Comparing Heel Stick Glucoses in Neonates

Newborn babies can develop low blood sugar (glucose) which can lead to brain injury and poor developmental outcomes. Therefore, it is important to accurately measure the blood glucose in babies. One way to measure the blood glucose is to test blood from the baby's heel with a bedside device called a point of care glucometer. This method is very common and easy; however, multiple factors can lead to an inaccurate reading. A false low reading may require additional blood testing and admission to the NICU. A false high reading could result in the medical provider missing the diagnosis of low blood glucose. Our team wants to know if there is a difference between blood glucose measurements taken from warmed and un-warmed heels of infants. Blood flow farther away from the heart, such as in the feet and heels, may be less than the rest of the body, and might move more slowly. This could cause the glucose level to be lower in the feet and heels. Therefore, sampling blood from an un-warmed heel may result in a falsely low glucose reading. There is some research that suggests warming the heel increases blood flow to the area; however, only one study that we know of has evaluated differences in blood glucose readings from warmed and un-warmed heels. They found significantly higher blood glucose readings from warmed heels compared to un-warmed heels in 57 babies. However, these babies were part of a larger study comparing different diets on blood glucose levels, and the heels were warmed using warm water which is no longer a current practice. The goal of this study is to compare the capillary blood glucose levels from warmed and un-warmed heels in about 100 infants who are breast and/or formula fed using the current practice of warming heels with gel heat packs. The null hypothesis is that there will be no difference between capillary blood glucose levels sampled from an infants warmed and un-warmed heel. The alternative hypothesis is that capillary blood glucose levels sampled from warmed heels will be higher than those samples from un-warmed heels.

Neonatal hypoglycemia is defined by the American Academy of Pediatrics as blood glucose level less than 47 mg/dL and is one of the most common diagnoses requiring admission to the neonatal intensive care unit (NICU).1,2 The incidence of hypoglycemia in healthy newborns is approximately 5-15%, and as high as 51% in infants with risk factors.3 The cost of hospitalization of neonatal hypoglycemia is estimated to range from $2,350-$12,755.4 Hypoglycemia in the newborn has been associated with permanent brain injury including neurodevelopmental delay, learning difficulties and seizures.5 Clinically important hypoglycemia is not always associated with clinical exam findings;6 therefore, accurate measurement of blood glucose in the newborn is of great importance. Laboratory testing of venous blood, typically plasma, is a widely accepted method of diagnosing neonatal hypoglycemia.6 However, point of care (POC) testing of capillary blood using a bedside glucometer is a common screening tool in newborn nurseries and NICUs. The POC testing method provides real time results and is an easier, less invasive sampling technique.6 Despite these conveniences, studies have found POC testing to be less accurate than venous testing.7-9 POC testing is also less accurate at lower glucose levels and readings are affected by polycythemia and hyperbilirubinemia, common findings in newborn babies.6,7,10 Inaccurate readings may result in unnecessary admissions to the NICU for additional testing and management. Therefore, efforts to optimize the accuracy POC glucose testing in screening for neonatal hypoglycemia is imperative. One common practice thought to increase the accuracy of capillary POC glucose testing is to warm the infant's heel before blood sampling. It has been proposed that warming the heel will increase blood flow to the capillary bed and improve peripheral blood stasis that may otherwise result in a falsely low glucose reading.11,12 Current practice is to warm the heel with a gel heat pack designed and marketed specifically for the purpose of warming the infant's heel prior to blood collection. These gel heat packs are an additional cost to the hospital, and currently it is unclear if there is any significant clinical benefit to warming the infant's heel before obtaining capillary blood glucose measurements. Various studies utilizing laser Doppler flowmetry have suggested that local warming results in increased skin blood flow and perfusion in infants.13-15 This increase in blood flow is thought to be reflected in the volume of blood collected and ease of collection. In practice, however, the effect of local warming on capillary blood collection in neonates appears to be minimal. One study found that heel warming in infants did not improve the volume of blood collected or the ease of blood collection.16 A subsequent study found that warming the heel decreased the need to squeeze the heel for blood collection, however, there was no difference in volume of blood collected or collection time.17 To the best of our knowledge, the only study that has directly compared the effect of warming the heel on capillary blood glucose in infants was conducted by Russell and McKay in 1966.18 The study compared simultaneously collected capillary blood glucose levels from a warmed and un-warmed heel in 57 infants. Overall, glucose levels from the warmed and un-warmed heels were highly significantly correlated. However, the mean glucose concentration from blood in warmed heels was 2.8 mg/dL higher than that from un-warmed heels. Additionally, of the 57 infants, 13 had a statistically significant higher glucose level measured in their warmed heel compared to their un-warmed heel. The Russell and McKay study suggested that the infant's heel must always be warmed before collecting capillary blood for glucose measurement. However, the study included only 57 infants, a subset of which were also assigned to a low calorie, high calorie, or hydrocortisone administration groups to determine the outcomes of these interventions on blood glucose levels. Additionally, the infant's heels were warmed by soaking in warm water. This is no longer the current practice as most institutions use gel heat packs designed specifically for warming the heels of infants. The goal of this study is to simultaneously compare the capillary blood glucose levels from warmed and un-warmed heels in approximately 100 infants who are being breast and/or formula fed using the current practice of warming heels with gel heat packs. The null hypothesis is that there will be no difference between capillary blood glucose levels sampled from an infants warmed and un-warmed heel. It has been proposed that warming the heel will increase blood flow to the capillary bed and improve peripheral blood stasis that may otherwise result in a falsely low glucose reading. Therefore, the alternative hypothesis is that capillary blood glucose levels sampled from warmed heels will be higher than those samples from un-warmed heels. After admission to the newborn nursery or NICU, the infant's mother will be approached regarding her participation and of her newborn baby. If participation is desired, mom will sign consent for herself and for the infant; PHI authorization will be obtained. Warmed and un-warmed heel capillary blood glucose will be obtained from the infant at the time of a clinically indicated point of care glucose check. A heating pad will be placed on one heel for 10 minutes. The other heel will not be warmed. A lancet will be used to produce blood at both heels, and blood glucose will be measured from both heels using the Nova Stat glucometer. Data from the hospitalization will be extracted from the Electronic Health Record from both mother and baby. Data will include demographic information, medications and health conditions, delivery information, and anthropometric data. Data analysis will be done using SPSS Statistic software to evaluate for differences between capillary blood glucose levels from warmed and unwarmed heels. 1. Thompson-Branch A and Havranek T. Neonatal Hypoglycemia. Pediatr Rev. April 2017;38(4):147-157. DOI: https://doi.org/10.1542/pir.2016-0063. 2. Alsaleem M, Saadeh L, and Kamat D. Neonatal Hypoglycemia: A Review. Clin Pediatr (Phila). 2019 Nov;58(13):1381-1386. doi: 10.1177/0009922819875540. Epub 2019 Sep 26. 3. Hosagasi NH, Aydein M, Zenciroglu A, Ustun N, and Beken S. Incidence of hypoglycemia in newborns at risk and an audit of the 2011 American academy of pediatrics guideline for hypoglycemia. Pediatr Neonatl. 2018 Aug;59(4):368-374. doi: 10.1016/j.pedneo.2017.11.009. Epup 2017 Nov 15. 4. Alemu BT, Beydoun HA, Hoch M, Van Lunen B, and Akpinar-Elci M. Hospitalization cost in infants with hypoglycemia. Curr Pediatr Res. Dec 2018:22(4). 5. Rozance PJ and Wolfsdorf JI. Hypoglycemia in the Newborn. Pediatr Clin North Am. Apr 2019;66(2):333-342. doi: 10.1016/j.pcl.2018.12.004. Epub 2019 Feb 1. 6. Beardsall K. Measurement of glucose levels in the newborn. Early Hum Dev. May 2010;86(5):263-267. doi: 10.1016/j.earlhumdev.2010.05.005. Epub 2010 Jun 9. 7. Hussain K and Sharief N. The inaccuracy of venous and capillary blood glucose measurement using reagent strips in the newborn period and the effect of haematocrit. Early Hum Dev. 2000 Feb;57(2):111-21. doi: 10.1016/s0378-3782(99)00060-2. 8. Ameur K et al. Evaluation of the measurement of capillary glucose concentration versus plasma glucose in the newborn. Arch Pediatr. 2016 Sep;23(9):908-12. doi:10.1016/j.arcped.2016.04.025. Epub 2016 Jun 28. 9. Balion C, Grey V, Ismaila A, Blatz S, and Seidlitz W. Screening for hypoglycemia at the bedside in the neonatal intensive care unit (NICU) with the Abbott PCx glucose meter. BMC Pediatr. 2006 Nov 3;6:28. doi: 10.1186/1471-2431-6-28. 10. Hamid MH, Chrishti AL and Maqbool S. Clinical utility and accuracy of a blood glucose meter for the detection of neonatal hypoglycemia. J Coll Physicians Surg Pak. 2004 Apr;14(4):225-8. 11. Short BL. Capillary blood sampling. In: Fletcher MA, MacDonald MG, eds. Adas of procedures in neonatology. Philadelphia: J B Lippincott Co, 1993: 97-100. 12. Meites S. Skin-puncture and blood-collecting technique for infants: update and problems. Clin Chem. 1988 Sep;34(9):1890-4. 13. Suichies HE, Brouwer C, Aarnoudse JG, Jentink HW, de Mul FFM, and Greve J. Skin blood flow changes, measured by laser doppler flowmetry, in the first week after birth. Early Hum Devel 1990; 23: 1-8. 14. Takayanagi T, Fukunda M, Nakazawa M, Tanaka S, and Yoshinaga M. Response of skin blood volume, velocity and flow to local warming in newborns, measured by laser Doppler flowmetry. Pediatr Int. 1999 Dec;41(6):624-30. doi: 10.1046/j.1442-200x.1999.01145.x. 15. Beinder E, Trojan A, Bucher HU, Huch A, and Huch R. Control of skin blood flow in pre- and full- term infants. Biol Neonate. 1994;65(1):7-15. doi: 10.1159/000244021. 16. Barker DP, Willetts B, Cappendjik VC, and Rutter N. Capillary blood sampling: should the heel be warmed? Arch Dis Child 1996;74:139-40. 17. Janes M, Pinelli J, Landry S, Downey S, and Paes B. Comparison of capillary blood sampling using an automated incision device with and without warming of the heel. J Perinatol. 2002;22:154-58. 18. Russell G and McKay E. Blood Glucose Concentration in the Perinatal Period. Arch Dis Child. 1966 Oct;41:496-502.s0378-3782(99)00060-2. 8\. Ameur K et al. Evaluation of the measurement of capillary glucose concentration versus plasma glucose in the newborn. Arch Pediatr. 2016 Sep;23(9):908-12. doi:10.1016/j.arcped.2016.04.025. Epub 2016 Jun 28. 9\. Balion C, Grey V, Ismaila A, Blatz S, and Seidlitz W. Screening for hypoglycemia at the bedside in the neonatal intensive care unit (NICU) with the Abbott PCx glucose meter. BMC Pediatr. 2006 Nov 3;6:28. doi: 10.1186/1471-2431-6-28. 10\. Hamid MH, Chrishti AL and Maqbool S. Clinical utility and accuracy of a blood glucose meter for the detection of neonatal hypoglycemia. J Coll Physicians Surg Pak. 2004 Apr;14(4):225-8. 11\. Short BL. Capillary blood sampling. In: Fletcher MA, MacDonald MG, eds. Adas of procedures in neonatology. Philadelphia: J B Lippincott Co, 1993: 97-100. 12\. Meites S. Skin-puncture and blood-collecting technique for infants: update and problems. Clin Chem. 1988 Sep;34(9):1890-4. 13\. Suichies HE, Brouwer C, Aarnoudse JG, Jentink HW, de Mul FFM, and Greve J. Skin blood flow changes, measured by laser doppler flowmetry, in the first week after birth. Early Hum Devel 1990; 23: 1-8. 14\. Takayanagi T, Fukunda M, Nakazawa M, Tanaka S, and Yoshinaga M. Response of skin blood volume, velocity and flow to local warming in newborns, measured by laser Doppler flowmetry. Pediatr Int. 1999 Dec;41(6):624-30. doi: 10.1046/j.1442-200x.1999.01145.x. 15\. Beinder E, Trojan A, Bucher HU, Huch A, and Huch R. Control of skin blood flow in pre- and full- term infants. Biol Neonate. 1994;65(1):7-15. doi: 10.1159/000244021. 16\. Barker DP, Willetts B, Cappendjik VC, and Rutter N. Capillary blood sampling: should the heel be warmed? Arch Dis Child 1996;74:139-40. 17\. Janes M, Pinelli J, Landry S, Downey S, and Paes B. Comparison of capillary blood sampling using an automated incision device with and without warming of the heel. J Perinatol. 2002;22:154-58. 18\. Russell G and McKay E. Blood Glucose Concentration in the Perinatal Period. Arch Dis Child. 1966 Oct;41:496-502.