© 2008 American Thoracic Society doi: 10.1164/rccm.200804-554ED
Arrhythmias in the ICUWhat Do We Know?
Universitätsklinik für Innere Medizin Management of arrhythmias is undoubtedly one of the major problems in emergency and critical care medicine; however, there are only limited data on the frequency of admissions with a primary arrhythmia diagnosis and on the incidence and type of arrhythmia during the intensive care unit (ICU) stay (1–5). In the ICU setting, an arrhythmia incidence of up to 78% has been reported (2). However, this retrospective study was too inclusive, including even patients with ventricular or supraventricular premature beats or bundle branch block, which do not represent the common perception of a significant, sustained arrhythmia. Other prospective observational studies reported an annual incidence of 14.9% (4) and between 15.7 and 19.7% (3) in a surgical and a medical ICU, respectively. In the current issue of the Journal (pp. 20–25), the article by Annane and coworkers (6) examines the incidence of arrhythmias in a general ICU population. The authors did a meticulous job in collecting data on a large number of patients and used rigorous criteria for the classification of sustained arrhythmias. The study confirms previous data on the incidence of arrhythmias and extends existing knowledge in several ways. Annane and colleagues report in this multicenter 1-month cohort study a 12% incidence of ventricular plus supraventricular arrhythmias for a general ICU population. This incidence is in line with the aforementioned data from surgical and medical–surgical ICUs. When comparing the incidence against previous studies, it should be recognized that the case mix may not have been so different when compared with previous work. Only 9 ICUs were mixed, whereas 14 were medical and 3 were surgical ICUs. Although the majority of patients were admitted for noncardiac disease (87%), 60% of all patients and 86% in the arrhythmia group had a cardiovascular history. Arrhythmia only rarely appears to be a diagnosis for primary admission to the ICU (1). Rather, arrhythmias occur during the ICU stay. It is evident that the occurrence of arrhythmia will depend on an underlying disease and thus on the case mix. It is therefore interesting that the incidence of arrhythmia is virtually identical in surgical, medical, and cardiologic ICUs when contemporary arrhythmia definitions are used. Moreover, and even more interesting, the median time to occurrence of specific arrhythmias in the "general" ICU population of this study was comparable to the time course of atrial fibrillation (AF) in noncardiac postoperative patients (7), being around Days 1–2. Several possible explanations exist for these intriguing findings. First, the case mix may have been identical in all studies, which is very unlikely. Second, very different etiologies, such as chronic obstructive pulmonary disease, acute respiratory distress syndrome, pulmonary embolism, and valvular heart disease, may lead to a final common pathway of arrhythmia as is the case for AF or flutter. Third, the ICU environment, with interventions such as mechanical ventilation, vasopressors, and inotropes or invasive procedures, may be the cause of arrhythmia. Although sympathetic tone definitely is a trigger for arrhythmias (8) and sudden cardiac death (9), conflicting data exist as to the influence of analgesia and sedation. The circadian occurrence of arrhythmia through the 24-hour day does not appear to be influenced by the presence or absence of sedation (10). Fourth, common pathophysiologic mechanisms operative in critical illness might predispose to arrhythmia. One process common to critical illness and multiorgan failure of different origin is systemic inflammation. Inflammation has also been suggested to be a possible mechanism for AF (11), which is the single most common arrhythmia in the ICU. However, the occurrence of arrhythmia episodes did not coincide with a peak in inflammation parameters in an uncontrolled study (12). In favor of this fourth possibility, Annane and colleagues, in the online supplement, report a 5.3% incidence of supraventricular arrhythmia, mainly AF, when excluding primary cardiac diagnoses and chronic arrhythmia. Even in trauma patients with a very different etiology of critical illness, AF was a predominant arrhythmia (5). Similar to previous data, AF was the single most frequent arrhythmia in the current study. AF carries a two- to six-fold risk of stroke, and the rate of ischemic stroke in nonvalvular AF averages 5% per year (13). The stroke rate secondary to new onset AF in critically ill patients in the ICU is essentially unknown. For the first time in a prospective study, data on neurological consequences of true AF in the ICU are reported by Annane and coworkers. The authors identified 4 of 87 patients with focal neurological deficits accounting for a rate of 4.6% during their 1-month observation period. This is in line with data on AF in the non–critically ill population. Thus, the current study presents important data on the stroke risk in the ICU, but certainly requires confirmation in a larger number of patients. Some studies suggested that supraventricular arrhythmias are associated with higher mortality (2, 14). It has been unclear so far whether this association merely reflected an association of arrhythmias with a higher severity of disease. The current study adds a new aspect to this discussion by adjusting for a variety of important covariates, among which were age and the Simplified Acute Physiology (SAPS) II score. The authors found that supraventricular arrhythmias were no longer associated with a poorer hospital/90-day survival after covariate adjustment and after propensity score matching. This finding shows that AF and supraventricular tachycardia are associated with higher comorbidity and are indicators of sicker patients but not necessarily a harbinger of a poorer outcome. FOOTNOTES Conflict of Interest Statement: G.H. has no financial relationship with any commercial entity that has interest in the subject of this manuscript. REFERENCES
Related articles in AJRCCM:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
