Hemoglobin A1c as screening for gestational diabetes mellitus in Nordic Caucasian women

Table 3 Sensitivities and specificities around 95, 97.5 and 100 % for diagnosing GDM with corresponding cut-offs for HbA1c and probabilities for GDM estimated by logistic regression models

Outcome	HbA1c alone^a					Model
Outcome	Cut-off, HbA1c ≥ (%)	Sensitivity (%)	Specificity (%)	Percentage of women avoiding OGTT	Percentage of women avoiding OGTT misclassified	Cut-off, probability of GDM ≥	Sensitivity (%)	Specificity (%)	Percentage of women avoiding OGTT	Percentage of women avoiding OGTT misclassified
GDM-WHO week 32–36	4.7	100	3.7	3.4	0	0.009	100	8.5	8.0	0
	4.8	97.3	8.0	7.7	2.0	0.011	97.3	11.2	10.7	1.5
	4.9	91.9	17.1	16.6	2.8	0.013	94.6	14.3	13.8	2.3
	5.5	29.7	95.0	6.4	73.2	0.16	29.7	95.0	6.4	73.2
	5.6	21.6	97.7	3.4	63.3	0.22	21.6	97.7	3.4	63.3
	5.9	0	100	0	0	0.50	0	100	0	0
GDM-WHO throughout pregnancy	4.4	100	0.5	0.5	0	0.018	100	2.9	2.7	0
	4.5	97.6	2.4	2.4	6.7	0.026	97.6	11.2	10.6	1.5
	4.6	95.2	7.7	7.5	4.3	0.027	95.2	11.8	11.3	2.8
	5.2	14.3	95.2	5.4	82.4	0.15	19.1	95.1	5.9	78.4
	5.3	9.5	97.8	2.7	76.4	0.18	16.7	97.4	3.5	68.6
	5.8	0	100	0	0	0.37	2.4	100	0.2	0
GDM-IADPSG week 18–22^b	4.7	100	16.6	16.2	0	0.010	100	16.0	15.7	0
	4.8	87.5	30.3	29.8	1.0	0.0153	93.8	36.0	35.3	0.4
	4.9	68.8	51.3	50.8	1.5	0.0154	87.5	36.5	35.9	0.8
	5.2	12.5	94.7	5.5	94.6	0.056	18.8	95.0	5.3	91.7
	5.3	6.3	97.4	2.7	94.5	0.067	0	97.4	2.5	100
	5.8	0	100	0	0	0.18	0	100	0	0
GDM-IADPSG week 32–36^b	4.6	100	1.0	1.0	0	0.003	100	2.0	1.9	0
	5.0	96.6	31.4	30.1	0.5	0.013	96.6	23.9	23.0	0.7
	5.1	82.8	48.2	46.7	1.7	0.018	93.1	36.0	34.6	0.9
	5.5	31.0	94.7	6.5	77.9	0.13	37.9	95.0	6.5	73.2
	5.6	20.7	97.5	3.3	71.4	0.18	24.1	97.5	3.5	68.2
	5.8	6.9	100	0.3	0	0.57	3.5	100	0.2	0
GDM-IADPSG throughout pregnancy^b	4.4	100	0.5	0.5	0	0.014	100	3.3	3.0	0
	4.6	97.8	2.4	2.4	6.6	0.027	97.8	16.6	15.6	1.0
	4.7	95.6	16.5	15.6	2.0	0.037	95.6	30.0	28.2	1.1
	5.2	13.3	95.2	5.4	82.3	0.18	26.7	95.0	3.2	75.2
	5.3	8.9	97.8	2.7	76.2	0.21	11.1	97.4	6.5	70.7
	5.8	0	100	0	0	0.52	0	100	0	0

See Table 2 for variables included in the various models. In addition, the percentage of women avoiding an OGTT by using cut-offs with high specificity to rule-in and high sensitivity to rule-out GDM and the percentage of those avoiding OGTT who are misclassified are shown
^aHbA1c at gestational week 18–22 was used for GDM-WHO throughout pregnancy, GDM-IADPSG at week 18–22 and GDM-IADPSG throughout pregnancy, while HbA1c at gestational week 32–36 was used for GDM-WHO at week 32–36 and GDM-IADPSG at week 32–36
^bModified IADPSG criteria were used, i.e. 1-h s-glucose was missing

ISSN: 1758-5996