- Email: [email protected]
Hemorrhagic fever with renal syndrome in Albania. Focus on predictors of acute kidney injury in HFRS
Journal of Clinical Virology 91 (2017) 25–30
Contents lists available at ScienceDirect
Journal of Clinical Virology journal homepage: www.elsevier.com/locate/jcv
Full length article
Hemorrhagic fever with renal syndrome in Albania. Focus on predictors of acute kidney injury in HFRS Elvana Rista a,∗ , Arben Pilaca b , Ilir Akshija c , Ariol Rama f , Endri Harja b , Edmond Puca d , Silvia Bino g , Vilma Cadri e , Majlinda Kota h , Thereska Nestor e , Harxhi Arjan d a
Department of Nephrology, Hygeia Hospital Tirana, Tirana, Albania Department of Internal Medicine, Hygeia Hospital Tirana, Tirana, Albania c Department of Statistics, University Hospital Center “Mother Theresa”, Tirana, Albania d Department of Infectious Diseases, University Hospital Center “Mother Theresa”, Tirana, Albania e Department of Nephrology, University Hospital Center “Mother Theresa” Tirana, Albania f Department of Clinical Laboratory, Hygeia Hospital Tirana, Tirana, Albania g Department of Public Health, University Hospital “Mother Theresa”, Tirana, Albania h Biologist/virologist National Virology Laboratory Infectious Disease Control Department Public Health Institute, Tirana, Albania b
a r t i c l e
i n f o
Article history: Received 20 October 2016 Received in revised form 23 March 2017 Accepted 28 March 2017 Keywords: HFRS AKI Albania
a b s t r a c t Background: Hemorrhagic fever with renal syndrome (HFRS) is a rodent borne zoonosis, caused by the members of the family Bunyaviridae, genus Hantavirus. The main clinical features of the infection by this virus family are fever, thrombocytopenia and acute kidney injury. Objective: The aim of our study was to identify, for the first time, characteristic features of HFRS in the Albanian population. Study design: The study comprised 33 consecutive patients admitted with suspected HFRS from April 2011–April 2016 at one center. Clinical diagnosis was confirmed by ELISA and real-time PCR. Statistical analysis was performed to identify prognostic markers and indicators of disease severity. Results: The virus strain causing HFRS was Dobrava type in all 33 cases. The disease outbreaks occurred during the period June–July. Mean hospital stay was 15.7 ± 6.9 days. 29 (88%) of the patients were male. The mean age was 39.7 ± 14.1. 16 (48.5%) patients were from Northeast Albania. 8 (24.2%) patients required dialysis. The strongest correlation was the inverse relationship of nadir platelet count with urea and creatinine, p < 0.0001, p < 0.0079 respectively. Creatinine and hyponatremia were inversely correlated p = 0.0007, whereas hyponatremia and nadir platelet count had the highest sensitivity and specificity for development of severe AKI, 92.6%, 100%; 88.9%, 83.3% respectively. Mortality rate was 9.09%. Conclusion: HFRS is a severe viral disease in Albania caused by Dobrava strain. It is associated with high mortality, 9.09% in our cohort. In our study, thrombocytopenia, urinary volume, hyponatremia were indicators of more severe disease. © 2017 Elsevier B.V. All rights reserved.
1. Background HFRS is a rodent borne zoonosis caused by the members of the virus family Bunyaviridae, genus Hantavirus. Infection with Hantavirus may lead to life threatening renal disease. The members of the Hantavirus genus i.e Hantaan, Puumala, Seoul, Dobrava cause different forms of clinical disease [1,2]. Disease spread and distribution depends on chronic infection of vertebrates or recirculation of the viruses in nature via arthro-
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (E. Rista). http://dx.doi.org/10.1016/j.jcv.2017.03.021 1386-6532/© 2017 Elsevier B.V. All rights reserved.
podes to vertebrates, which in turn are influenced by ecological and climate factors. The hosts of viruses responsible for HFRS are chronically infected striped field mouse, yellow necked mouse, and rat or bank vole which spread HFRS in humans [3]. Viruses are most commonly transmitted to humans by inhalation of aerosols released by rodent excreta, respiratory secretions, aerosolized droplets of saliva, urine from infected rodents or by inhalation of contaminated dust particles. Infection by mouse bite is very rare [4]. Humans are usually the final hosts of the transmission chain. No documentation to date of infection from human-to-human has been reported. Dobrava virus is the cause of a severe form of HFRS in Albania, Croatia, Greece, Slovenia and other countries of the Balkan area [5].
26
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
Data show that mortality and morbidity rates depend on the strain of virus. It ranges from 5 to 15% in cases of Dobrava and Hantaan virus infection [6]. On the other hand, the condition caused by Puumala type has the best prognosis with a mortality rate <1% [6]. There is no sex or race predilection of the infection. The higher incidence observed in males might be explained by lifestyle habits particularly outdoor activities. Several studies suggest that immune response may play an important role on the pathogenesis of the disease, but it still remains largely unknown. Once infected, there is a sequence of immune cellular events that determine the clinical stages of the condition. Marked cytokine production, kallikrein-kinin activation, complement pathway activation, or increased levels of circulating immune complexes are detected. These are the main determinants of the febrile and hypotensive phases. Damage to the vascular endothelium, capillary dilatation, and leakage are clinically significant features of the disease. Vascular injury, the hallmark of the HFRS is brought about by immune complexes, which activate complement and trigger mediator release from platelets and inflammatory cells [1,14]. Clinically, these events are manifested with fever, hemorrhage, an increase in hematocrit and hemoglobin on admission, a decrease in serum albumin and serum protein levels followed by acute kidney injury [1]. The clinical course of severe HFRS can be divided into five phases: febrile, hypotensive (shock), oliguric, diuretic and convalescent [7]. Patients typically start to improve by the second week. Symptoms resolve gradually and renal functional tests start to normalize. It requires weeks to months for the kidney to completely gain normal functional activity. Acute kidney injury is a severe complication of the syndrome. An important role in the development of renal failure is played by the insult of vascular endothelium, tubular and interstitial damage by cytokines and other humoral factors [8]. Carefully managed supportive care is critical to recovery from severe disease. Treatment includes but it is not limited to careful management of the patient’s hydration status, electrolyte (e.g., sodium, potassium, chloride) levels, maintenance of optimal oxygenation, blood pressure levels and appropriate therapy aimed towards any secondary infections. Renal replacement therapy has a pivotal role in the management of the most critical cases complicated with acute renal failure. Early institution of RRT as indicated has been shown to improve outcomes [5]. Idealistically, the invention of a cost-effective vaccine would be the best form of prevention. Unfortunately, this is almost impossible due to the high antigenic heterogeneity of the viruses associated with the low disease incidence in regard to the general population. 2. Objectives To identify, for the first time, characteristic features of HFRS in the Albanian population. 3. Study design This is a prospective study. 33 consecutive patients were enrolled in the cohort from April 2011 to April 2016 at Infectious disease Clinic, University Center “Mother Teresa” Tirana, Albania. All patients had laboratory and clinical evidence highly suggestive of viral infection responsible for HFRS. Serology identification was performed by indirect-ELISA using Hantavirus IgG and IgM Dx select TM , catalogue number EL 1600M and EL 1600G. The results were confirmed using real time PCR using ABI 7500 based on the fluorogenic 5’nuclease assay. The tests were carried out at the National
Table 1 General, clinical characteristics and laboratory test of the study population. Age—yr Male— no (%) Female— no (%) Pregnant— no (%) LOS —days
39.7 29 4 2 15.7
±14.1 87.8% 12.2 50 ±6.9
Occupational distribution— no (%) Farmer Shepherd Beekeeper Woodcutter Tourist Mortality— no (%)
13 8 4 4 4 3
39.3 24.2 12.1 12.1 12.1 9.09
Clinical sign and symptoms Fever— no (%) Malaise— no (%) Low urine output— no (%) Headache— no (%) Abdominal pain— no (%) Myalgia —no (%) Back pain— no (%) Vomiting— no (%)
33 31 25 26 24 23 21 21
100 93.9 75.7 78.7 72.7 69.6 63.6 63.6
Laboratory test on admission Urea— (mg/dl) Creatinine— (mg/dl) Potassium— (mEq/l) Hemoglobin— (g/dl) Hematocrit— (%) WBC— (/L) PLT— (/L) AST— (IU/L) LDH— (IU/L)
107.1 2.68 3.61 14.78 43.1 9.6 44.6 107.9 536.2
±70 ±1.7 ±0.4 ±2.69 ±6.8 ±5.1 ±26.7 ±87.6 ±266
Plus-minus values are means ±SD.
Reference Laboratory of the Institute of Public Health, Tirana, Albania. Relevant clinical data i.e. body temperature, blood pressure, heart rate, respiratory rate, urinary volume, etc upon admission were recorded. Routine laboratory investigation was performed including complete blood count, liver functional tests (AST, ALT, bilirubine), creatine kinase, albumin, total protein level, amylase, lactate dehydrogenase, electrolytes and coagulation panel, arterial blood gases. AKI was defined according to RIFLE classification: risk (R), injury (I), failure (F), loss (L), end-stage kidney disease (E) and Acute Kidney Injury Network (AKIN): AKIN I, AKIN II, AKIN III [9,10]. Patients were grouped according to RIFLE and AKIN classification. 3.1. Statistical analyses Results are presented in descriptive tables and graphs. Mean annual incidence was calculated per 100,000 people per region. Continuous variables are presented in mean values and standard deviation. Mann-Whitney U test is used in comparing continuous variables. AKIN and RIFLE ordinal scales were the basis of comparisons. Spearman correlation coefficient and graphical presentation were used to represent laboratory findings. Two sided P values < 0.05 were considered statistically significant. Data were analyzed using Medcalc 14.8.1 and IBM SPSS 20. 4. Results A total of 33 patients were enrolled during a five year time frame from April 2011 to April 2016. Map shows (Fig. 1) annual incidence per 100,000 of Hemorrhagic fever with renal syndrome (HFRS). The mean (SD) age was 39.7 ± 14.1, ranging from 15 to 59 years. The number of cases peaked from July to August during the whole time span of the study, which corresponds with the usual disease outbreaks of the specific viral strains (Table 1).
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
27
Fig. 1. Geographic distribution of cases of HFRS in Albania.
Patient distribution (Fig. 2) according to RIFLE classification was: R 2(6%) pts, I 4(12%) pts, F 27(82%) pts and according to AKIN criteria was: I 4 pts (12.1%), II 5 pts (15.1%), III 24(72.7%).
The clinical characteristics of the patients according the RIFLE and AKIN classification are shown in the Table 2. Characteristics of dialysis and non-dialysis groups are shown in Table 3.
28
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
Fig. 2. Patient distribution according to RIFLE and KIN classification.
Table 2 RIFLE and AKIN classification.
Age Male sex LOS(day) BP(mmHg)sys on admission BP(mmHg)dias admission Minimal urinary volume Hb(g/dl) admission Hct (%) admission Na(mmol/L) admission Platelet min(x103 /l) WBC(x103 /l) HCO3 − (mmol/L) Mortality Albumin(g/dl) LDH (IU/L)
Non severe (R + I)
Severe (F)
P value
Non severe (I + II)
Severe (III)
P value
49.3 ± 7.39 24.2% 10.1 ± 3.48 120 ± 11.4 77.6 ± 7.5 1216.6 ± 440 12.7 ± 1.26 39.6 ± 4.5 134.8 ± 1.16 62.5 ± 26.9 10.6 ± 2.8 20.4 ± 1.4 0 3.1 ± 0.18
37.6 ± 14.4 75.8% 16.9 ± 6.98 106.8 ± 19.7 67 ± 12.3 696.9 ± 532.9 15.2 ± 2.7 43.9 ± 7 125.8 ± 4.4 33.7 ± 18 14.9 ± 4.0 19.3 ± 3.3 3(11%) 2.6 ± 0.3
0.68
45.7 ± 13.6 24.2% 10.5 ± 3.2 116.2± 74.5 ± 9.0 1287 ± 476 13.08 ± 1.4 40.1 ± 4.2 132 ± 4.9 64 ± 23.6 11.5 ± 3.2 20.6 ± 1.3 0 3.1 ± 017 340 ± 130.9
37.8 ± 13.9 75.8% 17.4 ± 7 106.9 ± 20.4 67.2 ± 12.75 508 ± 398 15.3 ± 2.7 44.1 ± 7.2 125.7 ± 4.2 38.4 ± 24.9 14.9 ± 4.2 19.1 ± 3.4 3(12%) 2.6 ± 0.3 605.2 ± 269.3
0.1
0.001 0.029 0.035 0.03 0.01 0.01 0.0002 0.01 0.012 0.03 0.0005
0.0043 0.01 0.1 0.001 0.02 0.07 0.001 0.013 0.04 0.2 0.0004 0.02
Table 3 Characteristics of dialysis vs non-dialysis. Variable
Dialysis (n = 8)
Urine output min(ml/day) PLTa (x103 /l) WBCa (x103 /l) Albumina (g/dl) Hcta (%) Hba (g/dl) LDHa (IU/L) Na+ a (mmol/L) CPKa (IU/L)
n 8 8 8 8 8 8 8 8 8
a
No dialysis (n = 25) Mean ± SD 175 ± 175.2 29250 ± 24370 13675 ± 8656 2.45 ± 0.22 49.6 ± 6.4 17.6 ± 2.8 757 ± 275.4 123.4 ± 5.4 1276 ± 1045
P value Mean ± SD 864 ± 499 49668 ± 26049 8408 ± 2628 2.9 ± 0.31 41.1 ± 5.6 13.8 ± 1.9 391.6 ± 188.1 128.8 ± 4.7 398 ± 356
n 25 25 25 25 25 25 25 25 17
0.0004 0.033 0.09 0.0009 0.0025 0.001 0.003 0.023 0.04
value on admission.
Statistical analysis using Spearman correlation and linear regression showed strong association of platelet nadir with several clinical and laboratory data. In our study, thrombocytopenia was a ubiquitous and consistent finding in all enrolled patients with documented HFRS. The strongest correlation was the inverse relationship with urea (P < 0.0001, r = −0.686) and creatinine (P < 0.0079, r = −0.455). Other relevant associations among platelet count and WBC (P = 0.0176, r = −0.440), LOS (P = 0.045, r = −0.351), hypoalbuminemia (P = 0.0217, r = 0.398), hematocrit (P = 0.0176, r = 0.411), hyponatremia (P = 0.0021, r = 0.516) were found (Fig. 3). There was a very strong correlation of the inverse relationship between creatinine and hyponatremia (P = 0.0007, r = −0.560). ROC curves (Fig. 4) were used to assess the prediction accuracy for developing severe AKI of three markers, i.e.: sodium level, nadir platelet and zenith white blood cell count. Area Under the Curve (AUC) quantified the accuracy. AUC for nadir platelet count was 0.957. The nadir platelet count analysis showed a specificity of 83.3% and sensitivity of 88.9% at 52 × 103 /l cut-off value to pre-
Table 4 Sensitivity, specificity and predictive values for the development of severe AKI using sodium, nadir platelet and zenith WBC count cut-off values.
Admission Na+ (mmol/l) Nadir PLT(x103 /l) Peak WBC(x103 /l)
AUC
Cut-off value
Sensitivity
Specificity
0.957 0.833 0.830
≤133 ≤52 >10
92.6% 88.9% 96.3%
100% 83.3% 66.7%
dict development of severe AKI. AUC for sodium level on admission was 0.957. Sodium level on admission had a 100% specificity and 92.6% sensitivity at a cut-off value of 133 mmol/L. AUC for zenith WBC count was 0.830. At a cut-off value of 10 × 103 /l, sensitivity and specificity were 96.3% and 66.7%, respectively (Table 4). After adjustment for hemoglobin and hematocrit, nadir platelet count (odds ratio 0.9435, 95% CI, 0.8965–0.9929, p = 0.0255) remained an independent predictor of AKI (Table 5).
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
29
Table 5 Multivariable logistic regression for the prediction of severe AKI.
Hemoglobin minimum Hematocrit minimum PLT minimum
Odds ratio
95% CI
P value
0.8583 0.9880 0.9435
0.6052–1.2171 0.6965–1.4015 0.8965–0.9929
0.3911 0.9461 0.0255
Serological assay (indirect-ELISA) detected positive IgM and IgG anti Hantavirus. Virus serotype, detected by the real time (RT-PCR) was DOBRAVA. 5. Discussion
Fig. 3. Spearman correlation and linear regression analysis of laboratory and clinical data.
100
Sens itivity
80
60
40
Admission sodium Nadir platelets Peak leukocytes
20
0 0
20
40
60
80
100
100-Specificity
Fig. 4. Receiver-operating-characteristics (ROC) curve for laboratory data to predict the development of AKI.
HFRS is a severe clinical syndrome that requires highly specialized supportive care. The majority of cases were concentrated in the North-East region of the country, typical for the disease distribution. Interestingly, there were few cases, first-time diagnosed from other areas of the country, which may suggest virus habitat changes that should alert the physicians to consider HFRS in the differential diagnoses of patients from non-typical areas with specific symptoms of this condition. Most of the patients were relatively young and their occupational history suggested close contacts with the animals responsible for the spread of the zoonosis. Disease outbreaks were observed in July and August, starting from May throughout the second half of the year. Relevant clinical data, laboratory investigation upon admission were recorded and statistically analyzed in an effort to identify predictors of severity of AKI and need for renal replacement therapy. AKI was the most frequent and life-threatening complication of the patients with HFRS in this study. AKI occurred in 100% of patients with HFRS. Most of the patients were in the advanced stage of renal disease. Patients requiring hemodialysis had statistically significant thrombocytopenia (p = 0.009), higher hemoglobin on admission (p = 0.39), hypoalbuminemia (p = 0.003) and lower urinary volume ml/day (p = 0.004). Higher hemoglobin and hematocrit on admission values correlate well with the severity of disease. And of course urinary volume is a direct indicator of kidney function. Hyponatremia is an important risk factor for the development of acute kidney injury. It may indicate an early stage of the renal damage by the viral infection [11]. It is an early finding in the course of the disease and may guide treatment decision plan. Thrombocytopenia was a consistent finding along with renal function impairment in the entire patient cohort. Platelet count was a predictor and marker of disease severity and progression as well [12,13]. This may guide medical staff to treat this group of patients accordingly. An increase in platelet count associated with other clinical findings during the course of the disease may identify a positive turning point in the evolution of the condition status. Clinical data analysis showed that early institution of dialysis in cases of severe AKI contributed in reducing the rate of severe complications and mortality. However these variables should be taken into consideration in the general context of the condition, including other clinical manifestations, i.e. hemorrhage, fever, shock, etc. Our statistical analysis showed a strong correlation between creatinine and hyponatremia (r = 0.560, p = 0.0007), nadir platelet count and maximal urea levels (r = 0.686, p < 0.0001), platelet count and minimum hematocrit levels (r = 0.411, p = 0.0176), platelet count and Na+ levels on admission (r = 0.516, p = 0.0021), platelets and maximum WBC count (r = 0.440, p = 0.0104). As the results show, platelet count correlates well with several other blood tests. Thus, it serves as a marker of disease severity and may predict prognosis of the condition. There were three deaths in our cohort (9%), one female and two male patients. Dialysis was performed in two of them.
30
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
Additionally, there were two cases of pregnant women who both lost their fetuses in 16th week gestational age, as determined by ultrasound measurements. Fetal death coincided with peak severity of the disease and nadir platelet count. Neither of them needed dialysis. In our knowledge, there is a paucity of data on how to offer the appropriate available treatment to both mother and the baby [15]. Further studies are needed in order to define the best therapeutic approach in these rare cases. The study cohort is small and data should be interpreted carefully. However they are in line with other robust findings by other studies. There are several studies on HFRS in the Europe and the Balkan area more specifically. They show differences in viral strains prevalent in Europe, Northern and Southern Balkan. Dobrava is more prevalent in the south of Balkans, whereas Puumala strain is predominant in the rest of European countries. In Slovenia and Croatia, more than 70% of cases are caused by Puumala virus (PUUV). In Albania, Greece, Macedonia and Serbia, the condition is caused almost exclusively by Dobrava [16]. Dobrava infection is associated with higher mortality in comparison with PUUV. In our study, thrombocytopenia was identified as an independent predictor for developing acute kidney injury. In one study, in Croatia the authors did not include thrombocytopenia in their multivariate analysis. Instead, they identified hyponatremia and proteinuria as risk factors for oliguria and anuria [11]. Zupanc et al. showed that conjunctival bleeding, diarrhea, serum sodium of 133 mmol/L, dipstick urine value of >1.5 g/dL were risk factors for development of oliguria and anuria. [17]. Thrombocytopenia, urinary volume and hyponatremia were indicators of more severe disease in our study population. We believe these differences are related to the specific viral strain as well as genetic influence on immune response of Albanian patients. A risk score exists for calculating the risk of developing severe acute kidney injury depending on the presence or absence of three parameters (thrombocytopenia, elevated CRP, proteinuria) [18]. Given the specifics of the Albanian healthcare system, it cannot be ruled out that less severe or fulminant cases can go undetected in remote areas. Thus a selection bias toward moderate-to-severe cases may exist. There was a clear underestimation of creatinine values in patients who were treated with dialysis. 6. Conclusion HFRS is a severe viral disease in Albania, associated with high mortality, 9% in our cohort. Early diagnosis is paramount to treatment of HFRS. In our study, thrombocytopenia, urinary volume, hyponatremia were indicators of more severe disease. Nadir platelet count was an independent predictor for development of AKI. Interestingly, the two cases of pregnant patients resulted in fetal loss. Further studies are needed in order to elucidate the injury mechanism of the virus to the fetus. Funding None Competing interests None
Ethical approval Not required Acknowledgements The authors are grateful to the physician and nurses of Infectious Disease Clinic, Hemodialysis Clinic, University Center “Mother Teresa” Tirana, physician of the Institute of Public Health in Albania. References [1] T.M. Cosgriff, R.M. Lewis, Mechanisms of disease in hemorrhagic fever with renal syndrome, Kidney Int. Suppl. 2 (December (35)) (1991). [2] W. Muranyi, U. Bahr, M. Zeier, F.J. van der Woude, Hantavirus infection, J. Am. Soc. Nephrol. 16 (December (12)) (2005) 3669–3679. [3] C. Peters, California encephalitis, hantavirus pulmonary syndrome, and bunyavirid hemorrhagic fevers, in: G.L. Mandell, J.E. Bennett, R. Dolin (Eds.), Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 7th ed., Churchill Livingstone Elsevier, Philadelphia, Pennsylvania, 2010, pp. 2289–2293. [4] Rista Elvana, et al., Virol-mycol 5 (Suppl. 1) (2016), http://dx.doi.org/10.4172/ 2161-0517.C1.009. [5] C.F. Fulhorst, F.T. Koster, D.A. Enria, C.J. Peters, in: R.L. Guerrant, D.H. Walker, P.R. Weller (Eds.), Tropical Infectious Diseases Principles, Pathogens and Practice, 3rd edition, Elsevier, Edinburgh, 2011, pp. 470–480. [6] C. Schmaljohn, B. Hjelle, Hantaviruses a global disease problem, Emerg. Infect. Dis. 3 (April (2)) (1997) 95. [7] H. Jiang, H. Du, L.M. Wang, P.Z. Wang, X.F. Bai, Hemorrhagic fever with renal syndrome: pathogenesis and clinical picture, Front. Cell. Infect. Microbiol. 6 (2016). [8] M. Temonen, J. Mustonen, H. Helin, A. Pasternack, A. Vaheri, H. Holthöfer, Cytokines, adhesion molecules, and cellular infiltration in nephropathia epidemica kidneys: an immunohistochemical study, Clin. Immunol. Immunopathol. 78 (January (1)) (1996) 47–55. [9] R. Bellomo, J.A. Kellum, C. Ronco, Defining and classifying acute renal failure: from advocacy to consensus and validation of the RIFLE criteria, Intensive Care Med. 33 (March (3)) (2007) 409–413. [10] R.L. Mehta, J.A. Kellum, S.V. Shah, B.A. Molitoris, C. Ronco, D.G. Warnock, A. Levin, Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury, Crit. Care 11 (march (2)) (2007) 1. ´ I. Kuzman, J. Begovac, Clinical and [11] D. Turˇcinov, I. Puljiz, A. Markotic, laboratory findings in patients with oliguric and nonˇcoliguric Hantavirus haemorrhagic fever with renal syndrome: an analysis of 128 patients, Clin. Microbiol. Infect. 19 (July (7)) (2013) 674–679. [12] M. Wang, J. Wang, T. Wang, J. Li, L. Hui, X. Ha, Thrombocytopenia as a predictor of severe acute kidney injury in patients with Hantaan virus infections, PLoS One 8 (January (1)) (2013) e53236. [13] F.M. Rasche, B. Uhel, D.H. Kruger, W. Karges, D. Czock, W. Hampl, F. Keller, H. Meisel, L. von Muller, Thrombocytopenia and acute renal failure in Puumala hantavirus infections, Emerg. Infect. Dis. 10 (August (1)) (2004) 1420–1425. [14] D. Ferluga, A. Vizjak, Hantavirus nephropathy, J. Am. Soc. Nephrol. 19 (September (9)) (2008) 1653–1658. [15] B.N. Kim, B.D. Choi, Hemorrhagic fever with renal syndrome complicated with pregnancy: a case report, Kor. J. Internal Med. 21 (June (2)) (2006) 150–153. [16] Zupanc, et al., HFRS and hantaviruses in the Balkans/South-East, Europe-virus Res. 187 (1) (2014) 27–33. [17] Zupanc, et al., Virus Res. 187 (2014) 27–33. [18] Latus, et al., Acute kidney injury and tools for risk-stratification in 456 patients with hantavirus-induced nephropathia endemic, Nephrol. Dial. Transplant (2014) 1–7.
Contents lists available at ScienceDirect
Journal of Clinical Virology journal homepage: www.elsevier.com/locate/jcv
Full length article
Hemorrhagic fever with renal syndrome in Albania. Focus on predictors of acute kidney injury in HFRS Elvana Rista a,∗ , Arben Pilaca b , Ilir Akshija c , Ariol Rama f , Endri Harja b , Edmond Puca d , Silvia Bino g , Vilma Cadri e , Majlinda Kota h , Thereska Nestor e , Harxhi Arjan d a
Department of Nephrology, Hygeia Hospital Tirana, Tirana, Albania Department of Internal Medicine, Hygeia Hospital Tirana, Tirana, Albania c Department of Statistics, University Hospital Center “Mother Theresa”, Tirana, Albania d Department of Infectious Diseases, University Hospital Center “Mother Theresa”, Tirana, Albania e Department of Nephrology, University Hospital Center “Mother Theresa” Tirana, Albania f Department of Clinical Laboratory, Hygeia Hospital Tirana, Tirana, Albania g Department of Public Health, University Hospital “Mother Theresa”, Tirana, Albania h Biologist/virologist National Virology Laboratory Infectious Disease Control Department Public Health Institute, Tirana, Albania b
a r t i c l e
i n f o
Article history: Received 20 October 2016 Received in revised form 23 March 2017 Accepted 28 March 2017 Keywords: HFRS AKI Albania
a b s t r a c t Background: Hemorrhagic fever with renal syndrome (HFRS) is a rodent borne zoonosis, caused by the members of the family Bunyaviridae, genus Hantavirus. The main clinical features of the infection by this virus family are fever, thrombocytopenia and acute kidney injury. Objective: The aim of our study was to identify, for the first time, characteristic features of HFRS in the Albanian population. Study design: The study comprised 33 consecutive patients admitted with suspected HFRS from April 2011–April 2016 at one center. Clinical diagnosis was confirmed by ELISA and real-time PCR. Statistical analysis was performed to identify prognostic markers and indicators of disease severity. Results: The virus strain causing HFRS was Dobrava type in all 33 cases. The disease outbreaks occurred during the period June–July. Mean hospital stay was 15.7 ± 6.9 days. 29 (88%) of the patients were male. The mean age was 39.7 ± 14.1. 16 (48.5%) patients were from Northeast Albania. 8 (24.2%) patients required dialysis. The strongest correlation was the inverse relationship of nadir platelet count with urea and creatinine, p < 0.0001, p < 0.0079 respectively. Creatinine and hyponatremia were inversely correlated p = 0.0007, whereas hyponatremia and nadir platelet count had the highest sensitivity and specificity for development of severe AKI, 92.6%, 100%; 88.9%, 83.3% respectively. Mortality rate was 9.09%. Conclusion: HFRS is a severe viral disease in Albania caused by Dobrava strain. It is associated with high mortality, 9.09% in our cohort. In our study, thrombocytopenia, urinary volume, hyponatremia were indicators of more severe disease. © 2017 Elsevier B.V. All rights reserved.
1. Background HFRS is a rodent borne zoonosis caused by the members of the virus family Bunyaviridae, genus Hantavirus. Infection with Hantavirus may lead to life threatening renal disease. The members of the Hantavirus genus i.e Hantaan, Puumala, Seoul, Dobrava cause different forms of clinical disease [1,2]. Disease spread and distribution depends on chronic infection of vertebrates or recirculation of the viruses in nature via arthro-
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (E. Rista). http://dx.doi.org/10.1016/j.jcv.2017.03.021 1386-6532/© 2017 Elsevier B.V. All rights reserved.
podes to vertebrates, which in turn are influenced by ecological and climate factors. The hosts of viruses responsible for HFRS are chronically infected striped field mouse, yellow necked mouse, and rat or bank vole which spread HFRS in humans [3]. Viruses are most commonly transmitted to humans by inhalation of aerosols released by rodent excreta, respiratory secretions, aerosolized droplets of saliva, urine from infected rodents or by inhalation of contaminated dust particles. Infection by mouse bite is very rare [4]. Humans are usually the final hosts of the transmission chain. No documentation to date of infection from human-to-human has been reported. Dobrava virus is the cause of a severe form of HFRS in Albania, Croatia, Greece, Slovenia and other countries of the Balkan area [5].
26
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
Data show that mortality and morbidity rates depend on the strain of virus. It ranges from 5 to 15% in cases of Dobrava and Hantaan virus infection [6]. On the other hand, the condition caused by Puumala type has the best prognosis with a mortality rate <1% [6]. There is no sex or race predilection of the infection. The higher incidence observed in males might be explained by lifestyle habits particularly outdoor activities. Several studies suggest that immune response may play an important role on the pathogenesis of the disease, but it still remains largely unknown. Once infected, there is a sequence of immune cellular events that determine the clinical stages of the condition. Marked cytokine production, kallikrein-kinin activation, complement pathway activation, or increased levels of circulating immune complexes are detected. These are the main determinants of the febrile and hypotensive phases. Damage to the vascular endothelium, capillary dilatation, and leakage are clinically significant features of the disease. Vascular injury, the hallmark of the HFRS is brought about by immune complexes, which activate complement and trigger mediator release from platelets and inflammatory cells [1,14]. Clinically, these events are manifested with fever, hemorrhage, an increase in hematocrit and hemoglobin on admission, a decrease in serum albumin and serum protein levels followed by acute kidney injury [1]. The clinical course of severe HFRS can be divided into five phases: febrile, hypotensive (shock), oliguric, diuretic and convalescent [7]. Patients typically start to improve by the second week. Symptoms resolve gradually and renal functional tests start to normalize. It requires weeks to months for the kidney to completely gain normal functional activity. Acute kidney injury is a severe complication of the syndrome. An important role in the development of renal failure is played by the insult of vascular endothelium, tubular and interstitial damage by cytokines and other humoral factors [8]. Carefully managed supportive care is critical to recovery from severe disease. Treatment includes but it is not limited to careful management of the patient’s hydration status, electrolyte (e.g., sodium, potassium, chloride) levels, maintenance of optimal oxygenation, blood pressure levels and appropriate therapy aimed towards any secondary infections. Renal replacement therapy has a pivotal role in the management of the most critical cases complicated with acute renal failure. Early institution of RRT as indicated has been shown to improve outcomes [5]. Idealistically, the invention of a cost-effective vaccine would be the best form of prevention. Unfortunately, this is almost impossible due to the high antigenic heterogeneity of the viruses associated with the low disease incidence in regard to the general population. 2. Objectives To identify, for the first time, characteristic features of HFRS in the Albanian population. 3. Study design This is a prospective study. 33 consecutive patients were enrolled in the cohort from April 2011 to April 2016 at Infectious disease Clinic, University Center “Mother Teresa” Tirana, Albania. All patients had laboratory and clinical evidence highly suggestive of viral infection responsible for HFRS. Serology identification was performed by indirect-ELISA using Hantavirus IgG and IgM Dx select TM , catalogue number EL 1600M and EL 1600G. The results were confirmed using real time PCR using ABI 7500 based on the fluorogenic 5’nuclease assay. The tests were carried out at the National
Table 1 General, clinical characteristics and laboratory test of the study population. Age—yr Male— no (%) Female— no (%) Pregnant— no (%) LOS —days
39.7 29 4 2 15.7
±14.1 87.8% 12.2 50 ±6.9
Occupational distribution— no (%) Farmer Shepherd Beekeeper Woodcutter Tourist Mortality— no (%)
13 8 4 4 4 3
39.3 24.2 12.1 12.1 12.1 9.09
Clinical sign and symptoms Fever— no (%) Malaise— no (%) Low urine output— no (%) Headache— no (%) Abdominal pain— no (%) Myalgia —no (%) Back pain— no (%) Vomiting— no (%)
33 31 25 26 24 23 21 21
100 93.9 75.7 78.7 72.7 69.6 63.6 63.6
Laboratory test on admission Urea— (mg/dl) Creatinine— (mg/dl) Potassium— (mEq/l) Hemoglobin— (g/dl) Hematocrit— (%) WBC— (/L) PLT— (/L) AST— (IU/L) LDH— (IU/L)
107.1 2.68 3.61 14.78 43.1 9.6 44.6 107.9 536.2
±70 ±1.7 ±0.4 ±2.69 ±6.8 ±5.1 ±26.7 ±87.6 ±266
Plus-minus values are means ±SD.
Reference Laboratory of the Institute of Public Health, Tirana, Albania. Relevant clinical data i.e. body temperature, blood pressure, heart rate, respiratory rate, urinary volume, etc upon admission were recorded. Routine laboratory investigation was performed including complete blood count, liver functional tests (AST, ALT, bilirubine), creatine kinase, albumin, total protein level, amylase, lactate dehydrogenase, electrolytes and coagulation panel, arterial blood gases. AKI was defined according to RIFLE classification: risk (R), injury (I), failure (F), loss (L), end-stage kidney disease (E) and Acute Kidney Injury Network (AKIN): AKIN I, AKIN II, AKIN III [9,10]. Patients were grouped according to RIFLE and AKIN classification. 3.1. Statistical analyses Results are presented in descriptive tables and graphs. Mean annual incidence was calculated per 100,000 people per region. Continuous variables are presented in mean values and standard deviation. Mann-Whitney U test is used in comparing continuous variables. AKIN and RIFLE ordinal scales were the basis of comparisons. Spearman correlation coefficient and graphical presentation were used to represent laboratory findings. Two sided P values < 0.05 were considered statistically significant. Data were analyzed using Medcalc 14.8.1 and IBM SPSS 20. 4. Results A total of 33 patients were enrolled during a five year time frame from April 2011 to April 2016. Map shows (Fig. 1) annual incidence per 100,000 of Hemorrhagic fever with renal syndrome (HFRS). The mean (SD) age was 39.7 ± 14.1, ranging from 15 to 59 years. The number of cases peaked from July to August during the whole time span of the study, which corresponds with the usual disease outbreaks of the specific viral strains (Table 1).
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
27
Fig. 1. Geographic distribution of cases of HFRS in Albania.
Patient distribution (Fig. 2) according to RIFLE classification was: R 2(6%) pts, I 4(12%) pts, F 27(82%) pts and according to AKIN criteria was: I 4 pts (12.1%), II 5 pts (15.1%), III 24(72.7%).
The clinical characteristics of the patients according the RIFLE and AKIN classification are shown in the Table 2. Characteristics of dialysis and non-dialysis groups are shown in Table 3.
28
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
Fig. 2. Patient distribution according to RIFLE and KIN classification.
Table 2 RIFLE and AKIN classification.
Age Male sex LOS(day) BP(mmHg)sys on admission BP(mmHg)dias admission Minimal urinary volume Hb(g/dl) admission Hct (%) admission Na(mmol/L) admission Platelet min(x103 /l) WBC(x103 /l) HCO3 − (mmol/L) Mortality Albumin(g/dl) LDH (IU/L)
Non severe (R + I)
Severe (F)
P value
Non severe (I + II)
Severe (III)
P value
49.3 ± 7.39 24.2% 10.1 ± 3.48 120 ± 11.4 77.6 ± 7.5 1216.6 ± 440 12.7 ± 1.26 39.6 ± 4.5 134.8 ± 1.16 62.5 ± 26.9 10.6 ± 2.8 20.4 ± 1.4 0 3.1 ± 0.18
37.6 ± 14.4 75.8% 16.9 ± 6.98 106.8 ± 19.7 67 ± 12.3 696.9 ± 532.9 15.2 ± 2.7 43.9 ± 7 125.8 ± 4.4 33.7 ± 18 14.9 ± 4.0 19.3 ± 3.3 3(11%) 2.6 ± 0.3
0.68
45.7 ± 13.6 24.2% 10.5 ± 3.2 116.2± 74.5 ± 9.0 1287 ± 476 13.08 ± 1.4 40.1 ± 4.2 132 ± 4.9 64 ± 23.6 11.5 ± 3.2 20.6 ± 1.3 0 3.1 ± 017 340 ± 130.9
37.8 ± 13.9 75.8% 17.4 ± 7 106.9 ± 20.4 67.2 ± 12.75 508 ± 398 15.3 ± 2.7 44.1 ± 7.2 125.7 ± 4.2 38.4 ± 24.9 14.9 ± 4.2 19.1 ± 3.4 3(12%) 2.6 ± 0.3 605.2 ± 269.3
0.1
0.001 0.029 0.035 0.03 0.01 0.01 0.0002 0.01 0.012 0.03 0.0005
0.0043 0.01 0.1 0.001 0.02 0.07 0.001 0.013 0.04 0.2 0.0004 0.02
Table 3 Characteristics of dialysis vs non-dialysis. Variable
Dialysis (n = 8)
Urine output min(ml/day) PLTa (x103 /l) WBCa (x103 /l) Albumina (g/dl) Hcta (%) Hba (g/dl) LDHa (IU/L) Na+ a (mmol/L) CPKa (IU/L)
n 8 8 8 8 8 8 8 8 8
a
No dialysis (n = 25) Mean ± SD 175 ± 175.2 29250 ± 24370 13675 ± 8656 2.45 ± 0.22 49.6 ± 6.4 17.6 ± 2.8 757 ± 275.4 123.4 ± 5.4 1276 ± 1045
P value Mean ± SD 864 ± 499 49668 ± 26049 8408 ± 2628 2.9 ± 0.31 41.1 ± 5.6 13.8 ± 1.9 391.6 ± 188.1 128.8 ± 4.7 398 ± 356
n 25 25 25 25 25 25 25 25 17
0.0004 0.033 0.09 0.0009 0.0025 0.001 0.003 0.023 0.04
value on admission.
Statistical analysis using Spearman correlation and linear regression showed strong association of platelet nadir with several clinical and laboratory data. In our study, thrombocytopenia was a ubiquitous and consistent finding in all enrolled patients with documented HFRS. The strongest correlation was the inverse relationship with urea (P < 0.0001, r = −0.686) and creatinine (P < 0.0079, r = −0.455). Other relevant associations among platelet count and WBC (P = 0.0176, r = −0.440), LOS (P = 0.045, r = −0.351), hypoalbuminemia (P = 0.0217, r = 0.398), hematocrit (P = 0.0176, r = 0.411), hyponatremia (P = 0.0021, r = 0.516) were found (Fig. 3). There was a very strong correlation of the inverse relationship between creatinine and hyponatremia (P = 0.0007, r = −0.560). ROC curves (Fig. 4) were used to assess the prediction accuracy for developing severe AKI of three markers, i.e.: sodium level, nadir platelet and zenith white blood cell count. Area Under the Curve (AUC) quantified the accuracy. AUC for nadir platelet count was 0.957. The nadir platelet count analysis showed a specificity of 83.3% and sensitivity of 88.9% at 52 × 103 /l cut-off value to pre-
Table 4 Sensitivity, specificity and predictive values for the development of severe AKI using sodium, nadir platelet and zenith WBC count cut-off values.
Admission Na+ (mmol/l) Nadir PLT(x103 /l) Peak WBC(x103 /l)
AUC
Cut-off value
Sensitivity
Specificity
0.957 0.833 0.830
≤133 ≤52 >10
92.6% 88.9% 96.3%
100% 83.3% 66.7%
dict development of severe AKI. AUC for sodium level on admission was 0.957. Sodium level on admission had a 100% specificity and 92.6% sensitivity at a cut-off value of 133 mmol/L. AUC for zenith WBC count was 0.830. At a cut-off value of 10 × 103 /l, sensitivity and specificity were 96.3% and 66.7%, respectively (Table 4). After adjustment for hemoglobin and hematocrit, nadir platelet count (odds ratio 0.9435, 95% CI, 0.8965–0.9929, p = 0.0255) remained an independent predictor of AKI (Table 5).
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
29
Table 5 Multivariable logistic regression for the prediction of severe AKI.
Hemoglobin minimum Hematocrit minimum PLT minimum
Odds ratio
95% CI
P value
0.8583 0.9880 0.9435
0.6052–1.2171 0.6965–1.4015 0.8965–0.9929
0.3911 0.9461 0.0255
Serological assay (indirect-ELISA) detected positive IgM and IgG anti Hantavirus. Virus serotype, detected by the real time (RT-PCR) was DOBRAVA. 5. Discussion
Fig. 3. Spearman correlation and linear regression analysis of laboratory and clinical data.
100
Sens itivity
80
60
40
Admission sodium Nadir platelets Peak leukocytes
20
0 0
20
40
60
80
100
100-Specificity
Fig. 4. Receiver-operating-characteristics (ROC) curve for laboratory data to predict the development of AKI.
HFRS is a severe clinical syndrome that requires highly specialized supportive care. The majority of cases were concentrated in the North-East region of the country, typical for the disease distribution. Interestingly, there were few cases, first-time diagnosed from other areas of the country, which may suggest virus habitat changes that should alert the physicians to consider HFRS in the differential diagnoses of patients from non-typical areas with specific symptoms of this condition. Most of the patients were relatively young and their occupational history suggested close contacts with the animals responsible for the spread of the zoonosis. Disease outbreaks were observed in July and August, starting from May throughout the second half of the year. Relevant clinical data, laboratory investigation upon admission were recorded and statistically analyzed in an effort to identify predictors of severity of AKI and need for renal replacement therapy. AKI was the most frequent and life-threatening complication of the patients with HFRS in this study. AKI occurred in 100% of patients with HFRS. Most of the patients were in the advanced stage of renal disease. Patients requiring hemodialysis had statistically significant thrombocytopenia (p = 0.009), higher hemoglobin on admission (p = 0.39), hypoalbuminemia (p = 0.003) and lower urinary volume ml/day (p = 0.004). Higher hemoglobin and hematocrit on admission values correlate well with the severity of disease. And of course urinary volume is a direct indicator of kidney function. Hyponatremia is an important risk factor for the development of acute kidney injury. It may indicate an early stage of the renal damage by the viral infection [11]. It is an early finding in the course of the disease and may guide treatment decision plan. Thrombocytopenia was a consistent finding along with renal function impairment in the entire patient cohort. Platelet count was a predictor and marker of disease severity and progression as well [12,13]. This may guide medical staff to treat this group of patients accordingly. An increase in platelet count associated with other clinical findings during the course of the disease may identify a positive turning point in the evolution of the condition status. Clinical data analysis showed that early institution of dialysis in cases of severe AKI contributed in reducing the rate of severe complications and mortality. However these variables should be taken into consideration in the general context of the condition, including other clinical manifestations, i.e. hemorrhage, fever, shock, etc. Our statistical analysis showed a strong correlation between creatinine and hyponatremia (r = 0.560, p = 0.0007), nadir platelet count and maximal urea levels (r = 0.686, p < 0.0001), platelet count and minimum hematocrit levels (r = 0.411, p = 0.0176), platelet count and Na+ levels on admission (r = 0.516, p = 0.0021), platelets and maximum WBC count (r = 0.440, p = 0.0104). As the results show, platelet count correlates well with several other blood tests. Thus, it serves as a marker of disease severity and may predict prognosis of the condition. There were three deaths in our cohort (9%), one female and two male patients. Dialysis was performed in two of them.
30
E. Rista et al. / Journal of Clinical Virology 91 (2017) 25–30
Additionally, there were two cases of pregnant women who both lost their fetuses in 16th week gestational age, as determined by ultrasound measurements. Fetal death coincided with peak severity of the disease and nadir platelet count. Neither of them needed dialysis. In our knowledge, there is a paucity of data on how to offer the appropriate available treatment to both mother and the baby [15]. Further studies are needed in order to define the best therapeutic approach in these rare cases. The study cohort is small and data should be interpreted carefully. However they are in line with other robust findings by other studies. There are several studies on HFRS in the Europe and the Balkan area more specifically. They show differences in viral strains prevalent in Europe, Northern and Southern Balkan. Dobrava is more prevalent in the south of Balkans, whereas Puumala strain is predominant in the rest of European countries. In Slovenia and Croatia, more than 70% of cases are caused by Puumala virus (PUUV). In Albania, Greece, Macedonia and Serbia, the condition is caused almost exclusively by Dobrava [16]. Dobrava infection is associated with higher mortality in comparison with PUUV. In our study, thrombocytopenia was identified as an independent predictor for developing acute kidney injury. In one study, in Croatia the authors did not include thrombocytopenia in their multivariate analysis. Instead, they identified hyponatremia and proteinuria as risk factors for oliguria and anuria [11]. Zupanc et al. showed that conjunctival bleeding, diarrhea, serum sodium of 133 mmol/L, dipstick urine value of >1.5 g/dL were risk factors for development of oliguria and anuria. [17]. Thrombocytopenia, urinary volume and hyponatremia were indicators of more severe disease in our study population. We believe these differences are related to the specific viral strain as well as genetic influence on immune response of Albanian patients. A risk score exists for calculating the risk of developing severe acute kidney injury depending on the presence or absence of three parameters (thrombocytopenia, elevated CRP, proteinuria) [18]. Given the specifics of the Albanian healthcare system, it cannot be ruled out that less severe or fulminant cases can go undetected in remote areas. Thus a selection bias toward moderate-to-severe cases may exist. There was a clear underestimation of creatinine values in patients who were treated with dialysis. 6. Conclusion HFRS is a severe viral disease in Albania, associated with high mortality, 9% in our cohort. Early diagnosis is paramount to treatment of HFRS. In our study, thrombocytopenia, urinary volume, hyponatremia were indicators of more severe disease. Nadir platelet count was an independent predictor for development of AKI. Interestingly, the two cases of pregnant patients resulted in fetal loss. Further studies are needed in order to elucidate the injury mechanism of the virus to the fetus. Funding None Competing interests None
Ethical approval Not required Acknowledgements The authors are grateful to the physician and nurses of Infectious Disease Clinic, Hemodialysis Clinic, University Center “Mother Teresa” Tirana, physician of the Institute of Public Health in Albania. References [1] T.M. Cosgriff, R.M. Lewis, Mechanisms of disease in hemorrhagic fever with renal syndrome, Kidney Int. Suppl. 2 (December (35)) (1991). [2] W. Muranyi, U. Bahr, M. Zeier, F.J. van der Woude, Hantavirus infection, J. Am. Soc. Nephrol. 16 (December (12)) (2005) 3669–3679. [3] C. Peters, California encephalitis, hantavirus pulmonary syndrome, and bunyavirid hemorrhagic fevers, in: G.L. Mandell, J.E. Bennett, R. Dolin (Eds.), Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 7th ed., Churchill Livingstone Elsevier, Philadelphia, Pennsylvania, 2010, pp. 2289–2293. [4] Rista Elvana, et al., Virol-mycol 5 (Suppl. 1) (2016), http://dx.doi.org/10.4172/ 2161-0517.C1.009. [5] C.F. Fulhorst, F.T. Koster, D.A. Enria, C.J. Peters, in: R.L. Guerrant, D.H. Walker, P.R. Weller (Eds.), Tropical Infectious Diseases Principles, Pathogens and Practice, 3rd edition, Elsevier, Edinburgh, 2011, pp. 470–480. [6] C. Schmaljohn, B. Hjelle, Hantaviruses a global disease problem, Emerg. Infect. Dis. 3 (April (2)) (1997) 95. [7] H. Jiang, H. Du, L.M. Wang, P.Z. Wang, X.F. Bai, Hemorrhagic fever with renal syndrome: pathogenesis and clinical picture, Front. Cell. Infect. Microbiol. 6 (2016). [8] M. Temonen, J. Mustonen, H. Helin, A. Pasternack, A. Vaheri, H. Holthöfer, Cytokines, adhesion molecules, and cellular infiltration in nephropathia epidemica kidneys: an immunohistochemical study, Clin. Immunol. Immunopathol. 78 (January (1)) (1996) 47–55. [9] R. Bellomo, J.A. Kellum, C. Ronco, Defining and classifying acute renal failure: from advocacy to consensus and validation of the RIFLE criteria, Intensive Care Med. 33 (March (3)) (2007) 409–413. [10] R.L. Mehta, J.A. Kellum, S.V. Shah, B.A. Molitoris, C. Ronco, D.G. Warnock, A. Levin, Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury, Crit. Care 11 (march (2)) (2007) 1. ´ I. Kuzman, J. Begovac, Clinical and [11] D. Turˇcinov, I. Puljiz, A. Markotic, laboratory findings in patients with oliguric and nonˇcoliguric Hantavirus haemorrhagic fever with renal syndrome: an analysis of 128 patients, Clin. Microbiol. Infect. 19 (July (7)) (2013) 674–679. [12] M. Wang, J. Wang, T. Wang, J. Li, L. Hui, X. Ha, Thrombocytopenia as a predictor of severe acute kidney injury in patients with Hantaan virus infections, PLoS One 8 (January (1)) (2013) e53236. [13] F.M. Rasche, B. Uhel, D.H. Kruger, W. Karges, D. Czock, W. Hampl, F. Keller, H. Meisel, L. von Muller, Thrombocytopenia and acute renal failure in Puumala hantavirus infections, Emerg. Infect. Dis. 10 (August (1)) (2004) 1420–1425. [14] D. Ferluga, A. Vizjak, Hantavirus nephropathy, J. Am. Soc. Nephrol. 19 (September (9)) (2008) 1653–1658. [15] B.N. Kim, B.D. Choi, Hemorrhagic fever with renal syndrome complicated with pregnancy: a case report, Kor. J. Internal Med. 21 (June (2)) (2006) 150–153. [16] Zupanc, et al., HFRS and hantaviruses in the Balkans/South-East, Europe-virus Res. 187 (1) (2014) 27–33. [17] Zupanc, et al., Virus Res. 187 (2014) 27–33. [18] Latus, et al., Acute kidney injury and tools for risk-stratification in 456 patients with hantavirus-induced nephropathia endemic, Nephrol. Dial. Transplant (2014) 1–7.