Abstract
Metformin-associated lactic acidosis (MALA) and euglycaemic diabetic ketoacidosis (EKA) are both life-threatening endocrine emergencies. MALA is a well-documented complication of anti-glycaemic therapy in diabetics while EKA is an increasingly recognized disease entity with the advent of use of sodium-glucose co-transporter-2 (SGLT-2) inhibitors. However, the occurrence of concurrent metformin-associated lactic acidosis and euglycaemic ketoacidosis (MALKA) is uncommon and rarely reported in the literature. We report an unusual case of MALKA, in a 74-year-old gentleman with no previous history of chronic kidney disease and SGLT-2 inhibitor use, who presented with altered mental status, acute renal failure and profound high anion-gap metabolic acidosis (HAGMA) with lactaemia, ketonaemia and normoglycaemia (pH 6.965, bicarbonate 3.1, creatinine 522, glucose 6.6, lactate 17.5, ketones >8). The patient was initiated on intravenous insulin infusion with dextrose-containing drip and continuous renal replacement therapy (CRRT) in the intensive care unit (ICU). The patient had a prolonged hospital stay but was eventually discharged with a normalised renal function without need for long-term dialysis. The parallel occurrence of MALA and EKA suggest a metformin-associated inhibition of gluconeogenesis. This case highlights the importance of early recognition and investigation of concurrent diabetic ketoacidosis in the presence of MALA and vice versa as MALKA benefits from both prompt institution of parenteral glucose therapy and insulin infusion and consideration of initiation of haemodialysis.
Keywords
Introduction
Metformin is a biguanide and commonly used oral anti-glycaemic agent in diabetic management. It is generally considered to be safe, effective and inexpensive with a potential to reduce cardiovascular event risk and death. 1 However, metformin-associated lactic acidosis (MALA) is a rare but very severe complication with a great variation of incidence from 0 to 138 per 100 000 patient-years, and a mortality rate of about 25%–50%.2,3
Metformin exerts its anti-glycaemic effects by primarily inhibiting hepatic gluconeogenesis. It does this through various mechanisms, the most important being the inhibition of mitochondrial oxidative phosphorylation in the liver.4,5 This leads to an increase AMP-mediated protein kinase activity and downstream reduction of gluconeogenic enzymes gene transcription. 6 Metformin also enhances lactate levels by altering the hepatocellular redox state. This facilitates conversion of pyruvate to lactate and also inhibits utilisation of lactate towards gluconeogenesis by decreasing lactate and glycerol conversion to glucose.4,5 Metformin is not metabolised by the liver, and its elimination is prolonged in patients with renal impairment, hence safety labels carry warnings for higher MALA risk in those with renal or hepatic impairment, congestive heart disease and age.1,4
Euglycemic diabetic ketoacidosis, on the other hand, has been more commonly recognized as a subset of life-threatening endocrine emergencies which entails relative euglycemia together with metabolic acidosis (serum bicarbonate <18 mmol/L and pH <7.3) and ketosis. It is more so associated in the literature with type-2 diabetics who are insulin-deficient on sodium cotransporter-2 (SGLT-2) inhibitors as well as those with type-1 diabetes or latent autoimmune diabetics. 7 Other aetiologies include prolonged starvation, heavy alcohol consumption, chronic liver disease, pregnancy and glycogen storage disorders.
While there is established literature on MALA and SGLT-associated euglycaemic ketoacidosis (EKA) separately, the concurrent occurence of metformin-associated lactic acidosis and euglycaemic ketoacidosis (MALKA) has rarely been reported. 8 We here in report an unusual case of MALKA, in a patient with type-2 diabetes on Metformin, and no previous history of chronic kidney disease or SGLT-2 inhibitor use. Written consent was obtained from the patient’s next-of-kin (daughter).
Case presentation
A 74-year-old male patient presented to the Emergency Department (ED) via ambulance with sudden onset of altered mental status and witnessed seizure activity by his son. His comorbidities included type-2 diabetes mellitus (HbA1c 5.8%), hypertension, hyperlipidaemia, and recently diagnosed cognitive impairment from mixed dementia and right cortical sub-acute infarct for which he is on dual anti-platelet therapy (aspirin, clopidogrel). He had no history of chronic kidney disease and is on Metformin (850 mg three times daily) for his diabetes. His son reported that he was well prior and was noted to have labored breathing just before sudden onset of up-rolling of eyes. Rectal diazepam 10 mg was administered by the paramedics at the scene and on arrival to the ED his Glasgow Coma Scale (GCS) was E3V4M4. His vitals on arrival to the ED was: temperature 37°, heart rate 87 beats per minute, blood pressure 112/55 and oxygen saturation 100% on room air. In view of the acute onset of his altered mental status, the stroke team was activated. Computer tomography of the brain and multiphasic angiogram of the cerebral vessels were performed and they showed no established acute infarct, intracranial hemorrhage or large vessel occlusion. Subsequent laboratory investigations showed profound high anion gap metabolic acidosis (HAGMA) with partial respiratory compensation (arterial blood gas: pH 6.965, pO2 167.5, pCO2 9.0, base excess −27.3), hyperlactaemia (lactate 17.5), euglycaemia (glucose 6.6), acute kidney injury (urea 21.9, creatinine 522, bicarbonate 3.1, sodium 143, potassium 5.0, chloride 96; previous creatinine 79), and raised inflammatory markers (leukocytes 19.79, C-reactive protein 13.4, procalcitonin 0.76). His electrocardiogram was normal sinus rhythm, chest X-ray showed no active lung lesion and urinalysis was unremarkable. He was initially treated for presumptive severe sepsis from possible meningoencephalitis and started on empirical broadspectrum antibiotics. The patient subsequently turned hypotensive with systolic blood pressure dropping to 70 with further drop in his GCS to E2V3M4 while still at the ED. He was started on peripheral noradrenaline infusion, titrated to mean arterial pressure, and airway protection was accomplished through rapid sequence intubation in view of high risk of cardiopulmonary collapse. Additional bloods demonstrated ketonaemia (ketones > 8.0) and he was started on intravenous insulin with dextrose-containing drip. He was transferred to the intensive care unit (ICU) and started on continuous renal replacement therapy (CRRT) via a right internal jugular venous catheter. His lactate and ketones improved and normalized within 3 days of CRRT and intravenous insulin and dextrose-containing drip therapy respectively.
The patient remained in ICU for 6 days where he developed paroxysmal atrial fibrillation (pAF) on day two of his stay, precipitated by metabolic acidosis and electrolyte imbalance. He was subsequently transferred out to the general ward (GW) where he remained for 35 days. Whilst in GW, he had multifactorial delirium and deconditioning from his prolonged hospitalization and was transferred to a community hospital where he stayed for another 39 days for inpatient rehabilitation. He was discharged well and stable, with a normalized renal function and did not require long-term renal replacement therapy. He was also commenced on apixaban for his pAF.
Discussion
We describe a case of severe HAGMA and systemic shock from MALKA in a patient with type-2 diabetes on metformin only, with no previous chronic kidney disease or SGLT-2 inhibitor use, who presented with seizures and altered mental status. Conventionally, it is difficult to differentiate MALA from other more common causes of metabolic acidosis such as sepsis. This is especially so given that metformin serum levels are rarely obtained in the ED and most often, not practical. However, MALA is an important differential to be considered, in patients who presents with hyperlactaemia, especially if a diabetic patient on metformin presents with severe metabolic acidosis (pH 7.1, bicarbonate <10 mmol/L), raised lactate (>15 mmol/L), profound anion gap acidosis (>20 mmol/L) and kidney failure. 9 Initial management MALA usually involves fluid resuscitation, including alkalinization with sodium bicarbonate and treatment of the underlying cause. 10 In critically ill patients such as those with severe metabolic acidosis, hypovolemia and multi-organ failure, intensive care support with mechanical ventilation and cardiovascular support with vasopressors may be required.10,11 Furthermore, early consideration of haemodialysis in such patients, is useful, for drug clearance, buffer of acidosis and control of volemia.10,11 In our patient, acute kidney injury would have likely resulted in decrease renal clearance of metformin and the low oxygen delivery state of shock might have exacerbated the adverse effects of metformin, leading to hyperlactatemia and metabolic acidosis. Lactate production, from anaerobic glycolysis, is enhanced specifically by metformin-related inhibition of lactate to pyruvate and increase pyruvate to lactate conversion from inhibition of pyruvate entering the mitochondria for oxidative phosphorylation.4,12
Interestingly, in our patient, he had an anion gap of 43.6. This augmented anion gap was subsequently attributed to a secondary acid, serum ketone, demonstrating a concurrent EKA. Traditionally, diabetic ketoacidosis occurs due to either a relative insulin deficiency with insulin resistance or absolute insulin deficiency in patients with diabetes subjected to various physiological stress. The inability to use carbohydrates as substrates for energy production shifts reliance to fat oxidation.7,13 The upregulation of counter-regulatory hormones such as glucagon, cortisol, catecholamines, ultimately leads to free fatty acid oxidation and production of ketones (acetoacetate and beta-hydroxybutyrate), and the hyperglycaemia of DKA. However, in EKA, glucose levels tend to be low to normal. This is due to a few underlying mechanisms which results in normoglycaemia – depleted hepatic glucose production and glycogen stores in fasting states e.g. prolonged starvation/physical activities, chronic alcohol use, pregnancy, decreased caloric intake after bariatric surgery and enhanced urinary glucose excretion e.g. SGLT2-I use. 4 While the occurrence of EKA with SGLT-2 inhibitor use is relatively well-published, reports of metformin-related EKA is less widely reported. In our patient, further history from the patient’s family established that his oral intake had gradually declined with the progression of dementia. This explains a possible metformin-related suppressed hepatic gluconeogenesis, on a background of depleted glycogen stores, which could have led to normoglycaemia and ketonaemia and hence EKA in our patient.8,13
Nevertheless, it is important to note the considerable overlap between starvation ketoacidosis (SKA) and EKA. Diagnosis is often clinical. SKA patients are less likely to have a history of diabetes or major intercurrent illness and devleops a less profound metabolic acidosis. SKA management generally responds well to providing the substrate glucose, which stimulates endogenous production of insulin, restoring the insulin to glucagon ratio and reverses ketogenesis. 14 DKA/EKA, on the other hand, benefits from expeditious correction of dehydration and insulin infusion with dextrose containing drip, to reverse insulin insufficiency and suppress ketogenesis. 13 In our patient, the concurrent occurrence of lactic acidosis and ketoacidosis was reflected in an augmented anion gap, but his ketone levels far exceeded that of SKA, suggesting a possible metformin-related and starvation-independent ketogenesis.8,14,15
Conclusion
In conclusion, we would like to highlight the importance of early recognition of MALKA which is a rather rare clinical entity. The presence of MALA should trigger an investigation into concurrent diabetic ketoacidosis and vice versa. MALKA benefits from both prompt instituition of parenteral glucose therapy and insulin infusion and consideration of initiation of haemodialysis, ideally in an intensive care setting with other all-rounded aggressive supportive therapies, which will lead to better patient outcomes.
Footnotes
Author contributions
Writing – original draft: all authors. All authors reviewed and edited the manuscript and approved the final version of the manuscript.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
