Does previous COVID-19 severity affect propofol sedation safety during ERCP?: A prospective observational study

Abstract

Endoscopic retrograde cholangiopancreatography (ERCP) requires deep sedation, typically with propofol, because of its rapid onset and recovery time. However, propofol can cause respiratory and hemodynamic instability. Because Coronavirus Disease 2019 (COVID-19) can induce long-term pulmonary and cardiovascular dysfunction, previous COVID-19 infection may influence sedation safety. This study aimed to evaluate the effects of COVID-19 infection severity on respiratory, hemodynamic, and recovery parameters during ERCP with propofol sedation. This prospective observational study included adult patients who underwent ERCP using propofol. Patients were divided into those without a COVID-19 history (COVID−) and those with a polymerase chain reaction-confirmed prior infection (COVID+), stratified by severity (mild, moderate, or severe). The primary outcome was hypoxemia (peripheral oxygen saturation < 90%). Secondary outcomes included apnea, airway interventions, hemodynamic instability, vasopressor use, recovery time, and patient complications. Multivariate logistic regression was used to adjust for confounders, including the American Society of Anesthesiologists physical status. A total of 326 patients were analyzed (163 COVID− and 163 COVID+ patients). Patients in the COVID+ group were classified as mild (n = 105), moderate (n = 46), or severe (n = 12). No significant differences were observed between the COVID and mild COVID+ groups in terms of hypoxemia or complications. Conversely, patients with a history of moderate or severe COVID-19 infection exhibited significantly higher rates of hypoxemia, apnea, airway interventions, vasopressor use, prolonged recovery, and postoperative oxygen requirements. Multivariate analysis identified a history of moderate-to-severe COVID-19 as an independent risk factor for hypoxemia. In conclusion, patients with a history of moderate or severe COVID-19 infection may be at increased risk for respiratory complications during propofol sedation for ERCP. Preoperative risk stratification should prioritize the severity of previous COVID-19 infection rather than just the history of infection.

1. Introduction

Endoscopic retrograde cholangiopancreatography (ERCP) is a widely used advanced interventional endoscopic technique for the diagnosis and treatment of biliary and pancreatic disorders. Because of patient positioning and pain, stimulating procedural steps such as sphincterotomy, stone extraction, and stent placement, ERCP frequently requires deep sedation.^[1] Propofol is one of the most commonly preferred sedative agents for ERCP because of its rapid onset of action and short duration, allowing for faster and more predictable recovery than other sedatives. However, the potential adverse effects of propofol, including respiratory depression, apnea, and hypoxemia, make safe sedation management during ERCP critically important.^[2]

Following the Coronavirus Disease 2019 (COVID-19) pandemic, it has become evident that Severe Acute Respiratory Syndrome Coronavirus 2 infection can affect multiple organ systems not only during the acute phase but also in the long term. Within the spectrum of post-acute COVID syndrome, reduced pulmonary capacity, diffusion impairment, restrictive ventilatory patterns, cardiac dysfunction, autonomic nervous system abnormalities, exercise intolerance, and persistent fatigue have been reported to persist for months.^[3] In major elective surgeries, patients undergoing procedures within 0 to 4 weeks of infection exhibit a 3- to 6-fold increased risk of pneumonia and respiratory failure, and elevated pulmonary morbidity may persist for up to 8 weeks after recovery. While high sedative and analgesic requirements have been reported during the acute severe phase of COVID-19 due to increased tolerance and heightened respiratory drive, the primary driver of risk in the post-recovery period appears to be residual lung injury and reduced ventilatory reserve rather than a specific post-COVID sedative hypersensitivity.^[4,5] Pulmonary fibrotic changes, decreased ventilatory reserve, and dysautonomia affecting cardiovascular stability are particularly common in patients who experienced moderate-to-severe COVID-19.^[6] These pathophysiological alterations may increase the risk of sedation-related complications, including hypoventilation, apnea, hypoxemia, and hemodynamic instability.^[7]

Despite these concerns, data evaluating the impact of prior COVID-19 infection on endoscopic procedures requiring deep sedation, such as ERCP, remain limited. Most perioperative studies have evaluated COVID-19 history as a binary variable; however, the rationale for specifically examining disease severity stems from the observation that mild infections often result in complete functional recovery, whereas moderate-to-severe courses are more frequently associated with lasting structural lung damage and vascular changes. Furthermore, whether the severity of prior COVID-19 infection differentially affects sedation-related risk in the context of propofol-based endoscopic interventions remains unclear.

This study aimed to compare sedation-related respiratory complications during ERCP between patients with a history of COVID-19 and those without prior infection. We also evaluated whether the clinical severity of the initial infection influences these outcomes. We posited that patients with a history of moderate-to-severe COVID-19 may exhibit an elevated risk of respiratory complications during sedation compared with those with mild infection or no prior COVID-19, potentially attributable to diminished pulmonary and physiological reserve.

2. Materials and methods

2.1. Study design

This prospective observational cohort study was conducted in collaboration with the Gastroenterology Endoscopy Unit and the Department of Anesthesiology at Ankara City Hospital. The study was approved by the Ankara City Hospital Clinical Research Ethics Committee (date: March 30, 2022; Decision No: E2-22-1600) and was performed in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrollment. As this was an observational study without any interventional component, registration in a public clinical trial registry was not required. The study was conducted and reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines (see STROBE Checklist).

2.2. Study population

Adult patients scheduled for elective or semi-urgent endoscopic retrograde cholangiopancreatography (ERCP) under propofol-based sedation at the ERCP unit of Ankara City Hospital were included. The inclusion criteria were as follows: age ≥ 18 years, planned elective or semi-urgent ERCP, sedation primarily administered with propofol, and willingness to participate in the study. The exclusion criteria were as follows: active COVID-19 infection confirmed by positive polymerase chain reaction (PCR) or antigen testing within the previous 14 days, American Society of Anesthesiologists (ASA) physical status IV to V, hemodynamic instability, ERCP performed under general anesthesia with endotracheal intubation, severe pulmonary disease unrelated to COVID-19 (GOLD stage IV chronic obstructive pulmonary disease or severe restrictive lung disease), and pregnancy.

2.3. Group definitions

Patients were categorized according to their COVID-19 history. COVID-19 history and clinical severity were verified through a comprehensive review of institutional electronic medical records and the national health information system (e-Nabiz) to ensure data accuracy. Patients with a documented history of COVID-19 confirmed by PCR or antigen testing in hospital records or the national health database were classified into the COVID-positive (COVID+) group. These patients were further stratified according to disease severity as follows: Mild COVID-19, managed on an outpatient basis without oxygen requirement; Moderate COVID-19, required hospitalization and supplemental oxygen; Severe COVID-19, required intensive care unit admission, high-flow oxygen therapy, noninvasive ventilation, or invasive mechanical ventilation. Patients with no history of COVID-19 and no documented positive test results were classified as COVID-negative (COVID−).

2.4. Sedation protocol

Standard monitoring, including electrocardiography, noninvasive arterial blood pressure measurement, peripheral oxygen saturation (SpO₂), and respiratory rate, was performed in all patients. Supplemental oxygen was administered via a nasal cannula at 2 to 4 L/min, when necessary. Sedation was achieved using propofol in all patients. Lidocaine (1 mg/kg) was intravenously administered before propofol induction. The initial propofol bolus dose ranged from 0.5 to 1 mg/kg, followed by continuous infusion at 50 to 150 µg/kg/minute.^[2] Additional analgesia with fentanyl (25–50 µg) was permitted at the discretion of the endoscopist to optimize procedural conditions. Usage of adjunctive opioids was documented for each patient to assess its potential influence on respiratory outcomes. Sedation depth was assessed using the Ramsay Sedation Scale. The cumulative duration of bispectral index (BIS) values <40 was recorded in patients who were monitored using the BIS.^[8]

2.5. Outcomes and measurements

The primary outcome was the incidence of intraprocedural hypoxemia, defined as a sustained SpO₂ < 90% lasting for at least 10 to 15 seconds.^[9,10] Severe hypoxemia was defined as a sustained SpO₂ of <85%.^[7] The secondary outcomes were respiratory parameters, including the occurrence of apnea (cessation of chest wall movement requiring intervention), airway interventions (jaw thrust, head repositioning, nasal or oral airway insertion, mask ventilation, rescue with a supraglottic airway). Airway interventions were performed immediately upon clinical signs of hypoventilation or apnea, often preemptively to prevent frank desaturation; in cases where desaturation occurred despite early signs, interventions were recorded as rescue measures following the event. Oxygen requirements (baseline and mean intraprocedural flow rates), lowest recorded SpO₂ value, and suspected aspiration.^[11,12]

Hemodynamic parameters were recorded as follows: hypotension, systolic blood pressure < 90 mm Hg or mean arterial pressure < 65 mm Hg.^[10] Bradycardia (heart rate < 50 beats/min), and tachycardia (heart rate > 100 beats/min).^[13,14] New-onset arrhythmias (atrial fibrillation, supraventricular tachycardia, increased premature ventricular contractions). Vasopressor use (ephedrine, phenylephrine, and norepinephrine). Sedation-related parameters included the total propofol dose (mg and mg/kg), mean infusion rate (mg/kg/h), type and dose of additional sedative or analgesic agents, and sedation depth (Ramsay score and BIS values).^[15–17]

2.6. Postprocedural assessment

Postprocedural recovery variables included eye-opening time, response to verbal commands, time to achieve an Aldrete score of ≥9, length of stay in the post-anesthesia care unit (PACU), persistence of postoperative oxygen requirement, and incidence and treatment of postoperative nausea and vomiting.^[18] Postoperative pulmonary complications (within 7 days) were defined as pneumonia, new or increased oxygen requirements, the need for noninvasive ventilation or reintubation, and radiological evidence of new pulmonary infiltrates, accompanied by respiratory deterioration. Postoperative cardiovascular events (within 30 days) included myocardial infarction, arrhythmia, venous thromboembolism (deep vein thrombosis or pulmonary embolism), hospital readmission, and 30-day all-cause mortality.

The variables recorded for adjustment in the multivariate analyses included demographic data (age, sex, body mass index [BMI]), comorbidities (hypertension, diabetes mellitus, coronary artery disease, chronic obstructive pulmonary disease, asthma, obstructive sleep apnea, renal insufficiency, liver disease), ASA physical status, smoking history (current, former, never), baseline laboratory values (hemoglobin, creatinine, C-reactive protein), and ERCP-related factors (procedure duration, diagnostic or therapeutic intervention, and patient positioning).

Additional variables recorded for COVID+ patients included the interval between COVID-19 infection and ERCP, COVID-19 severity, and the presence of persistent post-COVID symptoms.

2.7. Sample size calculation

The sample size estimation was based on previously reported hypoxemia (SpO₂ < 90%) rates during propofol-sedated ERCP. Previous studies have reported an incidence of hypoxemia of approximately 20% in control groups receiving propofol sedation.^[15] Assuming a clinically significant absolute increase of 14% in the COVID+ group (expected incidence: 34%) with a power of 80% (1–β) and a two-sided alpha level of 0.05, the minimum required sample size was calculated as 145 patients per group. Accounting for an estimated dropout rate of approximately 12%, 163 patients per group (326 patients in total) were included in the study.

2.8. Statistical analysis

All statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk). The normality of continuous variables was assessed using the Shapiro–Wilk test. Normally distributed variables are expressed as mean ± standard deviation, whereas non-normally distributed variables are presented as medians (minimum–maximum values). Categorical variables are reported as counts and percentages. Between-group comparisons were performed using the independent samples t-test or Mann–Whitney U test for continuous variables, and the chi-square test or Fisher exact test for categorical variables. The association between the recovery interval (time from COVID-19 diagnosis to ERCP) and the primary outcome was evaluated using Spearman correlation.

A multivariable logistic regression model was constructed to identify the independent predictors of hypoxemia. Covariates included COVID-19 status, age, BMI, ASA class, presence of chronic obstructive pulmonary disease or asthma, obstructive sleep apnea, procedure duration, and smoking history. Statistical significance was set at P < .05. There were no missing data for the variables included in the final analysis. To minimize selection and information bias, all consecutive eligible patients were included, and data collection was performed using standardized monitoring and predefined outcome definitions. The calibration and goodness-of-fit of the multivariable logistic regression model were assessed using the Hosmer–Lemeshow test.

3. Results

3.1. Study population and baseline characteristics

A total of 358 patients were assessed for eligibility. Thirty-two patients were excluded (21 did not meet inclusion criteria, 5 declined participation, and 6 for other reasons). Consequently, 326 patients were included in the final analysis (163 COVID− and 163 COVID+). The patient selection process is illustrated in Figure 1. The baseline demographic characteristics and clinical features of the study population are shown in Table 1. There were no statistically significant differences between the groups with respect to sex distribution, BMI, ASA physical status classification, or prevalence of comorbidities (all P > .05). The mean age was significantly higher in the COVID group compared with the control group (63.9 ± 9.4 vs 60.2 ± 9.3 years, P < .001). The procedure duration and total propofol consumption were comparable between the 2 groups. Fentanyl administration rates were found to be statistically similar between the COVID− and COVID+ groups (P = .704), and notably, no fentanyl was administered to patients in the moderate or severe COVID-19 subgroups. There were no significant differences between the COVID− and COVID+ groups regarding the total dose of fentanyl administered (2.15 ± 11.2 mcg vs 1.23 ± 7.8 mcg, respectively; P = .359; Table 1). Furthermore, subgroup analysis revealed that no fentanyl (0%) was used in the moderate or severe COVID-19 groups, indicating that the observed respiratory complications were not associated with opioid administration (Table 3).

Figure 1.

Patient selection flow diagram.

Table 1.

Demographic characteristics and baseline clinical data of the study groups.

Variables	Control group (COVID−) (n = 163)	Study group (COVID+) (n = 163)	P value
Age (yr), mean ± SD	60.2 ± 9.3	63.9 ± 9.4	<.001*
Sex (female), n (%)	83 (50.9)	85 (52.1)	.825
BMI (kg/m2), mean ± SD	27.0 ± 3.9	27.1 ± 3.7	.933
ASA physical status, n (%)	0.747
I	30 (18.4)	29 (17.8)
II	94 (57.7)	89 (54.6)
III	39 (23.9)	45 (27.6)
Comorbidities, n (%)
Hypertension	61 (37.4)	63 (38.7)	.909
Diabetes mellitus	34 (20.9)	36 (22.1)	.892
COPD	11 (6.7)	21 (12.9)	.094
Procedure time (min)	43.9 ± 14.5	44.3 ± 15.4	.81
Total propofol dose (mg)	259.2 ± 67.3	264.9 ± 60.3	.418
Total fentanyl dose (mcg)	2.15 ± 11.2	1.23 ± 7.8	.359

Values are presented as mean ± standard deviation (SD) or number (percentage).

ASA = American Society of Anesthesiologists, BMI = body mass index, COPD = chronic obstructive pulmonary disease.

^*Statistically significant difference (P < .05). Student t test was used for continuous variables and the chi-square test for categorical variables.

Table 3.

Subgroup analysis of respiratory complications according to COVID-19 severity.

Variables	Mild (n = 105)	Moderate (n = 46)	Severe (n = 12)	P value*
Hypoxemia, n (%)	24 (22.8)	22 (47.8)	10 (83.3)	<.001†
Lowest SpO2 (%), mean ± SD	91.8 ± 5.7	89.3 ± 6.1	85.3 ± 6.8	<.001†
Airway intervention, n (%)	29 (27.6)	25 (54.3)	10 (83.3)	<.001†
Apnea, n (%)	7 (6.7)	9 (19.6)	2 (16.7)	.054
Recovery time (min), mean ± SD	61.8 ± 19.2	74.3 ± 18.5	84.5 ± 14.1	<.001†
Fentanyl used, n (%)	4 (3.8%)	0 (0%)	0 (0%)	.322

Values are presented as numbers (percentages) or means ± SD. Mild: outpatient follow-up; Moderate: hospitalized (ward); Severe: intensive care unit admission.

^*P value for trend (linear-by-linear association for categorical variables, ANOVA trend for continuous variables).

^†Statistically significant trend.

3.2. Primary and secondary sedation outcomes

The primary and secondary endpoints related to sedation safety and respiratory complications are presented in Table 2. The incidence of hypoxemia (SpO₂ < 90%), which was the primary outcome, was 34.4% (n = 56) in the COVID group and 24.5% (n = 40) in the control group; however, this difference did not reach statistical significance (P = .068). Similarly, no significant differences were observed between the groups in terms of the lowest recorded SpO₂ values, frequency of apnea, or incidence of hypotension. In contrast, postoperative recovery time, defined as the length of stay in the PACU, was significantly longer in the COVID group than in the control group (67.0 ± 20.1 min vs 60.5 ± 20.9 min, P = .005).

Table 2.

Comparison of sedation-related adverse events and recovery outcomes.

Outcomes	Control group (n = 163)	COVID+ group (n = 163)	P value	Effect size OR/MD (95% CI)
Hypoxemia (SpO2 < 90%), n (%)	40 (24.5)	56 (34.4)	.068	1.61 (0.99–2.60)
Lowest SpO2 (%), mean ± SD	91.7 ± 5.9	90.6 ± 6.2	.097	−1.10 (−2.41–0.21)
Apnea, n (%)	25 (15.3)	18 (11.0)	.326	0.69 (0.36–1.31)
Airway intervention, n (%)	61 (37.4)	64 (39.3)	.819	1.08 (0.69–1.69)
Hypotension, n (%)	22 (13.5)	31 (19.0)	.23	1.51 (0.83–2.73)
Bradycardia, n (%)	12 (7.4)	15 (9.2)	.681	1.28 (0.58–2.82)
Recovery time (min), mean ± SD	60.5 ± 20.9	67.0 ± 20.1	.005*	6.50 (2.05–10.95)

Values are presented as numbers (percentages) or means ± SD. For categorical outcomes (hypoxemia, apnea, airway intervention, hypotension, bradycardia), the effect size is reported as odds ratio (95% CI). For continuous outcomes (lowest SpO₂, recovery time), the effect size is reported as mean difference (95% CI).

Definitions: Hypoxemia was defined as a sustained SpO₂ < 90% lasting for at least 15 s. Airway interventions were performed immediately upon clinical signs of hypoventilation or apnea, often preemptively to prevent frank desaturation. Recovery time was defined as the duration from the end of the procedure until the patient achieved an Aldrete score of ≥9.

CI = confidence interval, MD = mean difference, OR = odds ratio, PACU = post-anesthesia care unit, SpO₂ = peripheral oxygen saturation.

^*Statistically significant difference (P < .05).

3.3. Subgroup analysis by COVID-19 severity

Subgroup analyses were performed according to the severity of the prior COVID-19 infection (Table 3). Patients in the COVID group were classified as having mild (n = 105), moderate (n = 46), or severe disease (n = 12). A strong association was observed between increasing disease severity and the incidence of sedation-related complications (trend, P < .001). The rate of hypoxemia was 22.8% in patients with mild COVID-19, which was comparable to that in the control group (24.5%), whereas it increased to 47.8% in the moderate group and 83.3% in patients with severe disease who had a history of intensive care unit admissions. Similarly, increased COVID-19 severity was positively correlated with the need for airway intervention and prolonged recovery times. The mean PACU stay was 84.5 minutes in the severe group compared with 61.8 minutes in the mild group.

3.4. Predictors of hypoxemia and multivariate analysis

To identify independent risk factors for intraoperative hypoxemia (SpO₂ < 90%), multivariate logistic regression analysis including age, BMI, ASA physical status, and severity of prior COVID-19 infection was performed (Table 4). The analysis demonstrated that a history of moderate-to-severe COVID-19 was independently associated with a significantly increased risk of hypoxemia compared with the control group (odds ratio [OR]: 4.35; 95% confidence interval [CI]: 2.21–8.57; P < .001). In contrast, a history of mild COVID-19 was not associated with an increased risk of hypoxemia (OR: 0.89; P = .704). Other variables included in the model, such as age (P = .579), BMI (P = .739), and ASA physical status (P = .114), were not identified as independent predictors of hypoxemia in the study cohort. These findings suggest that the severity of prior COVID-19 infection, rather than demographic characteristics or baseline clinical status, is the primary determinant of respiratory complications during propofol-sedated ERCP.

Table 4.

Multivariate logistic regression analysis of risk factors for hypoxemia.

Variables	Odds ratio (OR)	95% confidence interval (CI)	P value
Age	0.99	(95% CI: 0.96–1.02)	.579
BMI	1.01	(95% CI: 0.94–1.08)	.739
ASA-II	0.55	(95% CI: 0.29–1.05)	.071
ASA-III	0.54	(95% CI: 0.25–1.16)	.114
Mild COVID-19	0.89	(95% CI: 0.50–1.60)	.704
Moderate/Severe COVID-19	4.36	(95% CI: 2.22–8.58)	<.001*

Note: The multivariate model includes all variables listed in table. The reference category for COVID-19 history is the “control group.” For ASA classification, “ASA I” served as the reference. Bold P-values indicate statistical significance at the P < .05 level. Analysis was performed using a logistic regression model to determine the independent predictors of hypoxemia (SpO2 < 90%). Note: Model fit was validated using the Hosmer–Lemeshow test (χ² = 9.041, P = .339).

ASA = American Society of Anesthesiologists Physical Status Classification, BMI = body mass index, CI = confidence interval, OR = odds ratio.

The multivariate logistic regression model demonstrated high statistical reliability, as confirmed by the Hosmer–Lemeshow goodness-of-fit test (χ² = 9.041, P = .339), indicating no significant difference between observed and predicted outcomes. Furthermore, the time interval between COVID-19 infection and the ERCP procedure (mean: 12.8 months) was analyzed within the COVID+ group. No significant correlation was found between the recovery interval and the occurrence of hypoxemia (r = −0.038, P = .62). When included in the regression analysis, the time since infection was not an independent predictor of respiratory complications (P = .281).

4. Discussion

This study is one of the first prospective investigations to evaluate the impact of a previous COVID-19 infection on sedation safety and recovery parameters in patients undergoing ERCP with propofol sedation. Our findings indicate that a history of COVID-19 infection itself does not appear to be an independent risk factor; rather, sedation-related risk is directly associated with the severity of the prior infection. Patients with mild (outpatient-managed) COVID-19 exhibited a safety profile comparable to that of the control group, whereas those with moderate-to-severe disease, particularly those with a history of intensive care unit admission, demonstrated significantly increased rates of hypoxemia, airway interventions, and prolonged recovery. However, it is important to highlight from the outset that the subgroup of patients with severe COVID-19 was relatively small (n = 12). While the results in this group are clinically striking, this limited sample size necessitates a cautious approach to the generalizability and statistical robustness of the findings for the most severe cases.

Previous studies have reported that persistent respiratory symptoms following COVID-19 occur in approximately 10 to 50% of patients.16 In our cohort, the overall incidence of hypoxemia in the COVID-19 group (34.4%) was numerically higher than that in the control group (24.5%), although this difference was not statistically significant (P = .068). This finding may be explained by the predominance of patients with mild disease in the COVID-19 group (64.4%). Consistently, subgroup analysis revealed that the incidence of hypoxemia in patients with mild COVID-19 (22.8%) was nearly identical to that in the control group (24.5%), suggesting that mild infection does not result in persistent impairment of pulmonary parenchymal function or diffusion capacity. In contrast, the markedly higher complication rates observed in patients with moderate and severe COVID-19 (47.8% and 83.3%, respectively) are in line with the previously described pathophysiological mechanisms.^[19]

Severe COVID-19 pneumonia causes fibrotic changes, reduced gas exchange surface area, and ventilation/perfusion (V/Q) mismatch. In such patients, a predisposition to silent hypoxemia and reduced functional residual capacity, when combined with the respiratory depressant effects of propofol, may lead to more frequent and profound desaturation episodes.^[20] Furthermore, diaphragmatic weakness and respiratory muscle fatigue have been reported to persist for several months after severe COVID-19. This may explain the increased incidence of apnea and the need for airway interventions (jaw thrust or bag-mask ventilation) observed in our study.^[21]

ERCP inherently presents significant challenges to sedation safety due to the prone or semi-prone positioning, shared airway conditions, and prolonged procedure duration.^[22] These factors further narrow the respiratory safety margin, particularly in patients with reduced pulmonary reserve. Our findings suggest that post-COVID patients with a history of moderate-to-severe disease may represent a clinically relevant higher-risk subgroup during ERCP, in whom the combination of procedure-related and patient-related risk factors may synergistically increase the likelihood of respiratory complications.

The phenomenon of “silent hypoxemia” is also of major clinical relevance in this population group. Post-COVID impairment of chemoreceptor responsiveness and attenuation of hypoxic pulmonary vasoconstriction may delay the clinical recognition of hypoxemia.^[23] Deep sedation may result in sudden and severe desaturation events despite standard monitoring. These observations support the use of advanced respiratory monitoring modalities, such as capnography, in patients with a history of moderate-to-severe COVID-19.^[24] It is also important to consider the potential influence of adjunctive opioids on respiratory outcomes. In our study, fentanyl was used very sparingly across all groups, following our institutional restrictive opioid protocol. Our data confirmed that fentanyl usage did not differ between groups (P = .359) and, notably, was entirely avoided in the moderate and severe COVID-19 subgroups. Therefore, the higher incidence of complications in these patients is attributable to their underlying pathophysiology rather than sedative variability.

Another noteworthy finding of our study was the significant prolongation of recovery time. Patients in the severe COVID-19 group had a mean PACU stay that was approximately 24 minutes longer than that of the control group (P < .001), despite similar total propofol doses across the groups. This delay can be explained by 2 mechanisms: First, residual pulmonary impairment following severe COVID-19 may limit respiratory effort in the PACU, resulting in lower scores on the respiratory component of the Aldrete score and prolonged recovery times.^[18] Second, severe COVID-19 has been associated with increased blood–brain barrier permeability and neuroinflammation, potentially enhancing central nervous system sensitivity to sedative agents and delaying emergence from anesthesia.^[25] In addition, COVID-19–related autonomic dysfunction (dysautonomia) may slow hemodynamic recovery and restoration of homeostasis, thereby prolonging PACU discharge.^[26]

In addition to respiratory parameters, patients with moderate-to-severe COVID-19 had a significantly higher requirement for vasopressor support. This finding is consistent with reports of autonomic dysfunction and reduced cardiovascular reserve in patients with long COVID-19.^[27] The vasodilatory effects of propofol may precipitate hypotensive episodes that cannot be adequately compensated in patients with impaired autonomic regulation.^[20]

Multivariable logistic regression analysis adjusted for age differences between groups confirmed that our findings were independent of age and comorbidities (ASA score). Moderate-to-severe COVID-19 history emerged as an independent risk factor for hypoxemia during ERCP (OR, 4.35; 95% CI: 2.21–8.57, P < .001), whereas age and BMI were not independently associated with hypoxemia. These results support the concept that the observed respiratory complications are primarily attributable to the chronic effects of prior infection on pulmonary parenchyma and airway dynamics. Consistent with our regression model, these results underscore that demographic variables were not as influential as the clinical course of the viral infection itself. In our study, the severity of the initial COVID-19 infection was a more significant predictor of intraoperative hypoxemia than the time elapsed since the infection. Our analysis showed that even after an average recovery period of 12.8 months, patients with a history of severe disease remained at higher risk, suggesting that the physiological impact of severe COVID-19 on respiratory reserve may persist long-term.

From a clinical perspective, our results advocate for tailored sedation strategies for patients with a history of moderate or severe COVID-19 who are undergoing ERCP. Enhanced respiratory monitoring, conservative titration of sedative agents, preparedness for early airway intervention, and consideration of alternative sedation techniques may improve the safety of the procedure. Importantly, our data indicate that routine escalation of care is not necessary for patients with a history of mild COVID-19, thereby avoiding the unnecessary use of healthcare resources.

This study has several limitations that should be acknowledged. First, it was a single-center study, which may affect the generalizability of our findings. Second, as previously mentioned, the sample size in the severe COVID-19 subgroup (n = 12) was relatively small, potentially limiting the statistical power of this specific sub-group comparison and potentially inflating effect estimates. Third, we did not apply formal statistical adjustments (such as Bonferroni correction) for multiple comparisons during subgroup analyses. This omission increases the risk of Type I error; therefore, these secondary results should be interpreted with caution as exploratory. Fourth, objective preoperative pulmonary function tests, diffusion capacity measurements, or capnography monitoring were not routinely performed; such data could have provided more detailed insight into patients’ respiratory reserves and early hypoventilation events. Fifth, COVID-19 severity classification was primarily based on clinical history and medical records rather than physiological metrics. Finally, although we monitored immediate recovery, long-term postoperative respiratory outcomes beyond the hospital stay were not evaluated in this study. The prospective design of the study and the use of predefined clinical outcomes strengthen the internal validity of the findings.

5. Conclusions

Our findings indicate that anesthesiologists should stratify patients in the preoperative assessment not merely as “COVID-positive” or “COVID-negative,” but according to the severity of the prior COVID-19 infection. While patients with a history of mild COVID-19 can be safely managed using standard sedation protocols, those with moderate and severe disease should be considered at high risk. In this subgroup, capnography monitoring, proactive airway management strategies, and cautious titration of sedative agents may enhance procedural safety and reduce sedation-related complications. Future multicenter studies with larger cohorts are warranted to further clarify the long-term impact of severe COVID-19 on sedation safety during endoscopic procedures.

Author contributions

Conceptualization: Mehmet Sahap, Muge Cakirca.

Data curation: Mehmet Sahap, Onder Aydemir, Nihal Gokbulut Ozaslan, Gokhan Erdem.

Funding acquisition: Mehmet Sahap.

Investigation: Mehmet Sahap, Ayse Lafci.

Methodology: Mehmet Sahap.

Project administration: Mehmet Sahap, Ayse Lafci.

Resources: Mehmet Sahap, Ayse Lafci, Nihal Gokbulut Ozaslan, Gokhan Erdem.

Software: Mehmet Sahap.

Supervision: Mehmet Sahap.

Validation: Mehmet Sahap.

Visualization: Mehmet Sahap.

Writing – original draft: Mehmet Sahap.

Writing – review & editing: Mehmet Sahap.

Formal analysis: Onder Aydemir.