Proportion of Neutrophils in White Blood Cells as a Useful Marker for Predicting Bacteremic Acute Cholangitis

Objective A positive hemoculture in acute cholangitis is serious, but a blood culture result cannot be obtained at the initial diagnosis and so cannot be used for the severity assessment and decision-making concerning urgent/early biliary drainage. Accordingly, a predictor for bacteremia at the initial diagnosis of acute cholangitis would be particularly useful. We investigated the association between neutrophil proportions in white blood cell counts (%Neutro) and bacteremic acute cholangitis. Methods Of 166 patients with acute cholangitis who were diagnosed with the Tokyo Guidelines 2018/2013 from April 2015 to March 2017, a total of 94 underwent blood culture assessments and were divided into those with a positive hemoculture (n=48) and a negative hemoculture (n=46) and then compared. A receiver operating characteristic curve analysis was used to evaluate the predictive ability of %Neutro and other inflammatory markers. Results The %Neutro values were significantly higher in the positive hemoculture group than in the negative hemoculture group (91.7±4.0% vs. 82.5±9.0%, p<0.0001). A cut-off %Neutro value of 89.7% was strongly associated with bacteremia (area under the curve 0.86, sensitivity 77.1%, specificity 80.4%). A %Neutro of ≥89.7% was a predictor of a positive hemoculture in univariate (p<0.0001) and multivariate analyses (p<0.001). Patients with a %Neutro ≥89.7% needed early biliary drainage more frequently than others (30/46, 65.2% vs. 18/48, 37.5%, p=0.0063). Conclusion %Neutro is an independent predictor of bacteremia in patients with acute cholangitis and may contribute to decision-making concerning early biliary drainage.


Abstract
Background A positive hemoculture in acute cholangitis is serious, and accordingly, a predictor for bacteremia at initial diagnosis would be particularly useful. We investigated the association between neutrophil proportion in the white blood cell count (Neut%) and bacteremic acute cholangitis.

Methods
In 166 patients of acute cholangitis diagnosed with Tokyo guideline 2013 from April 2014 to March 2015, ninety-four patients underwent blood culture assessment and were divided into a positive hemoculture (n = 48) and negative hemoculture (n = 46) group and these two group were compared. A receiver operating characteristic curve analysis was used to evaluate the predictive ability of Neut% and other in ammatory markers.

Conclusions
Neut% is an independent predictor of bacteremia and prognosis in patients with acute cholangitis and may contribute to decision-making of early biliary drainage.

Background
Acute cholangitis can be fatal and requires biliary decompression at an appropriate time for recovery [1]. A positive hemoculture is a predictive factor for prognosis of acute cholangitis [2][3][4]. Unfortunately, a blood culture result cannot be instantly obtained upon admission and cannot be used to assess the severity of disease or to help in the decisionmaking process for urgent or early biliary drainage. Therefore, an excellent predictor of bacteremia would be useful for patients with acute cholangitis. Some biomarkers have been reported to predict prognosis and bacteremia in patients with acute cholangitis; for example, serum procalcitonin (PCT) levels [6][7][8]. Although, to the best of our knowledge, a credible predictive biomarker for bacteremia in patients with cholangitis has not yet been con rmed.
In our current practice, we have noticed that the neutrophil proportion in the white blood cell count (Neut%) is frequently higher in patients with a positive hemoculture than those with a negative hemoculture in acute cholangitis. Accordingly, we retrospectively investigated the association between Neut% and bacteremia and prognosis in patients with acute cholangitis.

Methods
This retrospective study included 166 patients diagnosed with acute cholangitis according to the Updated Tokyo Guidelines for acute cholangitis and acute cholecystitis 2013 (TG13) [9] between April 2014 and March 2015 at the National Hospital Organization Kure Medical Center and Chugoku Cancer Center. The number of patients who underwent hemoculture without pre-administration of antibiotics at the time of the rst medical examination was 115. Of these patients with acute cholangitis, we analyzed 94 patients who did not have any factors affecting white blood cell (WBC) count, serum C-reactive protein (CRP) level, body temperature, and Neut% (Fig. 1). This study was performed in accordance with the Declaration of Helsinki and was approved by our ethics committee Approval No. 2020-20 . Patients were not required to give informed consent to the study because the analysis used anonymous clinical data that obtained after each patient agreed to receive treatment for acute cholangitis. For disclosure, the details of study are posted on some walls in National Hospital Organization Kure Medical Center and Chugoku Cancer Center.

Examination on admission
All patients were checked for vital signs and underwent abdominal computed tomography scans. Blood samples were obtained on admission. Blood culture was assessed and a differential WBC count, including Neut%, CRP, PCT, and other laboratory data, was undertaken. The hematology tests were performed with a Sysmex Hematology XN Modular System (Sysmex Co. Ltd., Hyogo, Japan) [10].
Diagnosis of acute cholangitis, severity assessment, and indication for urgent or early biliary drainage TG13 was used for diagnosis and severity grading of acute cholangitis (Supplementary Tables 1, 2). Acute cholangitis was de ned using suspected or de nite criteria. Severity criteria involved organ dysfunction (hypotension, disturbance of consciousness, respiratory dysfunction, oliguria or serum creatinine> 2.0 mg/dl, prothrombin time international normalized ratio> 1.5, and platelet count< 100,000/mm 3 ). Moderate cholangitis-considered to pose a risk for severe cholangitis without early biliary drainage-was de ned in the presence of more than two of the following ve items: abnormal WBC count (> 12,000/mm 3 or < 4,000/mm 3 ), high fever (≥ 39℃), age≥ 75 years, hyperbilirubinemia (Total bilirubin≥ 5.0mg/dL), and hypoalbuminemia (Albumin < STD×0.7=2.92 at our institute) (Supplementary Table 2). Blood cultures were obtained twice in one session at initial diagnosis and one or two positive results were de ned as a positive hemoculture and the presence of skin-resident bacteria was treated as a negative hemoculture. We de ned the timing of biliary decompression as 'urgent' within 6 hours of admission and 'early' within 24 hours. Fundamentally, we performed urgent biliary drainage for severe cholangitis and early biliary drainage for moderate cholangitis using an endoscopic retrograde cholangiopancreatography technique.

Analysis of characteristics of patients with positive and negative hemoculture
We divided 94 patients into a positive hemoculture group and negative hemoculture group and compared the Neut%, WBC, CRP, PCT, and other characteristics.

ROC curve analysis
Receiver operating characteristic (ROC) curve analysis was performed to evaluate the ability of Neut%, WBC, CRP, and PCT to predict the presence of bacteremia and level of severity in patients with acute cholangitis.

Predictor for bacteremia in acute cholangitis
We compared the ability of Neut%, PCT, and other factors associated with the severity assessment criteria in TG13 to predict bacteremia.

Prognosis according to Neut%
We divided patients into higher Neut% group and lower group and compared their prognosis.

Usefulness of Neut% in mild acute cholangitis
Finally, we divided patients with mild cholangitis into a positive hemoculture and negative hemoculture group and evaluated the ability of Neut% to predict bacteremia in mild cholangitis.

Statistical analysis
A Fisher's exact test and χ 2 test were used for categorical variables and the Welch t-test and Wilcoxon signed-rank test were used for quantitative data, where appropriate. A binomial regression analysis was performed to identify independent predictors of positive hemoculture. To evaluate the usefulness of Neut%, WBC count, CRP, and PCT for prediction of positive hemoculture, ROC curves were plotted and areas under the curve (AUCs) were calculated with 95% con dence intervals (CI). Statistical evaluations of the AUC curve and the differences between each biomarker were performed using an χ 2 test. The optimal cutoff values for Neut% on the ROC curve were determined on the basis of the Youden index. A P value of <0.05 was regarded as signi cant. The odds ratios were reported together with their 95% con dence limits. All statistical analyses of recorded data were performed using the Excel statistical software package (Ekuseru-Toukei 2015 version; Social Survey Research Information Co., Ltd., Tokyo, Japan).

Patient characteristics
Forty-eight patients out of 94 had a positive hemoculture and 46 were negative (Fig. 1). The prevalence of species in a positive hemoculture and patient characteristics are shown in Table 1, 2, respectively. Mean age, body temperature, and heart rate were signi cantly higher in patients with a positive hemoculture than those with a negative hemoculture.
Platelet count and serum amylase levels were also lower in patients with a positive hemoculture. The levels of in ammatory markers including as WBC, CRP, and PCT did not differ signi cantly in the two groups. Neut% was signi cantly higher in the positive hemoculture group (91.7±4.0% vs. 82.5±9.0%, P<0.0001). In the multivariate analysis (Odds ratio(OR) 1.52 [1.16-1.15], P=0.0030), Neut% was identi ed as the only factor that differed signi cantly in the two groups. Patients diagnosed with a positive hemoculture were more likely to have moderate or severe cholangitis than those with a negative hemoculture (35/48, 72.9% vs. 17/46, 37%, P=0.0004). Urgent or early biliary drainage was performed in more patients with positive hemoculture than with negative hemoculture (31/48, 64.6% vs. 17/46, 37%, P=0.0065). Bacteremia and subsequent high Neut% were considered to be associated with patient prognosis. The data of procalcitonin were obtained from 35 patients with positive hemoculture and 34 patients with negative hemoculture. Data are presented as number of patients (%). Mean values are presented as mean ± SD.
WBC=white blood cells; Neut%=proportion of neutrophils in the white blood cells; Lymph%= proportion of lymphocytes in the white blood cells; Mono%=proportion of monocytes in the white blood cells; GOT=glutamic oxaloacetic transaminase; GPT= glutamic pyruvic transaminase; CRP=C-reactive protein; PT-INR=prothrombin time international normalized ratio Ability of Neut% to predict bacteremia and severity in acute cholangitis Fig. 2 illustrates the ROC curves for Neut%, WBC, CRP, and PCT for prediction of bacteremia (Fig. 2a), severe or moderate cholangitis (Fig. 2b), and severe cholangitis (Fig. 2c)

Comparison of ability to predict bacteremia with other factors
We compared the ability of Neut% for prediction of bacteremia with PCT and other factors associated with severity assessment of acute cholangitis in TG13. We set the cutoff value of Neut% at 89.7% using the Youden index in the ROC curve. There were more patients with bacteremia in the Neut% ≥89.7% patients than in the Neut% <89.7% patients in the univariate analysis (37/48, 77.1% vs. 9/46, 19.6%, P<0.0001) and Neut% ≥89.7% and body temperature were selected as independent predictors of bacteremia in the multivariate analysis (Table 3).  Duration of antibiotics and hospital days are presented as median values.

Ability of Neut% to predict bacteremia in mild cholangitis
In patients with mild cholangitis according to TG13, close observation without biliary drainage is recommended. However, 13 of 40 patients with mild cholangitis had a positive hemoculture in this study. As such, we studied the features of patients who had bacteremia even if they were diagnosed with mild cholangitis. The data of procalcitonin were obtained from 7 patients with positive hemoculture and 21 patients with negative hemoculture. Data are presented as number of patients (%). Mean values are presented as mean ±SD. The data of procalcitonin were obtained from 7 patients with positive hemoculture and 21 patients with negative hemoculture. Data are presented as number of patients (%). Mean values are presented as mean ±SD.

WBC=white blood cells; Neut%=proportion of neutrophils in the white blood cells; CRP=C-reactive protein; PT-INR=prothrombin time international normalized ratio
WBC=white blood cells; CRP=C-reactive protein; Neut%=proportion of neutrophils in the white blood cells

Discussion
To the best of our knowledge, this study is the rst to show the usefulness of Neut% for prediction of bacteremia in patients with focusing on only acute cholangitis. Neut% showed an excellent independent diagnostic ability for bacteremia in the ROC curve analysis (AUC in ROC analysis was 0.86, and diagnostic sensitivity was 77% and speci city was 80% for Neut% ≥89.7%), and there were more patients with moderate and severe cholangitis and more patients requiring urgent or early biliary drainages in the high Neut% group than in the low Neut% group. Based on these results, Neut% was thought to be an excellent predictor, not only for bacteremia, but also for prognosis of cholangitis.
Furthermore, measurement of Neut% is easy, quick, and inexpensive. Neut% may be useful for care of patients with acute cholangitis, especially in terms of the decision to perform early biliary drainage.
In our clinical practice, we have observed that patients with bacteremia and/or severe cholangitis often have a high Neut% value (for example, Neut% ≥90%). Nowadays, at any hospital in Japan, neutrophil levels, lymphocytes, eosinophil levels, and monocytes among WBCs (Neut%, Lymp%, Eosi%, and Mono%) are measured automatically and quickly in an easy method, and the results are extremely accurate [10]. In bacterial and fungal infections, the number of mature neutrophils increases in the blood, and the neutrophil level (the amount of mature and young neutrophils), Neut%, and WBC count are up-regulated [26,27]. If the in ammation becomes severe as in sepsis, the younger neutrophils are released into the peripheral blood and there is a neutrophil left shift. WBC count and neutrophil count are thought to be the earliest and the most sensitive in ammatory markers and can be used to evaluate the treatment effect [26,27].
Therefore, we studied the association between bacteremia and Neut%, and found that Neut% was excellent for diagnosis of a positive hemoculture. Some studies have reported that Neut% is a useful marker for bacteremia in infections. Fukui et al. [28] stated that a Neut% >80% was an independent risk factor for a positive hemoculture. In this time, we studied correlation between bacteremia and Neut% focusing on only acute cholangitis. Further, one report [29] concluded that a high neutrophil-to-lymphocyte ratio (NLR) was a very good independent predictor of lethal outcomes in critically ill patients with secondary sepsis and/or trauma. We presume that NLR is nearly equal to Neut%. The neutrophil left shift is used for diagnosis of sepsis and it is a possible predictive factor for bacteremia. Kim et al. reported the e cacy of measuring the delta neutrophil index (DNI) as a parameter of neutrophil left shift for prediction of prognosis in patients with acute cholangitis [24]. However, it is di cult to obtain results for the DNI and it is di cult to obtain a blood smear for all patients. In accordance with results of this study, the up-regulation of Neut% may show the increasing necessity of neutrophil and neutrophil left shift assessments.
WBC count is the most common in ammatory biomarker, and many authors have reported the usefulness of WBC count as a predictor of severe acute cholangitis. However, our study and some other authors [6,7] concluded that WBC count was not independently useful for prognosis and bacteremia in acute cholangitis. A reason for this may be that the reaction of the WBC count in response to bacterial or fungal infections differs in each patient, especially in the elderly, and another reason is that the WBC count may change from high to low over the course of treatment of a severe case, and in severe in ammation.
Many authors have suggested that serum PCT level is a superior predictor of the presence of severe acute in ammation compared to conventional markers such as WBC count and serum CRP levels [30]. In this study, PCT showed a good and equal ability for assessment of severity compared with Neut% in the ROC curve analysis, but the diagnostic ability for bacteremia was weaker than Neut%. In general, in most diseases, bacteremia occurs with extensive and prolonged in ammation, but bacteremia in acute cholangitis occurs early and easily with the onset of obstruction of the bile duct because the cause of bacteremia is cholangio-venous re ux by up-regulation of pressure in the bile duct. Therefore, in acute cholangitis, it is important to identify infection in the early phase after onset. Although PCT is readily produced within 6-12 h, and serum procalcitonin levels increase [30], we presume that the earliest biomarker for bacteremic cholangitis is not PCT, CRP, or WBC-it is Neut%.
The recommendation for patients with mild cholangitis is observation without biliary drainage in TG13/18; however, we had 13 bacteremic patients of 40 patients with mild cholangitis. These patients may have needed early biliary decompression after considering other examination results. In this study, Neut% was an excellent predictor of bacteremia in not only all patients with cholangitis, but also patients with mild cholangitis. Accordingly, high Neut% can be utilized as an indicator for early biliary decompression in patients with mild cholangitis in TG13/18.
This study had several limitations. First, this was a retrospective analysis of medical records at a single hospital. There were some patients with acute cholangitis who did not have a blood culture test and there were no criteria on which to base the decision to obtain a blood culture or not. Patients who underwent a blood culture test had acute cholangitis with suspicion of bacteremia and severe cholangitis, and as such, our analysis using this population was relatively reasonable. Second, there was some data de ciency in serum PCT levels; therefore, this analysis may not have provided a completely correct assessment of the association between Neut% and PCT. Third, in this study, patients with acute pancreatitis and usage of an anticancer drug were excluded because these conditions would affect WBC count, Neut%, and CRP. In acute pancreatitis, not only the severity assessment, but also the diagnosis of cholangitis is di cult, and Neut% is relatively high in acute pancreatitis. Thus, predicting bacteremia in patients with acute pancreatitis is challenging. The anticancer drugs for pancreato-biliary cancer, gemcitabine, S-1, nab-paclitaxel, and cisplatin, do not induce severe bone marrow suppression compared with drugs for other cancers, and as such, there is often a high Neut% with a normal or low WBC count in patients with bacteremic acute cholangitis (data not shown). We think that high Neut% is an important factor for consideration when bacteremia is suspected in a patient taking an anticancer drug.

Conclusions
In conclusion, the percentage of neutrophils in the WBC count is measured automatically with an easy and quick method and might be useful for predicting bacteremia and severity of acute cholangitis during initial medical examination. If the Neut% is high, the existence of bacteremia should be considered, and early biliary drainage should be performed.

Declarations Ethics approval and consent participate
This study was performed in accordance with the Declaration of Helsinki and was approved by our ethics committee Approval No. 2020-20 . Patients were not required to give informed consent to the study because the analysis used anonymous clinical data that obtained after each patient agreed to receive treatment for acute cholangitis.