Addition of platelet-lymphocyte ratio to risk factors to improve the early prediction of acute kidney injury and mortality in critically ill neonates

Introduction

Acute kidney injury (AKI), which is highly prevalent among hospitalized neonates, especially those who are in critically ill, has a negative impact on neonatal prognosis, forcing up the risk of in-hospital mortality and morbidity (1-3). Therefore, early prediction and diagnosis are necessary to prevent AKI. The diagnosis of neonatal AKI depends upon serum creatinine (SCr) and urine output, although they are not early and sensitive AKI markers, especially in neonates (1). Previous researchers have developed biomarkers for predicting AKI in neonates (4,5). However, the accuracy of these biomarkers in the clinical diagnosis, especially in more heterogeneous populations such as neonatal intensive care unit (NICU) patients, has not been clearly defined. None of these biomarkers is currently routinely used in neonatal clinical practice.

Numerous studies have revealed that inflammatory processes mediated by the immune system are crucial in the pathophysiology of AKI (6,7). The inflammatory cells, such as neutrophils and lymphocytes, have been detected in kidneys with AKI in the early stage and are expected to have an important role in the pathogenesis of renal injury (6-8). Therefore, the use of easily available parameters found in a complete blood count (CBC) to provide an early diagnosis of AKI and to predict associated patient outcomes has been considered in multiple settings in adults (9). In addition, since the interactions of neutrophils, lymphocytes, and platelets appear to be a critical step in inflammation, the neutrophil, lymphocyte, and platelet ratio (NLPR), neutrophil-lymphocyte ratio (NLR), and platelet-lymphocyte ratio (PLR) have been calculated and various attempts have been made to associate these ratios with the prediction of AKI and outcomes in adults and children (10-17). In addition to the fact that absolute neutrophil, lymphocyte, and platelet counts in newborns are different from those in children and adults, they also differ between preterm and full-term infants (18-20). The relationship between these ratios and AKI in neonates currently remains undefined. Therefore, evaluation of the association of these ratios with AKI and outcomes in a heterogeneous population of neonates with a wide range of gestational ages would be meaningful.

The aims of this study were to investigate the associations of the NLPR, NLR, and PLR on NICU admission with AKI and mortality and to determine whether these ratios could act as an early predictor in critically ill neonates. We present this article in accordance with the STARD reporting checklist (available at https://apm.amegroups.com/article/view/10.21037/apm-22-1075/rc).

Methods

Participants

A summary patient-level analysis was performed using pooled data from our previously published prospective studies conducted in critically ill neonates (4,21-23). All previous studies, which were originally designed to examine the diagnostic accuracy of urinary biomarkers and enrolled neonates admitted to the NICU prospectively, had clearly defined enrolment and exclusion criteria (4,21-23). The priori exclusion criteria of these previous studies were as follows: parental refusal to participate, neonates with severe congenital malformations, and failure to collect urine samples before discharge from the NICU or death. The parents or guardians of all participants provided written informed consent. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and approved by the Ethics Committee of Children’s Hospital of Soochow University (No. 2015LW005).

Clinical data collection

Neonatal clinical information and maternal information were collected. Clinical laboratory results were obtained during the first 24 h after admission as part of routine care. The score for neonatal acute physiology (SNAP) was used to assess the illness severity (4,21-23).

AKI definition and clinical outcome

Diagnosis of AKI that developed during the first 7 days after NICU admission was consistent with our previous studies (4,23), and was based on (I) the neonatal AKI Kidney Disease: Improving Global Outcome (KDIGO): SCr rise greater than or equal to 0.3 mg/dL (26.5 µmol/L) within 48 h; SCr rise greater than or equal to 1.5 times the reference SCr within 7 days; urine output less than or equal to 0.5 mL/kg/h for 6–12 h (4,23), or (II) SCr greater than 1.5 mg/dL (132.6 µmol/L) sustained for at least 48 h and the mother of the neonates had normal renal function (4,23). The SCr level was routinely measured, and all neonates underwent at least 2 SCr measurements during the first 7 days after admission.

CBC and calculation of the NLPR, NLR, and PLR

A sample of venous blood was routinely collected on admission for the measurement of the CBC using a Sysmex XN-1000 blood cell analyzer (Sysmex Corporation, Tokyo, Japan). In accordance with the previous study (9), the NLPR was defined as the neutrophil, lymphocyte, and platelet ratio and calculated as: (neutrophil count ×100)/(lymphocyte count × platelet count). The NLR was defined as the ratio of neutrophil count to lymphocyte count. The PLR was defined as the ratio of platelet count to lymphocyte count.

Statistical analysis

The data were analyzed using SPSS 16.0 software (IBM Corp., Armonk, NY, USA). Kolmogorov-Smirnov and Levene’s tests were performed to analyze normality and homogeneity of variance. Continuous variables were presented as the median [interquartile range (IQR)] and categorical variables as the total number (percentage). The Mann-Whitney U test and chi-square or Fisher’s exact test were used to analyze the differences between groups. Multivariate linear regression analysis was performed to investigate factors independently associated with these ratios. Continuous variables with skewed distribution were log₁₀ transformed for linear regression, and the multicollinearity was judged by variance inflation factor (VIF) and tolerance values. The VIF and tolerance values of less than 2 and greater than 0.5, respectively, indicated the absence of multicollinearity. To identify independent variables associated with AKI and outcome, univariate and multivariate (forward conditional) logistic regression analyses were performed. The odds ratio (OR), adjusted OR (AOR), and 95% confidence interval (CI) were calculated. Model fit was assessed by the Hosmer-Lemeshow goodness-of-fit test. The area under the curve (AUC) of the receiver-operating characteristic (ROC) curve was performed to assess discriminative ability of variables, and the sensitivity and specificity were calculated. The nonparametric method of Delong in SigmaPlot 10.0 software (Systat Software, Inc., San Jose, CA, USA) was used to compare differences between AUCs. All statistical analyses were 2-sided with an alpha level of 0.05.

Results

Patient characteristics

In total, 442 neonates were included in the study, and the study flow diagram is displayed in Figure S1. Of the neonates, 49 (11.1%) were identified as having AKI during the first 7 days after admission, including 22 with AKI stage 1, 17 with stage 2, and 10 with stage 3. The demographic and clinical characteristics, the first available laboratory results during the first 24 hours after admission, and the levels of NLPR, NLR, and PLR on admission were compared between neonates with and without AKI (Table 1).

In addition, of the neonates, 429 (97.1%) received antibiotics, including cephalosporin, penicillin, meropenem, and vancomycin. None of the neonates received aminoglycosides, indomethacin, or ibuprofen during their NICU stay. Of the 442 neonates, 2 were born from mothers with chronic kidney disease, and 25 were born from mothers who were diagnosed with diabetes mellitus during the pregnancy.

Correlation of the NLPR, NLR, and PLR with clinical variables

To investigate whether these ratios on admission were associated with gestational age, birth weight, sex, and illness severity as assessed by the SNAP, multivariate linear regression analyses were performed after checking the multicollinearity in Table 2. In addition, Spearman’s correlation analysis was used to investigate the correlation of these ratios with clinical factors (Table S1).

Association of the NLPR, NLR, and PLR with AKI

To explore whether these ratios on admission could be used for early prediction of the development of AKI during the first 7 days after admission, the demographic and clinical variables, including gestational age, birth weight, sex, and SNAP, and the first available laboratory results with P less than 0.05 in the comparison of Table 1 were considered confounding factors and analyzed. As shown in Table 3, the final forward multivariate regression model identified birth weight, SNAP, SCr, and PLR as independent factors significantly associated with AKI.

The ability of the PLR to predict AKI is displayed in Table 4. The PLR on NICU admission provided an AUC of 0.62, which was similar to that of SCr (AUC =0.61), for discriminating AKI. When combining PLR and SCr, the performance improved significantly (AUC =0.68), as compared with SCr alone (AUC =0.68 vs. 0.61, P=0.010). The combination of the PLR with birth weight, SNAP, and SCr displayed the best performance (AUC =0.78), which was significantly better than PLR (AUC =0.78 vs. 0.62, P=0.002) or SCr (AUC =0.78 vs. 0.61, P<0.001) alone in the early prediction of AKI.

Association of the NLPR, NLR, and PLR with NICU mortality

Of the 442 neonates, a total of 35 died during NICU stay. The comparisons of the demographic and clinical characteristics, the first available laboratory results during the first 24 h after admission, and the levels of NLPR, NLR, and PLR on admission between survival and non-survival neonates are displayed in Table S2.

To explore whether these ratios on admission were suitable for early prediction of NICU mortality, the demographic and clinical variables, including gestational age, birth weight, sex, and SNAP, and the first available laboratory results with P less than 0.05 in the comparison displayed in Table 1 were considered confounding factors and analyzed. The final forward multivariate regression model identified birth weight, SNAP, and PLR as independent factors significantly associated with mortality (Table 5).

The PLR provided an AUC of 0.63, which was not better than birth weight or SNAP, in predicting mortality (Table 6). The combination of the PLR with birth weight and SNAP displayed the best performance (AUC =0.79), which was significantly better than PLR (AUC =0.79 vs. 0.63, P=0.029) or SNAP (AUC =0.79 vs. 0.69, P=0.019) alone in the early prediction of NICU mortality.

Discussion

Our data demonstrate that the PLR on NICU admission, but not NLPR and NLR, is independently associated with AKI and mortality in critically ill neonates. The potential mechanism of low PLR might be related to decreased platelet production or excessive platelet destruction caused by diseases, therapeutic interventions, inflammation, and hypoxia. To our knowledge, the study was the first to explore the relationship between the NLPR, NLR, and PLR and risk of AKI and mortality in critically ill neonates. The OR for the association of the PLR with AKI and mortality remained significant after adjustment for body weight and illness severity, as assessed by the SNAP, demonstrating that an association between the PLR exists, calculated from CBC obtained on admission, and AKI or NICU mortality in critically ill neonates.

Our observation of NLPR and NLR, which were not associated with AKI and mortality in critically ill neonates, is not in accordance with previous studies (11-13,24). Unlike the PLR, the results of NLPR and NLR were largely related to the absolute neutrophil count. The most likely explanation for this discrepancy between our data and previous data conducted in adults might be that we evaluated the association of the ratios in a general and heterogeneous NICU population. In a prospective cohort of critically ill patients, Han et al. demonstrated a U-shaped relationship between WBC count and risk of AKI and mortality, and proposed that the higher risk of AKI in leucocytosis could be related to neutrophilia, and the higher risk of AKI in leucopenia could be related to lymphopenia and monocytopenia (14). Unfortunately, in critically ill neonates, the counts of neutrophil and lymphocyte are substantially influenced by postnatal age and depend strongly on the underlying conditions and the severity of the illness (18,19).

Interestingly, a negative correlation of the PLR with AKI and outcome was observed in the study, implying that the lower the PLR value, the higher the risk of AKI and mortality in critically ill neonates. Although a high PLR has been associated with an increased risk of AKI or mortality in previous studies conducted in a specific clinical setting, such as in adult patients undergoing contrast-induced AKI (15) or in selected sepsis patients (16), the negative correlation in our study is in line with a previous report conducted on 10,859 critically ill patients with AKI, where a U-shaped relationship was observed between the PLR and increased all-cause mortality, namely, not only high PLR but low PLR is correlated with increased mortality (10). Thus, the correlation between the PLR and the risk of AKI and outcome is diverse, which is determined by the difference in age, underlying conditions, and illness severity, as well as the stage of inflammation. The neonates in this study were critically ill in the early acute stage of illness. The lower PLR might result from the presence of thrombocytopenia that has been demonstrated to be associated with higher AKI and mortality risk in previous studies (25,26).

In this study, the AUC value for the PLR alone to predict AKI was less than 0.70, suggesting that the PLR on admission, similar to SCr, cannot effectively discriminate AKI that developed during the first week after NICU admission by itself in critically ill neonates. The poor results derived from a mixed heterogeneous NICU population might be related to the low specificity of PLR or SCr for AKI. Indeed, we know that SCr levels reflect the creatinine levels of the mother during the first few days after birth and are unstable during the postnatal period, and the percentages of lymphocyte and platelet in the blood routine examination change with postnatal age (18,19). Nevertheless, we identified that the combination of the PLR with birth weight, SNAP, and SCr was much better than any individual marker alone for early prediction of AKI in heterogeneous critically ill neonates. The significant difference in birth weight and SNAP between neonates with and without AKI has been reported by our previous study and in research conducted by others (4,27). Our finding suggests that the additional predictive value of combining PLR on admission with birth weight, SNAP, and SCr improves the quantification of the risk for prediction of neonatal AKI.

A similar situation was observed in prognosis. The combination of the PLR with body weight and SNAP appeared to be the best predictive model for death in critically ill neonates, with an AUC of 0.79. Gestational age, birth weight, and illness severity are well known to contribute independently to the risk of death in critically ill neonates (28,29). In the study, the body weight and SNAP displayed superior predictive performance to that of other perinatal risk factors, which is accordance with our previous study (4). Our data indicate that the PLR, as an early marker calculated based on easily available parameters found in a CBC, could add predictive value to other previously described perinatal risk factors for death.

This study has several limitations. Firstly, AKI diagnosis was made on the basis of SCr and urine output. Although the urine output is of great value in improving the diagnostic rate of AKI, the criteria of urine output for neonatal AKI are still ambiguous or even controversial (2,30). Secondly, the primary aim of the study was to evaluate the early predictive value of these ratios. The CBC was collected for analysis only on NICU admission rather than through serial measurements, which might affect the predictive power of the PLR. Thirdly, the study included neonates from a single center, only 16 (3.6%) neonates were born at less than 29 weeks of gestational age and the illness severity of the neonates assessed by the SNAP was less severe. Our findings may not be generalizable to neonates with extremely low gestational age or with more severe conditions, suggesting that the PLR needs to be validated as a diagnostic and prognostic marker in multiple centers before they can be extrapolated to other populations.

Conclusions

Our study indicates that low PLR on NICU admission is associated with increased risk for AKI and NICU mortality in critically ill neonates. Although the PLR alone is not predictive of AKI and NICU mortally, it adds predictive value to other risk factors for the prediction in the population.

Acknowledgments

We thank the staff in the biochemistry laboratory for their technical assistance.

Funding: This work was supported by grants from the National Natural Science Foundation of China (No. 81971432) and Jiangsu Province Science and Technology Support Program (No. BE2020660). The funders had no role in study design, data collection, preparation of the manuscript, or decision to publish.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://apm.amegroups.com/article/view/10.21037/apm-22-1075/rc

Data Sharing Statement: Available at https://apm.amegroups.com/article/view/10.21037/apm-22-1075/dss

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://apm.amegroups.com/article/view/10.21037/apm-22-1075/coif). YL reports that this work was supported by grants from the National Natural Science Foundation of China (No. 81971432) and Jiangsu Province Science and Technology Support Program (No. BE2020660). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The parents or guardians of all participants provided written informed consent. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013), and approved by the Ethics Committee of Children’s Hospital of Soochow University (No. 2015LW005).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Selewski DT, Charlton JR, Jetton JG, et al. Neonatal Acute Kidney Injury. Pediatrics 2015;136:e463-73. [Crossref] [PubMed]
2. Starr MC, Charlton JR, Guillet R, et al. Advances in Neonatal Acute Kidney Injury. Pediatrics 2021;148:e2021051220. [Crossref] [PubMed]
3. Gallo D, de Bijl-Marcus KA, Alderliesten T, et al. Early Acute Kidney Injury in Preterm and Term Neonates: Incidence, Outcome, and Associated Clinical Features. Neonatology 2021;118:174-9. [Crossref] [PubMed]
4. Li Y, Fu C, Zhou X, et al. Urine interleukin-18 and cystatin-C as biomarkers of acute kidney injury in critically ill neonates. Pediatr Nephrol 2012;27:851-60. [Crossref] [PubMed]
5. Kamianowska M, Szczepański M, Kulikowska EE, et al. The Tubular Damage Markers: Neutrophil Gelatinase-Associated Lipocalin and Kidney Injury Molecule-1 in Newborns with Intrauterine Growth Restriction. Neonatology 2019;115:169-74. [Crossref] [PubMed]
6. Jang HR, Rabb H. Immune cells in experimental acute kidney injury. Nat Rev Nephrol 2015;11:88-101. [Crossref] [PubMed]
7. Juncos LA, Wieruszewski PM, Kashani K. Pathophysiology of Acute Kidney Injury in Critical Illness: A Narrative Review. Compr Physiol 2022;12:3767-80. [Crossref] [PubMed]
8. Bonavia A, Singbartl K. A review of the role of immune cells in acute kidney injury. Pediatr Nephrol 2018;33:1629-39. [Crossref] [PubMed]
9. Gameiro J, Lopes JA. Complete blood count in acute kidney injury prediction: a narrative review. Ann Intensive Care 2019;9:87. [Crossref] [PubMed]
10. Zheng CF, Liu WY, Zeng FF, et al. Prognostic value of platelet-to-lymphocyte ratios among critically ill patients with acute kidney injury. Crit Care 2017;21:238. [Crossref] [PubMed]
11. Fan LL, Wang YJ, Nan CJ, et al. Neutrophil-lymphocyte ratio is associated with all-cause mortality among critically ill patients with acute kidney injury. Clin Chim Acta 2019;490:207-13. [Crossref] [PubMed]
12. Xiao W, Lu Z, Liu Y, et al. Influence of the Initial Neutrophils to Lymphocytes and Platelets Ratio on the Incidence and Severity of Sepsis-Associated Acute Kidney Injury: A Double Robust Estimation Based on a Large Public Database. Front Immunol 2022;13:925494. [Crossref] [PubMed]
13. Bu X, Zhang L, Chen P, et al. Relation of neutrophil-to-lymphocyte ratio to acute kidney injury in patients with sepsis and septic shock: A retrospective study. Int Immunopharmacol 2019;70:372-7. [Crossref] [PubMed]
14. Han SS, Ahn SY, Ryu J, et al. U-shape relationship of white blood cells with acute kidney injury and mortality in critically ill patients. Tohoku J Exp Med 2014;232:177-85. [Crossref] [PubMed]
15. Hudzik B, Szkodziński J, Korzonek-Szlacheta I, et al. Platelet-to-lymphocyte ratio predicts contrast-induced acute kidney injury in diabetic patients with ST-elevation myocardial infarction. Biomark Med 2017;11:847-56. [Crossref] [PubMed]
16. Shen Y, Huang X, Zhang W. Platelet-to-lymphocyte ratio as a prognostic predictor of mortality for sepsis: interaction effect with disease severity-a retrospective study. BMJ Open 2019;9:e022896. [Crossref] [PubMed]
17. Manuel V, Miana LA, Turquetto A, et al. The role of the neutrophil-lymphocyte ratio for pre-operative risk stratification of acute kidney injury after tetralogy of Fallot repair. Cardiol Young 2021;31:1009-14. [Crossref] [PubMed]
18. Amatuni GS, Sciortino S, Currier RJ, et al. Reference intervals for lymphocyte subsets in preterm and term neonates without immune defects. J Allergy Clin Immunol 2019;144:1674-83. [Crossref] [PubMed]
19. Yu JC, Khodadadi H, Malik A, et al. Innate Immunity of Neonates and Infants. Front Immunol 2018;9:1759. [Crossref] [PubMed]
20. Haley KM, Recht M, McCarty OJ. Neonatal platelets: mediators of primary hemostasis in the developing hemostatic system. Pediatr Res 2014;76:230-7. [Crossref] [PubMed]
21. Li Y, Xiao Z, Yan J, et al. Urine erythropoietin level is associated with kidney and brain injury in critically ill neonates. Neonatology 2013;104:87-94. [Crossref] [PubMed]
22. Wang Q, Li M, Li X, et al. Early urinary angiotensinogen excretion in critically ill neonates. J Renin Angiotensin Aldosterone Syst 2015;16:1010-20. [Crossref] [PubMed]
23. Fang F, Hu X, Dai X, et al. Subclinical acute kidney injury is associated with adverse outcomes in critically ill neonates and children. Crit Care 2018;22:256. [Crossref] [PubMed]
24. Manuel V, Miana LA, Jatene MB. Neutrophil-Lymphocyte Ratio in Congenital Heart Surgery: What Is Known and What Is New? World J Pediatr Congenit Heart Surg 2022;13:208-16. [Crossref] [PubMed]
25. Griffin BR, Jovanovich A, You Z, et al. Effects of Baseline Thrombocytopenia and Platelet Decrease Following Renal Replacement Therapy Initiation in Patients With Severe Acute Kidney Injury. Crit Care Med 2019;47:e325-31. [Crossref] [PubMed]
26. Wu M, Luan YY, Lu JF, et al. Platelet count as a new biomarker for acute kidney injury induced by hemorrhagic shock. Platelets 2020;31:94-102. [Crossref] [PubMed]
27. Stoops C, Boohaker L, Sims B, et al. The Association of Intraventricular Hemorrhage and Acute Kidney Injury in Premature Infants from the Assessment of the Worldwide Acute Kidney Injury Epidemiology in Neonates (AWAKEN) Study. Neonatology 2019;116:321-30. [Crossref] [PubMed]
28. Cao Y, Jiang S, Sun J, et al. Assessment of Neonatal Intensive Care Unit Practices, Morbidity, and Mortality Among Very Preterm Infants in China. JAMA Netw Open 2021;4:e2118904. [Crossref] [PubMed]
29. Gebreheat G, Teame H. Survival and mortality of preterm neonates in a neonatal intensive care unit in Northern Ethiopia: a retrospective cohort study. Sci Rep 2022;12:600. [Crossref] [PubMed]
30. De Mul A, Parvex P, Héneau A, et al. Urine Output Monitoring for the Diagnosis of Early-Onset Acute Kidney Injury in Very Preterm Infants. Clin J Am Soc Nephrol 2022;17:949-56. [Crossref] [PubMed]