Hematocrit correction: impact on neonatal screening (NS) for endocrinopathies

Silvia Marino; Florencia Tommasi; Alejandro VilcheJuarez; Alvarez Olmedo Maria Belen; Christian Arias; Gerard Zeida; Gustavo Dratler

Hematocrit was considered an uncontrollable variable that affects results of biomarkers in whole blood filter paper samples (FPS). NS methodologies are generally adjusted by assuming a standard hematocrit of 55%. Our group has previously presented a method to estimate hematocrit by measuring hemoglobin values in FPS. This correction has the potential to reduce the number of false positives, improve specificity in NS, and decrease the number of recalled patients.

Objective: To verify and quantify the impact of hematocrit correction in NS of endocrinopathies (Congenital Hypothyroidism and Congenital Adrenal Hyperplasia).

Materials and methods: The methodology was implemented in a NS program (February to September 2022), measuring TSH and 17OHP, in samples of three groups of patients: GA) National NS Program (n:9366): processed by ELISA (Zentech) and reconfirmed by DELFIA (PerkinElmer). GB) Buenos Aires NS Program (n:3185): processed and reconfirmed by DELFIA. GC) Neonatal intensive care unit (NICU) (n:207). Processed and reconfirmed by DELFIA.

Patients: Who exceeded cut-off values were recalled for a confirmatory serum sample. The recall rate with and without correction was evaluated, as well as the percentage of samples that avoid to be recalled. Linear regression was performed between gestational age and hematocrit (n: 443).

Results:


TSH	n	Positives	Recall (%)	Post Correction Positives	Post Correction Recall (%)	Avoid Recall (%)	True Positives
GA	9366	116	1,24	77	0,82	33,62	5
GB	3185	50	1,57	33	1,04	34,00	3
GC	207	3	1,45	2	0,97	33,33	0
17OHP
GA	9366	47	0,50	30	0,32	36,17	0
GB	3185	15	0,47	8	0,25	46,67	0
GC	207	2	0,97	2	0,97	0,00	0

Those groups with confirmed CH cases (n:8) were analysed.


Group	TSH	Sensitivity (%)	Specificity (%)	False Positives (%)	Positive Predictive Value (%)
GA	No correction	100	98.81	1.19	4.31
GA	Corrected	100	99.23	0.77	6.49
GB	No correction	100	98.52	1.48	6.00
GB	Corrected	100	99.53	0.47	16.67

Linear Regression Hematocrit(y) vs. Gestational Age(x): y=-0.5346+1.3052x

Conclusions: Hematocrit correction of NS biomarkers significantly decrease the recall rate, reducing the negative impact on families due to unnecessary alerts and the need of confirmatory tests. It improves specificity, reduces false positive rate and increases the positive predictive value of the tests. The positive gestational age-hematocrit correlation implies that this correction is especially useful in NS of preterm babies.

IMPE Abstracts

P67