• 
• 

Three Part Question

In [Adult patients admitted in the intensive care units with metformin associated lactic acidosis], is [hemodialysis better than supportive care], at [reducing mortality]?

Clinical Scenario

A 70-year-old male patient, weighing 80 Kg, with diabetes mellitus, coronary heart disease, congestive heart failure (NYHA III), essential Hypertension (stage 2) and renal dysfunction (serum urea: 80 mg/dL, serum creatinine: 1,6 mg/dL, creatinine clearance of 48,6 mL/min/1.73m²); His therapeutic regimen included isosorbide dinitrate 20 mg bid, furosemide 40 mg qd, enalapril 20 mg qd and metformin 1000 mg tid.
He was admitted in the emergency department of our Hospital, complaining of fever, malaise, respiratory distress, myalgias, disorientation and abdominal discomfort with positive right Murphy’s sign. He was hemodynamically unstable with MAP (mean arterial pressure) of 50 mmHg and tachycardia (120 bpm). Laboratory evaluation revealed leukocytosis 28000/mm³, severe renal failure (serum urea: 210 mg/dL, serum creatinine: 6 mg/dL, creatinine clearance of 4,9 mL/min/1.73m²), high anion gap metabolic acidosis in arterial blood gas analysis (pH 6,9; AG 30 mEq/L), a plasma lactate of 10 mEq/L, no ketonuria or evidence of ingestion of a toxic substance (such as ethylene glycol, methanol). A renal ultrasound confirmed right acute pyelonephritis. According to the patient’s medical and drug history, clinical and laboratory analysis (although confirmatory laboratory metformin levels were not obtainable), we suspected of a case of Metformin associated lactic acidosis (MALA), complicating a septic shock in the context of acute pyelonephritis. The patient was transferred to our intensive care unit and managed aggressively in accordance with the Sepsis Surviving Campaign Guidelines (2008), with mechanical ventilation, fluids and vasopressor agents; despite intravenous sodium bicarbonate therapy, the clinical scenario was deteriorating and therefore immediately continuous venovenous hemodiafiltration (HDFVVC) was started and went on during 23 hours. Ultimately, he was stabilized and progressive restoration of acid-base balance and renal function was observed.
Did hemodialysis, compared to supportive care, had a more positive effect on this patient’s outcome?

Search Strategy

MEDLINE: 1966 – July 2011 (Week 3) and citation review of relevant primary and review articles. Limit to English articles.
("adult"[MeSH Terms] OR "adult"[All Fields] OR "adults"[All Fields]) AND ("intensive care"[MeSH Terms] OR ("intensive"[All Fields] AND "care"[All Fields]) OR "intensive care"[All Fields]) AND ("metformin"[MeSH Terms] OR "metformin"[All Fields]) AND associated[All Fields] AND ("acidosis, lactic"[MeSH Terms] OR ("acidosis"[All Fields] AND "lactic"[All Fields]) OR "lactic acidosis"[All Fields] OR ("lactic"[All Fields] AND "acidosis"[All Fields])) AND ("haemodialysis"[All Fields] OR "renal dialysis"[MeSH Terms] OR ("renal"[All Fields] AND "dialysis"[All Fields]) OR "renal dialysis"[All Fields] OR "hemodialysis"[All Fields]) AND supportive[All Fields] AND care[All Fields] AND ("mortality"[Subheading] OR "mortality"[All Fields] OR "mortality"[MeSH Terms]).
Cochrane
Cochrane – all of the Cochrane library – “Intensive care” and “Metformin associated lactic acidosis” and “dyalisis” and “mortality”.

MEDLINE: 1966 – July 2011 (Week 3) and citation review of relevant primary and review articles. Limit to English articles.
("adult"[MeSH Terms] OR "adult"[All Fields] OR "adults"[All Fields]) AND ("intensive care"[MeSH Terms] OR ("intensive"[All Fields] AND "care"[All Fields]) OR "intensive care"[All Fields]) AND ("metformin"[MeSH Terms] OR "metformin"[All Fields]) AND associated[All Fields] AND ("acidosis, lactic"[MeSH Terms] OR ("acidosis"[All Fields] AND "lactic"[All Fields]) OR "lactic acidosis"[All Fields] OR ("lactic"[All Fields] AND "acidosis"[All Fields])) AND ("haemodialysis"[All Fields] OR "renal dialysis"[MeSH Terms] OR ("renal"[All Fields] AND "dialysis"[All Fields]) OR "renal dialysis"[All Fields] OR "hemodialysis"[All Fields]) AND supportive[All Fields] AND care[All Fields] AND ("mortality"[Subheading] OR "mortality"[All Fields] OR "mortality"[MeSH Terms]).
Cochrane
Cochrane – all of the Cochrane library – “Intensive care” and “Metformin associated lactic acidosis” and “dyalisis” and “mortality”.

Search Outcome

Medline - 1 paper found, that was critically appraised.
Cochrane – No relevant findings;

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Peters 2008 France	Retrospective analysis of 30 patients admitted to the intensive care unit (ICU) of Hospital Central of Nancy (France), from August 2002 to August 2007 (5 years), presenting with MALA (defined as lactic acidosis with lactate of greater than 5 mmol/L and bicarbonate of less than 22 mmol/L). Those were patients who were chronically taking metformin or in the setting of a metformin overdose (suicide attempt). No patient had MALA as their admission diagnosis. Clinical and laboratory features at admission and during ICU stay were studied: SAPS II, Charlson Index, age, sex, reason for ICU admission, blood pressure, respiratory rate, vasopressor requirement, acute renal failure defined according to Rifle criteria, and biological data (arterial pH, blood lactate, bicarbonate, glucose and creatinine, prothrombine time). The patients were divided according to their 28-day outcome in order to investigate if there were differences in relation to all the studied parameters. The population was also split regarding the use of haemodialysis. Due to the retrospective design, no rules precluded the use of haemodialysis. Patients were not enrolled if a limitation of care was decided on admission.	2b – Retrospective cohort study;	Assess the prevalence of MALA in a 16-bed, university-affiliated, intensive care unit (ICU), and the effect of dialysis on patient outcome;	MALA is of low (1%) prevalence in medical ICUs; MALA accounted for 0.84% of all admissions during the study period; MALA is associated with a 30% mortality rate; Prothrombin time on admission seems to be inversely related to survival; Although patientswho went on to haemodialysis had higher illness severity scores,as compared with those who were not dialysed, the mortalityrates were similar between the two groups (31.3% versus28.6%);	Small size of the population and therefore, the protective effect of dialysis remains hypothetical; The bias could not be corrected, since the study was retrospective: only the more severely ill patients were dialyzed; The exact role of metformin in explaining the degree of lactic acidosis could not be definitely ascertained as certain conditions (cardiac arrest or shock), may per se be associated with hyperlactataemia;

Comment(s)

Theoretical Considerations Metformin, an oral biguanide commonly used in diabetes mellitus type 2, has been used in Europe since the 1970s and is the sixth most frequently prescribed drug in the USA, since its introduction in May 1995, with over 50 million prescriptions in 2009(2) . Its usage increased following publication of the results of the UKPDS study (The United Kingdom Prospective Diabetes Study, 1998), which demonstrated impressive reductions in diabetes-related endpoints and mortality in overweight patients with type 2 diabetes who used this drug (3). Metformin is renally cleared and is known to accumulate in patients with chronic kidney disease. Metformin blood levels are known to correlate with serum creatinine concentration. It is highly water soluble, not bound to proteins and its apparent volume of distribution is usually reported to be higher than 3 L/kg, attesting to the predominance of the intracellular location. Considering these data, metformin can theoretically be extracted from blood by haemodialysis if dialysis is conducted for long enough to mobilize the intracellular form (4). Current guidelines stipulate that metformin should be used with caution in estimated glomerular filtration rates (eGFRs) of less than 60 mL/min/1.73 m² and not at all in eGFRs of less than 30 mL/min/1.73 m² (2). Metformin-associated lactic acidosis (MALA) is, relative to the very widespread use of the drug, a rare but potentially life-threatening complication with a high mortality rate, averaging 50%. It has an estimated incidence of 1 to 5 cases per 100,000 patient-years, but may be as high as 30 cases per 100,000 patient-years (2). It usually occurs in a previously metformin-treated type 2 diabetic patient, presenting with very severe lactic acidosis (LA), when contraindications of the drug have been overlooked or more often when a precipitating disease (chronic cardiac or respiratory failure, impaired liver function, sepsis or severe dehydration) induces an acute renal failure (AKI) and an incidental overdose. In most cases, patients who develop severe lactic acidosis under metformin treatment, suffer from acute renal failure or severe sepsis (4). Sometimes, medications that interfere with renal hemodynamic autoregulation (that is, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and nonsteroidal anti-inflammatory drugs) and volume depletion are frequently implicated in generating the AKI leading to MALA (2). Voluntary intoxication, usually as a suicide attempt, is rare (4). The physiopathology of MALA is complex and mostly unclear: presumably it suppresses biological oxidation and the enzymes of the citric acid cycle. It is also known that metformin impairs lactate clearance of the liver through the inhibition of complex I of the mitochondrial respiratory chain. Although increased lactic acid production may be induced by haemodynamic instability and/or tissue hypoxia associated with severe metformin overdose or any underlying unstable cardiovascular or respiratory condition, lactic acidosis is predominantly due to a lack of lactate's clearance than to an increased production. Salpeter et al, in a recent Cochrane systematic review (347 trials with 70,490 patient-years of metformin use) found no evidence that metformin is associated with an increased risk of lactic acidosis, or with increased levels of lactate, compared to other anti-hyperglycemic treatments: metformin alone, when it is used as labeled, does not cause LA (5). A recent study stated that diabetes is a risk factor for LA rather than metformin: being a vascular disease, may contribute through a predisposition to relative tissue hypoxia leading to anaerobic glycolysis more readily than in non diabetics (6). Lactic acidosis (LA), as it is known for decades, carries a poor prognosis. Being the most frequent cause of metabolic acidosis, it is characterized by an increase in the anion gap, arterial lactate concentration exceeding 5 mmol/L and pH ≤ 7,35. Lactate production is indeed an adaptive response to impending energy failure. This response provides some energy and chance for cells to survive; even when oxygen availability or utilization is defective it acts as an oxidative substrate exchanged between cells and tissues. The prognosis of LA primarily depends on the underlying mechanism and on its reversibility. When LA is due to metformin accumulation, then renal replacement therapy can efficiently remove the toxic substance and prognosis can be surprisingly good. The situation can be much more complex and less easily reversible when lactic acidosis is primarily due to severe hypoxia or tissue hypoperfusion (7). Therefore, the distinction between MALA and lactic acidosis (LA) is sometimes very subtle and metformin accumulation may coexist with other risk factors, all contributing to the pathogenesis of LA. The literature on the management of MALA is controversial and sparse. Current therapeutic approach of MALA includes correction of acidosis, acceleration of lactate metabolism, elimination of metformin and management of any concurrent disease: vital function support, activated charcoal 100g orally in the first four hours after admission (especially in cases of metformin overdose), bicarbonate infusion (if pH at least 7.15), although it’s use is known to worsen intracellular acidosis and hyperlactatemia and sodium overload (4). Renal replacement therapies (bicarbonate-buffered and high-sodium dialyzate), including conventional hemodialysis and continuous venovenous hemofiltration, initiated early, may possess a protective effect and have been successfully employed in MALA, to restore blood volume, enhancing renal blood flow, correcting metabolic acidosis and removing metformin and lactate, by diffusion through the dialyzer membrane (4). Unfortunately, this technique has not gained widespread acceptance due to the lack of well-conducted studies. Indeed, only case reports and retrospective design studies (with relatively low number of patients) have dealt with this subject. Compared to similarly severe lactic acidosis of another origin, the prognosis of MALA is significantly better (8). Death is primarily linked to the underlying health status and to the severity of the acute organ dysfunction (4). It is important to consider MALA in any metformin-treated patient and to start therapy promptly. Comment This paper that I have critically appraised is a retrospective cohort study (level of evidence 2b) over a five-year period (August 2002 to August 2007) of all patients (n=3556) who were either admitted to the intensive care unit (ICU) with metformin as a usual medication, or who attempted suicide by metformin ingestion. Within this population, 30 patients presenting with lactic acidosis, had MALA defining criteria (lactate of greater than 5 mmol/L and bicarbonate of less than 22 mmol/L), were selected and their clinical and biological features described. No patients had MALA as their admission diagnosis and most were admitted for management of shock or acute renal failure. Therefore, MALA was part of the clinical presentation in this patient cohort rather than an admission diagnosis. The patients were then divided according to their 28-day outcome in order to investigate if there were differences in relation to all the studied parameters. The population was also divided regarding the use of haemodialysis. Due to the retrospective design, no rules precluded the use of haemodialysis. Patients were not enrolled if a limitation of care was decided on admission. The goal was to assess the prevalence of MALA in a 16-bed, university-affiliated, ICU, and the effect of dialysis on patient outcome. The retrospective and non-interventional nature of this study waived the need for ethics committee approval. Appropriate statistical tests were used. The study’s key results: MALA is of low (1%) prevalence in medical ICUs and accounted for 0.84% of all admissions during the study period; it was associated with a 30% mortality rate. Eighty percent of these patients developed acute renal failure and 62.5% required hemodialysis. Only one patient with normal renal function was dialyzed because of severe acidosis. The only factors associated with a fatal outcome were the reason for admission in the ICU and the initial prothrombin time (as also revealed in the study by Seidowsky et al). The definition of MALA in this study did not duly account for people presenting primarily with tissue hypoperfusion as the likely cause of their lactic acidosis; although metformin may interfere with lactate clearance in a shock state, it is not thought to be the primary cause of the acidosis. The exact role of metformin in explaining the degree of lactic acidosis could not be definitely ascertained as most of the patients who died in this study were admitted with shock, suggesting that hypoperfusion, rather than metformin, was the principal cause of their lactic acidosis, as it may per se be associated with hyperlactataemia. Metformin is a factor that is detrimental to the outcome in the setting of an acute disease rather than the primary reason for referral to the ICU. MALA itself can present with hypotension due to negative inotropic effects and increased systemic vascular resistance with acidosis. Nevertheless, the recommended definition of MALA was strictly applied for the inclusion of the patients. The mortality rate of patients who received dialysis was similar to that of patients who were not dialyzed (31.3% versus 28.6%). However, it was the more acutely and chronically ill patients who actually received dialysis. This suggests that hemodialysis was beneficial in preventing a higher mortality rate in those who required renal replacement therapy. The mortality rate in MALA was not altered by hemodialysis. Several limitations of this study must be acknowledged. First, the small size of this study did not allow multiple comparisons and multivariate analyses. The patients who were dialyzed were more acutely ill as they had higher values on the SAPS II (Simplified Acute Physiology Score II), a larger burden of comorbidity (Charlson index) and more severe acidosis. Second, the current study was retrospective and therefore we couldn't correct the bias by which only the more severely ill patients were dialyzed.

Clinical Bottom Line

Indeed, only case reports and retrospective design studies with relatively low number of patients (obviously not included in the relevant papers in this Bet review, because the key words and outcomes were different than the ones used to formulate my three part question) have dealt with this subject, but in the great majority of those articles, dialysis techniques proved to be the adequate treatment, better than supportive care, in MALA patients treated in emergency and intensive care contexts. This goes along with my clinical experience, although small, in managing this entity. Unfortunately, although haemodialysis seems to possess a protective and benefit effect in MALA, it remains hypothetical, since there is no larger study dealing with MALA in the critical care setting, than the one revised above. So, larger and prospectively designed studies are clearly needed to draw firm recommendations on the treatment of MALA, especially concerning the possible beneficial effect of dialysis in the care of this disorder.

Level of Evidence

Level 2 - Studies considered were neither 1 or 3.

References

1. Peters et al; Metformin-associated lactic acidosis in an intensive care unit. Critical Care 2008; 12(6):1-5