Introduction and objective

Sepsis is a life-threatening organ dysfunction deriving from a dysregulated host response to infection [1]. The incidence of sepsis and septic shock have risen continually since the first consensus definition in 1991, reaching approximately 49 million cases worldwide in 2017, with 11 million sepsis-related deaths. These data led to sepsis being declared a global health priority by the World Health Organization (WHO) [2]. Early diagnosis and appropriate treatment in the first hours following the development of sepsis and septic shock improve outcomes [1]. Testing other methods of showing systemic inflammation for identifying high-risk septic shock patients may therefore be a useful guide. From the pathogenetic perspective, sepsis is currently regarded as the result of several mechanisms involving numerous pro- and anti-inflammatory mediators on a simultaneous basis [3]. The pan-immune-inflammation value (PIV) [neutrophil count (103/mm3) × platelet count (103/mm3) × monocyte count (103/mm3)]/lymphocyte count (103/mm3), a novel biomarker that combines neutrophil, platelet, monocyte, and lymphocyte counts, some of the most widespread indicators of systemic inflammation, is an index capable of evaluating patients’ immune and inflammatory status [4, 5]. PIV has recently begun being employed as an inflammatory markers for various. Studies have shown that a high PIV indicates poor prognosis in cancer patients [6], is a reliable predictor of clinical outcomes in cancer patients [7], constitutes a new prognostic biomarker in patients with metastatic colorectal cancer [8, 9], and is associated with increased mortality in hypertensive patients [5].

To the best of our knowledge, no previous studies have investigated the relationship between PIV and mortality and prognosis in patients with septic shock. The purpose of this study was to determine whether PIV can predict mortality and prognosis in patients admitted to the intensive care unit (ICU) with septic shock. In addition, PIV was compared with the Sequential Organ Failure Assessment (SOFA) score, the Acute Physiology and Chronic Health Evaluation II (APACHE II) score, lactate, C-reactive protein (CRP), and procalcitonin (PCT), routinely employed in clinical practice, to determine which may best predict mortality in septic shock.



This prospective study was performed in patients aged 18 or over who were diagnosed with septic shock while being followed up in the tertiary ICU of Marmara University Hospital between July 2022 and January 2023, and in whom informed and written consent was obtained from themselves or their relatives. The research was conducted in conformity with the Declaration of Helsinki and Best Clinical Practice guidelines. Approval was granted by the Marmara University Medical Faculty clinical research ethical committee (no: 09.2022.1023). The complete blood count (CBC) was obtained in patients who developed septic shock while hospitalised in the ICU. Diagnosis of septic shock was based on Surviving Sepsis Campaign criteria [1]. Patients with hematological diseases and neutropenia were excluded. CBC was performed to evaluate the patient’s neutrophil, platelet, monocyte, and lymphocyte counts. PIV was calculated from those data. For CBC measurement, blood specimens collected from the catheter in the patient’s radial artery were placed into ethylene diamine tetra-acetic acid (EDTA)-containing tubes. The washing solution was removed from the arterial catheter before blood sampling. Ninety-three patients were admitted to intensive care during the study period. However, clotting occurred in four blood specimens, four patients were neutropenic, and three patients had hematological diseases. These were excluded from the study, which was thus completed with 82 patients with septic shock (Fig. 1).

Fig. 1
figure 1

Study design flowchart

The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) case–control checklist was used in the writing of this article [10]. Age, sex, BMI, comorbid diseases, lactate, neutrophil, monocyte, platelet, lymphocyte, PIV, CRP, and procalcitonin values, Glasgow Coma Scale (GCS) and sequential organ failure assessment (SOFA) scores, and Acute Physiology and Chronic Health Evaluation II (APACHE II) values were recorded at the time of diagnosis of septic shock. PIV was calculated using the formula [neutrophil count (103/μL) × platelet count (103/μL) × monocyte count (103/μL)]/lymphocyte count (103/μL) [11]. Whether a correlation existed between PIV values and mortality was investigated. The patients were followed-up at least 28 days from admission to the ICU.

Sample size

The sample size estimated based on high- and low-PIV groups’ comparison of survivor functions. The median PIV value used as the cut-off value to form the groups, and the patients were allocated on 1:1. The estimated total sample size was 74 patients when the high- to low-PIV groups had hazard ratio (HR) of 2.0 based on previous studies on different fields [6, 11,12,13,14] along with alpha level of 0.05 and power of 80% (Stata Statistical Software: Release 15. College Station, TX: Stata Corp LLC).

Statistical analysis

Descriptive statistics were presented as frequency (%) for categorical variables, and as median (IQR) and mean (± SD) for continuous variables. Normal distribution assumptions were assessed using the Shapiro–Wilk test. The median PIV value was adopted as the cut-off between low and high PIV. Patient characteristics were compared between the low- and high-PIV groups using the Mann–Whitney U test and Pearson’s chi-square test. Cut-off values of the laboratory parameters predicting ICU mortality were determined using receiver operating characteristics (ROC) curve analysis and the maximum Youden Index value. Overall survival (OS) time and survival probability were estimated with the Kaplan–Meier method. The survival curves of the low- and-high PIV groups were compared using the Wilcoxon (Breslow) test. The clinical and laboratory parameters predicting ICU survival time were examined with the multivariate Cox proportional hazards (CPH) model, enter method. Schoenfeld’s residuals were investigated for proportional hazards assumption. Statistical significance was set at 0.05 level. Statistical analysis was performed on Jamovi version 2.3 software (The Jamovi project (2022) and R Core Team (2021)), and STATA 15 (Stata Statistical Software: Release 15. College Station, TX, USA: Stata Corp LLC.) for testing the proportional hazards assumption.


Patient characteristics

The clinical and laboratory parameters of the 82 ICU patients are presented in Table 1. The patients’ mean age was 62.63 (± 18.10) years, and 52.44% were male. Mean BMI was 28.86 (± 12.08). Twelve (14.63% patients were diagnosed with malignancy, 23 (28.05%) with respiratory diseases, 10 with (12.20%) operational diseases, and 37 (45.12%) with other diseases. Comorbidity was present in 58 (70.73%) patients. Hypertension (HT) was present in 35 (42.68%) patients, diabetes mellitus (DM) in 18 (21.95%), and cardiovascular disease (CVD) in 17 (20.73%) patients. Mean and median GCS, SOFA, and APACHE II, and lactate values, neutrophil, monocyte, platelet, and lymphocyte counts, and PIV, CRP and PCT values are presented in Table 1. The mortality rate was 39.02% (n = 32).

Table 1 Clinical and laboratory parameters

Clinical and laboratory parameters of the Low- and High-PIV Groups

Clinical and laboratory parameters were compared between the low- and high-PIV groups (Table 2 and 3). Median age, presence of HT, and APACHE II, neutrophil, monocyte and platelet values were lower in the low-PIV group than in the high-PIV group (p < 0.05).

Table 2 Clinical and laboratory parameters of the Low- and High-PIV Groups
Table 3 CBC Counts of the Low- and High-PIV Groups

Cut-off values of the laboratory parameters

The cut-off values of the laboratory parameters predicting ICU mortality and the test performance indicators are presented in Table 4. The highest area under the ROC curve (AUC) values were estimated for SOFA (0.94 (0.89 – 0.99)), followed by GCS (0.81 (0.70 – 0.91)), APACHE II (0.80 (0.69 – 0.91)) and lactate (0.77 (0.67 – 0.88)). The AUC values for CRP (0.64 (0.52 – 0.76)) and PCT (0.67 (0.55 – 0.79)) were lower than 0.70. The classification of the mortality status was not statistically significant for PIV (AUC (95% CI) = 0.58 (0.44 – 0.72); p = 0.227).

Table 4 Cut-off values for the laboratory parameters

Survival analysis

The OS values of the ICU patients are shown in Table 5 The OS rate among all the ICU patients was 61%, and the median survival time was 18 (8.65 – 27.35) days. Kaplan–Meier survival curves for all the ICU patients are shown in Fig. 2. Median survival was longer in the low-PIV group than in the high-PIV group (28 (15.25 – 40.76) vs 16 (9.46 – 22.55) days, respectively, p < 0.05). The survival probability for 28-day was 39.56% in all patients, while it was 42.70% and 37.30% in low- and high-PIV groups, respectively (Table 5). Kaplan–Meier survival curves for the low- and high-PIV groups are shown in Fig. 3.

Table 5 Overall Survival of the ICU patients
Fig. 2
figure 2

Kaplan–Meier Survival Curve Estimate for Overall Survival in the ICU

Fig. 3
figure 3

Kaplan–Meier Survival Curves of Overall Survival in the Low- and High-PIV Groups

Clinical and laboratory parameters associated with ICU mortality in the septic shock patients were investigated with the univariate CPH model presented in Table 6. High PIV (HR = 2.13 (1.03—4.38)), low GCS (HR = 3.31 (1.34 – 8.15)), high SOFA (HR = 9.41 (2.86 – 30.95)), high APACHE II (HR = 3.08 (1.47 – 6.45)), high lactate (HR = 6.56 (2.73 – 15.75)), and high PCT (HR = 2.73 (1.11 – 6.69)) were associated with decreased survival time (p < 0.05). In the multivariate model showed the age-adjusted risk estimates for these six laboratory parameters. High lactate (HR = 7.97 (2.19 – 29.08)) and high SOFA (HR = 4.85 (1.22 – 19.32)) were significantly associated with declined survival time (p < 0.05). Patients with high lactate levels (≥ 2.05) had 7.97 times higher risk of mortality than those with low lactate levels. Similarly, patients with high SOFA scores (≥ 8.50) had 4.85 times higher risk of mortality than those with low SOFA scores (Table 6).

Table 6 Cox Proportional Hazard Model Predicting ICU Mortality


Sepsis and septic shock are important health problems that affect millions across the world every year and that result in mortality in one in three and one in six, respectively, of those affected [1]. In the face of this significant risk of mortality the screening of patients with septic shock and early intervention are of critical importance. New diagnostic methods for identifying patients with septic shock may therefore need to be tested. The PIV, one of these novel methods, began employed as an inflammatory marker in recent years. The PIV is particularly advantageous as a low-cost, simple, and easily available parameter obtained from complete blood count tests CBC. Research has shown that sepsis and septic shock are an immune and inflammatory disease [15]. The PIV is an index that combines neutrophils, platelets. Monocytes, and lymphocytes, and has been regarded as representing a comprehensive evaluation of immune and inflammatory conditions in previous studies evaluation [5, 9]. Thrombocytosis emerges as the result of the stimulation of megakaryocytes by proinflammatory cytokines [16, 17]. Platelets are not solely associated with thrombosis, but also trigger and exacerbate inflammation by combining with endothelial cells and encouraging leukocyte migration and adhesion. Monocytes reflect the function of macrophages, and the numbers of these proinflammatory cells generally indicate the patient’s immune and inflammatory status [18, 19]. Since the reactions of leukocytes in circulation to various inflammatory events is generally characterized by an increase in neutrophil numbers and a decrease in lymphocytes, the ratio between them is employed as an inflammatory marker in the clinical setting in the ICU [20, 21]. The PIV appears to be a powerful indicator of mortality in patients diagnosed with ST-segment elevation myocardial infarction (STEMI) examined retrospectively [5, 22]. A study of patients with local advanced head and neck squamous cell carcinoma (HNSCC) observed shorter overall survival (OS) and disease-free survival in patients with high PIV values. The results suggested that PIV scores may represent a prognostic biomarker in HNSCC [23]. A meta-analysis involving patients with malignant tumors showed that patients with higher PIV values were at a significantly higher risk of mortality than those with low PIV scores (HR = 2.00, 95% CI: 1.51–2.64, p < 0.001) [6]. Patients with advanced triple-negative breast cancer (aTNBC) with high PIV values experienced poorer OS [adjusted hazard ratio (HR): 4.46, 95% confidence interval (CI): 2.22–8.99; adjusted p < 0.001] [12]. In a retrospective study of breast cancer patients with a PIV cut-off value of 310.2 [24], five year OS rates in the low- and high-PIV groups were 71.55% and 62.50%, respectively (hazard ratio (HR): 1.737, 95% CI: 1.096–2.755, log-rank test, p = 0.016) [5]. At multivariate Cox hazard model analysis of patients diagnosed with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) [13] whose medical records were investigated PIV ≥ 1011.3 (HR 2.689, 95% CI 1.156, 6.254) emerged as a significant and independent risk factor for all-cause mortality. PIV predicted mortality in these studies, the majority of which were conducted retrospectively.

In this prospective study, we aimed to examine the role of PIV for predicting the mortality and prognosis in ICU patients with septic shock. The median PIV value used for formation of the low- and high-PIV groups, and clinical and laboratory parameters of eighty-two patients were compared between the groups. The median age, presence of HT, APACHE II score, and neutrophil, monocyte, and platelet counts were significantly lower in the low-PIV group than the high-PIV group. This finding is in accordance with the literature except for APACHE II score [5]. There was no significant difference observed in lymphocyte counts. The similar lymphocyte level between low- and high-PIV groups may be related with PIV’s inability to predict the mortality in multivariate analysis. The prognostic PIV value estimated with ROC analysis showed that the AUC value was insignificant, thus it did not provide sufficient evidence to classify the mortality. When the survival analysis rerun with the prognostic value, it showed similar results with the median PIV value findings presented in this study. The median survival time was significantly higher in the low-PIV group than the high-PIV group, this finding was in accordance with the few studies [23,24,25,26]. The univariate Cox PH model showed that the high PIV value was a risk factor in survival, which was similar to those of previous studies [24, 26].

The multivariate analysis indicated that the lactate elevation and high SOFA scores associated with risk factors of survival, while PIV was an insignificant risk factor. The insignificance of high-PIV as a risk factor might be due to the confounding effect of age and other inflammatory factors that were included in the multivariate analysis, since these parameters were correlated with each other, and high PIV was also correlated with elevated APACHEE II level and older age. The insufficient sample size in the multivariate analysis could be another factor contributing to lack of evidence in demonstrating the association of PIV and mortality. According to our knowledge, this is the first study investigating PIV in the septic shock patients.


In conclusion, our finding showed that PIV could significantly predict the median survival time difference in patients with septic shock. Although PIV is capable of showing inflammation, it is not associated with mortality in the multivariate analysis. Further studies are needed to better understand the relationship between PIV and mortality in the septic shock patients.