Introduction

Two decades have passed since fulminant type 1 diabetes (T1D) was proposed as a new sub-type of T1D by Imagawa et al. [1]. Recently, a definitive distinction between fulminant type, which shows drastic and diffuse destruction of pancreas tissue with a non-autoimmune pathophysiology, and acute-onset type [2], demonstrating an autoimmune reaction to some peptide antigens in pancreatic β cells [3, 4], has been widely accepted based on their primary etiologies. However, some reported cases have shown characteristics of both, for example, fulminant type with temporary depletion of insulin secretion function or without complete depletion [5,6,7], and with the presence of islet-related autoantibodies [7,8,9]. Under the circumstances, very recent results have led to a hypothesis regarding the role of autoimmunity in fulminant T1D [10]. In addition, most patients with T1D in Japan are given a diagnosis of acute-onset type below the age of 30 years, and of fulminant type when in their 30 s or 40 s [11]. Nevertheless, an issue that has received focus is that newly developed T1D in elderly individuals should be given a higher priority as a result of the aging of society, though reports of patients older than 80 years are limited [12,13,14]. We present here a case report of an 82 year-old female patient in Japan newly diagnosed with acute-onset T1D accompanied by repeated mild increases of pancreatic exocrine enzymes, thus mimicking fulminant type.

Case report

An 82 year-old female was transported to our emergency department with deterioration of consciousness. Her past medical history included hypertension, chronic kidney disease (creatinine 1.5 mg/dL, creatinine clearance 19.9 ml/min noted several months prior), lacunar infarction in the right thalamus, which was well-controlled with amlodipine and clopidogrel, and mixed dementia. A reduced consciousness level (Glasgow Coma Scale score of 10: E3, V1, M6) was noted, while blood pressure was 166/100 mmHg, pulse rate 91 bpm, body temperature 36.6 °C, respiratory rate 25/min, and pulse oximetry 96% on 3 L of nasal cannula oxygen. In a physical examination, no abnormalities including paralysis were noted, except for a dried oral cavity. Height and weight were 150.0 cm and 44.3 kg, respectively (BMI = 19.7 kg/m2). Blood and urine test results showed hyperglycemia (584 mg/dL with 8.2% HbA1c) with metabolic acidosis (pH in blood gas analysis 7.32), a large blood anion gap (total ketone body production 2900 µmol/L, lactic acid 17.1 mg/dL), mildly increased exocrine pancreatic enzymes (amylase 543 IU/L, and lipase 59 U/L, elastase-1 479 ng/dL), inflammatory change (white blood cell 14,700/µL, C-reactive protein 2.1 mg/dL), pyuria, and negative with some tumor makers including carcinoembryonic antigen, carbohydrate antigen 19-9, α-fetoprotein, S-pancreas-1 antigen, pancreatic cancer-associated antigen-2. Computed tomography showed no remarkable findings, such as cerebral vascular accident, acute pancreatitis, or apparent tumor lesions.

Based on those findings, the diagnosis was diabetic ketosis caused by urinary infection, and following urgent hospitalization intensive transfusion therapy including intravenous injections of insulin and an antibacterial drug were started. The patient was successfully treated (Fig. 1) and intensive insulin use was shifted to daily oral hypoglycemic agents (linagliptin, glimepiride), after which she was discharged on hospital day 21. Fourteen days after discharge, improved blood test results were noted, as follows: HbA1c 7.6%, glycoalbumin 17.9%, fasting plasma glucose 117 mg/dL, serum connecting peptide immunoreactivity (CPR) 1.63 ng/mL, creatinine 1.27 mg/dL, creatinine clearance 23.7 ml/min, and normalized pancreatic exocrine enzymes.

Fig. 1
figure 1

Progress of blood glucose, amylase, lipase, elastase-1, and connecting peptide immunoreactivity (CPR) in serum and urine during second admission. Decreases in serum and urinary CPR were observed after amelioration of diabetic ketoacidosis. A glucagon-loading test revealed a CPR secretion defect

Full size image

At 38 days after discharge (73 days after first admission), the patient returned to the hospital because of diabetic ketoacidosis (DKA) and with weight reduction (39.9 kg). Blood test results showed hyperglycemia with ketoacidosis (blood glucose 910 mg/dL, pH 7.15 in blood gas analysis, bicarbonate 5.7 mmol/L, total ketone bodies > 7000 µmol/L), and a re-increase of pancreatic exocrine enzyme levels (amylase 382 IU/L, lipase 82 U/L, elastase-1 569 ng/dL). Unlike the previous hospitalization, decreased serum and urinary CPR (0.48 ng/mL and 6.1 µg/day, respectively) were detected after improvement of hyperglycemia and hypovolemic state (blood glucose 163 mg/dL, creatinine 1.32 mg/dL, creatinine clearance 22.9 ml/min), while a glucagon-loading test revealed a CPR secretion defect (Fig. 1). Test results for the anti-glutamic acid decarboxylase (GAD), anti-insulinoma antigen-2 (IA-2), anti-islet cell cytoplasmic (ICA), and zinc transporter 8 (ZnT8) antibodies were negative. According to the clinical course, the diagnosis was type 1B diabetes. Human leukocyte antigen (HLA) haplotyping was performed with a next generation sequencing method using HLA typing kits from Scisco Genetics, Inc. (Seattle, WA), which resulted in detection of A*11:01/26:01, B*51:01/54:01, C*01:01/14:02, DRB1*04:05/15:01, DRB4*01:03/-, DRB5*01:01/-, DQA1*01:02/03:03, DQB1*04:01/06:02, DPA1*02:02/-, and DPB1*05:01/-. Virus neutralization antibody tests were performed by BML, Inc. (Tokyo, Japan), which revealed high antibody titers for several coxsackie viruses (type A4, A7, A9, B3) and parainfluenza virus type B3 for a period of at least 145 days, with a significantly elevated antibody titer (more than two steps) against coxsackie virus type B3 (CVB3) (Table 1).

Table 1 Results of virus antibody titers examination
Full size table

Intensive insulin therapy was performed and normalization of blood glucose and pancreatic exocrine enzyme levels were achieved. We intended to provide continuous intensive insulin therapy for the patient in consideration of the low level of insulin secretion. Unfortunately, cognitive impairment (Category II in Japanese guidelines for elderly individuals with diabetes mellitus [15]) and lack of support from family members did not allow that after discharge. Alternatively, the patient was given basal support oral therapy with miglitol and insulin degludec with support provided by a home nursing care service, and she was successfully discharged after 20 days showing normalization of blood glucose and pancreatic exocrine enzyme levels. Thereafter, HbA1c level remained safe between 7.0% and 8.0%, a range recommended by the Japanese guidelines for elderly individuals [15], though serum CPR level was low at 0.17 ng/mL at 43 days and then 0.08 ng/mL at 286 days after discharge.

Discussion

We experienced a case of newly diagnosed acute-onset type 1B diabetes with mildly increased levels of serum pancreatic exocrine enzymes in an elderly female patient. Although the clinical progress of attenuation of innate insulin secretion function corresponded to that seen in acute-onset type cases [2], an increase of those enzymes is generally observed in fulminant type cases [16]. Interestingly, our patient showed a bimodal and milder increase of those enzymes following hyperglycemia as compared with a previous report of hyperglycemia patients [1]. While the possibility of ketoacidosis-onset noninsulin-dependent diabetes mellitus was a concern related to temporal amelioration with glimepiride, the present patient did not show typical characteristics, such as age at onset and gender [17], and the finding of weight loss at the time of recurrence was contrary to the “accelerator hypothesis” [18]. Together with negative findings for islet-related autoantibodies more often seen in fulminant type [16], the present case should be considered as mimicking fulminant type with more than a certain duration required for depletion, in which the impetus may be attenuated by a background feature, such as an environmental or genetic factor.

A similar case was reported by Tamura et al.: a 66 year-old patient with well-controlled type 2 diabetes suddenly developed acute hyperglycemia following flu-like symptoms along with elevated pancreatic exocrine enzymes, and negative findings for GAD and IA-2 antibodies, while insulin-secreting function was not severely damaged [6]. Findings in that case suggest attenuation of the impetus seen in typical fulminant type, similarly to the present case. In this pathophysiology, degree of β cell destruction may vary; mild damage may cause impairment of insulin secretion, showing the features of type 2 diabetes, which was seen in Tamura’s case and initial phase of the present case, although severe damage should result in depletion of insulin, leading to development of T1D. Onset of this type of diabetes, in which mild elevation of serum pancreatic exocrine enzymes can be observed, would be hardly detected without any unusual symptoms or routine blood test, and hence, potential patients with this type of diabetes might be more than those we recognize in newly-onset diabetes without any known causes, including obesity, inappropriate life-style, or strong genetic background.

Effect of viral infection should be considered as an environmental factor, in which a strong contribution towards devastation of β cells is suggested and the mechanism is assumed to vary from direct β cell destruction to indirect damage through various possible pathways activated by virus infection, including autoimmunity, inflammation, activated T cells, molecular mimicry, induction of dedifferentiation [19]. In the present case, continuous positive findings for CVA4, A7, A9, B3, and parainfluenza virus type B3 antibodies with high titer were observed (Table 1), which were all previously reported to be associated with the pathology of T1D [20,21,22,23,24]. These previous reports suggest that infection of viruses, whose antibodies were detected with high titer for a long period, might be involved in destruction of β cells directly or indirectly, although further analyses should be required to understand the mechanisms.

HLA haplotypes have genetic potential for both susceptible and protective effects on development of fulminant and acute-onset types of T1D. HLA-DRB1*04:05-DQB1*04:01 and DRB1*15:01-DQB1*06:02 were detected in the present case. According to the previous reports, DRB1*04:05-DQB1*04:01 is susceptible in acute-onset type as well as fulminant type negative for the anti-GAD antibody, while DRB1*15:01-DQB1*06:02 is protective in acute-onset but neutral in fulminant type [25, 26]. However, those with duplotypes of HLA-DRB1*04:05-DQB1*04:01 and DRB1*15:01-DQB1*06:02 reportedly have a lower possibility of developing acute-onset type (odds ratio 0.35, p value 2.7 × 10–3) and no significance was found for fulminant type as compared with normal control subjects [25]. Our patient having these duplotypes could not avoid development of acute-onset T1D without autoimmune pathophysiology. Considering the findings of these previous analyses [25, 26], the HLA haplotype detected in the present case may not have been strong enough to protect from an environmental factor to induce β cell destruction, though may have potential to attenuate the impetus seen in fulminant type.

Most patients with fulminant type T1D in Japan are diagnosed at an age below 30 or in their 40 s [11], with onset in younger adults typical for acute-onset type. The present case newly developed T1D at the age of 82 years old. With respect to onset of T1D in very elderly patients in Japan, knowledge is limited. Sixteen patients with T1D (12 type 1A, 4 type 1B) were found among 199 Japanese cases of new-onset diabetes aged over 65 years old and the usefulness of an IA-2 antibody test was reported, though this investigation unfortunately did not include the ZnT8 antibody [27]. Another retrospective study conducted in Japan consisted of 22 T1D patients diagnosed at over 75 years old (8 acute-onset, 12 slowly progressive type, 2 unknown) provided some interesting findings; female patients were dominant, a negative or low grade of anti-GAD antibody titer was observed in acute-onset type, and HLA-A24, DR4, and DR9 were related with both acute-onset and slowly progressive type [28]. The present case is also female and negative for anti-GAD antibody, but does not have these HLA haplotypes. To know genetic background and environmental risk factors for developing T1D at very elderly age, further analyses should be required in this aged society.

The present report has some limitations, as a histological examination was not performed and results of other genetic factors such as single-nucleotide polymorphisms including the CTLA4 gene would be helpful [29]. Nevertheless, it is interesting to note that the present elderly female patient showed characteristics of fulminant T1D but finally received a clinical diagnosis of acute-onset type. A future study and case accumulation to clarify the mechanism for giving a better understanding of T1D are anticipated and it might contribute to re-classification of T1D as the other subtype.