Perioperative hypoglycemia is associated with neurological complications, cognitive dysfunction, and increased mortality, highlighting the need for appropriate glycemic control. This case report describes a patient with suspected insulinoma and recurrent hypoglycemia, in whom a left atrial tumor was incidentally discovered and resected under cardiopulmonary bypass. Both hypoglycemia due to insulinoma and hyperglycemia related to surgery and cardiopulmonary bypass were considered. Stable glycemic control was achieved through blood glucose monitoring every 15 minutes and continuous infusion of 10% glucose. Although the outcome was favorable, individualized blood glucose management is essential during cardiac surgery for patients with suspected insulinoma.
Perioperative hypoglycemia is associated with neurological complications, cognitive dysfunction, and increased mortality, highlighting the need for glycemic control. Insulinoma, a rare pancreatic neuroendocrine tumor, is a well-known cause of recurrent hypoglycemia. In suspected insulinoma cases, anesthetic management during cardiac surgery with cardiopulmonary bypass (CPB) is challenging due to the dual risks of hypoglycemia from insulin overproduction and hyperglycemia from surgical stress. We report the case of a patient with suspected insulinoma who underwent resection of an incidentally discovered left atrial tumor under CPB. This case highlights the significance of individualized intraoperative glucose management in minimizing glycemic fluctuations. This manuscript adheres to the applicable EQUATOR guideline. Written informed consent was obtained from the patient for publication of this case report. Institutional Review Board approval was not required for this single case report. All potentially identifying information has been removed in accordance with HIPAA standards.
CASE DESCRIPTION
A 56-year-old woman (158 cm, 36 kg, body mass index, 14 kg/m) was noted to have hypoglycemia (40 mg/dL) during health checkups for 2 years, but had no symptoms. Two months before surgery, she was brought to the emergency department with aphasia, motor impairment, and altered mental status. On arrival, she had right hemiparesis with blood glucose level of 48 mg/dL. Head magnetic resonance imaging (MRI) and chest-abdominal-pelvic computed tomography (CT) were unremarkable. Her symptoms resolved rapidly with intravenous glucose, suggesting hypoglycemia as the primary cause. However, the subsequent discovery of left atrial myxoma raised the possibility of transient embolism.
A subsequent checkup revealed a fasting blood glucose (FBG) of 39 mg/dL and HbA1c of 4.8%, prompting referral to internal medicine. Upon evaluation, FBG was 40 mg/dL, C-peptide was 0.81 ng/mL, and serum insulin was 4 μU/mL, raising suspicion of insulinoma. Endoscopic ultrasonography revealed a 12-mm mass in the pancreatic head, and a rapid calcium infusion (RCI) test showed an increase in serum immunoreactive insulin (IRI). A selective arterial secretagogue injection (SASI) test was scheduled for definitive diagnosis. However, a cardiac evaluation incidentally revealed a left atrial mass, leading to cancelation of the SASI test.
Transthoracic and transesophageal echocardiography revealed a mobile mass measuring 42×58 mm in the left atrium, attached to the left atrial septum and anterior wall. The mass prolapsed into the left ventricle during diastole without evidence of mitral valve stenosis or regurgitation. Because of the suspected risk of embolism from the mass, surgical resection was planned.
PERIOPERATIVE MANAGEMENT
On the day before surgery, a continuous intravenous infusion of 10% glucose at 60 mL/h was initiated. On the morning of surgery, the FBG level was 65 mg/dL; therefore, 20 mL of 50% glucose was administered intravenously, increasing the level to 152 mg/dL. Subsequently, the infusion was changed to a crystalloid-based solution with added glucose (final concentration approximately 10%), and the patient entered the operating room while receiving a continuous intravenous infusion of 10% glucose at a rate of 100 mL/h. However, the glucose level dropped again to 50 mg/dL immediately at the start of anesthesia induction, requiring an additional 20 mL of 50% glucose to be added to the infusion bag. Since this occurred at the start of anesthesia induction, bolus administration was avoided to prevent excessive glucose administration.
Anesthesia was induced under 100% oxygen with midazolam (3 mg), fentanyl (100 μg), remifentanil (0.3 μg·kg·min), and rocuronium (30 mg). Maintenance anesthesia was achieved with sevoflurane (1.5%), remifentanil (0.2 μg·kg·min), and rocuronium (7 μg· kg·min). During CPB, propofol was administered via target-controlled infusion (TCI) at a concentration of 2 μg/mL.
Following the rapid administration of 150 mL of 10% glucose at induction, her blood glucose level increased sharply to 255 mg/dL. Immediately, the rapid infusion of 10% glucose was discontinued, and 10% glucose was initiated at 1 mL·kg·h. The 10% glucose was resumed at a fixed rate to maintain glycemic stability, while additional fluid requirements were managed using glucose-free intravenous solutions to prevent excessive glucose administration. Blood glucose was monitored every 15 min. During CPB, glucose infusion was temporarily halted. Except for a transient hyperglycemic episode (213 mg/dL) following cardioplegia, blood glucose levels remained within the controlled range (106-168 mg/dL) (Figure 1).
Surgery was performed via median sternotomy with arterial inflow through the ascending aorta and venous drainage from the superior and inferior vena cava. CPB was initiated under mild hypothermia (32°C) after aortic cross-clamping. Through a right-sided left atriotomy, a gelatinous mass originating from the interatrial septum was resected and the defect was patched. The mass was consistent with a left atrial myxoma. After aortic declamping, 10% glucose infusion at 1 mL·kg·h was resumed. The patient was weaned from CPB with dopamine and dobutamine at 3 to 4 μg·kg·min. Hemodynamics and glucose levels (146-158 mg/dL) remained stable.
Postoperatively, she was managed in the intensive care unit under mechanical ventilation with continued glucose infusion. Glucose remained around 160 mg/dL without hypoglycemia. Blood glucose was measured every 15 min until stable, allowing timely intervention if needed. One month postoperatively, insulinoma was confirmed by the SASI test, and is currently being managed with drug therapy using dapagliflozin.
DISCUSSION
Insulinoma is a rare neuroendocrine tumor of pancreatic β-cells that causes hypoglycemia due to inappropriate insulin secretion, with an estimated incidence of 1 to 4 cases per million annually. The diagnosis relies heavily on Whipple's triad -- symptoms or signs of hypoglycemia, documentation of low plasma glucose during those symptoms, and relief of symptoms following glucose administration -- and the presence of inappropriately elevated insulin levels during hypoglycemia. Hypoglycemic symptoms depend not only on glucose levels but also on individual threshold, rate of glucose decline, neuroadaptation, and other factors. Thus, in prolonged hypoglycemia, as in the present case, symptoms may not correlate with glucose levels. Although a definitive diagnosis was not confirmed in this case, the findings from diagnostic testing and an RCI test strongly suggested insulinoma. Therefore, meticulous glycemic management was required.
Glucose metabolism is influenced by various factors during cardiac surgery. Sevoflurane suppresses insulin secretion, while propofol increases insulin release but induces insulin resistance, particularly with lipid loading. These effects necessitate careful anesthetic selection.
Hypothermia during CPB also impacts glucose metabolism by reducing insulin secretion, impairing peripheral glucose utilization, and increasing resistance, all promoting hyperglycemia. In this case, mild hypothermia was maintained during CPB per institutional protocol. Additional CPB-related factors, including hyperoxia, neuroendocrine activation, heparin administration, hemodilution, and glucose-containing cardioplegia, also contribute to glycemic alterations. Consequently, glucose homeostasis during cardiac surgery is affected by dynamic factors, requiring close monitoring and flexible management. In our case, both hypoglycemic and hyperglycemic risks were anticipated, and predicting intraoperative glycemic fluctuations during CPB was challenging. Frequent blood glucose monitoring proved essential for successful control. Continuous glucose infusion contributed to stable glucose levels, and the decision to discontinue glucose at CPB initiation was considered appropriate. Discontinuing 10% glucose infusion represented a reasonable strategy that likely contributed to glycemic stability.
Although artificial pancreas monitoring is beneficial during insulinoma resection, its limitations include cost and operational complexity. Continuous glucose monitoring during surgery has been reported. However, due to factors such as intraoperative fluid shifts and vasopressor effects, sufficient accuracy has not been achieved. Therefore, the device was removed preoperatively.
Alternatively, continuous infusion of 10% glucose and frequent (every 15 minutes) blood glucose monitoring were used to maintain stability. While more frequent monitoring is ideal, 15-minute intervals were the practical limit.
The optimal glucose range for insulinoma patients remains unclear. Based on the NICE-SUGAR trial, a perioperative target glucose range of 70 to 200 mg/dL was established, increasing the upper limit from the originally recommended 180 mg/dL to allow greater flexibility during cardiac surgery and to reduce hypoglycemia. Intraoperatively, 10% glucose was infused at 1 mL·kg·h, with plans to increase the rate if blood glucose dropped below 70 mg/dL. However, no such drop occurred, and infusion rates remained unchanged except during CPB. No additional bolus doses were required. Glucose loading at induction is recommended in insulinoma. In our patient, continuation of preoperative glucose infusion into the induction period led to transient hyperglycemia, necessitating careful adjustments in both rate and dose.
Although intraoperative glucose metabolism impacts the glucose levels, frequent glucose monitoring avoids significant excursions. This case illustrates the complexity of perioperative glucose management during cardiac surgery in patients with suspected insulinoma. The coexistence of insulinoma and CPB-related hyperglycemia presents a unique challenge. No established guidelines exist. Individualized strategies balancing hypoglycemia risk and insulin secretion are essential. Our approach -- avoiding excessive glucose while maintaining safe levels during CPB -- was effective and may guide future management. More case accumulation and analysis are needed to develop standardized protocols.
CONCLUSION
This rare case involved a patient with suspected insulinoma and a tendency toward hypoglycemia who underwent surgical resection of an incidentally discovered left atrial myxoma under CPB. Intraoperatively, despite the dual risks of hypoglycemia and hyperglycemia, glucose levels were successfully maintained within an acceptable range through frequent blood glucose monitoring and tailored glucose administration. This case highlights the critical importance of perioperative glucose management in patients undergoing major cardiac surgery and underscores the need for individualized anesthetic planning based on thorough preoperative evaluation in patients with suspected insulinoma.
ACKNOWLEDGMENTS
We are grateful to the endocrinology team for their contribution to the patient's preoperative blood glucose control.
This manuscript was handled by: Kent H. Rehfeldt, MD, FACC, FASE.