Long QT Syndrome and Risk for Hypoglycemia in a Postbariatric Surgery Patient

| January 1, 2019

by Roselyn Cristelle Mateo, MD, MS, and Mary Elizabeth Patti, MD

Dr. Mateo is with the Division of Endocrinology, Beth Israel Deaconess Medical Center, Joslin Diabetes Center, and Harvard Medical School, Boston, Massachusetts. Dr. Patti is with the Division of Endocrinology, Beth Israel Deaconess Medical Center, Joslin Diabetes Center, and Harvard Medical School, Boston, Massachusetts

Funding: Dr. Patti gratefully acknowledges research grant support from National Institutes of Health (NIH) DK114156, DK101043, DK107114, DK106193, HD091974, and DK036836 (Joslin DRC), Helmsley Charitable Trust (New York, New York), Dexcom (San Diego, California), and Xeris Pharmaceuticals (Chicago, Illinois).

Disclosures: Dr. Mateo reports no conflicts of interest relevant to the content of this article. Dr. Patti has received investigator-initiated grant support from Janssen Pharmaceuticals (Titusville, New Jersey), Medimmune (Gaithersburg, Maryland), Sanofi (Bridgewater, New Jersey), Astra-Zeneca (Wilmington, Delaware), Jenesis Marketing Group (Reynoldsburg, Ohio), Xeris Pharmaceuticals, Dexcom, and Nuclea Biotechnologies (Pittsfield, Massachusetts); has been a site investigator for XOMA Corporation (Berkeley, California), and acknowledges clinical trial research support from Ethicon (Cincinnati, Ohio), Medtronic (Minneapolis, Minnesota), Novo Nordisk (Plainsboro, New Jersey), Nestlé Health Science (Bridgewater, New Jersey), and Dexcom within the past five years. Dr. Patti has consulted for Eiger Biopharmaceuticals (Palo Alto, California) and discloses a patent application for plasma proteins contributing to hypoglycemia.

Bariatric Times. 2019;16(1):10–12.

AbstractThe authors report the case of a 41-year-old female patient with a body mass index of 44.4kg/m2 who underwent sleeve gastrectomy with postoperative course complicated by cardiac arrest due to polymorphic ventricular tachycardia and prolonged QT interval (as long as 576msec). The patient subsequently developed hypoglycemia several weeks after surgery, with sensor glucose readings as low as 33mg/dL, largely in the postprandial state. She underwent a diagnostic 72-hour fast with results indicating appropriate suppression of insulin secretion with fasting, consistent with lack of autonomous insulin secretion in the fasting state. Given the coexistence of hypoglycemia and prolonged QT interval, we considered an underlying genetically determined long QT syndrome as a potential contributor to her hypoglycemia. Genetic testing revealed mutation in the KCNQ1 gene, which encodes subunits of the voltage-gated potassium channels (Kv7.1) and (Kv11.1) that are expressed in both cardiomyocytes and pancreatic beta cells. Patients with the KCNQ1 mutation have longer QT and lower serum glucose and potassium levels. Therapeutic goals to reduce the risk of recurrent arrhythmia included avoidance of medications that prolong the QT interval, as well as prevention of both hypokalemia and hypoglycemia. Medical nutrition therapy was aimed at reducing postprandial glycemic spikes, which can stimulate insulin secretion and increase risk for subsequent hypoglycemia after meals.

Keywords: Obesity, bariatric surgery,  sleeve gastrectomy, complications, cardiac arrest, long QT syndrome, hypoglycemia

Endocrinologists are increasingly encountering hypoglycemia in patients after bariatric procedures and other forms of upper gastrointestinal surgery, yet the condition remains poorly recognized by patients and physicians alike. There are differences in the diagnostic criteria, and its true incidence remains uncertain.1

Postbariatric hypoglycemia (PBH) typically emerges 1 to 3 years postoperatively, but it can be seen as soon as 6 to 12 months after surgery. In this setting, hypoglycemia is most commonly experienced 1 to 3 hours after meals, with no hypoglycemia after a prolonged fast of at least 12 hours. Typical symptoms of hypoglycemia, ranging from mild adrenergic or cholinergic symptoms (e.g., sweating, palpitations, tremors, and agitation, to debilitating neuroglycopenic symptoms (e.g., blurred vision, impairment of speech, memory, altered cognition, and even seizures). Importantly, some of these symptoms are nonspecific, making accurate diagnosis challenging because they might be initially attributed to dumping syndrome, anxiety, or cardiovascular disorders. To define hypoglycemia in this setting, venous blood glucose level at the time of symptoms should be less than 54mg/dL.1 Repeated episodes of hypoglycemia can result in blunted hypoglycemia awareness, thus putting the patient at risk for further severe hypoglycemia-related complications. While novel treatment approaches for PBH are being pursued, prevention by identification of patients at high risk for this condition prior to surgery is more desirable.

We present a case in which a female patient experienced cardiac arrest during postoperative care following sleeve gastrectomy.

Case Report

A 41-year-old female patient with a body mass index (BMI) of 44.4kg/m2 underwent sleeve gastrectomy. Her postoperative course was complicated by cardiac arrest due to polymorphic ventricular tachycardia (torsades de pointes) with corrected QT interval as long as 576msec. This was initially attributed to postoperative pharmaceutical treatments, including ondansetron (Zofran, GlaxoSmithKline, Research Triangle Park, North Carolina) and metoclopromide (Reglan, ANI Pharmaceuticals, Inc., Baudette, Minnesota).

Prior to surgery, at 20 years of age, symptoms of headache, dizziness, imbalance, sweating, and confusion would occur if the patient fasted for a prolonged period. At age 31, she noted the onset of episodic syncope and near-syncope and adapted the habit of eating frequently, always carrying food to prevent these symptoms. Blood glucose levels would drop as low as 40mg/dL. Her BMI was normal until her 20s, when she slowly started to gain weight. Despite attempts at lifestyle modifications, including low-calorie diet programs, over-the-counter dietary supplements, increased activity, and health lifestyle coaching programs, she was unable to achieve sustained weight loss and ultimately considered bariatric surgery. Given the possibility of hypoglycemia as a contributor to these symptoms, a masked continuous glucose monitoring (CGM) test was performed during the preoperative period; this showed minimum sensor glucose of 50mg/dL, especially between 8 p.m. to 4 a.m., with a concurrent diary indicating nocturnal sweating.

Past medical history included Raynaud’s disease without gangrene, reflux esophagitis, hypertension, polycystic ovary syndrome (PCOS) and migraine headaches. The patient’s family history was notable among several family members: coronary artery disease, dyslipidemia, diabetes; a brother with history of obesity who underwent uncomplicated sleeve gastrectomy without postoperative hypoglycemia; and a mother with gastric cancer. The mother reported that the patient had hypoglycemia at birth, which was attributed to poor feeding and which resolved once formula feeding started. There was no family history of arrhythmia, pituitary or parathyroid tumors, recurrent kidney stones, or other multiple endocrine neoplasia (MEN) syndrome components. The patient was employed full-time at time of surgery in a largely sedentary managerial position and lived with her husband and young child. She reported rare alcohol use.

Physical examination on postoperative Day 1 was remarkable for a BMI of 44kg/m2, pulse 89, blood pressure (BP) 110/62, normal cardiovascular exam, and intact surgical scar. Laboratory testing revealed mild reductions in potassium, ranging from 3.2 to 4mEq/L, which required supplementation, and plasma glucose levels from 77 to 102mg/dL in the immediate postoperative period.

Several weeks postoperatively, the patient continued to have adrenergic and cholinergic symptoms, including headache, excessive sweating, shakiness, and nightmares, despite following a diet of low-glycemic-index foods. A CGM test showed sensor glucose levels as low as 33mg/dL, with associated adrenergic and cholinergic symptoms. Clinical and laboratory analysis did not suggest noninsulin-mediated causes of hypoglycemia, such as low glycogen stores or impaired gluconeogenesis in association with malnutrition, kidney or liver disease, or sepsis. Thyroid disease and adrenal insufficiency were ruled out on the basis of history and laboratory studies. Due to the severity of recurrent hypoglycemia in the early postoperative state, a pattern highly atypical for post-bariatric hypoglycemia, a diagnostic 72-hour fast to assess insulin secretion and its suppressibility with fasting was performed. Blood glucose values decreased progressively with fasting, with nadir of 59mg/dL at 46 hours of fasting with associated headache, sweating, and shakiness. Concurrent laboratory analysis included insulin level 1.9uIU/mL (2.0–19.6uIU/mL), C-peptide 0.54ng/mL (0.80–3.85ng/mL), proinsulin less than 7.5pmol/L (<18.8pmol/L), beta hydroxybutyrate 2.5mmol/L (<4mmol/L), and glucagon 73pg/mL (8–57pg/mL). Glucose rose minimally to 67mg/dL after administration of 1mg glucagon, consistent with depletion of glycogen stores. Together, these results are consistent with lack of autonomous insulin secretion in the fasting state, confirming the clinical pattern of largely prandial insulin secretion.

Based on the presence of hypoglycemia in association with long QT syndrome, we hypothesized that hypoglycemia was related to an underlying genetic mutation in KCNQ1 or KCNH2, and potentially aggravated by increases in incretin and insulin secretion following sleeve gastrectomy. Genetic testing confirmed the loss of function mutation in KCNQ1. Therapeutic goals to reduce the risk of recurrent arrhythmia include avoidance of medications which prolong the QT interval, as well as prevention of both hypokalemia and hypoglycemia. Medical nutrition therapy was aimed at reducing postprandial glycemic spikes, which can stimulate insulin secretion and increase risk for subsequent hypoglycemia after meals; this included avoidance of simple carbohydrates, consumption of complex carbohydrates in controlled portions, bedtime cornstarch (a highly complex carbohydrate), and ensured adequate protein and micronutrient intake.8 CGM with alarms was used to allow the patient to detect hypoglycemia and treat before it became severe. Diazoxide, a potassium channel activator that reduces insulin secretion, was initiated at 25mg at bedtime. With this multipronged approach to therapy, symptoms improved. While beta blockers are effective in preventing cardiac events in approximately 70 percent of patients with long QT syndrome, providers should also be mindful that these drugs might blunt hypoglycemia awareness as a consequence of antiadrenergic effects. Moreover, beta blockade can suppress glycogenolysis, thus reducing correction of hypoglycemia. Long-term studies, however, have not been performed to date.9


One clue to the potential cause of hypoglycemia in our patient was the postoperative identification of long QT interval, which, upon further review, was present even in the preoperative period. While long QT syndrome can result from specific medications or metabolic abnormalities, such as hypocalcemia, hypomagnesemia, and hypokalemia, the identification of long QT interval should also prompt consideration of an underlying genetically determined long QT syndrome. Three major long QT syndrome genes (KCNQ1, KCNH2, and SCN5A) have been identified, together accounting for approximately 75 percent of individuals with the disorder.2

Interestingly, long QT syndrome due to mutations in KCNQ1 and KCNH2 genes encoding subunits of the voltage-gated potassium channels (Kv7.1) and (Kv11.1) has recently been linked to hypoglycemia. KCNQ1 and KCNH2 are expressed in both cardiomyocytes and pancreatic beta cells. In a study by Torekov et al,3 patients with the KCNQ1 mutation who underwent a 75g oral glucose tolerance test after an overnight fast had lower plasma glucose levels three hours after ingestion compared with age-matched controls (4.4+0.3 vs. 5.4+0.5mmol/L, P=0.03). Insulin responses during the oral glucose tolerance test were higher than in matched controls (area under the curve [AUC] 45.6+6.3 vs. 26.0+2.8 min*nmol/L, P<10−5.). Plasma levels of insulin and C-peptide and insulin secretion rate (ISR) were also higher than in controls (≈1.9 fold for insulin, 1.2 fold for C-peptide, 1.3 for ISR). Beta-cell sensitivity to glucose, which was evaluated as the slope of the relation between ISR and glucose, was also significantly greater in mutation carriers. Furthermore, continuous glucose measurements for 3 to 7 days demonstrated that patients spent 77±18 minutes per 24 hours with glucose levels less than 3.9 mmol/L, and 36±10 minutes with glucose less than 2.8 mmol/L), compared to zero minutes (<3.9 mmol/L glucose) for the control participants (P<0.05). In patients with the mutation, low glucose levels occurred 3 to 5 hours after meals. Both frequency and severity of hypoglycemia, as measured by a standardized hypoglycemia questionnaire, were also higher than control participants. Plasma glucagon levels were not significantly different in patients with the mutation, despite hypoglycemia, perhaps due to impaired counterregulatory response with recurrent hypoglycemia, suppression of glucagon secretion by increased insulin levels, or increases in somatostatin release associated with the KCNQ1 mutation. These responses did not differ in the patients taking beta-blockers. Interestingly, patients with the KCNQ1 mutation also had lower serum potassium levels, likely due to insulin activation of the Na-K ATPase, moving potassium intracellularly.3–5 Both hypoglycemia and hypokalemia can further prolong the QT interval, initiating a vicious cycle of risk for arrhythmia and death.

Similarly, loss-of-function mutations in the KCNH2 gene encoding Kv11.1, the alpha subunit of a potassium ion channel, also cause both long QT syndrome and hypoglycemia. Aside from cardiac muscles, these voltage-gated channels are also present in intestinal K and L cells that secrete incretins, such as glucagon-like peptide-1 (GLP1), in response to oral glucose stimulation and pancreatic alpha and beta cells.6 Hylten-Cavallius et al7 studied individuals with KCNH2 mutations with a prolonged oral glucose tolerance test. Sixty-three percent developed plasma glucose levels less than 70mg/dL and 18 percent had plasma glucose levels less than 50mg/dL. In parallel, there was a 56- to 78-percent increase in insulin, C-peptide, GLP-1, and GIP response.


In summary, our patient illustrates a unique presentation of hypoglycemia linked to underlying long QT syndrome due to mutation in KCNQ1. This was not recognized in the preoperative state, but it might have contributed to mild hypoglycemia. Postoperative medications and electrolyte shifts likely resulted in prolongation of the QT interval and cardiac arrest. Bariatric surgery-induced increases in gastric emptying and postprandial insulin secretion likely further contributed to KCNQ1-related postprandial hypoglycemia. A clue to the unique nature of this case is that hypoglycemia occurred rapidly (within weeks) after surgery, compared to the typical presentation of postbariatric hypoglycemia one or more years after surgery. Patients who present with hypoglycemia early after surgery require detailed workup for hypoglycemia.1 Moreover, we suggest that patients with known risk factors should be screened for long QT interval in the preoperative screening phase, so appropriate pre-, peri- and postoperative planning can be implemented.


  1. Salehi M, Vella A, McLaughlin T, Patti ME. Hypoglycemia after gastric bypass surgery: current concepts and controversies. J Clin Endocrinol Metab. 2018;103(8):2815–2826.
  2. Tester DJ, Ackerman MJ. Genetics of long QT syndrome. Methodist Debakey Cardiovasc J. 2014;10(1):29–33.
  3. Torekov SS, Iepsen E, Christiansen M, et al. KCNQ1 long QT syndrome patients have hyperinsulinemia and symptomatic hypoglycemia. Diabetes. 2014;63(4):1315–1325.
  4. Boini KM, Graf D, Hennige AM, et al. Enhanced insulin sensitivity of gene-targeted mice lacking functional KCNQ1. Am J Physiol Regul Integr Comp Physiol. 2009;296(6):R1695–1701.
  5. McTaggart JS, Clark RH, Ashcroft FM. The role of the KATP channel in glucose homeostasis in health and disease: more than meets the islet. J Physiol. 2010;588(Pt 17):3201–3209.
  6. Engelbrechtsen L, Mahendran Y, Jonsson A, et al. Common variants in the hERG (KCNH2) voltage-gated potassium channel are associated with altered fasting and glucose-stimulated plasma incretin and glucagon responses. BMC Genet. 2018;19(1):15.
  7. Hylten-Cavallius L, Iepsen EW, Wewer Albrechtsen NJ, et al. Patients with long-QT syndrome caused by impaired hERG-encoded Kv11.1 potassium channel have exaggerated endocrine pancreatic and incretin function associated with reactive hypoglycemia. Circulation. 2017;135(18):1705–1719.
  8. Suhl E, Anderson-Haynes SE, Mulla C, Patti ME. Medical nutrition therapy for post-bariatric hypoglycemia: practical insights. Surg Obes Relat Dis. 2017;13(5):888–896.
  9. Casiglia E, Tikhonoff V. Long-standing problem of beta-blocker-elicited hypoglycemia in diabetes mellitus. Hypertension. 2017;70(1):42–43.

Tags: , , , , , ,

Category: Case Report, Current Issue

Comments are closed.