A Case of Post-Bariatric Hypoglycemia With Factitious Hypoglycemia

Abstract

Background/Objective

Diagnosing factitious insulin administration is challenging, especially in the rare instance where coexisting endogenous hypoglycemia is present. Delay in diagnosis results in higher morbidity and increases health care costs. This case highlights specific characteristics of both post-bariatric hypoglycemia (PBH) and factitious insulin use.

Case Report

A 32-year-old woman with an extensive medical history including Roux-en-Y gastric bypass developed episodes of symptomatic hypoglycemia 8 months after bariatric surgery. She experienced postprandial hypoglycemia preceded by hyperglycemia, consistent with PBH, confirmed by a mixed meal tolerance test. During a 3-month period, she was hospitalized for hypoglycemia numerous times. During a 72-hour fast, she experienced hyperinsulinemic hypoglycemia due to surreptitious insulin use yet her c-peptide was not completely suppressed.

Discussion

This is a unique case of concurrent factitious insulin use and PBH. Her c-peptide during fasting hypoglycemia was not suppressed as is expected with exogenous insulin use, yet her insulin to c-peptide ratio was elevated and her proinsulin level was low supporting surreptitious insulin administration.

Conclusions

This case exemplifies that surreptitious hypoglycemia should be considered in the differential diagnosis, especially when clinical features consistent with factitious disorder are present. Although c-peptide levels are typically <0.6 ng/mL (or <0.2 nmol/L) in insulin-induced hypoglycemia, this case highlights that this is not always present. Clinical presentation, along with proinsulin levels and the insulin to c-peptide ratio can provide additional insight into the underlying cause.

Highlights

• •Hypoglycemia due to bariatric surgery is primarily postprandial and worse after simple carbohydrate ingestion. It is important to consider other potential etiologies in patients with symptomatic fasting hypoglycemia
• •Typical features of factitious disorder include work in a healthcare field, a history of frequent visits to hospitals and/or providers, self-injury, atypical presentation, inconsistent laboratory testing, and no response to treatment
• •The insulin-to-c-peptide ratio is a helpful diagnostic tool to determine endogenous versus exogenous insulin use when c-peptide values are ambiguous and proinsulin results are delayed

Clinical Relevance

Although c-peptide levels are typically suppressed in insulin-induced hypoglycemia, this case highlights that equivocal c-peptide results are possible and the insulin to c-peptide ratio can provide additional insight into the underlying cause. In addition, although rare, it is possible to have more than one cause of hypoglycemia.

Introduction

Hypoglycemia in patients without diabetes is often challenging to diagnose. Considering the differential diagnosis in terms of hypoglycemia presenting in well-appearing compared to ill-appearing patients can be useful.¹ In well-appearing individuals, hypoglycemia is caused by insulinoma, post-bariatric hypoglycemia (PBH), noninsulinoma pancreatogenous hypoglycemia, insulin or insulin secretagogue use, and insulin autoimmune syndrome.¹ Causes of hypoglycemia in ill appearing individuals include end organ failure, sepsis, starvation, chronic alcohol abuse, adrenal insufficiency, and non-islet tumors.¹

We present a case of a well-appearing patient found to have hypoglycemia from both PBH and factitious insulin administration. Recognizing both etiologies of the patient’s hypoglycemia required understanding the typical features of PBH and factitious insulin use. PBH onset normally occurs at least 6 months to a year after bariatric surgery such as a Roux-en-Y gastric bypass or vertical sleeve gastrectomy. Hypoglycemia due to bariatric surgery is primarily postprandial and worse after simple carbohydrate ingestion. While the pathophysiology is complex and multifactorial, it is understood that rapid nutrient entry into the gut elicits a strong glucagon-like peptide-1 response, resulting in hyperinsulinism and hypoglycemia.² Other pathophysiologic mechanisms include upregulated intestinal glucose transporters, inadequate glucagon response, altered microbiome, increased bile acids and fibroblast growth factor 19, decreased gluconeogenesis and lipogenesis, increased serotonin, altered cerebral blood flow, and dysregulated proinflammatory signaling.3, 4, 5, 6, 7, 8, 9, 10 Treatment includes a multidisciplinary approach incorporating expertise from endocrinologists, bariatric surgeons, and dieticians. Patients are advised to follow a high-protein, low-carbohydrate diet aimed at blunting the rise in glucose. Yet, if hypoglycemia persists after dietary changes, uncooked cornstarch and/or acarbose are added in attempt to delay glucose absorption, reduce the incretin release triggered by postprandial hyperglycemia, and attenuate subsequent insulin secretion. In severe cases, if patients fail medical management, surgical reversal is considered in eligible patients.

Factitious disorder is a behavioral health disorder in which patients feign illness or induce symptoms to gain attention, reduce loneliness, or receive sympathy.¹¹ The motivation or gain is internal as opposed to malingering where the motivation is external, such as receiving financial benefits.¹¹ Typical features of factitious disorder include a history of frequent visits to different hospitals, numerous providers, a patient history that is misleading or inconsistent, and atypical medical presentation often with no response to treatment.¹¹ In addition, self-induced injuries, worsening or new medical problems prior to hospital discharge, laboratory testing that does not align with patient history, and work in a health care field are also common in factitious disorder.¹¹

A prolonged fast is a helpful tool to decipher the etiology of hypoglycemia. Patients are asked to abstain from caloric intake for up to 72 h. If the serum glucose returns <55 mg/dL, insulin, c-peptide, proinsulin, beta-hydroxybutyrate, and a sulfonylurea and meglitinide screen are measured. An insulin level of 3.0 U/ml (18 pmol/L) or higher and a beta-hydroxybutyrate level of 2.7 mmol/L or less suggest an insulin mediated cause of hypoglycemia. Factitious insulin administration is distinguished from insulinoma through analysis of c-peptide, proinsulin levels, and an oral hypoglycemic agent screen. C-peptide is released by the pancreas in a 1:1 ratio with insulin. Proinsulin is a precursor of insulin that is also made in the pancreas. In the case of exogenous insulin use, endogenous insulin, c-peptide, and proinsulin production is halted. Therefore, a c-peptide level of <0.6 ng/ml (or <0.2 nmol/L) and a proinsulin level of <5.0 pmol/L suggest exogenous insulin administration.¹

The following case highlights typical clinical characteristics of PBH and factitious insulin use. It also demonstrates the importance of considering multiple etiologies when evaluating hypoglycemia and the utility of the insulin to c-peptide ratio in making a diagnosis.

Case Report

A 32-year-old woman with a history of Roux-en-Y gastric bypass, post-traumatic stress disorder (cleared by psychiatry prior to bariatric surgery), and numerous medical problems and hospitalizations presented with episodes of diaphoresis, fatigue, palpitations, blurred vision, and loss of consciousness due to hypoglycemia with initial onset 8 months after Roux-en-Y gastric bypass. She endorsed Whipple’s triad with glucose levels <55 mg/dL on multiple glucometer readings. Her hypoglycemic episodes occurred primarily after meals, but she mentioned intermittent overnight hypoglycemia. Laboratory evaluation and pancreatic imaging were normal (Table 1). She consulted a dietician who told her to increase carbohydrate ingestion, which exacerbated hypoglycemic episodes. Subsequently, she underwent hospitalization due to loss of consciousness from hypoglycemia occurring at a clinic visit. She began periodically wearing a continuous glucose monitor which depicted numerous instances of post-baritric hypoglycemia with blood sugar values over 200 mg/dL followed by hypoglycemia with blood sugars of 40-50 mg/dL clinically consistent with post-baritric hypoglycemia (PBH) (Fig. 1). She consulted a bariatric surgeon and an endocrinologist who jointly recommended dietary and medical management of her symptoms. She received extensive education pertaining to the PBH diet, yet her postprandial hypoglycemia with neuroglycopenic symptoms continued. She attempted treatment with cornstarch and acarbose to manage hypoglycemia; however, both were ineffective. A mixed meal tolerance test was performed with results typical of PBH (Table 2). Although she did not experience hypoglycemia, she had hyperglycemia with a peak serum glucose of 217 mg/dL and peak insulin level of 506 μU/mL (3036 pmol/L) with c-peptide level >10 nmol/L (>30 ng/mL) followed by a nadir glucose of 68 mg/dL (Table 2).

Table 1.

Results of Mixed Meal Tolerance Test and 72-Hour Fast

Evaluation	Result
CMP	Normal
CBC	Normal
TSH and free T4	Normal
Cosyntropin stimulation test (250 μg cosyntropin)	Normal (baseline cortisol of 15 μg/dL to 28 μg/dL at 60 min)
Insulin antibodies	<0.4 U/mL (reference range: <0.4 U/mL)
Abdominal MRI	No pancreatic lesion

Abbreviations: CBC = complete blood count; CMP = comprehensive metabolic panel; free T4 = free thyroxine; MRI = magnetic resonance imaging; TSH = thyroid-stimulating hormone.

The oral hypoglycemic agent screen was negative.

Fig. 1.

Case’s continuous glucose monitor (CGM) pattern is consistent with post-bariatric hypoglycemia.

Table 2.

Results of Mixed Meal Tolerance Test and 72-Hour Fast

	Serum glucose (mg/dL)	Insulin 2-23 μIU/mL (12-138 pmol/L)	C-peptide 0.8-5.2 nmol/L (2.4-15.7 ng/mL)
Mixed meal tolerance test – Ensure Plus given with serum glucose, insulin, and c-peptide levels measured at baseline and every 30 min for 3 h
Baseline	80	8 (48)	0.86 (2.6)
60 min	217 (peak)	506 (3036)	>10 (>30)
2 h	68 (nadir)	11 (66)	1.66 (5)

	Serum glucose (mg/dL)	Insulin 2-23 μIU/mL (12-138 pmol/L)	C-peptide 0.8-5.2 nmol/L (2.4-15.7 ng/mL)	Insulin (nmol/L) to c-peptide ratio (nmol/L)	Proinsulin (≤7.2 pmol/L)	Beta-hydroxy-butyrate (≤0.27 mmol/L)
72-h fast
Cutoffs suggestive of exogenous insulin per Cryer et al1	<55	>>3 (>> 18)	<0.2 (<0.6)	>1	<5	≤2.7
17 h into the fast	20	532 (3192)	0.364 (1.1)	8.8	2.6	0.1

The oral hypoglycemic agent screen was negative.

In the subsequent 3-month period, she was hospitalized for hypoglycemia multiple times. During those episodes, her continuous glucose monitor pattern was not consistent with PBH as persistent hypoglycemia was present without the classic peaks and falls in glucose seen in PBH (Fig. 2). Due to this and her endorsement of intermittent fasting hypoglycemia, which is atypical for PBH, a supervised inpatient fast was arranged; 17 h into the fast, she developed hypoglycemia with a serum blood glucose of 20 mg/dL, insulin of 532uLU/mL (3192 pmol/L), c-peptide of 1.1 ng/mL (0.364 nmol/L), and beta-hydroxybutyrate of 0.1 mmol/L (Table 2). The proinsulin result was unavailable during her hospitalization, given the laboratory's processing time of up to 2 weeks. Despite suspicion for exogenous insulin administration, the c-peptide level was not less than 0.6 ng/mL (0.2 nmol/L), though it was markedly low in relation to the insulin level. Her insulin to c-peptide ratio was significantly elevated at 8.8. No insulin vials, pens, or syringes were found upon searching her hospital room after the fast was complete. An inpatient meeting involving endocrinology, psychiatry, and internal medicine occurred to determine how to approach the patient. The patient was addressed in a nonpunitive and nonjudgemental manner, yet she denied insulin injection. After her hospital discharge, her proinsulin level returned low at 2.6 pmol/L supporting exogenous insulin injection. To date, she has reported significant improvement in hypoglycemia symptoms with intermittent hypoglycemia after meals consistent with PBH. She has not had readmissions for hypoglycemia and her primary care doctor has asked that she establish care with psychiatry.

Fig. 2.

Case’s continuous glucose monitor (CGM) pattern is not consistent with post-bariatric hypoglycemia (PBH) showing factitious basal insulin injection.

Discussion

Factitious insulin administration is often challenging to diagnose, and the concurrent presence of endogenous hyperinsulinemic hypoglycemia due to PBH presented a diagnostic challenge. As a result, this case offers valuable insights for clinical practice. Firstly, when a patient with PBH presents with fasting hypoglycemia, this is atypical, and other possible etiologies should be considered such as insulinoma, adrenal insufficiency, and factitious insulin use. In addition, education regarding PBH treatment is imperative as a high carbohydrate diet will make hypoglycemia worse in patients with PBH. The PBH diet consists of eating mixed meals high in protein and low in carbohydrates (<30 g per meal) with avoidance of added sugars.¹² In contrast to diabetes management, hypoglycemia resulting from PBH is treated with 10 g of simple carbohydrate followed by protein to avoid cyclical fluctuations of glucose levels.¹²

Since the proinsulin level often requires 2 weeks to process, we relied on serum glucose, insulin, c-peptide, and beta-hydroxybutyrate to guide our evaluation. Although the c-peptide is usually <0.6 ng/mL (0.2 nmol/L) in hypoglycemia due to insulin use, this case demonstrated that is not always the case. Patient presentation, proinsulin, and molar ratio of insulin to c-peptide ratio are other helpful tools in identifying the etiology. The insulin to c-peptide ratio is particularly helpful in determining if hypoglycemia is due to factitious insulin administration.¹³ C-peptide is mainly cleared by the kidney with a lower metabolic clearance rate than insulin which is cleared by the liver. Thus, insulin to c-peptide ratio is < 1 with endogenous insulin production and is > 1 with exogenous insulin use. Therefore, our patient’s insulin to c-peptide ratio of 8.8 was consistent with insulin injection.

Lastly, although our patient was not a healthcare worker, other potential indicators towards factitious insulin use included hypoglycemia uncharacteristic of PBH, many unexplained medical conditions, treatment at numerous health care facilities, and frequent hospitalizations. Recommendations for approaching a patient with factitious disorder include collecting evidence such as an insulin syringe, if possible, collaboration with psychiatry and the primary care provider, nonjudgemental and nonpunitive communication with the patient with proposal of ongoing support, and detailed documentation.¹³

In summary, this case presents a unique overlap of PBH and factitious insulin administration, reflecting hallmark features of both conditions. It reinforces the importance of comprehensive history-taking and the utility of additional laboratory tools such as the insulin-to-c-peptide ratio when c-peptide values are ambiguous and pro-insulin results are delayed.