Risk factors for readmission for hyperbilirubinemia in neonates with ABO hemolytic disease: a single-center retrospective cohort study

Abstract Objective ABO hemolytic disease of the newborn (ABO-HDN) is a major risk factor for severe hyperbilirubinemia, a common readmission reason for newborns. In this study, we aimed to assess the risk factors for readmission associated with hyperbilirubinemia in neonates with ABO-HDN. Methods A retrospective cohort study was conducted including newborns with gestational age ≥35 weeks and ABO-HDN in 2018. Among 291 newborns, 36 were readmitted for hyperbilirubinemia and defined as the readmission group. The remaining 255 cases were used as a control group. We then performed between-group comparisons of clinical conditions associated with hyperbilirubinemia. Logistic regression was used to select risk predictors of readmission associated with hyperbilirubinemia due to ABO-HDN. Results Baseline characteristics were similar between both groups (p > .05, respectively). However, total serum bilirubin (TSB) before initiating phototherapy was significantly higher in the readmission group when compared with that in the control group at 0–24 h, 24–48 h, and 48–72 h (183.70 µmol/L [interquartile range (IQR) 161.18–196.48] vs. 150.35 µmol/L [IQR 131.73–175.38], p = .005; 229.90 µmol/L [IQR 212.45–284.30] vs. 212.50 µmol/L [IQR 197.85–230.28], p = .026; 268.10 µmol/L [IQR 257.70–279.05] vs. 249.50 µmol/L [IQR 236.80–268.70], p = .045, respectively). The age of initiation of phototherapy in the readmission group was significantly lower than that in control group (30.0 h [IQR 18.0–49.00] vs. 42.0 h [IQR 23.0–61.0], p = .012). The rate of rebound hyperbilirubinemia after the first phototherapy treatment was significantly higher in the readmission group compared to that in the control group (9 [25%] vs. 13 [5.1%], p = .000), and the rate of positive direct antiglobulin testing was significantly higher than that in control group (17 [47.2%] vs. 74 [29.0%], p = .027). Logistic regression analysis showed that the age of initiation of photography, TSB level before the first phototherapy, and rebound hyperbilirubinemia after first phototherapy were independent risk factors for readmission in newborns with hyperbilirubinemia associated with ABO-HDN. Conclusions Earlier age of phototherapy initiation, higher TSB levels at the time of initiating phototherapy and rebound hyperbilirubinemia after the first phototherapy treatment may increase the risk of readmission for hyperbilirubinemia in neonates with ABO-HDN. These factors should be considered in discharge planning and follow-up for newborns with ABO-HDN associated hyperbilirubinemia.


Introduction
Hyperbilirubinemia is a major cause of neonatal readmission, with approximately 60% of term and 80% of preterm infants presenting with jaundice during the first several days life.Of these infants, 5-10% may be treated for hyperbilirubinemia [1].Although hyperbilirubinemia is usually benign, in some newborns severe hyperbilirubinemia may cause irreversible neurological damage without effective treatment [2].Therefore, timely and effective treatment is essential for neonates with hyperbilirubinemia [3].
While many disorders can cause neonatal hyperbilirubinemia, ABO hemolytic disease of the newborn (ABO-HDN) remains the primary etiology of severe hyperbilirubinemia [4][5][6][7][8].Several prolonged hospitalization strategies for these neonates have been recommended in previous reports [9,10].However, these strategies may raise concerns regarding mother-infant separation, which would impede the breastfeeding process.Therefore, instead of indiscriminately prolonged hospitalization strategies, it may be more reasonable to identify those ABO-HDN neonates who are likely to develop severe hyperbilirubinemia and to be at risk of readmission due to hyperbilirubinemia.
Thus, the aim of this study was to identify the predictive risk factors associated with the readmission of newborns with ABO-HDN.

Participants
This retrospective cohort study was conducted between 1 January 2018 and 31 December 2018 at the Department of Neonatology of the Women's Hospital at the Zhejiang University School of Medicine.Newborns !35 weeks of gestational age and a diagnosis of ABO-HDN were included (Figure 1).The study was approved by the Ethics Committee of Women's Hospital, Zhejiang University School of Medicine (IRB-20220062-R).The requirement for informed consent was waived due to the study's retrospective nature.

Inclusion criteria
Inclusion criteria were as follows: (1) gestational age between 35 þ0 and 41 þ6 weeks; (2) hyperbilirubinemia, treated by phototherapy and/or exchange transfusion; and (3) diagnosis of ABO-HDN within three days after birth.

Definition of ABO-HDN
All the following criteria must have been met for a diagnosis of ABO-HDN: (1) ABO group incompatibility between the mother and newborn; (2) hyperbilirubinemia with a positive direct antiglobulin test (DAT) result or hyperbilirubinemia with a negative DAT result and a positive elution testing result with high suspicion of a hemolytic condition such as anemia and/or reticulocytosis.

Methods
Perinatal data were collected for all enrolled newborns.The newborns were then divided into a readmission group and a control group according to readmission status.The following data were compared between the two groups: basic demographic characteristics, total serum bilirubin (TSB) levels before first phototherapy treatment, onset age of phototherapy, incidences of rebound hyperbilirubinemia, immunoglobulin applications, hematocrit upon admission, and DAT results.Logistic regression analysis was applied to estimate independent risk factors for readmission.

Bilirubin monitoring and treatment indications
Transcutaneous bilirubin (TCB) (JM105, Konica Minolta, Osaka, Japan) was measured at 12-h intervals beginning at birth.Total serum bilirubin testing was measured whenever TCB levels reached the 75th percentile of the bilirubin nomogram or exceeded 15 mg/dL [11].According to the 2004 American Academy of Pediatrics (AAP) guidelines [12], if the TSB exceeds the threshold for phototherapy, newborns should be admitted for treatment and primarily receive two-sided phototherapy treatment.When TSB approaches the threshold for exchange transfusion, intensive phototherapy and intravenous immunoglobulin administration are indicated for neonates with ABO-HDN [13].

Indications for termination of phototherapy
Phototherapy was discontinued if the TSB dropped below two standard deviations of the phototherapy threshold at the current age [12,13].

Definition of rebound hyperbilirubinemia
Rebound hyperbilirubinemia was defined as the TSB reaching or exceeding the hour-specific phototherapy threshold (AAP 2004) within 72 h after the termination of the first phototherapy treatment [14].

Discharge criteria and follow-up
Following the termination of phototherapy, all newborns were observed for a minimum of 72 h before discharge.Additionally, only neonates with TSB less than the 75th percentile of Bhutani nomogram were discharged.Once rebound hyperbilirubinemia was identified during the observation period, the discharge would be delayed after another phototherapy treatment.Follow-up plans after discharge were created based on the patient age and the serum bilirubin level according to the AAP management.The follow-up was operated by local hospital physicians through TCB measurement.

Readmission criteria
Newborns were readmitted for treatment whenever the level of TSB exceeded the threshold for phototherapy treatment after discharge.

Statistical analysis
All statistical analyses were performed by the Statistical Package for Social Sciences (SPSS) software, version 20.0 (IBM Corp., Armonk, NY).Depending on their distribution, the data were presented as mean and standard deviation or as median and interquartile range (IQR).Frequency measures were presented as numbers (n) and frequencies (%).Demographic characteristics and outcomes were compared between the readmission group and control groups.The independent sample t-test was used for normally distributed variables and the Mann-Whitney U-test was used for non-normally distributed variables to determine differences between the two groups.The Chi-square test was used for categorical variables.The risks of readmission were estimated by logistic regression analysis.p Values of less than .05were considered to indicate statistical significance.

Results
A total of 291 infants diagnosed with ABO-HDN between 35 þ0 and 41 þ6 weeks of gestational age were included.Of these, 36 newborns were in the readmission group, and the remaining 255 were in the control group.There were no significant differences in basic characteristics between groups including gestational age, birth weight, sex, feeding method, mode of delivery, multiple births, or maternal complications during pregnancy (p > .05).In addition, the incidence of small for gestational age, large for gestational age, cephalohematoma, and polycythemia was similar in both groups (p > .05)(Table 1).
The logistic regression analysis of variables affecting readmission for hyperbilirubinemia in the study population is displayed in Table 3.Among these variables, the onset age of phototherapy, the incidence of rebound hyperbilirubinemia, and TSB level before the first phototherapy were independent risk factors for readmission of newborns with ABO-HDN.

Discussion
In our study, neonates with hyperbilirubinemia accounted for 12.5% (2404/19,293) in our hospital in 2018.The readmission rate was 12.4% (36/291) in neonates with ABO-HDN.Although a low incidence of hyperbilirubinemia in infants after discharge was observed in this study that may have been associated with the prolonged discharge strategy used, as hyperbilirubinemia after discharge remains an issue in infants with ABO-HDN.For example, Chang and Newman [14] analyzed 7048 neonates with hyperbilirubinemia and found that approximately 4.6% of neonates, mostly with positive DAT results, had hyperbilirubinemia after phototherapy treatment.Additionally, a recent study identified a relationship between rebound hyperbilirubinemia in neonates and hemolytic disease, especially ABO-HDN [15].Previous research has focused more on infants during their first hospitalization than after discharge.In contrast, our study aimed to identify the predictive risk factors for readmission in infants with ABO-HDN, and the results may raise interest for individualized follow-up strategies.It has previously been recommended to prolong hospitalization for neonates with ABO-HDN [9,10].For example, Barak et al. [16] suggested that neonates with more risk factors should have prolonged hospitalization, and it has also been suggested that this strategy would reduce the risk of severe hyperbilirubinemia for neonates with ABO-HDN [14,15].In our hospital, this is a strategy that has long been applied.However, we must also consider the benefits of breastfeeding that have been emphasized in recent years.In line with this, Chang et al. [17] have suggested that prolonged hospitalization was detrimental to breastfeeding.
In our study, we strictly adopted the AAP statement on the treatment and follow-up of hyperbilirubinemia.During the study period, one attending physician was responsible for the management of hyperbilirubinemia to reduce treatment bias and strengthen the confidence level of the study.
Our results showed that there was a higher risk of readmission after discharge when the bilirubin rebounded within 72 h after the termination of the first phototherapy treatment.This was consistent with the results of Sachdeva et al. [18] that suggested that neonates with rebound hyperbilirubinemia after phototherapy may require close follow-up after discharge.We found that 34.4% (100/291) of neonatal hyperbilirubinemia cases occurred within 24 h after birth, and 12.4% of these patients were readmitted after discharge.ABO-HDN is well-known as the main cause of early-onset neonatal hyperbilirubinemia, and our findings were consistent with those of previous studies such as Tsujimae and Yamana [19].We also found that higher TSB levels before the first phototherapy treatment were an independent risk factor for readmission.Therefore, closer follow-up should be performed for neonates with early higher levels of TSB, higher incidence of rebound hyperbilirubinemia and earlier onset of phototherapy after discharge.
In our study, a positive DAT result was not an independent risk factor for readmission.We speculated that neonates with ABO-HDN and a positive DAT result may have had a longer duration of hospitalization and received more phototherapy.Additionally, the occurrence time of hyperbilirubinemia has been shown to be a valuable prediction for readmission in the neonates with ABO-HDN [20][21][22].
Individualized management of hyperbilirubinemia has developed based on the AAP guidelines.For example, Japan has developed its own guidelines for the management of hyperbilirubinemia in full-term infants and preterm infants.However, there is still no full consensus about the management of near-term or term infants with ABO-HDN.Our study may provide some information for these populations.We believe that not all neonates with ABO-HDN require extended hospitalization, despite a relatively higher rate of hyperbilirubinemia in this patient population.We speculate that it may be more reasonable to adopt the pronged hospitalization strategy for only those neonates with early-onset hyperbilirubinemia.In addition, closer follow-up after discharge should be more strongly considered [3,23].
There were some limitations in our study.First, it was a single-center retrospective cohort study, and the sample of our study was relatively small.Second, data were taken from a pre-pandemic era, but postpandemic hospital discharge and readmission management might have changed, and our study cannot provide a more valuable reference for post-pandemic risk factors and management.In addition, our study lacks long-term follow-up data.In the future, we may develop predictive models by expanding the sample size to find more risk factors and include data from the post-pandemic era.
In conclusion, we believe that it is necessary to extend the length of hospitalization and to implement a close follow-up strategy after discharge for those neonates with early-onset hyperbilirubinemia, rebound hyperbilirubinemia after phototherapy and those with a higher level of TSB at first phototherapy treatment.We speculated that such a strategy should not only reduce the total time of maternal-infant separation but also avoid increasing the incidence of severe hyperbilirubinemia by reducing the rate of repeated readmission due to phototherapy.

Figure 1 .
Figure 1.Flow of participation for the neonates with ABO-HDN.

Table 2 .
Comparison of examination test and treatment.

Table 3 .
Logistic regression model to predict readmission risk factors.

Table 1 .
Comparison of demographic features and complications.
a Data expressed as numbers (%) unless otherwise indicated.