The present study is the first to investigate the time-dependent increase or decrease in AChE-activity in septic patients with suspected SAE. Over a period of 5 consecutive days after admission to intensive care unit statistically significant changes occurred compared to baseline. In contrast, no change in AChE-activity was observed in non-septic patients even with delirium or cognitive impairment.
AChE-activity is downregulated due to a decrease of acetylcholine which is the substrate for acetylcholinesterase. However, despite a substrate deficiency AChE- activity can also be upregulated (see discussion below).
Decrease of AChE-activity
In the present study, the time-dependent decrease in AChE-activity in the majority of septic patients with suspected SAE is consistent with the results of Bitzinger et al. They reported a significant and time-dependent decrease in AChE-activity in a rat cecal ligation puncture (CLP) - model (21). In patients with sepsis and suspected SAE, inflammation appears to be a major cause for the alteration of cholinergic metabolism and thus AChE-activity. Sepsis leads to increased formation of oxygen radicals, which in turn can cause neuronal damage (34). If neuronal damage affects cholinergic neurons, the transmitter acetylcholine is subsequently reduced and the activity of the surrogate parameter AChE-activity is altered. A deficiency of acetylcholine can manifest itself as delirium, i.e. attention and memory deficits (35). Reduced concentrations of acetylcholine are also involved in the symptoms of Alzheimer's disease. Thus, similarities with the pathophysiological changes in patients with SAE can be assumed. Méndez-Garrido et al. were able to show that in patients with Alzheimer's disease, higher concentration of reactive oxygen species, e.g. H2O2, in central nervous system were detected. Oxidative stress decreases the AChE-activity and simultaneously increases the acetylcholine hydrolysis, which ultimately contributes to a central cholinergic deficiency (36).
Increase of AChE-activity
In the present study, a time-dependent increase in AChE-activity was demonstrated in about one third of the septic patients with suspected SAE. In a CLP-induced sepsis model, the surviving mice showed a decrease in cholinergic neurons in the basal forebrain, a significant increase in AChE-activity and an increase in expression of their coding gene in the hippocampus and cortex, probably caused by microglial activation (23). Enhanced AChE-activity leads to an increased breakdown of acetylcholine and ultimately to a cholinergic deficit, which is associated with characteristic symptoms such as memory disorders, disorientation, hypo- or hyperactivity (7,8). In this context of interest is the hippocampus as interface between short-term and long-term memory: changes in cholinergic transmission in the hippocampus of septic patients appear to play a central role in the pathogenesis of septic-associated encephalopathy. Many of the symptoms associated with SAE, such as memory disorders, attention deficits and consciousness disorders, can be attributed to changes in this particular area of the brain. Zivkovic et al. “identified the hippocampus as the site of dysfunction and pathology in sepsis induced delirium” by Magnetic Resonance Imaging (MRI) (20). Given the possible pathophysiological changes postulated in SAE, both an increase and a decrease in AChE-activity seem plausible.
Considering the current evidence on the importance of esterase activities in patients with sepsis, most studies refer to changes in butyrylcholinesterase activity (BChE-activity), also called non-specific plasma esterase (33,37–39). Together with AChE-activity, BChE-activity is responsible for the maintenance and regulation of central cholinergic transmitter homeostasis. BChE-activity is subject to many different impacts and has therefore proven to be more of an outcome parameter (39,40). As a surrogate for the central transmitter status it is too inaccurate (41).
Clinical-therapeutic relevance of AChE-activity in patients with suspected SAE
Until now, proof of efficacy of cholinesterase inhibitors for the treatment of delirium in critically ill patients could not be successfully demonstrated. However, in various experimental sepsis models an improved anti-inflammatory immune response was demonstrated by the administration of indirect parasympathopmimetic drugs crossing the BBB (40,42,43). Assuming that sepsis leads to increased permeability of the BBB, activation of microglia and damage to cholinergic neurons, the administration of indirect centrally acting parasympathomimetic drugs could have positive effects in patients with SAE. The ex-juvantibus administration of indirect parasympathomimetics which has been propagated up to now only leads to an improvement of cognitive symptoms in a few cases. Therefore, it seems to be crucial to identify those patients in whom a change in central cholinergic transmitter homeostasis is actually present as the cause of delirious symptomatology/SAE.
In summary, in septic patients with positive CAM-ICU and suspected SAE, there was a statistically significant increase or decrease in AChE-activity for at least 5 consecutive days compared to baseline. In contrast, in non-septic patients with delirium or cognitive dysfunction no statistically significant change in AChE-activity could be detected during the observation period. Therefore, in contrast to single measurements longitudinal measurement of AChE-activity in septic patients with delirium is able to diagnose SAE.
Limitations
The present study has some limitations that need to be discussed. The study was conducted in an interdisciplinary surgical intensive care unit and planned as a prospective observational study. Two patient groups were distinguished (septic and non-septic patients) which differ considerably from each other in terms of group size and need for admission. Further subgroup analysis was not possible with the present case load.
A major limitation is the different sample size between the septic group and the non-septic group. However, in a prospective observational study, a different group size is not uncommon and the study plan ruled out an observation period longer than 12 months. Therefore, it was not possible to boost the case number. Due to this the calculated statistical power is 60% which may lead to underpowered results. To increase the statistical power up to 80% the sample size should have been 100 patients in each group. From previous studies it was known that a statistically significant change in AChE-activity if present could be postulated in septic patients also in small patient numbers (28). If the statistically significant results of the present study are taken into account it can be assumed that the calculated power is not only 60% but underpowered results cannot be excluded with absolute certainty. Another point to discuss is the impossibility to define precisely the onset of sepsis in non-experimental studies. However, since sepsis can be classified into different phases it is conceivable that the increase or decrease in AChE-activity depends on the course of sepsis. The change in AChE-activity over the observation period may have been influenced also by other factors such as the application of anticholinergic drugs. Anticholinergic drugs (such as furosemide and opioids) have to be administered regularly and usually without alternative in ICU patients. In addition, SAE is a diagnosis of exclusion. Reliable diagnostic tools are missing. The CAM-ICU has been validated for the diagnosis of delirium but it is not known whether the high sensitivity and specificity also applies to patients with suspected SAE.