The severity of Puumala hantavirus induced nephropathia epidemica can be better evaluated using plasma interleukin-6 than C-reactive protein determinations

Nephropathia epidemica (NE) is a Scandinavian type of hemorrhagic fever with renal syndrome. The causative agent, Puumala virus (PUUV), is a member of the genus Hantavirus in the family Bunyaviridae [1]. Other hantaviruses causing more severe forms of HFRS include Hantaan, Seoul, and Dobrava viruses [2]. Many hantaviruses in North and South America, e.g. Sin Nombre, Andes, and Black Creek Canal viruses, cause hantavirus pulmonary syndrome (HPS) [2]. The natural host of PUUV is the bank vole (Myodes glareolus) [3].

Nephropathia epidemica is prevalent in Finland, elsewhere in Scandinavia, in Western Russia, in the Balkan region and also in many parts of Western Europe [2, 4]. Approximately 1000 serological diagnoses of PUUV infection are made in Finland annually [5]. However, the seroprevalence in the Finnish population is 5%, implying that most infections are subclinical or undiagnosed [5].

The clinical severity of NE varies greatly. Host genetics have been shown to influence the clinical picture [6, 7]. It is clinically characterized by acute fever, headache, back and abdominal pains, myalgia, nausea, vomiting, and transient myopia, while hemorrhages are uncommon [8, 9]. Renal involvement causes proteinuria, hematuria and oliguria, followed by polyuria [8, 9]. A minority of patients needs transient dialysis treatment [8], but complete recovery is the usual outcome [8, 9]. During the acute phase, an increase in the serum creatinine concentration, thrombocytopenia, anemia, leukocytosis, as well as moderately elevated erythrocyte sedimentation rate and C-reactive protein (CRP) values are typical laboratory findings [8, 9]. In addition, radiological pulmonary manifestations have been detected in 16-53% of the patients, and they have been associated with the degree of acute renal insufficiency [8–12].

The pathogenesis of NE is not completely understood. An important feature in hantavirus infections is universally increased capillary permeability [13], but the mechanisms of vascular leakage are unclear. PUUV causes no cytopathic effects in cultured cells but has a wide cell susceptibility in vitro [14]. It has been suggested that immunological factors including cytokines are involved in the pathogenesis of NE [2]. Increased cytokine levels have been found in the plasma, urine, and tissues of hantavirus infection patients [15–18]. Infection of cynomolgus macaques by PUUV also results in increased serum levels of several cytokines [19]. We have previously found plasma and urinary levels of interleukin (IL)-6, tumor necrosis factor-α, IL-1, and IL-1-receptor antagonist to be increased during the acute phase of NE, so that the observed levels of IL-6 were exceptionally high [15].

The IL-6 molecule is a multifunctional cytokine involved in immune responses and inflammation. IL-6 is the main inducer of CRP production in vitro in cultured human hepatoma cells [20], but data about the associations of IL-6 and CRP in vivo is scarce. The known main functions of CRP are complement activation, enhancement of phagocytosis, and induction of cytokine synthesis. Although plasma CRP level is widely used as an indicator of the severity of the disease in various infections, there are no reports associating high CRP levels to a severe disease in NE or other viral infections. On the other hand, IL-6 level has been found to be associated to the severity of the disease in some viral infections, e.g. influenza [21, 22]. Therefore, we studied whether plasma IL-6 or CRP levels are associated with the severity of NE, in order to evaluate if IL-6 or CRP are good markers for disease severity in NE.

The median age did not differ between patients with high plasma IL-6 and patients with low IL-6 (41 years, ranging 15-65 vs 39 years, ranging 17-71, P = 0.741). Forty-four (75%) of the patients with high IL-6 were males compared with 42 (71%) of the patients with low IL-6 (P = 0.679). In patients with high plasma CRP, the median age was higher than in patients with low CRP (46 years, ranging 25-71 vs 38 years, ranging 15-64, P < 0.001). Forty-two (72%) of the patients with high CRP were males compared with 44 (73%) of the patients with low CRP (P = 0.911).

The maximum level of plasma IL-6 associated strongly with several variables reflecting the severity of the disease (Figures 1 and 2). Patients who had high IL-6 had lower minimum urinary output (1040 vs 2180 ml/day, Figure 1A), lower minimum systolic blood pressure (110 vs 115 mmHg, Figure 1B), and a greater change in weight during hospital care (3.2 vs 2.0 kg, Figure 1C), and they also stayed longer in hospital than patients with low IL-6 (8 vs 6 days, Figure 1D). Furthermore, patients with high IL-6 had numerically (but not statistically significantly) higher maximum serum creatinine levels (242 vs 140 μmol/l, Figure 2A), higher maximum urinary protein excretion (2.51 vs 1.68 g/day, Figure 2B), higher maximum leukocyte count (11.9 vs 9.0 × 10⁹/l, Figure 2C), lower minimum platelet count (55 vs 80 × 10⁹/l, Figure 2D), and lower minimum hematocrit (0.34 vs 0.38, P = 0.001) than patients with low IL-6. The clinical or laboratory variables between the patients with high and low CRP did not differ (Figures 1F-H, 2E-H) with the exception of minimum urinary output. Patients with high CRP had slightly lower minimum urinary output when compared to patients with low CRP (1400 vs 1700 ml/day, Figure 1E).

The occurrence of a pathologic chest radiograph had no significant associations with high IL-6 or high CRP. Twenty patients (38%) with high IL-6 had a pathologic chest x-ray compared with 10 patients (24%) with low IL-6 (P = 0.147). Eighteen patients (37%) with high CRP had a pathologic chest x-ray compared with 12 patients (26%) with low CRP (P = 0.265).

There was a positive correlation between plasma CRP and plasma IL-6 levels (r = 0.323, P < 0.001). We also found a slight positive correlation between maximum plasma IL-6 and serum creatinine concentrations (r = 0.238, P = 0.010), whereas no correlation was found between maximum plasma CRP and serum creatinine levels.

Logistic regression analyses were then carried out to evaluate the associations of high CRP and high IL-6 with high serum creatinine, thrombocytopenia, or hospitalization exceeding seven days. High plasma IL-6 was revealed as an independent risk factor for high serum creatinine (OR 3.2, 95% CI 1.4-7.3, P = 0.005), whereas high plasma CRP was found to be a protective factor (OR 0.3, 95% CI 0.1-0.7, P = 0.009) in this model. High plasma IL-6 was also found to be an independent risk factor for thrombocytopenia and hospitalization exceeding seven days (OR 3.6, 95% CI 1.6-8.0, P = 0.002, and OR 4.5, 95% CI 1.9-10.8, P < 0.001, respectively) in models including high plasma IL-6, high plasma CRP, and age. High plasma CRP did not have a significant association with these factors (data not shown).

As the patients with NE sought medical assistance at different time intervals from the onset of fever, the plasma CRP and IL-6 samples were also taken at different periods from the onset. From the majority of patients (66%) we had both CRP and IL-6 samples taken 5 days from the onset of fever. In this subgroup of 78 patients the main results remained the same: when compared to patients with low IL-6, patients with high IL-6 had higher creatinine levels, higher leukocyte count, greater change in weight, as well as a lower platelet count, hematocrit, and urinary output. They also stayed longer in hospital than patients with low IL-6. High CRP five days after the onset of fever had no associations with the variables reflecting the severity of NE (data not shown).

The present study with 118 consecutive prospectively collected hospitalized patients is by far the largest study concerning inflammatory parameters, i.e. IL-6 and CRP, in acute Puumala hantavirus infection. The present data showed that high plasma IL-6 is associated with clinically severe NE. High IL-6 was found to be an independent risk factor for impaired renal function, thrombocytopenia, and longer hospitalization, when examined together with high CRP and age. Surprisingly, the results also suggested that high plasma CRP might have a protective effect on renal function, but the data must be interpreted with caution.

We found that maximum plasma IL-6 was associated with the severity of renal insufficiency, blood leukocytosis and thrombocytopenia. It also associated strongly with the duration of hospitalization. We have previously studied plasma IL-6 concentrations in a cohort of 70 NE patients and found IL-6 levels to be increased [15]. In that earlier study, there was no correlation between plasma IL-6 levels and serum creatinine, but in the present larger study we did find a correlation between the levels of these two variables.

Previously, Linderholm and co-workers have studied 15 NE patients and detected elevated IL-6 plasma levels in all cases [17]. They also found a significant correlation between maximum levels of IL-6 and serum creatinine in concordance with our results. Takala and co-workers have studied 19 NE patients and 13 patients with other viral infections and detected an inverse correlation of serum IL-6 concentrations in NE patients with mean arterial pressure and minimum platelet count [24]. Plasma levels of IL-6 have been reported to associate with the severity of the disease also in influenza and Japanese encephalitis [21, 22, 25]. Studies of IL-6 in respiratory syncytial virus infection and Dengue virus infection have produced controversial results [26–29]. In acute renal failure, it has been demonstrated that plasma IL-6 levels are elevated and that high levels predict mortality [30, 31].

As far as we know, this is the first report suggesting that CRP might act as a protective factor for renal function in infectious diseases. Previously, it has been shown in mice that CRP prevents and reverses proteinuria in accelerated nephrotoxic nephritis [32–34]. It has also been reported that genetics associated with reduced CRP production predispose to the development of systemic lupus erytematosus [35]. This has been attributed to the ability of CRP to prevent the deposition of immune complexes and enhance their phagocytosis. It should be noted that immune complexes have also been found in NE patients [36]. Reduced deposition and enhanced phagocytosis of immune complexes could be the mechanism by which high CRP protects renal function in NE. In viral infections, there are no reports of CRP concentrations affecting the severity of the disease. In bacteremia and sepsis, the results concerning CRP as a predictor of clinical outcome are controversial. A year 2005 review concludes that the ability of CRP level to reflect the severity of sepsis may be limited [37].

Finally, there was a positive correlation between maximum IL-6 and CRP levels in the present study, which can be explained by the fact that IL-6 induces the production of CRP. However, high IL-6 concentration was associated with more severe disease, whereas high CRP level, in contrast, was associated with less severe renal impairment. This finding can probably be explained by the diverse biological influences of IL-6 and CRP.

High plasma IL-6 concentration is associated with clinically severe acute NE and could be used as a marker of the severity of the disease. On the other hand, high CRP as such does not indicate a severe form of NE.