Effect of Duration of Pre-Procedure Fasting on Clinical Outcomes in Intensive Care Patients Undergoing Percutaneous Tracheostomy

Introduction

Tracheostomy is a frequently performed procedure in ICU patients requiring prolonged mechanical ventilation, particularly when intubation exceeds 7–10 days. Compared with the surgical method, percutaneous tracheostomy is less invasive, can be performed at the bedside, and is associated with lower complication rates [1–4]. One of the preparatory steps before tracheostomy is the discontinuation of enteral nutrition. Fasting has traditionally been used to reduce the risk of aspiration. The American Society of Anaesthesiologists (ASA) recommends a minimum fasting period of 2 hours for liquids and 6 hours for light solids before elective interventions [5]. However, there is no consensus regarding these periods in critically ill patients [6]. In ICU patients, the duration of fasting varies depending on several clinical factors such as the patient’s overall condition, ventilatory support, type of procedure, and aspiration risk [7,8]. Accordingly, practices are highly heterogeneous: in some centers, feeding continues until immediately before the procedure, while in others, fasting periods of up to 34 hours are applied [6].

Prolonged interruption of enteral nutrition may impair nutritional status, prolong ventilatory support, and increase morbidity. Therefore, clarifying the optimal fasting duration before tracheostomy is of critical clinical importance. Although some guidelines are available, there is insufficient evidence on whether different fasting durations before tracheostomy significantly affect patient outcomes such as pneumonia incidence, respiratory parameters, or the time required to resume feeding [6]. A recent study by Varghese et al highlighted significant variability in fasting practices but did not provide outcome-based data, underscoring the need for further investigation [6].

The primary aim of this study was to evaluate the effect of pre-procedural fasting duration on the development of aspiration pneumonia in critically ill patients undergoing percutaneous tracheostomy. Secondary aims include the assessment of postoperative respiratory parameters, time to achieve caloric targets, complication rates, length of stay, and mortality. Our hypothesis is that prolonged fasting does not reduce the risk of aspiration pneumonia, but delays the resumption of feeding and the achievement of caloric goals, thereby supporting the use of shorter and individualized fasting strategies in ICU practice.

Material and Methods

STUDY DESIGN AND ETHICAL APPROVAL:

This study was planned as a single-center retrospective observational study. Ethical approval was obtained from the University of Health Sciences Bursa Yüksek Ihtisas Training and Research Hospital Clinical Research Ethics Committee (Approval No: 2011-KAEK-25 2022/09–16, dated 07.09.2022). All procedures were conducted in accordance with the ethical principles of the Declaration of Helsinki. Owing to the retrospective design, individual patient consent was not applicable. To ensure confidentiality, all patient identifiers were removed during data extraction, and analyses were conducted on anonymized datasets. Patient privacy was strictly protected in accordance with institutional and national regulations.

STUDY POPULATION:

This retrospective study initially evaluated adult patients aged ≥18 years who underwent bedside percutaneous tracheostomy and received enteral nutrition support in the ICU of our hospital between January 1, 2017 and August 8, 2022.

Patients who had tracheostomy at the time of admission to the ICU, those who underwent surgical tracheostomy, those who had gastrostomy, those with inaccessibility of patient records, those who developed complications (such as major bleeding, tracheoesophageal fistula, pneumothorax, subcutaneous emphysema, respiratory complications like air leakage, or haemodynamic instability) that could prevent enteral nutrition within the first hour after the procedure, and those who were followed up with a diagnosis of COVID-19 were excluded from the study. All consecutive patients meeting the inclusion criteria during the study period were included to reduce selection bias. A total of 222 patients who met the specified criteria were analyzed. During the study period, a total of 274 patients were screened. After applying the exclusion criteria, 52 patients were excluded (details are provided in the flow diagram), and 222 patients were included in the final analysis. To minimize potential confounding, all consecutive patients meeting the inclusion criteria were enrolled, and baseline characteristics were compared across groups to ensure methodological consistency.

DATA COLLECTION:

Within the scope of clinical data, reasons for hospital and ICU admission, Acute Physiology and Chronic Health Evaluation-II (APACHE-II) score, Nutritional Risk Screening 2002 (NRS-2002) score, total duration of ICU and hospital stay, and duration of ICU stay after tracheostomy were analyzed. The APACHE-II score was calculated as described by Knaus et al [9], and the NRS-2002 scoring was performed according to Kondrup et al [10]. Mechanical ventilation (MV) modes – including synchronized intermittent mandatory ventilation (SIMV), pressure-controlled ventilation (PCV), continuous positive airway pressure with pressure support (CPAP-PS), adaptive support ventilation (ASV), bilevel positive airway pressure (bilevel), and T-tube spontaneous breathing – along with changes in ventilation mode, duration of mechanical ventilation, aspiration pneumonia development, complications, and mortality rates were analyzed.

Aspiration pneumonia was diagnosed based on the combined evaluation of clinical, radiological, and microbiological findings, including new pulmonary infiltrates, increased secretions, fever, leukocytosis, impaired oxygenation, and the need for changes in MV settings [11,12]. These parameters are consistent with widely accepted criteria for the diagnosis of pneumonia in critically ill patients and with the clinical features described for aspiration pneumonia [12]. All diagnoses were made by experienced intensive care physicians using this integrated approach. In cases of uncertainty, diagnoses were confirmed by consensus among at least 2 physicians to minimize interobserver variability.

Within the scope of respiratory evaluation, fractionated inspired oxygen (FiO₂) and peripheral oxygen saturation (SpO₂) levels, and MV parameters measured before tracheostomy and at 48 hours following the procedure, were compared on a group basis. Within the scope of nutritional parameters, the daily energy requirement calculated using the Harris-Benedict equation, the status and duration of discontinuation of enteral nutrition before tracheostomy, and the time to reach the target calories were analyzed.

All data were retrospectively extracted from patient files and the hospital information management system using a standardized data collection form. The hospital information management system included ICU daily follow-up charts, nursing observation records, medication and nutrition charts, and the electronic laboratory information system, which together ensured completeness and accuracy of the clinical data. Data were independently collected by 2 trained investigators, cross-checked by a third investigator, and verified by random audits to ensure accuracy and consistency. Records with missing data for key variables were excluded from relevant analyses.

The primary outcome of this study was the incidence of aspiration pneumonia following percutaneous tracheostomy, defined by a combination of clinical, radiological, and microbiological criteria. Secondary outcomes included respiratory parameters (FiO₂, SpO₂, and MV settings before tracheostomy and 48 hours after the procedure), nutritional outcomes (duration of enteral feeding interruption, time to resume feeding, and time to reach target caloric intake), and clinical outcomes (complication rates, ICU and hospital length of stay, and mortality).

ENTERAL NUTRITION PROTOCOL:

Enteral nutrition was administered via a nasogastric tube in the ICU according to patient tolerance. During the feeding process, all patients were kept in a head-up position of at least 30 degrees to reduce the risk of aspiration. Various standard and commercially available enteral nutrition formulas, including protein-enriched and disease-specific products, were used based on the clinical condition and tolerance of each patient, in line with institutional nutrition practices. All formulas complied with international standards for enteral nutrition.

Patients were classified into 4 groups (<1 hour, 1–4 hours, 5–12 hours, and >12 hours) according to the duration of enteral nutrition cessation before tracheostomy. Grouping was determined based on the fasting durations recorded in patient files, reflecting routine clinical practice rather than any predefined protocol or random assignment. This grouping scheme also considered the wide time range (0–34 hours) reported in the literature regarding the interruption of enteral nutrition [6] and the fasting durations commonly encountered in clinical practice. In routine practice, the resumption of enteral feeding after tracheostomy was performed via an infusion pump following confirmation of nasogastric tube placement, in order to reduce the risk of aspiration [13].

TRACHEOSTOMY PROCEDURE:

Percutaneous tracheostomy was performed at the bedside in the ICU using the Griggs technique under bronchoscopic guidance [14]. Patients were positioned in slight neck extension, and the tracheostomy site was infiltrated with 5–10 ml of 2% lidocaine for local anaesthesia. Under sterile conditions, a skin incision was made at the level of the second or third tracheal ring. The trachea was punctured with a needle, and a guidewire was advanced into the tracheal lumen. A dilating forceps was then introduced over the guidewire and gently opened to create a stoma. Finally, the tracheostomy tube was inserted over the guidewire and secured. Bronchoscopy was used to confirm correct placement. FiO2 was increased to 100% during the procedure, and ventilatory settings were adjusted to maintain adequate oxygenation. The mean duration of the procedure was approximately 10–15 minutes.

STATISTICAL ANALYSIS:

All data were analyzed using IBM SPSS Statistics for Windows, Version 28.0 (IBM Corp., Armonk, NY, USA). A 95% confidence level was set and P<0.05 was considered statistically significant. Quantitative variables were summarised as mean±standard deviation or median (minimum-maximum), as appropriate; qualitative variables were reported as frequency (n) and percentage (%). Distributional assumptions for quantitative variables were assessed with the Shapiro-Wilk test; non-parametric tests were used when normality was not met.

For comparisons across the 4 fasting-duration groups, Pearson’s chi-square test was used for categorical variables (Fisher’s exact test when >20% of expected cell counts were <5). For continuous variables, test selection depended on distribution: one-way analysis of variance (ANOVA) for normally distributed data and the Kruskal–Wallis H test for non-normal data. We also compared transitions from SIMV/PCV before tracheostomy to CPAP-PS/T-tube spontaneous breathing after tracheostomy across fasting-time groups using chi-square tests.

Missing data were handled using a complete-case approach: cases with missing values in key variables were excluded from the corresponding analysis. Effect-size estimates (eg, Cohen’s d for key continuous outcomes) and their 95% confidence intervals were reported to quantify the magnitude of between-group differences.

Results

DEMOGRAPHIC AND BASELINE CHARACTERISTICS:

Demographic and baseline clinical characteristics by fasting duration are presented in Table 1. The mean age was comparable across the 4 groups (P=0.940). Male predominance was observed in all groups, and gender distribution did not differ significantly (P=0.375). Similarly, no significant differences were identified in APACHE-II or NRS 2002 scores among the groups (P=0.866 and p=0.733, respectively).

Regarding clinical diagnoses, cerebrovascular disease was the most frequent condition, accounting for 42.8% of the overall cohort, with a similar distribution across the groups (P=0.207). Pulmonary diseases were present in 24.4% of patients, also with no significant intergroup differences (P=0.593). Other diagnoses – including multitrauma (6.1%), malignancy (9.3%), chronic renal failure (2.2%), neuromuscular disorders (2.1%), and cardiac diseases (13.1%) – were evenly distributed, with no statistically significant differences between groups (P>0.05).

RESPIRATORY PARAMETERS BEFORE AND AFTER TRACHEOSTOMY:

Tracheostomy-related respiratory parameters are shown in Table 2. The duration of stay until the tracheostomy procedure was similar across the groups (P=0.419). Pre-tracheostomy MV-FiO2 values ranged from 39.8% to 47.7% and did not differ significantly between groups (p=0.142). Pre-procedural SpO2 values were also similar across groups (P=0.278). During the first 48 hours after tracheostomy, MV-FiO2 values differed significantly among the fasting groups (p=0.001; Cohen’s d=0.81), with higher FiO2 levels observed in patients with longer fasting durations. In contrast, post-tracheostomy SpO2 values remained comparable between the groups (P=0.195).

NUTRITIONAL OUTCOMES:

Nutritional parameters related to tracheostomy are presented in Table 2. The time to reach the target calorie intake differed significantly among the fasting groups (P<0.001; Cohen’s d=1.05, 95% CI: 0.6–1.5). Patients in the longer fasting groups (Group 3 and Group 4) reached their caloric targets later compared with those in the shortest fasting group (Group 1).

CLINICAL OUTCOMES, LENGTH OF STAY, AND COMPLICATIONS:

When the groups were compared in terms of ICU and hospital length of stay, no statistically significant difference was found (ICU stay P=0.077; hospital stay P=0.302) (Table 3). Complications were observed at similar rates among the groups; only one patient in Group 3 developed pneumothorax (P=0.291). The incidence of aspiration pneumonia, as the primary outcome, did not differ significantly between the groups (P=0.421). Mortality rates were also similar across the groups (P=0.445).

The distribution of MV modes before and 48 hours after tracheostomy was compared according to fasting duration, and no statistically significant differences were observed between the groups in either period (all P>0.05) (Table 4).

The rates of transition from SIMV and PCV modes to CPAP-PS or T-tube spontaneous breathing after tracheostomy are shown in Figure 2. No statistically significant differences were found between the fasting duration groups for either mode (P=0.274 for SIMV; P=0.596 for PCV), and pairwise comparisons showed no significant differences between any group pairs (all P>0.05).

Discussion

LIMITATIONS:

The retrospective and single-center design of this study carries inherent risks of missing data, variability in clinical documentation, and potential selection bias. Although all consecutive patients were included to reduce bias, decisions regarding fasting duration may have partly reflected individual clinician preferences and patient-specific factors that could not be fully controlled. Nevertheless, the similar baseline characteristics among the groups may mitigate the impact of such bias.

In addition, the limited patient representation in the longest fasting category may have reduced the statistical power to detect significant differences for some outcomes. The predominance of patients with cerebrovascular and pulmonary diseases, as well as the generally elderly nature of the study population, also limits the generalisability of the findings to younger or non-neurological ICU populations. Finally, more detailed clinical parameters – such as measurements of muscle mass, assessments of gastrointestinal tolerance, or comprehensive microbiological data – could not be evaluated due to the retrospective nature of the study, preventing a broader assessment of the clinical effects of fasting duration.

As with any retrospective analysis, missing data may have introduced bias. However, missingness in primary outcomes was limited and non-differential across groups, and a complete-case approach was applied. While no additional sensitivity analyses were performed, the low and non-differential rate of missingness limits the likelihood that this issue substantially impacted the study findings.

Conclusions

This study found that prolonged pre-tracheostomy fasting did not have a meaningful effect on aspiration pneumonia, length of stay, or mortality. In contrast, longer fasting durations were associated with delays in achieving nutritional targets and higher postoperative oxygen requirements, whereas patients who fasted for less than 4 hours had markedly faster nutritional recovery and lower ventilatory support needs.

These findings indicate that the traditional practice of pre-tracheostomy fasting – historically maintained due to concerns regarding aspiration pneumonia and related complications – should be reconsidered. Moreover, this study addresses an important gap in the literature, provides a basis for re-evaluating existing clinical protocols, and offers a guiding framework for future large-scale, multicenter investigations. Such studies will help validate these results and support their stronger integration into clinical practice.