Advertisement

Performance assessment of the SAPS II and SOFA scoring systems in Hanta virus Hemorrhagic Fever with Renal Syndrome

Open AccessPublished:August 10, 2017DOI:https://doi.org/10.1016/j.ijid.2017.08.003

      Highlights

      • The sample comprised 384 patients with HFRS from January 2006 to February 2017.
      • A new scoring system specifically for HFRS was formulated, named H-SOFA.
      • PLT, PCT, TB, and FOBT were independent predictors of severe HFRS.
      • SAPS II, SOFA, and H-SOFA had high predictive value for the progression of severe HFRS, with H-SOFA being the highest.

      Abstract

      Background

      Hemorrhagic Fever with Renal Syndrome (HFRS), caused by the hantavirus, is a natural infectious disease characterized by fever, hemorrhage and renal damage. China is the most severely endemic area for HFRS in the world. In recent years, critical scoring systems based on quantitative classification have become an important clinical tool for predicting and evaluating the prognosis of critical illness, and provide guidelines for clinical practice.

      Methods

      The sample comprised 384 patients with HFRS treated in the Taizhou Hospital from January 2006 to February 2017. The patients were divided into the severe group and the mild group according to their clinical characteristics. By comparing the differences in clinical symptoms, signs and laboratory data between the two groups, the clinically relevant indicators of severe HFRS were explored. According to the previous studies, we incorporated the positive fecal occult blood test (FOBT) into the sepsis-related organ failure assessment (SOFA) tool and formulated a new scoring system specifically for HFRS, named H-SOFA. By comparing the simplified acute physiology score II (SAPS II), SOFA and H-SOFA scores of the two groups, their predictive values for the progression of HFRS were assessed.

      Results

      Compared to the mild group, patients in the severe group had longer hospital stays; higher frequencies of nausea, vomiting, abdomen pain, signs of congestion and hemorrhage; and more pronounced impairment of liver and renal function. The levels of PLT, PCT, TB, and FOBT were positively correlated with the progression of HFRS (P < 0.001). Patients with HFRS in the severe group got significantly higher scores on the SAPS II, SOFA, and H-SOFA scoring systems (P < 0.001). The values of SAPS II, SOFA and H-SOFA, were significantly correlated with the severity of HFRS, and the AUC values were 0.90, 0.96, and 0.98, respectively.

      Conclusion

      PLT, PCT, TB, and FOBT were independent predictors of severe HFRS; SAPS II, SOFA, and H-SOFA had high predictive value for the progression of severe HFRS, with H-SOFA being the highest.

      Abbreviations:

      AIDS (acquired immunodeficiency syndrome), ALB (albumin), ALT (alanine aminotransferase), AMV (Amur virus), APTT (activated partial thromboplastin time), ARDS (acute respiratory distress syndrome), ARF (acute renal failure), AST (aspartate transaminase), AUC (area under the ROC curve), B (Independent variable coefficient), BUN (blood urea nitrogen), Ca (serum calcium), CD3 (cluster of differentiation 3), CD4 (cluster of differentiation 4), CD8 (cluster of differentiation 8), CI (confidence interval), CK (creatine kinase), CRP (C reactive protein), CRRT (continuous renal replacement therapy), df (degrees of freedom), DIC (disseminated intravascular coagulation), DOBV (dobrava virus), ELISA (enzyme-linked immunosorbent assay), Fib (fibrinogen), FOBT (fecal occult blood test), GCS (glasgow coma scale), GFR (glomerular filtration rate), HB (hemoglobin), HCO3- (bicarbonate ion), HFRS (Hemorrhagic Fever with Renal Syndrome), HTNV (hantaan virus), ICU (intensive care unit), IHD (intermittent hemodialysis), K (serum potassium), LDH (lactate dehydrogenase), MODS (multiple organ dysfunction syndrome), MV (mechanical ventilation), Na (serum sodium), OR (odds ratio), PCT (procalcitonin), PLT (platelet), PT (prothrombin time), PUUV (Puumala virus), ROC (receiver operating characteristic), SAPS II (simplified acute physiology score II), SCr (serum creatinine), SE (standard error), SEOV (Seoul virus), SOFA (sepsis-related organ failure assessment), TB (total bilirubin), UA (uric acid), UOBT (urine occult blood test), WBC (white blood cells)

      Keywords

      Introduction

      Hemorrhagic fever with renal syndrome (HFRS), caused by hantaviruses, is a natural infectious disease characterized by fever, hemorrhage and kidney damage (
      • Ermonval M.
      • Baychelier F.
      • Tordo N.
      What do we know about how hantaviruses interact with their different hosts?.
      ). HFRS has been a major epidemic mainly in Asia and Europe; about 100,000 cases of HFRS are documented annually, most of which occur in China, Korea, and Russia (
      • Yu H.
      • Jiang W.
      • Du H.
      • Xing Y.
      • Bai G.
      • Zhang Y.
      • et al.
      Involvement of the Akt/NF-kappaB pathways in the HTNV-mediated increase of IL-6, CCL5, ICAM-1, and VCAM-1 in HUVECs.
      ). Among all the countries, China is the most seriously affected one and accounts for over 90% of the total number of HFRS cases around the world (
      • Jiang H.
      • Zheng X.
      • Wang L.
      • Du H.
      • Wang P.
      • Bai X.
      Hantavirus infection: a global zoonotic challenge.
      ). Due to positive prevention efforts of the government, the prevalence of HFRS had been lowered in recent years, but the high incidence persisted in some areas (
      • Zou L.X.
      • Chen M.J.
      • Sun L.
      Haemorrhagic fever with renal syndrome: literature review and distribution analysis in China.
      ). The prevalence of HFRS in China is characterized by a large number of affected patients and high mortality of critical cases. Also, the incidence of atypical cases with unusual clinical manifestations has increased, and some new hantavirus genotypes have been found recently, hindering the early diagnosis and treatment of HFRS (
      • Jonsson C.B.
      • Figueiredo L.T.
      • Vapalahti O.
      A global perspective on hantavirus ecology, epidemiology, and disease.
      ). Similar to other critical illnesses, exploring early and new biomarkers, and combining clinical features with laboratory parameters to detect the severity and prognosis of HFRS in advance are very important to guide clinicians to initiate effective treatment and improve the remedy achievement ratio.
      In recent years, severity scoring systems based on quantitative classification have become an important clinical tool for the prediction and evaluation of the prognosis of critically ill patients, and serve as a guide to clinical practice (
      • Routsi C.
      • Pratikaki M.
      • Sotiropoulou C.
      • Platsouka E.
      • Markaki V.
      • Paniara O.
      • et al.
      Application of the sequential organ failure assessment (SOFA) score to bacteremic ICU patients.
      ). The scoring systems commonly used in clinical settings include the Simplified Acute Physiology Score II (SAPS II) and the Sepsis-related Organ Failure Assessment (SOFA) score, which is widely used in many Western countries. For example, it has been reported that whereas the SAPS II is positively associated with mortality in acute renal failure (ARF) (
      • Strand K.
      • Strand L.I.
      • Flaatten H.
      The interrater reliability of SAPS II and SAPS 3.
      ), the SOFA score is widely used in the management of severe sepsis, and can be used to predict the duration of admission and risk of mortality (
      • Siddiqui S.
      • Chua M.
      • Kumaresh V.
      • Choo R.
      A comparison of pre ICU admission SIRS, EWS and q SOFA scores for predicting mortality and length of stay in ICU.
      ). Some studies showed that although both the SAPS II and SOFA score could predict the prognosis of septic shock, the SAPS II was a little worse in doing so compared to the SOFA score, which was more reflective of the patient’s circulatory system (
      • Kim Y.H.
      • Yeo J.H.
      • Kang M.J.
      • Lee J.H.
      • Cho K.W.
      • Hwang S.
      • et al.
      Performance assessment of the SOFA, APACHE II scoring system, and SAPS II in intensive care unit organophosphate poisoned patients.
      ,
      • Perren A.
      • Previsdomini M.
      • Perren I.
      • Merlani P.
      Critical care nurses inadequately assess SAPS II scores of very ill patients in real life.
      ). Currently, there are few reports about the application of the SAPS II and SOFA in the prediction of HFRS.

      Methods

      Study participants

      384 patients with HFRS who were treated in the Affiliated Taizhou Hospital of Wenzhou Medical University from January 2006 to February 2017 were selected randomly and reviewed. The diagnosis of HFRS was made based on the detection of specific IgM antibodies to hantavirus by enzyme-linked immunosorbent assay (ELISA).
      Based on the criteria for the clinical classification of HFRS (
      • Bai X.
      • Xu Z.
      Hemorrhagic fever with renal syndrome.
      ), the patients were divided into two groups: the severe group consisted of serious and critical cases, and the mild group consisted of mild and moderate cases (Table 1).
      Table 1The criteria for the clinical classification of HFRS.
      GroupTypeTemperatureEffusionHemorrhageShockKidney Injury
      Mild GroupMild<39 °Cmildskin and mucous membranesnonealbuminuria+−++

      without oliguria
      Moderate39–40 °Cmoderate

      chemosis
      ecchymosishypotension susceptibilityalbuminuria ++−+++

      and oliguria
      Severe

      Group
      Serious≥40 °Cserious and toxic psychiatric symptomsecchymosis and gastrointestinal hemorrhageshockoliguria ≤5 days

      anuria ≤2 days
      Criticalhave one or more of the following complications in addition to the criteria for serious cases:

      refractory shock ≥2days, visceral hemorrhage; oliguria >5 days or anuria >2 days, BUN >42.84 mmol/L; pulmonary edema; heart failure; severe infection; cerebral hemorrhage, brain edema or herniation.
      Abbreviations: BUN, blood urea nitrogen.

      Symptoms and signs

      The positive symptoms and signs of two groups were compared and analyzed, including cough, nausea and vomiting, dizziness and headache, abdominal pain, backache, diarrhea, conjunctival congestion, pharynges congestion, flush, subconjunctival hemorrhage, cervico-thoracic hemorrhage, underarm hemorrhage, renal percussive pain, and abdominal tenderness.

      Laboratory parameters

      Twenty-seven clinical laboratory parameters were detected and analyzed, including blood tests performed using an autoanalyzer (Sysmex-2100, Sysmex Corp., Japan), biochemical and immune examination performed using an autoanalyzer (Architect ci16200, Abbott Corp., USA), blood coagulation detected by hematology analyzers (STA Compact, Stago Corp., France), lymphocyte subgroup examination detected by flow cytometry (Becton-Dickinson FACSCalibur, BD Biosciences, CA, USA), and routine urine performed using an autoanalyzer (UF-1000i, Sysmex Corp., Japan). The detailed laboratory parameters were white blood cell (WBC) count; platelet count (PLT); levels of hemoglobin (HB); C-reactive protein (CRP), procalcitonin (PCT), alanine aminotransferase (ALT), aspartate transaminase (AST), total bilirubin (TB), serum sodium (Na), serum potassium (K), serum calcium (Ca), albumin (ALB), uric acid (UA), creatine kinase (CK), lactate dehydrogenase (LDH), urea, serum creatinine (SCr), glomerular filtration rate (GFR), prothrombin time (PT), activated partial thromboplastin time (APTT), fibrinogen (Fib), cluster of differentiation 3 (CD3), cluster of differentiation 4 (CD4), and cluster of differentiation 8 (CD8); albuminuria; urine occult blood test (UOBT); and fecal occult blood test (FOBT).

      SAPS II

      The SAPS II includes 17 variables, including age, physiological variables (heart rate, blood pressure, temperature, PaO2/FiO2 ratio, urine volume, blood urea nitrogen (BUN), WBC, K, Na, bicarbonate ion (HCO3), TB, and Glasgow coma scale (GCS) score), type of admission (emergency surgery, elective surgery, medical patient), and chronic diseases (acquired immunodeficiency syndrome (AIDS), metastatic cancer, hematological malignancy) (
      • Le Gall J.R.
      • Lemeshow S.
      • Saulnier F.
      A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study.
      ). The score for each item ranges from 0-26 points, and the total score is 0-163 points. The SAPS II scores of the two groups were collected respectively, then compared and analyzed by statistical tools.

      SOFA Score

      The SOFA scoring system comprises 7 indexes, including the respiratory system (PaO2/FiO2), hematological system (PLT), liver function (TB), cardiovascular system (systolic blood pressure and application of vasoactive drugs), central nervous system (GCS), renal function (SCr and urine volume) (
      • Vincent J.L.
      • Moreno R.
      • Takala J.
      • Willatts S.
      • De Mendonca A.
      • Bruining H.
      • et al.
      The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine.
      ). The score for each item ranges from 0-4 points, and the total score is 0-28 points. According to our previous studies, a positive FOBT was significantly correlated with the progression of HFRS. Thus, we constituted a new scoring system specifically for HFRS by juxtaposing the grade of FOBT positivity on the SOFA system, called the H-SOFA. The quantitative graduation for points is as follows: the result ‘-’ for FOBT equals 0 points, ‘+’ equals 1 point, ‘++’ equals 2 points, ‘+++’ equals 3 points, and ‘++++’ equals 4 points. The SOFA and H-SOFA scores for the two groups were collected respectively, then compared and analyzed by statistical tools.

      Statistical analysis

      Statistical analysis was performed using SPSS version 22.0 (IBM Inc., Armonk, NY, USA). The tables were created using Microsoft® Excel® 2016 (Microsoft Corp., USA). Continuous variables were presented as a mean ± standard deviation (SD). The variables with homogeneity of variance were analyzed by student’s t test, and variables with heterogeneity of variance were analyzed by the Welch-Satterthwaite t test. The frequencies and percentages were given for qualitative variables. Significant differences were tested by the chi-square test, and Fisher’s exact test was used when the numbers were too small. Binary logistic regression analysis was used to identify the clinical risk factors and scoring systems which correlated with the progression of HFRS. The predicting values were tested with receiver operating characteristic (ROC) curves, and quantified by calculating the area under the curve (AUC) and the 95% confidence interval (CI). A two-tailed P < 0.05 was considered to indicate a statistically significant difference.

      Ethics statement

      This retrospective study was reviewed and approved by the Institutional Review Board of the Affiliated Taizhou Hospital of Wenzhou Medical University.

      Results

      Of the 384 HFRS patients, 334 patients belonged to the mild group, and 50 patients belonged to the severe group. Compared to the mild group, patients in the severe group had longer hospital stays (P < 0.001). Also, the differences between the two groups in other parameters, like gender, age, career, the interval from onset to arrival at the hospital, duration of the febrile phase, and hospital stay, were not significant (P > 0.05) (Table 2). Compared to the mild group, patients in the severe group had higher frequencies of nausea and vomiting, abdominal pain, signs of congestion and hemorrhage, and abdominal tenderness (P < 0.05). However, there were no statistically significant differences in the other parameters between the two groups, like cough, dizziness and headache, backache, diarrhea, and renal percussive pain (P > 0.05) (Table 3).
      Table 2The characteristics of HFRS patients, and the differences between two groups.
      CharacteristicsSevere group

      (n = 50)
      Mild group

      (n = 334)
      P value
      Male/Female, n41/9235/990.088
      Age, years50.52 ± 14.1946.74 ± 14.400.084
      Career (farmer), n (%)44 (88.00%)273 (81.74%)0.276
      Interval from onset to arrival at hospital, days4.58 ± 1.975.17 ± 1.940.059
      Duration of febrile phase, days6.63 ± 6.206.00 ± 3.110.350
      Hospital stay, days16.92 ± 9.7011.21 ± 4.82<0.001
      Table 3Positive symptoms and signs in HFRS patients, and a statistical analysis of the differences between the two groups.
      Symptoms and signsSevere group

      (n = 50)
      Mild group

      (n = 334)
      χ2 valueP value
      Cough, n (%)11 (22.0%)43 (12.9%)2.9970.083
      Nausea and vomiting, n (%)38 (76.0%)152 (45.5%)16.175<0.001
      Dizziness and headache, n (%)27 (54.0%)159 (47.6%)0.7120.399
      Backache, n (%)17 (34.0%)133 (39.8%)0.6190.431
      Bellyache, n (%)20 (40.0%)59 (17.7%)13.277<0.001
      Diarrhea, n (%)16 (32.0%)73 (22.0%)2.4380.118
      Conjunctival congestion, n (%)36 (72.0%)161 (48.2%)9.8570.002
      Pharyngeal congestion, n (%)22 (44.0%)81 (24.3%)8.6410.003
      Flush, n (%)25 (50.0%)111 (33.2%)5.3450.021
      Subconjunctival hemorrhage, n (%)15 (30.0%)10 (3.0%)47.769<0.001
      Cervico-thoracic hemorrhage, n (%)13 (26.0%)13 (3.9%)30.262<0.001
      Axillary hemorrhage, n (%)15 (30.0%)11 (3.3%)44.999<0.001
      Renal percussive pain, n (%)23 (46.0%)122 (36.5%)1.6610.198
      Abdominal tenderness, n (%)10 (20.0%)26 (7.8%)6.2680.012
      Compared to the mild group, patients in the severe group had higher levels of WBC, CRP, PCT, ALT, AST, TB, UA, CK, LDH, urea, SCr, PT, and APTT, as well as lower levels of PLT, Na, Ca, and GFR (P < 0.05). However, there were no statistical differences in the other parameters between the two groups, like HB, K, Fib, CD3, CD4, and CD8 (P > 0.05) (Table 4). Compared to the mild group, patients in the severe group had higher frequencies of albuminuria, and positive UOBT and FOBT (P < 0.01) (Table 5).
      Table 4The Laboratory parameters of HFRS patients, and a statistical analysis of the differences between the two groups.
      Laboratory parameters

      mean ± SD
      Severe group

      (n = 50)
      Mild group

      (n = 334)
      t valueP value
      WBC (×109/L)25.24 ± 11.3213.31 ± 6.937.251<0.001
      PLT (×109/L)20.30 ± 18.9064.01 ± 51.61−11.24<0.001
      HB (g/L)128.84 ± 34.09138.64 ± 22.75−1.9680.054
      CRP (mg/L)56.55 ± 59.4120.95 ± 31.853.792<0.001
      PCT (μg/L)11.20 ± 11.281.62 ± 3.644.732<0.001
      K (mmol/L)3.68 ± 0.643.53 ± 0.491.4880.143
      Na (mmol/L)130.83 ± 9.38135.87 ± 5.79−3.6950.001
      Ca (mmol/L)1.739 ± 0.221.96 ± 0.18−6.729<0.001
      ALT (U/L)331.66 ± 719.8899.84 ± 183.022.2580.028
      AST (U/L)751.22 ± 1738.55113.31 ± 175.092.5910.013
      TB (μmol/L)28.24 ± 31.9916.84 ± 27.862.3880.02
      ALB (g/L)27.52 ± 4.6732.02 ± 4.15-6.635<0.001
      UA (μmol/L)711.35 ± 269.39480.57 ± 209.625.800<0.001
      CK (U/L)520.58 ± 1006.40142.22 ± 155.432.6540.011
      LDH (U/L)1089.02 ± 1378.01426.55 ± 227.293.3920.001
      urea (mmol/L)32.04 ± 16.8112.36 ± 8.028.142<0.001
      SCr (μmol/L)573.84 ± 217.18202.78 ± 141.8611.713<0.001
      GFR (mL/min)12.93 ± 11.0848.06 ± 37.70-13.56<0.001
      PT (sec)16.42 ± 6.2713.31 ± 1.123.4980.001
      APTT (sec)72.83 ± 39.8446.21 ± 9.544.294<0.001
      Fib (g/L)3.04 ± 1.574.16 ± 4.49-1.5660.119
      CD3 (/L)4264.05 ± 3161.013294.0 ± 2187.711.2950.208
      CD4 (/L)945.8 ± 536.17861.74 ± 455.570.710.480
      CD8 (/L)3296.1 ± 2920.532383.6 ± 1836.281.3310.196
      Abbreviations: WBC, white blood cells; PLT, platelet; HB, hemoglobin; CRP, C reactive protein; PCT, procalcitonin; K, serum potassium; Na, serum sodium; Ca, serum calcium; ALT, alanine aminotransferase; AST, aspartate transaminase; TB, total bilirubin; ALB, albumin; UA, uric acid; CK, creatine kinase; LDH, lactate dehydrogenase; SCr, serum creatinine; GFR, glomerular filtration rate; PT, prothrombin time; APTT, activated partial thromboplastin time; Fib, fibrinogen; CD3, cluster of differentiation 3; CD4, cluster of differentiation 4; CD8, cluster of differentiation 8.
      Table 5The routine urine and stool test results of HFRS patients, and a statistical analysis of the differences between the two groups.
      Results of routine urine and stool testsSevere group

      (n = 50)
      Mild group

      (n = 334)
      Z valueP value
      Albuminuria1 (2.0%)54 (16.2%)−3.4190.001
      +4 (8.0%)44 (13.2%)
      ++8 (16.0%)68 (20.3%)
      +++37 (74.0%)168 (50.3%)
      UOBT7 (14.0%)175 (52.4%)−6.279<0.001
      +11 (22.0%)80 (24.0%)
      ++10 (20.0%)40 (12.0%)
      +++22 (44.0%)39 (11.6%)
      FOBT9 (18.0%)238 (71.2%)−7.914<0.001
      +7 (14.0%)26 (7.8%)
      ++6 (12.0%)24 (7.2%)
      +++8 (16.0%)21 (6.3%)
      ++++20 (40.0%)25 (7.5%)
      Abbreviations: UOBT, urine occult blood test; FOBT, fecal occult blood test.
      We selected the symptoms, signs, and laboratory parameters with significant differences between the two groups, which consisted of 30 variables. Assignment of the dependent and independent variables were performed as follows: the severe and mild cases were defined as ‘1’ and ‘0’ respectively; the positive and negative symptoms and signs were defined as ‘1’ and ‘0’ respectively; and the laboratory parameters were defined as ‘1’, ‘2’, ‘3’ according to their respective levels. Single-factor binary logistic regression analysis was first used to explore the correlation between the variables and severe or mild HFRS. Only 10 indicators, nausea and vomiting, cervico-thoracic hemorrhage, axillary hemorrhage, PLT, PCT, Ca, TB, SCr, GFR, and FOBT yielded statistically significant results. Then multi-factor binary logistic regression analysis was used, and finally, low PLT and high levels of PCT, TB, and FOBT were identified as risk factors for the progression of HFRS (Table 6).
      Table 6Independent risk factors for the progression of HFRS.
      ParametersBSEWalddfP valueOR95% CI for OR
      LowerUpper
      PLT2.6460.45433.9601<0.00114.0915.78834.304
      PCT0.8370.2908.36710.0042.3101.3104.075
      TB0.9480.4454.52910.0332.5801.0786.176
      FOBT1.0560.27414.8381<0.0012.8731.6794.916
      Constant−12.0061.59356.801<0.0010.001
      Abbreviations: B, Independent variable coefficient; SE, standard error; df, degrees of freedom; OR, odds ratio; CI, confidence interval; PLT, platelet; PCT, procalcitonin; TB, total bilirubin; FOBT, fecal occult blood test.
      To explore the predictive value of the risk factors, ROC analysis with AUC measurement was used. The analysis revealed that the AUC values for PLT, PCT, TB, and FOBT were 0.814, 0.712, 0.657, and 0.782, respectively. By combining the 4 factors, the AUC value rose to 0.890, and the sensitivity and specificity were 72.0%, 89.25%, respectively (Table 7, Figure 1).
      Table 7Predictive values for progression based on laboratory parameters in patients with HFRS.
      ParametersAUCP valueCut-off valueSensitivitySpecificity95% CI for AUC
      LowerUpper
      PLT0.814<0.001268.0091.940.7580.861
      PCT0.712<0.001246.0093.550.6490.769
      TB0.657<0.001146.0085.410.5930.718
      FOBT0.782<0.001182.0067.740.7240.833
      Combination0.890<0.001772.0089.250.8430.927
      Combination means combining of the 4 risk factors, the level of PLT, PCT, TB, and FOBT. Abbreviations: AUC, area under the ROC curve; CI, confidence interval; PLT, platelet; PCT, procalcitonin; TB, total bilirubin; FOBT, fecal occult blood test.
      Figure 1
      Figure 1ROC analysis of PLT, PCT, TB, and FOBT revealed that the 4 parameters reached statistical significance for predicting HFRS progression (P < 0.001). By combining the 4 factors, the AUC value was increased. Combination means combining of the 4 risk factors, the level of PLT, PCT, TB, and FOBT. Abbreviations: ROC, receiver operating characteristic; PLT, platelet; PCT, procalcitonin; TB, total bilirubin; FOBT, fecal occult blood test.
      Compared to the mild group, patients with HFRS in the severe group got significantly higher scores on the SAPS II, SOFA, and H-SOFA scoring systems (P < 0.001) (Table 8). Single-factor binary logistic regression analysis was used to explore the correlation between the scoring systems and severe or mild HFRS. High scores on the SAPS II, SOFA, and H-SOFA scoring systems were identified as risk factors for the progression of HFRS (P < 0.001) (Table 9). Finally, the ROC analysis of the SAPS II, SOFA and H-SOFA scoring systems revealed that the values of the SAPS II, SOFA and H-SOFA significantly correlated with the severity of HFRS (P < 0.001), and the AUC values were 0.90, 0.96, and 0.98, respectively (Table 10, Figure 2).
      Table 8Scores on the SAPS II, SOFA, and H-SOFA scoring systems for HFRS patients, and a statistical analysis of the differences between the two groups.
      Scoring systems,

      mean ± SD
      Severe group

      (n = 50)
      Mild group

      (n = 334)
      t valueP value
      SAPS II48.14 ± 27.1819.78 ± 9.237.265<0.001
      SOFA13.06 ± 5.694.34 ± 2.1810.629<0.001
      H-SOFA15.86 ± 5.975.06 ± 2.6012.472<0.001
      H-SOFA was a new version of the SOFA scoring system specifically for HFRS, which was incorporated with the level of fecal occult blood test according to the previous studies. Abbreviations: SAPS II, simplified acute physiology score II; SOFA, sepsis-related organ failure assessment.
      Table 9Independent risk factors regarding scoring systems for disease progression in patients with HFRS.
      Scoring systemsBSEWalddfP valueOR95%CI for OR
      LowerUpper
      SAPS II0.1580.02345.4091<0.0011.1711.1181.226
      SOFA0.8230.12841.4051<0.0012.2781.7732.928
      H-SOFA0.9190.16232.2221<0.0012.5061.8253.441
      Abbreviations: B, Independent variable coefficient; SE, standard error; df, degrees of freedom; OR, odds ratio; CI, confidence interval; SAPS II, simplified acute physiology score II; SOFA, sepsis-related organ failure assessment.
      Table 10Predictive values for disease progression based on scoring systems in patients with HFRS.
      Scoring systemsAUCP valueCut-off valueSensitivitySpecificity95%CI for AUC
      LowerUpper
      SAPS II0.900<0.0013080.0087.100.8550.936
      SOFA0.960<0.001696.0083.870.9270.981
      H-SOFA0.980<0.001992.0093.550.9530.994
      Abbreviations: AUC, area under the ROC curve; CI, confidence interval; SAPS II, simplified acute physiology score II; SOFA, sepsis-related organ failure assessment.
      Figure 2
      Figure 2The ROC analysis revealed that the values of the SAPS II, SOFA, and H-SOFA were significantly correlated with the severity of HFRS (P < 0.001). H-SOFA was a new version of the SOFA scoring system specifically for HFRS, which was incorporated with the level of fecal occult blood test according to the previous studies. Abbreviations: ROC, receiver operating characteristic; SAPS II, simplified acute physiology score II; SOFA, sepsis-related organ failure assessment.

      Discussion

      In this study, there was no significant correlation between severe HFRS and patient characteristics such as age and duration of the febrile phase. However, patients in the severe group had longer hospital stays; higher frequencies of nausea, vomiting, abdominal pain, signs of congestion, and hemorrhage; and more pronounced impairment of liver and renal function compared to the mild group. Patients with mild HFRS without shock and oliguria usually have mild signs of toxicity, and commonly enter the polyuric stage directly after the onset of fever. Such patients usually have a quick resolution of the condition with a good prognosis and short hospital stay. Patients with severe HFRS with an overlap of the second and third stages usually have more serious signs of toxicity and extensive damage to the capillaries and small vessels. This is usually accompanied by nausea and vomiting, abdominal pain, and other gastrointestinal symptoms, as well as obvious signs of congestion and hemorrhage (
      • Ma C.
      • Yu P.
      • Nawaz M.
      • Zuo S.
      • Jin T.
      • Li Y.
      • et al.
      Hantaviruses in rodents and humans, Xi’an, PR China.
      ). The kidneys are the most commonly damaged organs in HFRS, with the underlying pathological mechanisms including the direct damage caused by the hantavirus, hypoxic-ischemic renal damage, renal tubular obstruction, and immunologic injury caused by the renal deposition of immune complexes (
      • Schorr C.A.
      • Zanotti S.
      • Dellinger R.P.
      Severe sepsis and septic shock: management and performance improvement.
      ). Parameters such as urine volume, BUN, and albuminuria are all important indices in the clinical classification of the serious or critical type of HFRS in China (
      • Bai X.
      • Xu Z.
      Hemorrhagic fever with renal syndrome.
      ). HFRS with liver function impairment is also very common; some patients suffer mainly from liver damage, and since the renal injury is not obvious, are easily misdiagnosed as having acute viral hepatitis (
      • Shim S.H.
      • Park M.S.
      • Moon S.
      • Park K.S.
      • Song J.W.
      • Song K.J.
      • et al.
      Comparison of innate immune responses to pathogenic and putative non-pathogenic hantaviruses in vitro.
      ). Du et al. revealed that the level of AST was an independent prognostic factor in predicting the prognosis of HFRS in affected patients (
      • Du H.
      • Wang P.Z.
      • Li J.
      • Bai L.
      • Li H.
      • Yu H.T.
      • et al.
      Clinical characteristics and outcomes in critical patients with hemorrhagic fever with renal syndrome.
      ). It has been reported that a disproportionality in T lymphocyte subsets, the obvious decrease in the absolute count and ratio of CD4+ T cells, and a dysfunction of the immune system, might be associated with the pathogenesis of HFRS. However, there were no significant differences in the levels of CD3, CD4 and CD8 between the mild and severe groups in our study (P > 0.05).
      This study observed that the levels of PLT, PCT, TB, and FOBT were positively correlated with the progression of HFRS (P < 0.001), and the AUC values were 0.814, 0.712, 0.657, 0.782, respectively, which indicated the better predictive efficacy of PLT and FOBT. HFRS is usually associated with a sharp decrease in the level of PLT, the pathogenesis being the direct damage to megakaryocytes caused by the hantavirus, lysis, and death of platelets caused by the deposition of immune complexes, and massive depletion of platelets owing to vascular injury and DIC (
      • Jiang H.
      • Zheng X.
      • Wang L.
      • Du H.
      • Wang P.
      • Bai X.
      Hantavirus infection: a global zoonotic challenge.
      ). Consistent with the results of our study, some scholars reported that the dynamic monitoring of the levels and function of PLT, as well as the increasing and declining patterns, were beneficial to assessing the condition of HFRS patients (
      • Sundberg E.
      • Hultdin J.
      • Nilsson S.
      • Ahlm C.
      Evidence of disseminated intravascular coagulation in a hemorrhagic fever with renal syndrome-scoring models and severe illness.
      ). PCT, a precursor of calcitonin, is a diagnostic marker of systemic inflammatory response, with a high sensitivity and specificity for infection (
      • Indino P.
      • Lemarchand P.
      • Bady P.
      • de Torrente A.
      • Genne L.
      • Genne D.
      Prospective study on procalcitonin and other systemic infection markers in patients with leukocytosis.
      ). The elevated level of PCT in HFRS may be associated with eosinophilia and immune activation caused by hantavirus (
      • Muranyi W.
      • Kehm R.
      • Bahr U.
      • Muller S.
      • Handermann M.
      • Darai G.
      • et al.
      Bovine aortic endothelial cells are susceptible to hantavirus infection; a new aspect in hantavirus ecology.
      ). In patients with sepsis, the level of PCT increases significantly in the early stage, and this is directly related to the severity of sepsis (
      • Castelli G.P.
      • Pognani C.
      • Cita M.
      • Stuani A.
      • Sgarbi L.
      • Paladini R.
      Procalcitonin, C-reactive protein, white blood cells and SOFA score in ICU: diagnosis and monitoring of sepsis.
      ). Of course, there is the need for more studies to ascertain whether elevated PCT is positively correlated with the severity of HFRS. Hantavirus can damage the structural integrity of the intestinal wall directly, and induce immune injury by releasing inflammatory factors (
      • Jiang H.
      • Du H.
      • Wang L.M.
      • Wang P.Z.
      • Bai X.F.
      Hemorrhagic fever with renal syndrome: pathogenesis and clinical picture.
      ); therefore, patients with HFRS commonly suffer from gastrointestinal symptoms, such as nausea, vomiting, abdominal pain and diarrhea. Also, gastrointestinal bleeding, intestinal obstruction, and intestinal perforation may occur in severe cases. In our study, patients in the severe group had higher frequencies of positive FOBT compared to the mild group. Patients with positive FOBT had a higher probability of progression to severe HFRS.
      The values of the SAPS II, SOFA, and H-SOFA systems were significantly correlated with the severity of HFRS (P < 0.001), and the AUC values were 0.90, 0.96, and 0.98, respectively. This indicated that all three scoring systems had good predictive value for progression to severe HFRS, with H-SOFA having the highest efficacy. The SAPS II is a system for assessing disease severity, monitoring intensive care unit (ICU) performance, and predicting prognosis. It has the advantages of convenient parameter assessment, relatively easy data acquisition, good operability, accuracy, and repeatability, and is not influenced by the kind of disease (
      • Pantet O.
      • Faouzi M.
      • Brusselaers N.
      • Vernay A.
      • Berger M.M.
      Comparison of mortality prediction models and validation of SAPS II in critically ill burns patients.
      ). As revealed in our study, patients with high SAPS II score had a higher probability of progression to severe HFRS, with the sensitivity and specificity of the diagnosis being 80.0% and 87.1%, respectively, and a cut-off value of 30 points. The SOFA scoring system is designed to assess the incidence and dynamics of single or multiple organ failure in patients, as well as the degree of organ failure (
      • Safari S.
      • Shojaee M.
      • Rahmati F.
      • Barartloo A.
      • Hahshemi B.
      • Forouzanfar M.M.
      • et al.
      Accuracy of SOFA score in prediction of 30-day outcome of critically ill patients.
      ). The SOFA has been widely used in western countries. Studies have shown that the maximum SOFA score after admission is consistent with the prognosis in critically ill patients, and is the most important predictor of patient outcome (
      • Saksida A.
      • Wraber B.
      • Avsic-Zupanc T.
      Serum levels of inflammatory and regulatory cytokines in patients with hemorrhagic fever with renal syndrome.
      ). In this study, the SOFA system was found to have a higher efficacy in predicting the progression to severe HFRS, the sensitivity, and specificity of the diagnosis being 96.0% and 87.1%, respectively, with a cut-off value of 6 points. Similarly, Kim et al. evaluated patients with severe organophosphate poisoning using the SOFA and SAPS II systems, and found that the SOFA score was simpler and more effective in predicting mortality (
      • Kim Y.H.
      • Yeo J.H.
      • Kang M.J.
      • Lee J.H.
      • Cho K.W.
      • Hwang S.
      • et al.
      Performance assessment of the SOFA, APACHE II scoring system, and SAPS II in intensive care unit organophosphate poisoned patients.
      ). However, the SOFA scoring system lacks the ability to assess gastrointestinal function. In this study, we incorporated the FOBT results into the SOFA scoring system, to constitute a new H-SOFA scoring system specifically for HFRS. As revealed in this study, H-SOFA possessed the highest efficacy, with the sensitivity and specificity of the diagnosis of severe HFRS being 92.0% and 93.55%, respectively, and a cut-off value of 9 points. Of course, there is the need for large-scale, multicenter prospective studies to support the clinical use of the H-SOFA scoring system.
      In summary, HFRS is still an important public health problem in China. The dynamic monitoring of the levels of PLT, PCT, TB, FOBT, and the clinical application of the SAPS II and SOFA systems will contribute to the early detection and recognition of severe HFRS. With the sustainable progress made in the development of supportive therapy in recent years, like continuous renal replacement therapy (CRRT), intermittent hemodialysis (IHD), blood purification treatment, and mechanical ventilation (MV), the treatment options available to patients with severe HFRS are continually improving, and are expected to increase the remedy achievement ratio.

      Conflicts of interest

      The authors have no conflicts of interest to declare.

      Funding

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

      Acknowledgments

      The authors would like to thank the staff of the Medical Research Center of Taizhou Hospital for their helpful advice and guidance.

      Appendix A. Supplementary data

      The following is Supplementary data to this article:
      Figure thumbnail mmc1

      References

        • Bai X.
        • Xu Z.
        Hemorrhagic fever with renal syndrome.
        People’s Medical Publishing House, Beijing2013
        • Castelli G.P.
        • Pognani C.
        • Cita M.
        • Stuani A.
        • Sgarbi L.
        • Paladini R.
        Procalcitonin, C-reactive protein, white blood cells and SOFA score in ICU: diagnosis and monitoring of sepsis.
        Minerva Anestesiol. 2006; 72: 69-80
        • Du H.
        • Wang P.Z.
        • Li J.
        • Bai L.
        • Li H.
        • Yu H.T.
        • et al.
        Clinical characteristics and outcomes in critical patients with hemorrhagic fever with renal syndrome.
        BMC Infect Dis. 2014; 14: 191
        • Ermonval M.
        • Baychelier F.
        • Tordo N.
        What do we know about how hantaviruses interact with their different hosts?.
        Viruses. 2016; 8
        • Indino P.
        • Lemarchand P.
        • Bady P.
        • de Torrente A.
        • Genne L.
        • Genne D.
        Prospective study on procalcitonin and other systemic infection markers in patients with leukocytosis.
        Int J Infect Dis. 2008; 12: 319-324
        • Jiang H.
        • Du H.
        • Wang L.M.
        • Wang P.Z.
        • Bai X.F.
        Hemorrhagic fever with renal syndrome: pathogenesis and clinical picture.
        Front Cell Infect Microbiol. 2016; 6: 1
        • Jiang H.
        • Zheng X.
        • Wang L.
        • Du H.
        • Wang P.
        • Bai X.
        Hantavirus infection: a global zoonotic challenge.
        Virol Sin. 2017; 32: 32-43
        • Jonsson C.B.
        • Figueiredo L.T.
        • Vapalahti O.
        A global perspective on hantavirus ecology, epidemiology, and disease.
        Clin Microbiol Rev. 2010; 23: 412-441
        • Kim Y.H.
        • Yeo J.H.
        • Kang M.J.
        • Lee J.H.
        • Cho K.W.
        • Hwang S.
        • et al.
        Performance assessment of the SOFA, APACHE II scoring system, and SAPS II in intensive care unit organophosphate poisoned patients.
        J Korean Med Sci. 2013; 28: 1822-1826
        • Le Gall J.R.
        • Lemeshow S.
        • Saulnier F.
        A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study.
        JAMA. 1993; 270: 2957-2963
        • Ma C.
        • Yu P.
        • Nawaz M.
        • Zuo S.
        • Jin T.
        • Li Y.
        • et al.
        Hantaviruses in rodents and humans, Xi’an, PR China.
        J Gen Virol. 2012; 93: 2227-2236
        • Muranyi W.
        • Kehm R.
        • Bahr U.
        • Muller S.
        • Handermann M.
        • Darai G.
        • et al.
        Bovine aortic endothelial cells are susceptible to hantavirus infection; a new aspect in hantavirus ecology.
        Virology. 2004; 318: 112-122
        • Pantet O.
        • Faouzi M.
        • Brusselaers N.
        • Vernay A.
        • Berger M.M.
        Comparison of mortality prediction models and validation of SAPS II in critically ill burns patients.
        Ann Burns Fire Disasters. 2016; 29: 123-129
        • Perren A.
        • Previsdomini M.
        • Perren I.
        • Merlani P.
        Critical care nurses inadequately assess SAPS II scores of very ill patients in real life.
        Crit Care Res Pract. 2012; 2012: 919106
        • Routsi C.
        • Pratikaki M.
        • Sotiropoulou C.
        • Platsouka E.
        • Markaki V.
        • Paniara O.
        • et al.
        Application of the sequential organ failure assessment (SOFA) score to bacteremic ICU patients.
        Infection. 2007; 35: 240-244
        • Safari S.
        • Shojaee M.
        • Rahmati F.
        • Barartloo A.
        • Hahshemi B.
        • Forouzanfar M.M.
        • et al.
        Accuracy of SOFA score in prediction of 30-day outcome of critically ill patients.
        Turk J Emerg Med. 2016; 16: 146-150
        • Saksida A.
        • Wraber B.
        • Avsic-Zupanc T.
        Serum levels of inflammatory and regulatory cytokines in patients with hemorrhagic fever with renal syndrome.
        BMC Infect Dis. 2011; 11: 142
        • Schorr C.A.
        • Zanotti S.
        • Dellinger R.P.
        Severe sepsis and septic shock: management and performance improvement.
        Virulence. 2014; 5: 190-199
        • Shim S.H.
        • Park M.S.
        • Moon S.
        • Park K.S.
        • Song J.W.
        • Song K.J.
        • et al.
        Comparison of innate immune responses to pathogenic and putative non-pathogenic hantaviruses in vitro.
        Virus Res. 2011; 160: 367-373
        • Siddiqui S.
        • Chua M.
        • Kumaresh V.
        • Choo R.
        A comparison of pre ICU admission SIRS, EWS and q SOFA scores for predicting mortality and length of stay in ICU.
        J Crit Care. 2017; 41: 191-193
        • Strand K.
        • Strand L.I.
        • Flaatten H.
        The interrater reliability of SAPS II and SAPS 3.
        Intensive Care Med. 2010; 36: 850-853
        • Sundberg E.
        • Hultdin J.
        • Nilsson S.
        • Ahlm C.
        Evidence of disseminated intravascular coagulation in a hemorrhagic fever with renal syndrome-scoring models and severe illness.
        PLoS One. 2011; 6: e21134
        • Vincent J.L.
        • Moreno R.
        • Takala J.
        • Willatts S.
        • De Mendonca A.
        • Bruining H.
        • et al.
        The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine.
        Intensive Care Med. 1996; 22: 707-710
        • Yu H.
        • Jiang W.
        • Du H.
        • Xing Y.
        • Bai G.
        • Zhang Y.
        • et al.
        Involvement of the Akt/NF-kappaB pathways in the HTNV-mediated increase of IL-6, CCL5, ICAM-1, and VCAM-1 in HUVECs.
        PLoS One. 2014; 9: e93810
        • Zou L.X.
        • Chen M.J.
        • Sun L.
        Haemorrhagic fever with renal syndrome: literature review and distribution analysis in China.
        Int J Infect Dis. 2016; 43: 95-100