| BackgroundAcute heart failure (AHF) describes the rapid onset of, or changes in, symptoms and signs of heart failure (HF), and needs immediate medical attention on the emergent condition [1]. In most cases, AHF is performed as the first presentation of HF. AHF may be caused by an inherited or acquired abnormality of any aspect of cardiac function.There is often a clear incentive or precipitant in patients with preexisting HF, such as severe anemia, arrhythmia, volume overload, oliguria and hypoxemia. In the ’acute’ phase, cardiac dysfunction can exacerbate in a period of days or even weeks, but others developing HF within hours to minutes (e.g. during the event of acute myocardialinfarction). However, until now there have been no methods for predicting the onset of AHF within hours to minutes. A prediction model could enable AHF therapy to be different from the traditional treatment of acute or chronic heart failure. AHF has emerged as a major public health problem over the past decade.AHFS can be defined as new onset or gradual or rapidly worsening HF signs and symptoms requiring urgent therapy [2]. Irrespective of the underlying cause (e.g., ischemic event) or precipitant (e.g., severe hypertension), pulmonary and systemic congestion due to elevated ventricular filling pressures with or without a decrease in cardiac output is anearly universal finding in AHFS [2]. Coronary artery disease (CAD), hypertension, valvular heart disease, and/or atrial fibrillation, as well as noncardiac conditions such as renal dysfunction, diabetes, anemia, and medications (i.e., nonsteroidal anti-inflammatory drugs, glitazones), may also contribute to these abnormalities [3,4]. The majority of AHFS patients have worsening chronic HF; after initial management resulting in stabilization, they should no longer be considered acute but chronic HF [5].Acute coronary syndrome (ACS) is one of commonest illnesses seen in chest pain centers (CPCs). In 2009, close to 1.2 million hospital discharges in CPCs of the United States had an ACS diagnosis [6]. ACS results in significant morbidity and mortality [7], with HF as a frequent complication [8,9]. Additionally, nearly 700,000 emergency department (ED) visits in 2009 were due to AHF [10]. According to the survey of American Heart Association, the prevalence of HF is increasing from 2.8% in 2010 to 3.5% in 2030, which means an additional 3 million patientswill be affected [11]. It produces a 215%increase in projected direct medical costs (from $24.7 billion to$77.7 billion) and an 80% increase in projected indirect costs (from $9.7 billion to $17.4 billion) from 2010 to 2030 [12].Early prediction and identification of the onset of acute heart failure (AHF) in high-risk patients is of great significance for preemptive treatment and a better prognosis [13]. We sought to find a scoring system to predict the onset of AHF in patients in the acute heart failure unit (AHFU).The modified early warning score (MEWS) was a simple physiological scoring system suitable for bedside application at the emergency room [14]. MEWS could be used to identify medical patients at risk of catastrophic deterioration in a busy clinical area. The MEWS is best regarded as a defined judgment on routinely recorded physiological data for triage in the ED. Similarly, a new scoring system for early warning the onset of AHF was needed in the acute heart failure unit (AHFU).Usually the treatment strategy of acute and chronic heart failure was different; chronic heart failure was often combined with liver and renal insufficiency, poor activity tolerance, high level of N-terminal B-type natriuretic peptide (NT-proBNP), and moderately elevated of cardiac tropin I (cTnI). In cases of factors such as mood changes, anemia, fatigue, or iatrogenic factors, AHF may be judged to have occurred according to clinical observation. Vital signs and other monitoring factors appear to be changed before the onset of AHF. For example, patients may experience increases in heart rate, breathing rate, blood oxygen saturation, hourly urine rate, or abnormal emotions (e.g. restlessness, excitement, agitation, overstimulation, delirium, depression, apathy, unresponsiveness, lethargy, drowsiness, and coma). Emergency treatment at the onset of AHF is expected to provide advance intervention and may be able to reduce a heart attack. The treatment of acute and chronic heart failure may be combined by an appropriate cutoff point, which could not only improve the quality of survival, but may improve the prognosis of patients with the risk of AHF.Methods1. Study design and patient populationThis study was performed at the AHFU and CPC of Qilu Hospital. Inclusion criteria were any patient admitted to the clinic room who was triaged to the intensive care unit of CPC or AHFU owing to a high-risk first assessment. Patients in the AHFU have a higher risk of death and heart failure. We recorded sex, age, history of coronary artery disease (CAD), hypertension, diabetes, primary percutaneous coronary intervention (PPCI), and then reviewed temperature, pulse, SpO2, respiratory rate, blood*pressure, urine volume, and emotional state every hour. All admission and follow-up data were retrieved from the hospital charts. The study conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by Hospital’s Ethics Committee. The goal of the study is to enroll a representative patient sample. For the purpose of this study, acute heart failure is defined as either new-onset heart failure or decompensation of chronic, established heart failure with symptoms sufficient to warrant hospitalization. Patients are identified for inclusion in the study from admissions given a discharge diagnosis of heart failure based on International Classification of Diseases, Tenth Revision (ICD-10) coding. An ’Acute heart failure’ in the medical records was described as a. the patients’complaint of dyspnea or orthopnea, b. physical examination of moist and dry rales in both bottom of lung, decreased oxygen saturation, and c. rescue measures include oxygen, non-/invasive ventilator, diuretics, morphine, and digitalis for emergency therapy of an AHF attack.The cases with none of chief complaint (unless the tracheal intubation), physical examination, or rescue measures in patients were excluded. Eligibility is not contingent on the use of any particular therapeutic agent or regimen. Patients may be male or female and must be at least 18 years old at the time of hospital admission. The study is accumulating data on individual hospitalizations, not individual patients, and it is possible that some patients may be enrolled in the study more than once.2. Study endpointThe main endpoint of this study was all-cause mortality (in-hospital or after discharge). The secondary endpoint was the onset of AHF, which had been descripted and must include the treatment of oxygen, non-/invasive ventilator, diuretics, morphine, and digitalis in the medical records.3. Statistical analysisStatistical analysis was performed according to the SPSS 17.0 statistical software package (SPSS Inc., Chicago, IL, USA). All numerical data were expressed as the mean±standard deviation or median depending on normality. Differences between groups were assessed by means of the Student’s t-test when normally distributed, and for count data we used the Fisher’s exact test. Comparisons of all proportions were made using a chi-square analysis.The baseline characteristics and monitored factors, including sex, age, and history of CAD, hypertension, diabetes, PPCI, temperature, pulse, SpO2, respiratory rate, urinary volume and emotional state were put into a binary logistic regression model. The significant variables were screened out to make the scoring model. The cutoff value of each index was determined according to the receiver operated characteristic (ROC) curve to calculate Youden’s index, in order to formulate the scale.We used the new scoring system to evaluate patients retrospectively, and compared with the MEWS, to compare the advantages of the two scoring methods in early warning of AHF. The risk of AHF occurrence can determine the low-, moderate-, high-, and extremely high-risk patients. All patients received an average follow-up of 6 to 24 months. The survival rates were analyzed between the groups assigned based on the cutoff values of the ROC curves for each scoring system using Kaplan-Meier curves, and significant differences were calculated using the log-rank test. A p-value of less than 0.05 was considered to be statistically significant.Results1. Baseline characteristicsData for 433 patients assessed in the CPC or AHFU of Qilu Hospital affiliated Shandong University from November 2011 to June 2014 were included; 83 patients died in hospital. A total of 420 AHF in-hospital events were recorded.The study population included 264 men (61.0%) and 169 women (39.0%), and the mean age was 64.1±15.7 years (18-89 years). All patients had one or more risk factors for AHF, such as NYHA class III to IV, advanced age, ACS, PPCI for ST segment elevation myocardial infarction (STEMI) and myocarditis, postoperative cardiopulmonary resuscitation (CPR), malignant arrhythmia, and accompanied by chronic kidney disease. There were 278 ACS patients, of which 120 received PPCI, and 21 underwent CPR.287 (66.3%),101 (23.3%),40 (9.2%) patients had a history of CAD, hypertension, and diabetes, respectively. A total of 163 patients had AHF 420 times (0.97 times for all patients on average). All-cause in-hospital mortality was 19.2%.2. Physiological parametersSex, age, history of CAD, hypertension, diabetes, PPCI, temperature, pulse, SpO2, respiratory rate, urine volume, and emotional state, were put into a binary logistic regression model (Table 1). Thus, significant variables including age (OR:1.027,95% CI:1.010-1.044, p<0.05), emotion (OR:1.519,95% CI:1.065-2.168, p<0.05), SpO2 (OR:0.883,95% CI:0.809-0.963, p<0.05), urine volume (OR:0.985,95% CI: 0.981-0.989, p<0.05), pulse (OR:1.014,95% CI:1.002-1.026, p<0.05), and respiratory rate (OR:1.075,95% CI:1.031-1.121, p<0.05) are screened out to make the scoring model. Because the age is relatively constant, we retained the other five variables. Using the various indicators of the ROC curve, we determined Youden’s index to determine the cutoff value of each index. This scoring system was fine-tuned for ease of use, so as to formulate a scale using the five parameters SpO2, urine volume, pulse, emotional state, and respiratory rate, the’SUPER’score.The SUPER score predicted the onset of AHF 3.90±1.94 h (1-17 h) earlier. We then compared the SUPER score of patients with the MEWS score. The area under the ROC curve values for SUPER+Age, SUPER, and MEWS were 0.820,0.811, and 0.662. Both SUPER+Age and SUPER predicted the onset of AHF better than MEWS (p<0.05), with no statistically significant differences between them. 3. In-hospital and post-discharge outcomesAll patients were followed up in hospital and post-discharge for 6-24 months. According to the level of increased risk informed by SUPER score, we divided the patients into four groups:group 1,0-1 points; group 2,2-3 points; group 3,4-5 points; group 4,6-10 points. The incidence of AHF by hours in groups 1,2,3 and 4 were 17.3%,61.3%,84.4%, and 94.0%, respectively (p<0.05). Survival rates for the groups were assessed using Kaplan-Meier curves, and significant differences were calculated using the log-rank test. The all-cause mortality of the four groups either in-hospital or post-discharge showed statistically significant differences (p<0.05). Consequently, the SUPER score may be used for risk stratification, to identify the low-, moderate-, high-and extremely high-risk patients.Conclusions1. In patients at high risk of AHF, the SUPER scoring system could predict the onset of AHF 2 to 6 hours earlier.2. Preemptive treatment according to the SUPER score may prevent or delay AHF occurrence to improve quality of life and reduce mortality.3. The higher the patient’s score, the higher was their mortality. We expect this to help in risk stratification for AHF. |
