Evaluation of hematological and biochemical parameters in the differential diagnosis of Crimean-Congo hemorrhagic fever and brucellosis: insights from a large patient cohort

Abstract

Background

In Türkiye, where both Crimean-Congo Hemorrhagic Fever (CCHF) and brucellosis are endemic, these infections share common presenting symptoms, including fever and musculoskeletal complaints, as well as overlapping laboratory findings such as thrombocytopenia and elevated liver enzymes, making early differential diagnosis challenging. Given their distinct infection control and management requirements, this study aimed to evaluate the diagnostic performance of routine hematological and biochemical parameters for differentiating these infections before confirmatory test results become available.

Methods

In this retrospective study, a total of 1,578 patients [1,171 (74.2%) with CCHF and 407 (25.8%) with brucellosis] followed at two tertiary care centers between January 1, 2015 and January 1, 2025 were analyzed. Demographic, clinical, and laboratory parameters were compared. Receiver operating characteristic (ROC) curve analysis was performed, and area under the curve (AUC) values were calculated. Optimal cut-off values were determined using the Youden index. Multivariable logistic regression analysis was subsequently conducted to identify independent predictors differentiating CCHF from brucellosis.

Results

White blood cell count (WBC), lymphocyte count, monocyte count, and platelet count demonstrated significant discriminatory performance for differentiating CCHF from brucellosis. The highest diagnostic performance was observed for lymphocyte count (AUC: 0.933; cut-off ≤ 1.09 × 10⁹/L, sensitivity 91%, specificity 83%) and platelet count (AUC: 0.927; cut-off ≤ 151 × 10⁹/L, sensitivity 89%, specificity 85%). Monocyte count (AUC: 0.888; cut-off ≤ 0.35 × 10⁹/L, sensitivity 81%, specificity 83%) and WBC (AUC: 0.880; cut-off ≤ 4.47 × 10⁹/L, sensitivity 89%, specificity 79%) also demonstrated high discriminatory ability. Among biochemical parameters, ferritin (AUC: 0.845) and creatine kinase (CK) (AUC: 0.826) showed the highest diagnostic performance. Multivariable logistic regression analysis identified lower lymphocyte count (OR = 0.601, 95% CI = 0.464–0.778, p < 0.001), lower platelet count (OR = 0.978, 95% CI = 0.972–0.985, p < 0.001), and elevated CK levels (OR = 1.003, 95% CI = 1.001–1.005, p = 0.005) as independent predictors of CCHF.

Conclusions

Findings from this large cohort demonstrate that routine hematological and biochemical parameters exhibit strong discriminatory performance in differentiating between CCHF and brucellosis. In particular, lymphocyte counts, platelet counts, and CK levels-which were identified as independent predictors favoring CCHF-may serve as valuable adjuncts to clinical evaluation, thereby facilitating the prioritization of confirmatory serological and molecular testing during the early diagnostic phase.

Background

Crimean-Congo Hemorrhagic Fever (CCHF) is an infectious disease observed in more than 30 countries across Africa, Asia, and Europe. The World Health Organization (WHO) has classified CCHF among the nine priority diseases for research and development due to its epidemic potential, high case fatality rate, and limited effective medical countermeasures [1]. The causative agent is the CCHF virus (CCHFV), which belongs to the genus Orthonairovirus within the family Nairoviridae [2]. Humans are most commonly infected through tick bites or contact with body fluids or secretions of infected animals [3]. CCHF is typically an acute illness characterized by the abrupt onset of symptoms such as fever, headache, fatigue, and muscle and joint pain. Laboratory findings include elevated liver enzymes, thrombocytopenia, and prolonged bleeding time [4].

Brucellosis is a systemic zoonotic infection caused by bacteria of the genus Brucella. It can affect multiple organs and systems. Transmission generally occurs through consumption of contaminated animal products such as unpasteurized milk and dairy products, direct contact with infected animals, or inhalation [5]. Although it occurs worldwide, it is endemic particularly in developing countries. The most common symptoms include fever, headache, fatigue, and musculoskeletal pain [6]. Fever may occasionally exhibit an undulating (“Maltese fever”) pattern, and the disease may follow a prolonged or chronic course if not adequately treated. Laboratory findings such as thrombocytopenia and elevated liver enzymes are also frequently observed [7].

In Türkiye, where both brucellosis and CCHF are endemic, these two zoonotic infections may present with similar clinical and laboratory findings, making differential diagnosis challenging. However, their distinct transmission dynamics and treatment approaches increase the importance of early and accurate diagnosis. This study aimed to contribute to the differential diagnostic process by comparatively evaluating clinical and laboratory parameters of both diseases.

Methods

Study design and ethical approval

This study was conducted retrospectively by analyzing data from patients diagnosed with brucellosis and CCHF who were followed either as inpatients or outpatients between January 1, 2015 and January 1, 2025 at the infectious diseases and clinical microbiology departments of two tertiary care hospitals located in Sivas, a province in central Türkiye. The region is characterized by extensive livestock farming, rural settlements, and large pasture areas, factors that contribute to the endemic occurrence of both brucellosis and tick-borne infections such as CCHF. Ethical approval for this study was obtained from the Ethics Committee of Sivas Cumhuriyet University Health Sciences Research Ethics Committee. All procedures were carried out in accordance with the Declaration of Helsinki.

Patient selection and diagnostic criteria

All patients aged 18 years and older who fulfilled the diagnostic criteria for CCHF or brucellosis, had complete clinical and laboratory data required for the predefined analyses, and did not meet any exclusion criteria were included in the study. The CCHF group consisted of cases confirmed by molecular methods (RT-PCR positivity for viral RNA) in addition to clinical findings. The brucellosis group included patients with compatible clinical symptoms and either positive blood culture for Brucella spp. and/or a standard tube agglutination (STA) test titer ≥ 1/160.

Patients with co-infection, those under 18 years of age, and those with incomplete data were excluded from the study.

Data collection

Patient data were obtained from hospital automation systems and medical records and recorded in standardized data forms. Demographic data included age, sex, and history of chronic diseases [diabetes mellitus (DM) and hypertension (HT)]. Clinical parameters included symptoms reported at presentation, such as fever and muscle and joint pain, together with the assessment of organomegaly on physical examination. These variables were selected because they represent common clinical manifestations of both CCHF and brucellosis and were consistently documented in the medical records of the study population.

Laboratory parameters at admission included complete blood count [white blood cell count (WBC), neutrophil, lymphocyte, monocyte, eosinophil, hemoglobin, platelet] and biochemical parameters [blood urea nitrogen (BUN), creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), lactate dehydrogenase (LDH), creatine kinase (CK), alkaline phosphatase (ALP), gamma-glutamyl transferase (GGT), total bilirubin, direct bilirubin, ferritin, and C-reactive protein (CRP)].

Laboratory analysis

Venous blood samples were collected from all patients at the time of hospital admission, prior to the initiation of specific therapy, and analyzed using standardized laboratory procedures in the central laboratories of the participating tertiary care centers.

Complete blood count parameters, including WBC, neutrophil, lymphocyte, monocyte, eosinophil counts, hemoglobin, and platelet levels, were measured using an automated hematology analyzer (XN-1000, Sysmex Corporation, Kobe, Japan), based on fluorescence flow cytometry and impedance methods.

Serum biochemical parameters, including BUN, creatinine, AST, ALT, LDH, CK, ALP, GGT, and total and direct bilirubin, were analyzed using enzymatic colorimetric assays on a Cobas c702 analyzer (Roche Diagnostics, Mannheim, Germany).

Serum ferritin levels were measured using an electrochemiluminescence immunoassay (ECLIA) method on a Cobas e801 analyzer (Roche Diagnostics, Mannheim, Germany), while CRP levels were determined using an immunoturbidimetric assay on a Cobas c702 analyzer (Roche Diagnostics, Mannheim, Germany). All assays were performed in accordance with the manufacturers’ instructions, and routine internal and external quality control procedures were applied.

The diagnosis of CCHF was confirmed by detection of viral RNA using a qualitative reverse transcription polymerase chain reaction (RT-PCR) assay (RealStar CCHFV RT-PCR Kit, Altona Diagnostics, Hamburg, Germany). RT-PCR analyses were performed in a national reference laboratory in accordance with standardized protocols.

The diagnosis of brucellosis was established based on compatible clinical findings together with serological and/or microbiological evidence. Serological testing was performed using the Rose Bengal test (Sero-Lam Brucella Rose Bengal antigen, Seromed, Türkiye) and STA tests (Wright and Coombs methods) using Sero-Tüp Brucella agglutination antigens (Seromed, Türkiye) and anti-human globulin (AGH MAESTRIA IGG+C3D, Diagast, France), in accordance with the manufacturers’ instructions. A titer of ≥ 1/160 was considered diagnostic.

For microbiological confirmation, blood cultures were obtained prior to antimicrobial therapy and processed using an automated blood culture system (BD BACTEC system, Becton Dickinson, Sparks, MD, USA). Identification of Brucella spp. was performed using conventional microbiological methods and/or matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS).

All laboratory analyses were conducted under standardized conditions with adherence to internal and external quality assurance programs.

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics 23 (IBM Corp., Armonk, NY, USA) software. Distribution of continuous variables was assessed using Kolmogorov–Smirnov and Shapiro–Wilk tests. Continuous variables were expressed as median and interquartile range (IQR), while categorical variables were presented as number and percentage. Comparisons between groups were performed using the Mann–Whitney U test for continuous variables and the chi-square test for categorical variables.

Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of laboratory parameters in differentiating CCHF and brucellosis. The area under the curve (AUC) and 95% confidence intervals (CI) were calculated. Optimal cut-off values were determined using the Youden index (J = sensitivity + specificity − 1). Variables potentially associated with the differential diagnosis between brucellosis and CCHF were initially evaluated using univariable analyses. Among the hematological and biochemical parameters, variables demonstrating the highest AUC values in ROC analysis were selected for inclusion in the binary logistic regression model. In addition, age, DM, and HT were incorporated into the multivariable model. Prior to model construction, multicollinearity among candidate variables was assessed, and the final model was established using variables deemed appropriate for inclusion. Logistic regression results were reported as odds ratios (ORs), 95% CIs, and p-values. A p-value < 0.05 was considered statistically significant.

Results

During the study period, 1,171 (74.2%) patients diagnosed with CCHF and 407 (25.8%) patients diagnosed with brucellosis were identified. When these patients were compared in terms of demographic and clinical characteristics, patients with CCHF were found to be older (p < 0.001). Generalized muscle and joint pain was more frequently observed in patients with brucellosis (p < 0.001). A history of DM (p = 0.029) and HT (p < 0.001) was more common among patients with CCHF (Table 1).

Table 1.

Comparison of demographic and clinical characteristics of patients with brucellosis and Crimean-Congo Hemorrhagic Fever (CCHF)

Variable	CCHF (n = 1171)	Brucellosis (n = 407)	p-value
Age, median (IQR)	51 (36–64)	44 (32–58)	< 0.001
Male sex, n (%)	750 (64.0)	278 (68.3)	0.121
Fever present, n (%)	977 (83.4)	159 (82.0)	0.611
Muscle and joint pain present, n (%)	1053 (89.9)	253 (98.4)	< 0.001
Diabetes mellitus (DM), n (%)	158 (13.5)	38 (9.3)	0.029
Hypertension (HT), n (%)	330 (28.2)	64 (15.7)	< 0.001
Hepatosplenomegaly present, n (%)	345 (29.5)	36 (24.0)	0.164

CCHF: Crimean-Congo Hemorrhagic Fever, DM: Diabetes mellitus, HT: Hypertension, IQR: Interquartile range (25th–75th percentile)

All evaluated hematological parameters differed between patients with brucellosis and CCHF. Among biochemical parameters, all tests showed significant differences except gamma-glutamyl transferase (GGT) (p = 0.745) and C-reactive protein (CRP) (p = 0.061) (Table 2).

Table 2.

Comparison of laboratory parameters in patients with brucellosis and Crimean-Congo Hemorrhagic Fever (CCHF)

Parameter	CCHF Median (IQR)	Brucellosis Median (IQR)	p-value
WBC (109/L)	2.8 (1.98–4.06)	6.92 (5.66–8.65)	< 0.001
Neutrophil count (109/L)	1.83 (1.13–3.07)	3.77 (2.74–5.00)	< 0.001
Lymphocyte count (109/L)	0.54 (0.36–0.88)	2.16 (1.69–2.82)	< 0.001
Monocyte count (109/L)	0.17 (0.09–0.29)	0.56 (0.40–0.73)	< 0.001
Eosinophil count (109/L)	0.00 (0.00–0.02)	0.05 (0.01–0.11)	< 0.001
HGB (g/dL)	14.2 (13.1–15.2)	13.9 (12.6–15.1)	0.005
PLT (109/L)	93 (53–130)	253 (198–309.5)	< 0.001
BUN (mg/dL)	15.5 (11.4–20.4)	13.1 (10.4–15.9)	< 0.001
Creatinine (mg/dL)	0.86 (0.72–1.05)	0.77 (0.65–0.89)	< 0.001
ALP (U/L)	74 (58–94)	96 (73–124)	< 0.001
ALT (U/L)	51 (26–107)	27 (17–50)	< 0.001
AST (U/L)	90 (44–213)	25 (18–42)	< 0.001
LDH (U/L)	379 (267–581.8)	239.5 (190–337.3)	< 0.001
CK (U/L)	279 (141.5–594)	73 (50–134)	< 0.001
GGT (U/L)	39 (19–87)	38.5 (23–74)	0.745
Total bilirubin (mg/dL)	0.41 (0.29–0.60)	0.50 (0.37–0.71)	< 0.001
Direct bilirubin (mg/dL)	0.16 (0.12–0.24)	0.20 (0.14–0.28)	< 0.001
Ferritin (µg/L)	1526 (502.8–8036)	202 (74–446)	< 0.001
CRP (mg/L)	13.66 (5.57–35.00)	12.56 (3.40–43.00)	0.061

WBC: White blood cell, HGB: Hemoglobin, PLT: Platelet, BUN: Blood urea nitrogen, ALP: Alkaline phosphatase, ALT: Alanine aminotransferase, AST: Aspartate aminotransferase, LDH: Lactate dehydrogenase, CK: Creatine kinase, GGT: Gamma-glutamyl transferase, CRP: C-reactive protein, IQR: Interquartile range (25th–75th percentile)

The diagnostic performance of complete blood count parameters was evaluated using ROC curve analysis (Table 3; Fig. 1). All evaluated hematological parameters demonstrated discriminatory ability in ROC curve analysis (p ≤ 0.004). Lymphocyte and platelet counts demonstrated the highest discriminatory performance among hematological parameters, followed by monocyte count and WBC. Detailed ROC analysis results, including AUC values and confidence intervals, are presented in Table 3.

Table 3.

Receiver operating characteristic (ROC) curve analysis of statistically significant complete blood count parameters

Variable	AUC	Std. Error	p-value	95% CI (Lower–Upper)
WBC (109/L)	0.880	0.010	< 0.001	0.861–0.899
Neutrophil count (109/L)	0.783	0.012	< 0.001	0.759–0.806
Lymphocyte count (109/L)	0.933	0.008	< 0.001	0.918–0.948
Monocyte count (109/L)	0.888	0.010	< 0.001	0.869–0.908
Eosinophil count (109/L)	0.779	0.014	< 0.001	0.751–0.807
PLT (109/L)	0.927	0.009	< 0.001	0.908–0.945
Hemoglobin (g/dL)	0.451	0.017	0.004	0.417–0.485

ROC: Receiver operating characteristic, AUC: Area under the curve, CI: Confidence interval, WBC: White blood cell, PLT: Platelet count

Fig. 1.

Receiver operating characteristic (ROC) curve analysis of statistically significant complete blood count parameters for differentiating Crimean-Congo Hemorrhagic Fever (CCHF) and brucellosis. Lymphocyte and platelet counts demonstrated the highest diagnostic performance, followed by monocyte count and white blood cell (WBC) count

The diagnostic performance of significant biochemical parameters was evaluated using ROC curve analysis (Table 4; Fig. 2). Ferritin and CK demonstrated the highest discriminatory performance among the evaluated biochemical markers. AST, LDH, and creatinine showed moderate discriminatory ability, whereas ALT and BUN provided only limited discriminatory value. ALP demonstrated a potential inverse discriminatory relationship, whereas total and direct bilirubin levels did not show meaningful discriminatory performance.

Table 4.

Receiver operating characteristic (ROC) curve analysis of statistically significant biochemical parameters

Variable	AUC	Std. Error	p-value	95% CI (Lower–Upper)
BUN (mg/dL)	0.608	0.034	0.008	0.541–0.675
Creatinine (mg/dL)	0.711	0.035	< 0.001	0.643–0.780
ALT (U/L)	0.642	0.038	< 0.001	0.568–0.716
AST (U/L)	0.786	0.033	< 0.001	0.721–0.850
LDH (U/L)	0.723	0.034	< 0.001	0.656–0.790
CK (U/L)	0.826	0.031	< 0.001	0.765–0.888
Ferritin (µg/L)	0.845	0.026	< 0.001	0.794–0.895
ALP (U/L)	0.323	0.041	< 0.001	0.243–0.402
Total bilirubin (mg/dL)	0.421	0.042	0.067	0.338–0.503
Direct bilirubin (mg/dL)	0.456	0.043	0.307	0.371–0.540

ROC: Receiver operating characteristic, AUC: Area under the curve, CI: Confidence interval, BUN: Blood urea nitrogen, ALT: Alanine aminotransferase, AST: Aspartate aminotransferase, LDH: Lactate dehydrogenase, CK: Creatine kinase, ALP: Alkaline phosphatase

Fig. 2.

Receiver operating characteristic (ROC) curve analysis of statistically significant biochemical parameters in patients with Crimean-Congo Hemorrhagic Fever (CCHF) and brucellosis. Ferritin and creatine kinase (CK) showed the highest discriminative performance among biochemical markers

The diagnostic performance of the hematological parameters with the highest AUC values (WBC, lymphocyte count, monocyte count, and platelet count) was further explored using ROC analysis, and optimal cut-off values were determined using the Youden index (Table 5; Fig. 3). All selected parameters demonstrated strong discriminatory performance in ROC analysis (p < 0.001).

Table 5.

ROC analysis and optimal cut-off values (Youden index) for selected parameters in differentiating Crimean-Congo Hemorrhagic Fever (CCHF) and brucellosis

Parameter	AUC	95% CI	p-value	Cut-off	Sens (%)	NPV	PPV	Spec (%)	Youden
WBC (109/L)	0.880	0.861–0.899	< 0.001	4.47	89	93	58	79	0.68
Lymphocytes (109/L)	0.933	0.918–0.948	< 0.001	1.09	91	94	63	83	0.74
Platelets (109/L)	0.927	0.909–0.945	< 0.001	151	89	94	66	85	0.74
Monocytes (109/L)	0.888	0.869–0.907	< 0.001	0.35	81	91	60	83	0.65

ROC: Receiver operating characteristic, AUC: Area under the curve, CI: Confidence interval, WBC: White blood cell

Fig. 3.

Receiver operating characteristic (ROC) curves of white blood cell (WBC), lymphocyte, monocyte, and platelet counts for differentiating Crimean-Congo Hemorrhagic Fever (CCHF) and brucellosis. Lymphocyte and platelet counts showed the highest area under the curve (AUC) values and Youden index

Lymphocyte count demonstrated the highest diagnostic performance (AUC: 0.933; 95% CI: 0.918–0.948). The optimal cut-off value was determined as 1.09 × 10⁹/L, with a sensitivity of 91% and specificity of 83%. Platelet count also showed high discriminative performance (AUC: 0.927; 95% CI: 0.909–0.945), with a cut-off value of 151 × 10⁹/L, yielding a sensitivity of 89% and specificity of 85%.

For WBC, the AUC was 0.880 (95% CI: 0.861–0.899), with a cut-off value of 4.47 × 10⁹/L corresponding to a sensitivity of 89% and specificity of 79%. Monocyte count also demonstrated high discriminative power (AUC: 0.888; 95% CI: 0.869–0.907). At a cut-off value of 0.35 × 10⁹/L, sensitivity was 81% and specificity was 83%. The highest Youden index values were observed for lymphocyte and platelet counts (both 0.74).

Multivariate logistic regression analysis was performed to identify independent predictors differentiating CCHF from brucellosis (Table 6). Lower lymphocyte count (OR = 0.601, 95% CI = 0.464–0.778, p < 0.001) and lower platelet count (OR = 0.978, 95% CI = 0.972–0.985, p < 0.001) were significantly associated with an increased likelihood of CCHF diagnosis. Conversely, elevated creatine kinase levels (OR = 1.003, 95% CI = 1.001–1.005, p = 0.005) served as a significant independent predictor for CCHF. Age, sex, diabetes mellitus, hypertension, and ferritin levels were not independently associated with disease differentiation in the multivariable model. (p > 0.05).

Table 6.

Multivariate logistic regression analysis for predictors differentiating Crimean-Congo Hemorrhagic Fever (CCHF) from brucellosis

Predictor	OR	95% CI	p-value
Age	1.022	0.993–1.051	0.135
Sex (Male)	1.785	0.805–3.958	0.154
Diabetes mellitus	0.885	0.233–3.358	0.857
Hypertension	1.537	0.503–4.697	0.451
Lymphocyte count	0.601	0.464–0.778	< 0.001
Platelet count	0.978	0.972–0.985	< 0.001
Creatine kinase	1.003	1.001–1.005	0.005
Ferritin	1.000	1.000–1.000	0.596

CI: confidence interval, OR: odds ratio

Discussion

In this study, hematological parameters demonstrated high diagnostic performance in differentiating CCHF and brucellosis. WBC, lymphocyte count, monocyte count, and platelet count were identified as the most discriminative hematological parameters, whereas ferritin and CK showed the highest diagnostic performance among biochemical markers. Furthermore, multivariable logistic regression analysis demonstrated that lower lymphocyte count, lower platelet count, and higher CK levels remained independently associated with CCHF after adjustment for potential confounding factors. No significant difference was observed between the two diseases in terms of CRP levels, while total and direct bilirubin levels did not demonstrate significant discriminative value.

CCHF and brucellosis share similar geographical distributions, and cases of co-infection have been reported [8]. Moreover, the considerable overlap in clinical and laboratory findings makes differential diagnosis challenging [9]. In endemic regions such as Türkiye, brucellosis has been reported as one of the most frequent diagnoses among patients initially suspected of having CCHF but later not confirmed (17.7%) [10]. These findings emphasize the importance of differential diagnosis. Considering the differences in management, public health implications, and infection control measures between CCHF and brucellosis, as well as potential delays in confirmatory diagnosis, early and accurate differentiation is crucial for appropriate patient management and timely implementation of preventive measures [11–13].

In our cohort, patients with CCHF were older and had a higher frequency of diabetes mellitus and hypertension, which may partly reflect the age-related distribution of these comorbidities. In contrast, muscle and joint pain was more frequent among patients with brucellosis, consistent with the well-known musculoskeletal involvement of this disease.

Leukopenia in CCHF is associated with leukocyte lysis resulting from increased myeloperoxidase expression. Additionally, CCHFV is known to induce apoptosis via caspase activation. These mechanisms contribute to leukopenia. Endothelial damage leads to activation of the coagulation cascade, resulting in thrombocytopenia and disseminated intravascular coagulation, which may progress to multiorgan failure. Hemophagocytosis has also been shown to contribute to the development of cytopenias [8, 14–16].

Brucellosis, on the other hand, is characterized by mild leukopenia, relative lymphocytosis, mild anemia, and thrombocytopenia. These findings are primarily attributed to hypersplenism and bone marrow involvement. In rare cases, disseminated intravascular coagulation, hemophagocytosis, or immune-mediated cellular destruction may also contribute [17].

When hematological parameters were compared, patients with CCHF had significantly lower WBC, neutrophil, lymphocyte, monocyte, eosinophil, and platelet counts, while hemoglobin levels were higher. This finding can be explained by the more pronounced mechanisms leading to cytopenias in CCHF. Since admission laboratory values were analyzed in this study, anemia is not expected in the early phase of CCHF. Anemia typically develops during the hemorrhagic phase and is secondary to bleeding [18]. In contrast, anemia is a common laboratory finding in brucellosis and may result from chronic inflammation, bone marrow involvement, and anemia of inflammation [19]. This difference likely explains the higher hemoglobin levels observed in patients with CCHF.

Lymphocyte count was identified as the parameter with the highest diagnostic performance. Notably, lower lymphocyte count remained independently associated with CCHF in the multivariable model, suggesting that lymphopenia represents one of the most reliable laboratory features distinguishing CCHF from brucellosis. This finding can be explained by the marked lymphopenia observed in the early phase of CCHF compared to the relatively higher lymphocyte levels seen in brucellosis [8, 17].

Platelet count also demonstrated high diagnostic performance. Moreover, lower platelet count remained independently associated with CCHF in the multivariable logistic regression analysis, highlighting thrombocytopenia as one of the most robust laboratory features for differentiating CCHF from brucellosis. In CCHF, thrombocytopenia develops due to endothelial damage, disseminated intravascular coagulation, and peripheral consumption. In contrast, thrombocytopenia in brucellosis is generally less severe and has been attributed to mechanisms such as hypersplenism, bone marrow suppression, hemophagocytosis, and immune-mediated platelet destruction [8, 17].

Among biochemical parameters, ferritin and CK exhibited the highest discriminative power. Elevated ferritin levels in CCHF have been associated with the cytokine storm observed in the disease [20]. In brucellosis, ferritin elevation reflects both the inflammatory response and the intracellular survival strategy of the pathogen affecting iron metabolism [21, 22]. From a pathogenetic perspective, ferritin elevation in brucellosis is associated with a more controlled inflammatory response, whereas in CCHF it is linked to a more severe and uncontrolled inflammatory process. This difference explains the high discriminative power of ferritin. Nevertheless, although ferritin demonstrated strong discriminative performance in ROC analysis, it did not retain statistical significance in the multivariable logistic regression model. This finding suggests that the apparent diagnostic value of ferritin may partially reflect its association with other laboratory abnormalities rather than an independent contribution to disease differentiation.

In CCHF, the hemorrhagic process and cytokine storm contribute to increased myocyte damage, leading to elevated CK levels [23]. In contrast, musculoskeletal involvement in brucellosis is typically characterized by arthralgia and localized inflammation rather than widespread muscle destruction [24]. Therefore, CK elevation is generally limited in brucellosis. This pathophysiological difference explains the diagnostic value of CK. Importantly, CK remained a significant independent predictor of CCHF in the multivariable logistic regression analysis. Together with lymphocyte and platelet counts, this finding supports the potential utility of CK as a robust laboratory parameter for distinguishing CCHF from brucellosis in routine clinical practice.

Previous studies have primarily investigated laboratory predictors of disease severity, organ involvement, or bacteraemia within individual cohorts of CCHF or brucellosis [25–27]. In a recent study from Türkiye, Arslan et al. directly compared patients with CCHF and acute brucellosis and demonstrated that routine laboratory parameters may assist in the differential diagnosis of these two infections, with CK showing particularly strong discriminatory performance [28]. Consistent with these findings, CK was also identified as one of the most informative laboratory parameters in our cohort. However, unlike the previous study, which was conducted in a relatively small patient population, our analysis included a substantially larger cohort and incorporated multivariable logistic regression analysis. These methodological strengths allowed us not only to evaluate the discriminatory performance of individual laboratory parameters but also to identify independent predictors of CCHF, thereby providing more robust and clinically applicable evidence for the differential diagnosis of CCHF and brucellosis in routine clinical practice.

From a clinical perspective, the cut-off values identified in this study may be particularly useful during the initial evaluation of patients presenting with suspected CCHF or brucellosis in endemic regions. Because complete blood count parameters are routinely available in emergency departments and outpatient settings, these markers may help clinicians prioritize confirmatory molecular and serological testing while awaiting definitive diagnostic results. However, the proposed cut-off values were derived from a specific patient population and laboratory setting. Therefore, they should be interpreted in conjunction with clinical and epidemiological findings and may require validation before being applied universally across different healthcare systems and laboratory platforms.

Several limitations of this study should be acknowledged. Although standardized and widely accepted diagnostic criteria were applied for both diseases, the confirmatory diagnostic methods used for brucellosis and CCHF are not without limitations, which may have affected case classification. In addition, laboratory parameters were assessed only at the time of hospital admission, and information regarding the exact interval between symptom onset and sample collection was not available. Therefore, potential temporal changes in laboratory findings during the course of illness could not be evaluated. Additionally, several potentially informative biomarkers, including cytokines and coagulation parameters, could not be evaluated because these measurements were not consistently available in both study groups throughout the study period. Similarly, detailed information regarding symptom duration, exposure history, bleeding manifestations, and disease severity was not available in a standardized format for all patients and therefore was not included in the present analysis. Furthermore, because CCHF and brucellosis share overlapping clinical manifestations and laboratory abnormalities, concomitant infection remains a diagnostic consideration in endemic settings. Although patients were evaluated using established diagnostic procedures, the possibility of unrecognized co-infection cannot be entirely excluded. Finally, the retrospective design of the study and the evaluation of diagnostic performance in a cohort from a specific endemic region may limit the generalizability of the findings. Therefore, external validation in independent populations and prospective studies are warranted to confirm the applicability of these results in different clinical settings.

Conclusions

Among the evaluated laboratory parameters, lymphocyte count, platelet count, and CK level emerged as the most robust markers for differentiating CCHF from brucellosis, demonstrating both high discriminatory performance and independent predictive value in multivariable analysis. These findings suggest that simple and readily available laboratory parameters may serve as practical tools for the early differential diagnosis of clinically similar zoonotic infections, thereby supporting timely clinical decision-making and appropriate public health responses. Nevertheless, definitive diagnosis should continue to rely on appropriate microbiological, serological, and molecular methods. These laboratory parameters should be regarded as complementary tools that may support, but not replace, established diagnostic approaches.

Abbreviations

CCHF

Crimean-Congo hemorrhagic fever

ROC

Receiver operating characteristic

AUC

Area under the curve

WBC

White blood cell

CK

Creatine kinase

CRP

C-reactive protein

WHO

World Health Organization

CCHFV

Crimean-Congo hemorrhagic fever virus

RT-PCR

Reverse transcription polymerase chain reaction

STA

Standard tube agglutination

BUN

Blood urea nitrogen

AST

Aspartate aminotransferase

ALT

Alanine aminotransferase

LDH

Lactate dehydrogenase

ALP

Alkaline phosphatase

GGT

Gamma-glutamyl transferase

IQR

Interquartile range

CI

Confidence interval

DM

Diabetes mellitus

HT

Hypertension

Author contributions

CO conceived and designed the study, developed the methodology, performed data curation, formal analysis and investigation, interpreted the data, and drafted the manuscript. CO also contributed to reviewing and editing the manuscript and supervised the study. MÖ performed formal and statistical analyses, contributed to data visualization and methodology, interpreted the data, and contributed to writing—review and editing. EK and SBD contributed to data curation, investigation, validation, and writing—review and editing. ÖA contributed to data curation, patient recruitment, investigation, validation, and writing—review and editing. YÇK contributed to data curation, investigation, validation, and writing—review and editing. HOD contributed to methodology, performed biochemical analyses, validated the results, interpreted laboratory data, and contributed to writing—review and editing. NE supervised the study, contributed to validation, critically revised the manuscript, and contributed to writing—review and editing. SAB supervised the study, provided conceptual support, contributed to validation, critically revised the manuscript, contributed to writing—review and editing, and approved the final version of the manuscript. All authors read and approved the final manuscript and agree to be accountable for all aspects of the work.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval for this study was obtained from the Ethics Committee of Sivas Cumhuriyet University Health Sciences Research Ethics Committee. Written informed consent was obtained from all participants included in the study. This study was conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.