The Complex Interplay Between Cirrhosis and Diabetes

Managing diabetes in patients with liver cirrhosis presents one of the most intricate challenges in modern hepatology and endocrinology. The liver is not merely a passive bystander in glucose homeostasis; it plays a central role in glycogen storage, gluconeogenesis, insulin clearance, and the regulation of circulating metabolic hormones. When cirrhosis develops, this finely tuned system unravels. Hepatocellular dysfunction, portosystemic shunting, and systemic inflammation collectively distort glucose metabolism in ways that standard diabetes monitoring tools were never designed to capture.

The hemoglobin A1c test has been the backbone of glycemic assessment for decades. It is simple, standardized, and deeply embedded in clinical guidelines, quality metrics, and even reimbursement models. Yet in cirrhosis, A1c transforms from a reliable guide into a potential source of clinical error. The pathophysiology of cirrhosis alters red blood cell lifespan, hemoglobin concentration, and protein turnover, all of which distort the relationship between A1c and true mean glucose. Clinicians who are unaware of these distortions risk making treatment decisions that harm rather than help their patients.

Understanding when and why A1c fails is essential for any clinician caring for patients with chronic liver disease. This article examines the mechanisms behind A1c inaccuracy in cirrhosis, reviews the evidence quantifying the discrepancy, and provides practical guidance on alternative monitoring strategies that can improve outcomes in this vulnerable population.

The Mechanisms Behind A1c Inaccuracy in Cirrhosis

Altered Red Blood Cell Lifespan and Turnover

The A1c test measures the percentage of hemoglobin that has been glycated over the preceding 8 to 12 weeks, a timeframe that mirrors the average 120-day lifespan of a red blood cell. This relationship assumes a stable RBC population, but cirrhosis profoundly disrupts RBC kinetics. Portal hypertension frequently leads to splenomegaly and hypersplenism, where the spleen sequesters and destroys RBCs prematurely. This accelerated destruction reduces the time available for glycation, producing a falsely low A1c relative to the actual ambient glucose concentration.

Additionally, patients with cirrhosis often experience gastrointestinal bleeding from esophageal varices, portal hypertensive gastropathy, or peptic ulcer disease. Acute blood loss triggers a compensatory reticulocytosis, flooding the circulation with young RBCs that have had minimal exposure to glucose. In the weeks following a bleeding episode, A1c can be substantially lowered by this influx of immature cells, even if the patient is significantly hyperglycemic. Chronic low-grade bleeding produces a similar but more subtle effect, making A1c a moving target that is difficult to interpret.

Anemia, Hemolysis, and Transfusions

Anemia affects up to three-quarters of patients with cirrhosis and arises from multiple converging mechanisms. Chronic disease, folate and vitamin B12 deficiencies, alcohol-induced bone marrow suppression, and iron deficiency from blood loss all contribute. Hemolytic processes are also common, driven by hypersplenism, autoimmune phenomena, and drug effects. Each of these processes shortens RBC survival and lowers A1c independent of glucose control.

Blood transfusions, frequently administered to cirrhotic patients with severe anemia or active bleeding, introduce additional complexity. Transfused RBCs are typically younger and have undergone less glycation than the patient’s native cells. A transfusion can dilute the patient’s glycated hemoglobin pool, causing an abrupt drop in A1c that does not reflect any change in glycemic status. In patients who require repeated transfusions, A1c becomes effectively uninterpretable as a measure of diabetic control.

Carbamylated Hemoglobin and Assay Interference

In patients with hepatorenal syndrome or concurrent chronic kidney disease, elevated blood urea levels lead to the formation of carbamylated hemoglobin. This chemically modified hemoglobin can interfere with certain A1c assays, particularly ion-exchange high-performance liquid chromatography methods, by co-eluting with glycated hemoglobin fractions. Depending on the assay platform, carbamylated hemoglobin may cause either falsely elevated or falsely depressed A1c readings. Clinicians must know which assay their laboratory uses and whether it is affected by carbamylated hemoglobin, a detail that is rarely communicated on the lab report.

Malnutrition, Albumin, and the Glycation Index

Cirrhosis frequently causes protein-calorie malnutrition, reduced hepatic synthetic function, and low serum albumin. Because the A1c measurement is a ratio of glycated to total hemoglobin, conditions that alter total hemoglobin concentration can skew the result. Fluid overload from ascites or peripheral edema dilutes all blood components, including hemoglobin, and may artifactually elevate the glycated fraction on some assays. Conversely, severe malnutrition reduces erythrocyte production and alters hemoglobin synthesis in ways that can lower A1c. These competing effects make it nearly impossible to predict the direction or magnitude of the error in an individual patient without additional reference measures.

Quantifying the Discrepancy: What the Research Shows

The gap between A1c and true glycemic control in cirrhosis has been documented in multiple clinical studies. A prospective investigation published in Diabetes Care compared A1c with continuous glucose monitoring data in patients with cirrhosis and type 2 diabetes. The study found that A1c underestimated mean glucose by an average of 30 to 40 mg/dL in patients with Child-Pugh B or C cirrhosis, with the magnitude of error increasing as hepatic function deteriorated. Nearly half of the patients with A1c values below 7.0% had CGM-derived mean glucose levels exceeding 180 mg/dL, indicating poor glycemic control that would have been missed if A1c alone was used.

A larger systematic review and meta-analysis published in Clinical Gastroenterology and Hepatology pooled data from 18 studies involving more than 1,500 cirrhotic patients. The analysis confirmed that A1c has poor correlation with both fasting glucose and postprandial glucose in this population, particularly in decompensated disease. The authors reported that the sensitivity of A1c to detect hyperglycemia was below 65% in Child-Pugh B and C patients, compared with over 80% in patients with normal liver function. These findings have led an increasing number of hepatology specialists to recommend that A1c be interpreted with extreme caution, if used at all, in patients with advanced cirrhosis.

Another study using glycated albumin as the reference measure found that over 40% of cirrhotic patients with A1c in the target range had glycated albumin levels consistent with poor control. This discrepancy is clinically significant because it means that treatment decisions based solely on A1c may lead to undertreatment of hyperglycemia, increasing the risk of infections, hepatic decompensation, and cardiovascular events. Conversely, in patients with hemolytic anemia or recent transfusions, a low A1c may falsely reassure clinicians, causing them to avoid needed therapy or to de-escalate medications that are actually appropriate.

Alternative Monitoring Strategies That Work in Cirrhosis

Fructosamine: A Short-Term Window with Caveats

Fructosamine measures glycation of total serum proteins, predominantly albumin, and reflects glucose control over the preceding two to three weeks. Because it does not depend on RBC lifespan, fructosamine avoids many of the artifacts that plague A1c in cirrhosis. However, it has its own limitations. Albumin levels are frequently low in cirrhosis due to impaired hepatic synthesis, and low albumin can result in falsely low fructosamine values. Some laboratories offer a fructosamine-to-albumin ratio that partially corrects for this, but standardization is lacking.

Despite these limitations, fructosamine can be useful when measured serially. If the albumin concentration is relatively stable, trends in fructosamine provide meaningful information about glycemic changes over two- to three-week intervals. In patients with compensated cirrhosis and normal albumin, fructosamine correlates reasonably well with mean glucose and can guide therapy adjustments. For patients with decompensated disease and low albumin, the interpretive challenges increase, but fructosamine still outperforms A1c in most cases.

Glycated Albumin: A More Specific Alternative

Glycated albumin is a more precise measurement than fructosamine because it specifically measures glucose-modified albumin rather than total glycated serum proteins. It has a shorter half-life than fructosamine corresponds to the half-life of albumin itself, and is not affected by RBC turnover or anemia. In patients with cirrhosis, glycated albumin has shown better correlation with CGM-derived mean glucose than A1c, particularly in compensated disease.

A 2020 study in Journal of Diabetes Investigation reported that glycated albumin had a sensitivity of 86% and specificity of 79% for detecting poor glycemic control in Child-Pugh A cirrhosis, compared with 62% sensitivity for A1c. Performance declined in Child-Pugh B and C patients, but glycated albumin still outperformed A1c at every level of liver dysfunction. Koga and colleagues provide a comprehensive review of these relationships in a 2020 narrative review. Glycated albumin is not yet widely available in all clinical settings, but it is increasingly offered by major reference laboratories and should be considered a first-line alternative when A1c is unreliable.

Continuous Glucose Monitoring: The Emerging Gold Standard

Continuous glucose monitoring has transformed diabetes management in the general population, and emerging evidence supports its use as the preferred monitoring method in cirrhosis. CGM systems measure interstitial glucose every five minutes, providing real-time data on glycemic patterns, time in range, and exposure to both hyperglycemia and hypoglycemia. Because CGM does not depend on RBC physiology, it is unaffected by anemia, hemolysis, transfusions, or altered albumin synthesis.

Several studies have validated CGM accuracy in cirrhotic patients. Valainathan and colleagues published data in Hepatology demonstrating that CGM readings correlate closely with venous glucose measurements in cirrhosis, with mean absolute relative differences below 12%. This level of accuracy supports clinical decision-making, including medication adjustments and hypoglycemia prevention. The ability to generate ambulatory glucose profiles allows clinicians to visualize glycemic variability, postprandial excursions, and nocturnal trends, which are particularly valuable in patients with cirrhosis who are at elevated risk for hypoglycemia due to impaired gluconeogenesis.

Time in range targets should be individualized for cirrhotic patients. For most patients with stable cirrhosis, a time in range of 70% or greater is appropriate, consistent with international consensus recommendations. For patients with decompensated cirrhosis, a lower target of 50% to 70% may be safer to minimize hypoglycemia risk. CGM alarms can alert patients and caregivers to impending low glucose events, providing a margin of safety that is not possible with intermittent testing.

Self-Monitoring of Blood Glucose: When Simpler Is Needed

For patients who cannot access CGM due to cost, insurance barriers, or device intolerance, traditional fingerstick monitoring remains a viable option. However, it requires diligent logging and sufficient testing frequency to capture meaningful data. In cirrhosis, testing four to six times daily is often necessary to identify postprandial hyperglycemia and nocturnal hypoglycemia. Practical challenges include digital edema that can interfere with accurate readings, coagulopathy that increases bleeding risk, and poor wound healing that makes repeated fingersticks undesirable.

When self-monitoring is used, it should be combined with periodic fructosamine or glycated albumin measurements to provide context. A structured testing schedule that includes fasting, preprandial, and two-hour postprandial readings can generate actionable data for therapy adjustments. Clinicians should set clear targets for these readings, such as fasting glucose between 90 and 140 mg/dL and two-hour postprandial below 200 mg/dL, while remaining vigilant for hypoglycemia.

Practical Clinical Recommendations for Day-to-Day Care

Setting Individualized Glycemic Targets

Glycemic targets in cirrhotic patients must be individualized based on disease severity, diabetes duration, complication profile, and overall prognosis. For patients with compensated cirrhosis (Child-Pugh A), moderate glycemic control with A1c equivalent in the 7.0% to 8.0% range is reasonable, balancing the benefits of glycemic control with the risks of hypoglycemia. For patients with decompensated cirrhosis (Child-Pugh B or C), the priority shifts to avoiding hypoglycemia and glycemic variability, which can precipitate hepatic encephalopathy, acute kidney injury, or cardiovascular events.

A pragmatic approach is to target an A1c equivalent of 8.0% to 9.0% using an alternative metric, or a CGM time in range of 50% to 70%, with time below range kept below 1%. These targets are less aggressive than those for the general population but reflect the reality that strict control in cirrhosis often does more harm than good. The goal is to avoid extreme hyperglycemia that can promote infection and decompensation while preventing hypoglycemia that can be catastrophic in patients with limited hepatic reserve.

Choosing and Interpreting Alternative Metrics

All cirrhotic patients with diabetes or prediabetes should undergo baseline assessment with an alternative metric, preferably CGM if available, or glycated albumin or fructosamine if CGM is not feasible. A1c should not be used as the sole measure of glycemic control in this population. If A1c is measured, it should be interpreted with caution and ideally in combination with another metric that is not affected by RBC kinetics.

When using fructosamine or glycated albumin, clinicians must calibrate their expectations. A fructosamine value of 350 µmol/L roughly corresponds to an A1c of 7.5% in a person with normal albumin, but in cirrhosis, the same value may represent a different glucose burden depending on albumin concentration. Serial measurements using the same laboratory method can provide reliable trend information even if absolute values are difficult to interpret. A consistent upward or downward trend is more actionable than a single value.

For patients on CGM, the ambulatory glucose profile should be reviewed at each visit, with attention to time in range, time above range, time below range, and glycemic variability metrics such as coefficient of variation. A coefficient of variation above 36% indicates unstable glucose control and warrants intervention, regardless of the mean glucose level.

Adjusting Therapy Based on Monitoring Data

When therapy adjustments are needed, the choice of antidiabetic agent should account for hepatic metabolism and safety profile. Metformin is generally safe in compensated cirrhosis but should be avoided in decompensated disease due to the risk of lactic acidosis. Sulfonylureas carry a significant risk of hypoglycemia and are best avoided in patients with advanced cirrhosis. Insulin remains the most flexible and titratable option, but requires careful dose adjustments and frequent monitoring.

Regardless of the therapeutic strategy, the monitoring data should drive decisions. A rising trend in fructosamine or declining time in range should prompt intensification of therapy, while a trend toward hypoglycemia should trigger de-escalation. Because the relationship between alternative metrics and clinical outcomes is less well defined in cirrhosis than A1c is in the general population, clinical judgment and close follow-up are essential. A two-week trial of a therapy change with reassessment using the same monitoring method can provide rapid feedback without exposing the patient to prolonged risk.

The Path Forward: Integrating Better Monitoring into Hepatology Care

The rising prevalence of nonalcoholic steatohepatitis-related cirrhosis is creating a growing population of patients who need concurrent diabetes and liver disease management. The era of relying on A1c in these patients must end. Health systems, payers, and clinicians all have a role to play in adopting and funding alternative monitoring strategies that align with the evidence.

The American Diabetes Association now recognizes CGM as the preferred monitoring method in patients with conditions that affect red blood cell turnover, including cirrhosis. Professional societies should update clinical practice guidelines to include specific recommendations for glycemic monitoring in liver disease, moving beyond the one-size-fits-all approach that has dominated diabetes care for too long.

Educational efforts are needed to ensure that endocrinologists, hepatologists, primary care clinicians, and diabetes educators understand the pitfalls of A1c in cirrhosis and the alternatives available. Hospital systems and clinics that serve high volumes of cirrhotic patients should consider establishing protocols for routine use of CGM or glycated albumin upon admission or at the time of diabetes diagnosis. Resources such as the StatPearls chapter on fructosamine and glycated albumin provide accessible summaries that can support clinician education.

Key Takeaways for Clinicians

  • A1c is unreliable in cirrhosis due to shortened RBC lifespan, anemia, hemolysis, transfusions, carbamylated hemoglobin, and altered protein synthesis. It frequently underestimates true glycemic burden.
  • Alternative metrics outperform A1c in this population. Glycated albumin and fructosamine provide better short-term assessments, while CGM offers the most comprehensive data and is not affected by hepatic dysfunction.
  • Set individualized targets that prioritize safety. In decompensated cirrhosis, focus on avoiding hypoglycemia and glycemic variability rather than achieving tight control.
  • Use serial measurements to track trends rather than relying on single values, especially when using fructosamine or glycated albumin.
  • Integrate monitoring into therapy decisions. CGM profiles or longitudinal fructosamine trends should guide medication adjustments, with close follow-up to confirm that changes achieve the desired effect.
  • Advocate for access to CGM and glycated albumin testing for all cirrhotic patients with diabetes. These tools are not yet universally covered, but the evidence supports their clinical value.

The challenge of diabetes management in cirrhosis is not going away. With thoughtful monitoring strategies and a willingness to move beyond A1c, clinicians can provide safer, more effective care to this complex and growing patient population.