Abstract
This case report describes a 25-year-old woman with type 1 diabetes mellitus (T1DM) and autoimmune thyroid disease (AITD) who developed Addison disease (AD). The diagnosis was suspected due to a significant decrease in insulin requirement and recurrent hypoglycemia despite adjustments in insulin pump therapy. The patient exhibited typical clinical features of AD, including hyperpigmentation and electrolyte imbalances. Early recognition and prompt treatment are crucial in preventing a life-threatening adrenal crisis. This case highlights the importance of vigilant monitoring of T1DM patients, especially those with other autoimmune diseases, for early signs of AD. The use of continuous glucose monitoring (CGM) and insulin pump therapy can aid in the management and early detection of this complication.
Keywords: autoimmune polyendocrine syndrome type 2, type 1 diabetes mellitus, Addison disease, automated insulin delivery system, hypoglycemia, adrenal crisis
Introduction
Autoimmune polyendocrine syndrome type 2 (APS2) is a polygenic disease characterized by type 1 diabetes mellitus (T1DM), autoimmune Addison disease (AD), and autoimmune thyroid disease (AITD) (at least 2 of these 3). Some authors divide the syndrome into autoimmune polyendocrine syndrome type 2a (APS2a) if AD is present and autoimmune polyendocrine syndrome type 2b (APS2b) if it is absent. Whereas APS2a is rare, APS2b is relatively common, as approximately 20% of patients with T1DM have AITD [1]. APS2 prevalence is very variable depending on the population studied, ranging from 1:1000 to 1:20 000 inhabitants. It appears in adults, especially in the third and fourth decade, and it most frequently affects women with a ratio of 3:1. At the time of diagnosis, several diseases of the syndrome (AD, T1DM, AITD) may be present, but in most cases the onset is gradual [2].
In patients with T1DM, reduced insulin requirements or hypoglycemic trends in a previously well-controlled patient are a warning that AD may be developing, especially if the patient also has other autoimmune disease. Glucocorticoids are known to induce insulin resistance and raise glucose levels. Conversely, in the absence of physiological levels of circulating glucocorticoids, insulin sensitivity is increased [3]. Recently, a case report described an adolescent girl with T1DM and subsequent latent AD, which was diagnosed based on a recurrent hypoglycemic trend detected by continuous glucose monitoring (CGM) [4].
In this case study, we describe a 25-year-old woman with long-standing T1DM and AITD who developed primary adrenal insufficiency, clinically suspected because of recurrent hypoglycemic episodes and significant reduction on insulin requirements detected through analysis of the data from her automated insulin delivery (AID) Minimed780g system (MM780g). To our knowledge, this is the first time that an onset of Addison disease in a patient using an AID has been described, demonstrating that the MM780g, together with continuous medical follow-up, is able to adapt to this situation in the weeks prior to the onset of an adrenal crisis.
Case Presentation
In January 2024, a 25-year-old woman with a history of T1DM and AITD attended our center for a routine endocrinology follow-up visit.
The patient was diagnosed with T1DM when she was 7 years old (1999). In 2016 she started using insulin pump therapy, and in 2021 she changed to AID MM780g, an advanced hybrid closed-loop insulin pump. Using AID, she had an excellent metabolic control of the T1DM (time in range [TIR] 88%, time below range [TBR] 3%, time above range [TAR] 9%, glycosylated hemoglobin [HbA1c] 6.3%) in February 2023 (see Table 1). As the only complication of her diabetes, she had mild retinopathy, which had been diagnosed postpartum, and she had delivered a healthy baby the previous year. Her family history was positive for primary hypothyroidism, and she had autoimmune hypothyroidism, taking 137 µg of levothyroxine daily.
Table 1.
Glycemic control and adjustments made during the weeks prior and after AD diagnosis
| Ambulatory glucose profile metrics | % Objectives | February 2023 | January 2024 | March 2024 | April 2024 | July 2024 |
|---|---|---|---|---|---|---|
| TIR (%) Glu < 54-70 mg/dL, 2.99-3.89 mmol/L | >70% | 88% | 82% | 75% | 80% | 74% |
| TBR (%) Glu < 70 mg/dL, 3.89 mmol/L | <5% | 3% | 6% | 4% | 1% | 1% |
| TAR (%) Glu > 180 mg/dL, 9.99 mmol/L | <30% | 9% | 12% | 21% | 19% | 25% |
| CV (%) | ≤36% | 33% | 34% | 36% | 37% | 32% |
| Ratio g carbs/UI | 10 g/UI | 11 g/UI | 17 g/UI | 20 g/UI | 12 g/UI | |
| TDD UI | 35 UI | 33 UI | 30 UI | 24 UI | 50 UI | |
| Number of low glucose sensor alerts/day | 3.8 | 6.2 | 5.5 | 1.6 | 4.1 | |
| Glucose target (mg/dL, mmol/L) | 100 mg/dL, 5.5 mmol/L | 100 mg/dL, 5.5 mmol/L | 110 mg/dL, 6.11 mmol/L | 120 mg/dL, 6.66 mmol/L | 100 mg/dL, 5.5 mmol/L | |
| AIT (hours) | 3 hours | 3 hours | 3 hours | 3 hours | 2 hours | |
| Grams of carbs introduced/day on AID | 203 g | 159 g | 215 g | 218 g | 209 g | |
Abbreviations: AID, automated insulin delivery; AIT, active insulin time; CV, coefficient of variation; TAR, time above range; TBR, time below range; TDD, total daily dose; TIR, time in range; UI, units of insulin.
At this visit, the patient complained that her glycemic control had deteriorated due to an increase in the percentage of hypoglycemia (TIR 82%, TBR 6%, and TAR 12% in January 2024, see Table 1). The patient's review of systems was negative, and the blood test showed no alterations, sodium (Na) 138 mEq/L (138 mmol/L) (normal reference range 135-145 mEq/L, 135-145 mmol/L) and potassium (K) 4.2 mEq/L (4.2 mmol/L) (normal reference range 3.5-5 mEq/L, 3.5-5 mmol/L), but the patient noted that often she had to input fewer grams of carbohydrate into the pump's algorithm than she ate to avoid hypoglycemia. Other causes of hypoglycemia, such as excessive or mis-timed insulin dosing, delayed meals, increased physical activity, and alcohol consumption, were ruled out. At this time, the insulin pump parameters were modified: the carbohydrate insulin ratio (g/UI) was increased by 20% in all segments, the duration of active insulin was maintained at 3 hours to avoid aggressive bolus-corrections, and the pump target was raised to 110 mg/dL (6.1 mmol/L). Successive appointments were made by video consultation every 2 weeks, and the g/UI ratio was gradually adjusted, with an improvement in hypoglycemia (TBR 1%, see Table 1) and a reduction in the number of daily hypoglycemic alarms (6.2 vs 1.6, see Table 1). However, the increase in the ratio was remarkable: the previous year, the patient required 1 unit of insulin per 10 g of carbs, and by April 2024, the ratio had doubled to 20 g per unit. The daily insulin requirement had been reduced by 30%, from 35 units during the previous year and up to the previous month, to 24 units in April 2024.
In May 2024, a videotelephone follow-up visit was scheduled. The patient's relatives had observed a worsening of the patient's general condition, and after checking her glucose data on CareLink, she was referred to the emergency department for suspected adrenal insufficiency. In the emergency department, the patient reported symptoms of approximately 3 weeks of progressive asthenia and tendency to lethargy, accompanied by epigastric abdominal pain, nausea, vomiting, and marked hypotension (blood pressure 80/40 mmHg, normal >90/60 mmHg) with repeated presyncopal symptoms. She also noted a tendency to hypoglycemia and anorexia, more pronounced than in previous days (see Fig. 1).
Figure 1.
Hyperpigmented scar and nail band.
Diagnostic Assessment
Emergency room vitals were blood pressure 85/50 mmHg and heart rate 91 beats per minute (bpm). She was afebrile, conscious, and alert. She was mildly tachypneic at rest and with signs of dehydration.
Physical examination revealed increased pigmentation of the skin and mucous membranes. Note the darkening of a previously normal scar and the appearance of a hyperpigmented nail band, highly suggestive of Addison disease (Fig. 2).
Figure 2.
Fasting plasma venous glucose 45 mg/dL (2.22 mmol/L) despite using an AID system.
In the venous blood gas measurement (pH 7.40 [normal range 7.35-7.45]; partial pressure of carbon dioxide [pCO2] 33 mmHg [4.40 kPa] [normal range 35-45 mmHg, 4.7-6.0 kPa]; arterial bicarbonate [HCO3] 29 mmol/L [normal range 22-26 mmol/L]; glucose 45 mg/dL [2.22 mmol/L]; Na 117 mEq/L [117 mmol/L]; K 5.7 mEq/L [5.7 mmol/L]), we could observe marked hyponatremia with hyperkalemia, with simultaneous hypoglycemia confirmed in the laboratory blood tests. There was no acidosis because the patient was compensating with hyperventilation.
Suspecting adrenal insufficiency, basal cortisol levels were obtained and treatment was initiated.
Early morning (approximately 8 Am) measurement of serum cortisol was 2 µg/dL (55.2 nmol/L) (normal range >10 µg/dL, > 300 nmol/L), and subsequent tests confirmed our diagnosis, showing adrenocorticotropic hormone (ACTH) levels of 1223 pg/mL (269 pmol/L) (normal range 10-50 pg/mL, 2-11 pmol/L) and markedly elevated renin (renin 314 000 mUI/L, normal range 4-46 mUI/L) with low aldosterone (aldosterone 2.58 ng/dL or 71 pmol/L [normal range 4-16 ng/dL, 110-440 pmol/L]).
Treatment
Given the high suspicion of AD, 200 cc of 10% glucose solution, 500 cc of normal saline solution and 100 mg of hydrocortisone were intravenously administered within 10 minutes. Following the initiation of glucocorticoid, great clinical improvement and resolution of laboratory parameters were achieved. On the hospital ward, the patient maintained stable blood pressure, no hypoglycemia, and felt well without any suggestive clinical signs of adrenal insufficiency, so the dose of hydrocortisone was gradually reduced until 24 hours before discharge. Basic education about adrenal insufficiency and the need for dose adjustment or intramuscular/intravenous injection in special situations was provided during the stay. Furthermore, the use of a medical alert bracelet was emphasized.
Outcome and Follow-Up
At follow-up visits, the patient reports feeling acceptably well but with persistent asthenia. Medication was modified until she was on a daily dose of 20 mg hydrocortisone and 0.1 mg fludrocortisone. In addition, the insulin pump parameters have been gradually adjusted by decreasing the g/UI ratio, lowering the target to 100 mg and reducing the duration of active insulin time to 2 hours. In October 2024, the ambulatory glucose profile report showed TIR 74%, TBR 1%, and TAR 24%, with total insulin requirements much higher than before diagnosis, from 30 units in March 2024 to 50 units daily 4 months later.
Autoimmunity studies were also carried out with positive 21-alpha hydroxylase antibodies.
Discussion
Patients with an autoimmune disorder, such as T1DM, are at an increased risk of developing another autoimmune disease, such as AD, AITD, celiac disease, pernicious anemia, or rheumatoid disease [5].
In its early stages, the symptoms of AD are nonspecific symptoms (eg, asthenia, appetite loss, nausea, orthostatic dizziness, weight loss), which can be shared with other conditions and usually occur insidiously. For this reason, the diagnosis and treatment of this disease may be delayed. Since the risk of developing AD is 10 times higher in patients with T1DM than in the general population, and the combination of concomitant T1DM and AD is associated with a more than 4-fold increase in the risk of premature death compared with patients with T1DM alone, awareness of this risk is of paramount importance to physicians managing patients with T1DM [6, 7].
The tendency toward hypoglycemia can alert the doctor to the onset of adrenal insufficiency. The mechanisms of hypoglycemia in primary adrenal insufficiency are primarily due to cortisol deficiency. Cortisol is a key counter-regulatory hormone that promotes hepatic gluconeogenesis and glycogenolysis while opposing insulin action in peripheral tissues. In its absence, glucose production is impaired during fasting and stress, insulin sensitivity is increased, and the mobilization of amino acids and free fatty acids is reduced, all of which predispose to hypoglycemia. A lack of cortisol also compromises the normal counter-regulatory response to hypoglycemia, since cortisol is necessary for catecholamines and glucagon to have their full effect in raising blood glucose [8].
As highlighted in this case, upon a periodic and careful evaluation of the AID data of our patient, the insulin requirements were decreasing significantly, and this allowed us to suspect the onset of AD. As a result, follow-up visits were more frequent, with special attention to insulin needs, blood pressure, and Na and K levels. Hyperpigmentation of the skin and mucosal surfaces was also observed, representing the most specific sign of latent chronic primary adrenal insufficiency. Acute presentation of AD can be precipitated by physiologic stress such as surgery, trauma, or intercurrent infection, and the symptoms include hypoglycemia, hypotension, and dehydration. Many patients are diagnosed in the emergency room in a very poor general condition, as was the case with this patient.
The patient was using a MM780g, an AID system that adjusts basal insulin delivery in real time and automatically delivers correction doses based on sensor-derived glucose levels. The endocrinologist can also make adjustments to improve the algorithm. As is demonstrated with this patient, hypoglycemia was very low during the period preceding the adrenal crisis. This demonstrates how well and quickly the Smartguard technology from the MM780g adapts to the patient's insulin needs [9].
After initiation of glucocorticoids, case reports often mention an increase in insulin requirements to previous levels and a reversal of hypoglycemia after initiation of glucocorticoid replacement therapy, which was also observed in this patient [10]. Less described is the increase in glycemic variability and the risk of hypoglycemia leading to increased coefficient of variation, as in this case. Also, a recently published case of a young patient with type 1 diabetes and concomitant adrenal insufficiency showed augmented glycemic lability due to exquisite insulin sensitivity [11]. There is also a significantly different pattern of insulin requirements in patients with primary adrenal insufficiency and T1DM compared with patients with T1DM alone. While the percentage of daily coverage with basal insulin was lower in patients with adrenal insufficiency, the insulin requirement for prandial glycemic coverage was higher in patients with APS2 [12]. In patients using insulin pump therapy, this is evidenced by an increased insulin-to-carbohydrate ratio and total daily insulin dose. This phenomenon can be better controlled when extended-release hydrocortisone is prescribed, but it is not yet commercialized in our country.
Learning Points
Patients with T1DM may develop AD even though it is rare, so it is important to be vigilant to prevent a life-threatening adrenal crisis.
The progressive decrease in total daily insulin doses together with a high risk of hypoglycemia should raise suspicion of adrenal insufficiency.
The MM780g system, together with continuous monitoring, can adapt to this situation in the weeks before and after the onset of an adrenal crisis.
Patients with T1DM and adrenal insufficiency usually require higher doses of postprandial insulin, related to the insulin resistance secondary to exogenous glucocorticoids.
Contributors
All authors made individual contributions. M.C.H., I.J.V., and P.M.N. were involved in the diagnosis and management of the patient. S.M.C., L.M.S., and N.S.M.G. performed the outpatient follow-up. S.M.C. was responsible for writing and submitting the manuscript. All authors reviewed and approved the final draft.
Abbreviations
- AD
Addison disease
- AID
automated insulin delivery
- AITD
autoimmune thyroid disease
- APS2
autoimmune polyendocrine syndrome type 2
- CGM
continuous glucose monitoring
- MM780g
Minimed780g system
- T1DM
type 1 diabetes mellitus
- TAR
time above range
- TBR
time below range
- TIR
time in range
Contributor Information
Sara Mera-Carreiro, Servicio de Endocrinología y Nutrición, Hospital HM Madrid Río, HM Hospitales, Madrid 28005, Spain; Instituto de Investigación Sanitaria HM Hospitales, Madrid 28005, Spain; Servicio de Endocrinología y Nutrición, Hospital Universitario Clínico San Carlos, Madrid 28040, Spain.
Inés Jiménez-Varas, Servicio de Endocrinología y Nutrición, Hospital Universitario Clínico San Carlos, Madrid 28040, Spain.
Laura Martínez-Suero, Servicio de Endocrinología y Nutrición, Hospital Universitario Clínico San Carlos, Madrid 28040, Spain.
Noelia Sánchez-Maroto-García, Servicio de Endocrinología y Nutrición, Hospital Universitario Clínico San Carlos, Madrid 28040, Spain.
Paz De-Miguel-Novoa, Servicio de Endocrinología y Nutrición, Hospital Universitario Clínico San Carlos, Madrid 28040, Spain.
Martín Cuesta-Hernández, Servicio de Endocrinología y Nutrición, Hospital Universitario Clínico San Carlos, Madrid 28040, Spain.
Funding
No public or commercial funding.
Disclosures
The authors have no relevant financial or non-financial interests to disclose. Informed Patient Consent for Publication.
Informed Patient Consent for Publication
Signed informed consent obtained directly from patient.
Data Availability Statement
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.
References
- 1. Sperling MA, Angelousi A, Yau M, et al. Autoimmune polyglandular syndromes. In: Feingold KR, Ahmed SF, eds. Endotext. MDText.com, Inc.; 2024. [Google Scholar]
- 2. Fernández Miró M, Colom Comí C, Godoy Lorenzo R. Autoinmune polyendocrinopathy. Síndromes pluriglandulares autoinmunes. Med Clin (Barc). 2021;157(5):241‐246. [DOI] [PubMed] [Google Scholar]
- 3. Armstrong L, Bell PM. Addison's disease presenting as reduced insulin requirement in insulin dependent diabetes. BMJ. 1996;312(7046):1601‐1602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Vinci F, d'Annunzio G, Napoli F, et al. Type 1 diabetes and Addison's disease: when the diagnosis is suggested by the continuous glucose monitoring system. Children (Basel). 2021;8(8):702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Chantzichristos D, Persson A, Miftaraj M, Eliasson B, Svensson AM, Johannsson G. Early clinical indicators of addison disease in adults with type 1 diabetes: a nationwide, observational, cohort study. J Clin Endocrinol Metab. 2019;104(4):1148‐1157. [DOI] [PubMed] [Google Scholar]
- 6. Dittmar M, Kahaly GJ. Polyglandular autoimmune syndromes: immunogenetics and long-term follow-up. J Clin Endocrinol Metab. 2003;88(7):2983‐2992. [DOI] [PubMed] [Google Scholar]
- 7. Betterle C, Dal Pra C, Mantero F, Zanchetta R. Autoimmune adrenal insufficiency and autoimmune polyendocrine syndromes: autoantibodies, autoantigens, and their applicability in diagnosis and disease prediction [published correction appears in Endocr Rev. 2002 Aug;23(4):579.]. Endocr Rev. 2002;23(3):327‐364. [DOI] [PubMed] [Google Scholar]
- 8. Lee SC, Baranowski ES, Sakremath R, Saraff V, Mohamed Z. Hypoglycaemia in adrenal insufficiency. Front Endocrinol (Lausanne). 2023;14:1198519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. American Diabetes Association Professional Practice Committee . 7. diabetes technology: standards of care in diabetes-2024. Diabetes Care. 2024;47(Suppl 1):S126‐S144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Chantzichristos D, Persson A, Eliasson B, et al. Mortality in patients with diabetes mellitus and Addison's disease: a nationwide, matched, observational cohort study. Eur J Endocrinol. 2017;176(1):31‐39. [DOI] [PubMed] [Google Scholar]
- 11. Burke G, Emanuel B. Addison's disease and diabetes Mellitus in an adolescent. Clin Pediatr (Phila). 1965;4:543‐547. [DOI] [PubMed] [Google Scholar]
- 12. Chantzichristos D, Eliasson B, Johannsson G. MANAGEMENT OF ENDOCRINE DISEASE disease burden and treatment challenges in patients with both Addison's disease and type 1 diabetes mellitus. Eur J Endocrinol. 2020;183(1):R1‐R11. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.


