A case report of insulin edema associated with insulin icodec

Abstract

Summary

Insulin edema is a rare complication that may occur after the initiation of insulin therapy. Although it usually resolves promptly after discontinuation of insulin, severe cases can lead to pulmonary edema, congestive heart failure, and acute renal failure. Insulin icodec is a recently marketed once-weekly, long-acting basal insulin, and its clinical use is increasing in Japan because it offers the convenience of once-weekly administration with an acceptable safety profile. In this report, we describe what is, to our knowledge, the first case of edema associated with insulin icodec. An 82-year-old man with glucocorticoid-induced diabetes mellitus in the setting of pre-existing impaired glucose metabolism was started on insulin icodec because of worsening glycemic control and developed bilateral leg edema a few days after the first injection. As he had no signs of heart failure, liver disease, or renal impairment, insulin icodec-induced edema was suspected, and insulin icodec was discontinued. Notably, it took more than one month after discontinuation of insulin icodec for the edema to improve. This case suggests that, compared with edema caused by conventional insulin preparations, insulin icodec-associated edema may show a delayed course of resolution. A closer and longer observation should be considered when initiating insulin icodec, particularly in insulin-naive people or in those with suboptimal glycemic management.

Learning points

• Insulin edema is a diagnosis of exclusion and is characterized by improvement following discontinuation of insulin.

• Insulin icodec is a once-weekly insulin that provides relatively gradual and safe glucose lowering, but insulin icodec can also cause insulin edema.

• Edema associated with insulin icodec does not differ from conventional insulin edema in terms of diagnosis or treatment.

• Edema associated with insulin icodec may appear to persist longer after insulin discontinuation than conventional insulin edema, and a longer period of observation may, therefore, be considered.

Background

Insulin edema is an uncommon complication that may occur shortly after the initiation or intensification of insulin therapy. Although its frequency is presumed to be even lower in the modern era owing to advances in the treatment of diabetes mellitus, some studies have reported the incidence of insulin edema to be 3–3.5% (1). However, the true incidence may be underestimated because of limited diagnostic awareness (2).

The pathophysiology of insulin edema is not fully understood, but several mechanisms have been proposed, including renal sodium retention, vasodilatory effects mediated by nitric oxide, microvascular injury from chronic hyperglycemia, activation of the renin–angiotensin–aldosterone system, and rapid osmotic shifts during the correction of hyperglycemia (1, 2, 3).

Insulin edema is clinically diagnosed by excluding other possible causes of edema, such as heart failure, renal failure, and liver failure, and is typically characterized by resolution of edema within a few days after discontinuation of insulin therapy (3). Reported risk factors for insulin edema include type 1 diabetes mellitus, newly diagnosed diabetes mellitus, failure to achieve glycemic targets, and rapid correction of hyperglycemia (2, 3, 4). In most cases, insulin edema is transient and self-limiting, but severe cases can result in pleural effusion, ascites, or congestive heart failure (2, 5). Thus, together with the possibility of missed or delayed clinical recognition, vigilant clinical observation is warranted to enable prompt diagnosis and treatment of insulin edema.

Insulin icodec is a novel once-weekly basal insulin analog designed to provide a stable glucose-lowering effect over seven days through strong, reversible albumin binding and a prolonged half-life of approximately 196 h (6). Clinical trials in patients with type 2 diabetes have demonstrated glycemic control comparable to that achieved with once-daily basal insulin, with an acceptable safety profile (7). Insulin icodec exerts a more gradual glucose-lowering effect because of its prolonged half-life and reduced insulin receptor binding affinity. To our knowledge, no cases of insulin edema associated with insulin icodec have been reported to date. In fact, a previous report described a person who developed edema with various types of insulin, in whom the introduction of insulin icodec led to an improvement in insulin edema (8). To date, to our knowledge, no cases of insulin edema associated with insulin icodec have been reported. Herein, we report a case of insulin edema that developed after the initiation of insulin icodec, highlighting its distinctive clinical course, which appears to differ from that of previously reported cases of insulin-induced edema.

Case presentation

An 82-year-old man was diagnosed with interstitial pneumonia at the age of 70 and started on oral prednisolone. At that time, his hemoglobin A1c (HbA1c) was approximately 6.0%, which was consistent with prediabetes. Four years after initiating steroid therapy, his HbA1c levels began to increase, and at the age of 78, his HbA1c rose to 8.0%. He was, therefore, referred to our clinic, and metformin in combination with repaglinide was initiated under the diagnosis of glucocorticoid-induced diabetes mellitus in the setting of pre-existing impaired glucose metabolism. Other major past medical history, in addition to interstitial pneumonia, included hypertension, dyslipidemia, atrial fibrillation, myocardial infarction, abdominal aortic aneurysm, and a thyroid tumor. He had no family history of diabetes. He had preserved cognitive function and was independent in activities of daily living. Oral prednisolone was titrated between 2 and 10 mg/day depending on the activity of interstitial pneumonia. After the age of 79, he was treated with metformin 500 mg/day and dulaglutide 0.75 mg/week, with the dose of repaglinide adjusted between 0.25 and 0.75 mg/day according to blood glucose levels, and his HbA1c was maintained in the low 6.0% range. Although his body weight fluctuated considerably, even at its highest, it was approximately 73 kg, corresponding to a BMI of about 24 kg/m². He had not been previously noted to have lower limb edema. At age 82, he was hospitalized twice for infections, during which steroid intravenous pulse therapy was administered and his oral prednisolone dose was subsequently increased (up to 70 mg/day). As hyperglycemia developed after the increase in steroid dosage, a sliding-scale insulin regimen using regular insulin (Humulin R) was initiated. This was his first exposure to subcutaneous insulin therapy, and no adverse events, including edema, were observed until discharge. At discharge, the dose of oral prednisolone had been tapered to 12.5 mg/day. After these hospitalizations, his HbA1c rose from 7.0 to 8.6% within one month, and insulin icodec was initiated after discharge in addition to metformin 500 mg/day, dulaglutide 0.75 mg/week, and repaglinide 0.25 mg/day. His concomitant medications at that time included mycophenolate mofetil, nintedanib, esomeprazole, ezetimibe, Clostridium butyricum, ursodeoxycholic acid, and warfarin. This was his first fixed-dose insulin regimen. One week before the initiation of insulin icodec therapy, he had no edema, whereas his HbA1c was 8.7% and fasting plasma glucose was 174 mg/dL (SI: 9.7 mmol/L) (reference range: 70–110 mg/dL (SI: 3.9–6.0 mmol/L)). In the first week, he received 40 units of insulin icodec on day 0, with a planned increase to 80 units from the second week onward. Several days after the initial injection of insulin icodec, bilateral lower-leg pitting edema suddenly appeared. At the outpatient visit on day 11, pronounced bilateral lower-limb edema was observed.

Investigation

Laboratory tests, imaging studies, and other examinations were performed. Laboratory findings on day 11 are summarized in Table 1. Proteinuria and renal dysfunction were not observed. Serum albumin levels were mildly low, but they had remained at a similar level for the past few years and were not low enough to explain the sudden onset of edema. N-terminal prohormone of brain natriuretic peptide (NT-proBNP) was within the normal range, and chest radiography and echocardiography showed no evidence of heart failure. Thyroid function was normal. At the time the edema appeared, he was taking 11 mg of prednisolone and was in the tapering phase, which made steroid-induced edema less likely. HbA1c was 8.7% on day −7 but decreased to 8.0% on day 11, indicating a relatively rapid improvement in glycemia over a short period. Consistently, the fasting plasma glucose on day 11 was 77 mg/dL. Following suspicion of insulin-induced edema, insulin icodec was discontinued after the second dose and was not resumed thereafter. However, the edema did not improve immediately. On day 35, the patient’s body weight was 75.5 kg (+2.2 kg compared with the pre-insulin baseline), and the extracellular water to total body water (ECW/TBW) ratio assessed by bioelectrical impedance analysis (InBody720; InBody Japan, Japan) was 0.403, indicating potential fluid overload. Because the edema progressively worsened, hospitalization for further evaluation was planned. On admission (day 47), his height was 173.5 cm and body weight 76.2 kg (BMI: 25.3 kg/m²). Vital signs were stable: temperature 36.3°C, blood pressure 121/62 mmHg, and heart rate 76 beats/min. Ultrasonographic examination showed no evidence of deep vein thrombosis. Blood tests showed normal serum immunoglobulin levels and negative autoantibodies. A 24-h urine collection performed from the day of admission to the following day showed a urinary sodium excretion of 2.5 g/day (reference range: 1.2–4.6 g/day).

Table 1.

Laboratory findings 11 days after the first administration of insulin icodec to investigate the cause of edema. No obvious abnormalities suggestive of an alternative cause of edema were identified.

Test	Values	Reference range
Complete blood count
WBC, ×109/L	9.63	3.30–8.60
Hemoglobin, g/L	136	130–176
Platelet, ×109/L	226	150–350
Biochemistry
AST, U/L	17	13–33
ALT, U/L	22	8–42
LDH, U/L	254	120–230
ALP, U/L	65	38–113
γ-GTP, U/L	128	10–47
Total protein, g/L	60	66–81
Albumin, g/L	36	41–51
BUN, mmol/L	6.1	2.9–7.1
Creatinine, μmol/L	64	57–95
Sodium, mmol/L	141	138–145
Potassium, mmol/L	3.5	3.6–4.8
LDL-C, mmol/L	5.41	<3.63
HDL-C, mmol/L	1.61	≥1.04
Triglycerides, mmol/L	2.33	<1.69
Glucose, mmol/L	4.27	3.9–6.0
HbA1c, %	8.0	4.6–6.2%
Thyroid function
TSH, mIU/L	1.600	0.5–5.0
Free T4, pmol/L	19.8	9.0–20.6
Free T3, pmol/L	5.27	3.53–6.14
Other tests
NT-proBNP, ng/L	84	<125
Urinalysis (qualitative)
Protein	Negative	—
Glucose	Negative	—
Occult blood	Negative	—

WBC, white blood cell count; AST, aspartate aminotransferase; ALT, alanine aminotransferase; LDH, lactate dehydrogenase; ALP, alkaline phosphatase; γ-GTP, γ-glutamyl transpeptidase; BUN, blood urea nitrogen; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; HbA1c hemoglobin A1c; TSH, thyroid-stimulating hormone; Free T3, free triiodothyronine; Free T4, free thyroxine; NT-proBNP, N-terminal pro-B-type natriuretic peptide.

Treatment

Other than discontinuing icodec after the second week, no therapeutic interventions, including diuretics, were administered.

Outcome and follow-up

On physical examination at admission, bilateral lower-leg pitting edema was still present, although the patient stated that it had begun to improve several days before admission. He was taking oral prednisolone 8 mg/day. On the day after admission (day 48), his body weight was 74.9 kg and ECW/TBW was 0.397, and the edema was confined to the distal parts of the lower extremities, below the ankles (Fig. 1A). On the fourth hospital day (day 50), his body weight was 73.9 kg, and ECW/TBW was 0.395; the edema had resolved spontaneously without any specific treatment (Fig. 1B). Given that the edema resolved without additional treatment after discontinuation of insulin icodec, the diagnosis of insulin icodec-induced edema was established. The clinical course up to this point is shown in Fig. 2. During hospitalization, no significant hyperglycemia was observed with his pre-admission antidiabetic medications. Therefore, no modifications to diabetes therapy were considered necessary. He has continued to be followed in the outpatient clinic, and at six months after discharge, no worsening of glycemic control or recurrence of edema has been observed.

Figure 1.

Clinical appearance of the toe edema before and after resolution. (A) Bilateral lower-limb pitting edema observed on day 49. (B) Resolution of edema on day 50.

Figure 2.

Clinical course showing glycemic control and the timeline of insulin therapy, steroid dose, and edema appearance.

Discussion

We report, to our knowledge, what appears to be the first case of insulin icodec-induced edema. The patient developed bilateral lower-leg edema a few days after initiating insulin icodec. Other causes of edema – including heart failure, renal failure, liver failure, endocrine disorders, and deep vein thrombosis – were excluded. Although he was taking prednisolone, the edema appeared during the tapering phase, and he had not developed edema when receiving higher doses in the past; therefore, steroid-induced edema was considered unlikely. Moreover, although it required more than one month, the edema resolved after discontinuation of insulin icodec without specific therapeutic intervention. Based on these observations, a diagnosis of insulin icodec-induced edema was established.

Insulin edema has been reported to occur more frequently in people with specific predisposing backgrounds. It is considered more likely to develop in people with type 1 diabetes, in those newly starting insulin therapy, in people whose glycemic targets have not yet been achieved or who have suboptimal glycemic management, and in those who experience a rapid decline in glucose levels after treatment initiation (2, 3, 4). Consistently, in the present case, insulin icodec was introduced as the first fixed-dose insulin regimen (excluding sliding-scale insulin), and a rapid reduction in blood glucose levels was observed after its initiation. Although insulin icodec is a long-acting insulin that is generally considered to exert a more gradual glucose-lowering effect than conventional insulins (7), determining the appropriate dose can be challenging when it is used as the initial insulin regimen. Consequently, inadvertent overdosing may have occurred, leading to excessive glucose reduction. In this respect, the present case shares characteristics with previously reported cases of conventional insulin-induced edema.

Importantly, the clinical course in this case suggests that edema caused by insulin icodec may share certain similarities with, yet also differ from, edema caused by conventional insulin therapy. Similar to typical conventional insulin-induced edema, insulin icodec-induced edema in this case appeared within a few days after the initiation of insulin and resolved completely within several days once improvement began. However, the major difference lies in the duration of edema before improvement became evident. Okamura et al. summarized previously reported cases of insulin edema (9). Among these cases, we reviewed those in which insulin was discontinued or switched to another insulin formulation and examined the time from insulin discontinuation or switching to improvement of edema. In all such cases, edema improved within several days, with the longest reported duration being approximately 9 days. In contrast, insulin icodec-induced edema in the present case required several weeks before improvement became evident. Interestingly, after discontinuation of insulin icodec, it took approximately 40 days for the edema to begin improving. This delayed improvement can likely be attributed to the prolonged pharmacokinetic profile of insulin icodec. As shown in Fig. 3, the estimated serum insulin icodec concentration in this case suggests that the drug remained detectable even 40 days after the initial injection, which corresponds to the timing of the rapid resolution of edema observed in this case. In addition, Fig. 3 merely represents the pharmacokinetic profile in a standardized individual. Serum drug concentrations can be influenced by individual differences in body composition. In patients with edema and presumed reduced lean body mass, even with the same dose per body weight, the effective volume of distribution may be smaller than that in a standardized individual, potentially leading to higher serum concentrations (10). In the present case, reduced lean body mass may have contributed to a more prolonged decline in serum insulin icodec concentrations than that estimated by the model shown in Fig. 3.

Figure 3.

Simulated serum concentration–time profile of insulin icodec following administration of 40 units on day 0 (week 1) and 80 units on day 7 (week 2) in a 75 kg patient, modeled over 50 days. The profile was generated using a one-compartment pharmacokinetic model with first-order absorption (ka = 0.175 h⁻¹) and elimination (t½ ≈ 196 h). Concentrations are displayed in relative units normalized by body weight (U/kg).

To our knowledge, edema associated with insulin icodec has not been previously reported; therefore, caution is needed when generalizing the findings of the present case. In this case, we think a relatively high dose of insulin icodec for the patient may have contributed to a marked reduction in glucose levels, leading to insulin edema. However, the possibility that the pharmacological properties of insulin icodec contributed to the clinical course cannot be excluded, and further similar cases need to be accumulated. Nevertheless, edema did occur after the initiation of insulin icodec in the present case. This case highlights the potential risk of insulin edema associated with insulin icodec. Clinicians should be cautious regarding insulin edema when initiating insulin icodec in insulin-naive people or in those with suboptimal glycemic management. In addition, insulin icodec-induced edema may require relatively long-term, close observation following discontinuation of insulin icodec.

Declaration of interest

DY and TM received consulting or speaker fees from Novo Nordisk. AN has nothing to declare.

Funding

This research did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.

Patient consent

Written informed consent for the publication of his clinical details and clinical images was obtained from the patient.

Author contribution statement

All authors made individual contributions to authorship. All authors were involved in the diagnosis and management of this patient. AN and TM were involved in writing the first draft and submission. TM and DY supervised the study, revised the manuscript critically for important intellectual content, and approved the final manuscript. All authors reviewed and approved the final draft.