Authors’ Reply to the Letter to the Editor: “Comprehensive Assessment of Inflammatory Indices to Predict Outcomes in Acute Pancreatitis”

To the editor,

We thank the authors for their interest in our study and the compliments and constructive criticisms made [1]. As they emphasized, we studied two novel inflammatory markers, the systemic immune-inflammation (SII) index and the systemic inflammation response index (SIRI) in patients with acute pancreatitis (AP) [2]. We found that both the SII index and the SIRI increased gradually in parallel to the severity of AP and also the SII index and the SIRI were higher in AP patients who developed acute kidney injury (AKI). We believe that since both the SII and SIRI are cheap biomarkers and also easy to calculate, they may have potential uses in the follow-up and treatment of AP patients. The Pan-Immune-Inflammation Value (PIV) mentioned by the authors in their letter is a novel index which can also easily be calculated using differential cell counts from complete blood count using the formula (neutrophil count × platelet count × monocyte count)/lymphocyte count). The PIV has been studied in several diseases, particularly in cancer patients, but not in patients with acute pancreatitis yet [3–7]. Although the PIV takes all four cell lines into account, it doesn’t necessarily mean that it is a better inflammation marker than the SII and SIRI. But to be honest, the idea of studying PIV in our dataset as suggested by the authors is a really good idea and we thank the authors for their constructive contribution. At the end of the day, we calculated the PIV in the same dataset of our study and searched for its association with AP severity and AKI development. The results were again impressive; The PIV was highly correlated with the SII (r = 0.832, p < 0.001) and the SIRI (r = 0.933, p < 0.001) in AP patients, and a significant increase in the PIV was observed in parallel to the severity of AP. Median PIV values for patients with mild, moderately severe, and severe acute pancreatitis were 489, 1367.3, and 2022.9, respectively (p < 0.001). Not surprisingly, statistical analyses also showed that the PIV was higher in AP patients who developed AKI compared to patients who did not, the same as it was observed with the SII and SIRI. The median PIV was 1537.4 in AP patients with AKI and 499.9 in patients without AKI revealing a statistically significant difference between the groups (p < 0.001). When comparing the value of the PIV with the SII and SIRI, PIV was not superior to both indices to predict disease severity and AKI development. The area under the ROC curve (AUC) to predict severe acute pancreatitis was 0.751 for the PIV, 0.809 for the SII index, and 0.782 for the SIRI. AUC for the PIV was not higher than the SII index or the SIRI (p > 0.05). The AUC in predicting AKI development was 0.747 for the PIV, 0.820 for the SII index, and 0.776 for the SIRI. Again, PIV was not superior to the SII index or SIRI (p > 0.05). The optimum cutoff values and calculated sensitivity and specificity values for the PIV are shown in Table 1.

Table 1. ROC curve of pan-immune-inflammation value in predicting AKI and severe acute pancreatitis.

	Cutoff value	AUC	p	CI 95%	Sensitivity	Specificity	PPV	NPV
AKI-PIV	>902.4	0.747	<0.001	0.697 − 0.793	74.49%	66.24%	68.81%	72.20%
SAP-PIV	>966.6	0.751	<0.001	0.701 − 0.797	80.00%	61.24%	67.36%	75.38%

AKI: Acute kidney injury, PIV: The Pan-Immune-Inflammation Value, SAP: Severe acute pancreatitis, PPV: Positive predictive value, NPV: Negative predictive value.

The authors also suggested that using a statistical modeling system to calculate more sophisticated indices using the cell counts in complete blood cell count parameters, the same used to calculate the SII and the SIRI, would theoretically work better. Although their assumption may be true, this is an area of further research and honestly, it is beyond our statistical knowledge.

Lastly the authors highlighted the analysis of the dynamic changes in the SII and SIRI at different time points during the hospitalization and search for possible associations with clinical improvement or deterioration, duration of hospital stays, and type of therapy as important areas of future research. We again agree with them in this aspect and indeed we partially did it in our research. We calculated the SII index and the SIRI at baseline (on admission) and also at the 24th and 48th hours of admission. The 24th- and 48th-hour SII index and SIRI were increased in concordance with the disease severity and the highest SII was observed at the time of the first admission in mild, moderate, and severe AP groups. The SII index decreased at 24th and 48th hours gradually in all three groups. The same was true for the SIRI in mild and moderate AP groups. But in the severe AP group, the highest SIRI was observed at the 24th hour.

Murat Biyik,Mehmet Asil and Muharrem Keskin
Department of Internal Medicine, Division of Gastroenterology, Necmettin Erbakan University Faculty of Medicine, Konya, Turkey
Zeynep BiyikDepartment of Internal Medicine, Division of Nephrology, Selcuk University Faculty of Medicine, Konya, Turkey