The timing of using IVIG for neonatal ABO hemolytic disease

Abstract

Background

ABO hemolytic disease of the newborn (ABO HDN) is the most prevalent cause of hemolytic disease of the newborn, often leading to hyperbilirubinemia. In some cases, ABO HDN can progress to significant hyperbilirubinemia. The role of intravenous immunoglobulin (IVIG) in treating ABO HDN remains controversial. Several key questions regarding significant hyperbilirubinemia remain unanswered: What are the risk factors of ABO HDN with significant hyperbilirubinemia? Can IVIG provide benefits for ABO HDN with significant hyperbilirubinemia? Where do we place IVIG in the treatment sequence paradigm of ABO HDN? How do we best stratify ABO HDN with significant hyperbilirubinemia for IVIG therapy?

Methods

We conducted a retrospective cohort study include 948 newborns from West China Second University Hospital, divided into two groups according to the use/nonuse of IVIG. The maternal and neonatal baseline clinical data were collected from digital medical record system. The risk factors of the significant hyperbilirubinemia associated with ABO HDN were investigated using univariate and multivariate analysis. The generalized additive mixed model (GAMM) was used to analyze the nonlinear relationship of bilirubin level with different treatment over time.

Results

A total of 948 newborns fulfilled the inclusion criteria, with 143 (15.1%) in the IVIG group, and 805 (84.9%) in the non-IVIG group. (1) The multivariate analysis found that gestational age ≥ 37 weeks (OR 4.19; p = 0.019), age at admission range 24–48 h (OR 9.69; p = 0.03) and age at admission >48 h (OR 50.31; p < 0.0001) were affect significant hyperbilirubinemia in neonates. (2) The total serum bilirubin (TSB) descent speed of IVIG group (0.69µmol/L/h, 95%CI, 0.48–0.89µmol/L/h) was higher than non-IVIG group (0.21µmol/L/h, 95%CI, 0.14–0.29µmol/L/h) among the neonates with age at admission ≤ 48 h (p < 0.0001);No adverse effects related to IVIG treatment were recorded in either group.

Conclusion

Regular monitoring of TSB levels starting 24 h after birth is an effective strategy to prevent severe hyperbilirubinemia in ABO HDN. For southwest Chinese neonates, IVIG should not be routinely used to treat ABO HDN in those admitted > 48 h after birth.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12887-025-06062-0.

Background

The imbalance between increased breakdown of red blood cells (RBCs) and decreased bilirubin excretion is the main cause of the hyperbilirubinemia [1–4]. Hyperbilirubinemia is the most common clinical disease in newborns, and is one of the leading causes of neonatal hospitalization within the first week of life [5]. While most newborns with jaundice have a benign prognosis, severe cases can lead to death or long-term neurological sequelae [6–8]. ABO hemolytic disease of the newborn (ABO HDN) is the most prevalent cause of hemolytic disease of the newborn, which may lead to hyperbilirubinemia [9, 10]. Due to the neurotoxicity of unconjugated bilirubin, neonates with significant hyperbilirubinemia are at the risk of developing acute bilirubin encephalopathy (ABE) and kernicterus, if tjaundice is not regularly monitored and treated [5, 11–13]. The overall global incidence of significant hyperbilirubinemia was reported to be 99 per 100,000 live births [5]. The incidence rate of kernicterus ranges from approximately 0.2 to 2.7 cases per 100 000 livebirths [14, 15]. According to statistics, neonatal hyperbilirubinemia was the 7th leading cause of neonatal death in South Asia, 8th in sub-Saharan Africa, 9th in Western European and 13th in North America [5].

Phototherapy (PT) and exchange transfusion (ET) are two traditional approaches for treating neonatal hyperbilirubinemia [3, 8]. In addition, IVIG has been used for the treatment of ABO HDN for many years [15–18]. IVIG therapy has the advantages of non-invasive, convenient and technically simpler compared to ET [10, 19, 20]. Despite this, there are still some countries whose guidelines do not provide clear recommendations regarding the use of IVIG, such as Queensland Clinical Guidelines in Australia and Swiss Society of Neonatology [21, 22]. The American Academy of Pediatrics (AAP) provides recommendations: IVIG may be provided to infants with isoimmune hemolytic disease (i.e., positive direct antiglobulin test, DAT) whose TSB reaches or exceeds escalation of care threshold (defined as 2 mg/dL below the exchange transfusion threshold) [18]. Besides, the use of IVIG also requires the assessment of the other factors including response to phototherapy, TSB rate of increase, and the challenge of providing timely ET. According to the latest AAP guidelines, the use of IVIG in treating ABO HDN warrants careful consideration, as its effectiveness in preventing the need for ET, its safety and optimal timing for administration remain subjects of debate [15–17, 20, 23–30]. Studies by Miqdad and Nasseri demonstrated that IVIG administration in ABO HDN not only reduced the need for ET but also showed no adverse effects [17, 29]. However, two other studies concluded that using IVIG did not reduce the need for ET [31, 32]. Additionally, studies by Figueras and Yang found that the use of IVIG was associated with a higher incidence of necrotizing enterocolitis (NEC) [26, 27]. Other studies have also reported some side effects such as higher need of top-up transfusions and longer hospital stay related to IVIG therapy [28]. Maternal IgG antibodies, which are reactive against fetal RBCs antigens inherited from the father, can be transferred across the placenta and destroyed fetal RBCs causing hemolytic disease in the fetus and neonates [9]. Since the maternal IgG antibodies are no longer transferred across the placenta after birth, the number of sensitized RBCs in the newborn’s circulation gradually decrease over time. Therefore, the use of IVIG in ABO HDN should be restricted to conditions where its benefits have been clearly demonstrated.

This study aims to address several unresolved questions concerning significant hyperbilirubinemia: What are the risk factors of ABO HDN with significant hyperbilirubinemia? Can IVIG provide benefits for ABO HDN? Where do we place IVIG in the treatment sequence paradigm of ABO HDN? How do we best stratify ABO HDN with significant hyperbilirubinemia for IVIG therapy?

Methods

Study design

A retrospective cohort study was conducted on newborns with ABO HDN at the West China Second University Hospital, Sichuan University region between March 2014 and January 2020. The neonate cohort was divided into two groups according to the use/nonuse of IVIG. The decrease rate and descent speed of TSB were evaluated during the hospitalization. All procedures were performed, and written informed consent was obtained in accordance with the relevant guidelines and regulations of the Declaration of Helsinki.

Participants

The maternal and neonatal baseline clinical data were collected from digital medical record system admitted to the neonatal intensive care unit (NICU). The inclusion criteria for this study were as follows: (1) positive direct anti-human globulin test or positive antibody release test; (2) meet the diagnostic criteria of neonatal significant hyperbilirubinemia; (3) diagnostic criteria for neonatal anemia: venous hemoglobin within one week of birth was less than 140 g/L. (1) and (2) were essential conditions. The exclusion criteria for this study were as follows: (1) neonates older than 7 days at admission (n = 208); (2) neonates with other diseases at admission (sepsis, inherited metabolic diseases, congenital heart disease, coagulation disorder, intracranial hemorrhage, cranial hematoma larger than 3 centimeter in diameter etc.) (n = 188); (3) length of hospitalization ≤ 1 day (n = 17). (4) neonates with incomplete medical records (n = 44).(5) neonates with gestational age < 35 weeks (n = 22). Newborns who met all study criteria were divided into two groups: newborns who treated with phototherapy and IVIG (IVIG), and newborns who did not treated with IVIG (non-IVIG).

Diagnosis

Neonatal significant hyperbilibinemia was confirmed as the level of neonatal TSB exceeding the 95th percentile of neonatal hourly bilirubin histogram made by Bhutani et al., or the level of TSB exceeding the American Academy of Pediatrics (AAP) recommended reference curve of phototherapy [17, 33]. Acute bilirubin encephalopathy is mainly based on the clinical manifestation and laboratory examinations, commonly seen in neonates with TSB > 342µmol/L(20 mg/dl), and (or) the rapidly increased bilirubin level (> 8.5µmol/L/h), gestational age > 35weeks. This study used the following criteria: abnormal muscle tone; abnormal brainstem auditory evoked potentials; hyperintense globus pallidus on T1-weighted MR imaging [18].

Treatment

The phototherapy machines used in this study could produce the light wavelength near 460ཞ490nm, and the irradiance intensity of intensive phototherapy was approximately 30 µw/cm²/nm at least.

Since there are no detailed guidelines for the use of IVIG, the decision of whether or when to use of IVIG were made by pediatricians depending on the guidelines from Chinese Pediatric Society (ChPS) Chinese medical Association [34], or neonates whose TSB rising in spite of intensive phototherapy or within 2–3 mg/dl (34–51µmol/L) of exchange level. Considering the associated risks of IVIG, some pediatricians may not choose IVIG especially for late newborns. In addition, there were some parents of the neonates who refused to use IVIG. In this retrospective cohort study, we analyzed the efficacy and safety of the IVIG in the management of ABO HDN in order to explore the optimal timing of IVIG.

Statistical analysis

All the analyses were performed with the statistical software packages R (http://www.R-project.org, The R Foundation) and EmpowerStats (http://www.empowerstats.com, X&Y Solutions, Inc., Boston, MA). Continuous variables with normal distribution were presented as mean ± SD, and categorical variables were presented as count and percentage. Descriptive analyses to evaluate differences between the two groups (IVIG vs. non-IVIG) included t-tests for continuous variables with normal distribution, X² tests or Fisher’s exact tests for categorical variables. Univariate and multivariate logistic regression analyses were used to explore the related factors that may lead to the occurrence of hyperbilirubinemia. The generalized additive mixed model was used to analyze nonlinear relationship between the bilirubin level and the duration of treatment, and to investigate whether there was a difference in the descent speed of bilirubin level between the two groups. In all analyses values of p < 0.05 were considered statistically significant.

Results

Patients’ selection and baseline clinical characteristics

We identified a total of 1427 newborns with ABO HDN. Then, we selected the patients according to the inclusion criteria and exclusion criteria. Finally, a total of 948 neonates were included in this study, of whom 263 (27.7%) had TSB level were within 2 to 3 mg/dl of the exchange level or above the exchange level. The flowchart of patients’ selection was summarized in Fig. 1. There were 143 (15.1%) neonates in the IVIG group, and 805 (84.9%) neonates in the non-IVIG group, and most neonates in both groups were breastfed. There were significant differences in terms of the age, bilirubin level at admission, and serum free antibody test between two groups. The mean age at admission of IVIG group was 38.17 ± 27.81 h, significantly earlier than 54.18 ± 35.02 h in non-IVIG group (p < 0.0001). The mean TSB level of IVIG group was 260.52 ± 112.75 µmol/L, significantly higher than that of 236.44 ± 85.72 µmol/L in non-IVIG group (p = 0.003). The positive rate of serum free antibody test in IVIG group was significantly higher than that in non-IVIG group (90.9% vs. 81.1%, p = 0.004). Baseline clinical characteristics of included patients were summarized in Table 1.

Fig. 1.

Patient flow diagram. ABO-HDN, ABO hemolytic disease; GA, gestational age; NICU, neonatal intensive care unit; IVIG, intravenous immunoglobulin G

Table 1.

Clinical characteristic of infants, by IVIG or Non-IVIG. Dates 2014–2020

Variables	IVIG (n = 143)	Non-IVIG(n = 805)	P Value
Age to admission (h)	38.17 ± 27.81	54.18 ± 35.02	<0.0001
Gestational age (wk)			0.461
<37	14(9.8%)	64(8.0%)
≥ 37	129(90.2%)	741(92.0%)
Sex			0.351
Male	59(41.3%)	366(45.5%)
Female	84(58.7%)	439(54.5%)
Birth weight (g)	3305 ± 467.5	3287 ± 455.7	0.078
Age at the time of jaundice discovered (h)	30.01 ± 23.49	31.95 ± 23.24	0.36
Bilirubin level at admission (µmol/L)	260.52 ± 112.75	236.44 ± 85.72	0.003
Hemoglobin levels at admission (g/L)	158.58 ± 25.86	160.60 ± 25.56	0.386
Phototherapy treatment time (h)	90.58 ± 34.82	88.06 ± 55.0	0.597
Hospitalization in the NICU(h)	95.84 ± 36.17	92.89 ± 40.40	0.414
DAT			0.708
Positive	91(63.6%)	499(62.0%)
Negative	52(36.4%)	306(38.0%)
Free antibody test			0.004
Positive	130(90.9%)	653(81.1%)
Negative	13(9.1%)	152(18.9%)
Antibody-released test			0.832
Positive	137(95.8%)	768(95.4%)
Negative	6(4.2%)	37(4.6%)
Exchange transfusion			0.31
Yes	7(4.90%)	23(2.86%)
No	136(95.10%)	782(87.8%)
Use albumin			0.567
Yes	15(10.5%)	98(12.2%)
No	128(89.5%)	707(87.8%)
Conventional phototherapy			0.067
Yes	59(41.3%)	399(49.6%)
No	84(58.7%)	406(50.4%)

IVIG Intravenous immunoglobulin G; NICU Neonatal intensive care unit, DAT Direct antiglobulin test

p values from t-tests for continuous variables with normal distribution, nonparametric statistic tests for continuous variables with non-normal distribution and X

2 tests for categorical variables

Risk factors of significant hyperbilirubinemia

The results of univariate and multivariate logistic regressions were provided as follow (Tables 2 and 3), that revealed the risk factors of significant hyperbilirubinemia in neonates with ABO-mediated hemolytic disease. In Table S1, univariate logistic regression shows that gestational age ≥ 37 weeks (OR 3.44; p = 0.04), age at admission range 24–48 h (OR 9.59; p = 0.03), age at admission >48 h (OR 48.34; p < 0.0001), and time to jaundice detection > 24 h (OR 4.25; p < 0.0001) are associated with significant hyperbilirubinemia. In Table S2, multivariate logistic regression reveals that gestational age ≥ 37 weeks (OR 4.19; p = 0.019), age at admission range 24–48 h (OR 9.69; p = 0.03) and age at admission >48 h (OR 50.31; p < 0.0001) were affect significant hyperbilirubinemia in neonates.

Table 2.

For admission age >48 h, TSB decreasing rate in different time period, by IVIG or non-IVIG

Time(h)	TSB decreased	IVIG	Non-IVIG	P value
12	Yes	32(100.0%)	221(92.5%)	0.219
	No	0(0.0%)	18(7.5%)
24	Yes	22(100.0%)	205(98.6%)	0.99
	No	0(0.0%)	3(1.4%)

IVIG The group use intravenous immunoglobulin G, Non-IVIG The group nonuse of intravenous immunoglobulin G

Table 3.

Comparison of TSB descent speed of different age to admission, by IVIG or non-IVIG

Age at admission (h)	Group	TSB descent speed (µmol/L/h)	95% CI(µmol/L)	P value
≤ 48	IVIG	0.69	0.48ཞ0.89	< 0.0001
	Non-IVIG	0.21	0.14ཞ0.29	< 0.0001
	IVIG vs. Non-IVIG	0.47	0.29ཞ0.65	< 0.0001
>48	IVIG	1.71	1.38ཞ2.04	< 0.0001
	Non-IVIG	1.44	1.29ཞ1.59	< 0.0001
	IVIG vs. Non-IVIG	0.28	−0.12ཞ0.67	0.17

IVIG The group use intravenous immunoglobulin G, Non-IVIG The group nonuse of intravenous immunoglobulin G

IVIG treatment in ABO HDN with significant hyperbilirubinemia

In Table S3, we can see the TSB decrease rates of bilirubin in IVIG group were 96.2%, 94.7% and 89.7%, respectively at 12 h, 24 h and 48 h after admission, and these decreased rates were 82.6%, 86.1% and 82.5%, respectively in non-IVIG group. Therefore, the bilirubin decrease rate of IVIG group was higher than that of non-IVIG group, and was statistically significant at 12 h after admission (p<0.0001). Further stratified analysis was performed according to the age at admission.

Table S4 showed the bilirubin decrease rates of neonates aged at admission ≤ 48 h in IVIG group were 94.4% and 92.6%, respectively at 12 h and 48 h after admission, and these decrease rates were 70.7% and 73.6%, respectively in non-IVIG group. The bilirubin decrease rate of IVIG group were higher than that of non-IVIG group of neonates aged at admission ≤ 48 h (all p<0.05).

Table 2 showed the bilirubin decrease rates of neonates aged at admission >48 h in IVIG group were 100.0% and 100.0%, respectively at 12 h and 48 h after admission, and these decrease rates were 92.5% and 98.6%, respectively in non-IVIG group. There were no significant difference between the two groups.

Age-stratified analysis indicated that the bilirubin decrease rate of IVIG group was higher than that of non-IVIG group, and was statistically significant when the age at admission was no later than 48 h. Based on a generalized additive mixed model, the therapeutic effects of descent speed of bilirubin were further evaluated between the two groups. Figures 2 and 3 showed the nonlinear relationship between the bilirubin level and time.

Fig. 2.

TSB change curve over time based on Generalized Additive Mixed Model, age at admission >48 h. The curve composed of hollow circles represent the group use intravenous immunoglobulin G, and the curve composed of solid circles represent the group nonuse of intravenous immunoglobulin G

Fig. 3.

TSB change curve over time based on generalized additive mixed model, age at admission ≤ 48 h. The curve composed of hollow circles represent the group use intravenous immunoglobulin G, and the curve composed of solid circles represent the group nonuse of intravenous immunoglobulin G

Table 3 showed the bilirubin descent speed of each groups of neonates at different admission age. There were 508 neonates aged ≤ 48 h at admission, 94 (18.5%) neonates had TSB level were within 2 to 3 mg/dl of the exchange level or above the exchange level. The descent speed of IVIG group and non- IVIG group were 0.69µmol/L/h (95%CI 0.48–0.89µmol/L/h, p < 0.0001) and 0.21µmol/L/h (95%CI 0.14–0.29µmol/L/h, p < 0.0001), respectively. The rate of bilirubin decrease in the IVIG group was 0.47 µmol/L/h (95% CI 0.29–0.65µmol/L/h, p < 0.0001) higher than that in the non-IVIG group for neonates aged ≤ 48 h at admission.

There were 440 neonates aged >48 h at admission, 169 (38.4%) neonates had TSB level were within 2 to 3 mg/dl of the exchange level or above the exchange level. The descent speed of IVIG group and non- IVIG group were 1.71µmol/L/h (95%CI 1.38–2.04µmol/L/h, p < 0.0001) and 1.44µmol/L/h (1.29–1.59µmol/L/h, p < 0.0001), respectively. The rate of bilirubin decrease in the IVIG group was 0.28µmol/L/h (95% CI −0.12−0.67µmol/L/h, p = 0.17) higher than that in the non-IVIG group for neonates aged > 48 h at admission.

There were 263 cases that met the AAP criteria for IVIG administration in our study, of whom 70 (26.6%) cases received IVIG. There were 20 cases required ET within the first 48 h of life, among them, there were 11 (55%) cases in the IVIG group, and 9 (84.9%) cases in the non-IVIG group.The results of the TSB decrease rates were shown in Table S5 in different time period from cases met AAP criteria for IVIG administration in IVIG group and non-IVIG group. It can be concluded from the above data, there was no difference in the TSB decrease rates between the IVIG group and the non-IVIG group for whose TSB reaches or exceeds 2 mg/dL below the exchange transfusion threshold. The rates of ET in the IVIG group and the non-IVIG were 4.3% and 3.6% respectively (p = 0.805).

Complications and treatment outcomes of ABO HDN

Table S6 showed the complications and treatment outcomes of ABO HDN between the two groups. None of the neonates developed NEC during hospitalization.There were no significant difference in the incidence of complications, the duration of phototherapy, the length of hospitalization and the readmission rate between the two groups (all p>0.05).

Discussion

Our study found that gestational a gestational age ≥ 37 weeks, age at admission > 24 h and time to jaundice detection > 24 h were associated with significant hyperbilirubinemia in southwest China. A study from Boo et al. also showed that delayed detection of TSB and commencing photography increased risk of significant hyperbilirubinemia [35]. This finding that delayed detection of TSB is associated with significant hyperbilirubinemia is consistent with our research. However, another study from Turkey on significant hyperbilirubinemia in term and near term infants revealed that gestational age was not a risk factors of significant hyperbilirubinemia, which was inconsistent with the present study [36]. There could be several reasons as follow: (1) the term infants discharge from the hospital with their mothers earlier than near term infants because their organ functions were more mature. Owing to the lack of regular monitoring of TSB in term infants after discharge, jaundice may be detected late, and the term infants may had higher TSB level at admission. The result of Bergmann et al. also proved that early discharge from hospital (< 36 h) was a risk factor of significant hyperbilirubinemia [37]; (2) In the diagnosis and treatment of neonatal jaundice in Turkey, it is emphasized that for neonates discharged within 72 h after birth, doctors need to evaluate the risk factors for hyperbilirubinemia, formulate a reasonable follow-up plan according to the age of the neonates at discharge and the related risk factors, and provided parents with disease education. Different monitoring of neonatal TSB level, and different parental awareness of the disease may have contributed to the differences in results between the two studies. Therefore, it is of great significance to strengthen the evaluation of neonates’ condition before discharge, carry out disease publicity and education for parents at discharge, and regular follow up and monitor the TSB level, so as to reduce the incidence of significant hyperbilirubinemia and prevent ABE. The AAP guidelines recommend that TSB level should be monitored every 8 to 12 h during hospitalization, earlier and more frequent follow-up should be provided after discharge for ABO HDN [18]. Based on the result of this study, delayed hospitalization was a risk factor for significant hyperbilirubinemia in southwest China, newborn transcutaneous bilirubin (TcB) level measurement should be peformened at least 1 to 2 time a day after discharge until the newborn has passed the peak of jaundice. At the same time, continuously plotting nomograms of TcB level or TSB level after birth can more accurately reflect changes in neonatal bilirubin levels, guilding the frequency of bilirubin monitoring and follow-up evaluations after discharge.

Additionally, univariate analysis revealed that neonate with negative DAT were more likely to develop significant hyperbilirubinemia in southwest China, which was different from other studies. Due to the late development of ABO blood group antigens in the fetal period, the expression of some RBCs surface antigens are weak, so the positive rate of DAT in diagnosis of ABO HDN is low, and the false negative rate is high [33]. Considering that the later the neonates were on admission, the higher the TSB level of the neonates, indicating increased destruction of the neonates’sensitized RBCs. And the number of sensitized RBCs were already small at these neonates with late admission, the positive rate of DAT detection was relatively low at this time. Therefore, hemolysis screening should be carried out as soon as possible for neonate suspected of hemolysis.

The efficacy and safety have investigated of IVIG treatment for ABO HND remain controversial, mainly including the effectiveness of IVIG to prevent the need for an exchange transfusion, the increased risk of NEC and the optimal timing of using IVIG [16, 17, 23, 24, 26, 27]. Futhermore, previous studies has not performed stratified analyses of neonates based on ABO HDN specific characteristics. The present study differs from previous studies on IVIG because it explored the nonlinear relationship between the bilirubin level and the duration of treatment. These results can guide us develop a strategy for the standardized recommendation of IVIG in ABO HDN in southwest China. Our study found that the bilirubin decrease rate and bilirubin descent speed of IVIG group were higher than that of non-IVIG group. Besides, there were no significant difference in the incidence of complications, the duration of phototherapy, the length of hospitalization and the readmission rate between the two groups. Therefore, IVIG is an effective and safe strategy in ABO HDN with signifcant hyperbilirubenemia. In this study, there were 263 neonates had TSB level were within 2 to 3 mg/dl of the exchange level or above the exchange level, 94 neonates aged ≤ 48 h at admission, and 169 neonates aged >48 h at admission, respectively. In neonates, TSB level within 2 to 3 mg/dl of the exchange level or require ET was more common in neonates aged >48 h at admission (38.4% vs. 18.5%), and was statistically significant (p < 0.0001). Therefore, we further performed stratified analyses of neonates according to the age. We found that for neonates aged > 48 h at admission, although the rate of bilirubin decrease in the IVIG group was 0.28µmol/L/h (95% CI −0.12−0.67µmol/L/h, p = 0.17) higher than that in the non-IVIG group, the difference between the two groups was not statistically significant. For these neonates with significant hyperbilirubinemia, aged > 48 h at admission, most of their sensitized RBCs have already been destroyed, making IVIG treatment of limited effect. This suggested that PT alone could effectively reduce TSB and IVIG use should be evaluated more carefully with neonates age at admission > 48 h.

Given the retrospective single-center nature of this study, observation bias may have been reduced, but missing data cannot be avoided, such as the changes of hemoglobin after discharge, the subsequet nervous system development of these neonates with significant hyperbilirubinemia, and the differences between regions. So, a prospective study is needed in the future to collect more data of ABO HDN, especially for neonates at high risks of ET, to detect the relationship between the IVIG treatment and decreased ET and ABE. Since it was a retrospective observational design without randomization, there were differences between the two groups, although we adjusted for baseline characteristics of neonates, unmeasured confounders could have remained such as the the intensity of phototherapy. Furthermore, the present study was conducted in a single-center. A multi-center, and prospective study is needed to validate the study of ours in the future taking different regions into account.

Conclusion

Therefore, we recommend that newborns with evidence of ABO hemolysis should be monitored for jaundice at least 2 ~ 3 times within 24 h after birth to promptly detect newborns with rapidly rising jaundice. This study also indicate that the use of IVIG can promote a rapid decrease in blood bilirubin levels for neonates with ABO HDN with severe hyperbilirubinemia and age ≤ 48 h at admission, but it is not very helpful for neonates with ABO HDN with admission age > 48 h in southwest Chinese neonates.

Abbreviations

HDN

Hemolytic disease of the newborn

IVIG

Intravenous immunoglobulin

GAMM

Generalized additive mixed model

TSB

Total serum bilirubin

RBCs

Red blood cells

ABE

Acute bilirubin encephalopathy

PT

Phototherapy

ET

Exchange transfusion

AAP

American Academy of Pediatrics

DAT

Direct antiglobulin test

NEC

Necrotizing enterocolitis

NICU

Neonatal intensive care unit

ChPS

Chinese Pediatric Society

TcB

Transcutaneous bilirubin

Authors’ contributions

Na Ma wrote the main manuscript text and Feihong Zhang prepared figures and tables. Yong Hu and Bin Xia designed and managed this manuscript. All authors reviewed and edited the manuscript.

Funding

Not applicable.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki(as was revised in 2013). The study was approved by Ethics Committee of the West China Second University Hospital, Sichuan University. We obtained written informed consent from legal guardians for all participant.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.