The G8 HPLC analyzer from Tosoh Biosciences Inc is an NGSP-certified device that provides direct determination of stable HbA1c in 1.6 minutes using the gold standard Ion-exchange HPLC testing method.

Earlier this year, The Endocrine Society released a clinical practice guideline outlining recommendations regarding glycemic testing and goals. One of the more controversial suggestions—controversial in part because it is challenging—is that all patients, independent of a prior diagnosis of diabetes, have their blood glucose levels tested upon admission to a hospital.

The reasoning is based on evidence, as stated in the new guidelines. “Observational and randomized controlled studies indicate that improvement in glycemic control results in lower rates of hospital complications in general medicine and surgery patients,” and that the implementation of a “standardized subcutaneous insulin order set promoting the use of scheduled basal and nutritional insulin therapy is a key intervention in the inpatient management of diabetes.”1In theory, the guideline has a positive message, and many support the idea of more testing. “I believe anything we can do that comes with a reasonable cost burden, brings a reduction in morbidity and mortality, and improves patient safety is worth pursuing,” says James Donnelly, PhD, MBA, DABCC, FCACB, chief scientific officer for Siemens Healthcare Diagnostics Inc, Tarrytown, NY.

However, putting the guidelines into practice presents a few logistical challenges, of which cost is only one. Although some hospitals have begun to perform glycemic testing routinely on inpatient admissions, the process has not yet been widely adopted. Some feel additional information is needed, particularly regarding what sort of test is best.

“I have heard of confusion regarding which test to run: glucose or A1c. The initiative is still in the early states, and people aren’t sure what they should be doing,” says Ranka Milojkovic, BS, MBA, senior HPLC products manager for Tosoh Bioscience Inc, South San Francisco, Calif.

There are also questions about laboratory versus point-of-care (POC) testing. The guideline panel itself noted that “the accuracy of most handheld glucose meters is far from optimal,” though it still recommended their use.1 But with diabetes becoming so prevalent among the population at large, the challenges will eventually need to be overcome, even if there are compromises made today.


James Donnelly

James Donnelly, PhD, MBA, DABCC, FCACB, chief scientific officer, Siemens Healthcare Diagnostics.

“Hyperglycemia is a common, serious, and costly health care problem in hospitalized patients,” according to the new guidelines’ authors.1 The document was created by an eight-member task force composed of a chair, who was selected by the Clinical Guidelines Subcommittee of The Endocrine Society, six additional experts, and a methodologist.1The Centers for Disease Control and Prevention (CDC) estimates that diabetes affects 25.8 million people in the United States, representing 8.3% of the population. Of these, approximately 7 million people are undiagnosed. CDC data on the prevalence of prediabetes in the population is no better. The organization found that in 2005 to 2008, 35% of US adults aged 20 years or older had prediabetes; for those 65 and older, that figure rose to 50%.

Ranka Milojkovic

Ranka Milojkovic, BS, MBA, senior HPLC products manager, Tosoh Bioscience Inc

The Endocrine Society panel found similarly alarming estimates for hyperglycemia in its research, noting that observational studies have found hyperglycemia in 32% to 38% of patients in community hospitals, and that a sizable minority of these patients had no previous history of diabetes.1 Yet high blood sugar levels can negatively impact patient outcomes and, therefore, a hyperglycemic diagnosis can help to guide treatment decisions.

“Hyperglycemia is associated with prolonged hospital stay, increased incidence of infections, and death in noncritically ill hospitalized patients,” Guillermo Umpierrez, MD, Emory University, Atlanta, and chair of the task force that authored the clinical practice guideline, told The Endocrine Society.


The guidelines aim to provide a means by which hospitals can improve hyperglycemic-related issues with specific recommendations for what the panel describes as “practical, achievable, and safe glycemic targets,” as well as “protocols, procedures, and system improvements” for patients in noncritical care settings.

In addition to its recommendation to test all inpatients, other suggested guidelines cover: glycemic targets (eg, a pre-meal glucose target of less than 140 mg/dL and a random blood glucose of less than 180 mg/dL for the majority of hospitalized patients with noncritical illness); a scheduled subcutaneous insulin regimen for all patients with diabetes treated with insulin at home while in the hospital; intravenous (IV) continuous insulin infusion or subcutaneous basal insulin with bolus insulin as required to prevent hyperglycemia during the perioperative period for all patients with type 1 diabetes and most patients with type 2 diabetes who undergo surgical procedures; transitioning to scheduled subcutaneous insulin therapy at least 1 to 2 hours before discontinuation of IV continuous insulin infusion for all patients with type 1 and type 2 diabetes; and bedside POC glucose testing in all patients with high glucose values on admission and in patients with or without a history of diabetes receiving nutrition either through IV or a feeding tube.

The additional care is detailed because patients with hyperglycemia and/or diabetes are at greater risk for complications, according to Donnelly. “The patient’s therapy, nutritional status, and stress can affect blood sugar,” he says. But the guidelines provide a program that can help to mitigate the risk, reducing the patient length of stay and, ultimately, improving the outcome. “These guidelines have taken elements from other academic organizations and expanded what hospitals can do in terms of hyperglycemia,” Donnelly says.



For some hospitals, however, cost is a prohibitive factor, despite the potential positive long-term impact. “Instead of accepting that these patients may have an increased length of stay and an increased risk for infection and myocardial infarction, they can consider shifting their resources,” Donnelly says. “And eventually, with medical and administrative support, it becomes part of the routine.”

Initially (or even as a long-term strategy), hospitals may also choose to test only certain patient populations in order to maximize their investment. Most often, they select those groups who are at greatest risk based on the organization’s patient demographics. “Some facilities are testing based on age—for example, all admissions of patients over the age of 65,” Milojkovic says.

Others are testing based on patient population. “For instance, it would make more sense for a facility that specializes in endocrinology, diabetes, or renal failure to test all incoming patients than a general care facility,” Milojkovic says.

Although the guidelines are meant to apply to all patients, each institution will have to decide how to approach the issue based on its resources and patient population. This process will also impact how the tests are performed.

“Hospitals with a predictable inpatient population, perhaps a community hospital with smaller emergency room intake, will have most patients arrive in an orderly, scheduled manner, and inpatient testing in the laboratory works fine. But if a hospital has a large intake from trauma, point-of-care testing will be important because it allows physicians to quickly triage patients,” Donnelly says.

POC testing is also convenient because the glucose can then be adjusted bedside in response to the test, Donnelly says. But whether the tests are completed bedside or in the clinical laboratory, Donnelly says the impact on the lab would be minimal. “For many of the inpatient tests, it would just be an additional glucose to run with tests,” Donnelly says. The real impact would be felt by nursing, pharmacy, and the clinicians who have to write the orders and manage the patients.


Ultimately, the decisions come down to the physician who is ordering the test. “Certain physicians will do blanket testing that includes a full panel; some will perform the full panel on the first visit with regular follow-up testing,” Milojkovic says.

In ordering a glucose-related test, the physician must also decide whether to run a glucose test or an A1c. Generally, glucose tests are used for ER and critical care patients as chemistry is the preferred method, but POC tests can provide greater convenience. The disadvantage is the concern regarding their accuracy.

A 2011 evaluation presented by the FDA on glucose meters cleared in the previous 2 years noted that approximately 72% of devices would meet the requirement for +10 mg/dL at less than 75 mg/dL and approximately half would meet +15 mg/dL at less than 75 mg/dL.

These numbers are better than in the past, but manufacturers are constantly looking toward ways to improve the technology. In the meantime, feature improvements focus on flexibility, simplicity, and automation. “We’re facilitating simple tools, like the CentraLink Data Management System, that allows change on the fly,” Donnelly says. Alert rules can also be altered. Laboratories could decide to call out glucose results that fall below their current alerting levels.

The results of glucose testing are helpful in that they alert a physician to a potential blood sugar condition—whether hyperglycemia, hypoglycemia, prediabetes, or diabetes. “The disadvantage is that the result captures a finite time point as opposed to an A1c test, which provides a view of the patient’s glucose levels over a 3-month time period,” Milojkovic says.

A1c testing, however, requires more laboratory resources than a glucose test. “With A1c testing, the onus is on the lab technologist to decide if a result is correct or needs to be repeated or used with other criteria,” Milojkovic says. But trained technologists are hard to find and even more difficult to keep as greater numbers approach retirement age.

Analyzers that help to automate the process are available, but Milojkovic suggests that a dedicated A1c device can provide additional clinically useful information. “Manufacturers understand the need for consolidated panels and have incorporated A1c testing, but we believe A1c is an important analyte, and a dedicated instrument can deliver chromatography results and other information, like interference from other substances that you wouldn’t see on an analyzer,” Milojkovic says.

A1c bedside testing would be an ideal solution, but like glucose monitoring devices, accuracy has been difficult to achieve. “There are so many different kinds out there,” Milojkovic says. “There are inconsistencies in accuracy between devices, reagent lots, and systems.” Tosoh, and likely many other manufacturers, are looking into developing an A1c device for POC testing, but will not move forward until it has a technology that is consistent and reliable. “We don’t want to put out a me-too product, but want to offer something that stands out,” Milojkovic says.

Currently, there are very few standout products in the market. A 2010 study found that only two of eight HbA1c analyzers met the tightened criteria set by the National Glycohemoglobin Standardization Program (NGSP) (having a total CV of less than 3% in the clinically relevant range); six others did not.2 The Endocrine Society task force did recognize and suggest within its guidelines that only devices certified by the NGSP program be used.

As companies such as Tosoh invest in research and development, they are keeping a close eye on the guidelines, competition, and market in general. Tosoh is not rushing its development of a new POC device, in part because as a Japanese-based company, its strategies are more closely tied to its domestic market—where all patients are given a laboratory-run A1c test when admitted regardless of their age or medical background.

Whether the proposed initiative will catch on in the United States remains to be seen. Donnelly suggests a cost-benefit analysis would provide additional information to allow facilities to develop appropriate care measures, but ultimately, the real concern is for patient outcomes. Compromise is acceptable, but not when it compromises care, so eventually it may be all for testing and testing for all.

Renee Diiulio is a contributing writer for CLP.For more information, contact Editor Judy O’Rourke,

  1. Umpierrez GE, Hellman R, Korytkowski MT, et al. Management of hyperglycemia in hospitalized patients in non-critical care setting: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2012:97(1)16-38.
  2. Lenters-Westra E, Slingerland RJ. Six of eight hemoglobin A1c point-of-care instruments do not meet the general accepted analytical performance criteria. Clin Chem. 2010:56(1)44-52.