Editorials
Clinical Chemistry 59:10
1427–1429 (2013)
Reporting Hemoglobin A1c:
Do the Units Matter?
David B. Sacks1*
Hemoglobin A1c (Hb A1c)2 is the analyte most commonly measured in clinical laboratories for patients
with diabetes mellitus. Physicians use Hb A1c to monitor long-term glycemic control, adjust therapy, and
predict complications of diabetes. Recently, Hb A1c was
added as a criterion for the diagnosis of diabetes. Unlike most analytes, many patients with diabetes know
their Hb A1c value and are familiar with the target value
they need to attain for good glycemic control.
Glycation of proteins was recognized over 100
years ago. Hb A1c was first isolated (by chromatography) in 1958 (1 ) and was identified 8 years later as a
Schiff base between hemoglobin and a ketone or aldehyde (2 ). Early commercial assays for Hb A1c were limited by the absence of standardization. For example, a
comparison of 7 methods in 1992 revealed very large
differences (⬎200%) in Hb A1c values among methods
(3 ). Standardization programs were developed soon
after 1993, when the Diabetes Control and Complications Trial (DCCT) documented the clinical value of
Hb A1c in patients with type 1 diabetes (4 ). The publication 5 years later of results for the UK Prospective
Diabetes Study (UKPDS), which revealed that decreasing Hb A1c in patients with type 2 diabetes reduced
complications (5 ), further emphasized the essential contribution of Hb A1c monitoring to patient care. Importantly, the Hb A1c results obtained in the UKPDS were
aligned with those in the DCCT. Therefore, appropriate
patient management requires that an individual’s Hb A1c
value be traceable to DCCT/UKPDS values.
The NGSP, previously known as the National Glycohemoglobin Standardization Program, was formed
to achieve this objective. The NGSP laboratory network assists manufacturers of Hb A1c assays in calibrating their methods so that patient values correlate with
those of the DCCT and the UKPDS (6 ). Analogous to
the clinical-outcomes trials, Hb A1c is reported as a
1
Department of Laboratory Medicine, National Institutes of Health, Bethesda,
MD.
* Address correspondence to the author at: Department of Laboratory Medicine,
Bldg. 10, Rm. 2C306, 10 Center Dr., Bethesda, MD 20892. Fax 301-402-1885;
e-mail sacksdb@mail.nih.gov.
Received July 10, 2013; accepted July 15, 2013.
Previously published online at DOI: 10.1373/clinchem.2013.211227
2
Nonstandard abbreviations: Hb A1c, hemoglobin A1c; DCCT, Diabetes Control
and Complications Trial; UKPDS, UK Prospective Diabetes Study.
percentage of the total hemoglobin. The NGSP, which
has markedly improved Hb A1c standardization (6 ),
has been adopted by numerous countries. A few years
after the establishment of the NGSP, the IFCC developed a higher-order reference method for Hb A1c. A
comparison of the NGSP and IFCC networks revealed
a linear relationship, but IFCC values were approximately 1.5% to 2% lower (7 ). An influential decision
was made to report IFCC results in SI units [millimoles
of Hb A1c per mole of Hb A0 (8 )], and several countries, predominantly in Europe and the Antipodes,
have adopted SI units for Hb A1c (9 ). One of the motivations for the use of SI units was that the large difference in results (e.g., an Hb A1c value of 6.5% is equivalent to 48 mmol/mol) would eliminate confusion
regarding the reporting schemes, but no evidence to
support this premise has previously been published.
The study reported by Kilpatrick et al. (10 ) in this
issue of Clinical Chemistry has addressed this topic by
evaluating the effect on glycemic control of changing
Hb A1c reporting from DCCT/NGSP units (a percentage) to SI units (millimoles per mole). Data were derived from all the patients on the diabetes register in
Hull and East Yorkshire, UK. In the UK, Hb A1c was
reported exclusively as a percentage until June 1, 2009,
when “dual reporting” of both DCCT/NGSP units and
the new units was instituted. Beginning October 1,
2011, Hb A1c results were reported only in SI units.
Kilpatrick et al. (10 ) compared Hb A1c values during
the last year of dual reporting (October 1, 2010, to September 30, 2011) to those obtained during the first year
after DCCT/NGSP units were eliminated (October 1,
2011, to September 30, 2012). The authors have drawn
2 general conclusions from their data analysis: (a)
There was no difference in the overall glycemic control
of the entire population before and after the units were
changed, and (b) the change in Hb A1c in the patients
who had poor glycemic control (defined in their study
as an Hb A1c value ⬎8%) was not altered by the units
used for reporting.
This concise—and timely—report, which provides valuable information, exhibits several positive attributes. The population was large, with 44 721 Hb A1c
measurements (21 880 in the first year and 22 841 in
the second year) obtained from 13 197 patients, all of
whom have diabetes. All samples were analyzed in a
single laboratory using a single method. This strategy
1427
Editorials
minimized any potential effect of interlaboratory imprecision or bias on the data. The analytical performance was excellent, with interassay CVs of 1.7% and
1.3% at Hb A1c concentrations of 5.4% (36 mmol/mol)
and 9.3% (80 mmol/mol), respectively. These results
meet the recommended intralaboratory imprecision of
⬍2% (11 ). The data demonstrate that the conversion
of units for reporting Hb A1c in the UK did not lead to
deterioration in glycemic control during the period of
the analysis.
Like all good studies, this report elicits numerous
questions. One is eager to know more about the patients. What was the breakdown between patients with
type 1 diabetes and those with type 2 diabetes? Were
there any differences between the 2 groups? Were the
findings for children and adults with type 1 diabetes
identical? Was there any difference between patients
who were managed by diabetologists and those treated
by general practitioners? Were DCCT/NGSP values
made available to some physicians and/or patients
when only SI units were on the printed reports? If so,
did this information influence patients’ Hb A1c concentrations? Other factors, such as the introduction of
new oral agents or insulins and changes in patient demographics (e.g., deaths or new individuals diagnosed
with diabetes), may have influenced the findings. The
most pertinent limitation of the study is the short period of analysis. The authors present no data regarding
Hb A1c values for the patients in the year preceding the
institution of dual reporting. More importantly, the
brief follow-up period does not guarantee that longterm glycemic control will remain unchanged. A prior
study in Sweden evaluated the consequences of altering
the scale of Hb A1c results when methods were changed
(12 ). Analogous to the present work, glycemic control
in the first year was essentially the same as before the
change; however, glycemic control in the Swedish patients deteriorated approximately 1 year after the reference interval was lowered. Long-term follow-up of the
patients evaluated by Kilpatrick et al. is required to address this possible concern.
Perhaps the most consequential question that
arises from this report is whether these findings can be
extrapolated to other countries. Several factors need to
be considered. The study was conducted in the UK, a
very wealthy country that devotes considerable resources (approximately 10% of gross domestic product
in 2010) to healthcare (13 ). A highly coordinated, wellthought-out campaign was undertaken to educate
healthcare professionals and patients. Dual reporting
was carried out for 28 months, and all laboratories in
the UK converted to SI units on the same day, thereby
minimizing the potential for confusion. That the standard of care was high is exemplified by the observation
that the mean number of days between successive Hb
1428 Clinical Chemistry 59:10 (2013)
A1c analyses for patients with poor glycemic control
(Hb A1c ⬎8%) was 99 days. This finding is consistent
with guidelines that recommend testing Hb A1c every 3
months for patients who are not meeting treatment
goals (11 ). I look forward to publications of glycemic
control from other countries that have changed Hb A1c
units.
The same data could have a different interpretation. All the effort invested to switch units failed to
produce any improvement in the glycemic control of
patients with diabetes. Some might be tempted to use
the data to invoke the counterargument that there is no
evidence for promoting a change of units. Many countries lack both Hb A1c standardization and the proficiency testing necessary to ensure that clinical laboratories measure Hb A1c accurately. Some diabetologists
from these countries have voiced the opinion that
the limited resources available should be used to enhance the quality of Hb A1c analysis, not change the
units for reporting. The units used for reporting Hb
A1c remains a highly contentious topic, and the hope
previously expressed “that the same Hb A1c values
should be reported globally” (14 ) will, unfortunately, not be realized.
Notwithstanding these issues, the report by Kilpatrick et al. (10 ) makes a valuable contribution. It
allays (at least in the short term) the anxieties of those
caring for patients with diabetes in countries that have
adopted SI units. For countries that plan to convert
from DCCT/NGSP units to SI units, it is vital that an
organized strategy with adequate education be adopted. In diabetes, patient self-management is critical
for improving treatment outcomes (15 ). Several influential organizations, including the American Diabetes
Association, the American College of Physicians, and
the NIH, encourage patients to “know your ABCs”
(where “A” represents Hb A1c). Therefore, any change
in reporting needs to be unambiguously communicated to patients. It is vital for countries that convert to
SI units to avoid jeopardizing the health of patients,
and the well-known adage primum non nocere (first do
no harm) should not be forgotten.
Author Contributions: All authors confirmed they have contributed to
the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design,
acquisition of data, or analysis and interpretation of data; (b) drafting
or revising the article for intellectual content; and (c) final approval of
the published article.
Authors’ Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form.
Disclosures and/or potential conflicts of interest:
Employment or Leadership: D.B. Sacks, Clinical Chemistry, AACC.
Consultant or Advisory Role: None declared.
Editorials
Stock Ownership: None declared.
Honoraria: None declared.
Research Funding: D.B. Sacks, Intramural Research Program of the
National Institutes of Health.
Expert Testimony: None declared.
Patents: None declared.
References
1. Allen DW, Schroeder WA, Balog J. Observations on the chromatographic
heterogeneity of normal adult and fetal human hemoglobin - a study of the
effects of crystallization and chromatography on the heterogeneity and
isoleucine content. J Am Chem Soc 1958;80:1628 –34.
2. Holmquist WR, Schroeder WA. A new N-terminal blocking group involving a
Schiff base in hemoglobin AIc. Biochemistry 1966;5:2489 –503.
3. Little RR, Wiedmeyer HM, England JD, Wilke AL, Rohlfing CL, Wians FH Jr,
et al. Interlaboratory standardization of measurements of glycohemoglobins.
Clin Chem 1992;38:2472– 8.
4. Diabetes Control and Complications Trial Research Group. The effect of
intensive treatment of diabetes on the development and progression of
long-term complications in insulin-dependent diabetes mellitus. The Diabetes
Control and Complications Trial Research Group. N Engl J Med 1993;329:
977– 86.
5. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose
control with sulphonylureas or insulin compared with conventional treatment
and risk of complications in patients with type 2 diabetes (UKPDS 33). UK
Prospective Diabetes Study (UKPDS) Group. Lancet 1998;352:837–53.
6. Little RR, Rohlfing CL, Sacks DB. Status of hemoglobin A1c measurement and
goals for improvement: from chaos to order for improving diabetes care. Clin
Chem 2011;57:205–14.
7. Hoelzel W, Weykamp C, Jeppsson JO, Miedema K, Barr JR, Goodall I, et al.
IFCC reference system for measurement of hemoglobin A1c in human blood
and the national standardization schemes in the United States, Japan, and
Sweden: a method-comparison study. Clin Chem 2004;50:166 –74.
8. Mosca A, Goodall I, Hoshino T, Jeppsson JO, John WG, Little RR, et al. Global
standardization of glycated hemoglobin measurement: the position of the
IFCC Working Group. Clin Chem Lab Med 2007;45:1077– 80.
9. Sacks DB. Measurement of hemoglobin A1c: a new twist on the path to
harmony. Diabetes Care 2012;35:2674 – 80.
10. Kilpatrick ES, Rigby AS, Atkin SL, Barth JH. Glycemic control in the 12 months
following a change to SI hemoglobin A1c reporting units. Clin Chem 2013;
59:1457– 60.
11. Sacks DB, Arnold M, Bakris GL, Bruns DE, Horvath AR, Kirkman MS, et al.
Guidelines and recommendations for laboratory analysis in the diagnosis and
management of diabetes mellitus. Clin Chem 2011;57:e1– 47.
12. Hanas R. Psychological impact of changing the scale of reported HbA1c
results affects metabolic control. Diabetes Care 2002;25:2110 –1.
13. Appleby J. Rises in healthcare spending: Where will it end? BMJ 2012;345:
e7127.
14. Sacks DB. Global harmonization of hemoglobin A1c. Clin Chem 2005;51:
681–3.
15. Berger M, Muhlhauser I. Diabetes care and patient-oriented outcomes. JAMA
1999;281:1676 – 8.
Clinical Chemistry 59:10 (2013) 1429