Discussion
To the best of our knowledge, the present study demonstrated for the first time using a large cohort of patients with KD in multicenter that pretreatment PNI was negatively associated with biomarkers of immunity/inflammation (NLR and PLR) and positively associated with age, platelet, serum sodium. Low PNI was significantly associated with higher levels of CRP and higher the incidence of IVIG resistance. As with NLR and PLR, PNI was an independent predictor of IVIG resistance. Pretreatment PNI was not inferior to NLR and PLR for predicting IVIG resistance.
In the present study, the percentage of IVIG-resistance was 14.9%, which was far higher the 6.25% determined by Xu-Hai Tan et al[7] and was consistent with the 10-20% reported by Fukunishiet al [21] and Sleeper et al [22]. The discrepancy could be attributed to the different study populations, the sample size and the definition of IVIG-resistance. IVIG-resistant KD has been observed to be associated with a higher incidence of CALs, which is detrimental to the outcome of KD. Early additional treatment before the initial use of IVIG could decrease the occurrence of CALs in IVIG resistant KD. Therefore, it is essential to investigate simple and feasible indictors for IVIG resistance.
Consistent with previous data [4-6], the markers of inflammatory and immunological activation, NLR and PLR, were associated with IVIG resistance in the present study. Systematic immunonutritional status has been refined by introduction of PNI, a continuous variable based on serum albumin concentration and total lymphocyte count in peripheral blood, which was associated with the mortality and morbidity in patients with cardiac diseases [23], acute kidney injury [24] and cancer [10]. The present study revealed that low PNI was significantly associated with high levels of CRP and low counts of platelet. Furthermore, an inverse correlation between PNI and the biomarkers of immunity/inflammation (NLR and PLR) was observed in the present study, which was in accordance with patients with osteosarcoma [10]. PNI was also demonstrated to be correlated with disease activity in patients with systemic lupus erythematosus [25, 26]. Besides, several previous studies reported that hyponatremia was observed in children with KD, particularly children with IVIG resistant KD [27-29] and was associated with the extent of inflammation [30, 31]. In line with these studies, low PNI was demonstrated to be closely associated with low serum sodium in the present study. These results clearly convey the notion that PNI may be associated with IVIG resistance.
PNI brings two important measurements together. On the one hand, serum albumin is known as an indicator of host inflammatory and nutritional status [8], and its levels are associated with higher mortality in several different types of tumor. Hypoalbuminemia was observed in IVIG resistant KD [32] and has been confirmed to be also closely associated with inflammatory markers, such as NLR and PLR. Additionally, research has shown that malnutrition was associated with the immune-suppressed condition [33, 34], impaired respiratory function, and poor wound healing [24], as well as increased risks of postoperative morbidity and mortality and prolonged hospital stays[24]. On the other hand, lymphocytopenia is regarded as the marker of a depressed cell-mediated immunity [35], which can lead to the development of systemic inflammatory response syndrome and is closely associated with immunity and inflammation. It has been reported that critically ill patients with shock and sepsis had marked lymphocytopenia, and that the severity of the clinical course was correlated with the divergence of the lymphocyte (lower) counts [35]. Lymphocytopenia was also associated with decreased survival in advanced carcinoma, sarcoma and lymphoma [36], similar to hypoalbuminemia. Numerous studies from large samples of multiple research centers in Asia have indicated that lymphocyte counts were significantly decreased in IVIG resistant KD when compared with IVIG responsive KD [7, 37, 38]. In the present study, low PNI was associated with decreased albumin and/or lymphocytes. The patients with lower PNI had significantly higher the incidence of IVIG resistance in the present study, which was consistent with previous findings [18]. According to these results, PNI, combined with albumin and lymphocyte levels, could be a superior index that reflects inflammation, immune system and nutritional status in IVIG resistant KD.
Valid prediction of the IVIG-treatment response is crucial to the patients with KD. A number of clinical and inflammatory indexes (age <12 months, platelet <300 ×109/L, higher total bilirubin levels and CRP levels, serum sodium levels, NLR and PLR) for predicting IVIG resistance have been reported [3, 7, 37, 39]. The results of the present study further confirmed those indictors (age <12 months, serum sodium, CRP, NLR and PLR), implying that the results of the present study are convincing. Regretfully, the aforementioned parameters had relatively low predictive power for IVIG resistance according to previous investigations. The prognostic value of PNI has been investigated in certain types of human cancer [9, 10], postoperative septic complications, ST segment elevation myocardial infarction [14]. PNI can also predict the clinical outcome of the pediatric population in the intensive care unit after cardiac operation [40].Accordingly, both univariate and multivariate analyses suggested that pretreatment low PNI was an independent predictor for IVIG resistance in the present study. Although the discriminatory ability of NLR and PLR to predict IVIG resistant KD has been demonstrated in previous reports [5, 6, 37], as well as in the present study, pretreatment PNI having the highest AUC was not inferior to NLR and PLR as an indicator for predicting IVIG resistance determined by ROC analysis. The investigation performed by Miriliet al [41] in malignant melanoma was in accordance with the results of the present study. In addition, pretreatment NLR, PLR and PNI are emerging as predictors of the treatment response to neoadjuvant chemoradiotherapy in rectal cancer [42]. Taken together, PNI is a potential as a candidate for predicting IVIG resistance. As PNI is routinely available and can be measured accurately, it may become a useful and inexpensive approach to the IVIG-treatment response prediction in KD. Accordingly, it has been suggested that the patients with KD having lower PNI levels may be a potential candidate for an aggressive treatment strategy. Nevertheless, further studies should be warranted to verify the findings.
However, it is necessary to recognize limitations in the present study. First, the results have the potential to be biased in terms of the population choice. Second, although a multivariable analysis was used, the possibility of residual unmeasured covariables that may influence the outcomes cannot be excluded. For example, genetic, ethnic and regional factors, and IVIG metabolic factors have not been taken into account. Finally, the data sample used in the present study remains relatively small and only located in western China. Thus, these limitations can potentially limit the accuracy of the results in the present study. Therefore, the ability and accuracy of the PNI in predicting IVIG resistance should be further validated by performing additional prospective, multicenter studies.