Diabetic Ketoacidosis in a Pediatric Patient with Cystic Fibrosis–related Diabetes

Abstract

Cystic fibrosis (CF), a genetic disorder caused by pathogenic variants in the CFTR gene, is associated with various complications including cystic fibrosis–related diabetes (CFRD). CFRD is an entity distinct from type 1 or type 2 diabetes. We report a rare case of diabetic ketoacidosis (DKA) in a pediatric patient with CFRD, occurring during a significant pulmonary exacerbation. The patient's management involved addressing fluid and electrolyte imbalances, careful monitoring of nutritional status, and correction of hyperglycemia with insulin. This case serves as a reminder to consider DKA in the differential diagnosis of patients with CF presenting with respiratory distress, even in the absence of typical symptoms such as polyuria and polydipsia.

Introduction

Cystic fibrosis–related diabetes (CFRD) affects a substantial proportion of individuals with cystic fibrosis (CF). It is present in 2% of children, 19% of adolescents, and 40% to 50% of adults with CF [1]. CFRD shares features with both type 1 and type 2 diabetes. Pathogenic variants of the CFTR gene cause thick secretions, which lead to β-cell destruction and pancreatic insufficiency, but typically CFRD differs from type 1 diabetes in that the loss of pancreatic islet β-cells is not complete [2]. Conversely, unlike the more severe insulin resistance in type 2 diabetes, in CFRD, there is a mild to moderate baseline insulin resistance that intermittently worsens during acute illnesses [2].

Patients with CFRD have a mortality rate 6 times higher than patients with CF alone [3]. Newer studies indicate a reduced mortality risk related to CFRD that is attributed to early detection and intervention with insulin. Insulin, a potent anabolic compound, enhances the nutritional well-being of individuals with CFRD. Furthermore, insulin aids in controlling blood glucose, potentially decreasing the occurrence of pulmonary infections and enhancing pulmonary function [4].

Whereas diabetic ketoacidosis (DKA) is a common complication of type 1 diabetes and, less frequently, type 2 diabetes, it is a rare complication of CFRD [5]. Patients with CFRD are less likely to develop ketosis than those with type 1 diabetes, presumably due to residual insulin production and relative glucagon deficiency [6]. We present a case of DKA in a pediatric patient with known CFRD in the absence of underlying type 1 diabetes [7].

Case Presentation

A 14-year-old girl with CF and CFRD was admitted to the hospital for a CF exacerbation after failing outpatient therapy. She had been diagnosed with CF after newborn screening and later genotyped as deltaF508/R75X. She had been admitted for exacerbations 22 times since her initial CF diagnosis. She initiated therapy with ivacaftor/tezacaftor/elexacaftor, a CFTR modulator, 3 years prior to admission. Her home oxygen saturation levels were 95% to 96% on room air, and she had not required supplemental oxygen at home since initiating ivacaftor/tezacaftor/elexacaftor. Her past medical history was complicated by a diagnosis at age 7 of CRFD requiring insulin, pancreatic insufficiency requiring pancreatic enzyme supplements, and mild obstructive sleep apnea. Her CFRD was managed with a regimen of insulin glargine (34 units at bedtime), an insulin-to-carbohydrate ratio of 1:12 using lispro or aspart, and “sliding scale” (correction) insulin of 1 unit for every 50 the blood glucose level was over 150 mg/dL (8.32 mmol/L) (normal levels 70-106 mg/dL or 3.88-5.88 mmol/L). Her home blood glucose levels were typically around 150 mg/dL (8.32 mmol/L) when well controlled but had been in the upper 200s for the 2 weeks prior to admission.

At an outpatient visit the day before admission, she reported approximately 10 days of low-grade fevers, with a maximum temperature of 100.9°F, productive cough with greenish sputum, shortness of breath, chest tightness, body aches, chills, decreased energy, and reduced food intake. She had lost more than 5 pounds over the previous month. She denied experiencing polyuria, polydipsia, or missed doses of insulin. She had received 14 days of trimethoprim-sulfamethoxazole but had stopped taking ivacaftor/tezacaftor/elexacaftor 1½ weeks prior to her presentation. Her evaluation revealed a CF exacerbation with a significant decrease on her pulmonary function tests in forced expiratory volume in 1 second of 46% from a baseline of 71%. Accordingly, she was admitted for escalation of treatment.

Diagnostic Assessment

On admission, her physical exam was significant for a heart rate of 129/min, respiratory rate of 26/min, and oxygen saturation 94% on room air. Her breathing was rapid and deep, with diffuse crackles and decreased aeration on auscultation. Her mental status was intact, and she was able to converse appropriately.

Due to worsening respiratory distress requiring use of a venturi mask at 40% fraction of inspired oxygen, further laboratory studies were obtained. These revealed a metabolic acidosis (pH: 7.27, pCO₂: 34 mmHg, HCO₃⁻: 16 mEq/L, base excess: −11 mmol/L) (normal values pH: 7.35-7.45, pCO₂: 35-59 mmHg, HCO₃⁻: 18-26 mEq/L, base excess: −4 to −2 mmol/L), hyperglycemia (serum glucose: upper 300s), and ketosis (β-hydroxybutyrate: 5.3 mmol/L), establishing a diagnosis of DKA. Her hemoglobin A1c (HbA1C) was 10.2% (88 mmol/L), an increase from 8.7% (72 mmol/L) and 7.8% (62 mmol/L) 6 and 9 months prior, respectively. Her renal function was unremarkable with a blood urea nitrogen of 6 mg/dL (normal 6-20 mg/dL) and creatinine of 0.72 mg/dL (normal 0.57-0.8 mg/dL), along with electrolyte levels within normal levels [sodium 134 mEq/L (134-146 mEq/L), potassium 3.8 mEq/L (3.5-5 mEq/L), chloride 106 mEq/L (98-106 mEq/L)] with anion gap of 12 mEq/L (7-16 mEq/L).

A sepsis workup was initiated with blood, urine, and CF cultures. Autoantibodies were obtained to rule out type 1 diabetes. Her initial chest X-ray showed bronchial wall thickening and bronchiectasis with consolidation in the right superior perihilar region and more subtly in the left lung base, concerning for a superimposed pneumonia.

Treatment

She was transferred from the pulmonary service to endocrinology for management of DKA with appropriate fluids and insulin according to our institutional protocol [8]. After showing clinical and laboratory improvement, she was converted to a subcutaneous basal-bolus insulin regimen with glargine 38 units nightly, insulin-to-carbohydrate ratio of 1:10, and sliding scale insulin of 1 unit for every 50 the blood glucose was over 150 mg/dL (8.32 mmol/L).

She was transferred back to the pulmonary service, her ivacaftor/tezacaftor/elexacaftor was restarted, and cefazolin was initiated pending final culture results, given her history of methicillin-sensitive staphylococcus aureus in previous CF cultures. Her sepsis workup was negative aside from her CF cultures, which grew achromobacter and methicillin-sensitive staphylococcus aureus sensitive to piperacillin-tazobactam. She was transitioned to piperacillin-tazobactam for 16 total days of antibiotics.

Outcome and Follow-up

She was discharged home after she completed the antibiotic course and showed improvement in her pulmonary function to near baseline. Her diabetes (glutamatic acid decarboxylase and islet antigen-2) autoantibodies and thyroid peroxidase and thyroglobulin antibodies were all negative. On subsequent clinic follow-up in 4 weeks, she had gained 3.0 kg and was clinically doing well.

Discussion

DKA in a pediatric patient with CF is a clinical anomaly as only 4 pediatric and adult cases have been reported (Table 1). The rarity of DKA in CF may be attributed to the residual insulin secretion that is often seen in CFRD, which may be sufficient to prevent ketosis [5]. Our patient tested negative for islet autoantibodies, thus ruling out type 1 diabetes. Of note, we did not assess insulin antibodies because they are present in almost all patients who have been treated with exogenous insulin, regardless of the type of diabetes [8]. We did not measure C-peptide during the hospitalization because DKA is invariably associated with insulin deficiency. There are too few cases of DKA in patients with CF to draw conclusions regarding specific risk factors, but poor glycemic control (high HbA1c) is a potent risk factor for DKA in children with type 1 diabetes and was likely a factor with our patient [11].

Table 1.

Current case reports of pediatric and adult patients with CF presenting with diabetic ketoacidosis

Case	Age	Sex	Presentation	Diagnosis	Treatment	Citation
1	11	Male	CF without history of CFRD presenting with ketoacidosis	New-onset type 1 diabetes (antibody-positive) with DKA	Insulin therapy, CF management	Atlas et al [9]
2	17	Female	CF and CFRD presenting with ketoacidosis	DKA complicated by cerebral edema in a patient with CFRD	Insulin therapy, antibiotics	Swartz and Laffel [10]
3	21	Female	Patients with CF with CFRD presenting with ketoacidosis	Pulmonary infection with laboratory tests consistent with DKA. HbA1c 12% at presentation. Ruled out type 1 diabetes	Insulin therapy, antibiotics	Eenkhoorn et al [5]
4	45	Male	Nausea and vomiting for 3 weeks in DKA without a previous diagnosis of CF	Sweat test performed after ruling out type 1 diabetes and genetic testing later revealed CF	Insulin therapy	Sugrue et al [7]

Abbreviations: CF, cystic fibrosis; CFRD, cystic fibrosis–related diabetes; DKA, diabetic ketoacidosis; HbA1c, hemoglobin A1c.

Although DKA is rare in CFRD, it is an important differential to consider when patients with CF present with respiratory distress, particularly when the initial blood gas indicates metabolic acidosis. Potential triggers of DKA in patients with CFRD include recent glucocorticoid use, infection, and insulin discontinuation. Patients with CFRD are often treated with glucocorticoids, which can lead to hyperglycemia due to increased hepatic gluconeogenesis, altered glucose uptake in adipose tissue, and receptor-related changes [12]. While our patient had no recent glucocorticoid use or reported insulin discontinuation, she had a significant infection and recent interruption of her ivacaftor/tezacaftor/elexacaftor. Lower respiratory tract infections are recognized triggers for DKA in pediatric patients with diabetes mellitus. Infections elevate counterregulatory hormones including cortisol and catecholamines, leading to insulin resistance and hyperglycemia, which can precipitate DKA [13]. The fact that patients with CF can develop DKA demonstrates that even partial insulin deficiency, when coupled with an acute stressor such as infection, may precipitate this critical condition by increasing insulin resistance.

Symptoms of diabetes such as polyuria, polydipsia, and weight loss are not reliable for diagnosing diabetes in patients with CF, as these patients may not show symptoms initially. The 2-hour oral glucose tolerance test is the standard screening method for CFRD [14]. HbA1c is not a sensitive screening test for CFRD because it tends to be lower than in other individuals for a given level of blood glucose due to high red blood cell turnover. However, tracking HbA1c levels in individual patients over time can be useful, as increasing values may indicate the need for more thorough evaluation of glycemic control.

The management of DKA in CF does not differ significantly from standard DKA treatment. However, it is important to address hyperglycemia management while avoiding rapid shifts in osmolality, given the potential for worsening pulmonary status from fluid overload.

Learning Points

• Practitioners should consider DKA in the differential diagnosis of patients with CF presenting with respiratory distress, hyperglycemia, and acidosis, even in the absence of typical symptoms such as polyuria and polydipsia.

• The interruption of ivacaftor/tezacaftor/elexacaftor, a CFTR modulator, may have contributed to this patient's worsening lung infection, leading to DKA.

• This case emphasizes the need for continuous medication adherence and close monitoring of patients with CF on CFTR modulators, especially when managing comorbidities like CFRD.

Contributors

All authors made individual contributions to authorship. S.H., T.M., and P.C.W. were involved in the diagnosis and management of this patient and manuscript submission. All authors reviewed and approved the final draft.

Funding

No public or commercial funding.

Disclosures

S.H., T.M., and P.C.W. have no disclosures.

Informed Patient Consent for Publication

Signed informed consent could not be obtained from the patient or a proxy but has been approved by the treating institution.

Data Availability Statement

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.