Pulse Oximetry Desaturation in the Postoperative Recovery Room in Patients with Obesity and Diabetes Using GLP-1 Agonists: A Retrospective Observational Study

Abstract

Background and objectives: Glucagon-like peptide-1 agonists (GLP-1), obesity, and diabetes may delay gastric emptying in some settings. We tested three hypotheses. First, the duration of GLP1-hold is associated with a larger difference between pulse oximetry at room air immediately preoperatively and one hour postoperatively, just prior to discharge home (DSpO2), as a marker of clinically important atelectasis and/or micro-aspiration. Second, we tested the hypothesis that the duration of GLP1-hold affects fasting glucose. Third, we tested the hypothesis that DSpO2 is linked to morbid obesity.

Materials and methods: In this retrospective observational cohort study, we screened the cohort of all 1571 patients undergoing urologic ambulatory surgery in one stand-alone center between September 2023 and September 2024. The inclusion criteria were diabetics using GLP-1 within 30 days, creatinine below 2 mg/dL, and age between 18 and 80 years. The outcomes of fasting glucose and DSpO2 were compared between the groups holding GLP-1 longer and shorter than seven days. The outcome DSpO2 was also compared between the groups with and without morbid obesity, defined generally as a body mass index (BMI) of over 35 kg/m² in the presence of one major comorbidity. All of our study subjects had diabetes mellitus, a major comorbidity.

Results: Among all 107 subjects, 56% had laryngeal mask airways and 9% endotracheal tubes. GLP-1 was held 12 ± 8 days (mean ± SD). DSpO2 was 1.5% ± 1.9 with short GLP-1-hold versus 1.8% ± 2.0 with long GLP-1-hold (p=0.41). The mean glucose was 130 mg/dL ± 49 with short GLP1-hold versus 138 mg/dL ± 39 with long GLP-1-hold (p=0.69). DSpO2 was 2.29% ± 1.78 with morbid obesity (N=38) versus 1.43% ± 1.99 without (N=69, p=0.0254, unpaired two-tailed t-test). Patients with morbid obesity showed a trend toward twice the incidence (at 24%) for DSpO2 > 4% (chi-square, p=0.14).

Conclusions: It may be prudent to protect the airway with endotracheal intubation for patients with morbid obesity and diabetes using GLP-1, especially in the lithotomy position.

Keywords: ambulatory surgery; desaturation; diabetes mellitus; gastroparesis; glucagon-like peptide-1; laryngeal mask airway; morbid obesity; subclinical aspiration pneumonia.

Figure 1. A flowchart outlining the inclusion and exclusion of participants in the study

Figure 2. Box plot on left: The Y axis shows fasting glucose concentration (mg/dL). The mean glucose was 130 mg/dL ± 49 in the group with < 7 days GLP1-hold (N=29 had available pre-Glc data) versus 138 mg/dL ± 39 in the group with ≥ 7 days GLP1-hold (N=67 had available pre-Glc data) (two-tailed unpaired t-test p=0.69).

Box plot on the right: The Y axis shows the final (just prior to discharge home) oxygen saturation. The first box represents subjects with < 7 days of GLP1-hold, with data available in all 31 patients. The second box represents subjects with ≥ 7 days of GLP-1 hold, with data available in all 76 patients. The central line in the box plot is the median, the top of the box is the third quartile, the bottom of the box is the first quartile, and the whiskers are 1.5 times the interquartile range. pre-Glc: fasting glucose from a fingerstick glucometer immediately preoperatively

Figure 3. Correlation coefficient by the R statistical program

The X axis is the number of days the GLP-1 medication was held. The Y-axis is the fasting glucose in mg/dL. The R value is the Pearson moment correlation coefficient, R.

Figure 4. Distribution of oxygen saturation among BMI categories

The Y axis is the final oxygen saturation recorded just prior to discharge from the recovery room to home. The x-axis is the BMI (weight in kg divided by meters squared of height).

Figure 5. Clinical impact of DSpO2 = 4%. The red arrow points to acute pulmonary aspiration of gastric contents within the right upper lobe.