Sub-clinical hepatic encephalopathy in cirrhotic patients is not aggravated by sedation with propofol compared to midazolam: A randomized controlled study

Background & Aims

The risk of exacerbating sub-clinical hepatic encephalopathy (HE) by propofol has not been established. The aim of this study is to determine whether the use of propofol, for upper endoscopy in patients with cirrhosis, precipitates sub-clinical HE.

Methods

Sixty-one patients with compensated HCV and HBV cirrhosis (CP score 5–6) were randomly selected and divided into two groups (intent-to-treat population) matched for age, gender, and BMI. The first group received a single propofol sedation (N = 31, age 57 ± 12, dose range 70–100 mg/procedure) and the second group (N = 30, age 56 ± 12, dose 3–6 mg/procedure) received a single midazolam sedation, all done by an anesthesiologist. All patients completed number connection test (NCT), cognitive function score, time to recovery, time to discharge sheets, and hemodynamic parameters before sedation, and at discharge from the endoscopy unit, 1 h post-procedure. Thirty control subjects without cirrhosis were matched to the cirrhotic patients who received sedation with regard to age, gender, BMI, and education level.

Results

A total of 58/61 cirrhotic patients (95%) had sub-clinical encephalopathy before the endoscopy (mean NCT 84.7 ± 77 s, normal <30 s). No patient developed overt HE after sedation. There were no differences between groups in the incidence of adverse effects, cognitive function, MELD score, CP score, oxygen saturation, or respiratory and heart rates before and after sedation. Propofol did not exacerbate minimal HE when compared to midazolam (NCT changed from 87.5 ± 62 s prior to sedation to 74.2 ± 58 s after sedation in the propofol group versus 72.8 ± 62 s before to 85.6 ± 72 s after sedation in the midazolam group; p <0.01). Time to recovery (4.1 ± 1.9 min vs. 11.5 ± 5.0 min, p <0.001), and time to discharge (38.0 ± 9 min vs. 110 ± 42 min, p <0.001) were significantly shorter with propofol than midazolam. Pre- and post-procedure NCT (from 25 ± 20 s to 24 ± 20 s), cognitive function score (from 25 to 26), time to recovery (3.5 ± 1.0 min), and time to discharge (35 ± 10 min) did not change in the healthy controls.

Conclusions

Sedation with propofol has a shorter time recovery and a shorter time to discharge than midazolam and does not exacerbate sub-clinical hepatic encephalopathy in patients with compensated liver cirrhosis.

Introduction

Hepatic encephalopathy (HE) is a complex neurological disorder associated with advanced liver disease of either acute or chronic nature [1]. A milder form, referred to as sub-clinical hepatic encephalopathy (minimal hepatic encephalopathy), is defined as a condition in which patients demonstrate quantifiable neuropsychological defects, yet appear to have a normal mental and neurological status on clinical examination [2]. These patients can work, operate machinery, and drive a car, yet may be at increased risk of injury [3], [4].

Endoscopy is routinely performed in patients with chronic liver disease to screen for complications related to portal hypertension such as esophageal and gastric varices and portal gastropathy [5], [6]. Sedation is frequently administered to facilitate patient tolerance [7]. Patients with hepatic dysfunction, who undergo endoscopy, are at increased risk for complications related to sedation. There are few studies quantifying the risk in this population [8]. Since most drugs are metabolized in the liver, the choice of endoscopic sedatives may place cirrhotic patients at risk for either enhanced effects of the medication due to higher plasma level or prolonged effects due to delayed clearance [9]. The standard medications, meperidine and benzodiazepines, are predominantly metabolized in the liver and cirrhosis impairs first pass clearance as well as metabolism [10]. The lower plasma levels of albumin may also decrease drug binding, leading to an increase of absorption in the tissues, prolonging clearance and thus exacerbating sub-clinical hepatic encephalopathy [11]. Midazolam is a benzodiazepine depressant of the central nervous system that is commonly used in synergy with the opioid analgesic meperidine for conscious sedation in the general population undergoing upper GI endoscopy [12], [13]. This combination is widely used because of its short-acting sedative (half-life less than 6 h for midazolam), anxiolytic, and amnesic properties [14], [15]. A major component of the half-life is the production of active metabolites that may have different half-lives. However, for midazolam, the initial metabolite, alpha-hydroxyl-midazolam, has a short half-life as well [16]. Thus, liver patients are at potentially greater risk for complications, including cardiopulmonary compromise, and possible precipitation or exacerbation of encephalopathy, including sub-clinical encephalopathy [8], [17]. Propofol is a short-acting anesthetic agent that can be successfully used as a sedative drug during GI endoscopy. It is rapidly metabolized in the liver by conjugation with glucoronic acid (phase 2 reaction) [18]. Structurally, it is a substituted phenol agent (2,6-diisopropyl phenol), which is believed to exert its effect on the central nervous system through promotion of γ-amino butyric (GABA) activity in the brain [18]. It has a favorable pharmacokinetic profile in comparison to the benzodiazepines and opioids with regard to rapid induction of sedation, faster recovery, and equivalent levels of amnesia [19], [20]. The depth of sedation produced is typically greater than with traditional sedatives used in the outpatient setting. Although propofol undergoes hepatic metabolism, no adjustments in dosage are required in patients with chronic liver disease [19], [20]. It has been used effectively and safely in a small study of cirrhotic patients who exhibited significantly shorter induction and recovery periods [21]. It was safe even in elderly individuals and in those with multiple co-morbid conditions [22]; however, the effect on hepatic encephalopathy was not investigated. More recently, a trial with a close design has been published by Riphaus et al. [23]. However, his study differs from our study by several points including the etiology of liver disease (alcohol vs. viral), selection of the control groups (healthy individuals without sedation and without endoscopy vs. healthy individuals with sedation and with endoscopy), severity of pre-existing liver disease (six patients with Child-Pugh’s Class C vs. none with Child-Pugh’s Class C), amount of propofol (50–320 mg vs. 70–100 mg), and finally the timing of the number connection test evaluation (2 h vs. 1 h). We have previously demonstrated that sedation with midazolam exacerbates sub-clinical hepatic encephalopathy in patients with Child-A or Child-B cirrhosis [17]. In this study, we have investigated the effects of propofol compared to midazolam on the degree of sub-clinical hepatic encephalopathy and mean time to achieve full recovery in patients with liver disease who are undergoing upper endoscopy screening.