Risk factors for suspected infectious fever after diagnostic upper endoscopy: propofol overdose in a retrospective study from Korea
Article information
Abstract
Background/Aims
Propofol is widely used for sedation during upper endoscopy; however, oversedation may increase the risk of complications, including aspiration pneumonia. This study aimed to determine whether propofol overdose is a risk factor for suspected postendoscopic infectious fever (SPIF).
Methods
We retrospectively analyzed 1,474 in-patients who underwent upper endoscopy at Inha University Hospital between October 2021 and October 2022. After excluding 400 patients who received interventional procedures, SPIF was defined as new-onset fever (≥38 °C) within three days post-endoscopy with at least one of the following: intravenous antibiotic initiation, leukocytosis (white blood cell≥10,000/mm³), positive blood or sputum cultures, or pneumonia on chest X-ray. The risk factors assessed included comorbidities, sedative type and dosage, procedure duration, and endoscopist’s experience.
Results
New-onset fever occurred in 40 patients (3.7%), with 20 (1.9%) meeting SPIF criteria. Multivariable analysis showed propofol overdose (>2.2 mg/kg) significantly increased risks of new-onset fever (odds ratio [OR], 1.03; 95% confidence interval [CI], 1.00–1.06; p=0.023) and SPIF (OR, 1.04; 95% CI, 1.00–1.04; p=0.034). Dementia (OR, 1.11, p<0.001) and prolonged procedure (>5 minutes; OR, 1.01; p=0.012) were also risk factors for SPIF.
Conclusions
Propofol overdose significantly increases the risk of SPIF. Careful sedation management is essential, especially for patients with dementia or prolonged procedures.
INTRODUCTION
Propofol, a short-acting sedative, is commonly used in gastrointestinal endoscopy due to its rapid recovery profile. However, its use is not without risk, as elevated doses of propofol have been linked to respiratory depression, hypotension, and other adverse outcomes.1,2 Despite guidelines recommending low doses for sedation during endoscopy, propofol use has been increasing, with a significant rise in its administration in Korea from 29.1% in 2014 to 36% in 2019 for upper endoscopy procedures.3,4 This rise in usage underscores the importance of understanding and mitigating the risks associated with propofol, particularly in terms of overdose.
Furthermore, the median dose ranges between 70 and 116 mg per upper endoscopy procedure, with no clear studies exploring the effects of overdose on patient outcomes.5 Given the growing reliance on propofol in clinical settings, it is crucial to investigate the potential risks associated with higher doses, particularly concerning infections after upper endoscopy, including aspiration pneumonia. Previous studies have reported respiratory complications during upper endoscopy after propofol sedation.6-8
Post-infection within 7 days of upper endoscopy has been reported at a rate of 3 per 1,000 procedures.9 Hospitalization is a known risk factor for postendoscopic infection because a significant number of upper endoscopic procedures are performed in inpatient settings in Korea. Upper endoscopy has a higher post-infection rate than endoscopic retrograde cholangiopancreatography (ERCP).10 Endoscopic procedures, including ERCP, esophageal variceal band ligation, and endoscopic submucosal dissection (ESD), have been identified as causes of post-endoscopy fever or infection.11 However, limited data regarding post-endoscopy fever or infection following diagnostic upper endoscopy is available.
This study explored whether a propofol overdose during diagnostic upper endoscopy increases the risk of suspected infectious fever, a condition defined by the onset of fever post-procedure that may require intravenous antibiotics. The findings of this investigation provide valuable insights for optimizing sedation practices and minimizing the risks associated with endoscopic procedures.
METHODS
Study population
Data were collected from 1,474 in-patients who underwent diagnostic upper endoscopy with sedative agents at Inha University Hospital between October 2021 and October 2022. After applying specific exclusion criteria, 1074 patients were included in the final analysis. Excluded patients (400 in total) were those who had undergone interventional endoscopic procedures such as ESD, endoscopic mucosal resection (EMR), endoscopic ultrasonography (EUS), percutaneous endoscopic gastrostomy (PEG) insertion, and bleeding control procedures. Therefore, only those patients who underwent diagnostic upper endoscopy were included in this study. This study was reviewed and approved by the Ethics Committee of the Inha University Hospital (approval no. IUH 2024-10-062). The requirement for informed consent was waived due to the study’s retrospective nature.
Definitions used in data collection
Several key definitions were used throughout the study to properly classify and categorize the data. The definition of new-onset fever was a body temperature of ≥38 °C within three days following the upper endoscopy procedure.12,13 Suspected infectious fever was defined as persistent new-onset fever within three days post-procedure, accompanied by at least one of the following: initiation of intravenous antibiotics, leukocytosis (white blood cell≥10,000/mm³), a confirmed positive blood culture, a confirmed positive sputum culture, or radiographic evidence of pneumonia within 1 week after endoscopy.13-15 We excluded fever, leukocytosis, or administration of antibiotics before upper endoscopy.
Data collection and categorization
Data were collected during the study and categorized based on demographic characteristics, procedural details, and clinical outcomes. Demographic data, including patient age, sex, weight, and comorbidities, were also collected, including the presence of conditions such as diabetes with or without complications, liver disease (mild and severe), chronic heart failure, chronic kidney disease, cerebrovascular diseases, hemiplegia, cancer with or without metastasis, dementia, peptic ulcer, and chronic obstructive pulmonary disease. The age-adjusted Charlson comorbidity index (CCI), a standardized tool used to predict the overall health burden of patients by quantifying the impact of their comorbid conditions, was calculated for each patient to assess the severity of their comorbidities.16
Detailed data related to the endoscopic procedures were also collected, including whether an endoscopic biopsy was performed as a potential risk factor for fever or infection.17 Endoscopic duration was categorized as procedures lasting either ≤5 minutes or >5 minutes. The type of sedation used during the procedure was recorded, particularly the use of sedation agents such as propofol, midazolam, and etomidate.18 One critical variable was the total propofol dose, which was classified into two categories: ≤2.2 mg/kg and >2.2 mg/kg (considered overdose). These procedure-related factors were analyzed to assess their impact on infection-related complications. Propofol 2.2 mg/kg is the recommended maximum dose when slowly titrated; therefore, we divided patients into two groups based on whether they received ≤2.2 mg/kg or >2.2 mg/kg of propofol.18-20
The data also included the clinical outcomes related to fever and infection. The study recorded any new-onset fever occurring within 72 hours post-endoscopy, intravenous antibiotic administration, positive sputum or blood cultures, leukocytosis, or pneumonia on chest X-ray within one week—collectively indicating suspected infectious fever.21-24 Additionally, pneumonia, as diagnosed by radiological examinations, was noted as an important infection-related outcome.25
Statistical analysis
Various statistical methods were used to analyze the data. Baseline characteristics were compared between the patients who developed a fever and those who did not. Continuous variables were summarized as medians with interquartile ranges, depending on distributional assumptions, whereas categorical variables were presented as frequencies and percentages. For between-group comparisons, continuous variables were examined using the Mann-Whitney U-test, and categorical variables were compared using the chi-square test or Fisher exact test. To identify independent factors associated with the development of new-onset fever and suspected infectious fever, multivariable analysis was used to identify key factors associated with the development of new-onset fever and suspected infectious fever, including the role of propofol overdose, patient comorbidities, and the duration of the procedure. Additionally, odds ratios (ORs) were calculated to assess the strength of the association between propofol overdose (>2.2 mg/kg) and the risk of developing a suspected infectious fever. All statistical tests were two-tailed, and a p-value <0.05 was considered statistically significant. Data analyses were conducted using R software ver. 4.2 (R Foundation for Statistical Computing).
RESULTS
In total, 1,474 in-patients who underwent upper endoscopy under sedation were included in the initial cohort. After excluding 400 patients who underwent interventional endoscopic procedures such as ESD, EMR, EUS, PEG insertion, and bleeding control, 1,074 patients who underwent diagnostic upper endoscopy were selected for the final analysis. The study flow is illustrated in Figure 1, which shows the selection process and exclusion criteria. The primary objective was to evaluate the association between propofol overdose (>2.2 mg/kg) and the risk of developing postprocedural complications, specifically new-onset and suspected infectious fever.
Flow sheet of patients’ selection. EUS, endoscopic ultrasonography; ESD, endoscopic submucosal dissection; PEG, percutaneous endoscopic gastrostomy.
The baseline characteristics of the patients with and without new-onset fever after upper endoscopy are shown in Table 1. A total of 1,074 patients were analyzed, of whom 40 developed new-onset fever. There was no significant difference in age or sex distribution between patients with and without new-onset fever, and the use of propofol at doses greater than 2.2 mg/kg was significantly higher in patients with new-onset fever (40.0% vs. 23.7%, p=0.016). Comorbidities such as hemiplegia (p=0.029), localized solid tumors (p=0.032), and higher CCI scores (p=0.024) were significantly associated with new-onset fever. The endoscopist’s career level (fellow or staff member) and biopsy during endoscopy did not significantly affect the development of fever (p=0.778 and p=0.595, respectively).
Baseline characteristics of patients according to new-onset fever
The baseline characteristics of the patients with and without suspected infectious fever following diagnostic upper endoscopy are summarized in Table 2. A total of 1,074 patients were analyzed, with 11 patients developing suspected infectious fever, which was defined as fever requiring intravenous antibiotic treatment (55.0% vs. 3.4%, p<0.001) or pneumonia on radiologic examination (25.0% vs. 4.8%, p<0.001). Patients who developed suspected infectious fever were significantly more likely to have received a dose of propofol greater than 2.2 mg/kg (45.0% vs. 23.9%, p=0.03). Although the median age (69.5 vs 67.0 years, p=0.292) and sex distribution (male 60.0% vs. 56.3%, p>0.915) were similar across groups, patients in the suspected infectious fever group tended to have lower weight (53.5 vs. 59.9 kg, p=0.055). Additionally, the comorbidity burden was higher, with a CCI of 3.0 vs. 1.0 (p=0.041), and hemiplegia (35.0% vs. 12.1%, p=0.007) and dementia (15.0% vs. 1.8%, p=0.001) were more frequently observed in the suspected infectious fever group. Endoscopic biopsy did not significantly influence the incidence of suspected infectious fever (40.0% vs. 42.9%, p=0.595).
Baseline characteristics of patients according to suspected infectious fever
Multivariable analysis was performed to identify independent risk factors for new-onset fever and suspected infectious fever after diagnostic upper endoscopy in Table 3, using only variables from Tables 1 and 2 that had a p-value <0.05 in the univariable analysis. Patients with a history of hemiplegia had an approximately 2% increased risk (OR, 1.02; p=0.065), and the same 2% increase was observed in those with localized solid tumors (OR, 1.02; p=0.059), although neither reached the standard cut-off for statistical significance. Notably, patients receiving more than 2.2 mg/kg propofol had a significantly higher risk of new-onset fever (OR, 1.03; p=0.023). For suspected infectious fever, a propofol dose above 2.2 mg/kg also elevated the risk (OR, 1.04; 95% confidence interval [CI], 1.00–1.04, p=0.034), as did endoscopic duration >5 minutes (OR, 1.01; 95% CI, 1.00–1.04; p=0.012). Dementia emerged as the strongest predictor of suspected infectious fever (OR, 1.11; 95% CI, 1.05–1.18; p<0.001). Meanwhile, the CCI was insignificant in either model (p>0.05).
Risk factors for new-onset fever and suspected infectious fever after diagnostic upper endoscopy
DISCUSSION
This study investigated the risk factors for new-onset fever and suspected infectious fever after diagnostic upper endoscopy, with a particular focus on the effects of propofol overdose (>2.2 mg/kg). Propofol overdose was a significant independent risk factor for new-onset and suspected infectious fevers. Patients who received more than 2.2 mg/kg propofol had a notably higher incidence of both fever types. Additional factors such as a longer endoscopic duration (>5 minutes) and dementia were significantly associated with an increased risk of suspected infectious fever. Although comorbidities such as hemiplegia and localized solid tumors demonstrated trends toward increased fever risk, they did not reach statistical significance in the multivariate analyses. These findings underscore the importance of careful sedation management, particularly limiting the propofol dosage and procedure duration, to minimize fever-related complications after diagnostic endoscopy.
Postendoscopic fever is caused by tissue manipulation, an inflammatory response, and minor mucosal injury from endoscopic procedures.13,14,21 The association between propofol overdose (>2.2 mg/kg) and an increased risk of both new-onset fever and suspected infectious fever may be explained by several potential mechanisms. Propofol, as a sedative, is known to suppress the central respiratory drive, leading to hypoventilation, which in turn can cause transient hypoxia.26-28 Additionally, deeper sedation may delay recovery, resulting in prolonged immobilization and an increased risk of hospital-acquired infections, including pneumonia, which has been identified in several studies as a common complication in patients receiving higher doses of sedatives.29,30 Moreover, higher doses of propofol may lead to a prolonged duration of endoscopic procedures, as deeper sedation requires more careful postprocedural monitoring.31 This could expose patients to greater procedural stress, including mucosal trauma and longer periods of gastrointestinal tract manipulation, thereby increasing the likelihood of postendoscopic inflammatory responses, such as fever. A longer nasogastric tube stay is known to be a risk factor for fever during endoscopic procedures.13,32 Propofol-induced drug fever can occur rarely, so fever possibly occurs with high-dose propofol induction.33 Therefore, the findings of the current study are consistent with the adverse effects of propofol at higher dosages, particularly concerning respiratory depression, mucosal damage causing bacterial location, and drug side effects.34
This study builds upon previous research examining the risks associated with propofol use during endoscopic procedures. It provides new insights into the specific impact of propofol overdose on the development of both new-onset fever and suspected infectious fever. Previous studies have focused primarily on respiratory complications and recovery times linked to propofol, such as those that highlight the risk of hypoxemia and respiratory depression.1,6 This study is one of the few to specifically investigate the relationship between propofol overdose and postprocedural infectious outcomes. By including a large sample of 1,074 patients and using multivariable analysis to adjust for confounding factors such as comorbidities and procedure duration, this study offers robust evidence that propofol overdose is an independent risk factor for fever and suspicion of infection, extending the understanding of propofol's adverse effects beyond respiratory issues. Moreover, including comorbid conditions such as hemiplegia as risk factors adds depth to this study, setting it apart from other studies that have not fully considered the interaction between patient health status and sedation-related complications. The relative strength of this study lies in its comprehensive approach, which accounts for both sedation dosage and patient-specific variables, thereby providing more actionable insights for clinical practice, especially in terms of sedation management and risk stratification.
This study has several limitations. As a retrospective observational study, the establishment of a cause-and-effect relationship between propofol overdose and postprocedural fever may be subject to selection bias or inaccuracies in medical records. The single-center nature of this study limits the generalizability of the findings, as variations in sedation practices across different institutions may have influenced the results. Additionally, the lack of a long-term follow-up prevents assessing delayed infections or other complications that may arise after the procedure. Although key confounders such as comorbidities and procedure duration were accounted for, unmeasured factors such as variations in endoscopic technique or patient-specific factors could have affected the outcomes. The exclusion of interventional procedures means that the findings may not apply to patients undergoing more complex endoscopic interventions, where sedation requirements and risks may differ. This study did not include recently introduced sedatives. Recent Korean research underscores the evolving sedation practices and the emergence of novel sedatives for gastrointestinal endoscopy, highlighting the importance of vigilant propofol dosing.35
Nevertheless, this study demonstrates that propofol overdose is a significant risk factor for new-onset fever and suspected infectious fever after diagnostic upper endoscopy. Patients with certain comorbidities also have a higher risk of developing infectious fever. Our study suggests that longer procedures may contribute to postprocedural complications. These findings highlight the importance of careful management of sedation, particularly in vulnerable populations, to minimize adverse outcomes. Further research should focus on prospective and multicenter trials to establish a dose-response relationship for propofol, investigate alternative sedative agents, and explore comorbidity-specific sedation protocols.
Notes
Conflicts of Interest
The authors have no potential conflicts of interest.
Funding
This study was supported by grants from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare, Republic of Korea (grant no. RS-2024-00410513).
Acknowledgments
The authors are grateful to all the patients for whom the data were used in this study. Funding agencies played no role in the study design.
Author Contributions
Conceptualization: SL, JHL; Data curation: JS, BC, JTH, KSK; Formal analysis: JH. Funding acquisition: JHL; Investigation: SL, JS, BC, JTH, KSK; Methodology: JHL; Project administration: JHL; Resources: JS, BC, JTH, KSK; Software: JHL; Supervision: JHL; Validation: JHL; Visualization: JHL; Writing–original draft: SL, JHL; Writing–review & editing: all authors.