Characterization of Deposition from Nasal Spray Devices Using A Computational Fluid Dynamics Model of The Human Nasal Passages

Julia S. Kimbell, Rebecca Segal, Bahman Asgharian, Brian A. Wong, Jeffry D. Schroeter, Jeremy P. Southall, Colin Dickens, Geoff Brace, Frederick J. MillerJournal of Aerosol Medicine 2007 · DOI

TL;DR

This study uses a computational fluid dynamics (CFD) model of the human nasal passages to investigate how nasal spray characteristics and airflow affect drug deposition. The findings reveal that smaller particles, lower spray velocities, and the presence of inspiratory airflow significantly improve drug penetration past the nasal valve, offering crucial insights for designing more effective nasal drug delivery devices.

What Problem Does This Paper Address?

Nasal drug delivery holds great promise for both local and systemic treatments, offering advantages like rapid absorption and avoidance of first-pass metabolism. However, its effectiveness is often limited by the high deposition of drug particles in the anterior part of the nose, particularly before reaching the nasal valve – a critical anatomical region for deeper penetration. This premature deposition leads to wasted medication and potential side effects. While previous studies have explored various factors influencing nasal spray deposition, a comprehensive understanding of how multiple spray characteristics (like particle size and velocity) interact with nasal airflow to impact penetration, especially past the nasal valve, has been lacking. This paper aims to fill this gap by using a detailed CFD model to systematically analyze these interactions.

Methods

The researchers employed a three-dimensional, anatomically accurate CFD model of the human nasal passages, constructed from MRI scans. This model was meshed to represent the complex geometry. Airflow simulations were performed using steady-state inspiratory airflow at 15 L/min. For particle transport, they utilized in-house software that solved the equations of motion for particles, considering inertial forces as dominant for the particle sizes studied (20- and 50-µm).

Crucially, they integrated experimental data from 18 commercially available nasal sprays, characterizing their spray angles and velocities. This experimental data was used to define the parameters for their simulations. They then systematically varied several factors:

  • Nozzle Position: 0.5, 1.0, and 1.5 cm into the nostril.
  • Particle Size: 20 µm and 50 µm.
  • Spray Velocity: 1 m/sec and 10 m/sec.
  • Spray Angle: 32° and 79°.
  • Inspiratory Airflow: Present (15 L/min), absent, or with a delayed onset (0.5 or 1 sec after particle release).

This systematic variation allowed them to explore 48 different simulation cases to understand the sensitivity of nasal valve penetration to these parameters.

Datasets

The study utilized experimental data from 18 commercially available nasal spray devices, which provided measurements for spray characteristics such as spray angle and velocity. This data was used to inform the parameters of the CFD simulations. Additionally, the CFD model itself was validated against published experimental measurements of airflow and particle deposition in nasal models and human subjects. Specifically, they compared their CFD predictions for passive particle release with experimental data for 6- to 10-µm particles from studies by Pattle (1961), Hounam et al. (1971), and Kelly et al. (2004).

Key Insights and Ideas

  • The Nasal Valve is a Major Hurdle: The study reinforces that the nasal valve is a significant anatomical barrier to effective nasal drug delivery, with most particles depositing anteriorly.
  • Airflow is King (for Penetration): The presence of inspiratory airflow dramatically improved nasal valve penetration, highlighting that active breathing plays a crucial role in drawing medication deeper into the nasal passages.
  • Smaller is Better (for Penetration): Smaller particle sizes (20 µm vs. 50 µm) showed a marked increase in their ability to penetrate the nasal valve.
  • Gentle Spraying is Key: Lower spray velocities were more effective at achieving nasal valve penetration, suggesting that forceful sprays can lead to premature deposition.
  • The "Sweet Spot" for Nozzle Placement: A nozzle position of 1.0 cm into the nostril appeared to be optimal for nasal valve penetration compared to 0.5 cm or 1.5 cm.
  • Timing Matters for Airflow: Delaying the onset of inspiratory airflow by even 0.5 seconds shifted deposition sites from dorsal to ventral locations. Delays longer than 1 second rendered the airflow's benefit negligible, essentially mimicking a no-airflow scenario.
  • "Sweet Spots" for Release: The study hinted at the existence of specific "sweet spots" within the nasal vestibule where particle release could lead to higher penetration, suggesting patient-specific variations might be important.

Key Results

  • High Anterior Deposition: For 20- and 50-µm particles, deposition fractions in the anterior part of the nose exceeded 90% in most simulated conditions.
  • Limited Nasal Valve Penetration: Under most simulated conditions, no more than about 20% of spray particles were predicted to pass the nasal valve.
  • Significant Factors for Penetration: Statistical analysis showed that decreased particle size, decreased spray velocity, and a nozzle position of 1.0 cm significantly improved nasal valve penetration (p < 0.001). The presence of inspiratory airflow was also highly significant (p < 0.001).
  • Spray Angle's Minor Role: While a slight increase in penetration was observed with larger spray angles, this effect was not statistically significant.
  • Airflow Delay Impact: A 0.5-second delay in inspiratory airflow shifted deposition from the middle turbinate to the inferior turbinate. Delays exceeding 1 second negated the benefits of airflow.
  • CFD Model Validation: The CFD model's predictions for passive particle deposition of 6- to 10-µm particles at the nostrils showed good agreement with experimental measurements.

Limitations and Future Work

The authors acknowledge several limitations:

  • Rigid Nasal Walls: The CFD model used rigid nasal walls, neglecting the dynamic nature of nasal tissues.
  • No Mucus or Hairs: The model did not account for mucus movement or nasal hairs, which can influence particle deposition and clearance.
  • No Heat/Humidity: The simulation did not include heat and humidity transfer, which can affect particle behavior.
  • Simplified Airflow: Steady-state airflow was used, not accounting for the unsteady nature of breathing.
  • Particle-Particle Interactions: The study assumed negligible particle-to-particle interactions, which might be relevant at higher particle concentrations.
  • Simplified Spray Dynamics: The duration of the spray, interactions with airflow during actuation, and particle-to-particle interactions were not simulated.

Future work suggested includes:

  • Investigating the effects of particle sizes smaller than 20 µm.
  • Exploring the impact of abnormal nasal geometries.
  • Further studying the influence of spray duration on particle release and interactions.
  • Examining the potential effects of time-dependent airflow, nonuniform airflow conditions, and regional heat/humidity.
  • Understanding inter-individual differences in nasal anatomy.
  • Investigating the effects of mesh refinements on simulation accuracy, particularly for smaller particles.
  • Exploring the concept of "sweet spots" for particle release in more detail, considering patient-specific anatomy.

Summary updated

Abstract

Many studies suggest limited effectiveness of spray devices for nasal drug delivery due primarily to high deposition and clearance at the front of the nose. Here, nasal spray behavior was studied using experimental measurements and a computational fluid dynamics model of the human nasal passages constructed from magnetic resonance imaging scans of a healthy adult male. Eighteen commercially available nasal sprays were analyzed for spray characteristics using laser diffraction, high-speed video, and high-speed spark photography. Steadystate, inspiratory airflow (15 L/min) and particle transport were simulated under measured spray conditions. Simulated deposition efficiency and spray behavior were consistent with previous experimental studies, two of which used nasal replica molds based on this nasal geometry. Deposition fractions (numbers of deposited particles divided by the number released) of 20- and 50-microm particles exceeded 90% in the anterior part of the nose for most simulated conditions. Predicted particle penetration past the nasal valve improved when (1) the smaller of two particle sizes or the lower of two spray velocities was used, (2) the simulated nozzle was positioned 1.0 rather than 0.5 or 1.5 cm into the nostril, and (3) inspiratory airflow was present rather than absent. Simulations also predicted that delaying the appearance of normal inspiratory airflow more than 1 sec after the release of particles produced results equivalent to cases in which no inspiratory airflow was present. These predictions contribute to more effective design of drug delivery devices through a better understanding of the effects of nasal airflow and spray characteristics on particle transport in the nose.
BibTeX
@article{kimbell2007characterization,
  author = {Julia S. Kimbell and Rebecca Segal and Bahman Asgharian and Brian A. Wong and Jeffry D. Schroeter and Jeremy P. Southall and Colin Dickens and Geoff Brace and Frederick J. Miller},
  title = {{Characterization of Deposition from Nasal Spray Devices Using A Computational Fluid Dynamics Model of The Human Nasal Passages}},
  journal = {Journal of Aerosol Medicine},
  year = {2007},
  doi = {10.1089/jam.2006.0531},
}
Figure