Abstract
Objective
This research aimed to analyze the impact of hemorrhagic fever with renal syndrome (HFRS) with acute pancreatitis (AP) on the severity and prognosis of patients, screen the risk factors of HFRS with AP, and establish a nomogram model.
Methods
Data were collected from HFRS patients at the First Affiliated Hospital of Dali University and Dali Prefecture People’s Hospital (2013-2023). Patients were divided into HFRS with AP (n = 34) and HFRS without AP groups (n = 356). Propensity Score Matching (PSM) and logistic regression analyzed the impact of AP on HFRS severity and short-term prognosis. LASSO-Logistic regression was used to screen risk factors and develop a nomogram model.
Results
After PSM, HFRS patients with AP had higher rates of Continuous Renal Replacement Therapy (CRRT) and/or mechanical ventilation use, , ICU admission, and 30-day mortalitycompared with those without AP (p < 0.05). Further analysis revealed that smoking (OR: 3.702), ferritin (OR: 1.002), white blood cell (OR), fibrinogen (OR: 0.463), and platelet (OR: 0.987) were risk factors for HFRS with AP (p < 0.05). A nomogram model was constructed based on these factors, to predict the risk of HFRS with AP, with an Area Under the Curve (AUC) of 0.90 (95% CI: 0.84–0.95). Additionally, the model calibration curve fit well according to the Hosmer-Lemeshow test (χ2=8.51, p = 0.39).
Conclusion
Patients with HFRS with AP exhibit higher disease severity and poorer prognosis. Smoking, elevated ferritin and white blood cell levels, decreased fibrinogen and platelet levels are more susceptible to developing AP.
Keywords: Haemorrhagic fever with renal syndrome, acute pancreatitis, clinical characteristics, risk factors, nomogram
1. Introduction
HFRS is an acute zoonotic infectious disease caused by Hantavirus infection [1]. HFRS poses a severe threat to human health worldwide [2]. Among these, China is the country with the most severe epidemic situation accounting for 90% of the total reported cases [3,4].
AP is an inflammatory response characterized by the activation of pancreatic enzymes within the pancreas, leading to autodigestion, oedema, haemorrhage, and even necrosis of pancreatic tissue, which can be caused by multiple aetiological factors [5]. It is well-known that gallstones, hypertriglyceridemia, and alcohol are common causes of AP. However, studies have also confirmed that viral infections are closely related to the occurrence of AP [6–9]. Despite this, research by Guo et al. [10]. suggests that the mortality rate is significantly elevated in patients with concurrent AP. Wang et al. [11]. conducted a logistic regression analysis to investigate the risk factors for HFRS with AP in central China. Currently, reported incidence rates of HFRS with AP vary [10–13]. However, there has been limited research on the incidence, prognosis, risk factors, and predictive modelling for patients with HFRS complicated by AP in southwestern China.
Based on the aforementioned factors, this study aims to analyze the disease severity and short-term prognosis of HFRS with AP in southwestern China, using clinical manifestations and laboratory data from patients with HFRS and AP. Furthermore, it intends to screen for risk factors associated with concurrent AP and construct a nomogram model. This will provide a reference for the local prevention and treatment of HFRS with AP.
2. Materials and methods
2.1. Ethics
This study was approved by the Ethics Committee of the First Affiliated Hospital of Dali University. Due to the retrospective nature of this study, the Medical Ethics Committee of the First Affiliated Hospital of Dali University has granted exemption from informed consent for all patients involved. All research procedures involving human participants were conducted in accordance with the 1964 Helsinki Declaration and its subsequent amendments or similar ethical standards (No: DFY20231220001).
2.2. Research subjects
The data of 390 patients diagnosed with HFRS at the First Affiliated Hospital of Dali University and Dali Prefecture People’s Hospital between 1 January 2013, and 30 December 2023, were selected. All HFRS patients were confirmed to have HFRS-specific IgM antibodies in their sera (detected by enzyme-linked immunosorbent assay) by the Yunnan Provincial Institute of Endemic Diseases Control and Prevention.
2.3. Research design
This study initially analyzes the differences in clinical manifestations and laboratory indicators between patients with HFRS with AP and those without AP. Secondly, PSM, Logistic regression analyses are employed to investigate the impact of concurrent AP on the severity and short-term prognosis of HFRS patients. Finally, LASSO-Logistic regression is further utilized to screen for risk factors associated with HFRS with AP, and a nomogram model is constructed.
2.4. Diagnostic criteria
The diagnosis of acute pancreatitis requires meeting 2 out of the following 3 criteria: (1) acute and persistent pain in the mid-to-upper abdomen; (2) elevation of serum amylase or lipase to 3 times or more the upper limit of normal; (3) abdominal computed tomography (CT), magnetic resonance imaging (MRI), or abdominal colour Doppler ultrasound showing typical changes of acute pancreatitis. The diagnostic HFRS based on the ‘Diagnostic Criteria for Epidemic Hemorrhagic Fever’ (WS278-2008) [14,15].
Inclusion Criteria: All patients with HFRS had positive serum-specific IgM antibodies. Patients with AP met the Atlanta diagnostic criteria [16].
Exclusion Criteria: Pregnancy (n = 4), combined with HIV (n = 4), chronic kidney disease (n = 7), hematological disorders (n = 9), hypertriglyceridemia (n = 9), chronic liver disease (n = 10), biliary tract disease (n = 12), and incomplete data (n = 56) (Figure 1).
Figure 1.
Schematic flow chart for HFRS with AP.
2.5. Data collection
Peripheral blood samples of all HFRS patients were collected on the first day of admission and sent to Yunnan Institute for Endemic Disease Prevention and Control for detection of HFRS-specific IgM and IgG. Upon admission, peripheral blood samples were immediately collected from all HFRS patients and sent to the Yunnan Provincial Endemic Disease Control and Prevention Institute for detection of HFRS-specific antibodies IgM and IgG. Demographic data of the patients, including age, gender, smoking status, alcohol consumption, underlying diseases, clinical manifestations, complications, and laboratory results (blood routine tests, blood biochemistry, inflammatory indicators: C-reactive protein (CRP), Procalcitonin (PCT), ferritin, coagulation function, myocardial enzyme spectrum, and urine routine tests (urine protein categorized as 1 for -, +, and 2+, and 2 for 3+ and 4+), were collected through the hospital information system.
2.6. Statistical analysis
In this study, data processing was conducted using R Studio and SPSS software version 26.0. For normally distributed measurement data, an independent-samples student test was used for comparison between two groups, which were expressed as mean ± standard deviation (SD). For measurement data that did not conform to a normal distribution, the Mann-Whitney U test was employed for comparison between groups, and expressed as median (interquartile range, IQR). For categorical data, comparisons between groups were conducted using the chi-square test or Fisher’s exact test. The p-value < 0.05 was considered statistically significant.
To exclude the influence of confounding factors on disease severity and short-term prognosis in patients with HFRS with AP, this study employed 1:1 PSM analysis to adjust for baseline characteristic differences between patients with and without concurrent AP. The adjusted covariates included age, smoking, gastrointestinal haemorrhage, and stomach-ache. Logistic regression analysis was conducted to assess the impact of concurrent AP on disease severity after PSM. Kaplan-Meier survival analysis was used to evaluate the 30-day mortality in patients with HFRS with AP, and the log-rank test was applied to compare differences in survival curves.
To further identify risk factors for concurrent AP, with the occurrence of concurrent AP as the outcome variable, LASSO regression was used to screen the predictor variables. LASSO regression compresses the estimation parameters by adding a penalty term to the least squares method, thereby selecting factors with significant influence on the dependent variable. Based on the variables obtained from LASSO regression, multivariate Logistic regression analysis was then conducted to investigate the risk factors for HFRS with AP and to construct a nomogram prediction model. Additionally, the predictive performance of the model was analyzed.
3. Results
3.1. Comparison of clinical characteristics and laboratory data between patients with HFRS with AP and those without AP
During the study period, a total of 390 patients diagnosed with HFRS at the First Affiliated Hospital of Dali University and Dali Prefecture People’s Hospital were enrolled, among which 34 patients had concurrent AP (with 33 cases diagnosed with AP after admission and 1 case diagnosed outside the hospital), representing an incidence rate of 8.72%. Patients in the HFRS with AP group were older, had a higher proportion of smokers, and exhibited higher rates of abdominal pain, need for CRRT and/or mechanical ventilation, ICU admission, and 30-day mortality compared to those without AP (p < 0.05) (Table 1). In terms of laboratory data for patients in the HFRS with AP group, ferritin, White blood cell (WBC), absolute neutrophil count (N#), urea, creatinine (CREA), uric acid (UA), glucose (Gluc), Lactic Dehydrogenase (LDH), Hydroxybutyrate -dehydrogenase (HBDH), and urine protein were significantly elevated compared to the group without AP (p < 0.05). Conversely, serum sodium (Na), fibrinogen (FIB), platelet (PLT), and albumin (ALB) were significantly decreased (p < 0.05) (Table 2).
Table 1.
Baseline characteristics and clinical outcome of hemorrhagic fever renal syndrome patients with and without acute pancreatitis.
| Variables | HFRS (n = 390) | HFRS without AP (n = 356) | HFRS with AP (n = 34) | p-value |
|---|---|---|---|---|
| Patient characteristics | ||||
| Age (years) | 45.73 ± 14.51 | 45.32 ± 14.76 | 50.06 ± 10.86 | 0.023 |
| Male sex (%) | 274 (70.26%) | 251 (70.51%) | 23 (67.65%) | 0.728 |
| Personal history | ||||
| Smoke (Yes, %) | 92(23.59%) | 75 (21.07%) | 17 (50%) | <0.001 |
| Drinking (Yes, %) | 113 (28.97%) | 102 (28.65%) | 11 (32.35%) | 0.649 |
| Days from symptom onset to hospitalization (day) | 5.00 (4.00, 7.00) | 5.00 (4.00, 7.00) | 5.00 (4.00, 7.00) | 0.520 |
| Vital signs | ||||
| Body temperature (oC) | 38.05 (36.50,39.00) | 37.85 (36.50,39.00) | 39.00 (37.10,39.08) | 0.105 |
| Breath rate (beats/min) | 20.00 (20.00,20.00) | 20.00 (20.00,20.00) | 20.00 (20.00,20.75) | 0.796 |
| Pulse rate (beats/min) | 84.94 ± 18.56 | 85.27 ± 18.56 | 81.50 ± 18.48 | 0.258 |
| Systolic blood pressure (mmHg) | 108.50 (98.00,123.00) | 108.00 (97.00,123.25) | 109.00 (100.00,116.00) | 0.904 |
| Diastolic blood pressure (mmHg) | 69.00 (61.00, 77.00) | 69.00 (60.75,77.00) | 69.50 (62.00,77.25) | 0.582 |
| Signs and symptoms | ||||
| Hepatosplenomegaly | 33 (8.46%) | 30 (8.43%) | 3 (8.82%) | 1.000 |
| Gastrointestinal haemorrhage | 52 (13.33%) | 42 (11.80%) | 10 (29.41%) | 0.009 |
| Intracranial haemorrhage | 3 (0.77%) | 3 (0.84%) | 0 (0.00%) | 1.000 |
| Stomach-ache | 167 (42.82%) | 144 (40.45%) | 23 (67.65%) | 0.002 |
| Fever | 340 (87.18%) | 312 (87.64%) | 28 (82.35%) | 0.540 |
| Nausea or vomiting | 95 (24.36%) | 89 (25.00%) | 6 (17.65%) | 0.340 |
| Headache | 167 (42.82%) | 152 (42.70%) | 15 (44.12%) | 0.873 |
| Backache | 49 (12.56%) | 44 (12.36%) | 5 (14.71%) | 0.902 |
| Muscular soreness | 62 (15.90%) | 58 (16.29%) | 4 (11.76%) | 0.490 |
| Comorbidities | ||||
| Hypertension | 41 (10.51%) | 37 (10.39%) | 4 (11.76%) | 1.000 |
| Diabetes | 15 (3.85%) | 14 (3.93%) | 1 (2.94%) | 1.000 |
| Pulmonary infection | 92 (23.59%) | 82 (23.03%) | 10 (29.41%) | 0.403 |
| Pleural effusion | 61 (15.64%) | 53 (14.89%) | 8 (23.53%) | 0.185 |
| Treatment and prognosis | ||||
| CRRT and/or mechanical ventilation | 49 (12.56%) | 30 (8.43%) | 19 (55.88%) | <0.001 |
| ICU admission | 23 (5.90%) | 5 (1.4%) | 18 (52.94%) | <0.001 |
| 30-day mortality | 18 (4.62%) | 9 (2.53%) | 9 (26.47%) | <0.001 |
CRRT: continuous renal replacement therapy; ICU: intensive care unit; HFRS: hemorrhagic fever with renal syndrome; AP: acute pancreatitis.
Table 2.
Laboratory results on hospital admission in hemorrhagic fever renal syndrome patients with and without acute pancreatitis.
| Variables | HFRS (n = 390) | HFRS without AP (n = 356) | HFRS with AP (n = 34) | p-value |
|---|---|---|---|---|
| Inflammatory index | ||||
| Procalcitonin (ng/ml) | 2.10 (0.59, 3.90) | 2.10 (0.58, 3.90) | 2.30 (0.99, 4.29) | 0.368 |
| C-reactive protein (pg/ml) | 45.97 (19.80, 65.06) | 45.48 (19.00,63.79) | 46.87 (29.12,69.09) | 0.182 |
| Ferritin (ng/ml) | 1049.50 (790.00,1764.75) | 985.50 (783.00,1645.00) | 1716.00 (1250.00,2073.50) | <0.001 |
| Blood routine examination | ||||
| White blood cell (×109/L) | 7.85(5.58, 10.59) | 7.57 (5.50, 9.85) | 10.88 (8.22, 12.14) | <0.001 |
| Neutrophil count (×109/L) | 5.75 (3.88, 7.52) | 5.50 (3.73, 7.25) | 8.05 (5.30, 10.55) | <0.001 |
| Lymphocyte count(×109/L) | 1.85 (1.19, 2.61) | 1.81 (1.19, 2.61) | 2.12 (1.29, 3.35) | 0.214 |
| Monocyte count (×109/L) | 1.11 (0.70, 1.54) | 1.10 (0.69, 1.54) | 1.25 (0.94, 1.71) | 0.113 |
| Eosinophil count (×109/L) | 0.06 (0.02, 0.13) | 0.06 (0.02, 0.12) | 0.09 (0.01, 0.21) | 0.226 |
| Basophil count(×109/L) | 0.05 (0.03, 0.08) | 0.04 (0.02, 0.08) | 0.06 (0.04, 0.09) | 0.109 |
| Red blood cell (×1012/L) | 4.62 (4.32, 5.13) | 4.61 (4.33, 5.04) | 4.96 (4.28, 5.62) | 0.058 |
| Hemoglobin (g/L) | 145.00 (129.00, 158.00) | 145.00 (129.00, 158.00) | 150.50 (135.00, 163.50) | 0.122 |
| Platelet (×109/L) | 65.00 (40.00, 138.50) | 69.00 (43.00, 138.50) | 44.50 (37.10,57.25) | <0.001 |
| Liver function | ||||
| Total bilirubin (umol/L) | 14.20 (10.10, 18.53) | 14.45 (10.10, 18.38) | 12.90 (10.10, 18.98) | 0.475 |
| Direct bilirubin (umol/L) | 5.00 (3.32, 7.68) | 5.00 (3.30, 7.60) | 5.80 (3.90, 8.75) | 0.315 |
| Alanine aminotransferase (U/L) | 102.50 (53.00, 181.00) | 100.00 (52.00, 174.25) | 120.00 (68.25, 268.75) | 0.113 |
| Aspartate aminotransferase (U/L) | 107.00 (52.00, 233.25) | 100.00 (48.75, 212.25) | 170.00 (95.00, 389.50) | 0.006 |
| Albumin (g/L) | 31.40 (27.50, 35.00) | 32.00 (27.80, 35.00) | 27.90 (25.70, 31.38) | <0.001 |
| Renal function | ||||
| Urea nitrogen (mmol/L) | 6.92 (4.63, 15.67) | 6.59 (4.50, 14.60) | 15.23 (9.52, 21.59) | <0.001 |
| Creatinine (umol/L) | 121.00 (76.50, 264.75) | 110.00 (74.75, 255.25) | 274.50 (122.00, 416.00) | <0.001 |
| Urine acid (umol/L) | 397.00 (257.25, 573.75) | 392.00 (252.00, 570.25) | 469.50 (373.50, 654.75) | 0.030 |
| Electrolyte | ||||
| Serum potassium (mmol/L) | 3.80 (3.43, 4.13) | 3.79 (3.42, 4.12) | 3.98 (3.57, 4.23) | 0.078 |
| Serum sodium (mmol/L) | 138.00 (135.00, 141.28) | 138.00 (135.00, 141.50) | 135.15 (130.00, 139.35) | 0.003 |
| Serum chloride (mmol/L) | 101.70 (97.60, 104.77) | 101.85 (97.70, 104.93) | 99.96 (95.10, 103.92) | 0.130 |
| Serum magnesium (mmol/L) | 0.92 (0.86, 1.00) | 0.92 (0.86, 0.99) | 0.98 (0.90, 1.06) | 0.055 |
| Serum phosphorus (mmol/L) | 1.03 (0.86, 1.26) | 1.03 (0.86, 1.26) | 1.12 (0.78, 1.27) | 0.826 |
| Serum calcium (mmol/L) | 2.17 (2.02, 2.38) | 2.16 (2.03, 2.37) | 2.21 (2.01,2.48) | 0.889 |
| Glucose (mmol/L) | 5.89 (5.01, 7.20) | 5.86 (4.94, 6.87) | 6.93 (5.59, 8.82) | 0.010 |
| Protein urine | ||||
| 1 | 286 (73.33%) | 269 (75.56%) | 17 (50.00%) | 0.001 |
| 2 | 104 (26.67%) | 87 (24.44%) | 17 (50.00%) | |
| Blood lipids | ||||
| Total cholesterol (mmol/L) | 3.27 (2.71, 3.68) | 3.27 (2.70, 3.70) | 3.08 (2.85, 3.42) | 0.258 |
| Triglyceride (mmol/L) | 2.57 (1.82, 3.44) | 2.57 (1.77, 3.29) | 2.68 (2.06, 3.80) | 0.198 |
| High density lipoprotein (mmol/L) | 0.65 (0.46, 0.73) | 0.67 (0.46, 0.73) | 0.60 (0.45, 0.71) | 0.320 |
| Low density lipoprotein (mmol/L) | 1.41 (1.06, 1.87) | 1.44 (1.08, 1.88 | 1.29 (0.93, 1.45) | 0.091 |
| Apolipoprotein A (g/L) | 0.68 (0.56, 0.81) | 0.68 (0.56, 0.81) | 0.68 (0.56, 0.71) | 0.361 |
| Apolipoprotein B (g/L) | 0.71 (0.58, 0.79) | 0.72 (0.58, 0.80) | 0.66 (0.55, 0.74) | 0.066 |
| Coagulation function | ||||
| Prothrombin time (s) | 12.50 (11.50, 13.67) | 12.40 (11.40, 13.40) | 14.40 (13.35, 16.57) | <0.001 |
| Activated partial thromboplastin time (s) | 36.65 (29.60, 43.10) | 36.45 (29.60, 42.40) | 41.10 (27.85, 47.12) | 0.130 |
| Thrombin time (s) | 20.75 (18.00, 26.20) | 20.70 (17.98, 26.20) | 22.20 (19.15, 27.90) | 0.124 |
| Fibrinogen (g/L) | 3.09 (2.42,4.39) | 3.16 (2.50, 4.53) | 2.21 (1.83,2.96) | <0.001 |
| Myocardial enzyme | ||||
| Creatine kinase (U/L) | 87.00 (45.25, 204.00) | 87.00 (44.75, 184.00) | 108.50 (66.00, 241.00) | 0.092 |
| Creatine kinase isoenzymes (ng/ml) | 21.50 (12.00, 34.00) | 21.00 (12.00, 34.00) | 22.00 (9.50, 30.50) | 0.324 |
| Lactic Dehydrogenase (U/L) | 494.50 (346.75, 753.75) | 482.00 (345.75, 741.75) | 674.00 (436.75, 1003.00) | 0.008 |
| Hydroxybutyrate dehydrogenase (U/L) | 368.50 (268.25, 468.75) | 365.00 (263.00, 459.00) | 451.00 (331.25, 517.00) | 0.016 |
HFRS: hemorrhagic fever with renal syndrome; AP: acute pancreatitis; protein urine (3+ and 4+ were assigned 2, and 2+, 1+ and – were assigned 1).
3.2. Concurrent AP is a factor contributing to adverse clinical outcomes in patients with HFRS
After undergoing PSM, the baseline characteristics and laboratory data of HFRS patients with and without acute pancreatitis were shown in Supplementary Tables 1 and 2. Next, we found that patients with AP group exhibited increased utilization rates of CRRT and/or mechanical ventilation (55.88% vs. 17.65%), ICU admission rates (52.94% vs. 2.94%), and 30-day mortality rates (26.47% vs. 5.88%) compared to the without AP group (p < 0.05) (Table 3). These findings indicate that patients in the concurrent AP group after PSM have more severe conditions and poorer prognosis.
Table 3.
Clinical outcomes of HFRS patients with and without AP in PSM.
| Variable | HFRS (n = 68) | HFRS with AP (n = 34) | HFRS without AP (n = 34) | p-value |
|---|---|---|---|---|
| CRRT and/or mechanical ventilation | 25 (36.76%) | 19 (55.88%) | 6 (17.65) | 0.001 |
| ICU admission | 19 (27.94%) | 18 (52.94%) | 1 (2.94%) | <0.001 |
| 30-day mortality | 11 (16.18%) | 9 (26.47%) | 2 (5.88%) | 0.021 |
CRRT: continuous renal replacement therapy, ICU: intensive care unit, HFRS: hemorrhagic fever with renal syndrome; AP: acute pancreatitis; PSM: propensity score-matched.
To assess whether concurrent AP is a risk factor for adverse clinical outcomes in patients with HFRS, univariate and multivariate Logistic regression analyses were conducted with the use of CRRT and/or mechanical ventilation, ICU admission, and 30-day mortality as dependent variables, respectively, and age, smoking, stomach-ache, and gastrointestinal haemorrhage as covariates. The results showed that concurrent AP was a risk factor for the use of CRRT and/or mechanical ventilation, ICU admission, and 30-day mortality (Table 4 and Figure 2). To further reduce the impact of data bias and confounding factors, a 1:1 PSM analysis was employed, successfully matching 34 patients with HFRS who had concurrent AP and 34 patients without concurrent AP. After adjusting for age, smoking, gastrointestinal haemorrhage, and stomach-ache, it was found that concurrent AP remained a risk factor for adverse clinical outcomes in patients with HFRS, specifically: the use of CRRT and/or mechanical ventilation (p = 0.002, OR: 5.91, 95% CI 1.94–17.97), ICU admission rate (p < 0.001, OR: 37.12, 95% CI: 4.54–303.27), and 30-day mortality rate (p = 0.034, OR: 5.76, 95% CI: 1.14–29.08) (Table 4).
Table 4.
Analysis of acute pancreatitis associated with 30-day mortality rate and supportive treatment in 390 patients with hemorrhagic fever renal syndrome.
| Cure OR |
Adjusted OR |
Propensity score matchedd |
||||
|---|---|---|---|---|---|---|
| OR (95% CI) | p-value | OR (95% CI) | p-value | OR (95% CI) | p-value | |
| CRRT or/and ventilation | 13.76 (6.35–29.83) | <0.001 | 12.98 (5.80–29.07)a | <0.001 | 5.91 (1.94–17.97) | 0.002 |
| ICU admission | 78.97 (26.02–239.71) | <0.001 | 73.72 (22.57–240.72)b | <0.001 | 37.12 (4.54–303.27) | <0.001 |
| 30-day mortality | 11.29 (4.28–29.78) | <0.001 | 9.45 (3.31–27.01)c | <0.001 | 5.76 (1.14–29.08) | 0.034 |
Note: Adjusted for variables (aage; smoke; bage; smoke; gastrointestinal haemorrhage; cage; smoke; gastrointestinal haemorrhage; stomach-ache) associated with 30-day mortality, treated in ICU, treated with CRRT, treated with mechanical ventilation and the propensity score of each patient’s likelihood of being diagnosed with acute pancreatitis. dOf 390 patients, 34 pairs were matched.
OR: odds ratio; CI: confidence interval; HFRS: hemorrhagic fever with renal syndrome; AP: acute pancreatitis.
Figure 2.
Kaplan-Meier survival curves for 34 patients with HFRS complicated by AP and those without AP after PSM.
3.3. Screening of risk factors and establishment of a nomogram model for concurrent AP in HFRS
Further refinement of risk factors for concurrent AP in HFRS was conducted. Using LASSO regression, non-zero coefficient predictors were selected from variables that differed between the HFRS with concurrent AP group and the without concurrent AP group (Tables 1 and 2; Figure 3A). The optimal λ value (0.029) was chosen through 10-fold cross-validation, resulting in the selection of 10 predictive variables with non-zero coefficients, including ferritin, FIB, ALB, CREA, Gluc, WBC, PLT, gastrointestinal haemorrhage, smoking, and urine protein (Figure 3B). The 10 variables selected by LASSO regression were then subjected to multivariate logistic regression analysis. The results showed that smoking (OR: 3.702, 95% CI: 1.475–9.291), ferritin (OR: 1.002, 95% CI: 1.001–1.003), WBC (OR: 1.301, 95% CI: 1.129–1.499), FIB (OR: 0.463, 95% CI: 0.284–0.755), and PLT (OR: 0.987, 95% CI: 0.975–0.999) were independent influencing factors for concurrent AP in HFRS (Table 5).
Figure 3.
LASSO regression used to screen the predictor variable.
Table 5.
Multivariate logistic regression analysis of risk factors for AP with HFRS.
| Variable | β | S.E | Z | p | OR (95% CI) |
|---|---|---|---|---|---|
| Ferritin | 0.002 | 0.001 | 4.602 | <0.001 | 1.002 (1.001–1.003) |
| Fibrinogen | −0.770 | 0.249 | −3.171 | 0.002 | 0.451 (0.276–0.738) |
| Albumin | −0.038 | 0.046 | −0.826 | 0.409 | 0.963 (0.880–1.054) |
| Creatinine | 0.002 | 0.001 | 1.490 | 0.136 | 1.002 (0.999–1.004) |
| Glucose | 0.063 | 0.071 | 0.876 | 0.381 | 1.065 (0.925–1.224) |
| White blood cell | 0.263 | 0.072 | 3.645 | <0.001 | 1.301 (1.129–1.499) |
| Platelet | −0.013 | 0.006 | −2.112 | 0.035 | 0.987 (0.975–0.999) |
| Gastrointestinal haemorrhage | 0.813 | 0.572 | 1.420 | 0.156 | 2.255 (0.734–6.925) |
| Smoke | 1.309 | 0.469 | 2.788 | 0.005 | 3.702 (1.475–9.291) |
| Proteinuria | 0.592 | 0.473 | 1.250 | 0.211 | 1.807 (0.714–4.570) |
OR: odds ratio; CI: confidence interval; HFRS: hemorrhagic fever with renal syndrome; AP: acute pancreatitis.
Subsequently, a nomogram model for predicting the risk of concurrent AP in HFRS was developed using the aforementioned screened risk factors (Figure 4). The AUC of the model was 0.90 (95% CI: 0.84–0.95) (Figure 5A). The model was calibrated using a calibration plot (Figure 5B) and the Hosmer-Lemeshow test to assess its goodness-of-fit (χ2=8.51, p = 0.39). Further analysis of the model’s clinical utility was conducted through a decision curve analysis (DCA) (Figure 5C), which indicated a threshold probability range of 15%-98% for the model.
Figure 4.
Nomogram for predicting the risk of AP caused by HFRS.
Figure 5.
Performance evaluation of nomogram. (A) The ROC curve for the nomogram. (B) The calibration curve for the nomogram. (C) The clinical decision curve for the nomogram.
4. Discussion
The aim of this study was to analyze the impact of concurrent AP on disease severity and prognosis in patients with HFRS, to explore the risk factors for HFRS with concurrent AP, and to construct a nomogram model, thereby providing a reference for the prevention and treatment of HFRS with concurrent AP in the local area. Based on the above objectives, we obtained the following findings: (1) The incidence of HFRS with concurrent AP in this region was 8.72%; (2) Patients with HFRS and concurrent AP had more severe disease and poorer prognosis, indicated by increased use of CRRT and/or mechanical ventilation, higher ICU admission rates, and elevated 30-day mortality; (3) Through LASSO-Logistic regression analysis, smoking, elevated ferritin levels, WBC, decreased FIB levels, and PLT were identified as influencing factors for HFRS with concurrent AP; (4) A nomogram model for HFRS with concurrent AP was further constructed based on the risk factors identified by LASSO-Logistic regression, and this model demonstrated good predictive performance.
Although the primary target organ for HFRS damage is the kidney, research by Park et al. [17]. has revealed that approximately one-third of HFRS patients exhibit extrarenal manifestations, with the gallbladder and pancreas being the most commonly affected organs. Zhu et al. [12]. reported an incidence rate of 2.8% for HFRS complicated by AP. However, recent research by Wang et al. [11]. indicated that the incidence of this complication is as high as 26.32%. In comparison, the present study found an incidence rate of 8.72% for HFRS complicated by AP, which differs from the aforementioned studies. This discrepancy may be attributed to the unique geographical location of the study area as a cross-plateau region and the diversity in living habits among the local ethnic minorities.
In HFRS, the virus directly invades pancreatic tissue, leading to increased vascular permeability and subsequent plasma extravasation, which causes oedema in the pancreatic tissue [18]. Additionally, when AP complicates HFRS, patients often exhibit nausea and vomiting [13], accompanied by retrograde flow of bile and duodenal contents into the pancreatic duct, which then activates pancreatic enzymes and induces pancreatitis. This series of pathophysiological processes constitutes the possible pathogenesis of HFRS complicated by AP. More seriously, the occurrence of AP can trigger tissue necrosis, prompting cells to release more inflammatory mediators [19], thereby exacerbating multi-organ dysfunction, worsening the disease, and even leading to patient death. Research by Fanet al. [13]. indicated that in HFRS patients with multi-organ dysfunction, timely measures such as renal replacement therapy and mechanical ventilation can effectively regulate volume load, maintain internal environmental homeostasis, eliminate cytokines and inflammatory mediators, thereby improving clinical symptoms and significantly reducing mortality. In this study, it was observed that among patients with HFRS complicated by AP, there was an increased proportion requiring CRRT and/or mechanical ventilation, an elevated ICU admission rate, and an increased 30-day mortality rate. This finding is consistent with previous research by Guo et al. [10]. Therefore, it can be inferred that patients with HFRS complicated by AP have more severe disease and poorer prognosis. In view of this, an in-depth exploration of risk factors for HFRS complicated by AP and the construction of a nomogram prediction model have important practical value for the prevention and treatment of HFRS complicated by AP in the local area.
Further research has identified smoking, elevated ferritin levels, WBC, reduced FIB levels, and PLT as risk factors for HFRS complicated by AP. Previous studies have consistently shown that smoking is a risk factor for the development of AP [20–22]. The underlying mechanism may be related to increased oxidative stress and enhanced digestive enzyme gene expression in pancreatic tissue due to smoke exposure, leading to pancreatic tissue damage [23,24]. This study found a significantly increased risk of AP among smokers (OR: 3.702, 95% CI: 1.475–9.291, p = 0.005). However, it is noteworthy that, to date, no studies have reported on the relationship between smoking and HFRS complicated by AP. Therefore, there is an urgent need for more high-quality research to verify the association between smoking and HFRS complicated by AP.
The study by Che et al. [25]. revealed a significant correlation between serum ferritin levels and the severity of disease and mortality in HFRS. This may be attributed to the immunosuppressive and pro-inflammatory roles of serum ferritin during viral infection [26]. The present study found that serum ferritin is an important risk factor for HFRS complicated by AP. Although the understanding of the specific mechanisms of serum ferritin in HFRS complicated by AP is currently insufficient, further in-depth research is still needed to accurately elucidate the underlying associations.
A study by Wang et al. [11]. found that high levels of fibrinogen degradation products (FDPs) and low levels of D-dimer are important risk factors for HFRS complicated by AP, suggesting that monitoring coagulation function has important clinical value for early identification of HFRS complicated by AP. The present study further found that HFRS patients with elevated WBC, PLT, and increased FIB consumption are more prone to developing AP, a phenomenon that may be closely related to the pathogenesis of HFRS. Specifically, high loads of hantavirus directly act on endothelial cells and induce immune responses [27–29], thereby affecting pancreatic tissue. The tight junctions of endothelial cells damaged by the virus lead to increased vascular permeability, causing tissue oedema. Meanwhile, changes in PLT activation status and reduction in PLT impair vascular integrity. These changes promote the activation of the coagulation system and hyperfibrinolysis, resulting in the cleavage of prothrombin into fibrinogen monomers [1,30], ultimately forming microthrombi within capillaries and causing a decrease in FIB levels.
This study employed multivariate Logistic regression analysis and constructed a Nomogram prediction model, revealing that smoking and elevated WBC are the two most significant factors influencing HFRS complicated by AP. In contrast, Wang et al. [11]. identified alcohol consumption history, lymphocyte percentage (lym%), proteinuria, FDPs, and D-dimer as risk factors for HFRS complicated by AP. This study also observed a higher incidence of proteinuria in patients with HFRS complicated by AP but did not confirm it as an independent risk factor. Regarding alcohol consumption history and L%, there were no significant differences between patients with and without HFRS complicated by AP, which may be related to the relatively small sample size included in this study, potentially leading to data bias. Additionally, since not all patients completed testing for FDPs and D-dimer, these two indicators were not included in the analysis of this study. It is worth mentioning that this study was based on clinical practice data from two large tertiary teaching hospitals in western Yunnan, ensuring good representativeness of the data and thus providing certain reference and guidance value for the population in this region.
4.1. Limitations
This study still has certain limitations: Firstly, as a retrospective analysis, it is limited by the relative scarcity of HFRS complicated by AP cases, which may have introduced confounding bias during the screening of risk factors using LASSO-logistic regression analysis, affecting the accuracy of the results. Secondly, although the nomogram model for predicting HFRS complicated by AP was successfully constructed in this study, it has not yet undergone external validation, and its predictive performance and generalization ability need further investigation. Furthermore, due to the specificity of the hospital where this study was conducted, not all admitted HFRS patients underwent comprehensive testing for blood amylase, lipase, and urinary amylase, resulting in the exclusion of these indicators from the study scope and potentially missing some potentially important information. Lastly, the results of this study only reflect the situation of HFRS complicated by AP in western Yunnan, China, and cannot fully represent the general phenomenon nationwide. Therefore, to enhance the scientific and practicality of the research, there is an urgent need to conduct multicentre, long-term, large-sample prospective studies in the future to more comprehensively reveal the risk factors for HFRS complicated by AP and conduct rigorous external validation of the constructed prediction model, in order to provide a more reliable basis for the prevention and treatment of HFRS complicated by AP.
5. Conclusion
Patients with HFRS complicated by AP exhibit severe disease conditions and poor prognosis. Further research has identified smoking, elevated ferritin levels, WBC, decreased FIB levels, and PLT as risk factors for HFRS complicated by AP. Additionally, the nomogram model constructed based on these risk factors demonstrates good predictive performance in the application for patients with HFRS complicated by AP in western Yunnan, China.
Supplementary Material
Acknowledgments
The authors would like to thank all the patients who participated in this study.
Funding Statement
This study was supported by the Sub-Centre Project of Infectious Diseases Clinical Medical Centre of Yunnan Province and the Construction Project of Key Laboratory of Infectious Disease of Yunnan Provincial Education Department (Yunnan Provincial Department of Education Notice No. 70). and the Foundation of Yunnan Provincial Department of Education (2024J0857).
Ethics approval and consent to participate
As this study was retrospective, all patients waived informed consent. This study conformed to the guidelines of the Helsinki Declaration. Ethics approval was obtained by the Research Ethics Committee of the first Affiliated Hospital of Dali University.
Author contributions
Lihua Huang, Min Xiao and Xiaoling Huang collected and analyzed the laboratory data, prepared the tables, and authored and approved the final draft. Jun Wu and Jiao Luo reviewed the drafts of the paper. Fuxing Li and Wei Gu conceived and designed the experiments, contributed to data analysis, and approved the final draft. All authors have read and approved the final work.
Disclosure statement
The authors report no conflicts of interest in this work.
Data availability statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
- 1.Sehgal A, Mehta S, Sahay K, et al. Hemorrhagic fever with renal syndrome in Asia: history, pathogenesis, diagnosis, treatment, and prevention. Viruses. 2023;15(2):561. doi: 10.3390/v15020561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Mir S. Hantavirus induced kidney disease. Front Med (Lausanne). 2021;8:795340. doi: 10.3389/fmed. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Shen L, Sun M, Wei X, et al. Spatiotemporal association of rapid urbanization and water-body distribution on hemorrhagic fever with renal syndrome: a case study in the city of Xi’an, China. PLoS Negl Trop Dis. 2022;16(1):e0010094. doi: 10.1371/journal.pntd.0010094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Tian H, Tie WF, Li H, et al. Orthohantaviruses infections in humans and rodents in Baoji, China. PLoS Negl Trop Dis. 2020;14(10):e0008778. doi: 10.1371/journal.pntd.0008778. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chen S, Zhu J, Sun LQ, et al. LincRNA-EPS alleviates severe acute pancreatitis by suppressing HMGB1-triggered inflammation in pancreatic macrophages. Immunology. 2021;163(2):201–219. doi: 10.1111/imm.13313. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ge P, Luo Y, Liu J, et al. Eliminating COVID-19 as the immediate culprit for igniting pancreatitis. J Med Virol. 2023;95(12):e29272. doi: 10.1002/jmv.29272. [DOI] [PubMed] [Google Scholar]
- 7.Forsmark CE, Vege SS, Wilcox CM.. Acute pancreatitis. N Engl J Med. 2016;375(20):1972–1981. doi: 10.1056/NEJMra1505202. [DOI] [PubMed] [Google Scholar]
- 8.Simons-Linares CR, Imam Z, Chahal P.. Viral-attributed acute pancreatitis: a systematic review. Dig Dis Sci. 2021;66(7):2162–2172. doi: 10.1007/s10620-020-06531-9. [DOI] [PubMed] [Google Scholar]
- 9.Kilit TP, Kilit C, Erarslan S.. A rare cause of acute pancreatitis: hantavirus infection. Acta Gastroenterol Belg. 2017;80(1):59–61. PMID: 29364099. [PubMed] [Google Scholar]
- 10.Guo Q, Xu J, Shi Q, et al. Acute pancreatitis associated with hemorrhagic fever with renal syndrome: a cohort study of 346 patients. BMC Infect Dis. 2021;21(1):267. doi: 10.1186/s12879-021-05964-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wang W, Fan D, Quan B, et al. Logistic regression analysis of risk factors for hemorrhagic fever with renal syndrome complicated with acute pancreatitis. Ann Med. 2023;55(1):2232355. doi: 10.1080/07853890.2023.2232355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zhu Y, Chen YX, Zhu Y, et al. A retrospective study of acute pancreatitis in patients with hemorrhagic fever with renal syndrome. BMC Gastroenterol. 2013;13(1):171. doi: 10.1186/1471-230X-13-171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Fan H, Zhao Y, Song FC.. Acute pancreatitis associated with hemorrhagic fever with renal syndrome: clinical analysis of 12 cases. Ren Fail. 2013;35(10):1330–1333. doi: 10.3109/0886022X.2013.828187. [DOI] [PubMed] [Google Scholar]
- 14.Department of Health of the People’s Republic of China . Diagnostic criteria for epidemic hemorrhagic fever: WS 278-2008 [Z]. Beijing: China Standards Press. 2008:1-11. http://www.nhc.gov.cn/wjw/s9491/200802/39043.shtml. [Google Scholar]
- 15.Jiang H, Huang C, Bai X, et al. Expert consensus on the prevention and treatment of hemorrhagic fever with renal syndrome. Infect Dis Immun. 2022;2(4):224–232. doi: 10.1097/ID9.0000000000000054. [DOI] [Google Scholar]
- 16.Banks PA, Bollen TL, Dervenis C, et al. Classification of acute pancreatitis–2012: revision of the Atlanta classification and definitions by international consensus. Gut. 2013;62(1):102–111. doi: 10.1136/gutjnl-2012-302779. [DOI] [PubMed] [Google Scholar]
- 17.Park KH, Kang YU, Kang SJ, et al. Experience with extrarenal manifestations of hemorrhagic fever with renal syndrome in a tertiary care hospital in South Korea. Am J Trop Med Hyg. 2011;84(2):229–233. doi: 10.4269/ajtmh.2011.10-0024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Puca E, Harxhi A, Pipero P, et al. Pancreatitis in patients with hemorrhagic fever with renal syndrome: a five-year experience. J Infect Dev Ctries. 2017;11(11):900–903. doi: 10.3855/jidc.9567. [DOI] [PubMed] [Google Scholar]
- 19.Ge P, Luo Y, Okoye CS, et al. Intestinal barrier damage, systemic inflammatory response syndrome, and acute lung injury: a troublesome trio for acute pancreatitis. Biomed Pharmacother. 2020;132:110770. doi: 10.1016/j.biopha.2020.110770. [DOI] [PubMed] [Google Scholar]
- 20.Pang Y, Kartsonaki C, Turnbull I, et al. Metabolic and lifestyle risk factors for acute pancreatitis in Chinese adults: a prospective cohort study of 0.5 million people. PLoS Med. 2018;15(8):e1002618. doi: 10.1371/journal.pmed.1002618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yuan S, Giovannucci EL, Larsson SC.. Gallstone disease, diabetes, calcium, triglycerides, smoking and alcohol consumption and pancreatitis risk: mendelian randomization study. NPJ Genom Med. 2021;6(1):27. doi: 10.1038/s41525-021-00189-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mao X, Mao S, Sun H, et al. Causal associations between modifiable risk factors and pancreatitis: a comprehensive Mendelian randomization study. Front Immunol. 2023;14:1091780. doi: 10.3389/fimmu.2023.1091780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wittel UA, Singh AP, Henley BJ, et al. Cigarette smoke-induced differential expression of the genes involved in exocrine function of the rat pancreas. Pancreas. 2006;33(4):364–370. doi: 10.1097/01.mpa.0000240601.80570.31. [DOI] [PubMed] [Google Scholar]
- 24.Sliwińska-Mossoń M, Milnerowicz H, Jabłonowska M, et al. The effect of smoking on expression of IL-6 and antioxidants in pancreatic fluids and tissues in patients with chronic pancreatitis. Pancreatology. 2012;12(4):295–304. doi: 10.1016/j.pan.2012.04.007. [DOI] [PubMed] [Google Scholar]
- 25.Che L, Wang Z, Du N, et al. Evaluation of serum ferritin, procalcitonin, and C-reactive protein for the prediction of severity and mortality in hemorrhagic fever with renal syndrome. Front Microbiol. 2022;13:865233. doi: 10.3389/fmicb.2022.865233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zhao X, Zhou Y, Zhang Y, et al. Ferritin: significance in viral infections. Rev Med Virol. 2024;34(2):e2531. doi: 10.1002/rmv.2531. [DOI] [PubMed] [Google Scholar]
- 27.Hepojoki J, Vaheri A, Strandin T.. The fundamental role of endothelial cells in hantavirus pathogenesis. Front Microbiol. 2014;5:727. doi: 10.3389/fmicb.2014.00727. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Schönrich G, Krüger DH, Raftery MJ.. Hantavirus-induced disruption of the endothelial barrier: neutrophils are on the payroll. Front Microbiol. 2015;6:222. doi: 10.3389/fmicb.2015.00222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tariq M, Kim DM.. Hemorrhagic fever with renal syndrome: literature review, epidemiology, clinical picture and pathogenesis. Infect Chemother. 2022;54(1):1–19. doi: 10.3947/ic.2021.0148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Koskela S, Mäkelä S, Strandin T, et al. Coagulopathy in acute Puumala hantavirus infection. Viruses. 2021;13(8):1553. doi: 10.3390/v13081553. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.