An overview of procalcitonin in Crimean-Congo hemorrhagic fever: clinical diagnosis, follow-up, prognosis and survival rates

Introduction

Procalcitonin is known as a biomarker protein for bacterial sepsis [1]. Following the upregulation of procalcitonin, the release of interferon-γ–a cytokine that is released in response to viral infections – is suppressed, so that procalcitonin levels do not increase, or increase only slightly [2], [3].The leading cause of procalcitonin production is bacterial infection, but with it, everything that causes tissue inflammation leads to an increase in procalcitonin production [4].

Crimean-Congo hemorrhagic fever (CCHF) is a tick-borne zoonosis that frequently causes a sudden onset of severe illness and death in humans [5]. It has been suggested that endothelial damage resulting from the endothelial activation of the CCHF virus (the genus Orthonairovirus of the family of Bunyaviridae) and viral antigen-triggered immune mechanisms are implicated in the ethiopathogenesis [6], [7].

Proinflammatory cytokines such as IL-6, IL-10 and TNF-alpha, which are mediators of acute inflammation, have been shown to play important roles in CCHF [8]. Procalcitonin is produced ubiquitously in response to these proinflammatory cytokines, which in turn are released in response to bacterial infections [9].

There have been various studies of disease-specific correlations and poor clinical outcomes related to procalcitonin [10], [11], [12], however there have to date been no studies in literature investigating the diagnostic threshold value or usefulness of procalcitonin in CCHF. The objective of the present study is to identify any association between procalcitonin levels and CCHF, and to determine a predictive cut-off level. The study also explores the association between procalcitonin and survival rates and the prognosis of CCHF patients. The study further investigates the behavior of procalcitonin levels during hospital stays and any correlations between procalcitonin, coagulopathy and blood count differentiation parameters.

Materials and methods

Results

Procalcitonin results

The minimum serum procalcitonin level was 0.072 μg/L in the survivor patients, while the minimum serum procalcitonin level in the non-survivor patients was 0.315 μg/L. The patients’ data given in Tables 1 was obtained from those who admitted to the hospital within the first 24 h after the onset of the first disease symptoms. A statistically significant increase in procalcitonin was observed in patients with CCHF when compared to the controls (mean ± SD: 0.397 ± 0.738 μg/L vs. mean ± SD: 0.058 ± 0.01 μg/L, p<0.001, respectively).

Table 1:

The laboratory parameters of CCHF and control patients, and survivor and deceased CCHF patients, at the time of first admission.

Parameters	Control median (min/max) n=61	CCHF median (min/max) n=100	p-Value	Survivor median (min/max) n=95	Deceased median (min/max) n=5	p-Value
Procalcitonin (µg/L)	0.040 (0.02/0.62)	0.173 (0.020–5.050)	<0.001	0.504 (0.072/5.050)	1.315 (0.315/3.190)	0.028
WBC (×109/L)	7.4 (2.4/16.8)	6.7 (0.48/27.9)	0.456	2.1 (0.48/279)	10.41 (3/13)	<0.001
PLT (×109/L)	158 (121/451)	32 (18.99/229)	0.025	44 (15/229)	18.99 (12/98)	0.028
Urea nitrogen (mg/dL)	12.4 (3.8/62)	15.2 (3.6/105)	0.325	11.8 (3.6/105)	26 (23/73)	<0.001
Creatinine (mg/dL)	0.78 (0.52/1.89)	0.82 (0.41/7.2)	0.242	0.77 (0.4/7)	1.21 (1/3)	0.005
CRP (mg/L)	6.12 (2.1/9.2)	7.8 (0.7/56.2)	0.472	6.74 (0.7/279)	42 (40/56)	0.021
AST (U/L)	48 (21/67)	145 (12/1,171)	0.041	87 (12/1,171)	275 (36/331)	0.343
ALT (U/L)	52 (42/244)	198 (21/2,779)	0.048	74 (21/2,779)	193 (26/113)	0.825
LDH (U/L)	280 (145/453)	652 (177/3,568)	0.047	558(177/3,568)	1233 (863/2,139)	0.005
Neutrophil (%)	52.4 (24/80)	57.8 (19.8/89)	0.448	56.4 (19.8/82)	81.85 (78/89)	<0.001
Lymphocytes (%)	37.4 (13.3/57.5)	33.3 (7.2/53.8)	0.256	34.7 (7.2/53.8)	11.15 (9/16)	0.050
Neutrophil (×109/L)	4.15 (1.9/22.6)	5.02 (0.21/18.1)	0.547	1.02 (0.21/18.1)	8.62 (3/11)	<0.001
Lymphocytes (×109/L)	0.93 (0.28/5.02)	0.61 (0.11/3.94)	0.286	0.61 (0.19/3.94)	1.02 (0/2)	0.221
Fibrinogen (mg/dL)	260 (160/410)	178 (36/367)	0.040	229.98 (36/367.23)	116 (26.11/305)	0.040
aPTT (sec)	31 (18–42)	48 (26/120.6)	0.019	42 (26/107)	67 (42/120.6)	0.011
INR	1.01 (0.72/1.31)	1.28 (0.78/4.62)	<0.001	1.13 (0.7/4.62)	1.93 (1.37/2.24)	<0.001
Ferritin (ng/mL)	630 (45/745)	2000 (42/3344)	<0.001	2000 (42/3344)	2000 (1,800/2,091)	0.465
D-dimer (mg/L)	0.482 (0.250/0.680)	1.38 (0.23/37)	<0.001	1.24 (0.23/13.3)	12.5 (5.6/37)	0.004

1. AST, aspartate aminotransferase; ALT, alanine aminotransferase; CRP, C-reactive protein; aPTT, activated partial thromboplastin time; INR, international normalised ratio; CK, creatinine kinase; LDH, lactat dehydrogenase; PLT, platelet; WBC, white blood cell.

2. Bold values: correlation is significant at the 0.01 level.

Similarly, the procalcitonin levels of the non-surviving patients were also found to be significantly higher than those of the survivors in the patient group (mean ± SD: 1.389 ± 1.310 μg/L vs. mean ± SD: 0.504 ± 0.786 μg/L, p=0.028, respectively).

The procalcitonin levels of CCHF patients reached their highest level (1.24 ± 2.5 μg/L) or 0.44(0.09/11.17 μg/L) on the second day of admission to the hospital, and followed a declining trend from day 3 onwards until a second elevation wave was observed on the 5th day. On the final day of follow-up (day 9), the procalcitonin reached its lowest level in the consecutive measurements (0.49 ± 0.47 μg/L) or 0.19 (0.17/3.9 μg/L). While procalcitonin levels were significantly different from the control group for eight consecutive days after the first admission day (p<0.001), it was found that the procalcitonin levels were not different from the control group on the 9th day (Figure 1).

Figure 1:

Overview of average procalcitonin levels by days in CCHF patients. *Time behavior of PCT values (µg/L) in patients with CCHF. The median (dot), quartile (box), and 25/75 percentile (bar) are given, white rectangle inside bar: group median value. Group median differences were analyzed by Kruskall Wallis post-hoc pairwase comparisons test. *p<0.05, **p<0.001.

The Spearman’s correlations of procalcitonin are shown in Table 2. LDH and ferritin were strongly correlated with procalcitonin in CCHF patients (p<0.001; r=0.673, p<0.001; r=0.526).

Table 2:

Spearman’s correlations of procalcitonin in patients with CCHF.

Parameters	rho	p-Value
LDH (U/L)	0.673	<0.001
Ferritin (ng/mL)	0.536	<0.001
aPTT (sec)	0.577	<0.001
ALT (U/L)	0.406	<0.001
AST (U/L)	0.254	<0.05
C-reactive protein (mg/L)	0.396	<0.001
Platelet (×109/L)	−0.398	<0.001
Creatinin kinase (U/L)	0.335	<0.01
BUN (mmol/L)	0.319	<0.01
Creatinin (µmol/L)	0.278	<0.05
Fibrinogen (mg/dL)	−0.271	<0.05
INR	0.261	<0.05
WBC (×109/L)	0.314	<0.01
Lymphocyte (#)	–	>0.05
Lymphocyte (%)	0.253	<0.05
Neutrophil (#)	0.312	<0.01
Neutrophil (%)	0.283	<0.02
D-dimer (mg/L)	–	>0.05

1. #; absolute count, %; differantiation ratio.

2. Bold values: correlation is significant at the 0.01 level.

ROC analysis

The ROC curve analysis showed that the best predictive procalcitonin level cut-off points for patients with CCHF was 0.560 μg/L, AUC was 0.872 with a specify of 97% and sensitivity of 27%, respectively (Figure 2).

Figure 2:

Receiver operating characteristic analysis of serum procalcitonin that predicts CCHF patients, confirmed by RT-PCR test.

Survival analysis

A Chi-square test (with one df) of the overall difference between the level of the two procalcitonin groups were performed. According to this, for the two procalcitonin based group on the cut-off value found in our study were showed significantly different (log-rank: 4.358, p=0.037) (Figure 3).

Figure 3:

Plot of survival distribution functions according to procalcitonin levels.

Discussion

In a single study reported from Turkey with fewer patients a difference was noted between the procalcitonin levels of the survived and deceased patients with CCHF [15].

This retrospective study, however, is the first to suggest procalcitonin as a new biomarker for CCHF, to investigate a cut-off value for prediction at the time of first diagnosis, and to examine its association with survival.

The present study has shown that procalcitonin was increased in patients with CCHF, and that this elevation was correlated significantly with survival time. One particular strength of the present research was its reporting of procalcitonin levels both at first admission and on consecutive days of infection (9 days), which is unique in literature.

CCHF is a serious tick-borne viral disease (Orthonairovirus from the family of Bunyaviridae) that causes loss of life and which is endemic to Turkey [16]. Patients present to healthcare centers with dramatic and rapidly progressing complaints such as myalgia, fever and hemorrhage [17]. Depending on the quality of the health services and the geographical distribution overlap of Hyalomma and Ixodid spp. (the most important vectors), outbreaks of CCHF with case mortality rates of up to 40% have been reported [18]. The mortality rate of CCHF was found to be 5% in the present study.

High viremia levels are encountered in the first 5 days of illness [19], while specific IgM levels become measurable in the serum from day 5 of CCHF [17]. In the present study, procalcitonin was measured at its maximum level on day 2 following the onset of symptoms, and the highest viremia levels in the blood occur on days 2 and 3 of post-infection [20], [21]. It can thus be suggested that this high-level procalcitonin peak on day 2 may be associated with viremia levels.

The second increase wave of procalcitonin on day 5 is synchronous with the onset of IgM formation. This finding may be foreground the role of immunological mechanisms, which are also implicated as a cause of vascular endhotelitis in the etiopathology of CCHF. However, it was viral load, not IgM, in CCHF that was proposed as an independent prognostic marker in another study [22]. In the present study, viral load and immunoglobulins were measured only to confirm the diagnosis at the time of first admission to our emergency service, and the lack of a daily viral load quantitation and IgM titration follow-up due to the retrospective nature of the study can be considered a limitation of the study. Another study may be planned to examine the relationship between viral load, IgM levels and procalcitonin in CCHF.

The Hantaan, Seoul, Puumala and Dobrava viruses are all Orthohantaviridae family subtypes that cause hemorrhagic fever with renal syndrome (HFRS). Procalcitonin levels in HFRS associated with Hantaan viruses have been identified as an independent risk factor affecting the severity of the disease [23]. High serum procalcitonin levels, which are uncorrelated to the severity of kidney damage, have been recorded in nephropathy epidemics (NE) caused by Puumala virus [24]. The procalcitonin median value in HFSR patients caused by the Dobrava virus has been measured at 0.74 μg/L, while this value has been recorded as 0.50 μg/L for Puumala virus infections [25]. Although these are different members of the same family, the endothelial cell is the primary viral pathogenicity target in HFRS disease, as in CCHF disease. The procalcitonin median values in the present study were more compatible with Puumala virus infections, with 0.560 μg/L.

It is not uncommon that bacterial co-infections occur in severely-ill patients with viral hemorrhagic fevers. The any possible co-infection may be a cause of concomitant increase in serum procalcitonin levels observed in the disease. The effect of the possible secondary bacterial infection on WBC and procalcitonin levels could be considered in the patients who died, in our study. The fact that two of the patients died due to Staphylococcus aureus sepsis supports this finding.

These two patients were deceased on the 13th and 18th days of the follow-up period. The procalcitonin levels of these patients’ death days were 1.54 and 2.63 µg/L, and the procalcitonin levels at the time of first admission were 0.315 and 0.679 µg/L, respectively. These procalcitonin concentrations were evaluated in accordance with CCHF patient first day group averages.

These procalcitonin concentrations were interpreted as compatible with the group averages on the first day. In daily consecutive follow-up for these two patients, the procalcitonin levels of these patients tended to decrease starting from Day 6, but almost doubled from days 11 and 15, respectively. Procalcitonin is an acute phase reactant with an induction period of 4–12 h, peaking in 6–12 h and a half-life of 22–35 h, in bacterial origin inflammatory response [26].

According to the procalcitonin daily behavior findings of the deceased patients, sepsis, which is accused of for the etiology of two deaths, was interpreted as not originating from an existing bacterial infection at the time of first admission to the hospital, but of a hospital-acquired seconder bacterial infection.

Due to the nature of used statistical analysis, the values at the time of admission are used during ROC analysis conducted to evaluate the diagnostic and prognostic properties of the test. Therefore, we can say that these two patient results are not effective on the test sensitivity given in this study. Also, in the findings of procalcitonin daily behavioral graph, the results of these two patients had no effect on statistical results. We have not encountered any literature study on how a bacterial co-infection affects procalcitonin levels in CCHF patients. So, we have interpreted a nearly two-fold increase in the case of secondary infection (sepsis) as a valuable literature finding.

In the present study, a strong correlation was noted between procalcitonin and certain laboratory parameters that were reported previously as biomarkers in CCHF [27], [28]. It is known that serum PCT levels increase if bacterial, fungal or parasitic infections but increase normally or slightly in viral infections and non-infectious inflammatory reactions. However, the severity of the degree of CCHF endhotelitis may not be sufficient to explain this significant increase in procalcitonin levels. CCHF is also a disease that can cause kidney, lung, and especially sudden liver cell damage and, accordingly, liver failure. Indeed, in this study, significant acute liver damage is observed in CCHF patients. It has been previously shown that, without any bacterial infection, procalcitonin levels can increase >2.0 μg/L in toxic liver injury, regardless of the cytokine inducing effect of procalcitonin [29], [30]. Therefore, it can attribute the procalcitonin CCHF disease specific predictive effect of procalcitonin in this study not only to endothelial inflammation response but also to be a direct viral cell damage effect.

We demonstrated that procalcitonin showed the strongest correlation with lactate dehydrogenase, ferritin, activated partial thromboplastin time and aspartate aminotransferase. It has been reported that lactate dehydrogenase and procalcitonin elevate simultaneously only in influenza pneumonitis from among the viral diseases [31]. If it could be identified which isoenzyme was elevated, an advanced commentary could be made on the correlation between LDH and procalcitonin. However, a virus type has been identified that has an increased effect on the immunosuppressive conditions that in particular raise LDH enzymes in mice (Lactate dehydrogenase-elevating virus), and the cause of LDH elevation has been found to be the persistent destruction of the subpopulation of macrophages responsible for the circulatory clearance of this enzyme by the LDV virus [32]. Similar mechanisms explaining the elevation of LDH in CCHF were not identified in the present study.

Increased ferritin levels, which are thought to play an important role in hematophagocytosis in the pathogenesis of CCHF, have been identified as a marker in the diagnosis and prognosis of CCHF [33]. There was a strong correlation between procalcitonin and ferritin, in our study, and so we suggest that these acute phase reactants may be used as complementary markers, acting in the same direction in CCHF illness.

Fibrinogen levels did not vary significantly during hospitalization, although the lowest levels were recorded at the time of admission (median/min-max; 178–36/367 mg/dL). Similarly, there was no difference in the fibrinogen levels of the survivors and non-survivors (median-min/max; 116–26/305 vs. 226–36/367 mg/dL). We concluded that despite prolonged aPTT and INR, fibrinogen levels did not decrease significantly because it was a positive acute phase reactant.

Procalcitonin, which can be detected in circulation 2–6 h after adequate stimulation, has a clinical half-life of 20–24 h [34], and so the measurement of the time behavior of procalcitonin concentrations can be used to determine the clinical course after an induction event of limited duration in CCHF patients. As with the follow-up of sepsis, we can also say that follow-up of the overall daily procalcitonin measurements may be effective in the successful treatment and good prognosis in CCHF.

Although viral isolation is the standard in CCHF, there is still a high level of contamination risks, and only a limited number of laboratories can use this technique [35]. There is a lack of consensus on the most effective molecular or serological test method. Although there have been many studies into the prognosis of CCHF and the relationship between various biomarkers, access to these tests as routine laboratory requests by clinicians is difficult, and most of these tests are for research purposes. In this context, the prominence of short-term reported, easy-to-reach assistive diagnostic tests that provide support for clinical suspicion at the time of hospital admission is increasing. Further studies investigating the effects of simultaneous use of disease-specific different protein tests, quantitative laboratory parameters, and patient specific clinic data on procalcitonin test sensitivity and cut-off levels should be considered.

Procalcitonin values may serve as a prognostic indicator of the course of disease and an auxiliary biomarker to rule out the disease, although molecular and serological evaluations are necessary in CCHF. We therefore suggest procalcitonin as a clinically useful and supportive biomarker both in the diagnosis and follow-up of the disease, given the advantages of low cost, easy access, rapid results and simple methodology.