Utilizing cell population data for diagnosis and intensive care unit requirement prediction in Fournier gangrene patients

Introduction

Fournier Gangrene (FG) is a rare but serious condition that involves the rapid spread of a bacterial infection in the genital and perianal regions. It is a necrotizing fasciitis that affects the subcutaneous tissue, fascia, and muscle in the genital area. FG can progress rapidly, leading to sepsis, multi-organ failure, and death [1]. Several predisposing factors, including diabetes, advanced age, smoking, chronic alcoholism, peripheral vascular disease, congestive heart failure, hypertension, obesity, renal failure, coagulopathy, liver failure, local trauma, paraphimosis, urinary extravasation, and perianal abscess, have been associated with FG [2]. The causative agent of the infection is generally facultative bacteria such as Escherichia coli, Klebsiella spp., and Enterococcus spp. or anaerobic bacteria such as Bacteroides sp. [3].

Because of the high mortality rate, urgent urological intervention should be performed. Delayed treatment can lead to septic shock, multi-organ failure, and death. The most important determinants affecting mortality are early diagnosis, extensive surgical debridement, and using broad-spectrum antibiotics [4]. However, patients may need intensive care even if diagnosed early or treated aggressively. The Fournier Gangrene Scoring Index (FGSI) and Uludağ Fournier Gangrene Scoring Index (UFGSI) are used in determining the severity and predicting mortality of cases with FG [5, 6]. These scores are a numerical scoring system that considers various clinical and laboratory parameters to determine the extent and severity of the infection. While they are valuable tools for assessing severity, they have some limitations, such as limited validation and assessment of wound extent and comorbidities. Besides, some parameters used in these indexes, such as heart and respiratory rates, can be subjective and influenced by various factors such as pain, anxiety, or medications.

Complete blood count (CBC) is a commonly ordered blood test that provides information about various blood components, including erythrocytes, leukocytes, and platelets. One of the components of CBC is the measurement of various parameters of white blood cells, also known as leukocytes. Leukocytes are an essential part of the immune system, and their blood levels and population parameters can provide important clinical information. The new generation Sysmex XN-9000 analyzer can perform an extended leukocyte analysis known as cell population data (CPD) parameters. The leukocyte scattergrams are used to calculate CPD parameters based on lateral light scatter, forward light scattered light, and fluorescence light intensity for each leukocyte subgroup. These parameters indicate granular structures, reactivity, DNA and RNA content, metabolic activity, and size of leukocytes.

During acute bacterial infections, the bone marrow exhibits a response characterized by an elevation in the circulating levels of neutrophils and immature granulocytes. However, beyond this response, the significance of morphological and metabolic activity alterations within leukocyte subgroups among FG patients remains unclear. The aim of this study was to investigate the CPD-based biomarkers for the diagnosis and severity assessment of FG, distinguishing patients from healthy controls and predicting disease progression.

Materials and methods

Concepts of cell population data

The CBC and CPD parameters were obtained with a Sysmex-XN 9000 (Sysmex, Inc. Kobe Japan) analyzer using fluorescence flow cytometry technology. Leukocyte and subgroup count, expanded inflammation parameters (neutrophil reactivity index (RI) and granularity index (GI), antibody-synthesizing (AS) lymphocytes and reactive (RE) lymphocytes counts and percentages), and cell population indexes (X, Y, Z) and distributions widths (WX, WY, WZ) were evaluated. An increase in the lateral scatter light value (SSC or X) indicates an increase in granular structures of the cell population and has been defined as the granularity index specifically for neutrophils (Ne-GI or Ne-X). Higher fluorescent light intensity (SFL or Y) is associated with increased DNA and RNA content and metabolic activity of the cell population. It is also known as neutrophil reactivity (Ne-RI or Ne-Y). Increases in both neutrophil reactivity (Ne-RI or Ne-Y) and granularity (Ne-GI or Ne-X) indexes suggest that cells are active. Increased forward light scatter value (FSC or Z) correlates with larger cell size. Low WX, WY, and WZ values indicate that the population is homogeneous and similar in terms of granularity, activity, and size, respectively. In contrast, high values show that the cell distribution is heterogeneous and different. The count and percentages of reactive lymphocytes (RE-Ly) and antigen-synthesizing lymphocytes (AS-Ly), which are among the extended inflammation parameters, are calculated from the population of lymphocytes with increased fluorescence light intensity. These lymphocyte indexes can be used to evaluate cell-mediated innate and antibody-responsive adaptive immunity. Figure 1 provides an illustration of these values.

Figure 1:

Scattergrams of white cell differential (WDF) channel for peripheral blood samples from a healthy person and a Fournier gangrene patient performed by Sysmex-XN 9000. Leukocyte subgroups are shown in different colors as follows: eosinophils (red), lymphocytes (pink), monocytes (green), and neutrophils (light blue). SFL, fluorescence light intensity; SSC, side scatter; Ne-GI or Ne-X, lateral scattered light intensity of neutrophils; Ne-RI or Ne-Y, fluorescent light intensity of neutrophils; Ne-WX, lateral scattered light distribution width of neutrophils; Ly-WY, fluorescent light distribution width of lymphocytes.

Results

The mean age of the patients was 59 ± 16 years, and only one (4 %) was female. Diabetes mellitus was present in 10 patients (42 %), and urinary tract infections were diagnosed in 15 patients (63 %). Orchiectomy (for male patients) was performed in 8 (35 %), colostomy in 3 (13 %), and cystostomy in 3 (13 %). The primary infection was urogenital in 16 (67 %) patients and anorectal region in 8 (33 %). The extended disease was determined in 9 (38 %) patients. The mean FGSI and UFGSI values of the patients on admission were 3.5 ± 2.7 and 6.2 ± 3.9, respectively. A high level of positive correlation was found between the two indexes (Pearson correlation coefficient r=0.897, p<0.001). For quick SOFA value and length of stay (day), the median and quartiles were found to be 1 (0–1) and 18 (9–32), respectively. Seven (29 %) patients required intensive care, and mortality developed in two (8 %). The demographic and descriptive information of the patients and group comparisons are shown in Table 1.

The patients were separated into two groups as those who needed intensive care (n=7) and those who did not (n=17). There was no difference between these two groups in terms of age and gender. The frequency of diabetes mellitus and urinary tract infections was low in intensive care patients, with a significant difference determined only for urinary tract infections (p=0.004). The odds ratio for urinary tract infection frequency in these patients was found to be 0.036 (p=0.008). In intensive care patients, the colostomy rate was 27-fold higher (OR, p=0.039). No significant difference was found in terms of orchiectomy and cystostomy procedures. Anorectal primary infection was 12 times more common (OR, p=0.019) in intensive care patients. The rate of extended disease involving or exceeding the pelvic region was 28-fold (OR, p=0.008) higher in intensive care patients. There was no difference in body temperature, pulse, and respiratory rate, but systolic arterial pressure was significantly lower in intensive care patients (p=0.040). The FGSI (p=0.002), UFGSI (p=0.001), and qSOFA (0.003) scores on admission were higher in intensive care patients. There was no difference in terms of length of hospital stay.

Statistically significant differences were observed in terms of some CPD values of lymphocytes, neutrophiles, and monocytes in patients compared to healthy controls. Neutrophile-Y or reactivity index (Ne-Y or RI) (p=0.004), neutrophile-X or granularity index (Ne-X or GI) (p=0.009), monocyte-X (Mo-X) (p<0.001), and lymphocyte-WY (Ly-WY) (p<0.001) were higher in patients than controls. Ne-Y (RI) (p=0.012), Mo-X (p=0.001), Mo-Y (p=0.022), and Ne-WY (p=0.025) levels were higher in ICU patients than in non-ICU patients (Table 2).

Table 3 shows ROC analysis results for predicting the diagnosis of FG. AUC values above 0.9 for Neutrophile-Lymphocyte Ratio (NLR), Ly%, immature granulocyte count (IG#), Ne#, Ne%, and Ne-WY. All parameters had 100 % negative predictive values. ROC analyses were also performed to evaluate the power of the tests to predict the need for intensive care (Table 4 and Figure 2). Ne-Y (RI) had the highest AUC value compared to all other CPD and scoring indexes. Sensitivity and specificity for intensive care needs were calculated as 71.4 % and 100 % for FGSI at a cut-off value >5 and 100 % and 80 % for Ne-RI at a cut-off value >52, respectively. It was noteworthy that the negative predictive value for Ne-RI and positive predictive value for FGSI was 100 %.

Figure 2:

ROC analysis plot and AUC values for estimation of the need for intensive care unit. FGSI, Fournier Gangrene Scoring Index; ly#, lymphocytes count; Mo-X, lateral scattered light intensity of monocytes; Mo-Y, fluorescent light intensity of monocytes; Ne-RI or Ne-Y, fluorescent light intensity of neutrophils; Ne-WY, fluorescent light distribution width of neutrophils; UFGSI, uludağ Fournier Gangrene Scoring Index.

Discussion

Fournier gangrene is characterized by fulminant necrotizing fasciitis in the perineum or genital area. Immunosuppression, diabetes mellitus, alcoholism, acquired immunodeficiency, trauma, and genitourinary infections constitute predisposing factors for FG [7]. The disease can develop in three ways: lower urinary tract infections, infections in and around the rectum, and infections caused by bacteria on the skin after injury. Previous studies have shown that it causes mortality, although at different rates. FG is associated with poor health conditions in low socioeconomic groups, diabetes mellitus, and alcoholic liver disease, and diabetes is recognized as the most important medical condition [8]. The FGSI and UFGSI are classification systems used to evaluate the severity and course of the disease [9]. Since FG is an infectious disease, it is known that it causes changes in blood count parameters, and therefore several scoring systems utilizing these changes are used in disease prognosis [10]. One of the objectives of this study is to enhance prognosis prediction by incorporating leukocyte population indices alongside existing scoring systems.

This study presents the data of 24 FG patients, comprising 1 female and 23 males, who were followed up prospectively over the past year. Among these patients, seven (29 %) required intensive care units, and two (8 %) did not survive. Sepsis and multiple organ failure were the main determinants of intensive care need and mortality. Without the appropriate treatment, the mortality risk is high in this serious infection. However, we believe that using broad-spectrum antibiotics, enhancements in intensive care facilities, and early surgical debridement have significantly improved overall survival. Similar to the rates reported in the literature, 42 % of the patients had diabetes mellitus. However, no statistically significant correlation was observed between diabetes mellitus and the need for intensive care or death rates. Cystostomy was performed in three patients due to the spread of gangrene to the penis and urethra. Although FG typically leaves the testicles intact, the severity of infection and repetitive debridement can cause irreversible damage due to circulatory disturbance. Therefore, orchiectomy was performed in eight males to optimize patient outcomes and minimize complications after evaluating the patient’s general condition and the severity of the infection. As a result of anal sphincter involvement, three patients required the colostomy. In five of the seven patients who required intensive care, the source of infection was anorectal, and in only two of these cases was urogenital.

A relationship has been found between extended disease and the patient’s need for intensive care. While extended disease involving and exceeding the pelvic region was present in three of 17 patients who did not need intensive care, it was present in six of seven patients who needed intensive care. As a result, it is evident that the severity of the clinical condition is related to the extent of the disease.

It can be stated that the UFGSI score is a more valuable predictor than FGSI according to the ROC curve made in terms of predicting the requirement for intensive care. Similar to these scoring systems, the percentage of monocytes and neutrophils and lymphocytes count parameters that are frequently used in daily practice also provide valuable information.

There was a significant difference between the patient and healthy group in respect of the NLR and systemic inflammation index (SII), but patients hospitalized in the ward and intensive care units did not differ significantly in terms of these indexes. Therefore, NLR and SII are not considered appropriate indicators for predicting intensive care admission and disease severity.

The FGSI, UFGSI, and qSOFA scores were found to be high in patients who needed intensive care and were correlated with each other. Roghmann et al. showed that FGSI and UFGSI are useful in determining mortality, and it was seen in the current study that the indexes could be used to indicate the intensive care unit requirements [9]. Doluoğlu et al. showed that high FGSI and Charlson Comorbidity Index scores can be associated with poor prognosis [11]. In that study, it was concluded that metabolic parameters and predisposing factors should be evaluated together to predict prognosis. Although the indexes have been shown to be effective in predicting prognosis, the host factors and source of infection can also be considered important.

Extended hemogram parameters are used with increasing frequency in the diagnosis and prognosis of diseases. In a study by Demir et al. platelet count and NLR were found to be negatively associated with mortality [10]. In the literature review, no study was found showing the relationship between expanded hemogram parameters and FG. A comparison of the patient group and healthy controls found a significant increase in favor of the patient group in terms of leukocytes, monocytes, neutrophils, and immature granulocytes. It was observed that the neutrophil and monocyte population indexes were elevated in the group requiring intensive care unit.

Additional significant alterations were identified in the extended hemogram parameters of FG patients. Cell population data such as Ne-RI, Mo-X, Ne-WY, and Mo-Y provide valuable information about a patient’s need for intensive care unit admission and the severity of the clinical condition. The Ne-RI and Mo-Y parameters are obtained with fluorescent dyes that mark the nucleic acids of neutrophils and monocytes. These indexes are proportional to DNA/RNA content and reflect increases in cell reactivity. The data obtained in the study revealed that the Ne-RI value was increased in FG compared to the healthy subjects. Furthermore, it was observed to be notably higher in patients who needed intensive care compared to those who did not, and this value was determined with the highest performance in ROC analysis.

Another parameter indicative of neutrophil activity, Ne-GI, demonstrates an increase in granule contents. The Ne-GI was seen to be elevated in the patient group; however, it did not effectively predict the requirement for intensive care. Consistent with the findings of this study, Ustyantseva et al. showed that Ne-RI was higher in patients with sepsis compared to patients without sepsis and the control group, but Ne-GI did not differ between patients with and without sepsis [12].

It has also been shown that in addition to Ne-RI, the Mo-X value may be of clinical benefit in diagnosing and managing sepsis in intensive care patients [13]. In this study, it was observed that Mo-X and Ne-RI were significantly increased in patients with FG compared to healthy individuals, and these parameters can be used to predict the need for intensive care.

In a different study conducted with sepsis patients, increases were found in both Ne-RI and Ne-WY, which shows neutrophil activity and maturation heterogeneity [14]. It has also been stated that Ne-WY can be used to differentiate viral and bacterial diseases in the etiology of fever [15]. The data of the current study showed that Ne-WY could be used to predict the need for intensive care in cases with FG, in line with previous studies.

C-reactive protein (CRP) and procalcitonin represent primary acute-phase reactants widely employed in diagnosing bacterial infections. While their diagnostic efficacy in identifying sepsis is acknowledged, their utility as therapeutic indicators is constrained by their relatively prolonged kinetics of decline following treatment initiation. In contrast, as an indicator of metabolic activity heterogeneity, Ne-WY shows rapidly decreasing characteristics in patients recovering from the onset of sepsis [16]. Ne-WY is an interesting marker in the context of sepsis management as a more dynamic indicator of response to treatment compared to traditional acute-phase reactants and has the potential to overcome the weaknesses of acute phase markers at this stage. Therefore, using Ne-WY alone or in combination with CRP and procalcitonin can be particularly valuable in monitoring the efficacy of interventions and guiding decisions regarding the duration or intensity of treatment.

Peripheral blood smear evaluation is frequently used to detect banded neutrophils and immature granulocytes, such as toxic granulation and vacuolization, which is considered a left shift in leukocyte populations during infections. However, it is labor-intensive and time-consuming due to manual processes. Another limitation is that the evaluation is prone to errors due to inter-individual variation. Blood counting systems automate these processes, saving time and providing more objective and precise results. These devices analyze thousands of cells for size, granularity, and metabolic activity for each cell subgroup in the leukocyte population. The CPD parameters can be used in clinical practice as a rapid alternative to peripheral blood smear evaluation to detect bacterial and postsurgical infections [17].

One of the most significant limitations of our study, which investigates potential new markers for FG diagnosis and prognosis, is the necessity for careful interpretation of the results due to the small sample size. The relatively low incidence of FG limits the study’s statistical power when conducted at a single center. However, this challenge can be mitigated by performing multicenter studies to increase the sample size and enhance the robustness of the findings. Additionally, using intensive care unit patients without signs of infection or sepsis as the control group helps improve the comparisons. Our study did not follow changes in CPD parameters throughout the disease process. Future research should focus on these dynamics to provide further insights into disease prognosis. Lastly, it is important to note that these markers are currently investigational and not yet part of routine blood count reports.

Conclusions

The severity of FG disease can be determined using CDP data. Ne-Y (RI) serves as a novel and reliable biomarker for determining disease severity. Additionally, NLR can be used to rule out FG, especially in combination with other well-known clinical and diagnostic parameters. However, our findings should be interpreted with caution due to the small sample size. While these results provide important preliminary data, larger studies with more participants are necessary to confirm these findings and improve the generalizability and statistical power of the study.