In the present study, we found that ECG and q-SOFA score have good value in evaluating the condition and prognosis of acute poisoning. It is proposed for the first time that the effect of combined application of ECG and q-SOFA in predicting the outcome of acute poisoning is better than that of PSS score (p<0.05), and has better sensitivity and specificity. As ECG and q-SOFA score can be obtained and completed quickly after emergency admission, the combined application of the two can provide a more objective and accurate method for assessing the condition of acute poisoning, which has the potential value of popularization and application in emergency medicine.
Acute poisoning is a systemic disease that can lead to multiple system involvement and even multiple organ dysfunction. As the most important pivotal organ, the cardiac system is easily affected by various states of the whole body, and electrophysiological changes and ECG changes are one of the most sensitive indicators of cardiac effects. It has been reported that electrophysiological changes and ECG changes in acute carbon monoxide poisoning patients appear earlier than the elevation of cardiomyocyte injury markers such as CK-MB and TNI[14]. The mechanism of acute poisoning leading to ECG changes mainly include: (1) Direct action, the toxic substance can cause ECG changes by interfering with cardiomyocyte metabolism[15], directly damaging sodium, potassium and calcium ion channels[16], and destroying blood vessels[17]. (2) Indirect effects, toxic substances can indirectly affect the cardiovascular system by regulating systemic pathological alterations[18–20]. For example, carbon monoxide altered hemoglobin oxygenation, and leading to ECG changes due to hypoxia in patients[21]. Moreover, some drugs poisoning could easily lead to systemic electrolyte abnormalities and thus alter ECG[22]. (3) Arrhythmia is induced by stimulation of sympathetic nerve activity. A report suggested that some organic solvents can affect sympathetic activity and alter ECG[23]. The present study showed that ECG changes in acute poisoning included rhythm changes (34.3%) and morphological changes (65.6%). The rhythm changes were dominated by HR changes, and morphological changes were most significant by QTc changes. Meanwhile, our study revealed that HR (OR = 1.035, 95%CI 1.026 ~ 1.044) and QTc (OR = 1.013, 95%CI 1.007 ~ 1.019) were the independent risk factors for predicting poor prognosis in poisoned patients. A report confirmed that QTc can be used as a predictor of the adverse outcomes in acute poisoning[24]. Likewise, another study demonstrated that ECG can be used as an indicator of cardiovascular prognosis in patients with acute drug poisoning[25]. These findings were in agreement with our study results, which were applicable to a broader population than the aforementioned studies.
Based on a score of consciousness, systolic blood pressure, and respiratory rate, the q-SOFA score is commonly used to screen suspected sepsis patients as recommended by the Third International Sepsis Consensus Definitions Task Force[26–28]. Williams JM et al. found that the q-SOFA score can be used to screen infected patients for organ dysfunction and risk of death with a high degree of specificity[29]. Meanwhile, Cho YS et al. found that the q-SOFA score is an accurate predictor of poor outcome due to organ damage in glyphosate herbicide poisoning[10, 11]. Our data suggested that the q-SOFA score has moderate value in predicting the prognosis of poisoned patients (AUC 0.736,95%CI 0.684 ~ 0.787). Compared to ECG indicators (AUC 0.777,95%CI 0.728 ~ 0.827), the q-SOFA score has a high sensitivity, but lower specificity. One study demonstrated that ECG combined with q-SOFA score improved risk stratification for patients with acute pulmonary embolism[30]. Moreover, several reports indicate that combining heart rate variability with q-SOFA can effectively improve predictive ability for mortality in patients with sepsis[31, 32]. These studies indicated that electrophysiological alterations might precede the appearance of cellular damage, ECG could be a useful supplement to the predictive efficiency of q-SOFA. Therefore, we explored the utility of combination of ECG and q-SOFA score for acute poisoning patients.
Our study revealed that ECG combined with q-SOFA score (AUC 0.808, 95%CI 0.763 ~ 0.852) had a better predictive value than PSS score (AUC 0.617, 95%CI 0.558 ~ 0.677, p<0.05). In fact, the PSS score is a widely accepted tool for predicting the prognosis for poisoning[33–35]. However, PSS clinical utility is limited by its complexity and subjectivity[36]. Meanwhile, the score of ECG combined with q-SOFA seems more easy, objective and accuracy, which could be suitable for the acute poisoning patients with less information in stages of pre-hospital or triage in emergency department.
Inevitably, our study has some limitations. Firstly, the data was obtained from single-center with its possible selection bias. Secondly, in order to eliminate the interference of previous cardiac diseases on ECG, we excluded patients with previous cardiac injury in this retrospective study. Thus, the results are not suitable for the acute poisoning patients with previous cardiac disease.