Increased Mortality Rate in Infants with Abnormal T-Cell Receptor Excision Circles

doi:10.21203/rs.3.rs-2085456/v1

Purpose: T-Cell Receptor Excision Circles based newborn screening (TREC-NBS) allows for early detection and improved disease outcome in infants with primary immunodeficiency disorders (PIDD). The utility of abnormal TREC-NBS in infants without PIDD is not well studied. To determine the value of abnormal TREC-NBS in predicting mortality and morbidity in infants without PIDD.

Methods: 365,207 newborn screens in the database from October 2011 to December 2014 were reviewed. 467 newborns had an abnormal TREC-NBS either during the initial or repeat screen. 1,390 newborns with normal TREC-NBS served as matched controls based on gestational age, birth weight, neonatal intensive care unit status (NICU), and race. Any infant with a PIDD diagnosis was excluded. Demographic data was obtained through NBS, birth certificates, and death certificates records from Michigan Department of Health and Human Services (MDHHS) databases.

Results: Overall, infants with positive TREC-NBS had higher mortality (14.6% versus 3.3% in controls) even when PIDD is ruled-out. Transient positive TREC-NBS was not associated with higher mortality, but unresolved or late TREC-NBS was associated with higher mortality. More Black infants had positive TREC-NBS than infants of other races. Children with positive TREC-NBS were more likely to have NICU stays and have low Apgar scores. Infants with unresolved or late TREC-NBS were more likely to have congenital anomalies.

Conclusion: Infants with positive TREC-NBS may be at a higher risk of morbidity and mortality and should be carefully followed, especially if discharged home before a repeat screen can be completed.

Newborn screen

TREC

Lymphopenia

Primary Immunodeficiency

Mortality

Congenital

In 1963, newborn screening (NBS) was first established to screen for phenylketonuria with dried blood samples through the guidance of Robert Guthrie^1–3. Over time, NBS expanded through various DNA and mass spectrometry analysis to allow medical providers to identify and expedite management in newborns with various metabolic and genetic disorders through heel-stick blood sample testing. Screening for severe combined immunodeficiency (SCID) using T-cell receptor excision circles (TREC) was proposed in 2005 and first implemented in the state of Wisconsin. In Michigan, the TREC screening assay for SCID was added to the NBS program in 2011^1–4. T cell lymphopenia is one of the hallmarks of SCID and can be identified by measuring TREC, a biomarker of naïve T cells. The use of T-Cell Receptor Excision Circles based newborn screening (TREC-NBS) was created to provide cost-effective and accurate screening of SCID. This method detects T cell lymphopenia in the peripheral blood of affected newborns through evaluation of excised DNA fragments that are joined together to form circular DNA byproducts which are known as TREC^3,5. Since December 2018, TREC analysis has been implemented in all 50 of the United States to detect SCID, with a sensitivity of 100% and specificity of 98.8%^2,3.

Prior to NBS for SCID, the vast majority of affected infants experienced significant clinical deterioration and death within the first year of life due to severe infection and failure to thrive^1–3. The severe impairment or lack of production of T cells and B cells causes SCID patients to have recurrent and severe infections. SCID NBS allows for early diagnosis, which leads to early intervention, ultimately improving disease outcome. The utility of TREC- NBS to detect primary immunodeficiency disorders with severe T cell lymphopenia has been well documented.

Our group is interested in the value of TREC-NBS in predicting infant mortality and morbidity in infants without a primary immunodeficiency diagnosis. To our knowledge, data on using TREC- NBS evaluations to predict infant morbidity and mortality is not well characterized. A single-site, retrospective review of 50 patients identified low TREC values as being associated with non-immunologic conditions, such as congenital cardiac and renal abnormalities, and increased infant mortality⁷. The authors concluded that infants identified with low TREC levels could benefit from specialized programs that included close follow up and education for signs/symptoms of immunologic and non-immunologic abnormalities. This study supported the importance of TREC screening and its effect on future management of patients who are found to have low TREC values. In this study, we report the statewide infant mortality rate and cause of death in newborns with at least one low TREC-NBS screening and their associated factors including maternal, pregnancy, and delivery conditions.

Study population

The study population was obtained through the review of 365,207 newborn screens in the Michigan Department of Health and Human Services (MDHHS) database from October 2011 to December 2014. From the screens, 467 newborns had positive TREC-NBS either during the initial or repeat screen without the diagnosis of a primary immunodeficiency disorders (PIDD). These newborns were matched to 1,390 controls that had a negative TREC-NBS (see next section for matching criteria). Newborns with PIDD, such as SCID and DiGeorge Syndrome, were excluded, as we were interested in examining newborns with unexplained T cell lymphopenia. The final study set encompassed all neonates who had a positive TREC-NBS result without PIDD. Michigan Department of Health and Human Services Institutional review board approved the study, number: 201511-03-XA

Data Collection, Data and Statistical Analysis

Demographic data was obtained from Michigan NBS records, Michigan birth certificates records and Michigan death certificate records. NBS data which included gestational age, birth weight, neonatal intensive care unit (NICU) status, and race was used to match the cases. Newborns with at least one low TREC screening (positive TREC-NBS cases) were matched to newborn with only normal TREC screening (negative TREC-NBS controls) in a ratio of 1:3 (abnormal TREC: normal TREC). Clinical characteristics collected through birth certificates included maternal age, mother’s education status, prenatal care status, insurance payer during delivery period, Apgar scoring, tobacco use, alcohol use, mother’s ancestry, birth place, pre-pregnancy diabetes, chronic hypertension, gestational hypertension, previous poor pregnancy outcome, premature rupture of membranes, steroid use during delivery, labor induction, congenital anomalies, surfactant use, meconium exposure, and delivery method. Furthermore, newborn cause of death and death date were obtained from death certificates records for additional review.

Newborns were divided into three groups based on initial and repeat TREC-NBS results: Group #1 (transient positive cases) consisted of newborns that initially had a positive TREC-NBS and a subsequent normal repeat screen, group #2 (unresolved positive cases) comprised of newborns who initially had a positive TREC-based NBS and did not have subsequent repeat TREC- NBS testing due to infant death or unknown reasons, and group #3 (subsequent positive) included newborns who initially had a negative TREC- NBS and had a subsequent positive TREC- NBS screening result. The reason for repeat TREC- NBS in this group is either due to prolonged NICU stay or because a repeat screen was requested due to prior positive NBS screen for a different disorder.

Statistical Analyses

SPSS Version 26.0 IBM, Inc., SAS 9.4 and MedCalc were used to perform all statistical procedures. Demographic data was reported using ratios and proportions for categorically scaled variables and means/standard deviations for continuously scaled variables. Crosstab procedures were performed throughout within each group between cases and controls. Proportional differences were compared using a non-parametric Fisher’s Exact test in 2 X 2 group comparisons and table reported. Lastly, mortality rates were also calculated for each of the three groups and compared between cases and controls.

Baseline Characteristics

A total of 365,207 neonates were born between October 2011 and December 2014 and had a newborn screen collected. Overall, the majority of this NBS population was born ≥ 37 weeks gestational age (n = 319,996; 87.6%), with birth weight ≥ 2500g (n = 328,649; 90.0%) and did not require neonatal intensive care unit/special care nursery (NICU/SCN) admission after birth (n = 308,272; 84.4%). The majority of the population was White infants (n = 231,642; 63.4%), followed by Black infants (n = 65,501; 17.9%).

A total of 467 newborns with positive TREC- NBS results, but without PIDD, were included. They were further categorized into three groups based on their final TREC- NBS result: Group #1 – Transient positive cases (n = 375; 80.3%), Group #2 – Unresolved positive cases (n = 36; 7.7%), and Group #3 – Subsequent positive cases (n = 56; 12%); See Table 1.

Table 1

Newborn Screen Demographic Data
NBS Demographics	Study Groups						NBS Population Oct 2011-Dec 2014
NBS Demographics	Group 1		Group 2		Group 3		NBS Population Oct 2011-Dec 2014
	N	%	N	%	N	%	N	%
Gestational Age
< 32 wks.	116	30.9	24	67	24	42.9	5,175	1.4
32–36 wks.	82	21.9	8	22	9	16.1	28,173	7.7
≥ 37 wks.	164	43.7	4	11	17	30.4	319,996	87.6
Missing	13	3.5	0	0	6	10.7	11,863	3.2
Birth Weight
< 1500 g	121	32.3	21	58	26	46.4	4,765	1.3
1500–2499 g	76	20.3	6	17	12	21.4	24,087	6.6
≥ 2500 g	176	46.9	9	25	17	30.4	328,649	90.0
Missing	2	0.5	0	0	1	1.8	7,706	2.1
NICU/SCN Status
No NICU/SCN	127	33.9	1	3	5	8.9	308,272	84.4
NICU/SCN	242	64.5	35	97	51	91.1	40,076	11.0
Missing	6	1.6	0	0	0	0.0	16,859	4.6
Race
White	186	49.6	16	44	29	51.8	231,642	63.4
Black	127	33.9	16	44	16	28.6	65,501	17.9
Other	33	8.8	1	3	7	12.5	41,063	11.2
Missing	29	7.7	3	8	4	7.1	27,001	7.4
Total	375		36		56		365,207
Newborn screen demographic data was categorized by gestational age, birth weight, NICU/SCN Status, and race for all three groups with positive for TREC-NBS in relation to the total newborn screening population. Group 1: Transient positive cases; Group 2: unresolved positive cases; Group 3: subsequent positive

In group #1, the highest percentage of the neonates were born ≥ 37 weeks gestational age (n = 164, 48.7%), birth weight ≥ 2500g (n = 176; 46.9%), and required NICU/SCN admission following birth (n = 242; 64.5%). In group #2, the majority of the neonates were born < 32 weeks gestational age (n = 24; 67%), birth weight < 1500g (n = 21; 58%), and required NICU/SCN admission following birth (n = 35; 97%). In group #3, the highest percentage of the neonates were born < 32 weeks gestational age (n = 24, 42.9%), birth weight < 1500g (n = 26; 46.4) and required NICU/SCN admission (n = 51; 91.1%). The highest percentage of the patients in both the group #1 and group #3 were White infants (n = 186; 49.6% and n = 29; 51.8%, respectively). In group #2, infants were evenly split between White infants (n = 16; 44%) and Black infants (n = 16; 44%). Interestingly, Black infants represent 34% of the positive TREC- NBS population, without PIDD (n = 159/467), but only 18% of the overall NBS population (n = 65,501/365,207).

Apgar score and congenital anomalies

Birth Apgar score and congenital anomalies were examined and compared to matched controls. Low Apgar score were defined by our study as a score of 0–3. The majority of infants with a positive TREC- NBS had Apgar score of at least 4 or more, similar to matched controls. See Table 2. However, there were a statistically higher percentage of infants with Apgar score of 3 or less in all groups with positive TREC- NBS when compared to matched controls (p < 0.001). Group #2 had the highest percentage of infants with low Apgar score (n = 13; 36.1%) while group #1 had the lowest percentage of infants with low Apgar score (n = 32; 8.5%). There were a higher percentage of infants with congenital anomalies in the group #2 (n = 10; 27.8%, p = 0.002) and group #3 (n = 16; 28.6%, p < 0.001) as compared to their matched controls. There was no significant difference in the percentage of congenital anomalies between infant cases and controls in group #1 (p = 0.10).

Table 2

Infant Characteristics
	Apgar Score	Cases		Controls		P-value	Congenital Anomalies	Cases		Controls		P-value
		N	%	N	%			N	%	N	%
Group 1	0–3	32	8.5	34	3	p ≤ 0.001	Yes	25	6.7	47	4.2	p = 0.10
	4–10	343	91.5	1084	97	p ≤ 0.001	No	342	91.2	1036	92.7	p = 0.10
Group 2	0–3	13	36.1	11	10.2	p ≤ 0.001	Yes	10	27.8	7	6.5	p = 0.002
	4–10	23	63.9	97	89.8	p ≤ 0.001	No	26	72.2	98	90.7	p = 0.002
Group 3	0–3	12	21.4	10	6.1	p ≤ 0.001	Yes	16	28.6	12	7.3	P < 0.001
	4–10	44	78.6	154	93.9	p ≤ 0.001	No	38	67.9	151	92.1	P < 0.001
Apgar score and congenital anomalies for positive TREC-NBS cases and their matched controls. Controls matched for gestational age, birth weight, NICU/SCN status, and race. Group 1: Transient positive cases; Group 2: unresolved positive cases; Group 3: subsequent positive

Maternal characteristics

Maternal characteristics including maternal age, maternal education, prenatal care, maternal insurance payment type, tobacco use, alcohol use and mother’s ancestry were examined and compared to matched controls. See Table 3. In both the case and control groups, the majority of mothers are of non-Hispanic race, less than 35 years of age, had more than a high school education, received adequate prenatal care, and did not report use of tobacco or alcohol. Interestingly, the majority of mothers of infants in the group #1 (n = 205/375, 54.7%, p = 0.004) and group #3 (n = 34/56, 60.7%, p = 0.10) had Medicaid insurance, whereas the majority of matched control mothers did not have Medicaid. The group #2 was split 50% between mothers with Medicaid and non-Medicaid. In all three groups there was no statistically significant difference between maternal age, maternal education, prenatal care, tobacco use, alcohol use, or maternal race between cases and matched controls.

Table 3

Maternal Characteristics
		Cases N %		Controls N %		P-value
Maternal Age
Group 1	< 35	324	86.4	961	86	0.08
	≥ 35	51	13.6	157	14	0.08
Group 2	< 35	27	75	89	82.4	0.60
	≥ 35	9	25	19	17.6	0.60
Group 3	< 35	50	89.3	132	80.4	0.20
	≥ 35	6	10.7	32	19.5	0.20
Mother's Education
Group 1	<High School	73	19.5	193	17.3	0.70
	>High School	302	80.5	925	82.7	0.70
Group 2	<High School	6	2.8	21	19.4	0.80
	>High School	30	69.4	87	80.5	0.80
Group 3	<High School	11	19.6	22	13.4	0.20
	>High School	45	80.4	142	86.5	0.20
Prenatal Care
Group 1	Inadequate	49	13.1	161	14.4	0.40
	Adequate	326	86.9	957	85.6	0.40
Group 2	Inadequate	3	8.3	15	13.9	0.70
	Adequate	33	91.7	93	86.1	0.70
Group 3	Inadequate	14	25	29	17.7	0.20
	Adequate	42	74.9	135	82.3	0.20
Insurance Payment
Group 1	Medicaid	205	54.7	518	46.3	0.004
	Other	170	45.3	600	53.6	0.004
Group 2	Medicaid	18	50	48	44.4	0.70
	Other	18	50	60	55.6	0.70
Group 3	Medicaid	34	60.7	69	42.1	0.10
	Other	22	39.3	95	57.9	0.10
Tobacco Use
Group 1	Yes	106	28.3	264	23.6	0.20
	No	267	71.2	847	75.8	0.20
Group 2	Yes	11	30.6	23	21.3	0
	No	24	66.7	85	78.7	0
Group 3	Yes	10	17.9	38	23.2	0.60
	No	45	80.4	124	75.6	0.60
Alcohol Use
Group 1	Yes	1	0.3	7	0.6	0.40
	No	373	99.5	1103	98.7	0.40
Group 2	Yes	0	0	0	0	N/A
	No	36	1	107	99.1	N/A
Group 3	Yes	1	1.8	3	1.8	1.00
	No	55	98.2	160	97.6	1.00
Mother’s Ancestry
Group 1	Hispanic	20	5.3	77	6.9	0.50
	Non-Hispanic	354	94.4	1039	92.9	0.50
Group 2	Hispanic	2	5.6	6	5.6	0.80
	Non-Hispanic	34	94.4	101	93.5	0.80
Group 3	Hispanic	4	7.1	7	4.3	0.50
	Non-Hispanic	52	92.9	157	95.7	0.50
Maternal characteristics were compared between TREC-NBC positive infants and their matched controls. Controls matched for gestational age, birth weight, NICU/SCN status, and race. Group 1: Transient positive cases; Group 2: unresolved positive cases; Group 3: subsequent positive

Pregnancy factors

Pregnancy factors examined included pre-pregnancy diabetes, gestational hypertension, history of previous poor pregnancy outcome, premature rupture of membranes, and steroid use during pregnancy and compared to matched controls. See Table 4. For both cases and controls, the majority of pregnancies were not complicated by maternal gestational hypertension, premature rupture of membranes, pre-pregnancy diabetes, requirement for steroid or surfactant use during pregnancy, and the mothers did not have history of poor outcome in a prior pregnancy. Interestingly, group #1 infants had significantly higher percentage of mothers with gestational hypertension (n = 26, 6.9%) as compared to their matched controls (n = 34, 3%; p = 0.01). This finding was not observed in the group #2 and group #3 infants, and there were no statistically significant differences between the cases and controls for any other pregnancy factors.

Table 4

Maternal Pregnancy and Perinatal Factors
		Cases N %		Controls N %		P-value
Pre-Pregnancy Diabetes
Group 1	Yes	11	2.9	17	1.5	0.10
	No	363	96.8	1093	97.8	0.10
Group 2	Yes	3	8.3	2	1.9	0.20
	No	33	91.7	105	97.2	0.20
Group 3	Yes	0	0	4	2.4	0.60
	No	56	100	159	97	0.60
Gestational Hypertension
Group 1	Yes	26	6.9	34	3	0.01
	No	348	92.8	1076	96.2	0.01
Group 2	Yes	4	11.1	12	11.1	0.80
	No	32	88.9	95	88	0.80
Group 3	Yes	5	8.9	16	9.8	1
	No	51	91.1	147	89.6	1
Previous Poor Pregnancy Outcome
Group 1	Yes	12	3.2	33	3	0.60
	No	362	96.5	1077	96.3	0.60
Group 2	Yes	1	2.8	2	1.9	0.80
	No	35	97.2	105	97.2	0.80
Group 3	Yes	2	3.6	8	4.9	1
	No	54	96.4	155	94.5	1
Premature Rupture
Group 1	Yes	37	9.9	110	9.8	0.80
	No	334	89.1	992	88.7	0.80
Group 2	Yes	37	9.9	110	9.8	0.70
	No	334	89.1	992	88.7	0.70
Group 3	Yes	5	8.9	32	19.5	1
	No	51	91.1	131	79.9	1
Steroid Use
Group 1	Yes	46	12.3	156	14	0.50
	No	329	87.7	960	85.9	0.50
Group 2	Yes	7	19.4	18	16.7	0.70
	No	29	80.6	90	83.3	0.70
Group 3	Yes	6	10.7	32	19.5	0.10
	No	50	89.3	132	80.5	0.10
Surfactant Use
Group 1	Yes	19	5.1	46	4.1	0.30
	No	356	94.9	1066	95.4	0.30
Group 2	Yes	19	5.1	46	4.1	1
	No	356	94.9	1066	95.4	1
Group 3	Yes	9	16.1	14	8.5	1
	No	46	82.1	149	90.9	1
Maternal pregnancy and perinatal factors were compared between TREC-NBC positive infants and their matched controls. Controls matched for gestational age, birth weight, NICU/SCN status, and race. Group 1: Transient positive cases; Group 2: unresolved positive cases; Group 3: subsequent positive

Delivery factors

Delivery factors examined included induction of labor, meconium present during delivery, delivery place, and delivery method, and compared to matched controls. See Table 5. Most infants were delivered in a hospital, did not require labor induction, and the delivery was not complicated by meconium in both cases and matched controls. In all three TREC-NBS positive groups, infants were mostly born via Cesarean section. The group #1 had significantly higher percentage of Cesarean section (n = 211, 56.3%) as compared to their matched controls (n = 557, 49.8%; p = 0.03). Group #3 also had significantly higher percentage of Cesarean section (n = 42, 75%) vs. their matched controls (n = 97, 59.2%; p = 0.03).

Table 5

Delivery Factors
		Cases N %		Controls N %		P-value
Labor Induction
Group 1	Yes	68	18.1	169	15.1	0.30
	No	307	81.9	947	84.7	0.30
Group 2	Yes	0	0	7	6.5	0.10
	No	36	100	101	93.5	0.10
Group 3	Yes	8	14.3	16	9.8	0.30
	No	48	85.7	148	90.2	0.30
Birth Place
Group 1	Hospital	371	98.9	1114	99.6	0.10
	Non-Hospital	4	1.1	4	0.4	0.10
Group 2	Hospital	36	100	107	99.1	0.60
	Non-Hospital	0	0	1	0.9	0.60
Group 3	Hospital	56	100	163	99.4	1
	Non-Hospital	0	0	1	0.6	1
Delivery Method
Group 1	Cesarean	211	56.3	557	49.8	0.03
	Vaginal	164	43.7	561	50.2	0.03
Group 2	Cesarean	27	75	65	60.2	0.10
	Vaginal	9	25	43	39.8	0.10
Group 3	Cesarean	42	75	97	59.2	0.03
	Vaginal	14	25	67	40.9	0.03
Meconium
Group 1	Yes	27	7.2	46	4.1	0.04
	No	348	92.8	1070	95.7	0.04
Group 2	Yes	1	2.8	2	1.9	0.70
	No	35	97.2	106	98.2	0.70
Group 3	Yes	3	5.4	8	4.9	0.90
	No	53	94.6	156	95.1	0.90
Delivery factors were compared between TREC-NBC positive infants and their matched controls. Controls matched for gestational age, birth weight, NICU/SCN status, and race. Group 1: Transient positive cases; Group 2: unresolved positive cases; Group 3: subsequent positive

The group #1 was noted to have a higher percentage of meconium present during the delivery (n = 27, 7.2%) as compared to their matched controls (n = 46, 4.1%; p = 0.04). Group #2 and group #3 also had higher percentage of meconium present during the delivery; however these were not statistically significant. See Table 5.

Mortality rates

Newborns with positive TREC- NBS results (n = 467) and matched controls (n = 1390) were compared on mortality rates (Table 6). Overall, there were a total of 68 TREC-NBS positive infants who died during this review period (n = 68/467; 14.6%) and a total of 46 matched controls who died (n = 46/1390; 3.3%). For group #1, there was no statistically significant difference in the mortality rates between cases and matched controls (2.4%, 2.4% respectively; p = 0.1). There was a significant difference in the mortality rates between cases and matched controls in both group #2(91.7%, 8.3%, respectively; p < 0.001) and group #3 (26%, 10%, respectively; p < 0.001) where low TREC was non-transient.

Table 6

Mortality Rates in Infants with Positive TREC-NBS Results
Mortality Rates	Total	Died		P-Value
Mortality Rates	N	N	%
Group #1
Cases	375	9	2.4	0.10
Controls	1118	27	2.4
Group #2
Cases	36	33	91.7	< 0.001
Controls	108	9	8.3
Group #3
Cases	56	26	46.4	< 0.001
Controls	164	10	6.1
Mortality rates were calculated for all three study groups who screened positive on TREC-NBS at least once in relation to the total population in which the study groups were obtained. Group 1: Transient positive cases; Group 2: unresolved positive cases; Group 3: subsequent positive

Term status

To assess the impact of term status on mortality rates, the cases and their matched controls were re-categorized into term (37 weeks gestation and greater) or preterm (less than 37 weeks gestation). Most cases of infant mortality in all three TREC-NBS positive groups were of preterm infants. See Tables 7a and 7b. In group #1, term infant mortality (n = 1, 0.6%) and preterm infant mortality (n = 8, 4%) was not statistically different when compared to the infant mortality in their matched term infant controls (n = 1, 0.2%, p = 0.4) and preterm infant controls (n = 25, 4.2%, p = 0.9). The term infant mortality for group #2 was 75% (n = 3/4) versus 0% (n = 0/12) for matched controls (p < 0.001). Preterm infant mortality for group #2 was 93.8% (n = 30/32) versus 9.4% (n = 9/96) for matched controls (p < 0.001). The term infant mortality for group #3was 53.9% (n = 9/17) versus 2% (n = 1/51) for matched controls (p < 0.001). Preterm infant mortality for group #3 was 39.4% (n = 13/33) versus 7.2% (n = 7/97) for matched controls (p ≤ 0.001).

Table 7

a: Mortality Data for Term Infants
Mortality Data for Term Infants (≥ 37 wks.)
	Total	Died		P-value
	N	N	%
Group 1
Cases	164	1	0.6	0.40
Controls	492	1	0.2
Group 2
Cases	4	3	75.0	< 0.001
Controls	12	0	0.0
Group 3
Cases	17	9	52.9	< 0.001
Controls	51	1	2.0
Mortality rates in term infants (gestational age greater than or equal to 37 weeks) for all three study groups who screened positive on TREC-NBS at least once in relation to the total population in which the study groups were obtained. Group 1: Transient positive cases; Group 2: unresolved positive cases; Group 3: subsequent positive

Table 7

b: Mortality Data for Preterm Infants
Mortality Data for Preterm Infants (< 37 wks.)
	Total	Died		P-value
	N	N	%
Group 1
Cases	198	8	4.0	0.90
Controls	591	25	4.2
Group 2
Cases	32	30	93.8	< 0.001
Controls	96	9	9.4
Group 3
Cases	33	13	39.4	< 0.001
Controls	97	7	7.2
Mortality rates in preterm infants (gestational age less than 37 weeks) for all three study groups who screened positive on TREC-NBS at least once in relation to the total population in which the study groups were obtained. Group 1: Transient positive cases; Group 2: unresolved positive cases; Group 3: subsequent positive

TREC-NBS is known to have great efficacy in sensitivity and specificity when screening newborns for SCID. A number of studies have shown that utilizing TREC-NBS has increased the number of positive screening results for SCID, as well as other forms of T cell lymphopenia, in newborns across the United States^1–3,8−11. There are limited studies evaluating the association of positive TREC-NBS and infant mortality. Through birth and death certificates, and NBS records, we retrospectively evaluated whether one or more abnormal TREC-NBS result would predict higher mortality rates despite the absence of an identifiable underlying etiology.

All of the neonates in our study with positive TREC-NBS were more likely to require NICU/SCN admission after birth. Among the infants with positive TREC-NBS, transient positive group infants were similar in gestational age and birth weight to the general NBS population. The unresolved positive TREC- NBS infants and the subsequent positive infants were more likely to have severe prematurity (< 32 weeks) and/or very low birth weight (< 1500g). Over 90% of infants in the unresolved positive cases group and the subsequent positive cases group required NICU/SCN admission. This finding was not surprising as we expected infants with stress related factors such as severe prematurity, low birth weight and ICU status to trigger T cell lymphopenia and newborn screening. Interestingly, this was not observed in the transient positive group infants.

The incidence of positive TREC-NBS was disproportionally higher in Black newborns. Black neonates represented 18% of the total NBS population but 34% of the total positive TREC-NBS population. Perhaps, Black newborns were more likely to be preterm, have low birth weight or require NICU admission following birth. It would be interesting to further investigate this finding by examining perinatal factors that could have impacted positive TREC-NBS for Black neonates.

Infants with positive TREC- NBS in all three groups were more likely to have lower Apgar scores (3 or less). Apgar score is known as a standardized marker for the health status of newborns upon maternal delivery and represents stress perinatally which can lead to T cell lymphopenia.

Infants in the groups with unresolved positive TREC-NBS and repeat positive TREC-NBS were more likely have congenital anomalies compared to their matched controls and when compared to the infants with transiently positive TREC screen. In this study we did not look at the specific anomalies that were found in these infants, but in the future, this may give us more insight about its relationship with positive TREC- NBS results. Overall, infants in the unresolved positive cases group and the subsequent positive cases group were much more ill at birth with higher percentage of severe prematurity, lower birth weight, lower Apgar scores, and higher percentage of congenital anomalies.

Interestingly, unresolved positive cases group TREC-NBS infants did have a significantly higher percentage of mothers with Medicaid payment status, gestational hypertension during pregnancy, birth via cesarean delivery, and delivery complicated by meconium when compared to their matched controls. The subsequent positive TREC-NBS infants also had significantly higher percentage of cesarean delivery but none of the other factors seen in the unresolved positive cases group were statistically different from the subsequent positive cases group paired controls. While with the unresolved positive cases group, there was no significant difference with paired controls for any of these factors. It is interesting that each of these maternal, pregnancy and delivery factors could cause newborn stress and T cell lymphopenia, but we found only a few factors that were actually associated with TREC- NBS positive results and this association was almost only found in unresolved positive cases group, for unknown reasons. Increasing the study population size for these groups may allow for the better identification of these risk factors.

In regard to mortality, our study demonstrated that infants with positive TREC-NBS carry a higher percentage of mortality even after adjusting for gestational age, birth weight, NICU status and race. In the unresolved positive cases group and the subsequent positive cases group, those who did not have transient T-cell lymphopenia, had significantly higher mortality rate compared to their matched controls. A positive TREC result with repeat negative result (group #1; transient cases) was not correlated with higher mortality and these cases had comparable mortality percentage as matched controls. This finding showed that transiently positive TREC-NBS cases, on its own, did not seem to be a risk factor for infant mortality. Although the infants in this group did not display a higher mortality rate than control, it is difficult to ascertain at the time of the first positive TREC-NBS result which infants will fall into this lower risk, transiently positive screening group. The higher mortality rate seen in the unresolved positive cases group was expected as the majority of these patient died before a repeat screen could be completed. The positive initial screen for this group may have been due to significant neonatal stress.

There was not an observable pattern in the cause of death for the transient positive TREC- NBS infants. See Tables S1-S3 in the online resources for the listed causes of death in each group. Interestingly, almost 19% of negative TREC-NBS matched control infants in the transient positive cases group (n = 5/27) died from sepsis but none in the TREC- NBS positive infants in the transient positive cases group. We expected more incidence of sepsis in TREC- NBS positive infants as they likely had T-cell lymphopenia and would be more susceptible to infections. Most deaths in the infants with positive TREC- NBS in the unresolved positive cases group were due to “hydrops fetalis not disease” (n = 7/33, 21%). None of the matched controls for the unresolved positive cases group had “hydrops fetalis not disease” as a cause of death. This finding may have contributed to the early demise of the unresolved positive cases group infants. Overall, the cause of death for many of the unresolved positive cases group and the subsequent positive cases group infants included significant cardiopulmonary issues and infections supporting previous reports that neonatal stress may trigger a positive TREC screen. Given the limited number of mortality cases in this study, no concrete conclusions can be drawn to compare the cause of death between TREC-NBS positive infants and matched controls. Perhaps pooling nation-wide data can help us find better trends.

A major limitation to our study is the type of data available from NBS records, and birth and death certificates. Additional factors such as infants’ hospital course, comorbid conditions, steroid doses and timing prior to delivery to assess for possible underlying etiology of T-cell lymphopenia could not be investigated.

Using a balance of demographic data, diagnoses upon expiration, as well as calculated mortality rates, and our data demonstrates that an abnormal TREC level on NBS, without a confirmed negative repeat screen, was associated with higher mortality. Similar to prematurity, low birth weight, and NICU status, a positive TREC- NBS should be considered a risk factor for mortality in newborns. Those newborns with a positive TREC- NBS should be monitored closely, especially if they are discharged home before a repeat TREC-NBS test can be repeated. We suggest close monitoring for patients sent home on a weekly basis during the first 2 months of life as the majority of the infant deaths occurred within 2 months. Those infants who receive a normal repeat TREC-NBS can then be monitored similarly to a TREC-NBS negative infant, as their risk of mortality is lower. Furthermore, this data gives us further insight into possible risk factors for true transient positive TREC- NBS cases that are unrelated to PIDD including gestational hypertension, maternal Medicaid status, C-section delivery, and delivery complicated by meconium. While our study showed that infants with positive TREC- NBS and with the factors HTN, Medicaid status, C-section, or delivery meconium were more likely to have transient TREC-based NBS, whether these factors can help us better predict mortality risk or be used as a guide for infant monitoring should be further explored. Although this data is retrospective and represents a single-state experience, it characterized a large positive TREC- NBS population and assessed the association of positive screens with infant mortality. Further studies will be needed to determine if the TREC assay can serve as a predictor for mortality in newborns with an abnormal screen.

ICU: Intensive Care Unit

MDHHS: Michigan Department of Health and Human Services

NICU: Neonatal Intensive Care Unit

PIDD: Primary Immunodeficiency Disorders

SCN- Special Care Nursery

SCID- Severe Combined Immunodeficiency

TREC- T-Cell Receptor Excision Circles

TREC-NBS: T-Cell Receptor Excision Circles based newborn screening

ACKNOWLEDGMENTS

We would like to thank the Michigan Primary Immunodeficiency Disorders Quality Improvement Committee (PIDQIC) members, especially Dr. Ulrich Duffner (Michigan State University/Helen DeVos Children's Hospital), Dr. Sureyya Savasan (Children’s Hospital of Michigan/Central Michigan University College of Medicine), Dr. Nicholas Hartog (Helen DeVos Children's), Mary Fredrickson (Children’s Hospital of Michigan) and Mary Kleyn (Michigan Department of Community Health) for their support of this project, insightful comments and review of the manuscript.

Funding Source:The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Conflict of Interest: The authors have indicated they have no potential conflicts of interest to disclose.

Data availability: The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Author Contributions: All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Jenny Huang, Ashwin Shankar, Isabel Hurden, Ronald Thomas and Pavadee Poowuttikul. The first draft of the manuscript was written by Jenny Huang and Ashwin Shankar, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Consent to participate: Not applicable

Consent for publication: Not applicable

Michigan Department of Health and Human Services Institutional review board approved the study, number: 201511-03-XA

Chan K, Puck JM. Development of population-based newborn screening for severe combined immunodeficiency. J Allergy Clin Immunol. 2005;115(2):391–8.
Kelly BT, Tam JS, Verbsky JW, Routes JM. Screening for severe combined immunodeficiency in neonates. Clin Epidemiol. 2013;5(1):363–9.
Van der Spek J, Groenwold RHH, Van der Burg M, Van Montfrans JM. TREC Based Newborn Screening for Severe Combined Immunodeficiency Disease: A Systematic Review. J Clin Immunol. 2015;35(4):416–30.
Puck JM. Neonatal screening for severe combined immunodeficiency. Curr Opin Pediatr. 2011;23(6):667–73.
Puck JM. Laboratory Technology for Population-based Screening for SCID in Neonates: The Winner Is T-cell Receptor Excision Circles (TRECs). J Allergy Clin Immunol. 2013;129(3):607–16.
Adams SP, Kricke S, Ralph E, Gilmour N, Gilmour KC. A comparison of TRECs and flow cytometry for naive T cell quantification. Clin Exp Immunol. 2018;191(2):198–202.
Nevid M, Richmond GW, Davies S, Patel A, Hackett A, Mahdavinia M. Non-immunologic Conditions Associated with Low TREC Values. J Allergy Clin Immunol. 2020;145(2):AB214.
Cossu F.. Ital. Genetics of SCID. J Pediatr. 2010;36(1):76.
Puck JM. Laboratory technology for population-based screening for severe combined immunodeficiency in neonates: The winner is T-cell receptor excision circles. J Allergy Clin Immunol. 2012;129(3):607–16.
Kwan A, Puck JM. History and current status of newborn screening for severe combined immunodeficiency. Semin Perinatol. 2015;39(3):194–205.
Thakar MS, Hintermeyer MK, Gries MG, Routes JM, Verbsky JW. A practical approach to newborn screening for severe combined immunodeficiency using the T cell receptor excision circle assay. Front Immunol. 2017;8(NOV).

Increased Mortality Rate in Infants with Abnormal T-Cell Receptor Excision Circles

Status:

Journal Publication

Version 1

Abstract

Introduction

Methods

Study population

Data Collection, Data and Statistical Analysis

Statistical Analyses

Results

Baseline Characteristics

Apgar score and congenital anomalies

Maternal characteristics

Pregnancy factors

Delivery factors

Mortality rates

Term status

Discussion

Abbreviations

Declarations

References

Supplementary Files

Status:

Journal Publication

Version 1