming the ANE group. The remaining 23 constituted the non-ANE group. We compared the clinical parameters and computed tomography (CT) between the groups. The mean age of participants was 57.7 ± 20.4 years, and their mean HbA1c was 11.1 ± 3.3%. Notably, ethanol intake was significantly higher in the ANE group (44.4 ± 25.4 g/day) compared to the non-ANE group (6.8 ± 14.0 g/day; p = 0.013). Additionally, sodium-glucose transport protein 2 inhibitor use was significantly more prevalent in the ANE group (p = 0.013). Gastrointestinal symptoms were also significantly more pronounced in the ANE group, with vomiting occurring in 85.7% of patients compared to only 13.0% in the non-ANE group. Admission CT scans revealed further distinguishing features, with the ANE group showing significantly higher rates of esophageal wall thickening, intra-esophageal effusion, and calcification of the celiac artery origin (p < 0.0001, 0.0038, 0.0038, respectively). In conclusion, our study suggests that heavy alcohol consumption and strong gastrointestinal symptoms in DKA patients warrant a heightened suspicion of ANE. Early consideration of CT or upper gastrointestinal endoscopy is recommended in such cases.

ACUTE NECROTIZING ESOPHAGITIS (ANE) is a rare and potentially life-threatening disease characterized by necrotic lesions of the esophageal mucosa, referred to as “black esophagus” [1]. It arises from ischemic necrosis due to a combination of factors: impaired blood flow, mucosal damage, increased intra-esophageal pressure from refluxed gastric contents or vomiting, and a weakened mucosal barrier system caused by malnutrition or physical irritation. Diabetic ketoacidosis (DKA) is a frequent association with ANE [2]. Treatment for ANE typically involves addressing the underlying condition, administering proton pump inhibitors, and implementing bowel rest with fasting [3]. We previously reported a case of ANE complicated by hyperglycemia [4]. Since then, the Division of Diabetes, Endocrinology and Metabolism at Kawasaki Medical School has encountered a significant number of DKA cases complicated by ANE between January 1, 2018, and September 30, 2022. While elevated blood sugar levels have been documented in ANE patients, few studies have investigated the clinical characteristics of DKA patients to identify those at higher risk of developing ANE [5]. This study aims to bridge that gap by analyzing the clinical features of DKA patients with ANE.

Materials and Methods

Study population and patient preparation

This retrospective study was conducted at the Division of Diabetes, Endocrinology and Metabolism, Kawasaki Medical School. Between January 1, 2018, and September 30, 2022, we analyzed data from adult subjects admitted to our department for DKA. The study was approved by the Institutional Review Board of Kawasaki Medical School (No. 5891-00). Consent was obtained from participants through an opt-out mechanism on the Kawasaki Medical School website. A flowchart summarizing patient selection is shown in Fig. 1.

Fig. 1

The flowchart of the participants and exclusions in this study.

We initially identified 30 patients with DKA admitted to the Kawasaki Medical School Hospital during the specified period. DKA was defined as a history of diabetes diagnosis, blood pH <7.30, anion gap >12.0, and the presence of ketosis. ANE was diagnosed via upper gastrointestinal endoscopy revealing a “black esophagus” or through characteristic symptoms like coffee-ground emesis combined with endoscopic findings suggestive of a healing “black esophagus” (Fig. 2). Among the 30 DKA patients, seven with endoscopic confirmation of ANE were categorized as the ANE group. The remaining 23 patients, who lacked any clinical suspicion of ANE, formed the non-ANE group. Notably, only three individuals in the non-ANE group underwent upper gastrointestinal endoscopy.

Fig. 2

Upper gastrointestinal (UGI) endoscopic findings in the acute necrotizing esophagitis (ANE) group before and after treatment (left and right panels, respectively). (A–F) Six patients in the ANE group had black esophagus at admission. (G) Case seven had esophageal perforation on admission and underwent upper gastrointestinal endoscopy after healing of ANE.

Study design

This single-center retrospective analyzes clinical differences between the ANE and non-ANE groups using medical records, including diabetes-related parameters and admission computed tomography (CT) scan. Demographic data like age, sex, height, weight, body mass index (BMI), medical history (alcohol consumption and smoking history), and vital signs (blood pressure, heart rate, and body temperature) were retrieved from admission records. All patients underwent blood tests and CT scans on admission. The Aquilion Prime SP system by Canon Medical Systems Inc. (Tochigi) was used for CT imaging.

All patients in the ANE group and 20 in the non-ANE group underwent admission CT scans, which were assessed for the three key features: esophageal wall thickening, intra-esophageal fluid retention (presence of absorbent fluid within the esophageal lumen, particularly in the lower segment), and calcified lesions at the celiac artery origin (high-density lesions on axial and sagittal images, extending from the aorta towards the celiac artery and the superior mesenteric artery trunks) (Fig. 3). Two physicians independently reviewed all CT findings.

Fig. 3

Characteristic findings in ANE patients. (A) Black esophagus on upper gastrointestinal endoscopy. (B) Esophageal wall thickening on computed tomography. (C) Intra-esophageal effusion. (D) Calcification of the celiac artery origin.

To evaluate the clinical characteristics of ANE, we identified the period of fasting prior to admission, upper gastrointestinal endoscopy findings (black esophagus, extent of ANE, gastric mucosal atrophy, erosion of gastric mucosa, and erosion of duodenal mucosa), and date of retest of upper gastrointestinal endoscopy, counting from the date of admission, in the medical records.

Statistical analysis

All statistical analyses were performed using JMP version 16.0.2 (SAS Institute Inc., North Carolina, USA). Microsoft Excel for Mac version 16.71 (Microsoft Co., WA, USA) was used to create tables. Data are expressed as mean and standard deviation. The primary objective of the study was to identify differences in clinical characteristics between the ANE and non-ANE groups. To achieve this, Mann-Whitney U tests were used for continuous variables, while chi-square tests were employed for categorical variables. Subjects with type 1 diabetes in the ANE group, subjects with type 1 diabetes in the non-ANE group, subjects with type 2 diabetes in the ANE group, and subjects with type 2 diabetes in the non-ANE group were divided into groups to evaluate differences in clinical parameters between subjects with type 1 and type 2 diabetes. The Tukey method and chi-square test were used to analyze the results. Upper gastrointestinal endoscopy was performed in all subjects in the ANE group.

Results

Clinical characteristics of this study subjects

Table 1 summarizes the clinical parameters of the study participants. Subjects had an average age of 57.7 ± 20.4 years, and glycated hemoglobin (HbA1c) levels of 11.1 ± 3.3%. Notably, four participants in the ANE group and seven in the non-ANE group reported habitual alcohol consumption, resulting in a significantly higher proportion of heavy drinkers in the ANE group (p < 0.01). Furthermore, average ethanol intake among these habitual drinkers was significantly higher in the ANE group (77.8 g/day) compared to the non-ANE group (22.3 g/day). Of these individuals, one in the ANE group and four in the non-ANE group had already stopped drinking before admission. While γ-glutamyl transpeptidase levels were slightly elevated in the ANE group (94.0 ± 42.8 U/L) compared to the non-ANE group (48.7 ± 23.6 U/L; p = 0.067), no significant differences were observed in HbA1c, ketosis, acidosis, or renal function. These findings suggest that the ANE group had a significantly higher history of alcohol consumption compared to the non-ANE group.

Table 1

Various clinical parameters in this study subjects

Parameters	All subjects (n = 30)	ANE group (n = 7)	Non-ANE group (n = 23)	p value
Male/female	15/15	2/5	13/10	0.18
Age (years)	57.7 ± 20.4	67.0 ± 15.8	54.9 ± 8.7	0.19
Body weight (kg)	58.2 ± 17.0	50.4 ± 9.4	60.6 ± 18.0	0.18
BMI (kg/m2)	22.5 ± 5.3	20.0 ± 2.8	23.3 ± 5.6	0.16
Brinkman index	300.2 ± 624.6	321.3 ± 727.9	293.7 ± 589.4	0.92
Percentage of habitual drinkers (n (%))	11 (36.7)	4 (57.1)	7 (30.4)	<0.01
Ethanol intake among habitual drinkers (g/day)	42.5 ± 50.3	77.8 ± 22.0	22.3 ± 16.7	0.08
Fasting period prior to hospitalization (days)	2.0 ± 3.6	4.9 ± 6.3	1.2 ± 1.6	<0.01
Type of diabetes (type 1/type 2)	10/20	1/6	9/14	0.20
Duration of diabetes mellitus (years)	13.2 ± 12.4	9.3 ± 7.7	15.0 ± 13.2	0.30
Length of hospital stay (days)	18.7 ± 11.6	29.7 ± 16.4	15.3 ± 6.6	<0.01
SGLT2 inhibitor use (n (%))	8 (23.3)	3 (42.9)	5 (13.0)	0.02
Systolic blood pressure (mmHg)	132.6 ± 23.6	124.1 ± 29.2	135.1 ± 21.0	0.30
Diastolic blood pressure (mmHg)	79.5 ± 19.9	80.7 ± 22.4	79.1 ± 19.0	0.86
Pulse rate (beats per minutes)	108.5 ± 19.1	103.9 ± 20.0	109.9 ± 18.5	0.48
Body temperature (°C)	36.7 ± 0.7	36.4 ± 0.6	36.8 ± 0.7	0.29
Blood glucose (mg/dL)	636.8 ± 251.9	508.1 ± 247.4	675.9 ± 250.9	0.13
HbA1c (%)	11.1 ± 3.3	10.2 ± 3.8	11.4 ± 0.7	0.40
Glycoalbumin (%)	37.1 ± 13.2	30.1 ± 11.1	39.1 ± 1.6	0.25
Urine ketone qualitative	2.2 ± 1.0	2.7 ± 0.5	2.1 ± 1.0	0.14
Total ketone body (μmol/L)	10,614.9 ± 5,289.0	10,874.3 ± 6,399.3	10,532.4 ± 4,880.1	0.89
Acetoacetic acid (μmol/L)	3,166.8 ± 1,683.9	3,309.1 ± 1,834.8	3,121.5 ± 1,630.3	0.81
3-hydroxybutyric acid (μmol/L)	7,423.2 ± 3,731.8	7,565.2 ± 4,610.1	7,380.0 ± 3,418.8	0.91
pH	7.17 ± 0.14	7.21 ± 0.11	7.16 ± 0.15	0.52
Anion Gap	25.2 ± 7.1	25.2 ± 5.5	25.3 ± 7.4	0.98
Serum osmotic pressure (mOsm/kg)	328.9 ± 26.9	326.6 ± 27.7	329.6 ± 26.6	0.80
Triglyceride (mg/dL)	185.8 ± 97.7	131.1 ± 63.1	204.9 ± 100.4	0.09
LDL-cholesterol (mg/dL)	115.7 ± 60.8	84.8 ± 39.4	125.0 ± 63.0	0.17
HDL-cholesterol (mg/dL)	44.9 ± 18.3	53.5 ± 16.7	42.4 ± 18.0	0.21
AST (U/L)	33.6 ± 24.9	28.0 ± 12.5	35.3 ± 27.5	0.52
ALT (U/L)	32.8 ± 24.9	29.4 ± 19.6	33.9 ± 26.3	0.69
γGTP (U/L)	59.2 ± 56.7	94.0 ± 69.1	48.7 ± 47.5	0.07
Urea nitrogen (mg/dL)	49.1 ± 33.4	42.3 ± 14.8	51.2 ± 37.1	0.55
Creatinine (mg/dL)	1.74 ± 1.19	1.75 ± 1.29	1.74 ± 1.16	0.98
CRP (mg/dL)	3.55 ± 5.64	3.87 ± 3.18	3.45 ± 6.22	0.87

Data are presented as mean ± standard deviation. ANE, acute necrotizing esophagitis; BMI, body mass index; SGLT2, sodium glucose cotransporter 2; LDL, low density lipoprotein; HDL, high density lipoprotein; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γGTP, gamma-glutamyl transpeptidase; CRP, C-reactive protein. Mann-Whitney U tests and chi-square test were used for analysis.

Gastrointestinal symptoms on admission in this study subjects

Upon admission, vomiting was the chief complaint for six (85.7%) patients in the ANE group, including four with black vomiting. Notably, only three (13%) patients in the non-ANE group presented with vomiting, and only one had black vomiting. Furthermore, the duration of fasting before hospitalization was significantly longer in the ANE group (4.9 ± 6.3 days) compared to the non-ANE group (1.2 ± 1.6 days) (Table 1). Table 2 details the upper gastrointestinal endoscopic findings in the ANE group. Additionally, both Table 2 and Fig. 2 provide a visual representation of these findings. Six patients had their first upper gastrointestinal endoscopy performed on the day of admission. Case 7 was diagnosed with ANE based on post-healing upper gastrointestinal endoscopy images and clinical information, as no upper gastrointestinal endoscopy was performed on admission due to esophageal perforation. Crucially, six patients in the ANE group exhibited the characteristic “black esophagus.” The extent of esophageal mucosal lesions varied: one case (Fig. 2G, Case 7) showed limited involvement in the lower esophagus, while others displayed lesions extending from the upper and middle to the lower esophagus. Notably, three ANE patients also presented with white moss on the esophageal mucosa (Fig. 2B, 2E, 2F, Cases 2, 5, and 6). Overall, six out of the seven ANE patients displayed gastric or duodenal mucosal involvement (Fig. 2F and 2G, Cases 6 and 7). Case five further revealed ileal necrosis on contrast-enhanced CT scan, highlighting the potential for additional gastrointestinal complications associated with ANE.

Table 2

Chief complaint and upper gastrointestinal (UGI) endoscopic findings in this subjects with ANE

Case	Age, Sex	Fasting period before admission (days)	Black vomiting	Range of ANE	Atrophy of the gastric mucosa	Erosions of the gastric mucosa	Disorders of the duodenal mucosa	Period after admission until retest with UGI endoscopy (days)	Treatment
1	57, Male	4	(–)	Cervical esophagus to EGJ	(–)	(+)	(+)	6	Bonoprazan, sodium alginate
2	79, Female	2	(+)	Middle esophagus to EGJ	(–)	(+)	(–)	14	Bonoprazan, sodium alginate
3	70, Male	3	(–)	Cervical esophagus to EGJ	(–)	(–)	(–)	6	Bonoprazan, sodium alginate
4	62, Male	19	(+)	Upper esophagus to lower esophagus	(–)	(–)	(+)	7	Bonoprazan, sodium alginate
5	81, Male	2	(–)	Middle esophagus to EGJ	(–)	(+)	(+)	25	Omeprazole
6	79, Female	2	(+)	Middle esophagus to lower esophagus	(+)	(–)	(+)	9	Omeprazole
7	41, Male	2	(+)	Lower esophagus	(+)	(+)	(–)	15	Omeprazole

ANE, acute necrotizing esophagitis; EGJ, esophagogastric junction; UGI, upper gastrointestinal

Characteristic CT findings in ANE

We evaluate the presence of specific imaging findings on admission CT scans: esophageal wall thickening (Fig. 3B), intra-esophageal effusion (Fig. 3C), and calcification of the celiac artery origin (Fig. 3D). The results are summarized in Table 3. In the ANE group, esophageal wall thickening was detected in 85.7% of patients, intra-esophageal effusion in 57.1%, and calcification of the celiac artery origin in 85.7%. All three findings were significantly more prevalent in the ANE group compared to the non-ANE group (p < 0.0001, p = 0.0038, and p = 0.017, respectively).

Table 3

CT findings in this study subjects

CT finding	ANE group (n = 7)	Non-ANE group (n = 20)	p value
Esophageal wall thickening (n (%))	6 (85.7)	0 (0)	<0.01
Intraesophageal effusion (n (%))	4 (57.1)	1 (5.0)	<0.01
Calcification of the celiac artery origin (n (%))	6 (85.7)	7 (35.0)	0.02

CT, computed tomography; ANE, acute necrotizing esophagitis. Chi-square test was used for analysis.

Treatment of this study subjects

The average length of hospital stay was significantly longer in the ANE group (29.7 ± 16.4 days) compared to the non-ANE group (15.3 ± 6.6 days) (p = 0.0027). All participants in this study received intravenous insulin therapy. Fortunately, no deaths occurred during the study period. This includes fasting, mucosal protectants (sodium alginate) and acid secretion inhibitors (bonoprazan, omeprazole) (Table 2), and extracellular fluid infusion. Mucoprotective agents were administered orally after diagnosis. Acid secretagogues were administered intravenously during the fasting period and orally once oral intake was resumed. In one case, (case five) involving ileal necrosis, additional surgical treatment of the involved intestine was necessary followed by the same ANE regimen. Approximately 1–2 weeks after initiating fasting, an upper gastrointestinal endoscopy was performed to assess the healing of the “black esophagus.” If satisfactory progress was observed, a liquid diet was then introduced, gradually progressing to a more solid diet over the following weeks.

Clinical background of subjects with type 1 or type 2 diabetes

Table 4 summarizes the clinical background of subjects with both type 1 and type 2 diabetes. Importantly, no significant differences were observed in age, sex, BMI, HbA1c, or degree of ketoacidosis between the ANE and non-ANE groups within each diabetes type. For both type 1 and type 2 diabetic subjects, habitual drinkers in the ANE group consumed significantly more ethanol compared to their non-ANE counterparts (p < 0.01). Furthermore, within each diabetes type, the ANE group exhibited characteristics suggestive of a more severe presentation. They experienced longer hospital stays and presented with emesis, including black vomiting, as the chief complaint more frequently than the non-ANE group (p = 0.02 for type 1, and p < 0.01 for type 2, respectively).

Table 4

Clinical background of subjects with type 1 and type 2 diabetes

Parameters	T1DM subjects (n = 10)		T2DM subjects (n = 20)		p value
	ANE group (n = 1)	Non-ANE group (n = 9)	ANE group (n = 6)	Non-ANE group (n = 14)
Male/female	1/0	2/7	4/2	8/6	0.19
Age (years)	41	56.9 ± 24.4	71.3 ± 10.1	53.6 ± 20.7	0.29
Body weight (kg)	62.2	53.1 ± 12.9	48.5 ± 9.5	65.4 ± 20.2	0.15
BMI (kg/m2)	22.3	21.7 ± 1.8	19.6 ± 2.2	24.3 ± 1.4	0.33
Brinkman index	0	176.7 ± 220.5	375.0 ± 270.0	368.9 ± 176.8	0.86
Percentage of habitual drinkers (n (%))	1 (100)	1 (11.1)	3 (50.0)	6 (42.9)	0.18
Ethanol intake in habitual drinkers (g/day)	24.0	3.0	95.7 ± 73.6	25.5 ± 21.8	<0.01
Fasting period prior to admission (days)	2.0 ± 3.3	0.8 ± 1.1	5.3 ± 1.3	1.4 ± 0.9	0.07
Length of hospital stay (days)	25	12.2 ± 3.2	30.5 ± 19.2	17.3 ± 7.8	0.02
SGLT2 inhibitor use (n (%))	0 (0)	3 (33.3)	3 (50.0)	2 (14.3)	0.34
Vomiting (n (%))	1 (100)	1 (11.1)	5 (83.3)	2 (14.3)	<0.01
Black vomiting (n (%))	1 (100)	0 (0)	3 (50.0)	1 (7.1)	<0.01
Blood glucose (mg/dL)	397	768.9 ± 296.7	526.7 ± 265.6	616.1 ± 206.3	0.22
HbA1c (%)	10.2	9.8 ± 1.1	10.2 ± 4.1	12.5 ± 3.8	0.25
Urine ketone qualitative	3	2.1 ± 1.2	2.7 ± 0.5	2.1 ± 1.0	0.52
Total ketone body (μmol/L)	14,360	10,797.6 ± 5,552.3	10,293.3 ± 7,382.1	10,348.9 ± 4,798.0	0.92
pH	7.17	7.08 ± 0.19	7.22 ± 0.13	7.21 ± 0.11	0.20
Anion Gap	26.3	28.0 ± 6.9	24.9 ± 6.7	23.5 ± 7.7	0.57
Serum osmotic pressure (mOsm/kg)	320	333.2 ± 25.0	327.7 ± 32.6	327.3 ± 29.3	0.95

Data are presented as mean ± standard deviation. ANE, acute necrotizing esophagitis; BMI, body mass index; SGLT2, sodium glucose cotransporter 2. ANOVA and chi-square test were used for analysis.

Discussion

This clinical study sheds valuable light on the clinical characteristics of DKA patients with ANE. Compared to the non-ANE group, ANE patients displayed significantly higher alcohol consumption and vomiting frequency. Additionally, we identified characteristic CT findings associated with ANE: esophageal wall thickening, intra-esophageal effusion, and calcification of the celiac artery origin. These findings suggest the presence of severe gastrointestinal mucosal damage in ANE.

It is important to note that ANE is relatively rare, occurring in only 0.28% of patients undergoing upper gastrointestinal endoscopy [6]. Prior research has established that 90% of ANE patients also experience hyperglycemia, and several risk factors have been identified, including vascular disease, hypertension, chronic kidney disease, cancer, malnutrition, diabetes, and alcohol abuse [7]. The underlying mechanism of DKA-induced ANE is thought to involve a combination of hyperglycemia-induced fluid loss and gastroesophageal reflux due to DKA-induced impaired gastric peristalsis [8]. Notably, our study found no difference in the degree of hyperglycemia, ketosis, or acidosis between the ANE and non-ANE groups in DKA patients. However, significant differences were observed in drinking history and the use of sodium-glucose transport protein-2 (SGLT2) inhibitors. This finding supports the notion that ANE arises from a combination of DKA and additional factors that exacerbate dehydration. While several reports have linked SGLT2 inhibitor use to ANE complications, the causal relationship remains unclear [4, 9, 10]. Our study uniquely demonstrates a statistically significant association between ANE and SGLT2 inhibitor use. In light of the increasing popularity of SGLT2 inhibitors, further research with larger patient cohorts is necessary to solidify this connection. Furthermore, the presence of intra-esophageal fluid retention observed on CT scans in ANE patients may reflect impaired peristalsis and gastroesophageal reflux associated with DKA.

Previous case series at other institutions suggest a link between ANE and duodenal ulcers, as both organs receive blood supply from the celiac artery [11]. To explore this connection, we used CT scans to investigate whether calcification of the celiac artery origin was associated with ANE. Notably, we found a high prevalence of calcification in the ANE group, particularly in patients with multiple duodenal ulcers. This finding strongly suggests that celiac artery narrowing and impaired blood flow may contribute to ANE development. Therefore, DKA patients with calcification of the celiac artery origin should be evaluated for gastrointestinal symptoms and, potentially undergo upper gastrointestinal endoscopy.

In cases of DKA complicated by ANE, there is no certain opinion as to which occurs first, ANE or DKA. This uncertainty stems from the fact that DKA and ANE are often diagnosed upon patient presentation. Previous research has shown that around 90% of ANE patients also have hyperglycemia, and hyperglycemia-induced hypotension can exacerbate lower esophagus ischemia [12, 13]. Interestingly, our study identified a significantly longer fasting duration in the ANE group compared to the non-ANE group. Additionally, six ANE patients reported vomiting, including black vomiting, as their chief complaint. These findings led us to hypothesize that DKA-related vomiting, a symptom in roughly half of DKA patients, [14], might act as a mechanical trigger for ANE. However, prolonged fasting itself is a risk factor for ketosis [15], suggesting the possibility that ANE-related black vomiting might have preceded the onset of DKA, triggering the condition. Due to the retrospective nature of our study, acquiring detailed information about symptom onset proved challenging. Therefore, recognizing ANE early remains crucial, especially in DKA cases with prolonged fasting, considering the bidirectional relationship between DKA and ANE, with ANE acting as a poor prognostic factor for DKA.

Treatment for ANE focuses on supportive care and addressing the underlying conditions. This typically includes antacids, fasting, and fluid management to alleviate digestive symptoms, while also addressing hyperglycemia in DKA patients. Antimicrobial therapy is only recommended in confirmed cases of esophageal perforation or infection [16]. Notably, our study participants (except for one case of ileal necrosis) did not receive antibiotics, as ANE improvement was observed solely with diabetes management and supportive measures.

Several limitations warrant consideration in interpreting the study findings. First, this single-center retrospective design limits its generalizability, as results may vary across different populations and healthcare environments. Second, the lack of endoscopies in most non-ANE participants makes comparisons between groups less conclusive, as mild gastrointestinal symptoms often don’t require endoscopic evaluations. Finally, this relatively small sample size and potential recruitment bias due to a higher proportion of ANE patients during the observation period might not accurately reflect the true prevalence of ANE. Further epidemiological studies are needed to provide a more robust estimate. is not usually performed. Finally, this study had a small number of participants, and there happened to be many ANE patients during the observation period. It’s important to acknowledge that biopsies were not performed for ANE diagnosis due to the risk of esophageal perforation. Consequently, potential infectious etiologies like cytomegalovirus, candidiasis, and herpes simplex virus cannot be entirely ruled out. However, many participants underwent blood tests and cultures to assess for systemic infections.

In conclusion, this study highlights the association between heavy alcohol consumption, severe gastrointestinal symptoms, and ANE in DKA patients Additionally, it identified characteristic CT findings associated with ANE. For heavy alcohol drinkers with DKA and prominent gastrointestinal symptoms, consider ANE as a potential complication and proceed with CT scans and endoscopy for a definitive diagnosis.

Acknowledgement

In the diagnosis and treatment of ANE, we were greatly indebted to Dr. Motoyasu Osawa in the Department of Gastroenterology, Kawasaki Medical School. Some of the results of this study were reported at the 128th Chugoku Regional Branch Meeting of the Japanese Society of Internal Medicine and received the Young Investigator Award. This manuscript was edited by Elsevier Language Editing Services.

Conflict of Interests

H.K. has received honoraria for lectures, received scholarship grants, and received research grant from Novo Nordisk Pharma, Sanofi, Eli Lilly, Boehringer Ingelheim, Taisho Pharma, Sumitomo Dainippon Pharma, Takeda Pharma, Ono Pharma, Daiichi Sankyo, Mitsubishi Tanabe Pharma, Kissei Pharma, MSD, AstraZeneca, Astellas, Novartis, Kowa and Abbott. K.K. has been an advisor to, received honoraria for lectures from, and received scholarship grants from Novo Nordisk Pharma, Sanwa Kagaku, Takeda, Taisho Pharma, MSD, Kowa, Sumitomo Dainippon Pharma, Novartis, Mitsubishi Tanabe Pharma, AstraZeneca, Boehringer Ingelheim, Chugai, Daiichi Sankyo and Sanofi. T.M. is a member of Endocrine Journal’s Editorial Board.

Author Contributions

Y.I. designed the study. Y.I., T.K., T.I., S.M., T.S., M.O., H.T., H.I., J.S., Y.F., Y.K., F.T., M.S., S.N., T.M., and H.K. collected data. Y.I. analyzed the data. Y.I., T.K., F.T., M.S., S.N., T.M. and H.K. contributed to discussion. K.K. supervised the project. Y.I. wrote the manuscript. H.K. reviewed and edited the manuscript.

References

• 1   Gurvits  GE (2010) Black esophagus: acute esophageal necrosis syndrome. World J Gastroenterol 16: 3219–3225.
• 2   Dias  E,  Santos-Antunes  J,  Macedo  G (2019) Diagnosis and management of acute esophageal necrosis. Ann Gastroenterol 32: 529–540.
• 3   Kondo  T (2018) Acute necrotizing esophagitis. Cleve Clin J Med 85: 912–913.
• 4   Iwamoto  Y,  Tatsumi  F,  Isobe  H,  Katakura  Y,  Shimoda  M, et al. (2021) Acute and extremely severe necrotic esophagitis accompanied by hyperglycemic hyperosmolar syndrome in a subject with type 2 diabetes mellitus. J Diabetes Investig 12: 1925–1926.
• 5   Yasuda  H,  Yamada  M,  Endo  Y,  Inoue  K,  Yoshiba  M (2006) Acute necrotizing esophagitis: role of nonsteroidal anti-inflammatory drugs. J Gastroenterol 41: 193–197.
• 6   Augusto  F,  Fernandes  V,  Cremers  MI,  Oliveira  AP,  Lobato  C, et al. (2004) Acute necrotizing esophagitis: a large retrospective case series. Endoscopy 36: 411–415.
• 7   Kitawaki  D,  Nishida  A,  Sakai  K,  Owaki  Y,  Nishino  K, et al. (2022) Gurvits syndrome: a case of acute esophageal necrosis associated with diabetic ketoacidosis. BMC Gastroenterol 22: 277.
• 8   Jeyalingam  T,  Shimamura  Y,  Teshima  C (2018) Endoscopic resolution of black esophagus with treatment of diabetic ketoacidosis. Clin Gastroenterol Hepatol 16: e68–e69.
• 9   Greco  S,  Giovine  A,  Rocchi  C,  Resca  R,  Bigoni  R, et al. (2023) Acute esophageal necrosis as a rare complication of metabolic acidosis in a diabetic patient: a case report. Am J Case Rep 24: e939624.
• 10   Avramidou  D,  Violatzi  P,  Zikoudi  DG,  Mourseloglou  A,  Panagaris  S, et al. (2022) Acute esophageal necrosis complicating diabetic ketoacidosis in a patient with type II diabetes mellitus and excessive cola consumption: a case report. Diabetol Int 13: 315–319.
• 11   Gurvits  GE,  Cherian  K,  Shami  MN,  Korabathina  R,  El-Nader  EM, et al. (2015) Black esophagus: new insights and multicenter international experience in 2014. Dig Dis Sci 60: 444–453.
• 12   Inayat  F,  Hurairah  A,  Virk  HU (2016) Acute esophageal necrosis: an update. N Am J Med Sci 8: 320–322.
• 13   Rehman  O,  Jaferi  U,  Padda  I,  Khehra  N,  Atwal  H, et al. (2021) Epidemiology, pathogenesis, and clinical manifestations of acute esophageal necrosis in adults. Cureus 13: e16618.
• 14   Westerberg  DP (2013) Diabetic ketoacidosis: evaluation and treatment. Am Fam Physician 87: 337–346.
• 15   Laffel  L (1999) Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes. Diabetes Metab Res Rev 15: 412–426.
• 16   Gurvits  GE,  Shapsis  A,  Lau  N,  Gualtieri  N,  Robilotti  JG (2007) Acute esophageal necrosis: a rare syndrome. J Gastroenterol 42: 29–38.