Article ID: 2024-0015-OA
Recent studies have indicated that diabetic ketoacidosis (DKA) can be the primary presenting symptom of pancreatic cancer. This comprehensive review assesses the existing research on the incidence of DKA as an initial symptom of pancreatic cancer, including its clinical characteristics, diagnostic challenges, and implications for treatment and prognosis. A comprehensive search was conducted across four electronic databases (PubMed, Scopus, Web of Science, and Cochrane), complemented by a manual search. The search criteria focused on original case reports of pancreatic cancer patients who presented with DKA. Among the 360 studies reviewed, 9 met the eligibility criteria. Among the cases, pancreatic adenocarcinoma was the most common type, followed by somatostatinoma and cystadenocarcinoma. Diagnostic modalities included computed tomography, ultrasound, biopsy, and endoscopic ultrasound. Elevated tumor markers such as CA19-9 were reported in several cases. Most patients presented with gastrointestinal and neurological symptoms, with high levels of glucose and ketone bodies. This review highlights that DKA can serve as a rare but significant initial presentation of pancreatic cancer. Identifying this association is critical for facilitating early diagnosis, which may improve the otherwise poor prognosis of pancreatic cancer. Our findings suggest that clinicians should maintain a high index of suspicion for pancreatic malignancy in patients presenting with unexplained DKA, particularly those without traditional risk factors or precipitating events. Early imaging and multidisciplinary evaluation are essential in such cases.
Diabetic ketoacidosis (DKA) is defined by the presence of uncontrolled hyperglycemia, metabolic acidosis, and elevated serum ketone levels. It represents a life-threatening complication primarily associated with type 1 diabetes mellitus (DM), although it can occasionally manifest in individuals with type 2 DM. DKA is caused by a relative or absolute deficiency of insulin and is exacerbated by hyperglycemia, dehydration, and acidosis.1 The occurrence of DKA as an initial manifestation of pancreatic adenocarcinoma in the context of chronic hyperglycemia is exceptionally rare, with only a limited number of documented cases.1,2,3,4,5,6,7,8,9,10,11,12 It is hypothesized that pancreatic adenocarcinoma impairs the functionality of insulin-producing beta cells, reducing their ability to compensate for insulin resistance.2,8 Pancreatic adenocarcinoma is frequently diagnosed at an advanced stage because of the absence of early clinical symptoms. The association between DM and pancreatic cancer remains a subject of ongoing debate within the medical community. It is extremely uncommon for DKA to be the first indication of pancreatic cancer, but this possibility should be considered after more common triggers of DKA have been ruled out. In the future, more data will be available regarding DKA as the primary presenting symptom of pancreatic adenocarcinoma and other types of cancer during the diagnostic process. Pancreatic cancer is associated with a poor prognosis and short survival rates, emphasizing the importance of early detection and appropriate treatment for improved outcomes. Case reports have indicated that DKA can serve as an initial manifestation of various types of pancreatic cancer, including somatostatinoma, pancreatic glucagonoma, metastatic liver carcinoma, metastatic renal cell carcinoma, and pancreatic cystadenocarcinoma.1,2 The aim of this comprehensive review is to include all articles related to the topic of DKA as a presenting symptom of pancreatic cancer.
We conducted an extensive computerized literature search utilizing PubMed, Cochrane, Scopus, and Web of Science. The search terms used included “DKA” and “pancreatic cancer,” either individually or in combination. Animal studies were excluded. Our objective was to identify case reports documenting patients with pancreatic cancer who presented for the first time with DKA. The search was limited to English-language publications, and we included all published case reports available until March 2023. After eliminating duplicate articles, five independent reviewers assessed the titles and abstracts for eligibility criteria. Full-text articles were obtained for further evaluation based on these criteria. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were used throughout the review process. The review flowchart is presented in Figure 1. We evaluated the included studies using the CARE checklist, and a further two reviewers reviewed the data.
Flowchart of process used to compile the case reports.
The literature search strategy was based on the following command structure: TITLE-ABS-KEY (DKA OR (diabetic OR diabetica) AND (acidosis OR ketoacidosis OR ketosis OR ketoacidosis)) AND TITLE-ABS-KEY (pancreas OR pancreatic) AND TITLE-ABS-KEY(tumor OR cancer OR neoplasm OR malignancy OR carcinoma). The case reports included in this review focused on the occurrence of DKA in patients diagnosed with primary pancreatic cancer. For inclusion in our study, case reports were required to: (1) provide a detailed account of the clinical manifestations presented by the patients, and (2) be documented and diagnosed by qualified healthcare practitioners. Case reports or abstracts that reported DKA in patients without pancreatic cancer or vice versa were excluded.
The search strategy identified 17 case reports that linked DKA to underlying pancreatic malignancy. Out of these, 8 case reports were excluded for the following reasons: 4 reports were inaccessible, 1 report involved an animal study, 2 reports discussed glucagonoma in which DKA is not unusual, and 1 report in which the pancreatic cancer was secondary to renal cell carcinoma. The necessary data were extracted from the remaining case reports, including information such as age, sex, race, site and type of cancer, presenting symptoms, diagnostic methods employed, carbohydrate antigen 19-9 (CA19-9) level, glycated hemoglobin (HbA1c) level, pH, bicarbonate level, glucose level, serum and urine ketone level, and specific therapies administered.
After a search of online databases, we included nine case reports with nine adult patients from six countries. Six of the patients were women ranging in age from 36 to 75 years old; five of them presented with pancreatic adenocarcinoma, one with pancreatic somatostatinoma. Three patients were men ranging in age from 47 to 69 years old; one of them presented with pancreatic cystadenocarcinoma and two patients with pancreatic adenocarcinoma.
The different diagnostic modalities of computed tomography (CT), ultrasound, biopsy, fine needle aspiration (FNA), microscopic examination, fine needle aspiration cytology (FNAC), and endoscopic ultrasound (EUS) were used to diagnose cases. CT and ultrasound were used for diagnosis in most cases. CT was used in five cases: it was combined with biopsy in two cases, and it was combined with biopsy and ultrasound in three cases. Microscopic examination was used to diagnose one case, while EUS with FNA was used for one case. Laboratory workup revealed high levels of CA19-9 tumor marker indicator in four cases and high levels of HbA1c in six patients (Table 1).
Author | Age (years) | Sex (M/F) | Race | Country | Number | Cancer | Site | Diagnostic modality |
CA19-9 (U/mL) | HbA1c (%) |
Menakuru et al.4 |
52 | F | Not given | USA | 1 | Pancreatic adenocarcinoma |
CT with IV contrast and biopsy | 251 | 13.9 | |
Willcox et al.6 |
47 | F | Not given | South Africa | 1 | Pancreatic somatostatinoma |
||||
Lin et al.3 | 75 | F | USA | 1 | Pancreatic adenocarcinoma |
Tail with atrophy of head and body | CT, ultrasound, and biopsy | 9.5 | ||
Mead et al.5 |
66 | F | Caucasian | USA | 1 | Pancreatic carcinoma |
Body and tail | Microscopic examination | ||
Zhong et al.7 |
47 | M | African American | Australia | 1 | Pancreatic adenocarcinoma |
Head | Contrast CT, ultrasound-guided biopsy | 983 | 10.9 |
Bush et al.9 |
69 | M | Not given | USA | 1 | Pancreatic adenocarcinoma |
Uncinate process | EUS/FNA | ||
Markabawi et al.2 |
59 | F | Not given | USA | 1 | Pancreatic ductal adenocarcinoma | Uncinate process | EUS and biopsy | 14.1 | |
Kaya et al.10 |
60 | M | Turkey | 1 | Pancreatic cystadenocarcinoma |
Head | CT, ultrasound, biopsy | 202 | 12.5 | |
Lee et al.1 |
36 | F | Not given | Korea | 1 | Pancreatic adenocarcinoma complicated by pancreatic abscess | CT, excisional biopsy | 30.36 | 9.7 |
CA19-9: carbohydrate antigen 19-9; HbA1c: glycated hemoglobin; CT: computed tomography; IV: intravenous; EUS: endoscopic ultrasound; FNA: fine needle aspiration.
Most of the cases presented varying symptoms because of systemic affection of multiple organs. Acid–base disturbance and high levels of glucose and ketone bodies were detected in all patients.
Neurological symptomsIn six cases,2,3,4,5,6,7 patients presented with neurological symptoms as outlined in the following discussion. A 75-year-old woman presented with confusion and lethargy with a 3-day history. She was disoriented to time, place, and person. When she presented to the emergency room (ER), there were no focal neurological signs; after she was investigated with ultrasound, abdominal CT, and FNA, she was diagnosed with pancreatic adenocarcinoma.3 A 52-year-old woman presented with confusion and lethargy. She was not alert or oriented to time, place, or person; a workup with abdominal CT and biopsy from the pancreas and lymph nodes revealed pancreatic adenocarcinoma.4 A 59-year-old woman presented to the ER with a 3-day history of lethargy and worsening confusion. A laboratory workup revealed metabolic acidosis with a high anion gap, and a workup using abdominal CT and EUS with biopsy revealed pancreatic adenocarcinoma.2 A 66-year-old woman presented to the hospital because of confusion and lethargy with a 1-day history. The neurological examination was normal except for confusion. She was diagnosed with carcinoma of the head and tail of the pancreas with laboratory workup and autopsy findings.5 A 47-year-old woman presented with drowsiness attributed to acid–base disturbance. Pancreatic somatostatinoma diagnosis was established using radioimmunoassay, immunochemistry, and tumor extraction with chromatography.6 A 47-year-old man presented with worsening confusion and a workup with contrast-enhanced triple-phase hepatic CT, which revealed pancreatic adenocarcinoma at the head of the pancreas. Neurological examination revealed that he was disoriented to time, place, and person with no focal neurological defect.7
Respiratory symptomsPatients in two case reports1,5 presented with respiratory symptoms. A 36-year-old woman presented to the ER complaining of dyspnea with a respiratory rate of 32 breaths/min; her physical examination was free of abnormal findings.1 After partial resection of the pancreas, a frozen biopsy showed nonspecific chronic inflammation. After her condition did not improve, abdominal CT revealed pancreatic adenocarcinoma.1 A 66-year-old woman felt mild dyspnea 1 week before her admission to the hospital. Physical examination after admission revealed regular and very deep respiration at 18 breaths/min.5
Cardiac symptomsOnly one patient presented with cardiac symptoms.4 A 52-year-old woman presented with mild distress and tachycardia. Her blood pressure was within normal limits (100/80 mm Hg), and cardiovascular examination and auscultation revealed no abnormality.
Gastrointestinal and urinary symptomsPatients in six case reports1,3,5,7,9,10 presented with gastrointestinal tract (GIT) symptoms without urinary symptoms. A 69-year-old man presented to the ER with fatigue, weight loss, and abdominal pain. Investigation with CT, bronchoscopy with cytology, and EUS/FNA for the pancreatic mass revealed poorly differentiated pancreatic carcinoma metastasized from primary colonic cancer.9 A 60-year-old man presented with weakness, nausea, loss of appetite (3-day history), and weight loss of nearly 10 kg during the past 2 months. Physical examination revealed dry tongue, skin turgor, and decreased tonus. Workup with ultrasound, abdominal CT, and biopsy with FNA revealed a pancreatic ductal adenocarcinoma.10 A 36-year-old woman presented with abdominal discomfort; physical examination revealed a soft, non-tender abdomen with no abnormality.1 A 75-year-old woman presented with complaints of occasional vomiting, weight loss, and suprapubic discomfort.3 Physical examination revealed hypertension, tachycardia, and dehydration.3 A 66-year-old woman presented with complaints of loss of appetite, vomiting, and mild diarrhea. Physical examination revealed loss of skin turgor, dry mucous membranes, pale skin, and soft abdomen with no tenderness.5 A 47-year-old man presented with complaints of significant weight loss and polydipsia; abdominal examination showed no abnormality.7
Patients in a further three cases2,4,6 presented with urinary symptoms in addition to GIT symptoms. A 59-year-old woman presented to the emergency department with diffuse abdominal pain, nausea without vomiting, polyuria, and polydipsia. Physical examination was remarkable for diffuse abdominal tenderness with no guarding or rigidity.2 A 47-year-old woman presented with vomiting, loss of appetite, weight loss, polydipsia, and polyuria (10-day history). Abdominal examination revealed a mildly enlarged liver with no other abnormalities.6 A 52-year-old woman presented with complaints of weight loss, polydipsia, and polyuria. Physical examination revealed dry mucous membranes, skin turgor, and decreased tonus.4
General and constitutional symptomsPatients in four cases1,2,3,7 presented with general and constitutional symptoms in addition to other multiple organ symptoms: a 75-year-old woman complained of anorexia and malaise3; a 47-year-old man with pancreatic adenocarcinoma presented with anorexia in addition to other symptoms7; a 36-year-old woman with pancreatic cancer complained of fever1; and a 59-year-old woman presented with myalgia in addition to other multiple organ symptoms.2 According to previously presented symptoms, most of the patients presented with GIT symptoms. Neurological symptoms such as confusion and lethargy were also reported in some cases. Dyspnea and tachycardia were less commonly reported, as well as urinary symptoms and constitutional manifestations.
Different therapeutic modalities were used to manage the cases. Intravenous (IV) fluids and insulin were used in seven cases,1,2,4,5,7,9,10 and one case1 also used empirical antibiotics. Insulin was used alone to manage DKA. Detailed information about pH, glucose, ketone bodies, symptoms, and management of the cases is given in Table 2.
Author | pH | Bicarbonate (mmol/L) | Glucose (mg/dL) | Ketones (mmol/L) a | Presenting symptoms | Specific treatment |
---|---|---|---|---|---|---|
Menakuru et al.4 |
7.18 | 14 | 6192 | 7.8 | Confusion, weight loss, distress, tachycardia, fruity-smelling breath, skin tenting, dry tongue, decreased tonus | IV fluids, insulin |
Willcox et al.6 |
7.33 | 9.5 | 590 | +++ | Vomiting, polyuria, thirst, dehydration, drowsiness, poor appetite, weight loss | |
Lin et al.3 |
7.28 | 450 | 25.82 | Confusion, vomiting, anorexia | Insulin (Levemir): 30 units daily | |
Mead et al.5 |
337 | Confusion, lethargy, weight loss, mild dyspnea | IV fluids, SC crystalline insulin, NPH insulin | |||
Zhong et al.7 |
7.21 | 13 | 1166.4 | 7.7 | Confusion, anorexia, weight loss | IV fluids, insulin |
Bush et al.9 |
656 | + | Fatigue, weight loss, abdominal pain | IV crystalloids and insulin | ||
Markabawi et al.2 |
7.1 | 8 | 419 | + | Three-day history of worsening confusion, lethargy, myalgia, diffuse abdominal pain, polyuria, polydipsia, nausea without vomiting | IV fluids and insulin |
Kaya et al.10 |
7.28 | 14.4 | 473 | +++ | Weakness, nausea, loss of appetite, weight loss, vomiting | IV fluids, electrolytes, and regular insulin infusion |
Lee et al.1 |
7.1 | 4.5 | 334 | +++ | Dyspnea, fever, abdominal discomfort | IV fluids, insulin infusion, empirical antibiotics |
a Color test for ketones in urine. +: Low ketone level; ++: moderate ketone level; +++: high ketone level.
IV: intravenous; SC: subcutaneous; NPH: neutral protamine Hagedorn.
Based on the CARE case report guidelines checklist with a scoring system ranging from 1 to 12 points, six case reports were of high quality.2,3,4,6,7,10 One report7 scored 12, and one report3 scored 11 because it did not mention the term “case report” in the title. Three studies2,4,6 scored 10, because they did not mention “case report” in the title, and the patients did not share their perspectives. One report10 had a score of 9, because it did not adjust follow-up, did not show dates and times, and the patient did not share his perspective.
Three case reports1,5,9 were of moderate quality. One report1 scored 8 because “case report” was not mentioned in the title, it did not provide times or dates, diagnostic methods, or challenges, and the patient did not share her perspective. One report5 scored 7 because it did not mention “case report” or the key elements of the case and did not mention diagnostic methods or challenges. Also, the patient did not share his perspective. One case report9 scored 6 points. It did not mention “case report” in the title or the key elements of the case, and a description of the relevant examination findings, as well as times, dates, and follow-up details, were not mentioned (Table 3).
CARE guidelines | Case report a | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1) Includes “case report” (or “case study”) in the title along with the phenomenon of greatest interest (e.g., symptom diagnosis, test, intervention) | Y | N | N | N | Y | N | N | Y | N |
2) Includes key elements of the case in 2–5 words | Y | Y | Y | N | Y | N | Y | Y | Y |
3) (a) Introduction describes contribution of case report to knowledge in the discipline; (b) case presentation describes main symptoms of patient, main clinical findings, main diagnoses and interventions, main outcomes; (c) conclusion describes main “takeaway lessons” from the case | Y | Y | Y | Y | Y | Y | Y | Y | Y |
4) Provides a brief background summary of this case referencing the relevant medical literature | Y | Y | Y | Y | Y | Y | Y | Y | Y |
5) Provides: (a) demographic information (e.g., age, sex, ethnicity, occupation); (b) main symptoms of the patient (their chief complaints); (c) medical, family, and psychosocial history, including diet, lifestyle, and genetic information whenever possible, and details about relevant co-morbidities, including past interventions and their outcomes | Y | Y | Y | Y | Y | Y | Y | Y | Y |
6) Describes the relevant physical examination (PE) findings | Y | Y | Y | Y | Y | N | Y | Y | Y |
7) Depicts important dates and times (in tables or figures) | Y | Y | Y | N | Y | N | Y | N | N |
8) Describes: (a) diagnostic methods (e.g., PE, laboratory testing, imaging, questionnaires); (b) diagnostic challenges (e.g., financial, language/cultural); (c) diagnostic reasoning, including other diagnoses considered; (d) prognostic characteristics (e.g., staging) where applicable | Y | Y | Y | N | Y | Y | Y | Y | N |
9) Describes: (a) type of intervention (e.g., pharmacologic, surgical, preventive, self-care); (b) administration of intervention (e.g., dosage, strength, duration); (c) changes in intervention (with rationale) | N | Y | Y | Y | Y | Y | Y | Y | Y |
10) Describes: (a) changes in therapeutic intervention; (b) important follow-up diagnostic and other test results; (c) intervention adherence and tolerability; (d) adverse and unanticipated events | N | Y | Y | Y | Y | N | Y | N | Y |
11) Describes: (a) strengths and limitations of case management; (b) relevant medical literature; (c) rationale for conclusions (including assessments of cause and effect); (d) main “takeaway lessons” of the case report | Y | Y | Y | Y | Y | Y | Y | Y | Y |
12) Describes the perspective or experience of the patient. If requested, did the patient give informed consent? | Y | N | Y | N | Y | N | N | N | N |
Score | 10 | 10 | 11 | 7 | 12 | 6 | 10 | 9 | 8 |
a Identification of case reports. 1: Menakuru et al.4; 2: Willcox et al.6; 3: Lin et al.3; 4: Mead et al.5; 5: Zhong et al.7; 6: Bush et al.9; 7: Markabawi et al.2; 8: Kaya et al.10; 9: Lee et al.1
Y: yes; N: no.
Our review analyzed nine cases in which DKA was the initial presentation of pancreatic cancer. The type of pancreatic cancer in seven cases was adenocarcinoma, with one case of pancreatic cystadenocarcinoma and one case of somatostatinoma. Because of the systemic affection of multiple organs, patients presented a variety of symptoms, with most exhibiting both gastrointestinal and neurological symptoms. Although the link between DKA and pancreatic cancer has been previously reported,11,12 our study provides a comprehensive synthesis of existing evidence, highlighting patterns in symptom presentation, diagnostic approaches, and outcomes. Recognizing DKA as a potential diagnostic clue for pancreatic cancer, particularly in the absence of other precipitating factors, could facilitate earlier detection and management of this malignancy. Early pancreatic cancer symptoms are ambiguous, and symptoms usually appear later in the course of the disease. Therefore, most pancreatic tumors are diagnosed at an advanced stage, resulting in a 5-year survival rate of approximately 6% to 7%.13 A case was reported where a patient developed new-onset hyperglycemia and elevated HbA1c within 1 year of being diagnosed with DKA. The patient was later diagnosed with stage 4 pancreatic adenocarcinoma and ultimately succumbed to the disease.4 The authors suggested that earlier awareness of the potential severity of new-onset DM could have significantly improved the prognosis. They concluded that abdominal imaging, such as CT and EUS, should be considered for patients with new-onset DM to exclude pancreatic carcinoma.4
Essentially, DKA has a role in the development of pancreatic adenocarcinoma and needs to be further characterized and understood. DKA can be an initial symptom of pancreatic adenocarcinoma, according to data from case reports. Various diagnostic modalities were used, including CT, ultrasound, biopsy, FNA, microscopic inspection, FNAC, and EUS (Table 1). Preliminary laboratory tests revealed that most patients had high levels of CA19-9 and HbA1c, showed acid–base imbalance, and had high amounts of glucose and ketone bodies (Table 2). Most of the patients had GIT symptoms, though some also reported neurological symptoms like confusion and lethargy. Less frequently reported symptoms included tachycardia, dyspnea, and urinary symptoms. In most cases, different therapeutic modalities were used, such as IV fluids and insulin. One case used empirical antibiotics, IV fluids, and insulin; whereas another case used insulin only (Table 2).
Bidirectional relationship between diabetes and pancreatic cancerThe intricate bidirectional relationship between DM and pancreatic cancer highlights the importance of considering DM as both a risk factor and a consequence of pancreatic malignancies. Long-standing type 2 DM has been proposed to increase the risk of pancreatic cancer through mechanisms such as direct islet cell alterations and diabetogenic factors secreted by cancer cells. DM and impaired glucose tolerance are present in roughly half of the newly diagnosed pancreatic cancer patients and 74% of patients with new-onset DM (within 2 years).12,13,14,15 Conversely, pancreatic cancer may induce DM through beta-cell dysfunction, insulin resistance, and hyperinsulinemia.12,13,14 Cersosimo et al.16 reported that the primary mechanisms contributing to the development of DM in the context of pancreatic cancer include peripheral insulin resistance, which induces hyperinsulinemia, along with a concurrent impairment in the ability of beta-cells to respond to glucose. A meta-analysis17 highlighted the significant association of DM diagnosis within 1 year prior to the detection of pancreatic adenocarcinoma, suggesting that new-onset DM or worsening glycemic control may serve as an indicator for pancreatic adenocarcinoma screening. When compared with the general population, patients with DM have twice the risk of developing pancreatic cancer.
Individuals with new-onset DM face a six- to eightfold increased likelihood of being diagnosed with pancreatic cancer within 3 years, with an estimated 36-month incidence of approximately 1%. Therefore, new-onset DM in thin, older adults, particularly those with additional risk factors such as weight loss or family history, should prompt screening and imaging studies to exclude pancreatic malignancy.13,14,15,16,17 However, the clinical implementation of such strategies is complicated by the high prevalence of DM relative to pancreatic cancer and the limitations of current screening tools.13
Role of screening and early detection of pancreatic cancer in DMScreening strategies for pancreatic cancer in high-risk populations, particularly those with DM, are continuously evolving. However, existing guidelines present differing recommendations. In the United Kingdom, the National Institute for Health and Care Excellence recommends abdominal imaging within 2 weeks for individuals over 60 who present with new-onset DM and unexplained weight loss.13,14,15 Conversely, the United States Preventive Services Task Force and the National Cancer Institute do not endorse routine screening for asymptomatic populations because of insufficient evidence of benefit. Instead, they emphasize the importance of exploring the relationship between DM and pancreatic cancer as a key area for research.13 Although population-wide screening for pancreatic cancer in older adults with new-onset DM who are otherwise asymptomatic is not advised—primarily because of the high prevalence of DM, the absence of a reliable and affordable screening test, and the relatively small number of pancreatic cancer cases detected—targeted screening in high-risk individuals holds promise for early diagnosis and improved outcomes. Identifying specific features that differentiate pancreatic cancer-associated DM from other forms of new-onset DM could refine screening strategies, enabling a more focused and effective approach for those most likely to benefit.13,14
Early detection of pancreatic cancer remains a significant challenge, driving recent efforts to identify high-risk individuals through targeted screening. Because early-stage pancreatic cancer is often asymptomatic, developing a specific and sensitive screening system is critical for timely diagnosis. Predictive models, such as the Enriching New-Onset Diabetes for Pancreatic Cancer (END-PAC) score, have shown promise in stratifying risk. These models incorporate age, weight changes, glycemic control, and family history to identify at-risk individuals. For instance, Sharma et al.18 reported that using an END-PAC score threshold of 3 significantly improved the sensitivity and specificity of pancreatic cancer detection in high-risk groups.
Among biological markers, CA19-9 is a widely used biomarker for pancreatic cancer detection. Although its efficacy in population screening is limited by specificity, a cutoff value of 75 U/mL has demonstrated a sensitivity of 69.5% and specificity of 98.2%. However, it is most effective in the advanced stages of pancreatic cancer. Engle et al.19 also proposed that CA19-9 could play a role in the etiology of pancreatic cancer and may even serve as a therapeutic target. In our study, CA19-9 levels were high in three cases, very high in seven cases, and normal in one case, highlighting its variable utility. Another promising marker is circulating tumor DNA (ctDNA), which offers a complementary approach to the detection of pancreatic cancer. Derived from dying cancer cells, ctDNA is detectable in the bloodstream and holds significant potential for early diagnosis. While high levels of ctDNA are observed in more than 85% of advanced cancer cases, its early-stage detection is less frequent, emphasizing its potential as a tool for early-stage screening.13 These emerging tools and models underline the importance of integrating predictive analytics and biomarkers for improving the early detection of pancreatic cancer in patients with DM.
In adult patients with DKA, 47% had a known diagnosis of type 1 DM, 26% had a known diagnosis of type 2 DM, and 27% were newly diagnosed with DM at the time of presentation.20 It was previously thought that the onset of DKA in patients with type 2 DM necessitates a stressful triggering event. Recent research, however, has indicated that up to 25% of people with DM who developed DKA had no evident triggers.21 The authors of one of the studies1 reported that a patient with a complicated adenocarcinoma had DKA as an initial presenting symptom and was diagnosed with type 2 DM. However, the unusual features of DKA, such as sudden onset and high fever, raised suspicion for further investigations that revealed pancreatic abscess on pancreatic adenocarcinoma. Another case with known type 2 DM and treated with metformin and empagliflozin presented with DKA with high blood glucose, excluding SGLT-2 inhibitors, as a cause of DKA with no other apparent cause.2 The patient needed 3 days of IV insulin to restore her acid–base balance.2 These unusual characteristics referred to further evaluation that diagnosed pancreatic adenocarcinoma. Although such clinical situations are unique, clinicians should not rule out pancreatic cancer in newly diagnosed or even known type 2 DM presenting with DKA without evident precipitating events and peculiar features. An African American man was admitted with first-onset DKA without any evidence of triggering factors and high pancreatic enzyme levels in his initial laboratory investigation. After a meticulous workup, he was diagnosed with pancreatic adenocarcinoma. Interestingly, his mother and grandmother had died early because of pancreatic cancer, but this was unknown at the time of his initial presentation.7 A positive family history of pancreatic cancer greatly raises the risk (OR = 2.09, 95% confidence interval: 1.01–4.33).22 Therefore, careful evaluation of family history could be valuable in such an instance.
Neuroendocrine neoplasms (NENs) represent a heterogeneous group of rare tumors with diverse morphological features, immunophenotypic patterns, genetic profiles, and clinical presentations. These tumors can originate in various body regions, including the lungs, small intestine, and pancreas. Pancreatic NENs, which arise from the islets of Langerhans, such as somatostatinoma, comprise approximately 12% of all NENs and account for 1%–3% of all pancreatic malignancies.23 Over a 4-year period, a 47-year-old woman was hospitalized four times for acute hyperglycemia with significant ketoacidosis.6 She developed weight loss and hepatomegaly, and ultrasonography revealed a pancreatic tumor that was a somatostatinoma with no stromal or vascular invasion. The surgery was curative, and somatostatin levels decreased to low levels after surgery.6
This comprehensive review provides an overview of case reports investigating the occurrence of DKA as an initial presenting symptom of various pancreatic cancers, including pancreatic adenocarcinoma, pancreatic cystadenocarcinoma, and somatostatinoma. Early diagnosis of pancreatic adenocarcinoma is often challenging, emphasizing the need to identify early symptoms that may indicate the presence of cancer and potentially improve survival rates. Although significant progress has been made in understanding this relationship, there is still much to uncover. The association between DKA and pancreatic adenocarcinoma is complex, and establishing a causal relationship is not always straightforward. However, healthcare practitioners must consider DKA as a significant initial symptom of pancreatic adenocarcinoma, because a majority of globally recorded cases indicate that patients are diagnosed with cancer shortly after being diagnosed with DKA. In addition to the patient’s history of DM and previous health records, maximum benefit can be obtained from case reports that include comprehensive details on DKA management, the nature and location of the cancer, and the examinations and follow-up procedures that were conducted from the moment of DKA diagnosis to cancer detection.
LimitationsOur review has some limitations. First, the number of cases included is small, reflecting the rarity of DKA as an initial presentation of pancreatic cancer. Second, the reliance on case reports introduces publication bias and limits generalizability. Finally, the lack of longitudinal data precludes definitive conclusions about the long-term outcomes of these patients. Future prospective studies and larger datasets are necessary to validate these observations and refine screening strategies for pancreatic cancer in high-risk populations.
This review highlights that DKA can serve as a rare but significant initial presentation of pancreatic cancer. Identifying this association is critical for facilitating early diagnosis, which may improve the otherwise poor prognosis of pancreatic cancer. Our findings suggest that clinicians should maintain a high index of suspicion for pancreatic malignancy in patients presenting with unexplained DKA, particularly those without traditional risk factors or precipitating events. Early imaging and multidisciplinary evaluation are essential in such cases.
The authors have declared that no conflict of interest exists.