Therapeutic benefits of zinc sulfate on neonatal hyperbilirubinemia

Abdulrahman Al-Matary 1 · Abdullah AlGhamdi 2,3,4 · Bandar Alenaze 2,3,4 · Rasha Abdulaziz Mandili 5 ·Dhuha Abdulaziz Alhawsawi 5 · Duha Magzoub 6 · Ahmed Abu-Zaid 8

Neonatal hyperbilirubinemia is a common phenomenon shortly after birth and a frequent cause of prolonged hospitalization [ 1]. Increased enterohepatic circulation ofindirect bilirubin is one of the elucidated mechanisms implicated in the pathophysiology of neonatal hyperbilirubinemia [ 2, 3].According to various in vitro and in vivo evidence from animal [ 3] and human [ 2] trials, zinc salts have been depicted to reduce bilirubin levels. Mechanistically, it has been shown that zinc sulfate (ZnSO) binds to indirect bilirubin in the intestinal lumen to assemble a non-absorbable ternary complex that will eventually be expelled out into feces, thus contributing to lower total serum bilirubin (TSB) levels [ 2,3]. Boskabadi and colleagues reported that zinc levels were lower in jaundiced infants compared with healthy ones, suggesting that zinc might exert protective roles against neonatal hyperbilirubinemia [ 4]. Several randomized controlled trials (RCTs) scrutinized the utility of ZnSO 4 supplementation in infants with neonatal hyperbilirubinemia [ 5- 15].However, the results were largely conflicting and limited by small sample size. Therefore, we aimed to conduct a systematic review and meta-analysis of all RCTs that examined the efficacy and safety of ZnSOsupplementation in infants with neonatal hyperbilirubinemia.

This investigation was conducted in compliance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [ 16], and the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions [ 17]. Supplementary Table 1 shows the exact syntaxes used for the literature search in five databases from inception until 12-February-2021. All studies that met the following evidence-based participants, intervention, comparison, and outcomes criteria were included: (1) patients:neonates with hyperbilirubinemia; (2) intervention: ZnSO;(3) comparator: placebo or nothing; (4) outcomes: any of the prespecified efficacy and safety endpoints; and (5) study design: RCTs. We excluded non-English articles, non-randomized studies, grey literature, animal studies, and articles with jaundiced neonates having concurrent complications.

Three steps were followed in selecting studies: first, omitting the duplicated citations; second, examining the titles/abstracts and readily excluded the irrelevant ones; third,conducting full-text screening of the remaining citations for final inclusion in the meta-analysis. Data were collected pertaining to baseline characteristics of the included studies,specific efficacy endpoints (namely-TSB levels at different time points after intervention, incidence of hyperbilirubinemia, proportion of neonates requiring phototherapy,and the mean duration of phototherapy), and specific safety endpoints (namely-nausea, vomiting, and skin rash).

The risk of bias of included studies were evaluated in accordance with the Cochrane risk of bias assessment instrument (version 1) [ 18]. For the data analysis, Review Manager Software version 5.4.0 was used for meta-analysis.Data were pooled as weighted mean difference (WMD) or odds ratio (OR), as appropriate, with 95% confidence interval (CI). The between-study heterogeneity was defined as Chi-square

P

＜ 0.1 and

I

-square test (

I

2 ) ＞ 50% [ 19]. The fixed- and random-effects models were used for analysis of homogeneous and heterogeneous data, respectively. Leaveone-out sensitivity analysis was performed to explore the robustness of pooled results, by omitting one RCT at a time and recalculating the overall effect size of the remaining RCTs. Publication bias was assessed qualitatively through visual inspection of funnel plots for asymmetry [ 20], and quantitatively examined through Egger linear regression test [ 20]. For all purposes,

P

values less than ＜ 0.05 were deemed statistically significant.

Supplementary Fig. 1 exhibits the PRISMA flowchart for the literature search. Overall, 11 RCTs were included in the meta-analysis [ 5- 15], which comprised a total of 1218 neonates (ZnSO 4 = 600 and control = 618). Supplementary Table 2 exhibits the detailed baseline characteristics of the included studies. For the risk of bias assessment, one [ 13],five [ 8, 9, 11, 14, 15], and five [ 5, 6, 7, 10, 12] studies were classified as unclear, low, and high risk, respectively. Supplementary Fig. 2 exhibits the risk of bias summary and graph of the included studies.

The pooled results revealed no significant difference between both groups with regard to TSB levels at 1 day(WMD = - 0.49 mg/dL, 95% CI - 1.50 to 0.53,

P

= 0.35),2 days (WMD = - 0.24 mg/dL, 95% CI - 0.68 to 0.2,

P

= 0.29), 4 days (WMD = - 0.28 mg/dL, 95% CI - 0.87 to 0.30,

P

= 0.35), and 6 days (WMD = - 0.27 mg/dL, 95% CI- 0.61 to 0.07,

P

= 0.12) after intervention (Fig. 1). Moreover, the pooled results revealed no significant difference between both groups with regard to the incidence of hyperbilirubinemia after intervention (OR = 1.14, 95% CI 0.74-0.76,

P

= 0.56) and the requirement for phototherapy after intervention (OR = 0.92, 95% CI 0.58-1.47,

P

= 0.56).Nevertheless, the pooled results revealed a significantly lower duration of phototherapy in favor of the ZnSO 4 group compared with the control group (WMD = - 7.49 hours,95% CI - 14.36 to - 0.64,

P

= 0.03) (Fig. 2). The pooled results also revealed no significant difference between both groups with regard to the frequency of vomiting (OR = 0.64,95% CI 0.22-1.84,

P

= 0.41), diarrhea (OR = 1.20, 95% CI 0.35-4.09,

P

= 0.77), and skin rash (OR = 1.30, 95% CI 0.31-5.48,

P

= 0.72) (Fig. 3). Overall, the included RCTs qualitatively reported that ZnSOwas safe and did not culminate in major adverse events [ 5- 15].

Fig. 1 Forest plot showing meta-analysis of the total serum bilirubin level after 1, 2, 4, and 6 days after intervention between zinc sulfate and control groups. SD standard deviation, CI confidence interval

Fig. 2 Forest plot showing meta-analysis of the incidence of hyperbilirubinemia ( a), requirement for phototherapy ( b), and duration of phototherapy ( c) between zinc sulfate and control groups. SD standard deviation, CI confidence interval

The results of leave-one-out sensitivity analysis demonstrated robustness of data for all outcomes, expect TSB level at 1 day and duration of phototherapy after intervention (Supplementary Figs. 3, 4, 5). In addition,the results revealed no publication bias for all outcomes(Supplementary Figs. 6, 7, 8). The subgroup analyses of effi-cacy endpoints based on the quality of studies are depicted on Supplementary Figs. 9, 10.

In our meta-analysis, ZnSOreduced the duration of phototherapy. This is clinically significant as prolonged exposure to phototherapy is associated with unfavorable aftermaths, such as potential DNA damage as well as increased worries pertaining to patent ductus arteriosus, dehydration,extended hospital stay, and disrupted neonate-mother bonding [ 1]. Nevertheless, our meta-analysis showed that ZnSOfailed to reduce the TSB level and the requirement for phototherapy. Potential illuminations for these observations could be ascribed to probable absorption of ZnSOin the proximal intestine, thus resulting in its disadvantageous unavailability in the distal intestine to bind indirect bilirubin and prevent its enterohepatic circulation [ 11]. Moreover, the shorter duration of therapy and early measurements of TSB levels could be additional illuminations as to why no significant reduction in TSB levels was observed. The latter is supported by the study of Elfarargy et al. which depicted that there was no substantial difference in TSB levels between ZnSOand placebo group during the first 7 days of intervention, however, the TSB levels were significantly lower in favor of the ZnSOversus placebo group during day 8 to day 10 after intervention [ 13].

Fig. 3 Forest plot showing meta-analysis of the adverse events between zinc sulfate and control groups. CI confidence interval

Our study possesses several strengths. First, our metaanalysis thus far offers the most comprehensive and up-todate evidence on the therapeutic effect of ZnSOon neonatal hyperbilirubinemia. All outcomes, except TSB level and phototherapy duration, showed homogeneity, supporting the reliability of the withdrawn conclusions. Lastly, we performed extensive leave-one-out sensitivity analysis, subgroup analysis based on quality of studies, and publication bias analysis to strengthen the methodological rigor of the present investigation.

Nevertheless, our study equally possesses limitations that ought to be acknowledged. First, this research is based on only 11 RCTs, the vast majority of which had sample sizes less than 120 (

n

= 8). Moreover, the included studies had variable demographics (e.g., gestational age, birth weight, and mode of delivery) and zinc salt supplementation schedules (e.g., dosage, duration, and time of administration). Moreover, the compliance of ZnSOand its level measurement before and after intervention were not reported. All in all, these factors could have somehow introduced bias in the pooled outcomes. Besides, this research excluded non-English articles, in addition to grey literature,as we wanted to include only published RCTs that have undergone peer-review. Therefore, publication bias could have been introduced in our analysis. Lastly, despite not mandatory, this research investigation did not use medical subject heading (MeSH) search terms in PubMed database,and did not have a prior review protocol and registration in PROSPERO.

In view of the lack of significant reduction in TSB levels after zinc salt supplementation, future search directions include examining whether long-term use of zinc supplementation (＞ 7 days) or different zinc salt formulations are likely to offer therapeutic benefits. Additional prospective research include examination of the dose-response relationship between ZnSOand the extent of reduction in TSB level. Moreover, it is critical to identify which groups of neonates are more likely to benefit the most from ZnSO 4 supplementation to prevent and treat neonatal hyperbilirubinemia. Lastly, large-sized RCTs are needed to validate our meta-analysis findings.

In conclusion, this meta-analysis highlights that ZnSOis safe and may substantially reduce the duration of phototherapy in infants with neonatal hyperbilirubinemia. However, it is not associated with reduced TSB levels, decreased occurrence of neonatal hyperbilirubinemia, or less requirement for phototherapy. These conclusions should be interpreted with caution due to key limitations, including the small number of meta-analyzed studies, the high risk of bias of some included RCTs, and the potential publication bias arising from language bias and exclusion of grey literature.

Supplementary Information

The online version contains supplementary material available at https:// doi. org/ 10. 1007/ s12519- 022- 00532-6.

Author contributions

AMA contributed to conceptualization, data curation, formal analysis, investigation, validation, writing of original draft, reviewing and editing. AGA, AB, MRA, ADA, and MD contributed to data curation, investigation, validation, reviewing and editing.AZA contributed to conceptualization, data curation, formal analysis,investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing of original draft,and reviewing and editing. All the authors approved the final version of the manuscript.

Funding

None.

Data availability

The original contributions presented in the study were included in the article and supplementary materials. Further inquiries can be directed to the corresponding author.

Declarations

Ethical approval

Not applicable.

Conflict of interest

No financial or non-financial benefits have been received or will be received from any party related directly or indirectly to the subject of this article. The authors have no conflict of interest to declare.