Postoperative delirium does not affect cardiopulmonary exercise testing in aged patients undergoing cardiac valve surgery – JA Clinical Reports

The principal finding of this study is that the CPX values including AT-VO₂ and the slope of VCO₂ and VE at discharge were not significantly affected by delirium in aged patients undergoing cardiac valve surgery.

It is well documented that the development of delirium after surgery is an important predictor to increase mortality [1, 4, 5], and this has also been confirmed in aged patients [4, 6]. Accordingly, recently published clinical practice guidelines in adult ICU patients show that delirium is an independent predictor of mortality [2, 7]. Furthermore, postoperative delirium is shown to be a risk factor that worsens the quality of life, such as activities of daily living, of aged surgical patients [8,9,10]. So far, however, there have been no clinical reports that demonstrate whether delirium in ICU affects cardiopulmonary function in aged patients undergoing cardiac surgery. Olofsson et al. [11] reported that delirium causes a negative impact on the rehabilitation outcome in elderly orthopedic patients, and Uthamalingam et al. [12] documented that delirium facilitates adverse effect on cardiac function in aged patients with heart failure. Based on these previous reports, we supposed that postoperative patients with delirium may show lower cardiopulmonary function compared to those without delirium. However, unexpectedly, the cardiopulmonary function we examined in this study was comparable between the patients with and without delirium (Fig. 2). Although we have no clear idea to explain this discrepancy between the previous data and our results, we are thinking of the possibility that early introduction of rehabilitation program might contribute to prevent adverse effect on the cardiopulmonary function in patients with delirium.

There are several parameters available with CPX to predict cardiac events in patients with heart disease. Peak VO₂ is well recognized as a strong predictor [13]. However, the maximal exercise test may be risky for cardiac patients. Thus, we performed a submaximal exercise test and measured AT-VO₂ instead of peak VO₂, because these two factors were previously reported to show a significant correlation [14, 15]. AT-VO2 is a condition to be transitioned energetically to an obligatory anaerobiosis and identified by the increase of the VCO2/VO2. On the other hand, the slope of VCO₂ and VE corresponds to the increasing ventilation in response to CO2 production, so it reflects increased ventilatory drive, and it has been reported to be a prognostic parameter for patients with heart disease [3, 16]. Thus, these parameters we used may be reasonable to predict future cardiac events of the patients we studied.

One may claim that the definition of delirium in this study may be doubtful because of the retrospective study design. We have to acknowledge that the definition was not so strict compared with that of a prospective study. However, as shown in Tables 2 and 3, the duration of sedative administration in ICU in the delirium group was significantly longer and serum CRP before discharge in the delirium group was significantly higher than those values in the no delirium group [17]. We think that these data may reflect the clinical characteristics of delirium, and our grouping may be reasonable in spite of the retrospective design.

There are several limitations in this study. First, although the present study showed that delirium did not affect cardiopulmonary function at discharge, our data did not guarantee a long-term outcome. Some reports documented that delirium in ICU may be associated with cognitive dysfunction after discharge [18, 19], and it is generally accepted that delirium during ICU stay is a risk factor of a long-term patient’s outcome [5, 6, 8]. Thus, it might be important to examine the effect of delirium in ICU on CPX values after 1 or more years after cardiac surgery, even if we did not find any difference between patients with delirium and without delirium just before discharge. Second, we did not consider the severity of delirium. Because of the retrospective design, we did not obtain the data required to estimate the severity. So, we have to acknowledge the possibility that a severe degree of delirium may affect the CPX values. Third, we did not examine the CPX values before the operation. So, we could not guarantee that the CPX values in the two groups were comparable before the operation. However, preoperative examination of CPX is not clinically and ethically available in patients scheduled for cardiac operation. Fourth, the subjects of our study were a small population of patients undergoing valve surgery, and we included all types of valve disease and operation. Thus, for example, if we limited the subjects to patients with aortic valve stenosis, another conclusion might be led. Now, we are accumulating the number of subjects and would like to present the data in the future.