Comparative analysis of videofluoroscopy and pulse oximetry for aspiration identification in patients with dysphagia after stroke and non-dysphagics

Dysphagia is a prevalent symptom observed in acute stroke. Several bedside screening tests are employed for the early detection of dysphagia. Pulse oximetry emerges as a practical and supportive method to augment the existing techniques utilized during bedside swallowing assessments. Desaturation levels, as measured by pulse oximetry, are acknowledged as indicative of aspiration by certain screening tests. However, the predictive capability of pulse oximetry in determining aspiration remains a subject of controversy. The objective of this study was to compare aspiration and oxygen desaturation levels by time and aspiration severity in dysphagic patients compared to healthy controls. It also aimed to evaluate the accuracy of pulse oximetry by comparing it with VFSS findings in detecting aspiration in both liquid (IDDSI-0) and semi-solid (IDDSI-4) consistencies.

Eighty subjects (40 healthy and 40 acute stroke patients) participated. Patients suspected of dysphagia underwent videofluoroscopy as part of the stroke unit’s routine procedure. Baseline SpO₂ was measured before VFSS, and stabilized values were recorded. Sequential IDDSI-0 and IDDSI-4 barium tests were conducted with 5 ml boluses. Stabilized SPO₂ values were recorded during swallowing and 3-min post-feeding. Patients with non-dysphagia received equal bolus monitoring. Changes in SPO₂ during, before, and after swallowing were analyzed for each consistency in both groups.

The study revealed a statistically significant difference in SPO₂ between patients with dysphagia and controls for IDDSI-4 and IDSSI-0. In IDDSI-4, 20% of patients experienced SpO₂ decrease compared to 2.5% in control group (p = 0.013). For IDDSI-0, 35% of patients showed SpO₂ decrease, while none in the control group did (p = 0.0001). Aspiration rates were 2.5% in IDDSI-4 and 57.5% in IDDSI-0. In IDDSI-0, SpO₂ decrease significantly correlated with aspiration (p = 0.0001). In IDDSI-4, 20.5% had SpO₂ decrease without aspiration, and showing no significant difference (p = 0.613). Penetration-Aspiration Scale scores had no significant association with SpO₂ decrease (p = 0.602). Pulse oximetry in IDDSI-4 had limited sensitivity (0%) and positive predictive value, (0%) while in IDDSI-0, it demonstrated acceptable sensitivity (60.9%) and specificity (100%) with good discrimination capability (AUC = 0.83).

A decrease in SPO₂ may indicate potential aspiration but is insufficient alone for detection. This study proposes pulse oximetry as a valuable complementary tool in assessing dysphagia but emphasizes that aspiration cannot be reliably predicted based solely on SpO₂ decrease.