Clinical implications of aminotransferase elevation in hospitalised infants aged 8‐90 days with respiratory virus detection

Abstract Background Fever and respiratory symptoms are the major causes of hospitalisation in infants aged 90 days or less. Respiratory viruses (RVs) are detected by multiplex reverse transcriptase‐polymerase chain reaction (mRT‐PCR) in up to 70% of infants tested in this population. Aminotransferase elevation is not uncommon in RV infections, and repeat laboratory investigations are frequent due to concerns regarding the occurrence of hepatic disease. Methods This retrospective observational cohort study included 271 infants aged 8‐90 days, with positive RV mRT‐PCR results. Data were obtained on demographics, laboratory results and final diagnoses of hepatobiliary disease. Results Fever (73.1%) and/or respiratory symptoms (75.6%) were the major presentations among the hospitalised infants. Aspartate aminotransferase (AST) or alanine aminotransferase (ALT) levels were elevated in 62 (22.9%) of the 271 infants. Twenty‐four of these 62 infants had their first follow‐up, and 19 (79.2%) showed persistent elevation. All 10 (100%) infants who had their second follow‐up showed persistently elevated aminotransferase levels. Eventually, none of the 10 infants were diagnosed with hepatic disease during the median follow‐up of 10 days (range 3‐232 days). Among the RVs of interest, parainfluenza virus type 1 was significantly associated with aminotransferase elevation (odds ratio: 2.95; 95% confidence interval [CI]: 1.11‐7.83). Conclusions RV‐related non‐specific hepatitis is occasionally observed in infants aged 8‐90 days, and ALT elevation is the most common abnormality. However, a final diagnosis of primary hepatobiliary disease appears to be rare. Therefore, regular follow‐ups and targeted testing may be recommended in this specific population.


| INTRODUC TI ON
Respiratory viruses (RVs) involved in childhood lower respiratory tract infections can be identified using the multiplex reverse transcriptase-polymerase chain reaction (mRT-PCR). 1,2 Despite the lack of clear guidelines for the use of mRT-PCR in the management of febrile infants aged younger than 90 days, this diagnostic modality is frequently used to reduce antibiotic use, length of hospital stay and laboratory investigations and for aiding decision-making by doctors and families in certain cases. [3][4][5][6] The guidelines for the management of febrile infants aged 90 days or less vary from one institution to another. In general, the mainstay of management involves hospital admission and the administration of empiric antibiotics. 7 The hospital admission panel or battery measures the basic laboratory parameters of a patient at admission, that commonly includes aminotransferase measurements. 8 RV infections are known causes of non-specific hepatitis, which present with aminotransferase elevation. 9 Non-specific hepatitis is not uncommon even with a wide range of "normal" aminotransferase levels in this specific age group. Owing to the diagnostic possibility of true hepatic disease, additional laboratory follow-ups are performed based on the care providers' preference. Such investigations may prolong the duration of hospital stay and impart undue stress to both, the baby and the caregiver. 10 The main objective of the study was to evaluate the true clinical implications of aminotransferase elevation in infants aged 8-90 days hospitalised with RV infections. The association between the type of RV and aminotransferase elevation was also investigated.

| Patient characteristics
This retrospective observational cohort study included infants Infants born in the hospital and those with positive results for multiple RVs were excluded. Infants aged less than 8 days were excluded based on the incubation period of the viruses. The upper limit (90 days) was set based on the well-recognized fact that febrile infants aged between 0 and 90 days are at a higher risk of severe bacterial infections (SBIs) than infants and children aged between 3 and 36 months. 11 Basic demographic data, laboratory and imaging results, data on the final diagnosis of hepatic disease, if any, and mortality data were collected. Information was also obtained regarding the initial presentations of the infants. Fever was defined as a tympanic temperature of at least 38.0°C, as measured by medical personnel in the institute. Perceptions of fever by the caregiver or temperatures measured outside the hospital were excluded. Complete blood cell counts, aminotransferase levels, bilirubin, albumin and the coagulation panel were tested. For infants with elevated aminotransferases, data on the subsequent follow-up and laboratory tests were collected. Aminotransferase levels of infants with aminotransferase elevation at each follow-up stage were compared. Aminotransferase levels were also compared between infants who did and did not require follow-up.

| Criteria for elevated aminotransferase levels
The normal ranges of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) varied with sex and age, and were defined as follows: for AST, less than 72 IU/L for infants younger than 1 month and less than 64 IU/L for infants older than 1 month, and for ALT, less than 41 IU/L for boys younger than 1 month, less than 33 IU/L for girls younger than 1 month and less than 46 IU/L for both, boys and girls older than 1 month. 12 The aminotransferase elevations in patients were defined as elevations of either AST or ALT, unless otherwise specified.

| Respiratory virus identification
The presence of fever and/or respiratory symptom(s) was the major indication of a RV, which was subsequently identified using mRT-PCR. The decision to perform the test was made independently by the attending physician. Nasopharyngeal swab specimens were

| Statistical analyses and sample size calculations
Statistical analyses were performed using SPSS Statistics for Windows version 25.0 (IBM Corp.). Demographic, clinical and laboratory data were evaluated and presented descriptively.
Continuous data were analysed using the paired sample t test or the Mann-Whitney U test. Categorical data were analysed using the chi-squared test or Fisher's exact test. P < .05 was considered statistically significant.
Minimum sample sizes for detecting mean differences with sufficient statistical power (α = 0.05, β = 80%) were calculated and assured at each stage of the laboratory test in groups with and without aminotransferase elevation. For viruses associated with significant aminotransferase elevation, statistical power for the sample sizes was calculated separately.

| Ethical consideration
The institutional review board of the Chungbuk National University Hospital approved the study protocol (IRB no.2019-01-002). The need for informed consent was waived due to the retrospective nature of the study and data anonymisation.

| Initial laboratory results
At admission, the mean AST was 55. The mean AST and ALT levels in infants without and with aminotransferase elevations were 34.1 ± 9.5 IU/L vs. 129.0 ± 218.7 IU/L (P = .001) and 23.2 ± 7.9 IU/L vs. 131.4 ± 194.0 IU/L (P < .001), respectively. No differences were observed between infants without and with aminotransferase elevation in terms of the mean haemoglobin levels, platelet counts, neutrophil or lymphocyte differential counts, total bilirubin and C-reactive protein levels. However, the mean white blood cell count (10 837.2 ± 4535.3/µL vs 12 565.6 ± 4935.7/ µL, respectively; P = .010) and albumin level (4.2 ± 0.3 g/dL vs 4.3 ± 0.2 g/dL, respectively; P = .029) were significantly lower in infants without aminotransferase elevation than in those with aminotransferase elevation.

| First laboratory follow-up
The mean aminotransferase levels from the initial laboratory test

| Further laboratory evaluations and other investigations
Among the 10 infants who persistently showed aminotransferase elevations, four underwent additional follow-up investigations; these were performed once in three infants and six times in an infant with suspected Wilson's disease, which was eventually found to be absent. None of the 10 infants who required more than three followups were diagnosed with hepatobiliary disease.
Among the 62 infants with initial elevated aminotransferase levels, nine (14.5%) and seven (11.3%) underwent serologic testing for hepatotropic pathogens and ultrasonographic evaluation of the hepatobiliary system, respectively. Serologic testing for hepatotropic pathogens included tests for the following viruses: hepatitis A/B/C virus in 6/9 (66.6%) infants, herpes simplex virus in 7/9 (77.8%) infants, cytomegalovirus in 9/9 (100%) infants and Epstein-Barr virus in 5/9 (55.6%) infants. Except for one infant who demonstrated a positive cytomegalovirus IgM test result, the serologic tests were found to be negative in all other cases. No hepatobiliary abnormalities were noted on ultrasonography.

| Respiratory virus types and aminotransferase elevation
The percentage of infants with aminotransferase elevation for each type of RV was calculated ( Table 2

| D ISCUSS I ON
The two major findings of this study are as follows:  Note: Values are presented as mean ± standard deviation, unless specified otherwise.
Non-specific hepatitis presenting as aminotransferase elevation is not uncommon in young children. Studies to understand the pathologic mechanisms underlying aminotransferase elevation in the paediatric population are limited. A study in 559 paediatric patients found that aminotransferase elevation was truly non-specific in 7.5%. 9 In the same study, infections were found to be the most common cause of aminotransferase elevations (57.8%), with viral respiratory infections (19.8%) being the most common aetiological factor, followed by acute gastrointestinal infections (10.5%). Rheumatologic, autoimmune and non-alcoholic fatty liver disease also caused elevation of aminotransferases. Another study in 72 young children with isolated elevation of serum aminotransferases for 3-36 months found that the condition was benign, and usually resolved within a year. 10 The authors speculate that such patients could be followed conservatively, and invasive approaches including liver biopsy rarely contribute to the diagnosis.    admitted longer for follow-up. A possible explanation is the availability of follow-up tests in the outpatient department. Infants with elevated aminotransferases also had high white blood cell counts and serum albumin; however, the reason for its occurrence remains to be explored.
Not all infants with an initial aminotransferase elevation had subsequent follow-ups. Only 38.7% of infants with initial elevation had the first follow-up. However, 52.6% of infants with elevation on the first follow-up underwent subsequent follow-ups. The periods between initial laboratory testing and first laboratory follow-ups and between first and second laboratory follow-up were 3.8 ± 3.5 and 8.4 ± 6.6 days, respectively, implying that additional follow-ups occur more infrequently. All infants with first and second follow-up had persistent aminotransferase elevation. It is possible that the follow-up duration was inadequate for observing the normalisation of aminotransferase levels.
Interestingly, similar to the results of previous studies, none of the infants with positive viral PCR had a SBI. 3,21 This study has several limitations. First, the sample size was limited, particularly in view of the variety of viruses; the study results may, therefore, not be extrapolated to a larger population. The recognised PIV1 also demonstrated a relatively low statistical power (α = 0.05, β = 73%) in view of the sample size; there is a possibility that viruses with a true tendency for causing aminotransferase elevation were not detected due to an inadequate sample size. Second, this study is limited by the absence of negative controls. However, the inclusion of the general population with non-specific aminotransferase elevation was not within the scope of this study. In addition, a negative mRT-PCR does not necessarily imply that the infection is non-viral. Therefore, the use of negative controls has its limitations. Further studies using negative controls are warranted, as more precise viral detection diagnostics become available. Nevertheless, this study has comprehensively reported the occurrence of non-specific aminotransferase elevation in febrile infants aged 90 days or younger.
In conclusion, non-specific hepatitis is occasionally observed with RV infection in infants aged 8-90 days. ALT elevation is the most common abnormality, and a final diagnosis of primary hepatobiliary disease is rare. Invasive investigations may be reduced by increasing the cut-off levels of aminotransferase in this paediatric group. Large sample studies are warranted to confirm our findings and evaluate the recommendations.