Case Report: Resolution of hypoglycemia following cessation of sacubitril/valsartan in a non-diabetic patient

Sacubitril/valsartan is a drug combination that is used in the treatment of chronic symptomatic heart failure with reduced ejection fraction. Existing literature suggests sacubitril/valsartan improves glycemic control in diabetic patients, but may be associated with an increased risk of hypoglycemia.^{,  ,  ,}  However, little is known about the effect of sacubitril/valsartan on glucose homeostasis in non-diabetic patients.

We present a case of hypoglycemia in a non-diabetic patient, which resolved following cessation of sacubitril/valsartan.

A frail, Caucasian, 84-year-old female patient (BMI 29.6 kg/m²) presented to the emergency department in December 2020 following a collapse. Her past medical history included hypertension and heart failure with reduced ejection fraction (HFrEF). Her regular medications included sacubitril 24 mg/valsartan 26 mg twice daily, spironolactone 50 mg once daily and furosemide 40 mg once daily. On physical examination, the patient was peripherally shut down (blood pressure 70/50 mmHg, pulse 140/minute, temperature 34.4 °C) and clinically dehydrated. Severe hypoglycemia was detected (2.2 mmol/L). The patient had no history of diabetes mellitus (DM) (HbA1c ranging from 5.2% - 5.4% [33 – 37 mmol/mol] in four years prior to admission). Chest, cardiovascular and abdominal examinations were unremarkable and no focal neurological deficit was observed.

Blood tests demonstrated acute kidney injury (AKI) with hyperkalemia and mildly raised inflammatory markers (creatinine 154 μmol/L [baseline 80 μmol/L]), potassium 7.3 mmol/L, white cell count 12.7 x 10⁹/L, c-reactive protein 44 mg/L). Liver function tests and thyroid function tests were normal. An electrocardiograph, chest radiograph and computed tomography of the head were unremarkable.

The patient received fluid resuscitation, broad-spectrum antibiotics and treatment for hypoglycemia and hyperkalemia. In view of her AKI, the patient was catheterised and her regular medications were held. During the first week of her hospital admission, the patient improved dramatically with resolution of sepsis and AKI. Sacubitril/valsartan was restarted on day 7 since admission. However, the patient continued to experience fasting hypoglycemic episodes ranging from 2.2 to 3.7 mmol/L.

A short synthetic-ACTH test was requested, which demonstrated a normal response (baseline cortisol 383 nmol/L, post-short-synthetic-ACTH cortisol 645 nmol/L). A 72-hour fast was not conducted as the plasma glucose levels remained ≥ 2.1 mmol/L. The hypoglycemic events were found to occur during the early morning (04:00 – 09:00) and thus bed time snacks were introduced. However, hypoglycemia persisted despite this intervention. An evaluation of the patient’s drug history revealed sacubitril/valsartan and furosemide had been restarted on day 7 and 10 since admission, respectively. Existing literature highlighted that sacubitril/valsartan had been associated with hypoglycemia in diabetic patients with HFrEF. Following this review and consultation with the cardiology team, we decided to stop sacubitril/valsartan, and start candesartan as an alternative. The afternoon dose of sacubitril/valsartan was held on day 14 and the drug was stopped indefinitely following the morning dose on day 15 since admission. No further hypoglycemic events occurred after day 13 since admission, corresponding with the cessation of sacubitril/valsartan (see timeline 1). Spironolactone was restarted on day 16 since admission. The patient was discharged home and regular follow-up arranged with the outpatient heart failure team. No further episodes of hypoglycemia have been recorded.

This case describes the resolution of hypoglycemia in a non-diabetic patient after stopping sacubitril/valsartan. Our observations suggest there may be a possible correlation with exposure to sacubitril/valsartan and hypoglycemia in a patient without DM.

Existing literature has highlighted an association between sacubitril/valsartan and improved glycemic control in diabetic patients, with a background of heart failure. A post-hoc analysis of patients with diabetes and HFrEF within the PARADIGM-HF trial (n=3778), demonstrated a significant long-term reduction in HbA1_c concentrations in those receiving sacubitril/valsartan, compared to enalapril. After one year of treatment, the HbA1c concentration decreased by 0.26% (SD 1.40) in patients receiving sacubitril/valsartan and 0.16% (SD 1.25) in patients receiving enalapril (p=0.0023). Similar trends were identified in years 2 and 3. Additionally, insulin-naïve diabetic patients receiving sacubitril/valsartan were less likely to be started on insulin therapy, compared to those receiving enalapril, during the three years of follow-up (HR 0.71, p=0.0052). Contrastingly, no significant change in HbA1_c concentration was observed in non-diabetic patients (p=0.29).

Published case reports have also described an association between sacubitril/valsartan and reduced insulin requirements in patients with type 1 and 2 DM. Barry et al described the onset of hypoglycemia in a patient with type 1 DM 18 days after starting sacubitril/valsartan. The basal dose of insulin lispro was subsequently reduced from 10 units to 9 units and no further hypoglycemic episodes occurred. Gamarra et al observed a reduction in the total daily insulin requirement (basal and bolus) in a patient with type 2 DM after starting sacubitril/valsartan. During four months of follow-up, the patient’s daily insulin requirement reduced by 22% (from 100 to 78 units daily), due to frequent post-prandial hypoglycemic episodes. There was a greater reduction in the bolus insulin requirement compared to the basal insulin requirement (-28% versus -14%, respectively).

Sacubitril/valsartan is a drug combination of sacubitril, an inhibitor of neprilysin that allows accumulation of natriuretic peptide and valsartan, an inhibitor of angiotensin receptor binding. The underlying mechanism by which sacubitril/valsartan may influence glycemic control remains unclear. It has been hypothesized that higher concentrations of natriuretic peptides may improve glycemic control. Animal studies in diabetic and obese mice found that increased plasma concentrations of natriuretic peptide improved glucose control and insulin sensitivity in skeletal muscle. Observational work in humans has also identified an inversely proportional relationship between the concentrations of NH₂-terminal pro b-natriuretic peptide (NT-proBNP), a cleave product of BNP, and risk of new diabetes. The mechanism underlying the effect of natriuretic peptide on glucose metabolism has not been fully elucidated. Possible explanations have included the effect of natriuretic peptides on adiponectin secretion, a hormone with insulin-sensitising properties. It has also been hypothesised that sacubitril/valsartan may induce hypoglycemia through its influence on glucagon-like-peptide-1 (GLP-1) and subsequent insulin secretion. GLP-1 is an incretin hormone that plays a significant role in glucose homeostasis through its enhancement of insulin secretion in response to enteral feeding. Neprilysin, like dipeptidyl peptidase, can cleave and inactivate GLP-1. In-vitro studies have demonstrated that inhibition of neprilysin in mouse islets can increase GLP-1 mediated insulin secretion.

To the best of our knowledge, no previous studies have reported an association between hypoglycemia and sacubitril/valsartan in patients without DM. Our case describes a non-diabetic patient who presented with sepsis and hypoglycemia. The patient experienced severe hypoglycemia on the day of admission. Further hypoglycemic events occurred on day 2 of the hospital stay and daily from days 5 to 13 of admission. Sacubitril/valsartan was held on admission, in view of the patient’s AKI and reintroduced on day 7 since admission. Thus, we did observe a period of hypoglycemia, in the absence of sacubitril/valsartan exposure. We may speculate that the patient’s glucose homeostasis was initially disrupted as a result of the severe infection. However, whether the patient had a borderline tendency to abnormal glycemic control or specific interactions with other drugs or disease is not known.

It is also important to highlight that the patient had been taking sacubitril/valsartan for one year prior to admission and no previous history of hypoglycemia had been noted. We observed that upon cessation of sacubitril/valsartan, no further hypoglycemic episodes occurred. Previous studies and case reports in diabetic patients have reported changes in glycemic control within the first few months of starting sacubitril/valsartan.^, Our observations suggests that a consideration of stopping sacubitril/valsartan should be made in patients with unexplained hypoglycemia, irrespective of the duration of treatment prior to onset of hypoglycemia, and that formal evaluation of this agent's effect on blood glucose is warranted, in diabetic and non-diabetic patients.

The authors have no potential conflicts of interest to disclose.

Funding for the publication fee was supported by charitable funds by Cwm Taf Morgannwg University Health Board.