What to do when there's a significant difference in Capillary Blood Glucose (CBG) readings between the upper and lower limb?

Significant Differences in Capillary Blood Glucose Between Upper and Lower Limbs

When encountering significant differences in capillary blood glucose (CBG) readings between upper and lower limbs, immediately switch to arterial or venous whole blood sampling from the upper extremity, as capillary measurements from lower limbs are unreliable in patients with peripheral hypoperfusion, shock, or peripheral arterial disease. 1

Understanding the Problem

Capillary blood glucose measurements can vary significantly between anatomical sites, particularly when comparing upper versus lower extremities. This discrepancy stems from fundamental physiological differences in perfusion states:

• Fingertip capillary blood most accurately reflects true blood glucose in stable patients, as it has the most reliable perfusion and fastest equilibration with arterial blood 1

• Lower extremity capillary samples are inherently unreliable in critically ill patients or those with peripheral vascular compromise, showing unpredictable variance that can be either falsely elevated or falsely low compared to arterial/venous samples 1

• Hypoperfusion increases glucose extraction in peripheral tissues, creating greater discrepancies between capillary whole blood and venous or arterial plasma glucose, with no consistent pattern to the variability 1

Immediate Action Algorithm

Step 1: Assess Patient Perfusion Status

Evaluate for conditions that compromise peripheral perfusion:

• Shock state or hypotension 1
• Vasopressor therapy 1
• Severe peripheral edema 1
• Mottled skin appearance 1
• Peripheral arterial disease (check for decreased ankle-brachial index, claudication history) 1

Step 2: Establish Sampling Site Hierarchy

Use this priority order for blood glucose measurement: 1

1. First choice: Arterial line sampling - Most similar to laboratory plasma values in paired samples 1
2. Second choice: Venous sampling - Generally acceptable if contamination from IV glucose-containing fluids is avoided 1
3. Third choice: Upper extremity fingertip capillary - Only in hemodynamically stable patients without edema 1
4. Avoid: Lower extremity capillary sampling - Unreliable in any compromised perfusion state 1

Step 3: Confirm with Laboratory Reference

When significant discrepancies exist (>20% difference between sites):

• Obtain laboratory plasma glucose measurement using arterial or venous sample to establish true glucose level 1
• Glucose meter accuracy standards allow up to 15 mg/dL variance for glucose <75 mg/dL and up to 20% variance for glucose ≥75 mg/dL, but these standards assume optimal conditions 1
• Meter performance can deviate >20% from laboratory control in critically ill patients regardless of blood source 1

Clinical Context: When Lower Limb Readings Are Particularly Problematic

Peripheral Arterial Disease Considerations

• Coarctation of the aorta causes decreased perfusion to lower extremities with increased gradient between upper and lower limbs 1
• Blood pressure measurements should be taken in both arms and one lower extremity in patients with suspected coarctation 1
• Decreased ankle-brachial index (abnormal gradient between upper and lower limbs) indicates significant arterial compromise 1

Alternate Site Testing Limitations

• Forearm and thigh blood glucose measurements lag behind fingertip values during rapid glucose changes 1
• After meals: forearm blood glucose rises more slowly and peaks lower than fingertip 1
• After exercise: thigh and forearm glucose levels fall lower than fingertip glucose 1
• During insulin treatment: alternate sites show delayed response 1

Common Pitfalls to Avoid

Do not average discrepant readings from different sites - this provides a meaningless value that doesn't reflect true glycemia 1

Do not continue using lower extremity capillary measurements in patients on vasopressors or with hypoperfusion, as this can lead to dangerous insulin dosing errors 1

Do not assume the higher reading is correct - the variability is unpredictable and can go in either direction 1

Avoid finger-stick testing as the primary method for any patient requiring prolonged insulin infusion, as it should be the site of last resort 1

Special Populations

Critically Ill Patients

• Establish institutional protocols that prioritize arterial or venous sampling for all patients in shock or on insulin infusions 1
• Minimize blood waste with catheter sampling devices to prevent iatrogenic anemia from frequent phlebotomy 1

Patients with Diabetes and Peripheral Arterial Disease

• Screen for peripheral arterial disease in all diabetic patients with unexplained glucose measurement discrepancies 1, 2
• Consider ankle-brachial index measurement if lower extremity perfusion is questionable 1, 2