Dilatation Therapy and Demographic Characteristics Significantly Influence the Amount of Propofol for Therapeutic Endoscopic Retrograde Cholangiography

Background and Study Aims Patients undergoing therapeutic endoscopic retrograde cholangiography (ERC) may require different amounts of sedative agents depending on demographic characteristics, indication of ERC, and/or endoscopic intervention. Patients and Methods We retrospectively analyzed all patients undergoing therapeutic ERC from 2008 – 2014 who received deep sedation with propofol ± midazolam. Results A total of 2448 ERC procedures were performed in 781 patients. The cumulative per procedure propofol dose in the different groups was as follows: PSC 479 mg (±256), bile duct stones 356 mg (±187), benign stenosis/cholestasis 395 mg (±228), malignant stenosis 401 mg (±283), and postliver transplant complications 391 mg (±223) (p < 0.05). Multivariable analysis showed that dilatation therapy (p = 0.001), age (p = 0.001), duration of the intervention (p = 0.001), BMI (p = 0.001), gender (p = 0.001), platelet count (p = 0.003), and bilirubin (p = 0.043) influence independently the propofol consumption. Conclusions Demographic characteristics and endoscopic interventions have a distinct influence on the amount of sedation required for therapeutic ERC. Although the sedation-associated complication rate is low optimization of sedative regimens is a prime goal to further reduce adverse events of therapeutic ERC.


Introduction
Endoscopic retrograde cholangiography (ERC) is an invasive procedure which may be associated with patient discomfort, abdominal pain, and potentially life-threatening complications [1,2]. Consequently, sedation is an essential aspect of ERC as it may reduce stress, anxiety, and pain of patients leading to higher acceptability and improved conditions for the procedure [3,4]. Main indications for therapeutic ERC are management of benign and malignant bile duct strictures, pancreatic diseases, removal of bile duct stones, treatment of biliary complications after liver transplantation and others [5][6][7][8]. One established sedation regimen for ERC includes the application of propofol ± midazolam [9].
Major factors influencing the pharmacokinetic profile and clinical effects of propofol are gender, weight, and age [3]. These factors and changes in the need of sedation for different endoscopic procedures are well described [10][11][12]. Increasing evidence suggest that distinct patient groups may require different amounts of sedative drugs for completion of therapeutic ERC independent of gender, weight, and age. In a prior study we could show that patients with primary sclerosing cholangitis (PSC) require more sedative agents for ERC compared to control groups [13]. These findings prompted us to further analyze the efficacy, safety, and factors influencing the required doses of anesthetic drugs in patients undergoing ERC at our institution. 2 International Journal of Hepatology study as well as patients receiving opioids during the intervention. In case of repetitive endoscopic examinations, all presentations were included in the analysis. Demographic characteristics, duration (minutes (min)) and time point of the intervention, underlying diseases, and the application rate of the anesthetics (amount of anesthetics) were extracted from the endoscopy database. Deep sedation was performed by intermittent bolus application of propofol with or without midazolam as premedication. No additional application of analgesics was performed. The sedation was administered and monitored by the endoscopists (regularly one endoscopist for examination and one endoscopist for sedation). Routinely, drugs were given intravenously. For induction, approximately 0,05 mg of midazolam per kg/body weight and 0,5 mg of propofol per kg/ body weight were given. For maintenance, regular incremental boluses were applicated. An infusion pump was not regularly used for propofol sedation. During ERC a deep sedation level (Ramsay sedation scale of 5-6) with maintained cardiovascular and respiratory function was targeted and controlled by a gastroenterologist who is expert and trained in advanced life support skills.
All physicians performing ERC at our institution are experienced endoscopists (> 300 examinations, > three years of regular ERC performance). The study was approved by the local institutional Ethics Committee (Ethics Committee of Hannover Medical School). Data were analyzed retrospectively and anonymously. Consequently, no written informed consent is available or necessary. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki, as reflected in a prior approval (June 2017) by the institution's human research committee.

Statistical Analysis
Baseline characteristics at the time of the first intervention are presented as absolute and relative frequencies for categorical variables and mean ± standard deviation, unless denoted otherwise. Our study analyzes the influence of different variables (endoscopic, demographic) on the propofol dose for completion of therapeutic ERC. Propofol dose served as the dependent variable. A logarithmic transformation was applied to all metric variables which showed a skewed distribution on the original scale. In case of more than 2 values for categorical variables a dummy transformation was performed.
In a first step we checked the influence of each independent variable separately in univariate models. Subsequently, significant (p < 0.05) variables were entered into a multivariable linear regression model and a backward selection based on the Wald statistic was performed. All analyses are of explorative character and no multiplicity correction was applied. The software used was the SPSS Statistical Package (version 19.0, SPSS Inc., Chicago, IL).  Table 1.

Results
A papillotomy was performed or had already been performed prior to the study period in 71% of the patients with PSC, 81% of patients with bile duct stones, 70% of patients with benign stenoses/cholestasis, 60% of patients with malignant stenoses, and 84% of patients with postliver transplant complications (p < 0,001). A dilatation therapy was applied in 30%, 3%, 13%, 27%, and 42% for the aforementioned groups, respectively. All endoscopic interventions and measures for the different patient groups are given in Table 2.
The mean duration of the endoscopic interventions was  Table 2.
The total propofol consumption in the different groups was as follows: PSC 479 mg (±256), bile duct stones 356 mg (±187), benign stenoses/cholestasis 395 mg (±228), malignant stenoses 401 mg (±283), and postliver transplant complications 391 mg (±223) (p < 0.05). Post hoc analysis showed that patients with PSC had a significantly higher demand of propofol compared to all groups (p < 0.05). Patients with malignant stenoses received significantly more propofol than patients who were treated because of biliary stones (p = 0.033). The propofol consumption was equal over time and revealed no progression in case of repetitive ERC procedures for all patients (up to 10 interventions). Patients with PSC and postliver transplant complications received significantly more midazolam compared to all other groups (p < 0.01; PSC 5 mg (±1), postliver transplant complications 5 mg (±1); other groups 4 mg (±1)).
In order to identify independent variables which influence the propofol consumption during ERC a multivariable      (Table 3). When the regression analysis was restricted to the first presentation of the patients (n = 781; exclusion of repeated measurements) dilatation therapy (p = 0.013), age (p = 0.001), duration of the intervention (p = 0.001), BMI (p = 0.05), gender (p = 0.008), platelet count (p = 0.017), and bilirubin (p = 0.036) were significantly associated with the propofol consumption. Patients undergoing dilatation therapy did not receive more midazolam compared to patients without dilatation (midazolam 4 mg (±1) for both groups; p > 0.05).

Discussion
Sedation for ERC is of paramount importance to facilitate complex endoscopic interventions [9]. However, sedation may be associated with severe adverse events and therefore an adequate monitoring and intervention-tailored sedation regimen is necessary [9,13].
To our best knowledge we systematically analyzed for the first time the influence of demographic and endoscopic measures on the required amount of sedation for completion of therapeutic ERC. Multivariable analysis showed that dilatation therapy is independently associated with an increased need for sedation. In our model dilatation therapy led to an additional application of approximately 30 mg of propofol. This finding is plausible as resolving of malignant or scarred benign strictures is generally linked to patient discomfort leading to application of higher doses of sedation/analgetics. Moreover we could show that depending on the indication for ERC (PSC, benign stenoses, malignant stenoses, removal of bile duct stones, and postliver transplant complications) a dilatation therapy is more or less likely which facilitates an adaptation of the sedative regimen. Other endoscopic interventions or measures such as papillotomy, stent placement, stent size, contrast injection in the pancreatic duct, and pancreas stent placement were not independently associated with the amount of sedation in our study. On one hand this may reflect the improvement of endoscopic interventions and techniques in high-volume ERC centers which lead to shorter intervention time and reduced need for sedation. On the other hand demographic characteristics have a stronger influence on the need for sedation compared to short endoscopic interventions which is reflected by the inclusion of demographic parameters in the multivariable analysis. Naturally, an ERC with papillotomy and stent placement will lead to a longer intervention time and therefore to higher doses of sedation compared to an ERC without intervention. However, demographic parameters predominate the required amount of sedation. As shown by other studies increase in age and BMI and male gender affect the amount of sedation for endoscopic interventions [9,[11][12][13]. For example, male patients required 42 mg more propofol than female patients in our model. Another important factor is the intervention time. Extension of the intervention time by one minute led to an increase in the propofol consumption of 6.5 mg. These findings are of interest and importance for an adequate dosage and adaption of the sedative regimen.
Interestingly, we could show that laboratory parameters (bilirubin and platelet count) may serve as surrogate parameters which indicate the amount of the required sedation. Normal or mildly elevated bilirubin levels associated with an undilated common bile duct are known risk factors for post-ERC pancreatitis as the examination and cannulation may be more challenging [14,15]. Comparably, we could show that higher bilirubin levels lead to lower application of sedative agents in our model. In contrast higher platelet count leads to higher doses of sedation in our model. In our centre mainly patients with advanced liver disease are treated and a low platelet count is a general observation. Lower platelet count may serve as surrogate parameter of disease severity 6 International Journal of Hepatology which could lead to lower amounts of sedation in these patients due to the reduced hepatic metabolism. However, these assumptions are clearly beyond the scope of our study and have to be addressed in future studies.
The adverse event rate (cardiorespiratory events and immediate endoscopic adverse events) in our study was low and comparable to other studies [10,16]. Nevertheless, optimization of sedative strategies to further decrease the sedation-associated adverse events for ERC remains a prime goal.
Our study is of explorative character and has some limitations. First of all the retrospective character of the study is a limitation. Our study did not analyze the postinterventional recovery time as it was not routinely documented. Moreover, the study design does not allow excluding potential confounders like concomitant use of nonsteroidal antiinflammatory drugs, selective serotonin reuptake inhibitors, or alcohol consumption of the patients which may have an impact on the required dose of sedation for ERC. However, we thoroughly analyzed our endoscopic and patient data base to reduce potential bias. We included repetitive examinations in our analyses but also analyzed the data with restriction to the first presentation at our endoscopic unit. The results were equal in both cases.
In summary, we could show that demographic characteristics and endoscopic interventions have a distinct influence on the amount of sedation required for therapeutic ERC. Although the sedation-associated adverse event rate is low optimization of sedative regimens is a prime goal to further reduce adverse events of therapeutic ERC.

Data Availability
Data were extracted from the clinical and endoscopy data base. Rare data can be supplied. The data used to support the findings of this study are available from the corresponding author upon request.

Ethical Approval
The study was approved by the local institutional Ethics Committee (Ethics Committee of Hannover Medical School).